## _gdds-assess-08

## Source details

**Canonical URL:** [_gdds-assess-08](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdds/_gdds-assess-08.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdds/_gdds-assess-08.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdds/_gdds-assess-08.pdf.json)

---

### Executive summary — adoption, achievements, limitations
- Adoption and participation:
  - GDDS has had 96 participants; cumulative participation to date: 96 participants, six of whom now subscribe to the SDDS, resulting in 90 current GDDS participants.
  - Approximately 81 percent of the membership now participates in the IMF’s data initiatives—GDDS together with the SDDS.
  - Evidence cited that countries can reduce borrowing costs significantly by subscribing to or participating in either the Special Standard or the General System.
- Achievements:
  - Participants have produced a material improvement in the comprehensiveness and quality of their statistical systems.
- Limitations and weaker outcomes:
  - Only six participants have progressed from the GDDS to the SDDS.
  - Participants often lag behind their timetables for meeting developmental objectives.
  - After about 10 years of GDDS experience, data dissemination—particularly periodicity and timeliness—remains weak.
  - GDDS requires metadata and plans but does not require data dissemination; the IMF does not monitor participants’ data dissemination practices.

### Causes and context
- Contributing factors to modest dissemination impact:
  - Scarcity of resources and low priority assigned to statistics in national development plans.
  - Initial GDDS emphasis on system development and data quality over dissemination.
  - Global changes (greater integration, transparency demands, Internet dissemination) not fully reflected in GDDS design.
- Supply creates demand:
  - Publishing data raises statistical agencies’ profile, elicits user input improving quality, and can increase resource allocation to statistics.

### Proposed reorientation and specific reform elements
- Strategic shift:
  - Emphasize data dissemination to the public substantially more and metadata updating less.
  - Import key SDDS dissemination elements into the GDDS to strengthen dissemination practices and provide clearer guidance toward SDDS subscription.
- Specific incorporations and changes:
  - Incorporate the National Summary Data Page (NSDP) and the Advance Release Calendar (ARC) elements of the SDDS into the GDDS.
  - Incorporate data categories developed to serve capital market needs, such as the International Reserves and Foreign Currency Liquidity Template (reserves template).
  - Simplify and reformulate GDDS data categories to align more closely with SDDS and current user needs.
  - Adopt recommended ranges of timeliness and periodicity (rather than SDDS prescriptiveness), giving countries options and guidance.
  - Intensify technical support for Africa and the Middle East.
  - Recast the GDDS so that the Special Standard becomes a special case of the General System.

### Organizational and practical recommendations
- Reorganize the GDDS to strengthen data dissemination and increase priority given to plans for improvement.
- Simplify and reformulate data categories to better align with SDDS and user needs.
- Intensify regional technical support, especially for Africa and the Middle East.

### Rationale and expected effects of reform
- A reformed GDDS would:
  - Be a more general case of the SDDS while allowing more flexible recommended timeliness and periodicity.
  - Better serve capital market needs for GDDS countries borrowing in international markets.
  - Leverage the Internet and electronic publication to raise dissemination standards and reduce costs.
  - Encourage faster progress toward SDDS subscription via clearer dissemination-oriented guidance and practical tools (NSDP, ARC, reserves template), while preserving countries’ choice of pace.

*Source: Executive Summary of the GDDS assessment document.*

---

### Design, implementation, participation, and coordinators
- Implementation phases:
  - GDDS implemented in two phases; second phase started in May 2000 when first metadata were posted on the DSBB.
  - Rapid early uptake: 39 participants joined during the first two years.
- Participation metrics and regional coverage (2007 summaries):
  - Cumulative participation: 96 participants; six have graduated to SDDS, leaving 90 current GDDS participants.
  - Combined GDDS and SDDS participation near 80 percent of IMF membership.
  - Regional representation (combined GDDS and SDDS):
    - Africa, Europe, Western Hemisphere: about 90 percent representation.
    - Asia Pacific: about 65 percent participation; 82 percent excluding seven small island economies not participating.
    - Middle East and Central Asia: 63 percent representation (least represented region).
- National coordinator profile and requirement:
  - Requirement to appoint a national coordinator responsible for metadata and plans for improvement and for IMF communication.
  - Senior-level managers (head or deputy head of agency) account for 56 percent of GDDS coordinators.
  - Coordinators from the statistical office account for 52 percent of GDDS coordinators.

### Assessment approach and methodology
- Assessment focuses:
  - Metadata and methodologies, plans for improvement, sociodemographic data, and dissemination practices (periodicity and timeliness).
  - Compilation assessed by the statistical methodologies countries use (GFSM 1986, GFSM 2001, 1993 SNA, MFSM 2000, BPM5).
  - Plans expected to include time frame, financing, and technical assistance and be updated annually.
  - Sociodemographic data analyzed using World Bank time-series on statistical capacity indicators.

### Adoption of international statistical methodologies (selected GDDS group of 55 countries, as of 2007)
- Number of countries adopting:
  - 1993 SNA: 35
  - GFSM 2001: 7
  - MFSM: 31
  - BPM5: 50
- Adoption rates (percentage of group):
  - 1993 SNA: 64
  - GFSM 2001: 13
  - MFSM: 56
  - BPM5: 91
- Regional adoption highlights:
  - BPM5 adoption by region: APD 100 percent, MCD 92 percent, AFR 91 percent, WHD 85 percent.
  - 1993 SNA adoption by region: MCD 75 percent, APD 63 percent, WHD 62 percent, AFR 59 percent.
  - GFSM 2001 adoption lowest: MCD 17 percent, WHD 15 percent, AFR 14 percent, APD 0 percent.
- Conclusion: significant progress but uneven and incomplete; resources and technical expertise are important determinants.

---

### Plans for improvement — priorities and aggregated gaps (analysis of 55 countries; total mentions = 2,118)
- Frequency of issues mentioned in plans for improvement (number of mentions):
  - Rank 1 — 3.1 Source data: 512 mentions
  - Rank 2 — 2.2 Scope: 302 mentions
  - Rank 3 — 0.2 Resources: 220 mentions
  - Rank 4 — 3.3 Statistical techniques: 218 mentions
  - Rank 5 — 5.1 Data accessibility: 178 mentions
  - Rank 6 — 2.1 Concepts and definitions: 176 mentions
  - Rank 7 — 4.1 Periodicity and Timeliness: 118 mentions
  - Rank 8 — 2.3 Classification/sectorization: 105 mentions
  - Rank 9 — 5.2 Metadata accessibility: 70 mentions
  - Rank 10 — 0.1 Legal and institutional environment: 57 mentions
  - Rank 11 — 4.2 Consistency: 46 mentions
  - Rank 12 — 2.4 Basis for recording: 37 mentions
  - Rank 13 — 3.4 Assessment and validation of intermediate data and statistical outputs: 20 mentions
  - Rank 14 — 0.4 Other quality management: 19 mentions
  - Rank 15 — 4.3 Revision Policy and Practice: 15 mentions
  - Rank 16 — All Other: 25 mentions
- Observations:
  - Plans are not always fully comprehensive; metadata may identify additional issues not captured in plans.
  - Plans reflect authorities’ priorities and may underrepresent some weaknesses.

### Major issues by data category (ROSC category breakdown; number of issues shown)
- National Accounts (529 issues):
  - Prerequisites of quality: 6
  - Methodological soundness: 28
  - Accuracy and reliability: 48
  - Serviceability: 10
  - 4.1 Periodicity and timeliness: 5
  - Accessibility: 8
- Prices (267 issues):
  - Prerequisites of quality: 9
  - Methodological soundness: 25
  - Accuracy and reliability: 54
  - Serviceability: 1
  - 4.1 Periodicity and timeliness: 1
  - Accessibility: 11
- Government Operations (333 issues):
  - Prerequisites of quality: 18
  - Methodological soundness: 43
  - Accuracy and reliability: 17
  - Serviceability: 10
  - 4.1 Periodicity and timeliness: 7
  - Accessibility: 13
- Government Debt (287 issues):
  - Prerequisites of quality: 36
  - Methodological soundness: 13
  - Accuracy and reliability: 15
  - Serviceability: 10
  - 4.1 Periodicity and timeliness: 7
  - Accessibility: 25
- Depository Corporations Survey (294 issues):
  - Prerequisites of quality: 11
  - Methodological soundness: 39
  - Accuracy and reliability: 27
  - Serviceability: 9
  - 4.1 Periodicity and timeliness: 7
  - Accessibility: 14
- Balance of Payments (408 issues):
  - Prerequisites of quality: 12
  - Methodological soundness: 26
  - Accuracy and reliability: 47
  - Serviceability: 9
  - 4.1 Periodicity and timeliness: 6
  - Accessibility: 5
- All Datasets (2,118 issues total):
  - Prerequisites of quality: 14
  - Methodological soundness: 29
  - Accuracy and reliability: 36
  - Serviceability: 8
  - 4.1 Periodicity and timeliness: 6
  - Accessibility: 12

### Regional differences in major issues (percent of total major issues; total issues by region shown)
- AFR (989 issues): top elements: 3.1 Source data: 25; 2.2 Scope: 11; 0.2 Resources: 16; 3.3 Statistical techniques: 9; 5.1 Data accessibility: 7.
- APD (354 issues): top elements: 3.1 Source data: 28; 2.2 Scope: 14; 0.2 Resources: 9; 3.3 Statistical techniques: 10; 5.1 Data accessibility: 8.
- MCD (348 issues): top elements: 3.1 Source data: 23; 2.2 Scope: 13; 0.2 Resources: 5; 3.3 Statistical techniques: 14; 5.1 Data accessibility: 9.
- WHD (427 issues): top elements: 3.1 Source data: 21; 2.2 Scope: 22; 0.2 Resources: 3; 3.3 Statistical techniques: 11; 5.1 Data accessibility: 12.
- Cross-regional common major issues: source data, scope, resources, statistical techniques, data accessibility, and concepts and definitions.
- Pattern noted: data dissemination issues (periodicity and timeliness) are assigned relatively low priority across regions (4.1 Periodicity and Timeliness accounts for 6 percent of all dataset issues).

---

### Sociodemographic data and Statistical Capacity Indicators (SCI)
- Priority assigned to sociodemographic data:
  - Participants assign upgrade of sociodemographic data a relatively low priority.
  - Staff used World Bank SCI (more than 70 percent of criteria are sociodemographic).
  - Only 9.1 percent of GDDS participants in the group adjusted their sociodemographic metadata to include the MDGs.
- SCI changes 1999–2006 (scale 0–100; a score of 100 indicates meeting all criteria):
  - GDDS countries as a group: SCI increased by 24.0 percent over 1999 to 2006.
  - Regional increases 1999 to 2006:
    - AFR: 13.7 percent
    - APD: 34.8 percent
    - MCD: 32.7 percent
    - WHD: 28.8 percent
  - Subperiod changes:
    - 1999 to 2004 increase for GDDS countries as a group: 19.1 percent
    - 2004 to 2006 increase for GDDS countries as a group: 4.3 percent
    - The SCI increase slowed: only 4.3 percent increase for 2004–2006.
- Absolute average SCI scores (selected years):
  - Total average score: 1999 = 51.3; 2004 = 61.1; 2006 = 63.7
  - AFR average score: 1999 = 47.6; 2004 = 52.5; 2006 = 54.1
  - APD average score: 1999 = 54.1; 2004 = 68.9; 2006 = 73.0
  - MCD average score: 1999 = 53.1; 2004 = 65.5; 2006 = 70.5
  - WHD average score: 1999 = 56.5; 2004 = 71.4; 2006 = 72.8
- Number of countries that included MDGs in metadata:
  - AFR: 3
  - APD: 0
  - MCD: 1
  - WHD: 1
  - Total: 5
- Attribution:
  - Strong improvement between 1999 and 2004 likely attributable to United Nations Millennium Declaration and MDG monitoring development.
  - Despite GDDS amendment in late 2003 to recognize MDG indicators, only 9.1 percent adjusted metadata to include MDGs—possibly because MDG data available on World Bank and UN websites.

---

### GDDS dissemination performance — periodicity and timeliness (Tables 7 and 7a summaries)
- Overall averages (Timeliness and Periodicity combined; "All"):
  - Simple average, All sectors: 58.2 percent
  - Population (Sociodemographic sector): 43.6 percent
- Examples of GDDS participants meeting GDDS recommendations (Timeliness and Periodicity; All):
  - National accounts: 72.7 percent
  - Depository corporations survey / Banking survey: 89.1 percent
  - Consumer price index: 92.7 percent
  - Official reserves: 34.5 percent
  - Balance of payments: 69.1 percent
- Periodicity-only performance (All sectors):
  - Simple average, All sectors: 72.5 percent
  - Population (Sociodemographic sector): 47.3 percent
  - National accounts periodicity met by: 98.2 percent
  - Consumer price index periodicity met by: 94.5 percent
  - Balance of payments periodicity met by: 100.0 percent
- Observations:
  - Timeliness is an important constraint; periodicity-only averages are more than 10 percentage points higher than when timeliness is included.
  - For employment data: periodicity met by close to 70 percent, timeliness met by about 55 percent.
  - Averaged over all datasets, about 62 percent of requirements are met by GDDS participants (text summary).

### Comparison with SDDS requirements (Tables 8 and 9 summaries)
- Overall capability to meet SDDS periodicity and timeliness (Table 8):
  - For GDDS countries in the group: about 31.7 percent meet SDDS periodicity and timeliness requirements (Simple average, All sectors = 31.7 percent).
  - Regional shares meeting SDDS periodicity and timeliness (Simple average, All sectors):
    - AFR: 23.7 percent
    - APD: 42.9 percent
    - MCD: 42.9 percent
    - WHD: 28.0 percent
  - Population (Sociodemographic sector) meeting SDDS periodicity and timeliness: 43.6 percent.
- Periodicity-only SDDS compliance (Table 9):
  - Simple average: 53.1 percent
  - Population (Sociodemographic): 47.3 percent
  - National accounts periodicity (SDDS): 20.0 percent
  - Consumer price index periodicity (SDDS): 94.5 percent
  - Central bank periodicity (SDDS): 94.5 percent
  - Official reserves periodicity (SDDS): 74.5 percent
- Sectoral observations:
  - Most serious problems in the real sector (excluding CPI).
  - Financial sector shows least problems for periodicity but fails on SDDS timeliness.
  - For GDDS countries as a group, less than 26 percent met SDDS requirements for any real sector data category; this increases to 40 percent if only SDDS periodicity is considered.
  - For the financial sector, less than 37 percent met SDDS requirements for both data categories, but close to 95 percent met periodicity for both data categories.

### Gaps and potential GDDS revisions to meet SDDS
- To meet SDDS data requirements, countries need to compile and disseminate:
  - The reserves template (not included in the GDDS)
  - General government sector statistics (encouraged extension)
  - External debt (encouraged extension)
  - International investment position (IIP) (encouraged extension)
- Recommendation: revise GDDS to include all SDDS data categories to define a clearer path to SDDS.

### Reasons GDDS countries do not meet SDDS requirements (paragraphs 52–55)
- Identified factors:
  1. SDDS timeliness requirements are considerably tougher than GDDS recommendations, particularly for GDP, labor market, central government operations, and balance of payments statistics.
  2. GDDS countries may not prioritize improving timeliness due to absence of demand, limited technical assistance emphasis, or data prepared primarily for internal government use.
  3. First data releases often are hard-copy publications, which take more time to produce.
- Additional considerations:
  - Large lags for real and external sector data may reflect reliance on expensive, resource-intensive source data.
  - Regional deviations may reflect differences in capacity to absorb technical assistance.
- Regional standings meeting SDDS periodicity and timeliness:
  - APD and MCD meet about 43 percent of SDDS requirements.
  - WHD meets about 28 percent.
  - AFR meets about 24 percent.

---

### Capital market access, sovereign ratings, and data coverage for rating agencies
- Sovereign ratings among GDDS countries:
  - 36 GDDS countries in the group (65 percent) have sovereign ratings by international rating agencies.
  - Many GDDS countries incur compilation costs for rating agencies but do not publicly disseminate these data.
  - Listings by region (as of November 1, 2007) provided in source (selected examples include Seychelles, Mauritius, Botswana, China, Sri Lanka, Qatar, Kuwait, Trinidad and Tobago, Venezuela, Panama, etc.).
- Moody’s data categories coverage (Table 11 summary):
  - From 30 selected indicators:
    - GDDS: 22 (73.3%) covered (fully, partially, or encouraged).
    - SDDS: 26 (86.7%) covered (fully, partially, or encouraged).
  - Examples of indicators not fully covered by GDDS:
    - GDP per capita (PPP): No (GDDS), No (SDDS)
    - General government debt/GDP: No (GDDS), Partially (SDDS)
    - External debt (US dollars) and external debt/GDP: Partially (GDDS), Yes (SDDS)
    - Financial Soundness Indicators: External vulnerability indicator: No (GDDS), Partially (SDDS, res. template)
    - Liquidity ratio (liabilities to BIS banks within one year/total assets held in BIS banks): No (GDDS), No (SDDS)
  - Conclusions:
    - SDDS is a closer match to Moody’s fiscal and external sector needs.
    - Neither GDDS nor SDDS fully covers Financial Soundness Indicators required by Moody’s.
    - Moody’s reports are not explicit on timeliness and periodicity requirements (e.g., no explicit quarterly national accounts requirement).

---

### Progress toward SDDS under GDDS participation — five-country assessments (Botswana, Cambodia, Jordan, Mauritius, Panama)
- Assessment framework: (1) new data compiled/disseminated; (2) coverage improvements; (3) periodicity improvements; (4) timeliness improvements.
- Average assessment time frame: about five years (average 5.1 years).
- Aggregate results (Table 12 totals/averages):
  - Assessment Time Frame (year) average: 5.1
  - New Data Compiled total: 8
  - Improvements in Coverage total: 13
  - Improvements in Periodicity total: 15
  - Improvements in Timeliness total: 29
  - Data Categories Improved (total): 51
  - New Data Coverage total: 14
  - Period. and/or Timeliness total: 12
  - Improvements Needed for Meeting SDDS total: 38
- Country highlights:
  - Botswana (5 years): Data Categories Improved (total): 4; Improvements Needed for Meeting SDDS: 11.
  - Mauritius (5.5 years): Data Categories Improved (total): 14; Improvements Needed for Meeting SDDS: 9.
  - Cambodia (4.5 years): Data Categories Improved (total): 10; Improvements Needed for Meeting SDDS: 2.
  - Jordan (5.5 years): Data Categories Improved (total): 15; Improvements Needed for Meeting SDDS: 6.
  - Panama (5 years): Data Categories Improved (total): 8; Improvements Needed for Meeting SDDS: 10.
- Interpretation:
  - All five countries made some progress in about five years, especially in timeliness.
  - Progress varied; some countries slower than ROSC mission expectations (examples: Jordan, Mauritius).
  - Factors affecting pace: available resources and authorities’ commitment.

---

### Assessment conclusions and design considerations
- Participation growth and recognition:
  - Combined GDDS/SDDS participation covers five sixths of IMF membership.
  - GDDS developmental aspects are widely recognized.
- Shortcomings and pace:
  - Progress slower than envisaged due to resource scarcity and low prioritization.
  - Only six countries graduated from GDDS to SDDS (predominantly transition countries with stronger statistical traditions).
  - Sociodemographic data improved since 1999, but most members did not update metadata to include MDGs.
- Dissemination performance:
  - About 60 percent of participants meet both periodicity and timeliness objectives of the GDDS.
  - Almost three quarters meet periodicity goals; fewer than one half meet timeliness objectives.
  - Against SDDS standards: slightly more than one half meet periodicity standards and about one third meet timeliness standards.
  - Some GDDS participants have credit ratings and market access but are distant from SDDS subscription requirements; meeting SDDS could significantly reduce borrowing costs.
- Design considerations:
  - Original GDDS emphasis: comprehensive statistical frameworks and metadata; no commitment to public dissemination.
  - Common problems by DQAF categories: source data, scope, resources, statistical techniques, concepts and definitions.
  - Resource constraints and technical assistance availability influenced methodological adoption (e.g., BPM5 adopted by 91 percent; GFSM 2001 by 13 percent).
  - Global changes since GDDS inception have not been fully reflected in its design and implementation.

---

### Recasting the GDDS — staff proposals (Section VII–VIII excerpts)
- Strengthening emphasis on data dissemination:
  - Recast GDDS to emphasize data dissemination to the public more and metadata updates less.
  - Staff proposals:
    - Expect GDDS participants to establish a centralized NSDP and disseminate GDDS data on it.
    - Expect establishment of an ARC listing data releases for the year and post it on the DSBB and domestic publications.
    - Encourage (but not require) hyperlinks among the DSBB, NSDP, and ARC.
    - Countries to make a good faith commitment to dissemination objectives while GDDS remains non-monitored.
- Realigning data categories:
  - Simplify and reformulate GDDS data categories to approximate the SDDS and ease progression to SDDS; introduce a “capital market track” for the General System.
  - Begin compiling certain SDDS-like categories within GDDS: reserves template, general government statistics, IIP, external debt.
  - Expand and align GDDS data categories with SDDS but with less demanding periodicity and timeliness.
  - Specific dataset changes (high-level):
    - Real sector/National accounts: require GDP in current prices and GDP volume by production or expenditure approach with disaggregated components.
    - Production index: manufacturing/industrial indices; primary commodity/agricultural indices as relevant.
    - Fiscal sector: revenue, expenditure, balance, financing with breakdowns (debt holder, instrument, currency) as relevant.
    - Financial sector: net external position, domestic credit, broad or narrow money.
    - External sector/BOP: imports and exports of goods and services, current account balance, reserves, overall balance.
    - IIP and external debt: gross official reserve assets in U.S. dollars; reserve-related liabilities encouraged; public and publicly guaranteed external debt broken down by maturity; external debt service schedules.
- Strengthening plans for improvement:
  - Continue developmental role of plans; streamline data categories to reduce metadata burden.
  - Drop requirement for annual metadata updates; permit updating on a “best effort” or “when merited” basis.
  - Retain expectation that plans be updated regularly and be comprehensive.
  - Assign higher priority and prominence to periodicity and timeliness within plans.
  - Use plans for interagency coordination of technical assistance and results-based management.
- Strengthening emphasis on graduation to the SDDS:
  - Streamlining, NSDP, and ARC expected to ease progression to SDDS; consider adopting SDDS progression as a presumptive end-goal while preserving voluntary nature.
- Achieving better regional balance:
  - Increase participation in Middle Eastern countries.
  - Intensify work in Africa with countries interested in moving to the SDDS; rationale: more SDDS subscribers in Africa could create “success stories.”

### Costs, implementation, and operational impacts
- IMF (STA) cost implications:
  - Strengthening dissemination not foreseen to add additional STA cost but will require reorientation of technical assistance priorities toward transparency.
  - Many GDDS participants likely require technical assistance to establish NSDP and ARC; assistance within existing programs.
  - Data dissemination issues would be added to terms of reference of technical assistance missions; regional outreach missions or workshops anticipated.
- Participant cost and operational estimates:
  - Initial startup costs best estimate: between 4–12 person weeks, mainly IT skills and GDDS coordinator effort.
  - Ongoing maintenance: GDDS coordinator might use about 10–15 percent of his/her time to update NSDP and ARC.
  - Benefits include a central data page with timely data for policymakers and potential savings if NSDP/ARC replace paper-based publications.
  - Streamlined metadata reduces time updating metadata; emphasis on up-to-date plans increases effort but enables better results-based management and donor negotiations.

---

### Proposed changes, consultation, and Executive Directors’ questions (Section on consultations)
- Staff view and consultation process:
  - Proposed options would result in important changes but not a fundamental change in GDDS objectives.
  - Staff judgment: changes go far beyond previous reviews and should be considered through consultative process involving Fund staff, Executive Directors, and participating countries.
  - Consultation sequencing:
    - Informal seminar to seek Executive Directors’ views.
    - If supported, broader member consultations.
    - Formal recommendations to the Board at the Seventh Review of the Data Standards (presently scheduled for fall of 2008).
  - Consultations to include transition procedures as in prior data standards adoption.
- Questions for Executive Directors (sample):
  - Do Directors agree to reorganize GDDS to place more emphasis on public data dissemination?
  - Should plans for improvement emphasize periodicity and timeliness?
  - Should GDDS participants be expected to establish a centralized NSDP and an ARC, as in the SDDS?
  - Should participants with ARCs and NSDPs post hyperlinks on the DSBB?
  - Do Directors support simplifying GDDS data structure to approximate SDDS and eliminate complex comprehensive/other dataset distinctions?
  - Should plans for improvement receive higher priority in metadata and less emphasis on describing existing systems?
  - Should emphasis on graduation to SDDS be strengthened?
  - Do Directors agree to continue efforts to obtain better regional balance in GDDS participation?

*Source: IMF Statistics Department; Fitch Ratings; Standard and Poor’s; Moody’s Investors Service; World Bank SCI; excerpts and data as presented in the supplied GDDS assessment content.*

### Executive Summary ......................................................................................................

### _gdds-assess-08 - Executive Summary

### Adoption and participation
- IMF member countries have enthusiastically adopted the General Data Dissemination System (GDDS or General System); to date, the GDDS has had 96 participants.
- Approximately 81 percent of the membership now participates in the IMF’s data initiatives—GDDS together with the SDDS.
- The data initiative has been effective by these participation metrics and is recognized by financial market participants (evidence cited that countries can reduce borrowing costs significantly by subscribing to or participating in either the Special Standard or the General System).

### Achievements and limitations
- Countries that have participated in the GDDS have met many developmental objectives, producing a material improvement in the comprehensiveness and quality of their statistical systems.
- Limitations and weaker outcomes identified:
  - Only six participants have progressed from the GDDS to the SDDS.
  - Participants often have lagged behind the timetables they established for meeting particular developmental objectives.
  - After about 10 years of GDDS experience, data dissemination—particularly periodicity and timeliness of data—remains weak.
  - The GDDS originally emphasized development of national systems and metadata over active data dissemination; there is no requirement for GDDS participants actually to disseminate data, nor does the IMF monitor participants’ data dissemination practices.

### Analysis of causes and context
- Contributing factors to modest impact on dissemination include:
  - Scarcity of resources and low priority assigned to statistics in national development plans.
  - The initial GDDS philosophical emphasis on data quality improvements and system development, with dissemination seen as a later-stage objective.
  - Changes in the global environment since GDDS inception—greater global integration, heightened emphasis on transparency and governance, increased time sensitivity of users, spread of the Internet and electronic publication—have not been fully reflected in GDDS design and implementation.
- The paper highlights that supply (publishing data) creates its own demand: publishing data can raise the profile of statistical agencies, elicit user input that improves quality, and increase allocation of resources to statistics.

### Proposed reorientation and specific reform elements
- General strategic shift recommended:
  - Place substantially more emphasis on data dissemination to the public and less emphasis on simply updating metadata descriptions of existing systems.
  - Import key dissemination elements of the SDDS into the GDDS to strengthen dissemination practices and provide clearer guidance toward SDDS subscription.
- Specific proposed incorporations and changes:
  - Incorporate the National Summary Data Page (NSDP) and the Advance Release Calendar (ARC) elements of the SDDS into the GDDS.
  - Incorporate data categories developed to serve capital market needs, such as the International Reserves and Foreign Currency Liquidity Template (reserves template).
  - Simplify and reformulate the GDDS data categories to more closely align them with those of the SDDS and current user needs.
  - Adopt recommended ranges of timeliness and periodicity (rather than the prescriptiveness of the SDDS), giving countries options and guidance while allowing them to choose their own pace of development.
  - Intensify technical support for data standards in Africa and the Middle East to achieve better regional balance.
  - Recast the GDDS so that the Special Standard becomes a special case of the General System.

### Organizational and practical recommendations (as identified by staff)
- Reorganize the GDDS to strengthen the data dissemination aspect and assign a higher priority to the plans for improvement.
- Simplify and reformulate data categories to better align with SDDS and user needs.
- Intensify regional technical support, especially for Africa and the Middle East.

### Rationale and expected effects
- A reformed GDDS would:
  - Be more truly a general case of the SDDS, including a larger number of data categories (owing to sociodemographic categories) while allowing more flexible recommended timeliness and periodicity.
  - Better serve capital market needs for GDDS countries that borrow in international capital markets and are subject to sovereign ratings.
  - Leverage the Internet and electronic publication to “raise the bar” on dissemination while reducing dissemination costs.
  - Encourage faster progress toward SDDS subscription by providing clearer dissemination-oriented guidance and practical tools (NSDP, ARC, reserves template), while preserving countries’ ability to choose their pace.

*Source: Executive Summary of the GDDS assessment document.*

### 11.      The design and implementation of the GDDS have benefited from close collaboration

### 11.      The design and implementation of the GDDS have benefited from close collaboration

### Participation: growth and regional coverage
- GDDS implemented in two phases: first phase focused on education and training; second phase started in May 2000 when the first metadata for countries participating in the GDDS were posted on the IMF’s Dissemination Standards Bulletin Board (DSBB).
- Rapid early uptake: 39 participants joined during the first two years.
- Cumulative participation to date: 96 participants, six of whom now subscribe to the SDDS, resulting in 90 current GDDS participants.
- Table 1 (summary statements):
  - Total GDDS participants (net) reported as 90.
  - Graduated to SDDS: 6.
- Regional representation (combined GDDS and SDDS) in 2007:
  - Overall participation near 80 percent of the IMF membership.
  - Africa, Europe, and Western Hemisphere: about 90 percent representation in the two initiatives.
  - Asia Pacific: about 65 percent participation; participation rate is 82 percent excluding seven small island economies not participating.
  - Middle East and Central Asia: 63 percent representation (least represented region). Within this region, participation is lowest among Middle East countries; Central Asian countries are well represented.
- GDDS national coordinator requirements and profile:
  - Requirement: appoint a national coordinator responsible for coordinating metadata and plans for improvement and for communicating with the IMF.
  - Senior-level managers (head or deputy head of agency) account for 56 percent of total GDDS coordinators.
  - Coordinators from the statistical office account for 52 percent of total GDDS coordinators.

### Performance of the GDDS: assessment approach
- Assessment focus:
  - Analyze metadata and methodologies, plans for improvement against developmental needs, and sociodemographic data.
  - Assess data dissemination practices: observance of periodicity and timeliness of data.
  - Compare country dissemination practices against SDDS benchmarks to highlight gaps and technical assistance needs.
- Key methodological framing:
  - Compilation assessed by the statistical methodologies countries use (e.g., GFSM 1986, GFSM 2001, 1993 SNA, MFSM 2000, BPM5).
  - Plans for improvement expected to be elaborated in technical assistance missions and updated annually.
  - Sociodemographic data analyzed using World Bank time-series on statistical capacity indicators.

### Adoption of international statistical methodologies (as of 2007)
- Aggregate adoption across selected GDDS countries (group of 55 GDDS participants) — number of countries:
  - 1993 SNA: 35
  - GFSM 2001: 7
  - MFSM: 31
  - BPM5: 50
- Adoption rates (percentage of all countries in the group):
  - 1993 SNA: 64
  - GFSM 2001: 13
  - MFSM: 56
  - BPM5: 91
- Regional adoption highlights:
  - BPM5 highest adoption in all regions; APD 100 percent, MCD 92 percent, AFR 91 percent, WHD 85 percent.
  - 1993 SNA adoption: MCD 75 percent, APD 63 percent, WHD 62 percent, AFR 59 percent.
  - MFSM adoption lower in AFR and APD: AFR 50 percent, APD 25 percent; higher in WHD 85 percent and MCD 58 percent.
  - GFSM 2001 adoption lowest across regions: MCD 17 percent, WHD 15 percent, AFR 14 percent, APD 0 percent.
- Conclusion: participants have made significant progress in adopting current best-practice methodologies, but progress varies across countries and regions and remains incomplete in many cases.
- Resources and technical expertise are important determinants of adoption and implementation.

### Countries’ GDDS plans for improvement: priorities and gaps
- Purpose of plans: reflect actions needed to meet GDDS recommendations; encouraged to include time frame, financing, and technical assistance.
- Analytical framework: plans categorized using IMF’s Data Quality Assessment Framework (DQAF).
- Aggregate analysis of 55 countries — frequency of issues mentioned in plans for improvement (number of times an issue is mentioned; total mentions = 2,118):
  - Rank 1 — 3.1 Source data: 512 mentions
  - Rank 2 — 2.2 Scope: 302 mentions
  - Rank 3 — 0.2 Resources: 220 mentions
  - Rank 4 — 3.3 Statistical techniques: 218 mentions
  - Rank 5 — 5.1 Data accessibility: 178 mentions
  - Rank 6 — 2.1 Concepts and definitions: 176 mentions
  - Rank 7 — 4.1 Periodicity and Timeliness: 118 mentions
  - Rank 8 — 2.3 Classification/sectorization: 105 mentions
  - Rank 9 — 5.2 Metadata accessibility: 70 mentions
  - Rank 10 — 0.1 Legal and institutional environment: 57 mentions
  - Rank 11 — 4.2 Consistency: 46 mentions
  - Rank 12 — 2.4 Basis for recording: 37 mentions
  - Rank 13 — 3.4 Assessment and validation of intermediate data and statistical outputs: 20 mentions
  - Rank 14 — 0.4 Other quality management: 19 mentions
  - Rank 15 — 4.3 Revision Policy and Practice: 15 mentions
  - Rank 16 — All Other: 25 mentions
- Observations on plans’ comprehensiveness:
  - Plans are not always fully comprehensive; metadata may identify issues not picked up in plans for improvement.
  - Plans reflect authorities’ priorities and may underrepresent some weaknesses.

### Major issues by data category (ROSC category breakdown)
- Distribution of major issues by DQAF dimension for datasets (percent of total major issues; total number of issues by dataset shown):
  - National Accounts (529 issues):
    - Prerequisites of quality: 6
    - Methodological soundness: 28
    - Accuracy and reliability: 48
    - Serviceability: 10
    - 4.1 Periodicity and timeliness: 5
    - Accessibility: 8
  - Prices (267 issues):
    - Prerequisites of quality: 9
    - Methodological soundness: 25
    - Accuracy and reliability: 54
    - Serviceability: 1
    - 4.1 Periodicity and timeliness: 1
    - Accessibility: 11
  - Government Operations (333 issues):
    - Prerequisites of quality: 18
    - Methodological soundness: 43
    - Accuracy and reliability: 17
    - Serviceability: 10
    - 4.1 Periodicity and timeliness: 7
    - Accessibility: 13
  - Government Debt (287 issues):
    - Prerequisites of quality: 36
    - Methodological soundness: 13
    - Accuracy and reliability: 15
    - Serviceability: 10
    - 4.1 Periodicity and timeliness: 7
    - Accessibility: 25
  - Depository Corporations Survey (294 issues):
    - Prerequisites of quality: 11
    - Methodological soundness: 39
    - Accuracy and reliability: 27
    - Serviceability: 9
    - 4.1 Periodicity and timeliness: 7
    - Accessibility: 14
  - Balance of Payments (408 issues):
    - Prerequisites of quality: 12
    - Methodological soundness: 26
    - Accuracy and reliability: 47
    - Serviceability: 9
    - 4.1 Periodicity and timeliness: 6
    - Accessibility: 5
  - All Datasets (2,118 issues total):
    - Prerequisites of quality: 14
    - Methodological soundness: 29
    - Accuracy and reliability: 36
    - Serviceability: 8
    - 4.1 Periodicity and timeliness: 6
    - Accessibility: 12

### Regional differences in major issues (Table 5 highlights)
- Major issues by DQAF element as percent of total major issues, by region (total number of issues by region shown):
  - AFR (989 issues): top elements
    - 3.1 Source data: 25
    - 2.2 Scope: 11
    - 0.2 Resources: 16
    - 3.3 Statistical techniques: 9
    - 5.1 Data accessibility: 7
  - APD (354 issues): top elements
    - 3.1 Source data: 28
    - 2.2 Scope: 14
    - 0.2 Resources: 9
    - 3.3 Statistical techniques: 10
    - 5.1 Data accessibility: 8
  - MCD (348 issues): top elements
    - 3.1 Source data: 23
    - 2.2 Scope: 13
    - 0.2 Resources: 5
    - 3.3 Statistical techniques: 14
    - 5.1 Data accessibility: 9
  - WHD (427 issues): top elements
    - 3.1 Source data: 21
    - 2.2 Scope: 22
    - 0.2 Resources: 3
    - 3.3 Statistical techniques: 11
    - 5.1 Data accessibility: 12
- Cross-regional common major issues: source data, scope, resources, statistical techniques, data accessibility, and concepts and definitions.
- Noted pattern: data dissemination issues (periodicity and timeliness) are assigned relatively low priority across regions.

### Implications and forward emphasis
- Current orientation of GDDS places relatively low priority on dissemination issues; hence periodicity and timeliness receive limited prominence in plans for improvement (4.1 Periodicity and Timeliness accounts for 6 percent of all dataset issues).
- If IMF focused more on dissemination aspects of the GDDS, periodicity and timeliness would be expected to become more prominent in countries’ plans for improvement.
- Sociodemographic data are recommended by the GDDS but are not included in the SDDS; earlier sections note use of World Bank indicators to analyze these data (analysis not detailed in the extracted content).

*Source: IMF Statistics Department (excerpts from _gdds-assess-08 - 11.      The design and implementation of the GDDS have benefited from close collaboration).*

### 38.      This section of the paper examines the developments in sociodemographic data over

### _gdds-assess-08 - 38.      This section of the paper examines the developments in sociodemographic data over

### Developments in sociodemographic data and priority
- Participants assign the upgrade of sociodemographic data a relatively low priority.
- The staff used the World Bank Statistical Capacity Indicators (SCI) as a proxy; more than 70 percent of the criteria included in the dimensions are sociodemographic data.
- Only 9.1 percent of the GDDS participants in the group have adjusted their sociodemographic metadata to include the MDGs.

### Statistical Capacity Indicators (SCI) — changes 1999–2006 (from Table 6)
- Scale: 0–100. A score of 100 indicates that a region meets all the criteria.
- For the GDDS countries as a group, the SCI increased by 24.0 percent over the seven years (1999 to 2006).
- Regional increases 1999 to 2006:
  - AFR: 13.7 percent
  - APD: 34.8 percent
  - MCD: 32.7 percent
  - WHD: 28.8 percent
- Changes by subperiod:
  - 1999 to 2004 increase for GDDS countries as a group: 19.1 percent
  - 2004 to 2006 increase for GDDS countries as a group: 4.3 percent
  - The SCI increase slowed significantly over the last two years (2004–2006), with only 4.3 percent increase for all countries as a group.
- Absolute average scores (selected years, Table 6):
  - Total average score: 1999 = 51.3; 2004 = 61.1; 2006 = 63.7
  - AFR average score: 1999 = 47.6; 2004 = 52.5; 2006 = 54.1
  - APD average score: 1999 = 54.1; 2004 = 68.9; 2006 = 73.0
  - MCD average score: 1999 = 53.1; 2004 = 65.5; 2006 = 70.5
  - WHD average score: 1999 = 56.5; 2004 = 71.4; 2006 = 72.8
- Number of countries which included MDGs in metadata (Table 6):
  - AFR: 3
  - APD: 0
  - MCD: 1
  - WHD: 1
  - Total: 5

### Attribution of SCI improvements
- The strong improvement in the SCI between 1999 and 2004 is likely most attributable to the commitment made by countries in 2000—the United Nations Millennium Declaration—and the subsequent development of statistics to track the eight Millennium Development Goals (MDGs).
- Although the GDDS was amended in late 2003 to explicitly recognize the United Nations’ MDG indicators and the development of appropriate statistical monitoring systems, only 9.1 percent of GDDS participants adjusted metadata to include the MDGs, possibly because extensive MDG data are available on World Bank and United Nations websites.

### GDDS dissemination performance: timeliness and periodicity (summary of Tables 7 and 7a)
- Overall averages (Timeliness and Periodicity combined; Table 7 — "All"):
  - Simple average, All sectors: 58.2 percent
  - Population (Sociodemographic sector): 43.6 percent
- Examples of GDDS participants meeting GDDS recommendations (Timeliness and Periodicity; Table 7, All):
  - National accounts: 72.7 percent
  - Depository corporations survey / Banking survey: 89.1 percent
  - Consumer price index: 92.7 percent
  - Official reserves: 34.5 percent
  - Balance of payments: 69.1 percent
- Periodicity-only performance (Table 7a, All):
  - Simple average, All sectors: 72.5 percent
  - Population (Sociodemographic sector): 47.3 percent
  - National accounts periodicity met by: 98.2 percent
  - Consumer price index periodicity met by: 94.5 percent
  - Balance of payments periodicity met by: 100.0 percent
- Observations:
  - Timeliness is an important constraint; periodicity-only averages are more than 10 percentage points higher than when timeliness is included.
  - For employment data: periodicity met by close to 70 percent, but timeliness met by about 55 percent.
  - Averaged over all datasets, about 62 percent of requirements are met by GDDS participants (text summary).

### Comparison with SDDS requirements (Tables 8 and 9)
- Overall capability to meet SDDS periodicity and timeliness requirements (Table 8):
  - For GDDS countries included in the group, about 31.7 percent meet SDDS periodicity and timeliness requirements (Table 8: Simple average, All sectors = 31.7 percent).
  - Regional shares meeting SDDS periodicity and timeliness (Table 8, Simple average, All sectors):
    - AFR: 23.7 percent
    - APD: 42.9 percent
    - MCD: 42.9 percent
    - WHD: 28.0 percent
  - Population (Sociodemographic sector) meeting SDDS periodicity and timeliness: 43.6 percent (Table 8).
- Periodicity-only SDDS compliance (Table 9):
  - Simple average: 53.1 percent
  - Population (Sociodemographic): 47.3 percent
  - National accounts periodicity (SDDS): 20.0 percent
  - Consumer price index periodicity (SDDS): 94.5 percent
  - Central bank periodicity (SDDS): 94.5 percent
  - Official reserves periodicity (SDDS): 74.5 percent
- Sectoral observations:
  - Countries experience the most serious problems in the real sector (excluding CPI).
  - Financial sector shows the least problems, but countries struggle to comply with SDDS timeliness requirements in that sector.
  - For GDDS countries as a group, less than 26 percent met SDDS requirements for any real sector data category; this increases to 40 percent if only SDDS periodicity is considered.
  - For the financial sector, less than 37 percent met SDDS requirements for both data categories, but close to 95 percent met periodicity requirements for both data categories.

### Gaps and potential GDDS revisions
- To meet SDDS data requirements, countries need to compile and disseminate data on:
  - The reserves template (not included in the GDDS)
  - General government sector (included as encouraged extension)
  - External debt (included as encouraged extension)
  - International investment position (IIP) (included as encouraged extension)
- The GDDS could be revised to include all SDDS data categories to help define the path for countries to graduate to the SDDS.

### Reasons why GDDS countries do not meet SDDS requirements (paragraphs 52–55)
- Identified factors:
  1. SDDS timeliness requirements are considerably tougher than GDDS recommendations, particularly for GDP, labor market, central government operations, and balance of payments statistics; GDDS participation by itself does not prepare countries to meet SDDS requirements.
  2. GDDS countries may not give high priority to improving timeliness of dissemination; possible causes include absence of demand, lack of emphasis on timeliness in technical assistance programs, and data prepared primarily for internal government use rather than public dissemination.
  3. First data releases often take the form of hard-copy publications, which take more time to produce (even if subsequently posted on the Internet).
- Additional considerations:
  - Large lags for real sector and external sector data may reflect reliance on expensive and resource-intensive source data.
  - Regional deviations in compliance could result from differences in availability of resources to absorb capacity-building technical assistance.
- Regional standings in meeting SDDS requirements (text summary):
  - Asia and the Pacific (APD) and Middle East and Central Asia (MCD) meet about 43 percent of the SDDS requirements.
  - Western Hemisphere (WHD) meets about 28 percent.
  - Africa (AFR) meets about 24 percent of the periodicity and timeliness requirements.

### Transition note — Capital market access (start of Section IV)
- The GDDS was developed for a broad group of countries not necessarily with access to capital markets; the SDDS was developed against the background of informational failures affecting capital markets.
- Virtually all SDDS subscribers are active borrowers in capital markets.
- Tables 10 and 11 (not included in this excerpt) are referenced as showing why the assumption about GDDS countries lacking capital market access is no longer fully justified.

*Source: IMF Statistics Department; World Bank SCI as reported in the provided GDDS assessment text.*

### 57.      Table 10 lists GDDS countries that have received commercial ratings. The table

### Table 10 lists GDDS countries that have received commercial ratings

### Sovereign Ratings among GDDS Countries
- 36 GDDS countries included in the group (65 percent) have sovereign ratings by international rating agencies.
- Rating agencies have assigned sovereign ratings despite possible data shortcomings in GDDS countries, indicating agencies had sufficient data at their disposal.
- Implication: many GDDS countries incur the costs of compiling information required by credit rating agencies but do not reap the benefit of public dissemination.
- The GDDS countries that have sovereign ratings are listed by region in Table 10 (as of November 1, 2007).*  
  - AFR: Seychelles (S&P), Mauritius (Moody’s), Botswana (S&P, Moody’s), Namibia (Fitch), Nigeria (S&P, Fitch), Senegal (S&P), Kenya (S&P), Mozambique (S&P, Fitch), The Gambia (Fitch), Uganda (Fitch), Madagascar (S&P), Rwanda (Fitch), Malawi (Fitch).
  - APD: China (S&P, Moody’s, Fitch), Sri Lanka (S&P, Fitch), Mongolia (S&P, Moody’s, Fitch), Vietnam (S&P, Moody’s, Fitch), Cambodia (S&P, Moody’s).
  - MCD: Qatar (S&P, Moody’s), Kuwait (S&P, Moody’s, Fitch), Oman (S&P, Moody’s), Macedonia (S&P, Fitch), Albania (Moody’s), Jordan (S&P, Moody’s), Azerbaijan (Moody’s, Fitch), Georgia (S&P, Fitch), Pakistan (S&P, Moody’s).
  - WHD: Trinidad and Tobago (S&P, Moody’s), Venezuela (S&P, Moody’s, Fitch), Panama (S&P, Moody’s, Fitch), Grenada (S&P), Belize (S&P, Moody’s), Guatemala (S&P, Moody’s, Fitch), Honduras (Moody’s), Bolivia (S&P, Moody’s, Fitch), Nicaragua (Moody’s).
- Note: Sources cited in the table include IMF Statistics Department, Fitch Ratings, Standard and Poor’s, and Moody’s Investors Service.  
- Footnote: * Countries in the selected group of 55 with sovereign ratings.

### Moody’s Data Categories Not Fully Covered by GDDS/SDDS (Table 11)
- Comparison focus: data categories used by Moody’s for sovereign bond ratings that are not fully covered in GDDS and SDDS.
- Coverage summary (from 30 selected indicators):
  - GDDS: 22 (73.3%) of data categories covered (fully, partially, or encouraged).
  - SDDS: 26 (86.7%) of data categories covered (fully, partially, or encouraged).
- Specific indicators not fully covered by GDDS (selected examples preserved from Table 11):
  - I. Economic structure and performance (Real sector)
    - GDP per capita (PPP basis): No (GDDS), No (SDDS)
    - Gross Investment/GDP (%): Partially (GDDS), Partially (SDDS)
    - Nominal and real exports and imports of goods and services (% change): Partially (GDDS), Partially (SDDS)
    - Net exports of goods and services/GDP (%): Partially (GDDS), Partially (SDDS)
    - Openness of the economy (exports + imports of goods and services/GDP) (%): Partially (GDDS), Partially (SDDS)
  - II. Government finance (Fiscal sector)
    - General government revenue/GDP, expenditure/GDP, and financial balance/GDP (%): Encouraged (GDDS), Yes (SDDS)
    - General government primary balance/GDP: No (GDDS), Encouraged (SDDS)
    - General government debt/GDP (%) and general government debt/general government revenue (%): No (GDDS), Partially (SDDS)
    - General government interest payments/general government revenue (%): No (GDDS), Encouraged (SDDS)
  - III. External payments and debt (External sector)
    - Real effective exchange rate (% change): No (GDDS), No (SDDS)
    - Relative unit labor costs (index): Partially (GDDS), Partially (SDDS)
    - External debt (US dollars) and external debt/GDP (%): Partially (GDDS), Yes (SDDS)
    - External debt/current account receipts (%): Partially (GDDS), Yes (SDDS)
    - Net foreign direct investment/GDP (%): Partially (GDDS), Yes (SDDS)
    - Net IIP/GDP (%): Encouraged (GDDS), Yes (SDDS)
  - IV. Monetary, vulnerability, and liquidity indicators (Financial sector)
    - Debt-service ratio (interest+current year principal/current account receipts) (%): Partially (ext. sector) (GDDS), Encouraged (ext. sector) (SDDS)
    - Dollarization ratio (total foreign currency deposits in domestic banks/total deposits in domestic banks) (%): Partially (GDDS), Partially (SDDS)
    - Dollarization vulnerability indicator (foreign currency deposits in domestic banks/official foreign exchange reserves+foreign assets in domestic banks) (%): Partially (GDDS), Partially (SDDS)
    - Financial Soundness Indicators: External vulnerability indicator: No (GDDS), Partially (SDDS, res. template)
    - Liquidity ratio (liabilities to BIS banks within one year/total assets held in BIS banks): No (GDDS), No (SDDS)
- Conclusion drawn in the source:
  - Both GDDS and SDDS broadly cover Moody’s data requirements; the SDDS is a closer match for some fiscal and external sector statistics (covers 86.7 percent of data categories).
  - The GDDS does not require dissemination of these data categories but recommends developing the datasets.
  - Neither GDDS nor SDDS covers the Financial Soundness Indicators (FSIs) required by Moody’s.
  - The data requirements in Moody’s reports are not explicit on timeliness and periodicity (for example, Moody’s reports do not mention the need for quarterly national accounts as required by the SDDS).

### Progress Toward SDDS under GDDS Participation
- Five GDDS participants were assessed: Botswana (AFR), Cambodia (APD), Jordan (MCD), Mauritius (AFR), and Panama (WHD).
- Assessment framework (four aspects): (1) new data categories compiled/disseminated; (2) improvements in coverage; (3) improvements in periodicity; (4) improvements in timeliness.
- Average assessment time frame: about five years.
- Overall finding: All five countries made some progress in about five years, especially regarding timeliness.
- Table 12 summary (Improvements in Statistical Practices — Number of data categories improved):
  - Botswana — Assessment Time Frame (year): 5; New Data Compiled: None; Improvements in Coverage: 1; Improvements in Periodicity: 1; Improvements in Timeliness: 2; Data Categories Improved (total): 4; New Data Coverage: 3; Period. and/or Timeliness: 1; Improvements Needed for Meeting SDDS: 11.
  - Mauritius — Assessment Time Frame (year): 5.5; New Data Compiled: None; Improvements in Coverage: 2; Improvements in Periodicity: 9; Improvements in Timeliness: 11; Data Categories Improved (total): 14; New Data Coverage: 1; Period. and/or Timeliness: 3; Improvements Needed for Meeting SDDS: 9.
  - Cambodia — Assessment Time Frame (year): 4.5; New Data Compiled: 2; Improvements in Coverage: 5; Improvements in Periodicity: 2; Improvements in Timeliness: 5; Data Categories Improved (total): 10; New Data Coverage: 8; Period. and/or Timeliness: 3; Improvements Needed for Meeting SDDS: 2.
  - Jordan — Assessment Time Frame (year): 5.5; New Data Compiled: 4; Improvements in Coverage: 3; Improvements in Periodicity: None; Improvements in Timeliness: 9; Data Categories Improved (total): 15; New Data Coverage: 0; Period. and/or Timeliness: 4; Improvements Needed for Meeting SDDS: 6.
  - Panama — Assessment Time Frame (year): 5; New Data Compiled: 2; Improvements in Coverage: 2; Improvements in Periodicity: 3; Improvements in Timeliness: 2; Data Categories Improved (total): 8; New Data Coverage: 2; Period. and/or Timeliness: 1; Improvements Needed for Meeting SDDS: 10.
  - Total/average: Assessment Time Frame (year) average: 5.1; New Data Compiled total: 8; Improvements in Coverage total: 13; Improvements in Periodicity total: 15; Improvements in Timeliness total: 29; Data Categories Improved (total): 51; New Data Coverage total: 14; Period. and/or Timeliness total: 12; Improvements Needed for Meeting SDDS total: 38.
- Interpretation and timing:
  - Progress varied: Cambodia, Jordan, and Mauritius made significant improvements; Botswana and Panama made relatively less progress.
  - Pace has been slow in some cases. Examples from ROSC mission expectations:
    - Jordan: 2002 data ROSC mission suggested meeting all SDDS requirements by February 2005.
    - Mauritius: 2001 data ROSC mission suggested SDDS subscription by July 2004.
  - Factors affecting progress: available resources and commitment of authorities.

### Assessment and Conclusions (Overall Assessment)
- Participation growth and recognition:
  - Combined participation/subscription in the GDDS/SDDS covers five sixths of the IMF membership.
  - The developmental aspects of the GDDS have been widely recognized.
- Shortcomings and pace:
  - Overall progress has been slower than envisaged, partly due to scarcity of resources and low priority assigned to statistics in national development plans.
  - Only six countries have progressed from GDDS subscribing to SDDS subscription (all transition countries mostly with a strong tradition and infrastructure in statistics).
  - Although sociodemographic data improved since 1999, member countries did not generally update sociodemographic metadata to include the MDGs.
- Dissemination performance:
  - About 60 percent of participants meet both periodicity and timeliness objectives of the GDDS.
  - Almost three quarters of participants meet the periodicity goals; fewer than one half meet the timeliness objective.
  - When judged against SDDS standards: slightly more than one half of GDDS participants meet periodicity standards and about one third meet timeliness standards.
  - Some GDDS participants have credit ratings and market access but remain some distance from meeting SDDS subscription requirements; meeting SDDS would result in a significant reduction in borrowing costs for these countries.
- Design considerations:
  - The GDDS initial emphasis was on developing comprehensive statistical frameworks with less immediate emphasis on dissemination; participating countries commit to (1) use the GDDS as a developmental framework, (2) designate a country coordinator, and (3) provide metadata describing current statistical practices and plans for improvement — there is no commitment to data dissemination per se.
  - Common problems identified (classified by DQAF categories): source data, scope, resources, statistical techniques, and concepts and definitions.
  - Resource constraints and availability of technical assistance have influenced the speed of methodological adoption (e.g., 91 percent of GDDS participants adopted BPM5; 13 percent adopted GFSM 2001).
  - Important global changes since GDDS inception (globalization, emphasis on transparency and good governance, spread of the Internet and electronic dissemination) have yet to be fully reflected in GDDS design and implementation.

*Italicized source attribution: IMF Statistics Department; Fitch Ratings; Standard and Poor’s; Moody’s Investors Service; data as presented in the supplied content.

### 74.      The foregoing assessment provides a basis to reconsider and fine-tune certain aspects

### VII. RECASTING THE GDDS

### Strengthening the emphasis on data dissemination
- Recast the General Data Dissemination System (GDDS) to give more emphasis to data dissemination to the public and less to generating and updating metadata descriptions of existing statistical practice.
- Rationale:
  - Original GDDS emphasized developmental processes because: (1) quality deficiencies could undermine usefulness of disseminated data; (2) data users’ needs in nonmarket borrowing countries were less time-sensitive than in SDDS countries.
  - Increased importance of data dissemination: dissemination creates demand for higher information quality and more extensive coverage, raises profile of statistical agencies, can lead to higher priority and resources for statistics, and responds to increased time sensitivity of users and rising governance/accountability standards.
  - Spread of the Internet and electronic publication has “raised the bar” and reduced the cost of data dissemination but this has yet to be reflected in the General System.
- Staff proposals:
  - Expect GDDS participants to establish a centralized National Summary Data Page (NSDP) and disseminate their GDDS data on it.
  - Expectation to establish an advance release calendar (ARC) listing data releases for the year ahead and post it on the DSBB and in domestic publications.
  - Encourage (but not require) hyperlinks among the DSBB, the NSDP, and the ARC.
  - Countries asked to make a good faith commitment to achieving dissemination objectives, while GDDS remains a non-monitored standard (unlike the SDDS).

### Realigning the data categories
- Objective: simplify and reformulate GDDS data categories to more closely approximate the SDDS to ease progression to the SDDS and better meet market/credit rating agency needs.
- Key proposals:
  - Begin compiling certain data categories as part of the GDDS, such as the reserves template, general government statistics, the IIP, and external debt.
  - Expand and bring GDDS data categories into conformance with the SDDS, with less demanding requirements for periodicity and timeliness.
  - Emphasize indicator series in “other data categories,” streamline comprehensive frameworks and combine them with some other data categories to eliminate redundancies and the distinction between comprehensive and other datasets — characterized as providing a “capital market track” for the General System.
- Specific dataset and component changes (only changes are shown in source):
  - Real sector / National accounts: require GDP in current prices and GDP volume by production approach with disaggregated components; or GDP in current prices and GDP volume by expenditure category with disaggregated components.
  - Production index: manufacturing/industrial indices; primary commodity/agricultural indices as relevant.
  - Fiscal sector: revenue, expenditure, balance, and financing with breakdowns (debt holder, instrument, currency) as relevant.
  - Financial sector: net external position, domestic credit, broad or narrow money (replacing some previous comprehensive framework elements).
  - External sector / Balance of payments: focus on imports and exports of goods and services, current account balance, reserves, overall balance.
  - International Investment Position: gross official reserve assets expressed in U.S. dollars; reserve related liabilities (encouraged); reserves template (encouraged).
  - External debt: public and publicly guaranteed external debt broken down by maturity; public and publicly guaranteed external debt service schedule; private external debt not publicly guaranteed (encouraged).

### Strengthening plans for improvement
- Continue developmental role: plans for improvement remain an important feature of the GDDS.
- Metadata burden and update frequency:
  - Streamlining data categories reduces the metadata burden.
  - Proposal to drop the requirement for annual updates of metadata and permit updating on a “best effort” or “when merited” basis.
- Expectations for plans for improvement:
  - Retain expectation that plans be updated regularly and be comprehensive.
  - Assign higher priority and greater prominence to periodicity and timeliness objectives within these plans.
  - More comprehensive and timely plans could form basis for interagency coordination of technical assistance and training.
  - More focused plans could emphasize results-based management and aid policymakers in tracking priorities and negotiating with donors.

### Strengthening the emphasis on graduation to the SDDS
- Streamlining and realignment proposals, introduction of the NSDP, and renewed emphasis on the ARC are expected to make progression to the SDDS easier and more natural.
- Consideration: adopt eventual progression to the SDDS as a presumptive end-goal for GDDS participants.
- Preserving voluntary nature: countries remain free to choose whether or not to “sign up” to the SDDS goal.

### Achieving better regional balance
- Strategic goals:
  - Increase participation in Middle Eastern countries.
  - Intensify work in the African region with countries interested in moving to the SDDS.
- Rationale:
  - Regional participation has increased unevenly; with currently only one SDDS subscriber in Africa (South Africa), increasing SDDS subscribers in Africa could generate “success stories” to spur statistical capacity-building.

### Costs and benefits of the proposed changes to the GDDS
- IMF (STA) cost implications:
  - Strengthening data dissemination is not foreseen to add additional cost to STA but will require a reformulation of STA’s technical assistance priorities toward more data transparency.
  - Many GDDS participants likely require technical assistance to establish a centralized NSDP and an ARC; assistance envisioned within existing technical assistance program.
  - Data dissemination issues would be added to terms of reference of technical assistance missions; some regional outreach missions or regional workshops would be conducted.
- Participant cost estimates and operational impacts:
  - Initial startup costs to set up the NSDP and ARC and ongoing maintenance costs will arise for GDDS participants.
  - Implementation one-time cost best estimates: between 4–12 person weeks, involving mainly technological (IT) skills and some efforts by the GDDS coordinator to compile data from various agencies.
  - Ongoing maintenance: GDDS coordinator might use about 10–15 percent of his/her time to update the NSDP and ARC.
  - Benefits: central data page with timely data for policymakers; potential savings if NSDP/ARC replace paper-based publications, which could offset or exceed additional costs.
- Metadata and plans workload:
  - Streamlined metadata reduces time spent updating metadata.
  - Emphasis on up-to-date plans for improvement increases effort required, but these plans could enable better results-based management and donor negotiations.

*Source: Excerpt from section VII–VIII of the provided GDDS assessment document.*

### 92.      The options and directions that have been proposed in this review would—if

### _gdds-assess-08 - 92.      The options and directions that have been proposed in this review would—if

### Proposed changes and consultation process
- Staff view: the proposed options and directions would—if implemented—result in important changes with implications for current and prospective GDDS participants, but would not result in a fundamental change in the objectives of the GDDS.
- Staff judgment: the changes go far beyond previous reviews of the Data Standards Initiative and should be considered in the context of a consultative process involving Fund staff, Executive Directors, and participating countries.
- Consultation sequencing described:
  - An informal seminar provides staff the first opportunity to seek Executive Directors’ views.
  - If Executive Directors support the broad thrust of analysis and recommendations, staff would next consult more broadly with the membership.
  - Formal recommendations would be brought to the Board at the time of the Seventh Review of the Data Standards, presently scheduled for the fall of 2008.
- Transition: consultations would include an appropriate procedure for transition, as was the case when the data standards were first formulated and adopted.

### Questions and matters for Executive Directors’ views
- Recasting the GDDS to emphasize data dissemination
  - Do Directors agree with staff’s overarching proposal to reorganize the GDDS to place more emphasis on data dissemination to the public?
  - Should plans for improvement emphasize periodicity and timeliness of data?
- Strengthening the data dissemination aspect of the GDDS
  - Do Directors agree that GDDS participants should be expected to establish a centralized NSDP and an ARC, as in the SDDS?
  - Do Directors agree that participants that have developed ARCs and NSDPs should post hyperlinks to these websites on the DSBB?
- Realigning the GDDS data categories
  - Do Directors support simplifying the structure of GDDS data by streamlining the data components of the comprehensive frameworks and combining the comprehensive frameworks with some of the data categories (i.e., to simplify and reformulate the data categories to more closely approximate that of the SDDS and to eliminate the somewhat complicated structure)?
- Strengthening plans for improvement
  - Do Directors support the staff’s view that a higher priority should be assigned to the plans for improvement in the metadata and less to updating the description of the existing system in the metadata?
- Strengthening emphasis on GDDS participants’ graduation to SDDS
  - Do Directors agree that emphasis on graduation to SDDS should be strengthened?
- Achieving better regional balance
  - Do Directors agree with the staff’s suggestion to continue with its efforts to obtain a better regional balance in the participation in the GDDS?

### Selection criteria and sample of GDDS participants (Appendix I)
- Frame and grouping:
  - Total GDDS participants (as of February 2007): 88
  - Grouping by IMF area departments: African (AFR), Asia and Pacific (APD), Middle East and Central Asia (MCD), Western Hemisphere (WHD). Two European countries (Albania and Macedonia, FYR) included in MCD.
- Exclusions and sample size:
  - 12 countries that had not updated metadata in the last four years (48 months) were excluded.
  - Metadata updated within 48 months: 76
  - Final group chosen: 55 countries (about 70 percent of the frame).
- Regional distribution of the 55-country group:
  - Africa: 22
  - Asia and Pacific: 8
  - Middle East and Central Asia: 12
  - Western Hemisphere: 13
- Appendix Table I.1 (selection of GDDS participants drawn by region) — key rows transcribed exactly as in source:
  - Total GDDS participants  39.0 12.0 17.0 20.0  88
    - Percent in total 44.3 13.6  19.3 22.7 100
  - Metadata updated within 48 months               29.0             12.0             17.0             18.0                           76
    - Percent in total 38.2 15.8 22.4 23.7 100
  - GDDS Participants in the group 
    22.0  8.0   12.0** 13.0  55
    - Percent in total 40.0 14.5 21.8 23.6 100
  - *   The sum of percent shares may not exactly equal to 100 percent due to rounding.
  - **  Includes Albania and Macedonia.

- Appendix Table I.2: list of the 55 GDDS participants by region (presented in the source). Example entries include:
  - AFR: Angola; Botswana; Central African Republic; Congo, Dem Rep. of; Congo, Rep. of; Ethiopia; Gambia, The; Kenya; Liberia; Madagascar; Malawi; Mauritius; Mozambique; Namibia; Nigeria; Rwanda; Senegal; Seychelles; Sierra Leone; Tanzania; Uganda; Zambia.
  - APD: Bangladesh; Cambodia; China; Kiribati; Mongolia; Nepal; Pakistan; Sri Lanka; Vietnam; Kiribati and others as listed.
  - MCD: Albania; Afghanistan; Azerbaijan; Georgia; Jordan; Kuwait; Macedonia, FYR; Oman; Qatar; Tajikistan; West Bank and Gaza; and others as listed.
  - WHD: Antigua and Barbuda; Belize; Bolivia; Dominica; Grenada; Guatemala; Honduras; Nicaragua; Panama; St. Kitts and Nevis; St. Lucia; Trinidad and Tobago; Venezuela.

### GDDS and SDDS coordinators (Appendix II)
- Participation requirement:
  - Participation in both the GDDS and the SDDS requires that a country appoint a national coordinator.
  - GDDS participation constitutes a commitment to updating metadata and plans for improvement at least once a year.
- Roles and responsibilities:
  - GDDS coordinators are responsible for updating metadata and can play an important role in advancing the reform agenda; their role is more strategic.
  - SDDS coordinators have more day-to-day responsibilities because data are posted on the website on an ongoing basis, and metadata are certified every quarter; the SDDS coordinator role is more technical and less strategic.
- Coordinator profiles (summary findings):
  - Senior-level managers (head or deputy head of agency) account for 56 percent of total GDDS coordinators.
  - Coordinators from the statistical office account for 52 percent of total GDDS coordinators.
  - Senior-level managers are predominant in the African, Middle East and Central Asian, and Western Hemisphere regions; different ranks are appointed more evenly in European and Asia and Pacific regions.
  - In contrast, SDDS countries (advanced economies) appoint mainly mid-level managers (head or deputy head of departments or divisions) in central banks and statistics offices.
- Appendix Table II.1 and II.2: coordinator-rank tables presented in the source (percent distributions by region and by institution). Selected transcriptions from the source:
  - Appendix Table II.1. Ranks of GDDS Country Coordinators (In percent of total number of GDDS coordinators as of February 2007)
    - Total number of GDDS coordinators 10056 25 19
    - By department 
      - AFR 4431   8 6
      - APD 14  5   5 5
      - EUR 3  1   2 0
      - MCD 16  9   6 1
      - WHD 2310   5 8
    - By institution 
      - Central bank 23  8 13 2
      - Ministry of finance 2513   5 8
      - Statistics office 5235   8 9
    - Source: Country authorities and Fund staff.
    - 1/ Senior management refers to head or deputy head of agency.
    - 2/ Middle management refers to head and deputy head of departments or divisions.
  - Appendix Table II.2. Ranks of SDDS Country Coordinators (In percent of total number of SDDS coordinators as of February 2007)
    - Total number of SDDS coordinators 10028 34 38
    - By department 
      - AFR 2  0   0 2
      - APD 16  5   8 3
      - EUR                                                                                    5516                                                                                    17                                                                                    22
      - MCD 9  3   5 2
      - WHD 19  5   5 9
    - By Institution 
      - Central bank 44  8 16 20
      - Ministry of finance 16  6   5 5
      - Statistics office 4114 14 13
    - Source: Country authorities and Fund staff.
    - 1/ Senior management refers to head or deputy head of agency.
    - 2/ Middle management refers to head and deputy head of departments or divisions.

### Statistical capacity indicators (Appendix III excerpt)
- The source presents country-level statistical capacity scores (scale 0–100) and percentage changes for selected countries and regional averages. Selected transcriptions of entries and averages exactly as presented in the source:
  - AFR entries (examples): 
    - Senegal 70 7575757.10.0
    - Uganda 52 60677340.421.7
    - Mozambique 62 6368689.77.9
    - Madagascar 62 6253631.61.6 X
    - Malawi 52 67606321.2-6.0
    - Mauritius 60 6363635.00.0
    - Kenya 65 655362-4.6-4.6
    - Tanzania 65 656562-4.6-4.6
    - Ethiopia 58 6363603.4-4.8
    - Rwanda 43 53536039.513.2
    - Gambia 38 60535339.5-11.7
    - Nigeria 53 405252-1.930.0
    - Republic of Congo 25 404050100.025.0
    - Namibia 50 5352500.0-5.7
    - Botswana 53 655847-11.3-27.7 X
    - Sierra Leone 22 273747113.674.1
    - Democratic Republic of Congo 42 3838432.413.2
    - Central African Republic 40 403838-5.0-5.0
    - Angola 27 33373529.66.1 X
    - Liberia 13 17171838.55.9
    - AFR average 47.6 52.552.154.113.73.1
  - APD entries (examples):
    - Bangladesh 60 73788033.39.6
    - Mongolia 60 70808033.314.3
    - Nepal 57 65737735.118.5
    - Vietnam 50 75757550.00.0
    - Sri Lanka 55 78727230.9-7.7
    - Cambodia 32 586365103.112.1
    - China 65 636562-4.6-1.6
    - APD average 54.1 68.972.37334.86.0
  - MCD entries (examples):
    - Albania 63 80808331.73.8
    - Pakistan 63 73808027.09.6 X
    - Azerbaijan 50 75777754.02.7
    - Macedonia 67 73777511.92.7
    - Tajikistan 45 63727566.719.0
    - Georgia 50 72737346.01.4
    - Jordan 77 737773-5.20.0
    - Afghanistan 10 152528180.086.7
    - MCD average 53.1 65.570.170.532.77.6
  - WHD entries (examples):
    - Guatemala 43 83808086.0-3.6
    - Bolivia 63 68707722.213.2
    - Venezuela 58 75777732.82.7
    - Nicaragua 52 82787544.2-8.5
    - Panama 58 75757529.30.0
    - Trinidad 58 58677020.720.7 X
    - Honduras 60 6255658.34.8
    - Dominica 60 6870635.0-7.4
    - WHD average 56.5 71.471.572.828.81.9
  - Total line as presented in source:
    - Total                                                                   51.3                                                                   61.162.363.724.04.3                                                                   5
  - Source attribution line in table: Source: World Bank.
  - Note in table: 1/ Scale of 0–100. A score of 100 indicates that a country meets all the criteria.

*Italic: Extracted from the IMF chapter content provided in the supplied content unit.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdds/_gdds-assess-08.pdf_