## 4.1 CONTENT ANALYSIS: OVERALL STRUCTURE AND LENGTH

## Source details

**Canonical URL:** [4.1 CONTENT ANALYSIS: OVERALL STRUCTURE AND LENGTH](https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026117-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2026/english/wpiea2026117-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2026/english/wpiea2026117-source-pdf.pdf.json)

---

### I. Introduction — role and evolution of fiscal communication
- Fiscal policy’s communicative role evolved:
  - Post-Second World War: budgets used to stabilize demand, support employment, and shape long-run growth.
  - 1980s–1990s: short-run stabilization shifted toward monetary authorities; fiscal policy associated with consolidation, rules-based budgeting, and medium-term debt control.
  - Past 15 years: fiscal policy reemerged as a central stabilization tool in response to the global financial crisis (GFC), the euro area sovereign debt crisis, the Covid-19 pandemic, and the inflation shock of 2021–2023.
- Noted communicative failures:
  - UK’s 2022 mini-budget crisis and recurrent U.S. debt ceiling standoffs.
- Contemporary importance:
  - Finance ministry narratives shape households’, firms’, and markets’ beliefs about government intentions and expectations formation.

### II. Purpose and scope of the paper
- Objective:
  - Characterize how governments narrate fiscal choices—signaling priorities, justifying trade-offs, and constructing rationale—rather than measure direct economic effects.
- Methodology:
  - Descriptive, politically grounded, unsupervised-learning style analysis using natural language processing tools focused on classification, clustering, and comparative characterization.
- Dataset:
  - Over 500 fiscal communications (budget documents, ministerial speeches, press releases) from the G7 countries between 2000 and 2024.

### III. Dataset construction and analytical tools
- Corpus scope:
  - Countries: Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States.
  - Time span: 2000–2023 or 2000–2024 depending on availability; corpus comprises several million words.
- Three channels included:
  - (1) the Budget Document
  - (2) the Budget Speech
  - (3) the Press Communiqué (or press release)
- Country-specific adjustments:
  - Japan: Japanese Public Finance Factsheet used as budget document proxy; Highlights of the Budget used as press-release analogue.
  - Germany: medium-term budget documents (current year plus three years) used for comparability.
  - Italy: historical budget documents with multi-year projections; binding authority applies only to the single fiscal year.
  - United States: Budget factsheet as press release proxy; president’s remarks used as sample of verbal fiscal communication.
- Translation:
  - Non-English documents translated using GPT-4o-mini; validated through parallel-text checks.
- Text preprocessing:
  - All texts converted to plain text; tables, charts, and non-substantive elements removed.
- Analytical toolkit and procedures:
  - Readability: Flesch–Kincaid index.
  - Rhetorical style: eloquence metrics (unique words ratio) and metaphorical richness via LLM informed by the VU Amsterdam Metaphor Corpus logic.
  - Topic modeling/classification: GPT-4o-mini to classify sentences into eight categories:
    1. growth and business climate
    2. tax and revenue
    3. debt and fiscal sustainability
    4. social spending
    5. infrastructure and investment
    6. defense and security
    7. climate and environment
    8. residual content
  - Spending/revenue classification: structured prompts and explicit rules to detect increases, decreases, or no change.
  - Sentiment on fiscal sustainability: sentence-level optimistic / neutral / cautionary classification benchmarked within countries.
  - Cross-format and cross-country disaggregation to isolate communicative choices from fiscal reality.

### IV. Key empirical finding: a stable but segmented communicative architecture
- Layered structure across formats:
  - Budget documents: technical blueprints heavy on macroeconomic assumptions and institutional commitments.
  - Ministerial speeches: reinterpret plans through emotionally resonant frames (examples: growth, fairness, responsibility) aimed at voters.
  - Press releases: distill key elements into sharp headlines for media uptake.
- Regularities:
  - Fiscal language across the G7 remains complex and frequently optimistic.
  - These patterns have been stable over two decades, reflecting institutional practices and political incentives.
- Risks from misalignment:
  - Selective emphasis, rhetorical optimism, or misalignment between formats can obscure trade-offs and undermine transparency.
  - Examples: UK 2022 mini-budget (market backlash); 2018 Italian statements that heightened investor attention.

### V. Structure and document length: format hierarchy and patterns (empirical specifics)
- Core stylized hierarchy in the G7 corpus (2000–2024):
  - Budget documents: mean ~60,000–80,000 words.
  - Speeches: ~3,000–5,000 words.
  - Press communiqués: ~500–1,500 words.
- Temporal and crisis-related patterns:
  - Budget documents lengthened significantly during the Covid-19 pandemic.
  - Press releases expanded during the Global Financial Crisis and evolved with media and digital norms.
  - Speeches remained relatively concise but showed slight growth during major reforms (examples: Japan’s 2014 VAT hike, Canada’s 2019 investment incentives).
- Country-specific variation:
  - UK and US: trend toward more concise budget documents in the early 2010s.
  - France: expanded technical annexes and narrative components in the early 2010s.
  - Canada: (2016–2018 and from 2022 onwards) dropped encyclopedic tax annexes and shifted toward narrative-focused presentation centered on the “middle class,” shedding hundreds of pages of technical detail.
  - Germany: has not yet introduced press releases as a communication tool; medium-term plans used for comparability are more concise than its annual budget documents.
  - Japan: concise but extremely stable formats.
- Sentence-structure differences by document type:
  - Budget documents: ~25–30 words per sentence.
  - Speeches: ~20 words per sentence.
  - Press releases: often contain surprisingly long sentences.
- Variance trends:
  - Average length relatively stable, but variance increases sharply beginning around 2016.

### VI. Linguistic clarity and readability (4.2)
- Readability metric:
  - Flesch–Kincaid (FK) index used; FK examples:
    - Score of 8: a 13–14-year-old could comprehend the content.
    - Score of 14: implies college-level proficiency.
- Empirical FK results (G7 corpus, 2020–23 and broader trends):
  - Budget speeches: cluster around FK levels of 10 to 12.
  - Budget reports and press communiqués: often between 13 and 15.
  - Monetary policy comparison:
    - Monetary policy statements in developed countries on average require 16 years of schooling (Bulíř et al. 2012).
    - ECB governors’ speeches: FK index on average at 14.5 over the last two decades (Bjerkander and Glas (2024)).
- Stability and institutional persistence:
  - FK scores show remarkable stability over time across G7 countries; minimal variation for budgets and press releases over the past two decades.
  - Ministerial speeches have broadly followed public-sector trends toward plain-language outreach.
- Jargon:
  - Terms such as “fiscal space,” “primary balance,” and program names are common in documents; speeches tend to explain or avoid jargon.
- Functional implications:
  - “Clarity gap” between speeches (political narrative) and documents (technical substance) can leave citizens missing constraints or analysts missing framing.
  - Many countries adopting user-friendly representations (interactive dashboards); France committed in March 2025 to include an explicit section on uncertainty and risk factors in the national budget implemented for the first time in the 2026 draft budget.

### VII. Eloquence and rhetorical style (4.3)
- Definition:
  - Eloquence = craft of persuasion; translates aggregates into narratives about priorities, fairness, intergenerational responsibility, and the future.
- Registers by format:
  - Budget documents: lowest rhetorical flourish and vocabulary diversity.
  - Budget speeches: higher vocabulary diversity; blend exposition and persuasion.
  - Press communiqués: historically highest eloquence scores; recently more restrained in some countries.
- Measurement:
  - Rhetorical tone via figurative language detection (LLM + VU Amsterdam Metaphor Corpus logic).
  - Linguistic richness proxied by vocabulary diversity (share of unique words to total words excluding stop-words).
- Empirical patterns:
  - Common-law countries (Canada, the UK, often the U.S.) use more narrative and metaphor.
  - Civil-law countries (France, Italy) favor structured exposition.
  - Japan: concise, formatted rhetorical mode.
  - Figurative language and metaphor usage peak during economic stress (post-GFC, COVID-19).
  - Climate references rise after the Paris Agreement, especially in Canada, France, Italy, and Germany.
- Quantitative note:
  - A sizeable portion of fiscal communication, typically 10–20 percent across formats, falls into the “other” category (procedural language, signposting, statutory references).

### VIII. Communicating policy goals (4.5)
- Role:
  - Public objectives signal priorities, anchor expectations, and set benchmarks for accountability (examples in source: “to halve the deficit within four years”; “to build an economy that works for everyone”).
- Methodology:
  - Sentences referencing aims, missions, priorities, or long-term intentions identified via a dictionary filter and grouped by topic using the topic-assignment approach; press releases excluded due to brevity.
  - Broader spending/revenue analysis identified over 190,000 fiscal statements across the G7 (see section 4.6).
- Findings — speeches vs documents:
  - Speeches: citizen-facing goals (social welfare, jobs, productivity, opportunity); broad forward-looking formulations.
  - Budget documents: institutional/technical vocabulary (fiscal sustainability, long-term debt management, tax reform); concrete targets (example: “reducing the debt-to-GDP ratio to 50 percent by 2030”).
- Climate goals:
  - Largely conveyed through documents (emission targets, green infrastructure); speeches reference climate more sparingly in moral terms.
- Cross-country distinctions:
  - Common-law countries: narrative, normative speeches.
  - Civil-law countries: procedural/legalistic goal language.
  - Germany: goals tied to fiscal rules (e.g., the Schuldenbremse).
- Observed communicative bias:
  - Structural bias toward "good news": communication weighted toward spending initiatives and emphasizes spending increases over discussion of revenue.
  - Spending cuts rarely discussed explicitly in speeches; often described vaguely (“streamlining,” “finding efficiencies,” “slowing the growth of expenditures”).
  - Pairing strategy: revenue increases are immediately linked to popular expenditures to soften perceived pain.

- Country-specific example and exact metric from source:
  - Canada: “60–70 percent of revenue-related sentences in Canadian speeches referred to tax cuts, credits, or reductions.”

### IX. Communication on public debt (4.7)
- Sentiment classification:
  - Debt-related sentences classified as optimistic, neutral, or cautionary.
- Aggregate tone:
  - Across the G7, ministries consistently adopt an optimistic tone on debt sustainability, including during periods of rising debt.
  - Crisis-era exceptions: 2009 communications acknowledged “record-high deficits”; 2020–21 Covid-19 communications adopted more negative wording but quickly paired warnings with optimism about recovery.
- Format differences:
  - Speeches: notably more upbeat than written documents.
  - Budget documents: more complete picture via tables and projections yet often framed reassuringly.
- Country-specific examples:
  - Canada: late 2000s and 2010s budgets emphasized having “the lowest debt in the G7,” even as debt rose after 2020.
  - UK: frequent projections that public debt would “peak next year and then decline,” with successive revisions.
  - Italy: cautionary framing during the euro-area sovereign debt crisis emphasizing structural vulnerabilities.
  - United States: comparatively muted debt discussion in formal budget documents; more references in speeches and fact sheets.
- Statistical tests and quantitative findings (t-statistics and topic shares reported exactly):
  - Paired t-tests (Table 1 — t-statistics: Budget vs Release | Budget vs Speech | Speech vs Release)
    - Readability: 4.7*** | 16.4*** | 13.9***
    - Sentence length: 2.2* | 12.3*** | 6.3***
    - Eloquence: 20.2*** | 29.4*** | 6.2***
    - Document length: 18.5*** | 17.8*** | 10.2***
    - Significance legend: *** p<0.001, ** p<0.01, * p<0.05
  - Table 2 — average differences in the share of sentences devoted to specified topics (Budgets | Releases | Speeches)
    - Spending > revenue (share in the text): 12.9*** | 22.2*** | 10.8***
    - Spending increase > Spending decrease: 24.6*** | 43.6*** | 27.4***
    - Revenue decrease > Revenue increase: -3.1 | -11.3 | 6.9**
    - Positive debt outlook > negative debt outlook: 9.5*** | 44.8*** | 30***
    - Note: one-sided t-tests used; stars indicate conventional levels of statistical significance.
- Empirical implications:
  - Policymakers devote more attention to spending than to revenue and emphasize spending increases and favorable debt dynamics.
  - Speeches place relatively greater emphasis on tax reductions than budget documents or press releases.

### X. Justifying trade-offs and distributional consequences
- Theoretical implications:
  - Procedural justice: explaining distributional impacts fosters legitimacy.
  - Crisis governance: communication should be iterative—monitor, adjust, learn.
  - Narrative economics: human-centered stories shape expectations and behavior.
  - Cultural sensitivity: tailor style and technicality to national norms.
  - Political economy: sequence disclosures and messages to balance short-term incentives and long-term credibility.
  - Institutional memory: standardize language and templates for continuity.
  - Participatory governance: two-way communication builds legitimacy.
- Practical guidance and policy recommendations (selected and exact phrasing preserved where present):
  - Explain distributional impacts explicitly: identify beneficiaries and payers, and justify why trade-offs are made.
  - Treat communication as process: monitor reactions, correct misperceptions, and update narratives as conditions change.
  - Build narrative capacity: combine data with human-centered stories and historical context to shape expectations.
  - Localize messaging: adapt tone, technicality, and formats to cultural and linguistic norms of target audiences.
  - Manage timing and staging: plan communications around budgetary cycles and partisan incentives to protect credibility.
  - Institutionalize practices: adopt standardized templates and cross-cycle continuity mechanisms for consistent signaling.
  - Engage stakeholders: implement consultations, digital Q&As, and briefings to solicit feedback and build legitimacy.

### XI. Principles and institutional measures for effective fiscal communication (excerpts)
- Adopt more conditional, staged communication: acknowledge short-run deterioration while clearly outlining medium-term consolidation paths.
- Ensure consistency in messaging across formats and time to align persuasion with clarity.
- Improve availability and quality of fiscal data: timely, reliable, and comprehensive reporting enables narratives to be evaluated and trusted.
- Adapt communication modes and pacing to digital media and AI environments; recognize social media fragmentation and generative AI scrutiny.
- Consider institutional reforms:
  - Structured fiscal communication regimes with explicit narrative anchors and plain-language summaries.
  - Expand the remit of independent fiscal councils to assess coherence and transparency of fiscal messaging (example noted: the UK’s Office for Budget Responsibility).

_Italicized source: Excerpted sections (4.1, Box 1, 4.2, 4.3, 4.5, 4.7, and related chapters) from the IMF Working Paper — Decoding Fiscal Messaging (Working Paper No. WP/2026/117)._

### 4.1 CONTENT ANALYSIS: OVERALL STRUCTURE AND LENGTH .....................................................................

### 4.1 CONTENT ANALYSIS: OVERALL STRUCTURE AND LENGTH

### I. Introduction — role and evolution of fiscal communication
- Fiscal policy historically used budgets to stabilize demand, support employment, and shape long-run growth trajectories in the decades following the Second World War.
- By the 1980s and 1990s, responsibility for short-run stabilization shifted toward monetary authorities; fiscal policy became more associated with consolidation, rules-based budgeting, and medium-term debt control.
- Over the past 15 years, fiscal policy reemerged as a central stabilization tool in response to: the global financial crisis (GFC), the euro area sovereign debt crisis, the Covid-19 pandemic, and the inflation shock of 2021–2023.
- High-profile episodes exposing communicative failures include the UK’s 2022 mini-budget crisis and recurrent U.S. debt ceiling standoffs.
- Fiscal communication now matters for expectations formation: narratives constructed by finance ministries shape households’, firms’, and markets’ beliefs about government intentions.

### II. Purpose and scope of the paper
- Objective: characterize how governments narrate fiscal choices—signaling priorities, justifying trade-offs, and constructing rationale—rather than measure direct economic effects (bond yields, inflation expectations, or household sentiment).
- Methodological stance: descriptive, politically grounded, unsupervised-learning style analysis using natural language processing tools focused on classification, clustering, and comparative characterization.
- Dataset: over 500 fiscal communications (budget documents, ministerial speeches, press releases) from the G7 countries between 2000 and 2024.

### III. Key empirical finding: a stable but segmented communicative architecture
- Layered structure across formats:
  - Budget documents: technical blueprints heavy on macroeconomic assumptions and institutional commitments.
  - Ministerial speeches: reinterpret plans through emotionally resonant frames (examples: growth, fairness, responsibility) aimed at voters.
  - Press releases: distill key elements into sharp headlines for media uptake.
- Regularities observed:
  - Fiscal language across the G7 remains complex and, in many cases, consistently optimistic.
  - These patterns have been remarkably stable over the past two decades, reflecting institutional practices and political incentives.
- Risks from misalignment:
  - Selective emphasis, rhetorical optimism, or misalignment between formats can obscure trade-offs and undermine transparency.
  - Examples of communicative failure with real effects: UK 2022 mini-budget (market backlash) and 2018 Italian government statements that heightened investor attention.

### IV. Conceptual framing: signaling, institutional constraints, and narratives
- Fiscal communication functions as signaling under incomplete information—announcements signal intentions, institutional strength, and credibility (references: Holmström 1982; Morris and Shin 2002).
- Institutional/political constraints produce a “communication constraint”:
  - Finance ministries face electoral cycles, coalition dynamics, bureaucratic turnover, fiscal rules, and distributional trade-offs.
  - Unlike central banks (with a focal operational target such as an inflation target), finance ministries lack a single operational anchor.
- Ambiguity as strategy:
  - Ambiguity can be purposeful—used to shield unpopular trade-offs and preserve flexibility (example reframings: “rebalancing,” “closing loopholes,” “efficiency gains”).
- Audience heterogeneity:
  - Markets: clarity on debt trajectories and financing plans.
  - Legislators: granular details.
  - Journalists: accessible soundbites.
  - Public: reassurance, fairness, national purpose.
- Narrative importance:
  - Frames (fairness, prudence, responsibility, renewal) shape acceptance; e.g., tax increases framed as “intergenerational fairness” versus “austerity” lead to different receptions.
  - Narratives mediate trade-offs and political support (Shiller 2017; Rodrik 2014).

### V. Communication environment and consequences
- Modern media dynamics:
  - Traditional budget presentations now compete with livestreaming, social media, and 24-hour news cycles—amplifying risks from slips or poorly worded statements.
- Macro consequences:
  - Fiscal announcements can move yields, exchange rates, and perceptions of creditworthiness; inconsistency can undermine confidence even when fundamentals are unchanged.
- Emergent view:
  - Coherent fiscal narratives become a form of macroeconomic capital; consistent, candid messaging tends to support public legitimacy and market access (Allen, Schiavo-Campo, and Killen 2015).

### VI. Organization of the paper (sections referenced)
- Section 2: conceptual framework (political economy, signaling theory, narrative economics).
- Section 3: dataset and methodology.
- Section 4: empirical findings (this chapter begins that presentation).
- Section 5: discussion and policy implications.
- Section 6: conclusion.
- Annex I: prompts used for GPT-4 analysis.
- Annex II: country level analysis.
- References follow (section and pagination indicated in source content).

*Source: IMF Working Paper excerpt (sections I–II and table of contents) from "Decoding Fiscal Messaging".*

### Box 1: Lessons from Elsewhere: What Fiscal Communicators Can Learn from Science, Health, and Crisis

### Box 1: Lessons from Elsewhere: What Fiscal Communicators Can Learn from Science, Health, and Crisis

### Response: key communication lessons from public health, climate science, and crisis management
- Communicating fiscal policy involves navigating uncertainty, anticipating reactions, and building trust across audiences with vastly different levels of expertise, interest, and skepticism.
- Start with trust, not technicality:
  - Cognitive research shows people assess messages based on perceived credibility of the source (Grimmelikhuijsen et al. 2013).
  - During the pandemic response, clear acknowledgment of uncertainty enhanced confidence in the communicator.
  - Recommendation: Fiscal institutions should be clear about assumptions, limits, and risks to build credibility, especially in uncertain environments.
- Frame complexity through story, not abstraction:
  - Behavioral and communication sciences show narratives, not data alone, shape how people make sense of information (Shiller 2019; Reyna and Brainerd 2007).
  - Example: Framing a fiscal deficit as either a threat to stability or an investment in the future yields radically different public responses.
  - Recommendation: Deploy metaphors, values, and reference points that resonate with citizens' lived experience.
- Different audiences, different messages:
  - Effective communicators use layered strategies: technical annexes for experts; press briefings for journalists; visual explainers for the public; real-time updates for social media.
  - Warning: One-size-fits-all messaging risks speaking to no one.
- Consistency matters as much as content:
  - Inconsistent institutional voices (between ministries, parliaments, and heads of government) can weaken message discipline and sow confusion.
  - Recommendation: Coordinate and maintain temporal consistency across institutional communications, even when policy is evolving.
- Show the values behind the numbers:
  - People seek competence and care (Haidt 2012).
  - Trust is earned by acknowledging trade-offs, speaking to fairness, and engaging moral concerns.
  - Recommendation: Explain distributional rationales for prioritizing or protecting specific groups to improve public acceptance.
- Make communication a dialogue, not a monologue:
  - Climate adaptation, disaster relief, and health policy have evolved to incorporate feedback loops and community engagement (Dietz and Stern 2008).
  - Recommendation: Adopt listening mechanisms (e.g., public consultations, citizen summaries, social media analysis) to adjust messaging and address misperceptions in real time.
- Overarching point:
  - These lessons demand rigor, empathy, and strategic clarity — not spin — because how ministries speak can determine whether policies are understood at all.

### Data and Methodology: corpus construction and analytical tools
- Corpus scope and rationale:
  - New corpus of fiscal communication covering the G7 economies from 2000 to 2024.
  - Three channels included as the most consistent components of the annual budget cycle:
    - (1) the Budget Document
    - (2) the Budget Speech
    - (3) the Press Communiqué (or press release)
  - Rationale: Each item conveys the same underlying fiscal policy but is tailored to a distinct audience and exhibits different rhetorical structures and informational priorities.
- Definitions and roles of the three channels:
  - Budget document:
    - Most technical; typically several hundred pages.
    - Lays out revenue and expenditure projections and breakdown, macroeconomic assumptions, debt-management strategies, and detailed architecture of policy measures.
    - Targets parliamentarians, economists, credit rating agencies, and international institutions (e.g., IMF, World Bank).
    - Focus: initial budget drafts; budget revisions during the year are excluded.
  - Budget speech:
    - Explicitly narrative; delivered by the finance minister.
    - Distils complex fiscal measures into a story for the broader public and political stakeholders.
    - Often more optimistic and simpler linguistically than technical documents; selectively emphasizes themes to persuade.
  - Press communiqué (press release):
    - Leanest format; sometimes just a page or two summarizing headline measures and framing.
    - Designed for rapid dissemination to news outlets and market analysts; provides concise, quotable content.
    - Acts as a gateway shaping how intermediaries frame and broadcast the fiscal narrative.
- Coverage and languages:
  - Countries: Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States.
  - Time span: nearly a quarter century (2000–2023 or 2024, depending on availability).
  - Corpus size: comprising several million words.
  - Non-English documents (notably France, Germany, Italy) were translated using GPT-4o-mini, validated through parallel-text checks to ensure consistency in linguistic and topical patterns1.
  - Caveat: inherent differences between native English texts and auto-translated communication may remain and do not necessarily reflect differences in effectiveness in the original language.
- Country-specific adjustments:
  - Japan:
    - Used the Japanese Public Finance Factsheet as a proxy for the national budget document.
    - Highlights of the Budget serve as the closest analogue to a press-release3.
  - Germany:
    - Used medium-term budget documents (current year plus three years) as the most comparable descriptive content to budget documents in other countries.
  - Italy:
    - Budget documents used historically featured multi-year fiscal projections; binding authority applies only to the single fiscal year4.
  - United States:
    - Budget factsheet treated as a proxy for press release; president’s remarks announcing the budget used as a sample of U.S. verbal fiscal communication5.
  - Note on archival gaps:
    - In Germany, France, and Italy, press releases were not consistently available in a standardized, retrievable format across 2000–2022; future work could recover these through targeted archival collection.
- Text preprocessing:
  - All texts converted to plain text and pre-processed to remove tables, charts, and non-substantive elements (e.g., tables of contents, legal footers).
  - Goal: focus on communicative substance rather than administrative artifacts.
- Analytical toolkit and procedures:
  - Readability metrics (Section 4.2):
    - Flesch–Kincaid index used to assess linguistic complexity across countries and formats.
  - Rhetorical style (Section 4.3):
    - Eloquence metrics: ratio of unique words to total words.
    - Metaphorical richness assessed using LLM with a prompt reflecting the VU Amsterdam Metaphor Corpus logic.
  - Topic modeling and classification (Section 4.4):
    - Unconstrained topic identification for each sentence using GPT-4o-mini.
    - Aggregated classification into eight broad categories:
      1. growth and business climate
      2. tax and revenue
      3. debt and fiscal sustainability
      4. social spending
      5. infrastructure and investment
      6. defense and security
      7. climate and environment
      8. residual content
    - GPT-4o-mini used to classify each sentence into one of the predefined categories (see Annex I).
  - Spending and revenue classification (Section 4.6):
    - Spending-related sentences drawn primarily from topics 4-6; revenue from topic 2; supplemented by debt-related sentences (topic 3).
    - Structured prompts and explicit classification rules determine whether each sentence signals an increase, decrease, or no change in spending or revenue.
    - Rationale: Without explicit rules, large language models misinterpret numerical content, introducing systematic noise.
  - Sentiment and tone of fiscal sustainability communication (Section 4.7):
    - Sentence-level sentiment analysis to classify fiscal risk portrayal as optimistic, neutral, or cautionary.
    - Detailed prompt guides GPT-4o-mini; tone benchmarked within countries to account for stylistic norms.
  - Cross-format and cross-country disaggregation:
    - Compared how ministries publish fiscal content and how presentation structures vary (e.g., UK and Canada more integrated and accessible ecosystems with online portals and plain-language summaries).
- Analytical premise and empirical strategy:
  - Fiscal communication encodes political incentives, institutional constraints, and strategic considerations.
  - Disaggregate by country and communication format6 to examine whether ministries systematically adjust tone, complexity, or thematic emphasis depending on audience.
  - Within-country, within-year variation across the three channels is used as a core empirical strategy to isolate communicative choices from underlying fiscal realities.

### Empirical findings: preview and analytical framing
- Analytical approach:
  - Descriptive but systematic quantification of textual features across countries, formats, and time.
  - Primary methods: unsupervised learning techniques, including topic modeling and sentiment analysis.
  - Purpose: diagnose ideological framing in fiscal communication rather than predict economic responses.
- Four-stage analysis roadmap:
  1. Examine how fiscal communication is structured: document length, sentence structure, reflection of intended audiences.
  2. Assess linguistic clarity using readability metrics.
  3. Explore substantive content: dominant themes and goals, attention shifts across policy domains, differences across formats.
  4. Analyze how fiscal stances are framed: portrayal of spending, taxation, and public debt in directional terms and sentiment used to characterize fiscal risk.
- Key empirical observation:
  - Fiscal communication is patterned, layered, and remarkably stable, yet responsive to moments of crisis and shifts in political economy.
  - Ministries of Finance frame economic choices, justify trade-offs, and attempt to shape expectations through these structured narratives.
- Next analytical step:
  - The paper proceeds to present evidence beginning with document length and sentence-level characteristics as the simplest dimension of communication.

*IMF WORKING PAPERS Decoding Fiscal Messaging — Box 1: Lessons from Elsewhere: What Fiscal Communicators Can Learn from Science, Health, and Crisis.*

### 4.1 Content Analysis: Overall Structure and Length

### 4.1 Content Analysis: Overall Structure and Length

### Structure and document length: format hierarchy and patterns
- Core stylized hierarchy in the G7 corpus (2000–2024):
  - Budget documents: mean ~60,000–80,000 words.
  - Speeches: ~3,000–5,000 words.
  - Press communiqués: ~500–1,500 words.
- Temporal and crisis-related patterns:
  - Budget documents lengthened significantly during the Covid-19 pandemic.
  - Press releases expanded during the Global Financial Crisis and evolved with media and digital norms.
  - Speeches remained relatively concise but showed slight growth during major reforms (example references: Japan’s 2014 VAT hike, Canada’s 2019 investment incentives).
- Country-specific and institutional variation:
  - Trend toward more concise budget documents in the United Kingdom and United States in the early 2010s (partly driven by digital presentation formats).
  - France expanded technical annexes and narrative components during the same period.
  - Canada (2016–2018 and from 2022 onwards) dropped encyclopedic tax annexes and shifted toward narrative-focused presentation centered on the “middle class,” shedding hundreds of pages of technical detail.
  - Germany has not yet introduced press releases as a communication tool.
  - Japan maintains concise but extremely stable formats.
  - Germany’s medium-term budget documents chosen for cross-country comparability are much more concise than its annual budget documents; the latter exceed the length of budget documents in other G7 countries.
- Sentence-structure differences by document type:
  - Budget documents: ~25–30 words per sentence (information-dense, legal-administrative prose).
  - Speeches: ~20 words per sentence (shorter, rhetorically calibrated).
  - Press releases: often contain surprisingly long sentences, compressing maximum information into minimum space.
- Variance trends:
  - Average length of fiscal communications remains relatively stable, but variance increases sharply beginning around 2016.
  - The rise in variance does not coincide with the Global Financial Crisis (GFC); appears linked to post-crisis institutionalization of fiscal responses and divergent administrative choices.

### Communicative functions and implications
- Functional mapping of formats to audiences:
  - Budget documents: technocratic scaffolding for credibility and compliance; targeted to experts.
  - Speeches: legitimize policy choices through storytelling and emotive framing; targeted to the public.
  - Communiqués: gatekeepers of first impressions; targeted to journalists and analysts; can now exceed length of typical monetary policy statements in some countries.
- Interpretive trade-offs:
  - Scope and structure are rhetorical choices revealing how governments seek to be understood.
  - Document length and sentence complexity encode communicative intent and audience segmentation.
  - High word counts and syntactic clarity do not guarantee substantive transparency; they can coincide with rhetorical complexity that obfuscates fiscal risks.
- Institutional innovations and responses:
  - France’s March 2025 action plan committed to include an explicit section on uncertainty and risk factors in the national budget — implemented for the first time in the 2026 draft budget — as a step to improve communication of fiscal risks.
  - Many countries (examples: Canada, the UK, Germany, France) are addressing user-friendliness through interactive dashboards that summarize key fiscal data.

---

### 4.2 Linguistic Clarity and Readability

### Readability metrics and cross-format findings
- Readability operationalized using the Flesch-Kincaid (FK) index (estimates years of schooling required to understand a text).
  - FK interpretation examples:
    - Score of 8: a 13–14-year-old could comprehend the content.
    - Score of 14: implies college-level proficiency.
- Empirical FK results for the G7 corpus (2020–23 and broader trends):
  - Budget speeches: cluster around FK levels of 10 to 12.
  - Budget reports and press communiqués: often between 13 and 15.
  - Monetary policy statements and related monetary communication context for comparison:
    - Monetary policy statements in developed countries on average require 16 years of schooling (Bulíř et al. 2012).
    - ECB governors’ speeches: FK index on average at 14.5 over the last two decades (Bjerkander and Glas (2024)).
- Stability and institutional persistence:
  - Remarkable stability of FK scores over time across G7 countries; minimal variation for budgets and press releases over the past two decades.
  - House styles, legal drafting conventions, internal templates, and regulatory requirements reinforce inertia in linguistic form.
  - Only ministerial speeches have broadly followed public-sector trends toward plain-language outreach.
- Role of jargon and acronyms:
  - Finance ministries widely use jargon (examples: “fiscal space,” “primary balance,” country-specific program names).
  - Speeches tend to explain or avoid jargon; documents lean into jargon assuming knowledgeable readers.
  - Analyses of jargon and acronyms reinforced FK score indications.

### Functional division of linguistic labor and policy implications
- Division of labor:
  - Speeches as translators: render technical content intelligible to the public and provide political rationale.
  - Budget documents as formal accountability instruments: dense, comprehensive, rule-bound, targeted to experts.
  - Press releases occupy a middle ground: concise but not necessarily simple; often aimed at journalists and analysts.
- Consequences for understanding and accountability:
  - Clarity matters for public understanding, trust, expectation formation, and accountability.
  - A “clarity gap” exists between speech (political narrative) and document (technical substance); citizens who hear only the speech may miss key constraints, while analysts who read only the document may miss framing.
- Policy questions and practice:
  - Should governments simplify technical materials via structured citizen summaries or layered formats?
  - Many countries are already adopting more user-friendly representations (interactive dashboards), though degree of detail and focus vary markedly.
  - No single optimal model identified; trade-offs exist between precision required for experts and accessibility needed for broader public comprehension.

---

*Source: IMF Working Paper — Decoding Fiscal Messaging (4.1 Content Analysis: Overall Structure and Length; 4.2 Linguistic Clarity and Readability).*

### 4.3 Eloquence and Rhetorical Style

### 4.3 Eloquence and Rhetorical Style

### Definition and significance
- Eloquence is defined as the craft of persuasion: the capacity to articulate fiscal choices in a way that engages, convinces, and reassures audiences.
- Fiscal rhetoric is constitutive: it translates budgetary aggregates into narratives about national priorities, fairness, intergenerational responsibility, and the future.
- Political economy literature cited: Habermas (1984); Rodrik (2014); Shiller (2017); Praet (2017); Assenmacher et al. (2021).

### Comparative rhetorical registers and institutional context
- Finance ministries must balance technical macroeconomic language with persuasive democratic accountability; unlike central banks, they operate within the democratic arena where legitimacy stems from contestation rather than insulation.
- Three primary registers across document types:
  - Budget documents: highly technical, standardized, constrained by legal and administrative templates; minimal rhetorical engagement.
  - Budget speeches: blend exposition and persuasion, often beginning with values (examples: “opportunity,” “resilience,” “discipline”); higher vocabulary diversity than budget documents.
  - Press communiqués: typically the most rhetorically dense and headline-ready; historically highest eloquence scores, though recent trends show more restrained language in some countries.

### Measurement approach
- Two empirical features used:
  - Rhetorical tone: persuasive devices such as metaphor, repetition, and emotive appeals, traced with ChatGPT and a prompt designed following algorithms from the VU Amsterdam Metaphor Corpus (see Annex I and Figure 5 in the source).
  - Linguistic richness: proxied by vocabulary diversity, the share of unique words relative to total words after excluding common stop-words.
- Operational caveat: figurative language detection combines lexicon tagging and semantic distance metrics and may exhibit time-varying sensitivity; manual validation and robustness checks are necessary.

### Empirical patterns and cross-country variation
- Consistent hierarchy across G7 (2000–2023):
  - Budget documents exhibit the lowest rhetorical flourish and vocabulary diversity (examples: France’s annex-heavy budgets; Japan’s Public Finance Factsheets; Germany’s medium-term plans).
  - Speeches show higher vocabulary diversity and narrative emphasis (examples: Canada mid-2010s framing around strengthening the middle class; UK during austerity and post-Brexit framing national transformation).
  - Press releases historically show the highest eloquence scores but have become more restrained in countries such as the UK and Canada.
- Format dependence: press releases most rhetorically sharpened; speeches mix narrative and exposition; budget documents remain formal.
- Cross-country stylistic differences:
  - Common-law countries (Canada, the UK, often the U.S.) use more narrative and metaphor.
  - Civil-law countries (France, Italy) favor structured exposition and codified style.
  - Japan maintains concise, highly formatted rhetorical mode.

### Dynamics over time and during crises
- Figurative language and metaphor usage peak during economic stress (examples: aftermath of the GFC and the COVID-19 pandemic).
- Shifts in rhetorical tone often coincide with political transitions, policy inflection points, or periods of economic stress.
- Climate and green investment references rise after the Paris Agreement, particularly in Canada, France, Italy, and Germany; the United Kingdom shows sustained attention to climate-related spending and policy.

### Topics, thematic emphasis, and format interaction (transition to Section 4.4)
- Topic modeling classified sentences into themes: growth and business climate; tax and revenue; debt and fiscal sustainability; social spending; infrastructure and investment; defense and security; climate and environment.
- Format-driven thematic allocation:
  - Budget documents: devote more space to taxation and expenditure mechanics.
  - Speeches: prioritize growth, jobs, and social spending.
  - Press releases: emphasize new initiatives (spending increases, tax relief, flagship projects) and tend to omit complex policy trade-offs.
- Presence of optimistic bias: press releases frequently highlight new initiatives while omitting trade-offs, contributing to a narrative of expansion and responsiveness.

### Quantitative and descriptive findings
- A sizeable portion of fiscal communication, typically 10–20 percent across formats, falls into the “other” category (procedural language, structural signposting, statutory references, narrative filler).
- Country-specific emphases:
  - US and UK: emphasize tax cuts and pro-growth measures in speeches.
  - France and Germany: greater weight on rules-based fiscal governance (debt brakes, Maastricht criteria).
  - Japan: consistent emphasis on demographic pressures, social stability, and long-term debt consolidation.

### Implications and risks
- Effective eloquence can:
  - Bolster public confidence.
  - Promote understanding.
  - Strengthen perceived coherence of fiscal policy and help navigate reforms and crises.
- Risks of excessive rhetorical polish:
  - Potential disconnect between narrative and substance.
  - Erosion of credibility when fiscal reality fails to match public storytelling.
- Trade-off: too little rhetorical engagement reduces public engagement; too much flourish risks credibility loss.
- Recent trend toward plainer press releases suggests ministries are recalibrating rhetorical strategies in response to media scrutiny and demands for credibility.

### Institutional factors and future research directions
- Fiscal rules may influence rhetorical discipline by constraining both behavior and rhetorical strategies; presence of rules could correlate with clearer communication of trade-offs and reduced optimistic framing.
- The operational definition of figurative language requires robustness checks and manual validation due to potential definitional drift over time.
- Future work could formally analyze whether fiscal rules and independent fiscal institutions correlate with rhetorical quality, coherence across formats, and alignment between narrative and arithmetic.

### Policy recommendations
- Improve fiscal transparency through more balanced communication that:
  - Celebrates initiatives while clearly explaining costs, trade-offs, and constraints.
  - Aligns narrative with technical fiscal projections to preserve credibility.
- Strengthen institutional mechanisms that bolster coherence:
  - Independent fiscal councils and rule-based frameworks can help align narrative and arithmetic.
- Monitor and validate rhetorical measures:
  - Use manual validation and robustness checks for automated figurative language detection to ensure temporal comparability.

*Source: IMF Working Paper section "4.3 Eloquence and Rhetorical Style" (authors’ calculations based on official public fiscal documents).*

### 4.5 Communicating Policy Goals

### 4.5 Communicating Policy Goals

### Role and purpose of goal communication
- Publicly stated objectives signal priorities, anchor expectations among households and markets, and set benchmarks for accountability.
- Goals provide narrative meaning to budgetary decisions (examples in source: “to halve the deficit within four years”; “to build an economy that works for everyone”).
- Clear goals:
  - reassure markets,
  - guide citizens,
  - reduce uncertainty about strategic direction.
- Theoretical rationale:
  - Economics: clearly communicated goals reduce policy uncertainty, shape forward-looking behavior, and strengthen credibility (analogy to inflation targeting).
  - Political science: goals legitimize executive decisions, bridge elites and citizens, reduce informational asymmetries, and enhance perceived fairness and transparency.

### Methodology used in the analysis
- Identification approach:
  - All sentences in budget documents and speeches across the G7 explicitly referencing aims, missions, priorities, or long-term intentions were identified using a simple dictionary method for filtering those words.
  - Sentences were grouped by goal topic using the same topic-assignment approach applied to all sentences in the texts.
  - Press releases were excluded due to brevity and emphasis on measures rather than overarching aims.
- Scope of statements analyzed:
  - The broader analysis of spending and revenue statements identified over 190,000 fiscal statements across the G7 (see section 4.6 for details on classification prompts in Annex I).

### Findings — differences between speeches and budget documents
- Speeches:
  - Overwhelmingly frame goals in citizen-facing terms: improving social welfare, creating jobs, boosting productivity, expanding opportunity, supporting families, investing in future generations.
  - Frequently use broad, forward-looking formulations: examples from source include “building an economy that works for everyone,” “laying the foundations for long-term prosperity,” “strengthening the middle class.”
  - Employed across Canada, the United Kingdom, France, Germany, and Japan to justify fiscal measures through appeals to fairness, resilience, or opportunity.
- Budget documents:
  - Tend to use institutional and technical vocabulary: fiscal sustainability, long-term debt management, tax reform, administrative efficiency, compliance with fiscal rules.
  - Examples cited: “reducing the debt-to-GDP ratio to 50 percent by 2030,” “meeting medium-term balance targets in line with the fiscal framework.”
  - Japan, Germany, and Italy embed such objectives in multi-year planning documents and legal mandates.
- Communicative divergence:
  - Speeches prioritize mobilizing sentiment and connecting with lived experiences.
  - Documents prioritize establishing credibility with markets, oversight institutions, and policy experts.

### Climate-related goals and communicative placement
- By the late 2010s and early 2020s, environmental themes gained prominence.
- Communicative asymmetry:
  - Climate goals were largely conveyed through documents (targeting emission reductions, green infrastructure investment, tax incentives).
  - Speeches referenced climate more sparingly or in moral terms (example: “protecting the planet for future generations”).
- Strategic salience:
  - During economic stress, ministries may downplay climate themes in speeches to foreground immediate concerns like cost-of-living or employment.

### Cross-country and administrative tradition differences
- Common-law countries (Canada, the UK, the United States):
  - Tend to frame objectives narratively and normatively in speeches, using accessible language emphasizing societal impact.
- Civil-law countries (France, Italy):
  - More often articulate goals in procedural or legalistic language, though sometimes use narrative tropes.
- Japan:
  - Fiscal goals consistently concise and structured, reflecting bureaucratic conventions.
- Germany:
  - Goals tied closely to fiscal rule frameworks (e.g., the Schuldenbremse), emphasizing compliance over vision.
- Conclusion:
  - Even under similar macroeconomic conditions (rising debt or inflation), countries frame goals through different institutional and cultural lenses.

### Integrating citizen-facing and expert-facing goals
- Two interrelated planes of goal-setting in fiscal communication:
  - Citizen-facing: social and economic aspirations.
  - Expert-facing: sustainability, efficiency, accountability.
- Effective communication challenge: integrate these perspectives into a unified, credible message.
- Successful goal articulation reinforces credibility, transparency, and public acceptability; it gives the public purpose and markets stability.

---

### Framing of spending, revenue, and fiscal stance (summary from adjacent section 4.6 as it relates to goal communication)
- Fiscal framing shapes perceptions of fairness, responsibility, and strategic intent; fiscal stances are both economic postures and rhetorical constructs.
- The analysis finds a structural bias toward "good news":
  - Across all G7 countries and across formats (budgets, speeches, press releases), communication is weighted toward spending initiatives and emphasizes spending increases over discussion of revenue.
  - Spending cuts are rarely discussed explicitly in speeches and are often described vaguely (“streamlining,” “finding efficiencies,” “slowing the growth of expenditures”) rather than named.
- Specific patterns:
  - Governments emphasize new or expanded initiatives with positive framing (“the largest investment in decades,” “support for families,” “unprecedented funding for innovation”).
  - Tax cuts receive prominent attention in press releases, especially in the United States, United Kingdom, and Canada.
  - When revenue increases occur (e.g., Japan’s VAT increases), rhetoric emphasizes purpose and pairs increases with mitigation measures or socially beneficial spending.
  - Revenue-neutral measures are frequently framed as wins (simplification presented as fairer or more efficient; maintaining tax rates recast as a positive choice: “we are not raising taxes”).
  - Pairing strategy: unavoidable revenue increases are immediately linked to popular expenditures to soften perceived pain.
- Country-specific notes and figures:
  - Canada: post-2015 budgets were branded around the “middle class,” with heavy emphasis on new investments; mentions of deficit concerns nearly disappeared from speeches. Exact metric from source: “60–70 percent of revenue-related sentences in Canadian speeches referred to tax cuts, credits, or reductions.”  
  - United Kingdom example: during 2010–2011 consolidation, rhetoric avoided the term “cuts,” favoring words such as “responsibility” and “opportunities for efficiency savings.”
  - Japan: VAT hikes in 2014 and preparation for 2019 were framed around securing budgets for social security and stabilizing long-term finances.
- Transparency implications:
  - Absence of explicit discussion of cuts and hikes can obscure trade-offs; citizens relying solely on public-facing communication may miss what was sacrificed to finance new spending.
  - Analysts and journalists face a heavier burden to uncover less favorable budget aspects.
- Central hypothesis supported:
  - Fiscal communication systematically favors expansionary, popular measures and deemphasizes contractionary, unpopular ones; quantification of this bias illuminates narrative strategies underpinning modern fiscal communication.

*Source: Excerpt from IMF Working Paper chapter "4.5 Communicating Policy Goals" (section and adjacent findings provided in the supplied content).*

### 4.7 Communication on Public Debt

### 4.7 Communication on Public Debt

### Overview
- Analysis identifies all sentences relating to debt, fiscal sustainability, borrowing needs, or medium-term fiscal trajectories and classifies sentiment as positive (reassuring or optimistic), negative (warning or concern), or neutral.
- Aggregated assessments across formats and over time reveal consistent patterns in G7 fiscal communication.

### Sentiment and Tone on Debt
- Across the G7, ministries consistently adopt an optimistic tone when communicating about debt sustainability, including in periods of rising debt.
- Typical formulations emphasized in the corpus include that the debt-to-GDP ratio will soon “stabilize,” that borrowing remains “manageable,” or that fiscal positions are “under control.”
- Recurrent optimism may reflect credible medium-term consolidation goals or favorable macroeconomic assumptions, but persistent optimism that outpaces outcomes can affect perceptions of credibility.
- A more somber tone appears during acute crisis moments:
  - 2009 communications acknowledged “record-high deficits.”
  - 2020–21 communications around the Covid-19 shock adopted more negative wording.
- Even in crisis years, negative wording is quickly paired with optimism about recovery (e.g., “borrowing will fall as the economy reopens”) or about institutional safeguards.

### Format Differences: Speeches, Budgets, Press Releases
- Speeches are notably more upbeat than written documents accompanying them.
  - Ministers often highlight the most flattering indicators (e.g., temporary dip in borrowing or favorable international comparison) and downplay less favorable dynamics.
- Budget documents provide a more complete picture with tables and projections that sometimes acknowledge rising debt or persistent deficits, but still often wrap facts in a reassuring narrative (examples: debt “remains manageable given prevailing interest rates”; deficits “reflect necessary investments”).
- The contrast between oral and written communication is consistent: public-facing messages emphasize positive narratives while technical documents retain complex details.
- Country-specific patterns:
  - Canada: budget communications in late 2000s and 2010s emphasized having “the lowest debt in the G7,” even as debt began to rise after 2020.
  - United Kingdom: frequent projections that public debt would “peak next year and then decline,” with successive revisions as macro developments required updates.
  - Italy: combined cautionary framing during the European sovereign debt crisis, emphasizing structural vulnerabilities and external euro-area influences.
  - United States: comparatively muted debt discussion in formal budget documents, with more references in speeches and fact sheets.

### Statistical Tests and Key Quantitative Findings
- Paired t-tests examine differences in readability, sentence length, lexical richness (eloquence), and document length across document types; results reported as t-statistics with conventional significance stars.

- Table 1 — t-statistics (Budget vs Release | Budget vs Speech | Speech vs Release)
  - Readability: 4.7*** | 16.4*** | 13.9***
  - Sentence length: 2.2* | 12.3*** | 6.3***
  - Eloquence: 20.2*** | 29.4*** | 6.2***
  - Document length: 18.5*** | 17.8*** | 10.2***
  - Significance legend: *** p<0.001, ** p<0.01, * p<0.05

- Table 2 — average differences in the share of sentences devoted to specified topics (Budgets | Releases | Speeches)
  - Spending > revenue (share in the text): 12.9*** | 22.2*** | 10.8***
  - Spending increase > Spending decrease: 24.6*** | 43.6*** | 27.4***
  - Revenue decrease > Revenue increase: -3.1 | -11.3 | 6.9**
  - Positive debt outlook > negative debt outlook: 9.5*** | 44.8*** | 30***
  - Note: one-sided t-tests used; stars indicate conventional levels of statistical significance.

- Empirical implications of the statistical results:
  - Budget documents are more technical; speeches and press releases are more accessible.
  - Policymakers tend to devote more attention to spending than to revenue and emphasize spending increases and favorable debt dynamics.
  - Speeches place relatively greater emphasis on tax reductions compared to budget documents or press releases.

### Implications for Credibility, Expectations, and Distribution
- Persistent emphasis on positive measures risks obscuring the full fiscal picture; omission of corresponding cuts or revenue hikes can lead the public to misunderstand distributional and macroeconomic effects.
- Prior literature suggests persistent optimism in official communication, if not matched by outcomes, can erode credibility over time.
- The divergence across formats (speeches/press releases vs technical documents) may create mixed messages: speeches target voters; documents target experts.
- Communicative procyclicality: rhetoric remains stronger than fundamentals in both good and bad times, risking complacency in booms and understatement of challenges in downturns.
- Distributional considerations: emphasizing benefits while underplaying delayed or indirect costs can disproportionately burden lower-income households and erode public trust.

### Policy Recommendations and Institutional Measures
- Adopt more conditional, staged communication: acknowledge short-run deterioration while clearly outlining medium-term consolidation paths to enhance multiplier effects and reduce precautionary behavior.
- Ensure consistency in messaging across formats and time to align persuasion with clarity.
- Improve availability and quality of fiscal data: timely, reliable, and comprehensive reporting enables narratives to be evaluated and trusted.
- Adapt communication modes and pacing to digital media and AI environments, recognizing that social media fragments message control and generative AI can parse and critique fiscal messaging in real time.
- Consider institutional reforms:
  - Structured fiscal communication regimes with explicit narrative anchors and plain-language summaries.
  - Expand the remit of independent fiscal councils to assess the coherence and transparency of fiscal messaging (example noted: the UK’s Office for Budget Responsibility as an illustrative institutional variation).

### Principles for Effective Fiscal Communication (selected from Box 2)
- Clarity of Message and Medium
  - Implication: offer layered communication—technical annexes for experts, clear summaries for citizens, timely accessible formats for media.
- Internal Consistency and Institutional Coherence
  - Implication: align fiscal messages with broader government narratives; avoid conflicting signals across institutions.
- Transparency with Uncertainty
  - Implication: report ranges, assumptions, and scenario analysis, not just point estimates, to build credibility and allow recalibration when shocks occur.

*Italicized source: Excerpt from IMF Working Paper chapter “4.7 Communication on Public Debt” (Decoding Fiscal Messaging).*

### 4. Justifying Trade-offs and Distributional Consequences

### 4. Justifying Trade-offs and Distributional Consequences

### Key theoretical insights and implications for fiscal communication
- Procedural justice theory (Tyler 2006): perceived fairness in process drives legitimacy, even when outcomes are unequal.
  - Implication: Finance ministries should explain distributional impacts, highlighting who benefits, who pays, and why.
  - Policy recommendation: Acknowledge equity trade-offs to foster social buy-in and reduce political backlash.

- Crisis governance (Ansell et al. 2010): successful communication requires monitoring, adjustment, and learning.
  - Implication: Fiscal communication should be iterative; track public and market reactions, clarify misunderstandings, and recalibrate messaging as facts or conditions evolve.
  - Policy recommendation: Embed internal feedback mechanisms to improve credibility and policy learning.

- Narrative economics (Shiller 2019): stories, not just data, shape expectations and collective behavior.
  - Implication: Humanize abstract trade-offs through narrative framing, analogies, and historical parallels.
  - Policy recommendation: Invest in storytelling capacity as a strategic imperative.

- Cultural and linguistic sensitivity (Hall 1976): high- vs. low-context cultural distinctions affect message interpretation and trust.
  - Implication: Tailor fiscal communication to national linguistic and cultural norms.
  - Policy recommendation: Avoid highly coded, technical language where it may alienate audiences; adapt style and format to improve resonance.

- Political economy constraints (Alesina and Perotti 1996): budgetary cycles and partisan incentives influence timing and content of communication.
  - Implication: Use strategic timing, staging, and narrative framing to navigate constraints without sacrificing long-term credibility.
  - Policy recommendation: Sequence disclosures and messages to align short-term political realities with long-term fiscal credibility.

- Institutional memory and continuity (Mahoney and Thelen 2010): legitimacy accrues through consistent behavior and signaling over time.
  - Implication: Standardize language, templates, and communication formats across political cycles to maintain continuity.
  - Policy recommendation: Institutionalize communication practices to build long-term credibility beyond party or personality.

- Participatory governance (Fung 2006): two-way communication builds legitimacy and compliance.
  - Implication: Fiscal communication should include public engagement mechanisms, consultations, digital Q&As, or stakeholder briefings.
  - Policy recommendation: Create feedback loops with citizens and intermediaries to reinforce trust and improve policy design.

### Practical guidance for finance ministries
- Explain distributional impacts explicitly: identify beneficiaries and payers, and justify why trade-offs are made.
- Treat communication as process: monitor reactions, correct misperceptions, and update narratives as conditions change.
- Build narrative capacity: combine data with human-centered stories and historical context to shape expectations.
- Localize messaging: adapt tone, technicality, and formats to cultural and linguistic norms of target audiences.
- Manage timing and staging: plan communications around budgetary cycles and partisan incentives to protect credibility.
- Institutionalize practices: adopt standardized templates and cross-cycle continuity mechanisms for consistent signaling.
- Engage stakeholders: implement consultations, digital Q&As, and briefings to solicit feedback and build legitimacy.

### Observed patterns and risks (from empirical analysis)
- Many advanced economies have stable communicative practices rooted in longstanding institutional routines; emerging markets may adopt integrated communication approaches earlier.
- Advanced economies (example cited) can still innovate by integrating behavioral insights and plain-language framing.
- In crisis episodes (pandemics, financial collapses, geopolitical shocks) communicative burden intensifies; governments often pivot quickly to reassurance.
  - Recommendation: Embrace scenario planning, contingency disclosure, and honest articulation of risks without inducing panic (Leeper 1993; Debrun and Kinda 2017).
- Systematic emphasis on policy benefits with limited discussion of costs can distort public understanding of fiscal trade-offs.
- Persistent optimism bias in official communications can undermine credibility if rhetoric diverges from macroeconomic outcomes.
- Divergences across communication formats (technical documents vs. public-facing narratives) can fragment public understanding and market interpretation.
- Presence of independent institutions such as fiscal councils has been linked to more consistent and transparent messaging (Debrun et al. 2013).
- Language that downplays distributional trade-offs or uses euphemistic terminology (e.g., “consolidation” or “efficiency savings”) can obscure full implications of fiscal measures (Kickert and Randma-Liiv 2015).

### Strategic priorities for strengthening fiscal communication
- Develop guidance on narrative best practices in addition to rules on fiscal frameworks (role for international organizations such as the IMF and OECD).
- Prioritize coherence across channels: align budget documents, ministerial speeches, and press releases to avoid mixed signals.
- Institutionalize contingency communication: prepare scenario-based messaging and contingency disclosures for crises.
- Enhance engagement tools: expand consultations and digital interaction to create two-way communication and policy feedback.

_Italic: Source — IMF Working Paper chapter: "4. Justifying Trade-offs and Distributional Consequences" (extracted from the PDF content provided)._

### References

### wpiea2026117-source-pdf - References

### Major scholarly themes represented
- Fiscal communication, debt, and public expectations
  - Aksoy, Yunus, Henrique S. Basso, Tobias J. Klein, and Christian Moser. 2025. “Speaking of Debt: Framing, Guilt, and Economic Choices.” CESifo Working Paper No. 12060.
  - Caldas Montes, G., R.T. da Fonseca Nicolay, and T. Acar. 2019. “Do fiscal communication and clarity of fiscal announcements affect public debt uncertainty? Evidence from Brazil.” Journal of Economics and Business, Vol. 103, pp. 38–60.
  - Beetsma, Roel, Massimo Giuliodori, and Francesco R. Pappa. 2021. “Fiscal Policy, Debt, and Economic Growth: The Role of Expectations.” Journal of Monetary Economics 118: 123–139.
  - Estefania-Flores, J., D. Furceri, S. Kothari, and J. D. Ostry. 2023. “Worse Than You Think: Public Debt Forecast Errors in Advanced and Developing Economies.” Journal of Forecasting, Vol. 42, No. 3, pp. 685–714.
  - Ostry, Jonathan D., Atish R. Ghosh, and Raphael Espinoza. 2015. When Should Public Debt Be Reduced? IMF Staff Discussion Note No. SDN/15/10.
  - Frankel, Jeffrey. “Over-Optimism in Forecasts by Official Budget Agencies and Its Implications.” Oxford Review of Economic Policy 27, no. 4 (2011): 536–62.
  - Frankel, Jeffrey, and Jesse Schreger. “Over-Optimistic Official Forecasts and Fiscal Rules in the Eurozone.” Review of World Economics / Weltwirtschaftliches Archiv 149, no. 2 (2013): 247–72.
  - Poplawski-Ribeiro, M. and J.C. Rülke. 2011. Fiscal Expectations Under the Stability and Growth Pact: Evidence from Survey Data. IMF Working Paper, WP/11/48, March.

- Fiscal rules, fiscal councils, and institutional design
  - Alonso, V., Arroyo, C., Aydin, O., Balasundharam, V., Davoodi, H.R., Hegab, G., Nguyen, A.M., Salazar Ferro, N., Sher, G., Solovyeva, A., and Tchelishvili, N. 2025. “Fiscal Rules and Fiscal Councils: Recent Trends and Revisions Since the Pandemic,” IMF Working Paper, WP/25/198, October.
  - Debrun, Xavier, Tidiane Kinda, Teresa Curristine, Luc Eyraud, Jason Harris, and Johann Seiwald. 2013. The Functions and Impact of Fiscal Councils. IMF Policy Paper.
  - Beetsma, R. W.R. Lam, E. Leonards, M. Poplawski-Ribeiro, and A. Solovyeva. 2026. “On the Role and Effectiveness of Fiscal Rules and Independent Fiscal Councils.” IMF Working Paper, forthcoming.

- Central bank communication, transparency, and market effects
  - Blinder, Alan S., Michael Ehrmann, Marcel Fratzscher, Jakob De Haan, and David-Jan Jansen. 2008. “Central Bank Communication and Monetary Policy: A Survey of Theory and Evidence.” Journal of Economic Literature 46 (4): 910–945.
  - Geraats, Petra M. 2002. “Central Bank Transparency.” Economic Journal 112 (483): F532–F565.
  - Woodford, Michael. 2005. “Central-Bank Communication and Policy Effectiveness.” In The Greenspan Era: Lessons for the Future, 399–474.
  - Ehrmann, Michael, and Jonathan Talmi. 2020. “Starting from a Blank Page? Semantic Similarity in Central Bank Communication and Market Volatility.” Journal of Monetary Economics 111 (2020): 48–62.
  - Hansen, Stephen, and Michael McMahon. 2016. “Shocking Language: Understanding the Macroeconomic Effects of Central Bank Communication.” In Proceedings of the NBER International Seminar on Macroeconomics 2015.
  - Bjerkander, Lena and Alexander, Glas. 2024. “Talking in a language that everyone can understand? Clarity of speeches by the ECB Executive Board.” Journal of International Money and Finance, Volume 149.
  - Assenmacher, Katrin & Glöckler, Gabriel & Holton, Sarah & Trautmann, Peter & Ioannou, Demosthenes & Mee, Simon & Alonso, Conception & Argiri, Eleni & Arigoni, Filippo & Bakk-Simon, Klára & Bergbauer, 2021. "Clear, consistent and engaging: ECB monetary policy communication in a changing world," Occasional Paper Series 274, European Central Bank.

- Text-as-data, sentiment analysis, and narrative methods
  - Grimmer, Justin, and Brandon M. Stewart. 2013. “Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts.” Political Analysis 21 (3): 267–297.
  - Ceron, Andrea, Luigi Curini, and Stefano Maria Iacus. 2016. “iSA: A Fast, Scalable and Accurate Algorithm for Sentiment Analysis of Social Media Content.” Information Sciences 367–368 (November): 105–124.
  - Beauchamp, Nicholas. 2017. “Predicting and Interpolating State-Level Polls Using Twitter Textual Data.” American Journal of Political Science 61 (2): 490–505.
  - D’Acunto, Francesco, Daniel Hoang, David Schoar, and Michael Weber. 2022. “Narrative Economics and Household Behavior: The Case of Fiscal Policy.” Review of Financial Studies 35 (9): 4427–4467.
  - Shiller, Robert J. 2017. “Narrative Economics.” American Economic Review: Papers and Proceedings 107 (4): 967–1004.
  - D’Acunto, Francesco, Yuriy Gorodnichenko, and Michael Weber. 2022. “What Do the Data Tell Us About Inflation Expectations?” ifo Working Paper No. 9602 (August).

- Political economy, accountability, and communication strategy
  - Alesina, Alberto, and Guido Tabellini. 2007. “Bureaucrats or Politicians?” Journal of Public Economics 91 (1–2): 187–213.
  - Besley, Timothy J., and Andrea Prat. 2006. “Handcuffs for the Grabbing Hand? Media Capture and Government Accountability.” American Economic Review 96, no. 3 (June): 720–736.
  - Hood, Christopher. 2011. The Blame Game: Spin, Bureaucracy, and Self-Interest. Oxford: Oxford University Press.
  - Weaver, R. Kent. 1986. “The Politics of Blame Avoidance.” Journal of Public Policy 6 (4): 371–398.
  - Rodrik, Dani. 2014. When Ideas Trump Interests: Preferences, Worldviews, and Policy Innovations. Journal of Economic Perspectives 28 (1): 189–208.
  - Rogoff, Kenneth. 1990. “Equilibrium Political Budget Cycles.” American Economic Review 80 (1): 21–36.
  - Kickert, Walter, and Tiina Randma-Liiv. 2015. Europe Managing the Crisis: The politics of fiscal consolidation London: Routledge.

- Forecasting, expectations, and measurement issues
  - Beetsma, Roel, Benjamin Bluhm, Massimo Giuliodori, and Peter Wierts. 2013. “From Budgetary Forecasts to Ex-Post Fiscal Data: Exploring the Sources of Revision Errors in the European Union.” Contemporary Economic Policy 31 (4): 795–813.
  - Chocobar, B., P. Claeys, and M. Poplawski-Ribeiro. 2025. “Fiscal Forecasting Rationality Among Expert Forecasters.” Journal of Forecasting, Vol. 44, No. 3, pp. 941–959.
  - Ramey, Valerie A. 2011. “Identifying Government Spending Shocks: It’s All in the Timing.” Quarterly Journal of Economics 126 (1): 1–50.
  - Cimadomo, J., P. Claeys, and M. Poplawski-Ribeiro. 2016. “How do experts forecast sovereign spreads?” European Economic Review, Vol. 87, pp. 216–235.

- Foundational theories and communication models cited
  - Grice, H. Paul. 1975. “Logic and Conversation.” In Syntax and Semantics, Vol. 3: Speech Acts, edited by Peter Cole and Jerry L. Morgan, 41–58.
  - Shannon, Claude E., and Warren Weaver. 1949. The Mathematical Theory of Communication.
  - Halliday, M. A. K., and Christian M. I. M. Matthiessen. 2014. Halliday’s Introduction to Functional Grammar. 4th ed.
  - Kahneman, Daniel. 2003. “Maps of Bounded Rationality: Psychology for Behavioral Economics.” American Economic Review 93 (5): 1449–1475.
  - Tversky, Amos, and Daniel Kahneman. 1981. “The Framing of Decisions and the Psychology of Choice.” Science 211 (4481): 453–458.
  - Thaler, Richard H. 2016. “Behavioral Economics: Past, Present, and Future.” American Economic Review 106 (7): 1577–1600.

### Key methodological and empirical strands present
- Use of text-analysis and sentiment algorithms to measure clarity, tone, and narratives in fiscal and monetary communications.
- Comparative institutional analysis of fiscal rules, fiscal councils, and their effectiveness across countries and time, including post-pandemic revisions (IMF Working Paper, WP/25/198).
- Assessment of forecast accuracy, forecast biases (over-optimism), and implications for debt management and policy credibility (IMF Working Paper, WP/11/48; SDN/15/10).
- Investigation of how central bank and government communication affect market volatility, inflation expectations, and macroeconomic outcomes.

### Representative policy and research directions implied by the literature cited
- Strengthening independent fiscal institutions and clarifying fiscal rules to improve forecasting accuracy and debt credibility (multiple IMF and academic sources).
- Leveraging text-as-data methods and sentiment analysis to monitor and evaluate the clarity and market impact of official communications.
- Incorporating behavioral and narrative approaches when designing communication strategies to shape public expectations and economic choices.
- Evaluating the role of transparency, semantic consistency, and framing in reducing uncertainty and fostering accountability in fiscal and monetary policy.

*Decoding Fiscal Messaging: How G7 Finance Ministries Communicate — Working Paper No. WP/2026/117*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026117-source-pdf.pdf_
