Sixth Review of the Fund's Data Standards Initiatives

The Data Standards Initiatives, the SDDS and the GDDS, have achieved the goals the Executive Board set in its Fifth Review of July 2003. The staff sees the next three years as a period of consolidating these gains by maintaining the credibility of the SDDS through improved monitoring of countries’ observance of its requirements, and further integrating both the SDDS and GDDS under the Fund’s Data Quality Program (DQP) by aligning their structure with the Fund’s Data Quality Assessment Framework (DQAF). The staff proposes to include no new data categories in the SDDS and GDDS. Instead, the staff proposes to deepen descriptive information on how countries cover oil and gas activities and products in selected existing data categories.

Abstract

The Data Standards Initiatives, the SDDS and the GDDS, have achieved the goals the Executive Board set in its Fifth Review of July 2003. The staff sees the next three years as a period of consolidating these gains by maintaining the credibility of the SDDS through improved monitoring of countries’ observance of its requirements, and further integrating both the SDDS and GDDS under the Fund’s Data Quality Program (DQP) by aligning their structure with the Fund’s Data Quality Assessment Framework (DQAF). The staff proposes to include no new data categories in the SDDS and GDDS. Instead, the staff proposes to deepen descriptive information on how countries cover oil and gas activities and products in selected existing data categories.

Executive Summary

The success of the Fund’s Data Standards Initiatives is manifest as their decennial approaches in 2006. The Special Data Dissemination Standard (SDDS) has helped subscribing countries improve data transparency. Further evidence has accumulated since the Fifth Review of the Data Standards Initiatives on the positive connection between data transparency and increased access to capital markets.[k1] The General Data Dissemination System (GDDS) has assisted participants to build statistical capacity. As endorsed by the Board in the Fifth Review, the Fund staff has continued its outreach effort, integrated with technical assistance, to promote SDDS subscription or GDDS participation among Fund member countries. As of May 2005, 139 Fund member countries (75 percent of the membership) have become either SDDS subscribers or GDDS participants, up from 111 (or 60 percent of the membership) since the Fifth Review. There were 61 SDDS subscribers and 78 GDDS participants.

The Data Standards Initiatives, the SDDS and the GDDS, have achieved the goals the Executive Board set in its Fifth Review of July 2003. The staff sees the next three years as a period of consolidating these gains by maintaining the credibility of the SDDS through improved monitoring of countries’ observance of its requirements, and further integrating both the SDDS and GDDS under the Fund’s Data Quality Program (DQP) by aligning their structure with the Fund’s Data Quality Assessment Framework (DQAF). The staff proposes to include no new data categories in the SDDS and GDDS. Instead, the staff proposes to deepen descriptive information on how countries cover oil and gas activities and products in selected existing data categories.

The actions Executive Directors endorsed in the Fifth Review of the Data Standards Initiatives (Public Information Notice No. 03/86, July 23, 2003 (PIN 03/86)), as well as developments on each since the Fifth Review, were:

  • Sustaining further increases in the rates of participation in the Data Standards Initiatives and members graduating from the GDDS to the SDDS. The staff’s integrated outreach and technical assistance effort has been an important contributor to improvement in data transparency by promoting SDDS subscription and GDDS participation. Since the Fifth Review in 2003, 8 countries have become SDDS subscribers and 24 have become GDDS participants. Four countries have graduated from GDDS participation to SDDS subscription as their statistical capacity has improved.

  • Completing implementation of the external debt data category in the SDDS.Since end-September 2004 all but two subscribers have provided the full SDDS prescribed detail by sector and by type of instrument on their National Summary Data Pages (NSDPs), and some detail was provided in each of these two cases. The dissemination of external debt data was facilitated by a series of five IMF regional training seminars directed at mid-level compilers from 76 countries held during May 2002 through July 2004. All but one SDDS subscriber and 67 of 78 GDDS participants disseminate descriptions of methods and practices (metadata) for external debt statistics as provided under the Third Review of the Data Standards Initiatives of March 2000.

  • Enhancing the Fund’s Dissemination Standards Bulletin Board (DSBB) and the exchange of statistical information under the Statistical Data and Metadata eXchange (SDMX) Initiative. The DSBB was redesigned in 2003 and now offers a dynamic querying capability into the metadata it contains. A user survey conducted on the DSBB during May 31, 2003 through January 7, 2004 showed respondents find that the new features of the DSBB make it a much more useful portal to information on countries’ statistical practices. Several labor-intensive functions supporting the DSBB, such as monitoring SDDS observance and GDDS metadata maintenance also have been partially automated. The SDMX consortium issued the first version of its data exchange standards and work advanced on a metadata exchange standard compatible with the DQAF.1

  • Adding a targeted timeliness flexibility option to the SDDS for data on central government operations (CGO). Under this option, if a subscriber disseminates with a one-quarter lag quarterly general government operations (GGO) data, in line with the Government Finance Statistics Manual 2001 (GFSM 2001) or an equivalent standard, the SDDS allows flexibility in the timeliness of monthly central government operations (CGO) data (whether on a cash, accrual, or modified cash/accrual basis). Since the Fifth Review, Australia has availed itself of this option.

  • Suggesting that SDDS subscribers take advantage of the possibility to provide—on a strictly voluntary basis—additional metadata on their practices regarding core inflation, forward-looking indicators, and interest rates used as operating targets. To inform subscribers of this possibility, the staff will include discussion of core inflation and interest rates as operating targets for financial policy in the “forward-looking indicators” section of the revised SDDS Guide (forthcoming 2005).

  • Advancing the role of the GDDS as a catalyst in the mobilization and coordination of technical assistance by the Fund, the World Bank, and other donors. The National Strategies for Development of Statistics Framework of the Partnership for Development of Statistics in the 21st Century (PARIS21) has incorporated GDDS into its core structure. PARIS21 is a major consortium of bilateral and multilateral technical assistance providers, including the Fund. The GDDS thus has helped bilateral and international providers, including the Fund’s regional technical assistance centers, coordinate delivery of statistical technical assistance. Also noted is the pilot integration of the GDDS into the Poverty Reduction Strategy Paper (PRSP) for Sierra Leone, encouraging national public expenditure planning to take account of recommendations from statistical technical assistance.

  • Incorporating the Millennium Development Goal (MDG) indicators into the GDDS. Directors noted that the incorporation of these indicators within the GDDS structure would not require major changes and would be helpful in closely collaborating with the World Bank as well as other institutions and donors. In October 2004 Fund staff, in consultation with the World Bank, incorporated selected Millennium Development Goals (MDGs) and associated indicators into the GDDS Guide 2004, which was posted on the DSBB in November 2004. Five[k2] GDDS countries now provide the metadata for these indicators, the bulk of which are in the GDDS socio-demographic data sector for which the World Bank provides statistical technical assistance to participating countries.

  • Balancing the preparation of “new” data modules of the Reports on the Observance of Standards and Codes (Data ROSCs) with follow-up ROSCs.[k3] Developments in this area will be covered in a separate Board paper on the Review of the Standards and Codes Initiatives (forthcoming).

  • Continuing implementation of the Data Quality Program. Since the Fifth Review, the staff has begun parsing the existing metadata on the DSBB into detailed DQAF topics or “granules.” Once completed, the newly “granularized” DSBB metadata can be linked to other DQAF-organized metadata bases such as the Data ROSCs and the Fund’s Project Management System for statistical technical assistance, integrating and streamlining the Fund Statistics Department’s (STA’s) surveillance and technical assistance work with the Fund’s data transparency initiatives.

The Public Information Notice (PIN 03/86) for the Fifth Review also noted Directors would have a future opportunity to discuss possible inclusion of Financial Soundness Indicators (FSIs) into the SDDS based on further experience with making FSIs operational. To accumulate this experience, the staff has initiated an FSI coordinated compilation exercise (CCE), which will yield results in 2007.

The staff proposes to take the following steps to advance countries’ data dissemination practices in the near and medium term:

  • Upholding the credibility of the SDDS as a monitored standard to sustain its benefits to subscribing countries, international capital markets, and the public. Evidence has accumulated that the SDDS has improved subscribers’ access to international capital markets by reducing borrowing costs. The continued benefits of the SDDS to subscribers and the public depend on its credibility which, in turn, rests on a culture of observance with its provisions among subscribers. To encourage and sustain this culture, the staff proposes to further standardize and automate the channels through which SDDS subscribers report information for posting on the DSBB. This will reduce processing lags and streamline staff monitoring of countries’ observance of the SDDS, facilitating the staff to prepare annual reports assessing subscribing countries’ observance of their SDDS undertakings and post them on the DSBB. STA and the relevant area departments would continue to work together to include serious SDDS observance issues on the agenda for discussion with senior country officials.

  • Strengthening the role of the GDDS in building countries’ statistical capacity. The staff proposes to encourage countries to use the “plans for improvement” integral to the GDDS to manage the development of their statistical systems and to coordinate donor technical assistance and explore whether and how to integrate the GDDS into the Poverty Reduction Strategy Paper (PRSP).

  • Assisting countries with a strong commitment to data transparency. Resource constraints and increased demand for support from an enlarged number of SDDS subscribers and GDDS participants characterize the medium-term environment for STA. Of the remaining 46 Fund member countries that are not SDDS subscribers or GDDS participants, five are working toward SDDS subscription and 18 are working toward participation in the GDDS. The rest of this group (23 countries) have not indicated plans to join in the SDDS or the GDDS. The staff believes official commitment to data transparency should be an important factor in setting technical assistance priorities. Countries desiring participation in the GDDS and GDDS participants and other Fund member countries desiring to become SDDS subscribers demonstrate this commitment.

  • Providing guidelines deepening the metadata to address coverage of oil and gas activities and products for selected data categories of the SDDS and GDDS. This would help to shed light on the measurement of oil and gas activities in relevant SDDS/GDDS macroeconomic datasets of both oil exporting and oil importing member countries. The relevant data categories in the SDDS and GDDS are (1) national accounts, (2) production index, (3) merchandise trade, (4) balance of payments, (5) central and general government operations, (6) consumer price index, and (7) producer price index. The staff proposes to work with SDDS subscribers and GDDS participants to provide this information on an “encouraged” basis. “Encouraged” coverage is desirable and may be strongly promoted by the staff, but is not a requirement for SDDS subscription or GDDS participation.

  • Advancing the Fund’s Data Quality Program. An important component of this work is to align countries’ SDDS/GDDS metadata with the DQAF, allowing the Fund’s SDDS and GDDS, the Data ROSC, and statistical technical assistance work to more closely complement one another. This is being undertaken while imposing no additional reporting burden on SDDS subscribers and GDDS participants.[k4] It will strengthen the role of the GDDS in guiding member countries’ development of quality statistics. The integration of the DQAF into the SDDS and GDDS also complements the Fund’s participation, along with other international organizations, in the SDMX consortium. A supplement to this Board paper, titled “Metadata Standardization in the Data Quality Program,” describes the proposed way forward.

To undertake these initiatives, the staff asks Executive Directors’ support for

  • Making subscriber observance of automated monitoring arrangements an SDDS undertaking to secure needed increases in efficiency given the increasing number of subscribers.

  • Making progress toward participation in the data transparency standards and advancing from GDDS to SDDS important criteria in prioritizing technical assistance.

  • Working with SDDS subscribers and GDDS participants to provide on an “encouraged” basis deepened metadata for oil and gas activities and products.

  • Examining at the next review whether to include selected Financial Soundness Indicators in the SDDS based on experience with the coordinated compilation exercise.

  • Conducting the Seventh Review of the Fund’s Data Standards Initiatives in the second half of 2008.

I. Introduction

1. The Fund’s work on data dissemination standards began in October 1995, in the aftermath of the Mexican financial crisis. The Interim Committee (now the International Monetary and Financial Committee) endorsed the establishment by the Fund of standards to guide members in the dissemination to the public of their economic and financial data. The standards were to enhance the availability of timely and comprehensive statistics, thereby contributing to the pursuit of sound macroeconomic polices, as well as the improved functioning of financial markets.2 The standards were to consist of two tiers: the Special Data Dissemination Standard (SDDS) to guide countries that have, or that might seek, access to international capital markets; and the General Data Dissemination System (GDDS) to assist all other member countries. The SDDS was approved by the IMF Executive Board in March 1996, and the GDDS in December 1997. (Background information on the establishment of the SDDS and the GDDS is provided in Appendix I.)

2. The SDDS is a monitored standard focusing on dissemination of data used by financial markets. It requires subscribing countries to observe the standard and to provide certain information about their practices (metadata) in disseminating economic and financial data. Subscribers must agree to (1) post information about their data dissemination practices on the Fund’s external website on an electronic bulletin board known as the Dissemination Standards Bulletin Board (DSBB) and (2) establish an internet site containing the actual data called a National Summary Data Page (NSDP) which is linked to the DSBB.

3. The GDDS is a framework to guide countries to develop sound statistical systems as the basis for dissemination of data to the public. The GDDS calls for participating countries to prepare metadata on their current statistical practices, develop their plans for improvement in the short and medium term, and identify associated needs for assistance in implementing these plans. Participating countries must update (at least annually) and revise their metadata to describe how their data compilation and dissemination activities compare with the best statistical practices as set forth in the GDDS.

4. The success of the Fund’s Data Standards Initiatives is manifest as their decennial approaches in 2006. The SDDS has helped subscribing countries improve data transparency. Further evidence has accumulated since the Fifth Review of the Data Standards Initiatives on the connection between data transparency and increased access to capital markets.[k5] The GDDS has assisted participants to build statistical capacity. Appendix II details these broad developments. As endorsed by the Board in the Fifth Review, the staff has continued its outreach effort, integrated with technical assistance, to promote SDDS subscription or GDDS participation among Fund member countries. As of May 2005, 139 Fund member countries (75 percent of the membership) have become either SDDS subscribers or GDDS participants,3 up from 111 (or 60 percent of membership) since the Fifth Review.

5. Since their inceptions, the SDDS and the GDDS have been reviewed by the Executive Board on a regular basis. Directors reviewed a progress report on the implementation of the SDDS in March 1996 and undertook a Second Review of the SDDS in December 1998. These were followed by Directors’Third Review of the Fund’s Data Standards Initiatives in March 2000, Fourth Review of the Fund’s Data Standards Initiatives in July 2001, and Fifth Review of the Fund’s Data Standards Initiatives in July 2003. This paper is prepared for Directors’ Sixth review of the Fund’s Data Standards Initiatives.

6. Directors endorsed the following initiatives in the Fifth Review (PIN 03/86):

  • Sustaining further increases in the rates of participation in the data standards initiatives and members graduating from the GDDS to the SDDS.

  • Completing implementation of the external debt data category in the SDDS.

  • Balancing the preparation of “new” data modules of the Reports on the Observance of Standards and Codes (data ROSCs) with follow-up data ROSCs.4

  • Enhancing the Fund’s DSBB and the exchange of statistical information under the Statistical Data and Metadata eXchange (SDMX) Initiative.

  • Adding a targeted timeliness flexibility option to the SDDS for data on central government operations (CGO).

  • Suggesting that SDDS subscribers take advantage of the possibility to provide—on a strictly voluntary basis—additional metadata on their practices regarding core inflation, forward-looking indicators, and interest rates used as operating targets.

  • Advancing the role of the GDDS as a catalyst in the mobilization and coordination of technical assistance by the Fund, the World Bank, and other donors.

  • Incorporating the Millennium Development Goal (MDG) indicators into the GDDS.

  • Implementing the Data Quality Program.

7. The rest of this paper is organized in five sections. Section II reports on developments in the data standards initiatives since the Fifth Review; Section III presents current and proposed initiatives; Section IV discusses future developments; Section V deals with the resource implications of current and proposed initiatives; and Section VI presents issues for Board discussion. The paper also contains four appendices. Appendix I provides background on the history of the SDDS and GDDS; Appendix II provides an overview of the role of the SDDS in enhancing data transparency and market access and the role of the GDDS in building statistical capacity; Appendices III and IV report on projects relating to Fund data standards initiatives.

II. Developments in The Data Standards Initiatives Since the Fifth Review

A. Expanding SDDS Subscription and GDDS Participation

SDDS subscription and GDDS participation

8. As noted earlier, three-fourths of the Fund membership have become either SDDS subscribers or GDDS participants. Figure 1 traces the growth in SDDS subscription and GDDS participation since 1996. SDDS subscription now stands at 61 countries, up eight from 53 at the Fifth Review. The rate of GDDS subscription thus far has exceeded the staff’s projection at the time of the Third Review of about 16 participants a year. To date, 83 countries have participated in the GDDS, five of which have graduated to SDDS subscription. Four of the five have graduated since the Fifth Review. Of the 24 countries that have become GDDS participants since the Fifth Review, 11 are from Africa, four from Asia and the Pacific, four from the Middle East and Central Asia, four from the Western Hemisphere, and one from Europe.

Figure 1.
Figure 1.

Number of SDDS Subscribers and GDDS Participants

Citation: Policy Papers 2005, 010; 10.5089/9781498331456.007.A001

9. Currently, 46 Fund member countries are neither SDDS subscribers nor GDDS participants. Figure 2 and Table 1 show the geographical distribution of these countries. The staff expects that about 18 additional member countries and one nonmember will become GDDS participants in the near and medium term.5 Of the remaining nonparticipating/ nonsubscribing countries, some do not disclose certain key economic data as a matter of policy. Others are constrained by their limited statistical capacity and need substantial technical assistance. These factors could contribute to an eventual slowing of the rates of subscription to the SDDS and of participation in the GDDS than experienced in the past. As shown in Table 1, however, most members not yet participating in the data transparency standards are working toward participation in the GDDS or subscription to the SDDS.[k6]

Figure 2.
Figure 2.

SDDS Subscribers and GDDS Participants in the Fund Membership by Region

Citation: Policy Papers 2005, 010; 10.5089/9781498331456.007.A001

Table 1.

IMF Member Countries Not GDDS Participants or SDDS Subscribers

As of End-May 2005

article image

Working toward GDDS participation.

Working toward SDDS subscription.

West Bank and Gaza, although not an IMF member country, also is working to participate in the GDDS.

Assisting countries with a strong commitment to data transparency

10. Since the Fifth Review, eight countries, all part of the SDDS outreach program,6 have subscribed to the SDDS. These are Armenia, Belarus, Bulgaria, Egypt, the Kyrgyz Republic, Romania, Russia, and Uruguay.7 Table 2 shows the current group of countries considered likely candidates for SDDS subscription within two years. The SDDS outreach program has been conducted through regional seminars and close collaboration between STA staff and the designated SDDS/GDDS country coordinators. Following the Fifth Review, SDDS outreach seminars were held in Uruguay in April 2004, South Africa in February 2005, and Thailand in April 2005. These seminars have helped countries assess their data dissemination practices relative to the SDDS and provided guidance on how they can meet SDDS requirements. These seminars also have provided information on the roles and responsibilities of country coordinators for the SDDS and other operational aspects of the standard.

Table 2.

Member Countries in the Program to Promote SDDS Subscription

As of End-May 2005

article image

11. For some countries in the set judged capable of meeting SDDS requirements in the near to medium term, national dissemination policy is as much a factor as statistical capacity in progress toward subscription. Eighteen of the group in Table 2 are participants in the GDDS and thus have made a significant commitment to data transparency. Five are working toward SDDS subscription. Of this group, the staff expects the rate of graduation of the 18 current GDDS participants to the SDDS to continue at a steady pace. The speed with which this group meets SDDS requirements depends significantly on technical assistance as well as official commitment to the more stringent transparency requirements of the SDDS. The Fund is able to supply technical assistance only in the statistical areas it traditionally covers (national accounts, price, external, fiscal, and financial statistics) and only in limited amounts, given the significant and competing capacity building needs for member countries at lower stages of statistical development. Progress toward meeting SDDS requirements in labor market statistics depends entirely on other donors, for example. Even with adequate technical assistance, data on quarterly national accounts, monthly central government operations and financing, and quarterly general government operations generally take time and commitment of counterpart resources to develop. Building and improving critical national accounts statistics rely on the development of source data such as establishment, labor market, and household surveys. The Fund has considered these areas beyond the scope of its own statistical technical assistance (TA). They require mobilization of, and coordination with, other TA providers. [k8]

12. The monthly reserves template, like the development of quarterly national accounts and labor market statistics, is among the most challenging requirements for the current group of prospective subscribers. As a detailed and high frequency component of the SDDS whose strict timeliness requirements are closely monitored, the reserves template requires a strong and sustained commitment to data transparency. This commitment is reflected both in disseminating the template data and in developing the capacity to produce the information called for in the template. However, it also is this challenge to existing and potential subscribers alike that makes dissemination of the reserves template a marker of statistical achievement. Countries also find the SDDS requirements for the external debt data category add to the challenge presented by the SDDS subscription process.

The SDDS Guide 2005 and the GDDS Guide 2004

13. The SDDS Guide 2005 will replace the 1996 provisional Guide to the Data Dissemination Standards (Module 1: The Special Data Dissemination Standard). The new Guide will support countries working toward SDDS subscription and guide existing subscribers in maintaining their observance of the SDDS. It will reflect the evolution of the SDDS over the past nine years, including the addition of the external debt and the reserves template data categories, and the prescribed dissemination of the international investment position. The format of the SDDS Guide 2005 will be more user-friendly and its coverage of operational issues, such as the provision of release dates for the advance release calendars (ARCs), will be expanded. The SDDS Guide 2005 also will provide detailed information on the subscription procedure to the SDDS, specify the duties and responsibilities of the national SDDS coordinator, and describe the procedure used by the Fund staff to monitor subscribers’ observance of the SDDS. The discussion of flexibility options has also been updated to include the targeted timeliness flexibility options for central government operations, which was approved during the Fifth Review. The staff expects publication of the Guide in the second half of 2005.

14. The GDDS Guide 2004, which superseded the version updated in March 2002, was posted on the DSBB in October 2004. The new GDDS Guide has benefited from the suggestions and recommendations of member countries that have been involved in implementing the GDDS. The GDDS Guide 2004 is expected to continue to evolve as the GDDS evolves, and will continue to benefit from the comments and suggestions of those who use it as a tool for participation in the GDDS. As examples of the evolution of the GDDS, the GDDS Guide 2004 incorporates decisions taken by the Executive Board to add the external debt data category (March 2000), to articulate the socio-demographic component of the GDDS (November 2001), and to incorporate the Millennium Development Goals (MDGs) as specific elements of the socio-demographic component (November 2003).8 In this regard, it is important to highlight the valuable contribution of the World Bank in articulating the socio-demographic component of the GDDS. A future version of the GDDS Guide will include the integration of the Fund’s Data Quality Assessment Framework (DQAF) into the GDDS metadata model.

B. New Data Category—External Debt

15. SDDS subscribers have made significant progress in disseminating external debt statistics under the SDDS framework. The first dissemination for the prescribed series9 of the new quarterly external debt data category was required by end-September 2003; and by end-December 2003, 53 of the 55 SDDS subscribers disseminated the data on their National Summary Data Pages (NSDPs). The two subscribers not providing the full SDDS prescribed detail by sector and by type of instrument nevertheless provided some detail. In addition to the prescribed detail, over one quarter of subscribers also publish some of the encouraged components. The dissemination of the external debt data category was greatly facilitated by a series of five IMF regional training seminars directed at mid-level compilers from 76 countries held during May 2002 through July 2004. (See also Box 1.)

External Debt

Under the Third Review of the Fund’s Data Standards Initiatives the IMF’s Executive Board, in March 2000, approved the introduction of the new external debt data category to provide a comprehensive and consistent measure of a country’s gross external debt. The category prescribes dissemination of quarterly external debt statistics covering four domestic sectors (the general government, monetary authorities, banking sector, and other sectors) with a time lag of one quarter after the end of the reference period. Data are broken down by maturity (short term and long term) on an original maturity basis and by instrument as set out in the Balance of Payments Manual, FIFTH EDITION (BPM5). Countries are also encouraged to disseminate a domestic-foreign currency breakdown of external debt, as well as supplementary information on prospective debt service obligations, in which principal and interest are separately identified.

In light of the results of staff consultations with data compilers and data users, and taking into account the importance of the new SDDS data category, the Executive Board introduced a three-year transition period ending on March 31, 2003. It meant that quarterly data for end-June 2003 had to be disseminated by September 30, 2003. No flexibility options on periodicity or timeliness were permitted for the external debt data category. However, once subscribers met the quarterly periodicity and timeliness for disseminating external debt, they could take up to nine months―instead of six months―to disseminate their international investment positions.

The dissemination of external debt statistics was facilitated by the Fund’s technical assistance and training. In early 2002, the Statistics Department undertook a series of seven seminars targeted at managers and senior external debt compilers in SDDS subscribing countries. Since May 2002, five additional seminars directed at mid-level compilers also were organized. In June 2003, the Fund published External Debt Statistics: Guide for Compilers and Users (Debt Guide). It was prepared by the Interagency Task Force on Finance Statistics and provided a comprehensive conceptual framework that linked external debt statistics to the international investment position.

On September 30, 2003 85 percent of all subscribers disseminated external debt statistics in SDDS format on their NSDP. Subscribers gained further experience with the quarterly dissemination of the new data category during the last quarter of 2003. By the end of the year, 54 of 55 subscribers met the periodicity and timeliness requirements for this data category. By May 15, 2005, 59 of 61 subscribers have fully disseminated external debt statistics as prescribed by the SDDS (see table below). The staff is working with the remaining two subscribers to disseminate all of the required components.

article image

Interagency cooperation on external debt

16. Interagency cooperation on external debt statistics continues to be productive in promoting a common approach among international agencies to the dissemination of external debt statistics. The Interagency Task Force on Finance Statistics (TFFS)10 has served a critical role in coordinating the development of internationally comparable and widely disseminated external debt statistics. Following the joint preparation of the External Debt Statistics: Guide for Compilers and Users (Debt Guide), published by the Fund in 2003, TFFS members have continued to cooperate on a number of projects. These projects include the joint BIS-IMF-OECD-World Bank Statistics on External Debt, which presents selected external debt data from creditor sources; the recently launched World Bank Quarterly External Debt Statistics database (QEDS), the external debt pilot project under the Statistical Data and Metadata eXchange (SDMX initiative), and the development of a comprehensive public debt data reporting system, including a draft template for compiling public debt data from country authorities, consistent with the recommendations of the Debt Guide.

Quarterly external debt database

17. The World Bank, in cooperation with the IMF and in consultation with SDDS subscribers, launched the QEDS in November 2004. The database, a significant step in broadening the dissemination of external debt statistics, brings together in a central location the detailed quarterly external debt data published by SDDS countries. It facilitates both time-series analysis and cross-country data comparisons.11 Countries’ participation in the QEDS database is voluntary. Currently 54 SDDS subscribers have agreed to participate in this database, of which 49 have provided SDDS-prescribed data as of December 2004.

C. Enhancements to the Dissemination Standards Bulletin Board

18. The DSBB currently supports metadata dissemination for 61 SDDS subscribers, 78 GDDS participants, and the Data Quality Reference Site (DQRS). The DSBB is an important instrument in providing information to all users on the data dissemination practices of SDDS subscribers and GDDS participants. The DSBB was redesigned in 2003 and now offers a dynamic querying capability into the metadata it contains. Content management system (CMS) technology has been implemented to promote operational efficiency and accuracy in metadata maintenance. Automated templates have been developed for the processing of both ARCs for the SDDS and updates to metadata for the GDDS. These templates have facilitated the exchange of data and metadata between the Fund and member countries. Several labor-intensive functions, such as the monitoring of SDDS observance and GDDS metadata maintenance also have been partially automated.

DSBB user survey

19. A user survey was conducted on the DSBB during May 31, 2003 through January 7, 2004. A total of 136 responses were gathered. Overall, the survey results show that respondents find the features of the new version of the DSBB useful, especially in: (a) Accessibility—information is easier to find; (b) Output—information is easier to organize and/or print; (c) Navigation—it is easier to move about the DSBB; (d) Page loading—speed of viewing/downloading pages is faster; (e) Use of links—access to related sites is easier; and (f) Reliability—fewer error messages.

20. Most respondents valued the new DSBB website capabilities. The capabilities cited most frequently included (a) ability to compare information on coverage, periodicity, timeliness, prior access, etc., by data categories and countries; (b) ability to compare more detailed information on data categories by countries (such as consumer price indices’ reference periods); (c) ability to compare ARCs for selections of data categories and countries; (d) ability to compare summary of observance information by data categories and countries; (e) increased access to statistics on countries’ NSDP; (f) increased access to other countries’ statistics through the DSBB; and (g) increased access to other country-specific websites through the DSBB.

Statistical Data and Metadata Exchange Initiative

21. The SDMX initiative brings together seven international organizations, including the Fund, to foster greater efficiencies in data and metadata exchange. It has developed technical standards—the SDMX Version 1.0 Standards, an approved International Organization for Standards technical specification—that provide, inter alia, the specification for the exchange formats of SDMX-structured data and metadata. The SDMX initiative is developing a complementary set of “content standards” to harmonize the use of concepts and terminologies when exchanging statistical information. These standards are being implemented in a pilot project—the Joint External Debt Hub (JEDH)—that will be completed by end-2005 and serve as a proof of concept for the standards. Additional information on the SDMX initiative is available on its website (see http://www.SDMX.org).

22. The staff aims to gradually implement SDMX protocols within the Fund’s data standards initiatives. In addition to the Fund’s contributions to the JEDH, the staff is working to implement SDMX standards in metadata exchange. The objectives of this work are facilitating access to and repurposing of SDDS and GDDS metadata, and achieving efficiency gains in transmitting metadata among international organizations and their member countries.

D. Encouraging Dissemination of Fiscal Data Following GFSM 2001

23. The targeted timeliness flexibility option for central government operations (CGO)[kdz9] was envisaged as a means to promote international best practice in fiscal statistics. To support implementing the Government Finance Statistics Manual 2001 (GFSM 2001) and to pursue dissemination of quarterly general government operations (GGO) data, the Board approved at the Fifth Review adding a targeted timeliness flexibility option12 for data on CGO (whether on a cash, accrual, or modified cash/accrual basis). The option leaves SDDS requirements unchanged for those subscribers that are not yet ready to implement the accrual recording that is the hallmark of the GFSM 2001 standard. The staff believes that this option should remain in place for subscribers implementing the GFSM 2001 or equivalent accrual-based reporting systems for fiscal data. If a subscriber disseminates quarterly GGO data, with a one-quarter lag, in line with the GFSM 2001 or an equivalent standard, this flexibility option would be available for monthly CGO data. It allows up to three months dissemination lag (rather than one month) for the last month of the fiscal year and up to two months lag (rather than one month) for the first month of a new fiscal year. In order to use this flexibility option, a subscriber should begin disseminating quarterly GGO data for at least the last quarter of the fiscal year in which the option is exercised.13 (See also Appendix III.)

E. Inflation Targeting Indicators

24. The Board recommended at the Fifth Review that SDDS subscribers following inflation targeting monetary policy regimes have the option to provide additional metadata. The Board directed that provision of these metadata be strictly voluntary. The intent is to promote public knowledge and understanding of country practices with respect to measuring core inflation and the use of “forward-looking indicators” and interest rates as operating targets. The staff notes that, although subscribers have this option of reporting inflation targeting indicators under the SDDS’s “forward looking indicators” data category, none has so far done so. To increase awareness of this option among SDDS subscribers the staff will describe it in the revised SDDS Guide 2005.

F. The GDDS as a Catalyst in Mobilizing and Coordinating Technical Assistance

25. The Fifth Review directed incorporation of the MDGs into the metadata of the GDDS, which facilitates integration of the GDDS with the Poverty Reduction Strategy Paper (PRSP). The MDGs comprise a set of targets and indicators, classified under both the macroeconomic and socio-demographic data categories of the GDDS, that form the generally accepted framework for measuring development progress. Countries’ GDDS metadata now should make explicit reference to selected MDGs and associated indicators, including their periodicity and timeliness.14 As noted above, the revised GDDS Guide 2004 incorporating the MDGs was posted in October 2004. To date, five countries have shown MDG indicators in their GDDS metadata.15 Because most of the MDGs are under the socio-demographic component of the GDDS, the speed with which MDG indicators are incorporated into GDDS metadata thus far has depended mostly on the involvement of World Bank missions. It is also clear that some countries will need extensive guidance to produce these metadata. The World Bank’s statistical TA on the MDG indicators is proceeding under the Marrakech Action Plan for Statistics (MAPS).16 Because the World Bank has the technical expertise in the socio-demographic field and has played a leading role in the development and maintenance of countries’ socio-demographic metadata, the staff will continue to work with the Bank in editing and posting new and updated GDDS metadata on the socio-demographic MDGs.[k10]

26. Both the countries and donor community broadly recognize the GDDS as the core framework for statistical capacity building. Seventy-eight countries participate in the GDDS and others are preparing to participate. Moreover, the National Strategies for Development of Statistics (NSDS) Framework of PARIS2117 has recognized this role of the GDDS. PARIS21 has launched a major project to introduce the NSDS for most developing countries as a part of the MAPS. The NSDS complements the GDDS. It covers areas beyond the expertise and mandate of the Fund, such as developing statistical infrastructure and conducting administrative surveys, that are critical to developing the source data underlying the macroeconomic statistics that are the Fund’s principal focus.[k11]

27. Using a regional approach, the Fund has leveraged scarce technical assistance resources to assist countries that participate or wish to participate in the GDDS. The regional GDDS projects18 have benefited from close collaboration with the World Bank. Technical assistance is provided by the projects’ resident technical advisors, short-term experts, or STA staff.

28. The GDDS has been the focus of the statistical work at the regional technical assistance centers. Since the Fifth Review, the Middle East Technical Assistance Center (METAC) has been launched. In addition, in Central America, in collaboration with the Central American Monetary Council (CAMC) and with financial support from Japan, STA has initiated a program of technical assistance for countries that wish to participate in the GDDS. These countries participated in a GDDS seminar/workshop in April 2004. Two of these countries (Guatemala and Nicaragua) have started to participate in the GDDS, while two (the Dominican Republic and Honduras) are currently finalizing their metadata and other preparatory work for GDDS participation.

G. Incorporating The Gdds Into The Poverty Reduction Strategy Papers

29. The possibility of more fully incorporating the GDDS into the PRSP framework has assumed greater importance since the Fund’s Independent Evaluation Office’s review of overall Fund technical assistance and the Executive Board’s approval of its key recommendations.19 One of these recommendations is that, for low-income countries (LICs), the PRSP should be the framework guiding Fund technical assistance priorities. This is particularly appropriate for technical assistance in statistics, which has a medium-term strategic focus derived from the GDDS. Such an approach is an essential step toward transforming the PRSP into an evidence-based performance framework and, if appropriately costed, statistical reforms would stand a better chance to be fully funded within the PRSP medium-term expenditure and macroeconomic frameworks.[k12]

30. The authorities in Sierra Leone, working with the Fund and the World Bank, used an innovative approach for incorporating elements of the GDDS in the PRSP. This effort comprised a systematic diagnosis of the statistical system along with the formulation of a capacity-building plan for statistical development.20 The GDDS plans for improvement have been integrated with the other statistical requirements of the PRSP, which have been costed and will be reflected in the annual budgets.[k13] This should help identify the funding requirements for GDDS/PRSP statistical reforms. The PRSP acknowledges in a separate section, “Empowerment with Statistics,” the central role of statistics in supporting the overall poverty reduction strategy, including the pursuit of the MDGs. The staff believes integration with the PRSP will strengthen the GDDS as the core framework for statistical capacity building, and will explore ways to carry that forward.[k14]

H. Continuing Implementation Of The Data Quality Program (DQP)

31. SDDS metadata within each of the four dimensions already are parsed into “key concepts” corresponding broadly to a set of “prompt points” aiding subscribers in drafting and editing metadata. Similar prompt points apply to the GDDS metadata.21 In early 2005 the staff revised the prompt points to align with DQAF three-digit topics and has begun parsing the SDDS and GDDS metadata on the DSBB into these detailed DQAF topics or “granules.” [kdz15]Once completed, the newly “granularized” DSBB metadata can be linked to other DQAF-organized metadata bases such as the Data ROSCs and the Fund’s Project Management System for statistical technical assistance, integrating and streamlining STA’s surveillance and technical assistance work with the Fund’s data transparency initiatives. This is being undertaken while imposing no additional reporting burden on SDDS subscribers and GDDS participants.[k16] See section III.D. below and the supplement to this paper on metadata standardization22 regarding the staff’s proposals for this DQAF integration project going forward.

III. Advancing the Data Standards Initiatives

A. Maintaining the Credibility of the SDDS

32. Empirical evidence is accumulating on the benefits the SDDS provides to subscribers in lowered borrowing costs. These studies, described in Appendix II, consistently show narrowed interest rate (yield) spreads on securities issued in emerging markets for countries subscribing to the SDDS. An article forthcoming in IMF Staff Papers will report that primary market spreads were lower during the period studied by close to 20 percent (or by an average of 55 basis points) for SDDS subscribers.23

33. The continued benefits of the SDDS to subscribers and the public depend on its credibility. The credibility of the SDDS rests on a culture of observance with its provisions among subscribers. A culture of observance rests on a general understanding among subscribers of the importance of adhering to SDDS requirements, but also importantly on the comprehensiveness, accuracy, and timeliness of the staff’s monitoring of subscribers’ observance of the SDDS.

34. The staff plans to start disseminating, on the DSBB, reports assessing subscribing members’ observance of their SDDS undertakings.24 The reports would focus on issues related to punctuality of data releases (how closely subscribers adhere to their ARCs), timeliness of data release (how closely subscribers comply with SDDS timeliness requirements), whether subscribers regularly certify the currentness of their metadata, and coverage issues such as persistent deviations from the reserves template guidelines and the external debt prescriptions. A draft format for the annual monitoring report is shown in Appendix IV. In early 2007 the staff intends to post on the DSBB reports on subscribers’ observance of the SDDS during 2006.

35. STA and area departments would continue to work together when serious SDDS observance issues arise. Serious issues central to effective surveillance would relate, among other things, to undercoverage of released statistics relative to SDDS requirements. Reserves template coverage issues generally fall into this category, for example. The SDDS contains precise guidelines to which subscribers must adhere in disseminating the template data because the Fund considers them critical for financial surveillance.25 For the other data categories, recurrent deviations on the data dimension (regarding coverage, periodicity, and timeliness of the data) or the access dimension (regarding the provision of an ARC and dissemination of a NSDP based on STA’s guidelines) could lead staff to identify these issues as serious deviations. In all instances, the staff tries to resolve minor deviations with the subscriber. If the issue is not resolved and its recurring nature leads the staff to identify it as a serious deviation, the staff would inform the subscriber when it has started to be considered as such. The subscriber notification would trigger the established SDDS procedure for dealing with serious deviations.[k17]26

36. In an environment of tight resources and increased demand for support from an enlarged group of SDDS subscribers, the staff is standardizing and automating SDDS monitoring. Subscribers now can transmit ARC information using electronic templates and have their NSDPs scanned to facilitate accurate updating. Subscribers shortly will be able to update their metadata using standard text tables in electronic templates downloaded from the DSBB and editable with word processing software. Subscriber cooperation in using ARC templates is the first important factor in the success of this initiative. Universal use of ARC templates will increase accuracy by reducing transcription errors. It also will reduce processing lags between receipt of calendar information from subscribers and posting ARC information on the DSBB. Subscriber cooperation with staff-designed NSDP guidelines is the second important factor in the success of this initiative. The NSDP is the responsibility of the subscriber. Accurate electronic scanning requires the NSDP adhere to formatting standards set by the staff. Once the NSDP information is successfully scanned, accuracy also requires the subscriber to promptly update the NSDP when new data are released. Subscriber cooperation with using metadata templates is the third important success factor for accurate and efficient monitoring. Accurate capture of metadata updates and short metadata processing lags are important in monitoring the SDDS-prescribed quarterly certification of metadata. Successfully implemented, these improvements will allow more effective staff monitoring of countries’ observance of their SDDS undertakings and streamline preparation of observance reports on the DSBB. Further improvements should be possible when countries adopt the emerging SDMX standard for encoding information in data files and in tables on websites created using eXtensible Markup Language (XML).

37. The staff proposes to include subscriber observance of automated monitoring arrangements as an SDDS undertaking. Monitoring observance is a critical factor in promoting sustained adherence to SDDS requirements, and sustained adherence to SDDS requirements is critical to the credibility of the SDDS to capital markets and the public. This paper has noted the growing evidence that, as a monitored standard, the SDDS may confer monetary benefits to subscriber country governments. More and more member countries may have access to these transparency benefits as SDDS subscribership grows. However, accurate monitoring will become increasingly difficult to sustain without universal subscriber observance of automation-compatible arrangements for reporting ARC information, for presenting data releases on NSDPs, and for reporting metadata for quarterly certification and update. The most recent subscribers to the SDDS have been able to comply with automated monitoring guidelines on NSDP format and use of the ARC template. The staff will continue to work with subscribing countries in designing automated arrangements minimizing reporting burden as well as maximizing efficiency.[k18]

B. Setting Priorities for Technical Assistance

38. Resource constraints and increased demand for support from an enlarged number of SDDS subscribers and GDDS participants characterize the medium-term environment for STA. Of the remaining 46 Fund member countries that are not SDDS subscribers or GDDS participants, five are working toward SDDS subscription and 18 are working toward participation in the GDDS. The rest of this group (23 countries) have not indicated plans to join in the SDDS or the GDDS. The staff believes official commitment to data transparency should be an important factor in setting technical assistance priorities.27 [k19]Countries desiring participation in the GDDS and GDDS participants desiring to become SDDS subscribers demonstrate this commitment.

C.Improved Metadata for Oil and Gas Activities

39. The staff proposes to work with SDDS subscribers and GDDS participants to deepen metadata on oil and gas activities and products on an “encouraged” basis. 28 In a seminar on recent oil market developments and issues in March 2005, Executive Directors agreed that the timeliness, accuracy, and comprehensiveness of data on both oil supply and demand should be improved, including data on inventories. The responsibility for oil statistics is well-entrenched among international organizations. While the Fund is a user rather than producer or disseminator of such statistics, it is the central international organization promoting data transparency. In this latter role, the Fund could consider how oil and gas statistics could be represented in the SDDS and GDDS. The staff proposes to deepen the metadata for selected data categories in both the SDDS and the GDDS. The deepened metadata would comprise descriptions of how oil and gas production and consumption, trade and other external transactions, and government revenues are represented in several data categories: national accounts, production index, merchandise trade, balance of payments, central and general government operations, and producer and consumer price indices.[k20]

40. Based on external trade data reported to the United Nations, some 15 SDDS subscribers and 10 GDDS participants are net exporters of oil and gas while the rest of SDDS subscribers and GDDS participants are net oil importers.29 [k21] Of the exporters, 44 SDDS subscribers and 25 GDDS participants are oil producers. Tables 3, 4, and 5 present, respectively, oil producing countries, oil and gas net exporting countries, and oil and gas net importing countries grouped by subscription to the SDDS or participation in the GDDS.

Table 3.

Oil Producing Countries

article image
Source, except +: Countries having positive production of oil in 2002 based on data from the Energy Information Administration, U.S. Department of Energy (http://www.eia.doe.gov/pub/international/iealf/table22.xls).

Source: Monthly Oil Market Report of the International Energy Agency, June 2005, data for selected oil producing countries for 2003. (All countries marked + also show positive production for 2002 in the Energy Information Administration source, except Chad.)

Countries belonging to the Organization of Petroleum Exporting Countries (OPEC) (see http://www. opec.org/ aboutus/).

Table 4.

Net Exporters of Oil and Gas

article image
Source: Countries having positive net exports of oil and gas based on exports and imports of oil and gas (HS 2709—Petroleum oils, oils from bituminous minerals, crude, HS 2710—Oils petroleum, bituminous, distillates, except crude, and HS 2711—Petroleum gases and other gaseous hydrocarbons) reported to the United Nations during 2001-2004. (http://unstats.un.org/unsd/comtrade/). These data exclude oil producers Equatorial Guinea, Timor-Leste, Iraq, and Libya, for which there are no reports for trade in these commodities classified using the HS system in the UN COMTRADE database for this period. Note that GDDS coverage also is incomplete in this source during 2001-2004: 58 of 78 countries participating in the GDDS are represented.

Countries belonging to the Organization of Petroleum Exporting Countries (OPEC) (see http://www.opec.org/aboutus/).[k22]

Table 5.

Net Importers of Oil and Gas

article image
Source: Countries having positive net imports of oil and gas based on exports and imports of oil and gas (HS commodities 2709-2711) reported to the United Nations during 2001-2004 (http://unstats.un.org/unsd/comtrade/). Note that GDDS coverage is incomplete in this source during 2001-2004: 58 of 78 countries participating in the GDDS are represented.

Special Administrative Region of the People’s Republic of China.

Nonmember of the Fund.

[k23]

41. It is important that indicators of demand, such as oil and gas products in merchandise imports, as well as supply be included in the deepened metadata on oil and gas, as relevant to the country. All countries subscribing to the SDDS or participating in the GDDS would be affected by this initiative. The initial impacts will be primarily to improve metadata on the demand side, as a dominant share of oil and gas is produced by countries that are neither subscribers to SDDS nor participants in the GDDS. Seven of the eleven member countries of the Organization of Petroleum Exporting Countries (OPEC) are not participants in either of the Fund’s data transparency frameworks.

42. The staff proposes metadata be deepened in seven of the data categories covered in the SDDS and GDDS. The affected categories are (1) national accounts, (2) production index, (3) merchandise trade, (4) balance of payments, (5) central and general government operations, (6) consumer price index, and (7) producer price index. The following paragraphs contain the staff’s proposal for deepening the oil and gas metadata in these data categories.

National accounts

43. National accounts estimates would include oil and gas products in several accounts and, to the extent that these products are significant to the country’s economy, their inclusion and source of data would be described in the metadata. This pertains specifically to oil and gas industrial activities in GDP by production, and for GDP by expenditure, to oil and gas consumption and net exports expenditures, as well as changes in oil and gas inventories in capital formation expenditures.

Production index

44. Most countries’ production indices provide data on changes in volume for industrial activities, including oil and gas production. To the extent that oil and gas activities are important in a country’s production index, their coverage and source data would be described in the metadata.

Merchandise trade

45. Where exports or imports of oil and gas products are important commodities in a country’s external trade, metadata on merchandise trade would include a discussion of how petrochemical products are treated in reporting exports or imports. If they are excluded, require special treatment, or are based on estimates, this information would appear in the metadata. It is important that countries state clearly how oil imported for processing/exported after processing is treated.

Balance of payments

46. Cross-border transactions linked to oil and/or gas may be found in several items of the current account. Where these transactions are significant, the metadata would include a description of their treatment in the balance of payments with regard, in particular, to the residence, time of recording, and valuation criteria used. Depending on the circumstances of the economy, the items involved can be imports/exports of goods (including oil/gas drilling rigs and production platforms), of goods for processing, of transportation services (including the issue of pipelines), and of merchanting services, as well as royalties and direct investment income.

Central and general government operations

47. In taking a comprehensive approach to transparency, national governments would compile and disseminate data and metadata on all revenues and other receipts derived from the oil and gas sector. A common way of taxing the oil and gas sector involves a combination of tax and royalty payments which could include levies, such as royalties to secure minimum payments, and the regular income tax that is applicable to all companies. In addition, the imposition of indirect taxes, such as customs duties and VAT, is widespread. However, governments can collect revenue from the oil and gas sector using a variety of tax and nontax instruments and information on the different forms of taxes and the resources that they yield would need to be tailored to take account of the individual national circumstances.

48. In many countries, revenues from taxation of the oil and gas production accrue to different levels of government (federal, state, local). Since these entities individually fulfill the functions of government as their primary activity and together comprise the general government sector, based on the methodological framework of the GFSM 2001, the receipts of the different government entities would be consolidated for comprehensive coverage of the resources derived from the oil and gas sector, and this would be described in the metadata for general government operations.

Consumer price index

49. Consumer price indices (CPIs) cover price developments for the oil and gas products the household sector consumes in categories such as gas and liquid fuels under the category electricity, gas, and other fuels, and fuels and lubricants for personal transport equipment under the transport category. The coverage and source data of oil and gas products within this component of the CPI would be specifically described in the metadata.

Producer price index

50. Most countries’ producer price indices (PPI) provide data on changes in prices for industrial activities, including oil and gas production. To the extent that oil and gas activities are important in a country’s PPI, their coverage and source data would be specifically described in the metadata.

D. Integrating SDDS/GDDS With the Data Quality Assessment Framework Under the Fund’s Data Quality Program

51. The DQP promotes the systematic integration of the various Fund data initiatives. This integration centers on the DQP’s DQAF. The DQAF is a standard topical structure for assessing the quality of countries’ economic and financial data, and is the underlying framework for the Data ROSC. The DQAF comprises a three-level hierarchy (with an associated three-digit coding system) whose top-level topics are called dimensions, whose second-level topics are called elements, and whose third-level topics are called indicators.30

52. The staff is working on structurally aligning countries’ SDDS and GDDS metadata with the DQAF. As noted in Section II.H, this is being undertaken while imposing no additional reporting burden on SDDS subscribers and GDDS participants.[k24] STA’s Data Quality Program, detailed in a supplement to the Fifth Review, envisaged bringing all of the data standards initiatives under a data quality improvement umbrella.31 The DQAF not only underlies the Data ROSC, but also the Project Management System for tracking Fund statistical technical assistance. Identifying three-digit DQAF indicators within the current DSBB metadata will allow links between the DSBB and the data ROSCs, STA’s technical assistance project management system, and, prospectively, data notes published in the Fund’s International Financial Statistics (IFS) and other statistical publications. It will also permit the DSBB to provide users with a DQAF presentation of SDDS and GDDS metadata, in addition to the standard metadata structure of the SDDS and GDDS.

53. Integrating the DQAF into the SDDS and the GDDS has two important implications for the effectiveness of statistical surveillance. First, it will make the SDDS and GDDS more effective in promoting the dissemination of quality information by organizing them under a data quality framework. Second, it will also link SDDS and GDDS metadata, which represent the prevailing level of a country’s statistical attainment, with the country’s data ROSC which assesses the quality of the statistical processes underlying a country’s data. These linkages would enhance the usefulness of SDDS and GDDS metadata, along with the metadata from the data ROSC, for including summary data quality assessments of the datasets from which the Table of Common Indicators Required for Surveillance (TCIRS) is drawn in the Statistical Issues appendices of Article IV documents. As noted in the Review of Data Provision to the Fund for Surveillance Purposes (SM/04/56), these metadata could help strengthen discussions of the analytical implications of data deficiencies and potential remedies for these deficiencies.[k25]

54. Integrating the DQAF into the SDDS and the GDDS will streamline technical assistance work. Section III (Detailed Assessments) of the Data ROSC, already structured using the DQAF, will be integrated with the metadata of the SDDS and GDDS. Further, action plans developed with the authorities on technical assistance missions, also already structured using the DQAF, will be integrated with the GDDS “plans for improvement.”32 Because these GDDS metadata are publicly disseminated on the DSBB, improving them will improve coordination among donors of technical assistance.33[k26]

55. Integrating the DQAF into the SDDS and the GDDS will improve the usefulness of Fund statistical publications. Consistent with STA efforts to improve data consistency among published datasets and with recommendations of the 2004 Board paper on data provision, STA proposes to develop online tables of the IFS time series, as available, that correspond to those from the NSDP of SDDS subscribers, linked to each country’s DSBB “country” page, which also contains a link to the country’s NSDP. STA would develop similar tables of available IFS time series corresponding to those referenced in GDDS metadata in the same NSDP format as IFS data for SDDS subscribers. The staff believes these IFS tables in NSDP format would substantially enhance the functionality of the DSBB to users.

IV. Future Developments in The Data Standards Initiatives

56. Development of Financial Soundness Indicators (FSIs) is progressing as planned. Following the finalization of the Compilation Guide on Financial Soundness Indicators, the Coordinated Compilation Exercise (CCE) was launched in 2004 with 62 participating countries. Fifty-four of 61 SDDS countries or territories and 5 of 78 GDDS countries are among the 62 countries or territories participating in the CCE.34 For the CCE, these countries will compile the 12 core FSIs as of reference date end-December 2005 and provide these data and related metadata by end-July 2006 to the Fund for dissemination, using standard report forms developed for this purpose. Countries are encouraged to compile some or all of the encouraged FSIs. The implementation of the CCE work program, which was finalized in consultation with CCE participating countries, is well on track.35 It is envisaged that by mid-2007, Fund staff will report to the Board on the outcome of the CCE and future work of the Fund in FSIs. The countries’ experience in compiling FSIs and the outcome of the CCE will inform future discussion and consideration by the Executive Board of including FSIs in the SDDS.

57. In considering FSIs as a future SDDS data category, the staff notes that, unlike other types of data, it may be difficult to achieve full standardization of FSIs across countries. Many FSIs build on supervisory data whose definitions vary across countries, depending on supervisory needs and the complexity of financial systems. Although the CCE will identify a core set of 12 FSIs within a consistent conceptual framework, the selection and details of the FSIs countries use in financial policy still will vary across countries. Should FSIs be added to the data categories of the SDDS, this aspect of FSIs would highlight the importance and utility of incorporating the CCE metadata for FSIs in the SDDS.[k27]

58. Work on public debt statistics is ongoing. Consistent with the Review of Data Provision to the Fund for Surveillance Purposes, the staff is working in the TFFS to develop a template for reporting public debt. The work remains at an early stage. In addition, the SDDS Guide 2005 will clarify the alternative breakdowns of central government debt the SDDS sanctions under its existing requirements. The staff will keep the Board apprised of significant milestones in this area and may suggest discussion of the role of public debt in the data standards initiatives in a future review.[k28]

V. Resource Implications

59. Noting the increased and sustained pressures on staff resources associated with the increase in SDDS subscription and GDDS participation, the staff expects to continue to meet data standards program objectives with the existing staff resources through automating SDDS monitoring as well as metadata updates for both SDDS and GDDS. Exploiting these opportunities depends largely on the pace at which SDDS subscribers can implement automation-compliant formats in reporting their ARCs and posting their NSDPs, and both SDDS and GDDS countries adopt planned improvements in electronically reporting metadata. Progress in this area would be assisted by requiring SDDS subscribers to follow the staff’s procedural guidelines in providing their ARCs and the NSDP format to meet the SDDS and provide for electronic monitoring.

VI. Issues for Discussion

A. Concerning Progress Made on the Data Standards Initiatives Since the Fifth Review

60. As discussed in Section II and consistent with Directors’ endorsement during the Fifth Review, the staff has taken steps to update the SDDS and GDDS to maintain their relevance and reflect evolving international statistical best practices. (Paragraphs 8-31.)

61. Are Executive Directors satisfied with the progress made since the Fifth Review of the Data Standards Initiatives?

B. Concerning the Proposal to Make Subscriber Observance of Monitoring Arrangements an SDDS Undertaking

62. Monitoring observance is a critical factor in promoting sustained adherence to SDDS requirements, and sustained adherence to SDDS requirements is critical to the credibility of the SDDS to capital markets and the public. Accurate monitoring of an enlarged number of subscribers will become increasingly difficult to sustain without universal subscriber observance of automation-compatible arrangements for reporting ARC information, for presenting data releases on NSDPs, and for reporting metadata for quarterly certification and update. (Paragraph 37.)

63. Do Executive Directors agree with making observance of automated monitoring arrangements an SDDS undertaking?

C. Concerning According Official Commitment to Data Transparency Due Importance in Setting Technical Assistance Priorities

64. “Conclusions of the Task Force on IMF Technical Assistance” (PIN 05/114) noted the importance of country ownership of technical assistance. The staff believes official commitment to data transparency and the GDDS statistical development framework, including particularly its plans for improvement, should be an important factor in setting technical assistance priorities. Countries desiring participation in the GDDS or subscription in the SDDS and GDDS participants desiring to become SDDS subscribers demonstrate this commitment. (Paragraph 38.)

65. Do Executive Directors agree with including participation or working toward participation in the data transparency standards as an important factor in setting priorities for statistical technical assistance?

D. Concerning Deepened Metadata for Oil and Gas Activities

66. Oil prices, production, and consumption can have a significant impact on the global economy and the economies of individual member countries. The staff proposes to deepen SDDS and GDDS metadata by including descriptions of how oil and gas production and consumption, trade and other external transactions, and revenues are measured in the statistics for several data categories: national accounts, production index, merchandise trade, balance of payments, central and general government operations, consumer and producer price indices. (Paragraphs 39-50.)

67. Do Executive Directors agree that the staff should work with SDDS subscribers and GDDS participants to deepen metadata for oil and gas activities on an “encouraged” basis?

E. Concerning Future Inclusion of New Data Categories

68. The staff considers that indicators of financial stability are receiving growing attention, but notes that whether selected financial soundness indicators could be included in the SDDS would be considered best in light of the results of the coordinated compilation exercise that will have been completed at the time of the Seventh Review. (Paragraphs 57-58.) The staff also proposes the Board consider at the Seventh Review whether to include discussion of public debt as a new data category in the Eighth Review, in light of countries’ progress by then in reporting information according to the public debt template currently being developed. (Paragraph 58.)

69. Do Executive Directors agree with the staff’s plan to examine at the next review whether selected FSIs should be included the SDDS?

F.Concerning Timing of the Next Review of the Fund’s Data Standards Initiatives

70. In view of the expected developments over the next three years in countries’ compilation of financial soundness indicators, in the proposed deepening of metadata on oil and gas activities, and in developments flowing from integration of the data standards initiatives under the Data Quality Program, the staff believes that the next review of data standards initiatives should take place in the second half of 2008.[k29] The staff will continue to report to the Board, as needed, on critical observance issues and, as necessary, the Board could address any further strengthening that may be necessary owing to unforeseen circumstances.

71. Do Executive Directors agree with the proposed timing of the second half of 2008 for the Seventh Review of the Fund’s Data Standards Initiative?

72. Depending on the outcome of Executive Directors’ discussion of the above issues, amendments to the SDDS Annex (Selected Decisions and Selected Documents of the International Monetary Fund, Furnishing of Information, Article VIII, Section 5, as amended, and its Annex, the Scope and Operational Characteristics of the Special Data Dissemination Standard, as amended) and GDDS Document, as amended), if required, will be prepared and circulated to the Board for approval.[k30]

The SDDS and the GDDS: Some Background Information and Lists of Subscribers and Participants

1. This appendix has several components. It first presents some background information on the SDDS and the GDDS covering their purposes and key features. This is followed by lists of SDDS subscribers and GDDS participants. Countries that have subscribed to the SDDS or participated in the GDDS since the Fifth Review of the Data Standards Initiative are also identified.

Some background information on the SDDS and the GDDS

2. The Fund’s work on data dissemination standards began in October 1995, in the aftermath of the Mexican financial crisis, when the Interim Committee (now the International Monetary and Financial Committee endorsed the establishment by the Fund of standards to guide members in the dissemination to the public of their economic and financial data. The standards were to consist of two tiers: the Special Data Dissemination Standard (SDDS) to guide countries that have, or that might seek, access to international capital markets, and the General Data Dissemination System (GDDS) to assist all other member countries. The SDDS was approved by the IMF Executive Board in March 1996, and the GDDS in December 1997. The design and implementation of the SDDS and the GDDS have benefited from close collaboration with member countries and other international organizations.

3. Both the SDDS and the GDDS are intended to enhance the availability of timely and comprehensive statistics, thereby contributing to the pursuit of sound macroeconomic policies. In addition, the SDDS is expected to contribute to the improved functioning of financial markets. A country’s subscription to the SDDS or its participation in the GDDS is voluntary. The Fund has set forth guidelines for SDDS subscription and GDDS participation.

4. The key differences between the SDDS and the GDDS are as follows: (1) The SDDS prescribes specific standards that must be observed by countries that subscribe to it, whereas the GDDS provides guidelines on good practices and is generally less demanding than the SDDS; (2) SDDS subscribers must fully meet the requirements of the SDDS at the time of subscription, whereas the GDDS does not set fixed dates by which participants must improve their existing practices; (3) the focus of the SDDS is on data dissemination in countries that already meet high data quality standards, whereas the primary goal of the GDDS is to assist countries to develop their statistical systems through setting up plans for improvement and identifying technical assistance needs; and (4) the SDDS principally covers macroeconomic and financial data (for the real, fiscal, financial, and external sectors) while including only population among the socio-demographic indicators, whereas the GDDS more comprehensively covers socio-demographic data (population, health, education, and poverty), in addition to macroeconomic and financial data.

Monitoring of SDDS observance

5. There are four dimensions of data dissemination: (1) the data (coverage, periodicity, and timeliness), (2) access by the public, (3) integrity of the disseminated data, and (4) quality of the disseminated data. For each of these dimensions, the SDDS prescribes two to four monitorable elements—good practices that can be observed, or monitored, by the users of statistics.

6. The data dimension lists 18 data categories that provide coverage for the four sectors of the economy, and it prescribes the periodicity (or frequency) and timeliness with which data for these categories are to be disseminated. In recognition of differences in economic structures and institutional arrangements across countries, the SDDS provides flexibility. Certain categories are marked for dissemination on an “as relevant” basis. Further, some data categories or components of data categories are identified as “encouraged” rather than “prescribed.” With respect to periodicity and timeliness, a subscribing member may exercise certain flexibility options while being considered in full observance of the SDDS.

7. The monitorable elements of the SDDS for access, integrity, and quality emphasize transparency in the compilation and dissemination of statistics: (1) to support ready and equal access, the SDDS prescribes (a) advance dissemination of release calendars and (b) simultaneous release to all interested parties; (2) to assist users in assessing the integrity of the data disseminated under the SDDS, the SDDS requires (a) the dissemination of the terms and conditions under which official statistics are produced and disseminated; (b) the identification of internal government access to data before release; (c) the identification of ministerial commentary on the occasion of statistical release; and (d) the provision of information about revision and advance notice of major changes in methodology; (3) to assist users in assessing data quality, the SDDS requires (a) the dissemination of documentation on statistical methodology and (b) the dissemination of component detail, reconciliations with related data, and statistical frameworks that make possible cross-checks and checks of reasonableness.

8. Consistent with this comprehensive view of data dissemination, dissemination itself is broadly to include electronic dissemination in addition to the more traditional formats.

9. A subscriber is expected to submit information about its data and its dissemination practices—its metadata—to the IMF for presentation on the IMF’s Dissemination Standards Bulletin Board (DSBB) on the Fund’s website (www.imf.org). Subscribers’ metadata are reviewed by the IMF for comprehensiveness and international comparability. The responsibility for the accuracy of the metadata, including timely updates, and for the economic and financial data underlying the metadata rests with the subscriber. In addition, subscribers are required to certify the accuracy of all metadata posted on the DSBB on a quarterly basis. Metadata certification is required three days after the end of the quarter.

10. The DSBB is maintained by the IMF. SDDS metadata provide insights into the usefulness and limitations of the published data. Their presentation on the DSBB facilitates monitoring of observance of the standard by IMF staff, the financial markets and other data users. As national summary data pages (NSDP) play a critical role in monitoring observance of the SDDS by the IMF staff, the Executive Board of the IMF decided in December 1998 to make hyperlinks to NSDP mandatory. The DSBB provides hyperlinks between the SDDS metadata and actual country data shown in the NSDP for all subscribers.

11. A member’s presence on the DSBB as a subscriber to the SDDS indicates its intention to observe certain tenets of good statistical citizenship. Serious and persistent nonobservance of the SDDS will be cause for action. Procedures to be followed in instances of such nonobservance have been approved by the Executive Board of the IMF.

Commitment called for in the GDDS

12. The GDDS provides members a basic framework for their national statistical development. The GDDS addresses the full range of issues critical for compiling and disseminating data and making explicit plans for developments to align national procedures with best practices. As in the SDDS, four dimensions of data compilation and dissemination are covered in the GDDS framework: (1) the data: coverage, periodicity, and timeliness; (2) access by the public; (3) integrity of the disseminated data; and (4) quality of the disseminated data.

13. The GDDS calls for the national authorities to set their own priorities and timing to develop their statistical systems. It allows the data-producing agencies to take control of their statistical development programs in a structured manner and to coordinate effectively among producing agencies, users, and the international community. The IMF supports these efforts by providing technical assistance and by catalyzing support from other sources.

14. In participating in the GDDS, member countries prepare metadata on their current statistical practices, develop their plans for improvement in the short and medium term, and identify associated needs for assistance in implementing these plans. Participating countries must update (at least annually) and revise their metadata to describe how their data compilation and dissemination activities are keeping pace with the best statistical practices. The DSBB also disseminates metadata of GDDS participating countries.

SDDS Subscribers as of End-May 2005

article image

Participating country in the coordinated compilation exercise (CCE) for Financial Soundness Indicators.[kdz31] Cyprus and Luxembourg, which are working toward SDDS subscription, also participate in the CCE.

GDDS Participants as of End-May 2005*

article image

Note: As of May 2005, 83 countries have participated in the GDDS. Five now subscribe to the SDDS.

Kazakhstan was an initial participant in the GDDS, having metadata first posted to the DSBB on May 22, 2000. In March 2003, Kazakhstan became the 53rd subscriber to the Special Data Dissemination Standard (SDDS).

Armenia was initially a participant in the GDDS, having metadata first posted to the DSBB on March 29, 2001. In November 2003, Armenia became the 54th subscriber to the Special Data Dissemination Standard (SDDS).

Bulgaria was initially a participant in the GDDS, having metadata first posted to the DSBB on May 22, 2000. On December 1, 2003.

Bulgaria became the 55th subscriber to the Special Data Dissemination Standard (SDDS).

The Kyrgyz Republic was initially a participant in the GDDS, having metadata first posted to the DSBB on February 14, 2001. On February 26, 2004, Kyrgyz Republic became the 57th subscriber to the Special Data Dissemination Standard (SDDS).

Romania was initially a participant in the GDDS, having metadata first posted to the DSBB on February 14, 2001. On May 4, 2005, Romania became the 61st subscriber to the Special Data Dissemination Standard (SDDS).

Participating country in the coordinated compilation exercise (CCE) for Financial Soundness Indicators. Ghana, which is working toward GDDS participation, also participates in the CCE.

New SDDS subscribers since the Fifth Review

15. The Republic of Armenia on November 7, 2003 became the 54th subscriber to the SDDS, marking a major step forward in the development of the country’s statistical system. The Armenian authorities view subscription as the culmination of many initiatives on the statistical front. The Republic of Armenia is the third country of the Commonwealth of Independent States to subscribe to the SDDS. The Republic of Armenia was among the first group of countries participating in the IMF’s GDDS to have metadata posted to the IMF’s DSBB, serving as an example for other countries in the region. On this occasion The Republic of Armenia is the second country to move from the GDDS to SDDS subscription, again providing an example to be followed by GDDS participants.

16. Bulgaria on December 1, 2003 subscribed to the SDDS. The country showed strong commitment to develop its macroeconomic and socio-demographic statistics in line with the IMF’s Data standards initiatives in 1999 when it participated in one of the outreach GDDS regional seminars. This commitment materialized in May 2000, when Bulgaria became one of the first participants in the GDDS. After that the country intensively used the GDDS as a framework for further improvement of its national statistical system to meet all the requirements of the SDDS. Bulgaria participated in the regional SDDS seminar in Athens in 2002 and hosted a ROSC Data Module mission in January 2003. On the basis of its recommendations as well as with the technical assistance and advice of the SDDS assessment mission that visited Sofia in September 2003, Bulgaria successfully finalized its preparation for subscription and became the third GDDS country to graduate to the SDDS. On December 3-4, 2003 the country hosted a regional seminar for SDDS coordinators. Bulgaria considers the subscription to the SDDS as an important step in its ongoing integration to the European Union.

17. Uruguay on February 12, 2004 became the 56th subscriber to the SDDS, marking a major step forward in the development of the country’s statistical system and underscoring its commitment to transparency. The Uruguayan authorities view subscription as the culmination of many initiatives to improve their data dissemination practices. Uruguay is the 10th country of the Latin American region to subscribe to the SDDS, and the third country of the South Common Market (MERCOSUR) after Argentina and Brazil. Uruguay was one of the countries that attended the regional seminar in Mexico for potential SDDS subscribers in Latin American and the Caribbean region, that was held in April 2003.

18. The Kyrgyz Republic on February 27, 2004 became the 57th country to subscribe to the SDDS. The Kyrgyz Republic is the fourth country of the Commonwealth of Independent States to subscribe to the SDDS. The Kyrgyz Republic has participated in the IMF’s GDDS since February 2001 when its metadata were first posted to the IMF’s DSBB, serving as an example for other countries in the region. The Kyrgyz Republic is the fourth country to move from participation in the GDDS to SDDS subscription, again providing an example to be followed by GDDS participants. The Kyrgyz Republic was one of the countries that attended the regional seminar in Athens for potential SDDS subscribers that was held in November 2002.

19. Belarus on December 22, 2004 became the 58th subscriber to the IMF’s SDDS, marking a major step forward in the development of the country’s statistical system. On February 1, 2005, a Report on Observance of Standards and Codes (ROSC) data module for Belarus, which was based on an assessment made by the IMF’s Statistics Department during March 23-April 7, 2004, was disseminated on the IMF’s external website. Belarus’s authorities view SDDS subscription and the dissemination of the data ROSC as the culmination of many initiatives on the statistical front.

20. Egypt on January 31, 2005 subscribed to the SDDS. The country began its preparation for subscription to the SDDS as early as 1998 when metadata for some data categories were informally submitted to the Fund. By mid-2002 metadata for almost all data categories were drafted. However, major difficulties remained related to the coverage of the data categories international reserves and foreign currency liquidity (reserves template), international investment position, and external debt. Egypt hosted a ROSC Data Module in October 2003 and its recommendations played an important role in the development of the national statistical system to meet the SDDS requirements. All issues related to the reserves template were resolved during an IMF mission in December 2004, which enabled the country to finalize its preparation for subscription in January 2005. Egypt is the second country in the Middle East to subscribe to the SDDS. The authorities view the subscription as a strong commitment to transparency and the culmination of many initiatives on the statistical front.

21. The Russian Federation on January 31, 2005 became the 60th subscriber to the IMF’s SDDS, marking a major step forward in the development of the country’s statistical system. On May 7, 2004, a Report on Observance of Standards and Codes (ROSC) data module for the Russian Federation, which was based on an assessment made by the IMF’s Statistics Department during October 8-23, 2003, was disseminated on the IMF’s external website. The Russian Federation’s authorities view subscription as the culmination of many initiatives on the statistical front.

22. Romania subscribed to the SDDS on May 4, 2005. The country joined the GDDS on February 14, 2001 and since then began using it as a framework to elaborate its macroeconomic statistics in line with best international practices. In 2003 Romanian authorities showed strong commitment to transparency by appointing a SDDS coordinator, which started their preparation to subscribe to the standard. In November 2004 a SDDS assessment mission from the IMF Statistics Department visited Bucharest and provided technical assistance and advice to the National Institute of Statistics, the Ministry of Public Finance and the National Bank of Romania in their joint work to meet all SDDS requirements. Romania is the fifth country to graduate from GDDS to SDDS, setting an example for other GDDS participants.

New GDDS Participants Since the Fifth Review

23. Countries listed below have become participants in the International Monetary Fund’s General Data Dissemination System (GDDS) since the Fifth Review, marking a major step forward in the development of these countries’ statistical systems. Comprehensive information on statistical production and dissemination practices (metadata) of these countries was published on the IMF’s Dissemination Standards Bulletin Board on the date shown.

article image

Improving Transparency, Market Access, and Statistical Capacity

The SDDS in improving transparency and market access

Impact on the functioning of financial markets

1. The SDDS was designed to inform financial decisions in international capital markets. In June 1995, in recognition of the “public-good” character of economic and financial data, the G7 countries asked the IMF to establish benchmarks of good statistical practices and a means to identify countries that observe them. The Fund responded by establishing the SDDS (see also Appendix I). The SDDS was designed to encourage subscribers to disseminate to the public more timely, accurate, and comprehensive macroeconomic and financial data and, equally important, metadata about the usefulness and limitations of published economic information.

2. The SDDS has enhanced data transparency and contributed to market stability. Subscribers’ provision of comprehensive, frequent, timely, and accurate data to market participants facilitates sound risk management, because it allows market participants to better assess country risks in their investment decisions. Subscribers’ adoption of internationally accepted standards of good statistical practices also reassures market participants of the quality of country data.36 Sound risk management contributes to market stability.

3. Enhanced data transparency has facilitated capital flows across countries. A study entitled “Transparency and International Investor Behavior”37 found that (1) international investment funds prefer to hold assets in more transparent markets; (2) herding among funds is more prevalent in less transparent countries; and (3) during a crisis, international investors tend to remove funds from less transparent markets, with transparency gauged from the IMF’s ROSCs, including the Data ROSC.38

Impact on early warning and economic policy reorientation

4. The lack of data transparency was recognized as a key weakness during the financial crisis of the mid-1990s. Studies have suggested that greater data transparency could have led to timely corrections of unsustainable policies in the Mexican financial crisis of 1994-95. Studies in the literature on financial crises have also shown that the lack of data transparency in the Asian banking systems made it difficult for policymakers to gauge the weaknesses of the systems and to respond with timely improvements.39 It has been observed that increasing transparency not only can help prevent and mitigate crises but also manage them when they happen. Making transparent, timely, and accurate information public can serve as an economic stabilizer, enabling investors to assess accurately various investment risks and allowing the authorities to maintain their credibility and cushion overreactions from markets in times of financial stress.

Impact on reducing borrowing costs and favorable effects on credit ratings

5. Studies have found that the SDDS has had a positive impact on subscribers’ access to international capital markets. For example, a study by the Institute for International Finance indicated that SDDS subscription substantially reduced credit spreads for emerging market economies.40 Two other studies completed in 2003, “The Link Between Adherence to International Standards of Good Practice, Foreign Exchange Spreads, and Ratings” (IMF WP/03/74) and “Is Transparency Good for You, and Can the IMF Help?” (IMF WP/03/132) provided evidence that Emerging Market Bond Index (EMBI) spreads in secondary markets were reduced by about 15 percent and 4-12 percent, respectively, following SDDS subscription.41 More recently, the paper “Does SDDS Subscription Reduce Borrowing Costs for Emerging Market Economies?” (IMF WP/04/58), which examined the impact of the SDDS on spreads on foreign currency (U.S. dollar, euro, yen) bonds issued in primary markets by seven emerging economies, found that spreads declined by some 75 basis points (or about 19 percent). A revised version of this paper (forthcoming in IMF Staff Papers) reports results from an enlarged sample of 17 countries, with broadly similar results. It found that primary market spreads declined by close to 20 percent (or by an average of 55 basis points) for SDDS subscribers.42 These studies provide empirical evidence of potentially significant savings in borrowing costs for countries subscribing to the SDDS.

The GDDS in statistical capacity building

A framework for developing countries’ statistical systems

6. The GDDS is a development framework for statistics that takes account of the statistical capacities of participating countries. Specifically, the GDDS (1) recommends best practices for the orderly development of statistical systems, (2) contains tools for diagnosing areas in the system that require priority attention, and (3) supports the formulation and implementation of short- and medium-term “plans for improvement” as well as flexibly adapting them to changing circumstances.

7. Since its inception, the GDDS has been a useful tool for countries to develop their statistical systems. Participation in the GDDS involves a structured process through which member countries commit voluntarily and publicly to improve the quality of the data they compile and disseminate to the public.43 The GDDS allows for a comparison of country practices with best international practices and facilitates the development of country strategic plans for statistical improvement. Countries that participate in the GDDS determine for themselves the priorities they will pursue and present their priorities in a set of statistical development plans. In their “plans for improvement” disseminated to the public on the Fund’s DSBB, GDDS participants also indicate their needs for external technical and financial assistance. By participating in the GDDS, countries also commit themselves to keep users informed of any new developments by updating the metadata and “plans for improvement” at least once a year. The Fund supports these efforts by providing technical assistance its areas of expertise and by catalyzing assistance from other sources.

8. SDDS subscription is an important goal for a GDDS statistical development program. One objective of statistical capacity-building is to achieve a level of data transparency meeting SDDS subscription requirements. Graduation from GDDS participation to SDDS subscription thus is a logical progression in statistical development. To date, Armenia, Bulgaria, Kazakhstan, the Kyrgyz Republic, and Romania have graduated from the GDDS to become SDDS subscribers.

A core framework for coordinating statistical technical assistance

9. The donor community broadly recognizes the GDDS as the core framework for statistical capacity building. For example, the National Strategies for Statistical Development of Statistics (NSDS) Framework of PARIS2144 has recognized this role of the GDDS. The NSDS complements the GDDS. It not only covers areas beyond the expertise and mandate of the Fund, it goes into areas such as statistical infrastructure and administrative surveys, recognizing these areas as critical to developing source data underlying the macroeconomic statistics that are the Fund’s principal focus.

10. Using a regional approach, the Fund has leveraged scarce technical assistance resources to assist countries that participate or wish to participate in the GDDS. The regional GDDS projects45 have benefited from close collaboration with the World Bank. Technical assistance is provided by the projects’ resident technical advisors, short-term experts, or STA staff.

11. The GDDS has been the focus of the statistical work at the regional technical assistance centers. Since the Fifth Review, the Middle East Technical Assistance Center (METAC) has been launched. In addition, in Central America, in collaboration with the Central American Monetary Council (CAMC) and with financial support from Japan, STA has initiated a program of technical assistance for countries that wish to participate in the GDDS. These countries participated in a GDDS seminar/workshop in April 2004. Two of these countries (Guatemala and Nicaragua) have started to participate in the GDDS, while two (the Dominican Republic and Honduras) are currently finalizing their metadata and other preparatory work for GDDS participation.

Developments in Countries’ Migration to New Manuals

GFSM 2001

1. The Government Finance Statistics Manual 2001 (GFSM 2001) was published in December 2001, replacing A Manual on Government Finance Statistics, 1986 (GFSM 1986). The GFSM 2001 introduces a comprehensive analytical framework that records the resource flows on an accrual basis and defines several balancing items appropriate for fiscal analysis. The GFSM 2001 analytic framework, though conceived from an accrual perspective, nonetheless can be used to present data generated by a variety of accounting practices. A significant improvement of the GFSM2001, compared to the GFSM 1986, is that it integrates stocks and flows (transactions and other economic flows) in a balance sheet framework that is useful for measuring the net worth of the general government and conducting fiscal sustainability analysis over time. The GFSM 2001 also requires that data be compiled on a cash basis for measuring the liquidity position of the general government. Moreover, the concepts and principles set out in the GFSM 2001 are harmonized with the other macroeconomic statistical standards46 to facilitate consistency of statistical analysis.

2. Subsequent to its introduction, staff has been providing technical assistance to member countries to implement migration to the GFSM 2001 methodology as the basis for compiling GFS for fiscal analysis. With respect to the dissemination of data in the Government Finance Statistics Yearbook (GFS Yearbook), beginning in 2003, staff reclassified all historical data in the GFS database going back to 1990 to conform as close as possible to the GFSM 2001 format. These reclassified data—along with newly reported country data—were initially published in the 2003 GFS Yearbook. Continuing this practice, the 2004 GFSY, presented data for 126 countries, 58 of which reported data according to the GFSM 2001 methodology as table II.1 (below) illustrates. The trend toward reporting data according to this framework (85 percent of all reporters in the 2004 GFSY) is expected to improve further in 2005.

Table 6.

Countries Reporting GFSY Data in GFSM 2001 Format, 2002-04

article image
Source: GFS Yearbook database (as of December 2004).

The 2002 GFS Yearbook was published using the GFSM 1986 presentation. A 2002 GFSY supplement covering 22 countries was published using the GFSM 2001 presentation.

The 2003 GFS Yearbook was published using the GFSM 2001 presentation; the GFSM 1986 presentation was discontinued.

SDDS: Sample Format for Annual Monitoring Report

1. The annual monitoring report would assess each subscribing member’s observance of its undertakings under the SDDS. These undertakings relate to the punctuality and timeliness of data releases, coverage requirements, and quarterly certification of members’ metadata. The annual monitoring report comprises a summary report showing the performance of each subscribing member for all data categories with hyperlinks to individual country reports by data category.

2. The assessment of each subscribing member’s SDDS observance would be made on the following basis:

  1. How closely subscribers adhere to their advance release calendars (punctuality) and comply with SDDS timeliness requirements (timeliness) would be assessed in terms of total number of days of delay for all data categories in the summary report as well as in terms of days of delay for each data category for the individual country reports;

  2. Coverage issues would include data categories for which there are persistent deviations from SDDS specifications; and

  3. How closely subscribers meet the SDDS metadata certification requirements would be assessed in terms of the number of quarters for which subscribers did not certify as compared with the timeframe specified in the legal text of the SDDS47 and the accuracy of the metadata posted on the DSBB.

3. Starting in early 2007, staff plans to disseminate on the DSBB annual reports on subscribers’ observance of the SDDS. The first such report will reflect subscribers’ SDDS observance during 2006.

Sample Format for Annual Monitoring Report

article image

Hyperlinked to individual country annual reports.

Sample Format for Annual Monitoring Report

article image
1

The SDMX consortium comprises the Bank for International Settlements, the European Central Bank, the IMF, the Organisation for Economic Co-operation and Development, the Statistical Office of the European Communities (Eurostat), the United Nations Statistics Division, and the World Bank.

2

In the wake of the 1994 crisis in Mexico, the international community recognized the essential role of data transparency for meeting the challenges and risks of globalization and reducing the likelihood of financial crises. It called for timely dissemination of reliable macroeconomic and financial data and an improved early warning system that would permit a swifter response to financial shocks.

3

In May 2005, there were 61 SDDS subscribers and 78 GDDS participants.

4

Developments in this area will be covered in a separate Board paper on the Review of the Standards and Codes Initiatives (forthcoming).

5

Afghanistan, Algeria, Bhutan, Burundi, Djibouti, Dominican Republic, Eritrea, Ghana, Haiti, Honduras, Iraq, Lao PDR, Palau, Qatar, Samoa, Seychelles, and Tonga. West Bank and Gaza, although not a member of the Fund, is also working toward GDDS participation.

6

At the Fifth Review, this program was called the “25/50 program.” It referred to a data standards outreach program identifying countries considered capable of meeting the SDDS requirements within the next two to three years (approximately 25 countries) and a large group of 50 countries that could meet the requirements within about five years. The group of 50 has enlarged to the remaining approximately 60 GDDS countries, in a group now called Tier 1. The Tier 2 group comprises nonparticipating IMF member countries and selected territories of these members.

7

Four of the eight new subscribers noted in this Review were previously participants in the GDDS (Armenia, Bulgaria, the Kyrgyz Republic, and Romania). There have been a total of five graduates from GDDS to SDDS. As noted in the Fifth Review, Kazakhstan was the first graduate in March 2003. For details see Appendix 1.

8

The Fifth Review of the Data Standards Initiatives referred to a June 2003 Supplement by the Fund and the World Bank that also identified MDG indicators already present in the macroeconomic data sectors of the GDDS.

9

The prescribed series are quarterly external debt data covering four sectors (general government, monetary authorities, the banking sector, and other), further disaggregated by maturity (short and long term), and provided on an original maturity basis and by instrument, as set out in BPM5.

10

The current members of the TFFS are the Bank for International Settlements, Commonwealth Secretariat, Eurostat, European Central Bank, IMF, OECD, Paris Club Secretariat, UNCTAD, and the World Bank.

11

The database is available on the World Bank website at http://www.worldbank.org/data/working/QEDS/sdds_main.html.

12

The option is detailed in Supplement 3, Fifth Review of the Fund’s Data Standards Initiatives—Government Finance Statistics Manual 2001—Adjusting the Special Data Dissemination Standard Requirements for the Fiscal Sector, available at http://www.imf.org/external/np/sta/dsbb/2003/eng/gfs.htm.

13

For example, when the fiscal year is the same as the calendar year, to use the option for monthly CGO data for December 2004 a subscriber would need to disseminate quarterly GGO data for the last quarter of 2004.

14

Supplement 1 to the Fifth Review prepared jointly by the World Bank and the IMF staff, The General Data Dissemination System and the Millennium Development Goals, set the context for and relationship between the GDDS and the series of statistical indicators for monitoring countries’ progress toward meeting the MDGs. The joint paper encourages participants to compile MDG indicators and to include in the GDDS metadata information that would assist data users in understanding the underlying sources and methods for the derivation of the indicators.

15

Central African Republic, Madagascar, Mauritania, Pakistan, and Trinidad and Tobago.

16

The MAPS resulted from the Second International Roundtable on Managing for Development Results of February 2004. The World Bank manages a Trust Fund for Statistical Capacity Building and provides loans (STATCAP) for countries to implement their statistical master plans. As one of the actions agreed in the MAPS, the World Bank also leads the development of International Household Survey Network, an international initiative to improve availability, accessibility, and use of survey data, an important source for the socio-demographic MDG indicators. See http://www.internationalsurveynetwork.org/home/.

17

The PARIS21 consortium’s founding institutions are the OECD, the World Bank, the United Nations, the IMF, and the European Commission; it also comprises a significant number of other major providers of statistical technical assistance worldwide, including several major bilateral and international agencies.

18

Regional GDDS projects exist for member countries of West African Economic and Monetary Union (WAEMU), Anglophone African countries, Lusophone African Countries, and Pacific Island countries and territories.

19

“Conclusions of the Task Force on IMF Technical Assistance” (PIN 05/114) noted the importance of country ownership of technical assistance.

20

Republic of Sierra Leone, Poverty Reduction Strategy Paper, Joint Staff Advisory Note, paragraph 10, footnote 5

21

See the GDDS Guide 2004, Appendix V for the prompt points.

22

Metadata Standardization in the Data Quality Program.

24

The legal text of the SDDS (Selected Decisions and Selected Documents of the International Monetary Fund, Article VIII, Section 5, Furnishing of Information, as amended) provides for the dissemination of such reports: “Finally, an annual report that assesses each subscribing member’s observance of its undertakings under the SDDS will be posted on the DSBB.”

25

In addition to the usual coverage, periodicity, and timeliness requirements, prescriptions for the reserves template cover the provision of detailed components compiled according to guidelines provided in: International Reserves and Foreign Currency Liquidity – Guidelines for a Data Template, A. Kester, IMF, 2001.

26

Selected Decisions and Selected Documents of the International Monetary Fund, Article VIII, Section 5, Furnishing of Information, as amended, and its Annex, the Scope and Operational Characteristics of the Special Data Dissemination Standard, as amended.

27

“Conclusions of the Task Force on IMF Technical Assistance” (PIN 05/114) noted: “Directors saw particular merit in greater involvement by country authorities in the design and follow up of technical assistance activities, based on the needs and priorities set out, where possible, in PRSPs….In this connection, most Directors agreed that countries’ track records in implementing TA recommendations should guide future TA allocations by the Fund. However, a few Directors cautioned that such a criterion amounted to a form of conditionality, and suggested that TA allocations be determined on a case-by-case basis.” While the staff proposes that commitment to the data transparency initiatives and the GDDS in particular an important factor for prioritizing TA, the staff is not proposing that this be the sole criterion.

28

The term “encouraged” carries a specific meaning in the data standards initiatives. “Encouraged” coverage is desirable and may be strongly promoted by the staff, but is not a requirement for SDDS subscription or GDDS participation.

29

The staff also could provide hyperlinks in STA’s Data Quality Reference Site to relevant international oil data initiatives such as the Joint Oil Data Initiative and the Extractive Industries Transparency Initiative (EITI). Participation in the EITI would facilitate countries’ preparation of oil and gas metadata.

30

See the Fund’s Data Quality Reference Site at http://dsbb.imf.org/vgn/images/pdfs/dqrs_Genframework.pdf.

31

Supplement 2 to the Fifth Review of the Fund’s Data Standards Initiatives on the Data Quality Assessment Framework and Data Quality Program.

32

See the Supplement attached to this board paper.

33

An early pilot conducted in May 2005 of drafting GDDS metadata for two GDDS candidates using the DQAF structure suggests that countries will find the metadata easier to compose in the updated framework. With the DQAF indicators incorporated into the GDDS template tables used in the drafting process, prompts for the relevant information under the data, quality, integrity, and access dimensions, as well as the aspects of the comprehensive framework, are thus built into the templates. See Appendix II of the Supplement to this paper for the DQAF items under each GDDS dimension and aspect of the GDDS comprehensive framework.

34

See the listing of SDDS subscribers in Appendix I, which identifies the countries both subscribing to the SDDS and participating in the CCE.

35

A detailed progress report on the work program on FSIs was submitted in May 2005 for information of the Executive Board.

36

The quality of data available to the market also affects credit ratings. See, for example, Leonel Price, Standards and Codes—Their Impact on Sovereign Ratings, Fitch Ratings, July 2002, and John Chambers, The Importance of the IMF’s Work on Standards and Codes, Standard and Poor, October 2002.

37

National Bureau of Economic Research, Working Paper No. 9260; republished as IMF Working Paper WP/02/174. The paper presents measures of transparency and lack of transparency (opacity) and evidence that during the Asian and Russian crises, emerging markets funds withdrew more strongly from less transparent countries after controlling for other risk factors.

39

See, for example, Tara Vishwanath and Daniel Kaufmann, “Toward Transparency: New Approaches and Their Application to Financial Markets,” The World Bank Research Observer, vol. 16 no. 1 (Spring 2001) and McHale, John, “The Korean Currency Crisis, A Report on the Third Country Meeting of the NBER Project on Exchange Rate Crisis in Emerging Market Countries,” February 2000 (www.nber.org/crisis/korea_report.html).

40

Institute of International Finance, “Does Subscription to the IMF’s Special Data Dissemination Standard Lower a Country’s Credit Spread?” (Appendix D on Action Plan Proposals and Dialogue with the Private Sector), 2002, Washington, D.C.

43

A country’s participation in the GDDS is determined by the fulfillment of the following criteria: (1) commit to using the GDDS as a framework for statistical development (through sending a letter to the STA Director); (2) nominate a country GDDS coordinator; and (3) provide metadata and “plans for improvement” for publication on the DSBB.

44

The PARIS21 consortium’s founding institutions are the OECD, the World Bank, the United Nations, the IMF, and the European Commission; it also comprises a significant number of other major providers of statistical technical assistance worldwide, including several major bilateral and international agencies.

45

Regional GDDS projects exist for member countries of West African Economic and Monetary Union (WAEMU), Anglophone African countries, Lusophone African Countries, and Pacific Island countries and territories.

47

See Selected Decisions and Selected Documents of the International Monetary Fund, Article VIII, Section 5, Furnishing of Information as amended) and its Annex, the Scope and Operational Characteristics of the Special Data Dissemination Standard, as amended

Sixth Review of the Fund's Data Standards Initiatives
Author: International Monetary Fund