## draft-bpm7-ch20-v7

## Source details

**Canonical URL:** [draft-bpm7-ch20-v7](https://www.imf.org/-/media/files/data/statistics/bmp7/draft-chapters/draft-bpm7-ch20-v7.pdf)

## Other formats

- [Markdown version](/-/media/files/data/statistics/bmp7/draft-chapters/draft-bpm7-ch20-v7.pdf.md)
- [Structured JSON version](/-/media/files/data/statistics/bmp7/draft-chapters/draft-bpm7-ch20-v7.pdf.json)

---

### Communication and dissemination: purpose and overarching guidance
- Users benefit from comprehensive, consistent, accurate and reliable information communicated and disseminated on a timely basis in an accessible and understandable manner (Draft BPM7 — Chapter 20, IMF).
- Communication objectives:
  - Enhance analytical usefulness, comparability through time and across economies, and maximise policy relevance.
  - Use terminology and presentation that, where sensible, reflect and align with the language of business, governments and the public.
  - Tailor communication strategies to diverse user segments (policymakers, businesses, researchers, general public); different terms can be used for the same content to assist understanding.
  - Combine dissemination (technical accessibility for specialised users) with communication (narratives, key messages, visualisations) to improve understanding and reduce misrepresentation.
- Institutional benefits of communication:
  - Demonstrate relevance, justify public outlay, improve relationships with data providers, and gain visibility for statistical products.

### Dissemination strategy, standards and operational practices
- Dissemination strategy and communication policy (Section B):
  - Production, analysis and dissemination should be transparent, accessible and provided through different channels to meet diverse user needs.
  - Dissemination policies should form part of the vision, mission, principles and values of statistical organisations and be consistent with the UN Fundamental Principles of Official Statistics.
  - Good dissemination supports simultaneous release to all users according to previously announced release calendars.
  - Recommended dissemination channels and formats include websites, printed materials, social media, tables, charts and raw datasets; publishing capability should support digital dissemination, mobile devices and open formats such as common separated value files.
- Data dissemination standards:
  - The SDMX initiative (launched 2001) defines an information model including descriptor concepts; packaging structure; dimensions and attributes; keys; code lists; and data structure definitions.
  - Countries are encouraged to subscribe to IMF SDDS (SDDS or SDDS Plus) or the Enhanced General Data Dissemination System (e-GDDS).
  - The National Summary Data Page (NSDP) serves as a “data portal” for SDDS Plus, SDDS and e-GDDS participants and enables automatic exchange and sharing of statistical data and metadata in SDMX for SDDS Plus and e-GDDS participants.
- Release calendar and timing:
  - Publish a release calendar at the beginning of each year, or at least well in advance of the release date, to promote transparency and provide evidence against political or inappropriate interference.
  - Release calendars should be realistic for compilers and useful for users; failure to meet announced timings frustrates users.
- Revision policies:
  - Revisions reflect new/improved data sources, methodologies or corrections and arise from the timeliness–quality trade-off.
  - Essential features of a good revision policy:
    - A predetermined schedule.
    - Reasonable stability from year to year.
    - Openness and advance notice of reasons for revisions (possibly with indicative size).
    - Easy access for users to analyse long time series of revised data.
    - Adequate documentation on revisions in publications and databases.
  - Special revisions that reassess coverage or compilation methods should be announced in advance with reasons and an assessment of possible impact.

### Statistical value chain, organisational links and media relations
- UN GSBPM mapping: Design; Specify needs; Build; Collect; Process; Analyse; Disseminate; Evaluate — dissemination manages release of static and dynamic products, product updating, promotion and user support.
- Organisational placement:
  - External communication functions should secure high attention and investment for data communication and technological infrastructure.
  - Strong links between compilers and communications teams improve translation of technical nuances; media training for compilers and macroeconomic training for communications personnel recommended.
- Media relations guidance:
  - Media act as main distributors to the broader public; limited, time-limited pre-release access may be granted only for narrowly specified reasons and should be publicly disclosed.
  - Contacts with media should promote open relationships, mutual professional respect, meet media needs and provide information quickly and completely in a truthful and responsible manner.

### Tailoring dissemination to user needs and channels (Section C)
- User categorisation:
  - General data users: general journalists, students, teachers, small businesses, ordinary citizens — simple data needs across a wide range.
  - Analytical users: government departments, local authorities, researchers, economic journalists, central banks, international organisations — complex requirements for detailed variables, time series and breakdowns.
- Forms of dissemination:
  - Scheduled regular releases (online, press releases) for media and general public.
  - Special topic publications or methodological papers with time series, metadata and occasional analysis.
  - Comprehensive annual datasets or yearbooks.
  - Social media posts or short forms to highlight newsworthy features.
- Channel and format considerations:
  - Printed (press releases, newsletters, infographics) versus electronic (pdfs, Excel, XML), video, blogs, podcasts, live streaming, databases, charts, animations, social media.
  - Ensure accessibility across desktop, laptop, tablet and smart phones; use multiple machine-readable formats to support independent visualisation and analysis.
  - Use press coverage analysis, user satisfaction surveys, seminars/webinars and training to monitor changing user demands.
- Communication with data suppliers:
  - Objectives: minimise burden, communicate purpose and confidentiality, build trust, value suppliers’ time.
  - Principles: provide choice in submission modes; minimise duplication; maintain high standards of respondent communication; use supplier-appropriate language; test questionnaires and use bridge tables where needed.
  - Examples: survey calendar for suppliers; sharing survey results; personalised statistical feedback showing suppliers’ contributions.
- Confidentiality and researcher access:
  - Statistical confidentiality protections, anonymisation and disclosure checking are required before publication (Fundamental Principles of Official Statistics: Principle 6).
  - Confidential data may be provided to a limited number of users under strict conditions for validation; such cases should be publicised.
  - Alternatives for researcher access: secure data labs; signed data access agreements permitting limited-time access to secure online data areas.

### Vintages, terms, and communicating revisions (Box 21.2)
- Definitions and key terms:
  - Data vintage: a data value or dataset for a reference period released at a particular release period; a new vintage is established when the same data are released at a different release period.
  - Update period: the time-period over which revisions to a data value have been applied.
  - Update: the numerical difference between two vintages of the same data point.
- Types of vintages and revisions:
  - Routine (regular) update: scheduled incorporation of more complete source data, improved models or minor methodological adjustments; may include seasonal adjustment or correction of compilation errors.
  - Benchmark estimate: final vintage compiled using the most comprehensive and highest quality source data and methods; not expected to be further revised except for changes in definitions or standards.
  - Benchmark update: revisions from incorporation of benchmark estimate(s).
  - Comprehensive update: benchmark update that also integrates new concepts, accounting treatments, classifications, substantially improved methods or base-year updates; generally occur with major changes to statistical standards and may cause discontinuities requiring back-casting.
- Recommended terminology and practice:
  - Use: routine updates, benchmark updates, comprehensive updates to communicate revision extent/substance.
  - Phase out other terms as part of the 2025 SNA implementation.
- Recommended minimum vintage metadata to communicate:
  - (1) substance of the release; (2) timeliness; (3) frequency; (4) the reference period; (5) the update period.
- Illustrative vintage chronology example for 2023 Q1:
  - May 30th, 2023 — initial release of 2023 Q1 (example release date).
  - June 30th, 2023 — 2023 Q1 may be revised due to more data becoming available [and seasonal adjustment].
  - September 30th, 2025 — 2023 Q1 may be revised due to benchmarking to annual estimates [through supply and use tables as well as revised seasonal adjustment analyses].
  - May 30th, 2026 — 2023 Q1 may be revised due to the results of an [economic census] /[benchmark survey].
  - September 30th, 2030 — 2023 Q1 may be revised due to the implementation of the new international standards for macroeconomic statistics [SNA]/[BPM].
- Example: communicating successive vintages (Quarterly GDP) — illustrative assumptions:
  - First, second and third vintages published 30, 60 and 90 days after the reference period, respectively; each vintage is a routine revision.
  - Recommended labels:
    - Quarterly National Accounts release, first quarter Year t - 30-day
    - Quarterly National Accounts release, first quarter Year t - Routine update – 60-day - (Year t-2 Q1 revised)
    - Quarterly National Accounts release, first quarter Year t - Routine update – 90-day - (Year t-2 Q1 revised)
  - Example real-time recording (illustrative numbers):
    - Release Date: June 30th, Year t — Type of update: Routine — Vintage values (Q1 2021, Q4 2020, Q3 2020, Q2 2020, Q1 2020): 99 95 90 85 80
    - Release Date: May 30th, Year t — Type of update: Routine — Vintage values: 102 95 90 85 80
    - Release Date: April 30th, Year t — Type of update: Routine — Vintage values: 100 95 90 85 80
    - Note: the above numbers are for illustrative purposes only.

### Taxonomies, metadata and product quality
- Metadata expectations:
  - Metadata are “data about data” including administrative facts, concept definitions and descriptions of collection and processing; metadata should be accessible, comprehensive and linked to the data they describe.
  - Implement metadata management systems that retain links between metadata and data throughout production.
- Statistical product quality taxonomy (Box 21.3):
  - Level 1:
    - Official Statistics: estimates that incorporate recommended nationally or internationally adopted concepts, methods, accounting rules and classifications and meet all required standards.
    - Official Provisional Estimates: early estimates that incorporate adopted concepts, methods and classifications but precede more comprehensive data.
  - Level 2:
    - Experimental Estimates: estimates that vary in limited ways from recommended standards, often research or indicative in nature, released to seek feedback.
  - Lifecycle notes:
    - Provisional estimates are expected to “graduate” to revised, more mature official statistics as methods and source data meet standards.
    - Experimental estimates may be ad hoc in update frequency and are released with a “proof of concept” notion; improvements can enable transition to Level 1.
- Taxonomy of accounts and tables (Box 21.4):
  - Accounts/tables placement relative to the SNA:
    - Sequence of Economic Accounts: core accounts/tables underpinning the main sequence of economic accounts within the SNA boundary.
    - Supplementary Accounts/Tables: operate within the SNA boundary and supplement the main sequence.
    - Thematic Accounts/Tables: operate within the SNA boundary and provide thematic detail.
    - Extended Accounts/Tables: operate beyond the SNA boundary and provide information beyond the sequence.
  - Presentation recommendations:
    - Publish breakdowns using agreed international classifications and their hierarchies (examples: ISIC Revision 5 and COICOP 2018).
    - Provide additional tables that support aggregates (e.g., GVA by institutional sector; household final consumption by COICOP divisions).
    - Align national classifications preferably one-to-one or many-to-one to international classifications and use correspondence tables linking different classifications.

### Alignment frameworks, comparability and presentation rubric (sections 21.91–21.114)
- Purpose of alignment frameworks:
  - Provide structured, systematic and consistent methods to assess an economy’s alignment to standards (2025 SNA / BPM7) and to signal basis for cross-country comparability.
  - Alignment frameworks are voluntary, stand-alone tools based on self-assessment and encouraged to be documented in metadata.
- Alignment framework structure (for the 2025 SNA / BPM7):
  - Four high-level components: Concepts and definitions; Methods; Classifications; Accounts / tables.
  - Practical focus on a subset of items that impact interpretation of levels and growth rates.
- Presentation rubric — degrees of alignment (recommended degree bands and interpretation):
  - Fully aligned — meaning that between 95-100 per cent of the guidelines and standards are implemented.
  - Highly aligned — meaning that between 75-95 per cent of the guidelines and standards are implemented.
  - Broadly aligned — meaning that between 50-75 per cent of the guidelines and standards are implemented.
  - Partially aligned — meaning that between 25-50 per cent of the guidelines and standards are implemented.
  - Not aligned — meaning between 0-25 per cent of the guidelines and standards are implemented.
  - Not applicable — meaning that for issues of materiality or relevance, the standard is not implemented. Materiality in this context is subjective but a steer would be less than 0.05 per cent (and/or a monetary equivalent) of GDP.
- Measurement notes:
  - The "per cent aligned" notion introduces subjectivity; a range approach is adopted because defining 100 per cent alignment is difficult.
  - Categories are appropriate for concepts, definitions, methods and accounting rules but not for classifications, tables or accounts; for tables and accounts use timeliness (days after reference period) and granularity (number of detailed classes) to quantify alignment.
- Presentation and dissemination recommendations:
  - Use a dashboard approach rather than a weighted scoreboard and present digitally as part of sources and methods documentation.
  - Recommended colour coding: Fully aligned = Green; Highly aligned = Light green; Broadly aligned = Yellow; Partially aligned = Light yellow; Not aligned = Red; Not applicable = Black.

### Net measures, clearer terminology and the Common Glossary (sections 21.91–21.119)
- Prominence of net measures:
  - Net measures (NDP, NNI, household disposable income, consumption, saving, net worth) are increasingly important for well-being and sustainability analysis and may be conceptually preferred for reflecting production costs, but measurement challenges impede comparability.
  - Presentation guidance: present net measures alongside gross measures, supplementing rather than replacing gross measures.
- Example gross-to-net articulation (conceptual change in 2025 SNA):
  - Gross Domestic Product (B1g) = 1 854
  - Depreciation (−) (P51c) = 222
  - Depletion (−) (K21) = 14
  - Net Domestic Product (B1n_P51c+K21) = 1 618
  - Showing both depreciation and depletion components allows user flexibility.
- Clarification of "net" usage:
  - Use "net" and "gross" only in two areas:
    - In balancing items for each account by institutional sector and the whole economy, where the difference is due to depreciation and depletion.
    - In financial accounts: (i) aggregating acquisitions less disposals for each financial instrument on assets and liabilities; (ii) aggregating changes in assets and changes in liabilities and showing both on a net basis.
  - Replace legacy uses of "net" (e.g., "net exports") with explicit expressions such as "exports less imports".
- Harmonisation of terminology and Common Glossary (2025 SNA and BPM7):
  - Common Glossary aims to harmonise labels and concepts across macroeconomic statistical standards and provide alternative easier-to-understand terms alongside technical definitions.
  - Examples of renamed accounts and terms (retained verbatim where given):
    - Generation of income account → The generation of earned income account
    - Allocation of primary income account → The allocation of earned income account
    - Secondary distribution of income account → Transfer income account
    - Compensation of employees → Remuneration of employees
    - Constant prices → In volume terms
    - Consumption of fixed capital → Depreciation
    - Imputed rental of owner-occupied dwellings → Owner-occupied housing services
    - Net errors and omissions → Statistical discrepancy
    - Primary income → Earned income
    - Uses → Expenditures
  - Intended benefits: facilitate communication across producers, improve comprehension for non-specialists and provide alternative easier terms.
  - Statistical producers are urged to target communication because some technical descriptions remain not user-friendly.

### Key recommendations (explicit)
- Use routine updates, benchmark updates and comprehensive updates to communicate extent/substance of revisions.
- Adopt and apply a consistent taxonomy over time and across countries to communicate data quality; adopt the two-tier product quality taxonomy (Level 1/Level 2).
- Communicate vintage metadata including series, reference period, timeliness, frequency, type of revision and update period using a standard format.
- Decompose revisions by source (conceptual, methodological, presentational) and distinguish level and growth-rate effects to aid interpretation.
- Where comprehensive changes create discontinuities, consider back-casting to adjust historical data.

*Source: Draft BPM7 — Chapter 20 (draft-bpm7-ch20-v7), International Monetary Fund.*

### 21.1 The way in which macroeconomic statistics are communicated and disseminated has a significant impact

### 21.1 The way in which macroeconomic statistics are communicated and disseminated has a significant impact on users’ understanding and utilisation of the data and should be viewed as a key component of the production chain of official statistics

### Purpose and overarching guidance
- Users benefit from comprehensive, consistent, accurate and reliable information communicated and disseminated on a timely basis in an accessible and understandable manner.
- Communication and dissemination should enhance analytical usefulness, comparability through time and across economies, and maximise policy relevance.
- Terminology and presentation of macroeconomic aggregates and concepts should, where sensible, reflect and align with the language of business, governments and the public.
- Consistent standards increase comparability and help users understand the basis of the data (for example, which version of the SNA and BPM is used).
- Tailor communication strategies to diverse user segments (e.g., policymakers, businesses, researchers, and the general public); different terms can be used for the same content to assist understanding.
- Dissemination covers technical accessibility for specialised users, while communication includes narratives, key messages, and visualisations to improve understanding and reduce misrepresentation.
- Effective communication conveys factual messages based on data suppliers’ explanations, comments and feedback to explain what, when, where, why and how events occurred.
- Statistical organisations can use communication to demonstrate relevance, justify public outlay, improve relationships with data providers, and gain visibility for their products.
- The chapter provides principles, guidelines and new recommendations to improve communication to enhance comparability, understanding and user experience.

### Dissemination strategy and communication policy (Section B)
- Production, analysis and dissemination of official statistics should be undertaken in a transparent and accessible way and provided through different channels to meet different user needs and uses.
- Dissemination policies and strategies should form part of the vision, mission, principles and values of statistical organisations and be consistent with the UN Fundamental Principles of Official Statistics.
- Principal aim of statistical communication: reveal more of the information contained in statistical data and make statistical information easier to interpret.
- Good dissemination policies support access to consistent and coherent statistical data to all users and ensure transparency and impartiality, including simultaneous release to all users according to previously announced release calendars.
- Dissemination channels and formats include websites, printed materials, social media, tables, charts and raw datasets; publishing capability should support digital dissemination and mobile devices and make data available in open formats such as common separated value files.

### Statistical value chain and GSBPM
- The UN Generic Statistical Business Process Model (GSBPM) describes the set of business processes forming the statistical value chain; Level 1 stages are referenced (Design, Specify needs, Build, Collect, Process, Analyse, Disseminate, Evaluate).
- The dissemination phase manages release of statistical products and includes assembling and releasing static and dynamic products via multiple channels and managing user support.
- For regularly produced outputs, dissemination occurs in each iteration and comprises processes such as updating output systems, producing/managing/promoting dissemination products, and managing user support (including feedback and satisfaction surveys).
- The collection phase relies on effective communication with suppliers and includes design and creation of frames, sample selection, and setting up collection processes.
- Supplier relationship details are expanded in section D.

### Organisational structure and media relations
- The placement of external communication functions within statistical systems affects effectiveness and should secure high attention and investment for data communication and technological infrastructure.
- Increased supply of data from non-official producers raises the need to improve communication quality, content, timeliness and channels, particularly for media, policymakers, and hard-to-reach groups.
- Strong link between economic statistics compilers and communications teams ensures technical nuances are effectively translated for users; media training for compilers and macroeconomic training for communications personnel can enhance this link.
- Media typically act as main distributors to the broader public; limited, time-limited pre-release access may be granted for specific reasons but such access should be narrowly time-limited and publicly disclosed.
- Communications teams commonly manage press conferences, expert interviews, journalist requests, media crises, social media, website management, digital design and data visualisation.
- Contact with media should:
  - Promote an open relationship.
  - Foster mutual professional respect.
  - Meet the needs of media.
  - Treat journalists as legitimate representatives of the public.
  - Provide information as quickly and completely as possible, in a truthful and responsible manner in line with statistical producer interests.

### Principles and standards
- Official statistics should be reliable, objective and relevant for decision making.
- Dissemination strategies should align with the UN Fundamental Principles of Official Statistics; Principle 1 states that "...official statistics are to be compiled and made available on an impartial basis by official statistical agencies to honour the entitlement of citizens to public information."
- Statistical organisations should provide users with maximum access to official statistics in accordance with confidentiality guidelines.
- Existing codes of practice (e.g., European Statistics Code of Practice, revised 2017) and national statistics codes of practice provide guidance on dissemination practices.

### UNECE guidance
- UNECE has produced guidance within the framework of UNECE Work Sessions on the Communication and Dissemination of Statistics to aid statistical organisations in communication and dissemination.

### Data dissemination standards
- In 2001, seven institutions launched the SDMX initiative to develop common dissemination standards for exchange of statistical information; the SDMX information model covers descriptor concepts; packaging structure; dimensions and attributes; keys; code lists; and data structure definitions.
- The data structure definition conveys the data classification scheme specifying the set of concepts required to describe and identify statistical data items.
- Countries are encouraged to subscribe to IMF SDDS (SDDS or SDDS Plus) or the Enhanced General Data Dissemination System (e-GDDS) to improve data quality.
- The National Summary Data Page (NSDP) is a “data portal” for economies participating in SDDS Plus, SDDS and e-GDDS allowing access to data, metadata or links to online datasets; for SDDS Plus and e-GDDS participants, the NSDP enables automatic exchange and sharing of statistical data and metadata in SDMX.
- Under the G-20 data gaps initiatives, templates have been developed for G-20 members to report data for various recommendations (for example, templates on shadow banking and institutional sector accounts).

### Release calendar
- Availability of a release calendar in advance is important to inform user expectations and enable planning of activities and topical analyses.
- The compilation and release schedule should be realistic for compilers and useful for users; failure to meet announced timings frustrates users.
- The release calendar should be published at the beginning of each year, or at least well in advance of the release date, on statistical producers’ websites to promote transparency and provide evidence against political or inappropriate interference.

### Data revision and revision policies
- Revisions are essential in macroeconomic statistics and typically reflect new or improved data sources and methodologies or corrections of past errors.
- Revisions result from the trade-off between timeliness and quality (accuracy and comprehensiveness); provisional data are often revised when more accurate source data become available.
- Regular communication with users in advance of expected updates and improvements will develop better public understanding of why revisions occur and reduce negative reactions.
- Countries are encouraged to develop a well-designed revision policy that is managed and coordinated across related statistical domains and communicated to users in advance; such a policy should enable users to understand revisions systematically.
- Essential features of a good revision policy include:
  - A predetermined schedule.
  - Reasonable stability from year to year.
  - Openness.
  - Advance notice of reasons for revisions (possibly with indicative size).
  - Easy access for users to analyse long time series of revised data.
  - Adequate documentation on revisions in publications and databases.
- Analysis of revisions is useful and may be published to help users understand revisions.
- Special revisions that reassess data coverage or compilation methods (leading to significant historical changes) should be announced in advance with reasons and an assessment of possible impact; see related guidance on national accounting backcasting methodology.
- Published revisions should be evaluated to identify persistent revisions and potential systematic bias, leading to improvements in sources and methods and reduced revision bias.
- Composition of revisions and explanations for users are covered in section F.

### Communication with users (Section C)
- Macroeconomic statistics should be designed to meet needs of a wide variety of users making different uses of the statistics.
- The user community includes government, business, academia, analysts, economists, researchers, journalists, international statistical agencies, the media and the public.
- Users can be grouped into two main categories:
  - General data users (e.g., general journalists, students, teachers, small businesses, ordinary citizens) with simple data requirements across a wide range of information.
  - Analytical users (e.g., government departments, local authorities, researchers, economic journalists, central banks, international organisations) with complex requirements for detailed variables, time series and regional or institutional sector breakdowns.

*Source: draft-bpm7-ch20-v7*

### 21.36 An understanding of the possible user needs is vital in identifying effective ways to disseminate the

### 21.36 An understanding of the possible user needs is vital in identifying effective ways to disseminate the statistical information

### Dissemination tailored to user needs
- Understanding who users are guides content and language so statistics are accessible.
- Different dissemination forms to meet diverse demands:
  - Scheduled regular statistical releases, typically made available on-line and sometimes also featured as press releases, suitable for the media and the general public focused on main findings.
  - Special topic-related publications or methodological-type papers, including time series, detailed data, metadata, and occasional short economic analysis.
  - Highly comprehensive detailed macroeconomic statistics presented as scheduled annual datasets (or yearbooks) and made available online.
  - Social media posts or similar short forms to supplement formal releases and highlight newsworthy features to broader audiences.
- Good standards of data visualisation in tables and charts aid effective dissemination.
- Statistical producers should publish extensive definitions and make datasets available in technically compatible ways (for example, common separated value files) to encourage independent innovation in online visualisations.
- Producers can use online crowdsourcing and artificial intelligence providers to collect data otherwise unavailable through usual channels.

### What to communicate (content considerations)
- Key content distinctions to communicate clearly:
  - Data – estimates versus projections; convey increasing data content as release vintages evolve to reduce data uncertainty.
  - Current data versus historical (archived) data.
  - Level of aggregation.
  - Micro data versus macro data.
  - Metadata.
  - Story or knowledge-adding explanations to aid understanding of data movements.

### How to communicate (channel and format considerations)
- Channels and formats that should be considered:
  - Printed form (press releases, newsletters, infographics) versus electronic (pdfs, Excel files, infographics, XML) or both.
  - Different machine-readable formats to suit users’ needs.
  - Internet release addressing website design, search facilities, etc.
  - Video releases, blogs, podcasts, presentations, live streaming.
  - Databases, tables, charts, animations.
  - Social media.
- Ensure releases are widely accessible across devices: desktop, laptop, tablet, smart phones; avoid limiting to a single online format.
- Consider freely available detail versus charged bespoke analyses requested by users.
- Use regular analysis of press coverage and feedback, user satisfaction surveys, seminars/webinars/workshops/conferences, and training/education to monitor and respond to changing user demands.

### Communication with data suppliers
- An effective data supplier engagement strategy is essential; suppliers include business surveys, administrative data, household surveys, etc.
- Key objectives: minimise burden, communicate purpose and confidentiality, build trust, and value suppliers’ time.
- Principles to reflect:
  - Provide choice to data suppliers (telephone data entry, secure electronic file transfer, paper submission alternatives).
  - Minimise impact or burden (fair selection frequency in survey samples, ask for information once, minimise duplication).
  - Maintain high standards of communication with respondents (standardised responses, phone assistance, timely communication).
- Use supplier-appropriate language (company accounting or administrative terms) rather than macroeconomic statistical standard terminology; test questionnaires and use bridge tables where needed to map supplier data to statistical concepts.
- Accountants often respond; ensure links between business accounting standards and macroeconomic statistical standards and monitor changes to business accounting rules.
- Examples of communications to improve supplier experience:
  - A survey calendar indicating when suppliers can expect questionnaires.
  - Sharing of survey results to which suppliers have contributed.
  - Personalised statistical feedback, including tables showing suppliers’ contributions to totals.
- Data collectors should provide feedback on data quality (accuracy and reliability) to allow suppliers to address issues.

### Statistical confidentiality
- Most official statistics are collected under legislation for statistical purposes only; dissemination must not permit identification of particular businesses or households.
- Ensure data confidentiality policies, anonymisation techniques, and disclosure checking procedures are in place before publication.
- Principle 6 of the Fundamental Principles of Official Statistics: individual data collected for statistical compilation are to be strictly confidential and used exclusively for statistical purposes.
- Permissions must be sought from a business to publish information that would otherwise be confidential.
- Confidential information may, in specific cases, be provided to a limited number of users under strict conditions for validation and quality assurance; such cases should be adequately publicised.
- Aim to maximise dissemination as a public good while meeting confidentiality obligations; increased granularity may lead to more disclosure cases and suppression or aggregation of cells.
- Global challenges: multi-national enterprise (MNE) group activities require data exchange and reconciliation across jurisdictions, with appropriate legal agreements and secure channels to enable data exchange.
- Alternative approaches for researcher access to micro data:
  - Secure data labs allowing analysis without revealing confidential data.
  - Signed data access agreements permitting access to secure online data areas for limited-time research or analyses.

### Taxonomies and metadata
- Producers must provide product and contextual information to enable proper use and interpretation of macroeconomic statistics.
- Metadata defined as “data about data” that enable sharing, querying, understanding, and use across collection, compilation, and dissemination stages; apply at micro to macro aggregation levels.
- Metadata include administrative facts (who created data and when), definitions of concepts applied, and descriptions of collection and processing steps.
- Accessible and comprehensive metadata promote data literacy and facilitate linking and integration across statistical domains.
- Metadata dissemination should be integral to dissemination strategy; implement metadata management systems that retain links between metadata and data throughout production.
- Good practice: actively link metadata to the statistical data they describe and vice versa.

### Releases and vintages of data
- National producers use practices and terminology (provisional, first, preliminary, second, final; initial estimates, mature estimates, final estimates, data revisions, benchmark revisions, rebased estimates, improvements to methods, corrections, experimental estimates) to communicate vintages and methodological changes.
- Producers should balance users’ needs for timely, high-frequency data with needs for high accuracy, and reconcile long consistent time-series expectations with agile statistical standards.
- Vintages are regular, anticipated updated releases for the same time period due to new data availability or methodological improvements and should be consistently described with pre-announced release dates in a release calendar.
- Example sequence for a 2023 Q1 release (illustrative chronology of potential revisions):
  - May 30th, 2023 — initial release of 2023 Q1 (example release date).
  - June 30th, 2023 — 2023 Q1 may be revised due to more data becoming available [and seasonal adjustment].
  - September 30th, 2025 — 2023 Q1 may be revised due to benchmarking to annual estimates [through supply and use tables as well as revised seasonal adjustment analyses].
  - May 30th, 2026 — 2023 Q1 may be revised due to the results of an [economic census] /[benchmark survey].
  - September 30th, 2030 — 2023 Q1 may be revised due to the implementation of the new international standards for macroeconomic statistics [SNA]/[BPM].
- Adopt a common approach when communicating different releases or vintages: at minimum include information about (1) substance of the release; (2) timeliness; (3) frequency; (4) the reference period; and (5) the update period.
- Defining, describing, and communicating vintages is complex and requires standardised terminology and definitions to improve interpretation and cross-country comparability.

*Source: draft BPM7 chapter 20 (IMF).*

### Box 21.2 Terms and definitions related to different vintages of economic data releases

### Box 21.2 Terms and definitions related to different vintages of economic data releases

### Definitions of basic terms
- A regular / routine update can be referred to as either regular or routine with the same meaning; the box uses the term routine.
- Data item: a discrete unit that can be represented numerically; sources include data collected, measurement based on data collected, and model-based estimates.
- Time-series: a series of regular time-ordered data values of a quantitative characteristic taken at successive, in most cases equidistant, periods/points of time.
- Data vintage: a data value or a dataset (sequence of values) for a given reference period that has been released for use at a particular point in time (release period). A new vintage is established when the same set of data for the same reference period or some overlapping portion is released at a different release period.
- Reference period: the time-period represented by the data.
- Update period: the time-period over which revisions to a data value have been applied.
- Release period: the calendar date when the data are released to the public.
- Update: a revision defined as the numerical difference between two vintages of the same data point.

### Types of vintages and revisions
- Routine (regular) update:
  - Incorporation of scheduled, more complete (not necessarily final) source data, improved models, or other iterations of the compilation process.
  - Occur for both sub-annual and annual estimates and can occur throughout the year, at regular intervals, or as new information becomes available.
  - May include seasonal adjustment, correction of compilation errors, or minor methodological adjustments made outside the benchmark or comprehensive revision process.
- Benchmark estimate:
  - The final vintage of a dataset compiled using the most comprehensive and highest quality source data and the most advanced methods at that point in time.
  - Not expected to be further revised and often referred to as the “final” estimate, except if definitions or macroeconomic statistical standards change or a new data source is used.
- Benchmark update:
  - Revisions from the incorporation of a benchmark estimate(s) into a given set of macroeconomic statistics or accounts.
- Comprehensive update:
  - A special case of benchmark update where revisions incorporate final source data and also integrate new or updated concepts, new accounting treatments, classifications, substantially improved methods, or updating a base year.
  - Generally occur with major changes to macroeconomic statistical standards and may cause discontinuities requiring back-casting considerations.

### Recommended terminology and communication practice
- The recommended terms to communicate the “extent” or “substance” of revisions are: routine updates, benchmark updates, comprehensive updates.
- The first two terms mainly reflect the vintage of source data entering the compilation process; the term comprehensive revision reflects additions or substantial changes to concepts, methods, classifications, or presentation.
- All other terms should be phased out as part of the 2025 SNA implementation.

### Example: communicating successive vintages (Quarterly GDP)
- Assumptions used in illustrative example:
  - First, second and third vintages of quarterly GDP are published 30, 60 and 90 days after the reference period, respectively.
  - Estimates are based on incomplete source data such that each vintage is a result of a routine revision.
- Recommended communication labels:
  - Quarterly National Accounts release, first quarter Year t - 30-day
  - Quarterly National Accounts release, first quarter Year t - Routine update – 60-day - (Year t-2 Q1 revised)
  - Quarterly National Accounts release, first quarter Year t - Routine update – 90-day - (Year t-2 Q1 revised)
- Example real-time recording (illustrative numbers):
  - Release Date: June 30th, Year t — Type of update: Routine — Vintage values (Q1 2021, Q4 2020, Q3 2020, Q2 2020, Q1 2020): 99 95 90 85 80
  - Release Date: May 30th, Year t — Type of update: Routine — Vintage values: 102 95 90 85 80
  - Release Date: April 30th, Year t — Type of update: Routine — Vintage values: 100 95 90 85 80
  - Note: the above numbers are for illustrative purposes only.

### Sources and decomposition of revisions
- Macroeconomic statistical standards define concepts, outline quantification methods and accounting rules, and identify classification systems, accounts, and table structures; any of these features can be sources of revisions.
- Recommended categories for decomposing sources of updates:
  - Conceptual changes: alignment to an updated set of macroeconomic statistical standards.
  - Methodological changes, including:
    - coverage adjustments (for example, exhaustiveness);
    - changes to source data (for example, new results based on improved survey response rates, replacing modelling algorithm with a survey-based estimate);
    - quality improvements (for example, data validation, consistency of source data results, seasonal adjustment);
    - changes in accounting rules (for example, changes from cash to accrual accounting).
  - Presentational changes: new tables, charts, revision triangles, granular detail where aggregates may not change but the presentation of components does.
- Decomposition should reflect source and size of revisions; may be broken down further when a revision combines multiple issues or affects multiple accounts. Distinguish levels and growth-rate effects of revisions and consider showing all components of the revision(s) for a single period or across all periods revised.

### Statistical product quality taxonomy (Box 21.3)
- Two-level taxonomy recommended:
  - Level 1:
    - Official Statistics: Estimates that incorporate recommended nationally or internationally adopted concepts, methods, accounting rules and classifications and meet all the standards required.
    - Official Provisional Estimates: Provisional estimates incorporate nationally or internationally adopted concepts, methods, accounting rules and classifications but represent an early estimate before more comprehensive data becomes available.
  - Level 2:
    - Experimental Estimates: Estimates that vary in limited ways from recommended concepts, methods, accounting rules or classifications but where the producer has good confidence in their validity; often research or indicative in nature, based on modelling assumptions or untested source data, released to seek feedback.
- Characteristics and lifecycle:
  - Provisional estimates are expected to “graduate” to revised, more mature official statistics status as methods and source data meet required standards.
  - Experimental estimates may be ad hoc in update frequency and are released with a “proof of concept” notion; improvements can enable transition to Level 1.

### Taxonomy of accounts and tables (Box 21.4) and presentation guidance
- Accounts/tables placement relative to the SNA:
  - Sequence of Economic Accounts: core accounts/tables underpinning the main sequence of economic accounts operating within the SNA boundary.
  - Supplementary Accounts/Tables (e.g., extended SUTs, IOTs, pensions, regional, environmental): operate within the SNA boundary and supplement the main sequence.
  - Thematic Accounts/Tables (e.g., health, tourism, sport, creative sector): operate within the SNA boundary and provide additional thematic detail.
  - Extended Accounts/Tables (e.g., unpaid household work): operate beyond the SNA boundary and provide information beyond the sequence of economic accounts.
- Accounts vs tables:
  - Accounts have opening and closing items linking resource and use flows and stocks and are interrelated across the sequence.
  - A table is distinguished from an account in that tables do not have a balancing item and are not part of the sequence of economic accounts.
- Presentation recommendations:
  - Publish breakdowns using agreed international classifications and their hierarchies (examples given for ISIC Revision 5 and COICOP 2018).
  - Provide additional tables that support aggregates, e.g., GVA by institutional sector; household final consumption by COICOP divisions; gross fixed capital formation by industry or asset type.
  - Align national classifications preferably one-to-one or many-to-one to international classifications to aid comparability and analysis.
  - Use correspondence tables linking different classifications to enhance user linking and analytical capability.

### Key recommendations (explicit)
- Use routine updates, benchmark updates, and comprehensive updates to communicate revision extent/substance.
- Adopt and apply a consistent taxonomy over time and across countries to communicate data quality; adopt the two-tier product quality taxonomy (Level 1/Level 2).
- Communicate vintage metadata including series, reference period, timeliness, frequency, type of revision and update period using a standard format (illustrated by the quarterly GDP example).
- Decompose revisions by source (conceptual, methodological, presentational) and distinguish level and growth-rate effects to aid user interpretation.
- Where comprehensive changes create discontinuities, consider methods such as back-casting to adjust historical data.

*Draft BPM7 — Box 21.2, International Monetary Fund*

### 21.91 An important feature of the macroeconomic statistical standards is their ability to develop internationally

### draft-bpm7-ch20-v7 - 21.91 An important feature of the macroeconomic statistical standards is their ability to develop internationally consistent macroeconomic statistics which in turn facilitate the comparison of estimates across countries

### Purpose and nature of alignment frameworks
- Macroeconomic statistical standards aim to produce internationally consistent macroeconomic statistics to facilitate cross-country comparison.
- Countries retain discretion in implementing recommendations to accommodate specific circumstances (e.g., limited resources, data availability, systems constraints, user needs, policy demands).
- Non-implementation may occur when an activity or concept is economically immaterial for a given economy, resulting in varying degrees of "alignment" across countries.
- Users need assurance that estimates are compiled on the same basis to be confident in cross-country comparisons.
- A set of internationally accepted alignment frameworks (e.g., for the SNA, BPM and GFS) provides structured, systematic and consistent methods to assess an economy’s alignment to these standards.
- The alignment frameworks draw on existing assessment frameworks and tools such as the IMF's Data Quality Assessment Framework (DQAF) and Reports on the Observance of Standards and Codes (ROSCs), the UN's Data Quality Assessment Framework (UN-NAQ), and the ISWGNA's Minimum Required Data Set (MRDS).
- Alignment frameworks are stand-alone tools intended for national statistical authorities and international agencies to assess macroeconomic statistical methodologies and processes at country or country-group levels.
- Alignment frameworks are voluntary and based on self-assessment; statistical producers are encouraged to document deviations or adaptations in metadata and methodological notes and to make results publicly available.

### Alignment framework for the [2025 SNA]/[BPM7]
- Structured around four high-level components:
  - Concepts and definitions — what gets measured.
  - Methods — how a compiler implements an accounting rule or measures a concept.
  - Classifications — level of detail and conformity with the 2025 SNA / BPM7 classification schemes (e.g., by industry, product, region, function).
  - Accounts / tables — how information is presented; the SNA and BPM accounts/tables form the basis of the framework.
- Practical implementation:
  - Only a subset of concepts, methods, classifications and accounts/tables is expected to be included, focusing on items that impact interpretation and assessment of levels and growth rates.
  - Examples of metadata questions: "Last benchmark year for GDP?", "Latest period for which balanced SUTs are available?", "Latest period for which institutional sector accounts are available?", "Do you have a published revision policy?"
- Examples of alignment elements (indicative, illustrative questions and categories):
  - Concepts and definitions: Units of the economy; Production boundary covers Informal economy, Underground economy, Illegal activities; IPPs.
  - Methods: Accounting rules such as Valuation, Time of Recording; seasonal adjustment for sub-annual series; deflation of Output and intermediate consumption by appropriate price indexes at basic prices or at producer prices; volume indexes chained-weighted.
  - Classifications: ISIC, CPC, COICOP; institutional sector, asset and transaction classifications.
  - Accounts / tables: Timeliness (e.g., T+30 days, T+3 months); Granularity (Number of industries or number of transaction lines); Annual value added by industry and GDP in current prices and in volume terms; Annual GDP by expenditure in current prices and in volume terms; Annual GDP by income in current prices; Annual sequence of accounts for the total economy (until net lending / borrowing); Annual rest of the world accounts (until net lending / borrowing).

- BPM-specific examples:
  - Metadata questions: "Do you have a published revision policy?", "Is the external accounts revision policy consistent with the national accounts revision policy?"
  - Concepts and definitions: Units of the economy; BOP coverage; IIP coverage; Structure; Functional categories; Sectorization.
  - Methods: Valuation; Time of Recording; Grossing / Netting; Investment income obtained directly (rather than estimated); Direct investment relationships identified by applying the Foreign Direct Investment Relationship (FDIR).
  - Classifications: Classification of institutional sectors; Earned income account, financial account, and IIP classified according to functional categories; Classification of financial assets and liabilities by instrument; Classification of services.
  - Accounts / tables: BOP standard components and memorandum items; IIP standard components and memorandum items; Other flows standard components and memorandum items; Reserve-related liabilities; Non-performing loans separately at fair value; Currency composition of assets and liabilities and institutional sector.

- Presentation rubric — degrees of alignment (recommended degree bands and interpretation):
  - Fully aligned — meaning that between 95-100 per cent of the guidelines and standards are implemented.
  - Highly aligned — meaning that between 75-95 per cent of the guidelines and standards are implemented.
  - Broadly aligned — meaning that between 50-75 per cent of the guidelines and standards are implemented.
  - Partially aligned — meaning that between 25-50 per cent of the guidelines and standards are implemented.
  - Not aligned — meaning between 0-25 per cent of the guidelines and standards are implemented.
  - Not applicable — meaning that for issues of materiality or relevance, the standard is not implemented. Materiality in this context is subjective but a steer would be less than 0.05 per cent (and/or a monetary equivalent) of GDP.
- Notes on measurement and categories:
  - The "per cent aligned" notion introduces subjectivity; a range approach is adopted because defining 100 per cent alignment is difficult.
  - The fully/highly/partially/not aligned categories are appropriate for concepts, definitions, methods and accounting rules but not for classifications, tables or accounts.
  - For tables and accounts, use timeliness (days released after the reference period) and granularity (number of detailed classes) to "quantify" alignment.

- Recommended presentation and dissemination:
  - Use a dashboard approach rather than a weighted scoreboard.
  - Present dashboard digitally and include as part of sources and methods documentation.
  - The assessment can cover entire macroeconomic statistics or be completed for individual accounts or tables.
  - Recommended colour coding:
    - Fully aligned = Green
    - Highly aligned = Light green
    - Broadly aligned = Yellow
    - Partially aligned = Light yellow
    - Not aligned = Red
    - Not applicable = Black

### Benefits to users and producers
- Cross-country comparability:
  - Different countries may conform to different editions (e.g., 1993 SNA, 2008 SNA, 2025 SNA; BPM5, BPM6, BPM7), producing data that may not be directly comparable due to variations in concepts, methodologies and coverage.
  - The proposed alignment framework provides a structure for users to assess national statistical practices in a comparable way.
- Signals to users:
  - Degree of alignment signals the quality of cross-country comparisons and the extent to which major revisions should be expected where an economy is not aligned with the latest standards.
  - Example scenario: Country A and Country B both use the 2025 SNA and BPM7 but differ in recording smuggling activity. Users comparing gross value added, balance sheets, productivity and cross-border flows need to understand these differences; Country A should be encouraged to disclose whether smuggling is not recorded because it is immaterial or due to lack of data.
- Benefits for producers:
  - Alignment information helps producers identify areas for improvement, prioritise resources, and formulate strategic plans to improve alignment to macroeconomic statistical standards.
  - Helps users make appropriate adjustments to achieve comparability and increases users’ capacity to provide feedback on priorities and demands.
- Limitations:
  - Alignment frameworks do not provide a comprehensive evaluation of the quality of an economy’s macroeconomic statistics; a comprehensive quality evaluation would require a different, more rigorous investigation.
  - The design and application of alignment frameworks are expected to evolve over time.

### Prominence of net measures and clarification of "net" usage
- Growing importance:
  - The role and prominence of indicators other than GDP and of net measures (e.g., net domestic product (NDP), net national income (NNI), household disposable income, consumption, saving and net worth) have increased as users demand outputs that better support well-being and sustainability analysis.
  - Net measures can be more important than gross counterparts for capturing sustainability considerations and are conceptually preferred because they better reflect actual production costs; however, measurement challenges impede comparability across countries.
- Presentation guidance:
  - Net measures should be presented alongside gross measures, supplementing rather than replacing gross measures.
  - Improved quality and timeliness are needed to place greater emphasis on net measures.
- Conceptual change in the 2025 SNA:
  - The 2025 SNA improves articulation between depreciation (produced assets) and depletion (non-produced assets, particularly natural resources); both are treated as costs of production.
  - Gross-to-net adjustment now has two components, e.g., from GDP to NDP:
    - Gross Domestic Product (B1g) = 1 854
    - Depreciation (−) (P51c) = 222
    - Depletion (−) (K21) = 14
    - Net Domestic Product (B1n_P51c+K21) = 1 618
  - Showing both components (depreciation and depletion) allows users flexibility to use either version according to needs.
- Clarification of terminology:
  - "Net" and "gross" should be used by statistical producers only in two areas:
    - In conjunction with balancing items for each account by institutional sector and the whole economy, where the difference is due to depreciation and depletion.
    - In financial accounts: (i) aggregating acquisitions less disposals for each financial instrument on the asset side and on the liability side; (ii) aggregating changes in assets and changes in liabilities and showing both on a net basis (further details in paragraphs 4.268-4.276 [to confirm]).
  - Other legacy uses of "net" (e.g., "net exports") should be replaced by explicit expressions such as "exports less imports" to avoid confusion; removal of the word "net" is reflected in examples shown in Table 21.9.

### Use of clearer terminology for users
- Presentation and communication:
  - Macroeconomic statistics rely on technically precise terminology that may not be widely understood outside economic statistics; inconsistent internal usage may also occur.
  - Statistics should be presented and communicated to maximize analytical usefulness, quality, scope, comparability and policy applications for a wide user base.
  - Terminology and branding of macroeconomic frameworks need an international communication strategy that aligns with the latest technology and current cultural norms.

*Source: draft-bpm7-ch20-v7 (sections 21.91–21.114).*

### 21.115 The macroeconomic statistical standards have lots of similar technical descriptions of concepts and

### 21.115–21.119 Harmonisation of terminology and the Common Glossary in the 2025 SNA and BPM7

### Purpose and overarching changes
- The 2025 SNA and BPM7 introduce a Common Glossary of terms and definitions used in macroeconomic statistics to improve consistency, readability and understandability for non-specialists.
- The Common Glossary reflects input from: SNA; BPM; European System of Accounts (ESA); Government Finance Statistics; Monetary and Financial Statistics; System of Environmental-Economic Accounting (SEEA); International Public Sector Accounting Standards (IPSAS); and guidance from the Bank for International Settlements (BIS).
- The Common Glossary aims to:
  - harmonise labels and concepts across macroeconomic statistical standards;
  - deliver a further level of harmonisation across macroeconomic statistical standards;
  - provide users with a clearer and more consistent understanding of key economic terms and definitions together with some alternative easier to understand terms.
- The effort includes alignment of terminologies (affecting variables, account labels, etc.), improved branding of the statistical standards, and does not change technical definition(s) where those remain unchanged.

### Changes to account and group names (examples)
- Changes affecting names of the SNA accounts (examples):
  - The production account — No change
  - The generation of income account — The generation of earned income account
  - The allocation of primary income account — The allocation of earned income account
  - The secondary distribution of income account — Transfer income account
  - The use of income account — No change
  - The capital account — No change
  - The financial account — No change
  - Other changes in assets account — Other changes in assets and liabilities account
  - Balance sheets — No change; presented as Balance Sheets (needs to be cross-checked/amended against final Editorial Team agreement on Glossary Issue Note discussions)

- Changes affecting names of the BPM accounts (examples):
  - Goods and services account — Goods account / services account
  - Current account — (present as the group heading)
  - Primary income account — Earned income account
  - Secondary income account — Transfer income account
  - Capital account — No change (Capital account)
  - Financial account — No change
  - Other changes in financial assets and liabilities account — No change
  - International investment position — No change (Balance sheets)

### Specific terminology changes (examples retained verbatim where given)
- Accumulation accounts — Accumulation of economic assets account
- Adjusted disposable income — Disposable income adjusted for social transfers in kind
- Allocation of (other) primary income account — Allocation of (other) earned income account
- Balance on goods/services/goods and services — Balance of international trade in goods/services/goods and services
- Balance of primary incomes — Balance of earned incomes
- Compensation of employees — Remuneration of employees
- Constant prices — In volume terms
- Consumption of fixed capital — Depreciation
- Distribution of income account — Earned income account
- Financial intermediation services indirectly measured (FISIM) — Implicit financial services on loans and deposits
- Financial lease — Finance lease
- Generation of income account — Generation of earned income account
- Imputed rental of owner-occupied dwellings — Owner-occupied housing services
- Net errors and omissions — Statistical discrepancy
- Net fees — Fees less service charges
- Net guarantees — Guarantees less service charges
- Net (non-life) insurance premiums — Non-life insurance premiums less service charges
- Net re-insurance premiums — Re-insurance premiums less service charges
- Net social contributions — Social contributions less service charges
- Other changes in assets account — Other changes in assets and liabilities account
- Other changes in the volume of assets account — Other changes in the volume of assets and liabilities account
- Primary income — Earned income
- Product balance — Balance of the supply and use of products
- Purchases less sales of goodwill and marketing assets — Acquisitions less sales of goodwill and marketing assets
- Redistribution of income account — Transfer income account
- Redistribution of income in kind account — Social transfers in kind account
- Resource lease — Natural resource lease
- Resources — Revenues
- Secondary distribution of income account — Income transfers other than social transfers in kind account
- Trade margin — Distribution margin
- Use of adjusted disposable income account — Use of disposable income adjusted for social transfers in kind account
- Uses — Expenditures

### Intended benefits and user considerations
- The harmonisation and the Common Glossary are intended to:
  - facilitate communication across the producers’ community by using common labels and concepts;
  - improve comprehension and usability of macroeconomic statistics for non-specialists;
  - provide alternative, easier to understand terms alongside technical definitions.
- Statistical producers are urged to target their communication because some technical descriptions and underlying definitions remain not user-friendly or not understood by users.

---


_Source: https://www.imf.org/-/media/files/data/statistics/bmp7/draft-chapters/draft-bpm7-ch20-v7.pdf_