Second Review of the Special Data Dissemination Standard

Second Review of the Special Data Dissemination Standard

Abstract

Second Review of the Special Data Dissemination Standard

Executive Summary

This paper reviews the experience under the Special Data Dissemination Standard (the SDDS or the Standard), discusses issues related to the closing of the transition period at the end of 1998, and presents proposals for refinements of the SDDS. The background to development of the SDDS and experience since its establishment in March 1996 are outlined in Box 1. The proposals concern the requirements for international reserves and external debt, a shift to making hyperlinks1 from the Dissemination Standards Bulletin Board (DSBB) to country data sites an integral part of subscription, the approach to monitoring observance, and procedures for dealing with nonobservance. The paper also indicates the staff resource implications of the suggested work program. Finally, the paper briefly describes proposed changes to the structure of the General Data Dissemination System, consistent with those proposed for the SDDS, regarding international reserves and external debt.

Background on Development and Experience

  1. Preparation of the Fund’s Special Data Dissemination Standard (the SDDS or the Standard) began in the wake of economic turbulence in 1994/95. These events underscored the role that information deficiencies could play in contributing to market turmoil. Within many countries, there are codes—promulgated by accounting organizations, stock exchanges and so on—which outline the financial information that economic entities should disclose on a regular and timely basis to maintain their standing in the capital markets. As of 1994/95, there was no counterpail in the international domain concerning countries’ economic and financial data. In this vein, the SDDS was envisaged as providing a voluntary set of data dissemination standards, representing an effort to codify existing good practice, to which countries participating in international capital markets, or aspiring to do so, could subscribe.

  2. After wide-ranging consultations by staff with users and producers of data and feedback from the Executive Board, the SDDS was established in March 1996, encompassing four key dimensions of statistical practice: data (within which coverage, periodicity and timeliness), access, integrity and quality. A transition period, to close at end-1998, was established, during which time subscribing countries would make necessary changes to their data practices Under the SDDS, subscribing countries would provide more comprehensive, reliable, accessible, and timely economic and financial information. The SDDS was of course not expected by itself to prevent, economic crises. Nevertheless, the disturbances that began in Asia in 1997 have further underscored the role that information deficiencies can play and have indicated the need for refinements to the Standard. In particular, enhancements are seen as essential concerning the specifications for international reserves and external debt.

  3. The international community is now engaged in an effort to improve the architecture of the international monetary system. Much of this work involves translating to the international domain frameworks and codes of practice that help to support efficient markets within individual countries. The areas where such work is underway include banking, accounting standards, bankruptcy procedures, and fiscal and monetary transparency. The SDDS can be seen as an early contribution by the Fund. As part of the evolving SDDS process, the Fund has undertaken a wide-ranging process of consultation with countries and other international agencies, its first experiments with public discussion papers of Fund policy proposals, and with the use of an Internet website to communicate information on data standards. All of these activities provide useful lessons for future activities in the area of standards.

  4. There has been a large volume of SDDS-related activity over the 2½ years since the Standard’s inception. By end-November 1998,47 countries were subscribers, including major industrial countries and many emerging market economies. Given the target audience—countries participating in international capital markets or aspiring to do so—this represents a high subscription rate, consistent with expectations when the SDDS was first established. Encouraged by the SDDS, subscribers have introduced wide-ranging improvements in statistical practices, to the benefit of the countries and the international community. In September 1996, the Fund opened the Dissemination Standards Bulletin Board (DSBB) on the World Wide Web. This electronic bulletin board provides easily accessible information about the statistical practices of subscribers, available at the click of a mouse.

  5. From the beginning, it was clear that users of the DSBB would find it more useful if there were a direct link to actual economic and financial data, in addition to the information it provided about data practices. To that end, hyperlinks from the DSBB to country websites—with the data on the country site to correspond to the data described on the DSBB—were put in place, beginning April 1997. By November 1998, hyperlinks to country data sites were in place for 17 subscribers. Virtually all subscribers have indicated their intention to introduce hyperlinks. Usage of the DSBB has surged over time, from some 25 thousand hits in September 1996 to more than 130 thousand by October 1998, with the cumulative number of hits by October 1998 at 1.8 million. The goal for the period ahead is the further enhancement of the DSBB as a user-friendly way to access information concerning the SDDS, the statistical practices of subscribers, and actual economic and financial data for subscribers.

A substantial volume of work has been undertaken by countries to improve their statistical practices in line with the SDDS, underscoring that the SDDS is serving the intended purpose of encouraging improvements in statistical systems. Some results are already available, such as improved coverage and more frequent and timely dissemination of economic and financial data; other results are apparent in the introduction of more transparency in practices, such as calendars of release dates. In recent contacts with the staff, subscribers have shown a high degree of commitment to completing the tasks still needed to bring them into full observance of the Standard by the end of the transition period. While the transition period formally closes at the end of 1998, evidence that countries are observing the requirements of the Standard will continue to become available well into 1999, some only by mid-year or later.2

With respect to international reserves, experience has revealed the need for a more comprehensive and disaggregated treatment, including reserve-related liabilities, financial derivative positions, and other supplementary information that would cast further light on countries’ reserve positions and potential demands on reserves. Most of this information is already compiled by country authorities and what is required now is the commitment to make the data publicly available in a timely manner. This paper proposes decisions for Board consideration that would add the main components of international reserves, covering these items, to the specifications of the SDDS, with periodicity of one week and a one-week lag, together with a one-year transition period for implementation. A more detailed breakdown of international reserves, organized in the form of a template, would be prescribed for dissemination with monthly periodicity and a one-month lag.

With respect to external debt, recent events in Asia and elsewhere have underscored the international community’s need for more comprehensive, frequent, and timely debt data. However, in contrast to international reserves, for many countries the necessary data collection and compilation systems for external debt are not in place. Inevitably, establishing new data systems—especially for private sector external debt—is likely to be an arduous and lengthy process. On the other hand, a substantial amount of information on debt is available from existing creditor-side databases.

Against this background, the paper proposes two initiatives aimed at achieving the necessary improvements in the availability of data for external debt. One initiative—to meet as fast as possible the international community’s desire for better external debt data that is both more accessible and more timely—consists of efforts already underway to improve and better coordinate existing creditor-side databases, such as those of the BIS and OECD, together with information from the World Bank and the Fund.

As a second parallel initiative, the paper proposes to enhance the specifications of the SDDS concerning external debt to improve coverage, periodicity, and timeliness. Because this work is likely to involve building new statistical infrastructure for many countries, it will take time. The staff proposes to establish a three-year transition period for the dissemination of annual IIP data with two-quarter timeliness as prescribed. It also proposes to prescribe the periodicity for dissemination of data on external debt as quarterly, with the prescribed information to be disaggregated among debt of the general government, the monetary authority and banks, and other (nonfmancial public corporations and the private sector). Dissemination of forward-looking information on debt payments falling due would also be prescribed (quarterly for the first year ahead, annual for the second). Reflecting the agreed strategy for this area, the staff would propose to consult further with data users, countries and other international agencies on the transition periods for these changes on external debt. A subsequent paper would provide proposed decisions.

The DSBB has successfully met its initial objectives although, as indicated below, there are new challenges to be met. The usage figures have been rising sharply and feedback from users is positive. One key ingredient has been the move to direct hyperlinks to appropriately formatted country websites, thus making actual data and information about the data accessible, for those subscribers with hyperlinks in place. Staff contacts indicate a step up in usage of those country data sites with DSBB-hyperlinks. Against this background, the staff proposes to make the establishment of a hyperlink to a country website, containing data covered by the SDDS, a prescribed feature of the SDDS. The staff proposal envisages a transition period of one year for this to be put in place. Further, in the evolving world of Internet technology, the initial experience with the DSBB, while gratifying, does not provide a guarantee of future success. Accordingly, in addition to the proposal for mandatory hyperlinks to country data sites, work is underway, in consultation with private sector experts, on developing ways to enhance and market the DSBB in light of rapid changes in technologies and user requirements for data. This would include the possibility of further developing the DSBB into a user-friendly database system..

Concerning the monitoring of observance of the Standard, the staff proposes to implement monitoring of the data dimension and advance release calendars. Staff monitoring would be facilitated by the introduction of mandatory hyperlinks. Monitoring of the remaining components of the access dimension together with the integrity and quality dimensions would be on a self-declaration basis by subscribers; information provided to the staff in the context of Article IV consultations could play a helpful supplementary role. The paper also spells out for Board consideration the procedures for dealing with cases of nonobservance.

From the beginning, the work program associated with the SDDS has drawn intensively on staff resources and the work program has expanded over time. If the work is to continue as planned and the credibility of the Standard is to remain high, sufficient staff resources will need to be provided The paper envisages that eight additional staff would need to be deployed to work on SDDS-related issues in the Statistics Department. There will also be a need for additional resources for technical assistance and for work on these issues elsewhere in the Fund.

I. Introduction and Scope of the Review

1. When the Executive Board established the Special Data Dissemination Standard3 in March 1996, Directors emphasized that in the initial phase its implementation should be both flexible and evolutionary. Directors called for reviews of the SDDS, at the end of 1997 and the end of 1998, that would provide the opportunity to make adjustments that might be needed. The first review took place in December 1997. The Board also addressed SDDS-related issues in its July 1998 meeting on international standards and Fund surveillance and in its September 1998 discussion of data availability, dissemination, and data provision to the Fund.4 The conclusions of these discussions are summarized in Box 2.

Recent Board Discussions on the SDDS: A Summary of Key Conclusions

The First Review of the SDDS (December 1997)

  • Endorsed proposed procedures for modifying the SDDS.

  • Agreed that consideration be given to modifying the international reserves data category.

  • Decided modification to include indicators of financial soundness should await development of standards in relevant areas.

  • Decided that a more precise timetable for implementation of the IIP be agreed at the second review.

  • Agreed with preliminary proposals for dealing with nonobservance after the end of the transition period. Agreed that a refined proposal be discussed at second review.

  • Decided that the DSBB remain free to users.

International Standards and Fund Surveillance (July 1998)

  • Concluded that the Fund should monitor compliance in those areas of most direct concern to the Fund and where it had relevant technical expertise. These core areas would comprise data dissemination, monetary and financial policies, and fiscal transparency The most efficient way to monitor would be through the Article IV consultation process.

  • Differed in views on whether the Fund’s evaluation of members’ compliance with standards in core areas should be made public.

  • Noted that staffing implications were potentially significant. Staff should provide detailed estimates of resource implications.

Data Availability, Dissemination, and Provision to the Fund (September 1998)

  • Staff to take lead in developing guidelines on reserves, working closely with other international fora. Staff to develop template on reserves. DSBB metadata on reserves should be strengthened. Options on auditing to be elaborated, including prescribed provision of metadata.

  • Identified improvements in external debt data as a high priority. Staff to continue to work closely with other international agencies. Urged intensification of Bank/Fund efforts to help countries through training and technical assistance.

  • Endorsed role of Inter-Agency Task Force on Financial Statistics. Asked for periodic progress reports.

  • Asked staff to seek wide range of views on strengthening external debt.

  • Agreed that hyperlinks should be prescribed and looked forward to staff s proposals on this.

  • Endorsed inclusion of data on short-term foreign currency debt of central government, with same periodicity and timeliness as international reserves.

  • Agreed to aim for weekly data on reserves with one week lag.

  • Looked forward to amplification/elaboration of policy on nonobservance at time of second review.

  • Endorsed suggestions to look at ways to enhance the DSBB, including options to make it a user-friendly database system.

2. This paper provides input for the second review of the SDDS. A brief sketch of the background to the SDDS and its evolution is provided in the Executive Summary; more detailed information on the experience is contained in Appendix II. Section II presents the staff proposals for modifications of the SDDS in various areas, including refinements as regards international reserves, external debt, and the DSBB; reviews the experience during the transition period and proposes some temporary additional flexibility for subscribers; and discusses monitoring of observance and procedures for dealing with nonobservance. It also briefly discusses the experience with the SDDS in the context of the broader work program on standards and Fund surveillance. Section III makes proposals for modifications to the General Data Dissemination System (GDDS) stemming from the proposed changes to the data dimension of the SDDS. Section IV briefly deals with staff resource issues. Section V presents issues for discussion.

II. Proposals for Refinement of the SDDS and Future Directions

A. Refining the SDDS

3. In its September meeting, the Board called for further discussions on strengthening the data dimension of the SDDS in the areas of reserves and external debt. The staff s proposals are set out below. This sub-section also briefly discusses the incorporation of indicators of financial soundness into the SDDS—where little progress has been made thus far.

International reserves

4. In September, most Directors supported expanding the prescribed SDDS category for international reserves to include reserve-related liabilities, financial derivative positions, and other supplementary information that would cast further light on countries’ net international reserves positions.5 Directors noted that disclosure to the public of reserves data on a weekly basis—and in some cases daily—was emerging as best practice and generally agreed that the SDDS should aim at the dissemination of reserves information weekly, with a lag of one week or less. They urged the staff to work quickly to develop a template for the disclosure to the public of data on reserves and related items. Most Directors observed that supplementary data on the short-term foreign currency debt of the central government should be prescribed, with the same periodicity and timeliness as that for international reserves.

5. Since the September meeting, the staff has worked, in the context of the Interdepartmental Task Force on External Debt and Reserves, to develop a template for the disclosure of reserves data to the public. The latest draft of this template is attached as Appendix III to this paper. The staff has worked closely with a working group established by the Euro-currency Standing Committee (ECSC) of the G-10 central banks to examine shortcomings in available data that were identified in the wake of the Asian financial crisis. This group presented its final report, Enhancing Transparency Regarding the Monetary Authorities’ Foreign Currency Positions, on September 29. The recommendations contained in this report, as in the G-22 Report on Transparency and Accountability, with respect to the coverage of reserves data, generally complement and reinforce the statistical directions envisaged at the September Board meeting (see Box 3), The ECSC report envisaged follow-up work by a group of technical experts to refine and test the reserves disclosure template. Fund staff participated in this group, which made proposals that would align the template more closely with that developed by the staff. Nevertheless, it still remains to be seen whether complete alignment will be feasible. The detailed template is being circulated to the ECSC and G-10 Governors in advance of their December 1998 meeting.

Disclosure of International Reserves: Main Differences between the ECSC and Staff Proposals

The staff has worked closely with a working group, constituted by the Euro-Currency Standing Committee (ECSC) of G-10 central banks, to refine a template for the disclosure of reserves and related items. The final version of the ECSC template, produced in late November, is very similar in most respects to the template proposed by the staff (Appendix III). However, a few important differences remain. Most significantly:

  • Periodicity. The G-10 Governors have agreed that, as a first step, by June 1999 data on reserves and related items should be published monthly, with a one-month lag. The staff proposes weekly disclosure of a subset of key items, with a one-week lag, and monthly disclosure of the full template; the transition period would be through end-1999. Weekly publication was endorsed by the Board in September.

  • Derivatives with a maturity of over one year. The ECSC template calls for disclosure of reserve-related liabilities and related items with a maturity of up to one year. The staff proposal also includes forwards and futures contracts with a maturity of more than one year, on the grounds that these are generally subject to margin calls (in response to the depreciation of the exchange rate) that can create a drain on reserves at any time while the contract is outstanding.

  • Liabilities that are not settled in foreign currency. Most reserve-related liabilities are denominated and settled in foreign currency. However, other transactions that are priced on the basis of changes in the exchange rate may have macroeconomic effects that are essentially the same. For instance, in the absence of exchange controls, the macroeconomic effects of nondeliverable forwards and domestic securities that are indexed to the exchange rate are essentially the same as those of outright forwards and foreign-currency-denominated securities. The ECSC includes all transactions priced in relation to the exchange rate but not settled in foreign currency in its template as a memorandum item. The staff proposal is similar, except that it calls for disclosure of nondeliverable forwards and indexed securities as two separate memorandum items.

  • Treatment of claims on domestic residents. The ECSC template adopts a definition of reserve assets based on whether the assets are denominated in foreign currency. The staff template bases the definition of reserve assets mainly on whether they are claims on nonresidents. It is clear, in any event, that neither reliance on a currency-based definition nor a strict reliance on the residency criterion will yield a satisfactory understanding in all cases of the actual availability of resources to meet balance of payments financing needs

  • A residency-based approach has the disadvantage that it would permit countries to include claims denominated in domestic currency as reserve assets; such claims may not be fully available in situations of currency crisis. At present, few monetary authorities invest their international reserve holdings in domestic currency denominated assets and the instances in which such assets have caused problems are thus extremely rare. On the other hand, a currency-based approach has the disadvantage of including as reserves foreign currency denominated claims on residents. In recent months, the inclusion of such assets in reserves has contributed to misstatement of the actual availability of resources to meet balance of payments financing needs in some cases, since these claims were neither liquid nor readily available. There have been similar problems, albeit in fewer cases, with the availability of assets held abroad in foreign branches or subsidiaries of banks headquartered in the reporting country.

  • In the ECSC template, these issues are addressed by requesting that data on deposits included be broken down between those that are held with resident institutions, those that are held with offices overseas of institutions headquartered in the reporting country, those held with foreign monetary authorities, and other deposits. The staff template has adopted the same breakdown In line with previous interpretations under the fifth edition of the Balance of Payments Manual (BPM5), the staff has proposed that countries include in reserve assets only domestic deposits that are backed by a counterpart asset abroad that is readily available and under the control of the monetary authorities.

6. The staff proposals regarding international reserves (and also to some extent external debt) are a shift of approach for the SDDS. They represent an effort to advance international practice in the dissemination of data on reserves (and also external debt) in a direction viewed as extremely important by the Fund—and its partners elsewhere in the international community—and thus go beyond the approach adopted under the SDDS so far, namely the codification of existing best practice. This shift is motivated by two main factors: the vital importance of improving data availability in these two areas, as evidenced by recent economic and financial turmoil, and the Fund’s unique position as both a locus of expertise and a focal point for discussions on these two data categories. The effort to use the SDDS as a vehicle to support improvements beyond what is currently existing best practice is expected to be confined to areas that exhibit such factors.

7. The staffs proposals on a framework for assessing a country’s reserves adequacy are summarized in the template shown in Appendix III. The components that the staff proposes be prescribed in the SDDS for most frequent dissemination are identified in that appendix. The staff proposes that periodicity be set at weekly, with weekly timeliness, for these main components. For the full detail of the template, the staff proposes dissemination with a monthly periodicity and with a one-month lag. The staff is aware that these proposals on timeliness—which reflect the approach supported by the Board last September—go beyond those likely to be adopted, as a first step, by the G-10 and represent a step up from current practice for many countries. In part, the EC SC view reflects the desire of some to preserve confidentiality in reserve management and exchange market intervention. Nevertheless, the staff believes that there are compelling reasons to aim for very frequent and timely dissemination of data on international reserves and related items to the public, a view which has been reinforced by the crisis experience of recent months. The timely availability of such information is essential for informed market assessment and reaction. Large-scale shifts in capital markets can result in very rapid changes in reserves which would not be captured by the release of data with a monthly periodicity.6 Comprehensive, frequent, and timely release of information on reserves serves to demonstrate to the public that the authorities themselves are well informed about the status of reserves and provides an opportunity for national authorities to explain their policy stance and the reasons for it. Finally, these data are already generally available to the authorities, at least in provisional form, on a daily basis. What is required now is a decision to disclose the data, consistent with the general presumption that data should be disseminated promptly once they are available.

8. In addition to these proposals on data coverage, periodicity, and timeliness, the staff proposes that the following be prescribed practices in the Standard: (a) the prescribed components should be compiled in conformity with BPM5 and associated operational guidelines that would be prepared by the staff; (b) to provide assurance to users that reporting is on the basis of BPM5, subscribers would post on the DSBB detailed metadata on the type and usability of foreign assets held as official reserves and on the institutions that hold reserves; and, (c) to cast further light on the reserve assets data, subscribers would post on the DSBB metadata on the currency composition of reserves and the quality of counterparties. The metadata on the final currency composition of reserve assets would give an up-to-date indication of the proportion of reserve assets that were held in the currencies comprising the SDR basket. The metadata on the quality of counterparties in reserve assets would quantify the share in foreign exchange reserves of deposits and securities rated A or better and the share of deposits and securities rated less than A through BBB-.

9. The staff proposes a one year transition period, ending on December 31,1999, for the modified reserves data category. During the early part of this period, the staff would work, in collaboration with other interested groups, on developing operational guidelines that would support the new SDDS prescriptions.7

External debt and the international investment position

10. In September, Directors welcomed the staffs plan to consult widely on external debt issues. They agreed that there was a need to prescribe the dissemination of frequent and timely data on external debt—in particular for short-term debt and debt of the private sector. Directors urged that consultations begin immediately on prescribing that UP statistics be developed in two stages: first the compilation and dissemination of external debt statistics, within the IIP framework, and subsequently the development of annual data on the remaining IIP components.8

Responses to the consultation paper on external debt

11. A consultation paper on external debt and related issues was posted on the DSBB on October 6, 1998, inviting comments. In the same vein, the staff has had the benefit of the views of members of the IMF Balance of Payments Statistics Committee, which met during October 21-23, 1998, and the Inter-Agency Task Force on Finance Statistics, which met during October 29-30.9

12. Noting that recent financial crises demonstrated the pressing need for more timely and comprehensive data on external debt, data users responding to the consultation paper thought that much more could be done to fill what they regard as major data gaps. The following points were made:

  • Existing sources do not provide timely or comprehensive data on nonbank private sector external debt or short-term external debt.

  • Annual data on the country’s IIP provided with a two-quarter lag was not sufficiently timely, and quarterly data with a one-quarter lag were needed to monitor more rapidly changing developments

  • Timely, forward-looking information on total debt payments falling due in the shortterm is crucial.

  • While application of the residency criterion remains important for consistency of external debt data with other macroeconomic data systems, supplementary data on foreign currency debt and debt indexed to foreign currency was also considered useful for vulnerability analysis.

  • The most useful breakdown of external debt information was by sector, then by instrument and functional category.

  • Different dissemination schedules for different components of external debt would be acceptable.

  • Some judged that data on total short-term external borrowing and deposits, including for the private sector, should, if possible, be disseminated on a monthly basis with no more than a two-month lag.

  • Information on the exposure of major international institutional investors, especially in emerging markets, should be made transparent and be available on a timely basis.

13. Data compilers acknowledged the emerging needs of data users, especially in light of recent global financial developments, and, in principle, supported the objective of promoting greater data transparency. However, they were concerned about resource implications and reporting burdens. Some noted the conflict between, on one hand, the general move in recent years toward elimination of exchange restrictions and liberalization of capital account transactions and, on the other hand, the thrust for improved data on these transactions.10 Compilers expressed concerns about additional difficulties they would face if data coverage were expanded and increased periodicity and greater timeliness were required. Some advocated improving the quality of existing data before undertaking the contemplated initiatives on coverage and periodicity. Several compilers also cited technical difficulties, including:

  • Identifying external debt on a residency basis, especially with regard to securities, when the residency of owners/holders often is unclear;

  • Capturing the many forms of private nonbank external debt on a comprehensive basis;

  • Recording complex transactions such as repurchase agreements and financial derivatives;

  • Adjusting data collection methods to compile external debt statistics on a remaining-maturity basis, rather than the original maturity basis called for in the IIP framework;

  • Compiling data on countries’ foreign currency exposures under the residency concept of the IIP framework; and

  • Gaining compliance with additional data reporting by official and private sector entities by putting in place an appropriate reporting framework or otherwise achieving adequate reporting despite the absence of strong legal and regulatory frameworks.

14. With regard to countries that already compile annual IIP data, some compilers indicated that it would be feasible to produce a preliminary IIP on a quarterly basis with a one-quarter lag, provided that the full reconciliation of stock and flow data, as required in a complete IIP, would still be undertaken on an annual basis. Compilers also indicated that having available quarterly international banking statistics and international securities data published by the Bank for International Settlements (BIS) would be useful to validate IIP data and could facilitate the compilation of a quarterly IIP. More generally, it was felt that the development of debtor source data would take a long time.

Staff proposals on external debt

15. Against this background, this paper proposes two initiatives to improve the availability of data on external debt. One initiative—to meet quickly the international community’s need for better and more accessible external debt data—entails an effort that is already underway to improve and better coordinate existing creditor-side databases, such as those of the BIS and OECD, together with information from the Fund and World Bank (see next paragraph). This initiative, while desirable, is not a substitute for the important but necessarily long-term task of improving external debt information compiled from national sources. In this vein, the second initiative is to enhance the specifications of the SDDS concerning external debt to improve coverage, frequency, and timeliness, to provide a vehicle for improvements in countries' external debt systems. Because this work program involves strengthening the statistical infrastructure for many countries and involves the cooperation of several international organizations, it will take considerable time.

16. With respect to the first initiative, at a meeting of the Inter-Agency Task Force on Finance Statistics in late October, the Task Force called on creditor countries and the various agencies to improve the frequency and timeliness of their reports. An initiative was taken to collaborate on the construction and publication of a database with quarterly data on selected components of external debt, mainly based on creditor source data supplied by the BIS (international banking and securities positions) and the OECD (data on official credits and nonbank export credits) and including related data from the World Bank (Brady bonds) and the Fund (international reserves). The tentative objective is to make the database available on the websites of the participating agencies beginning in February 1999, with data through September 1998. Although an improvement, this would still mean a lag of four to five months in the availability of creditor-side debt data. Once some experience has developed concerning this system, consideration could be given to establishing a hyperlink from the DSBB to a website containing the creditor-side information on external debt.

17. With respect to the second initiative, the staff proposes a transition period of three years, closing at end-2001, for dissemination of annual data for the HP as a prescribed category of the SDDS. Annual periodicity and two-quarter timeliness would be prescribed, with quarterly periodicity and one-quarter timeliness encouraged, as at present.

18. The experience of the last 18 months underscores the problems that can arise when external debt data are weak. While it will take time, improving country external debt information is vital for both countries and the international community. Accordingly, the staff proposes to introduce a separate data category for external debt in the SDDS. Within this category, the prescribed frequency for dissemination of data on external debt would be quarterly, with the information to disaggregate among debt of the general government, the monetary authorities and banks, and other (nonfinancial public corporations and the non-bank private sector).11 Prospective debt service obligations by quarter for the first year ahead and annually for the second year ahead would also be prescribed.12 The staff would consult further with data users, countries, and other agencies concerning transition periods for these debt items. These steps would be pushed forward at maximum speed and the staff would present proposals for Board consideration in the course of 1999.13 Subscribers would be encouraged, meanwhile, to disseminate components as appropriate and feasible.

Prudential-type bank indicators

19. The first review of the SDDS raised the issue of including prudential-type indicators in the coverage of the SDDS as a means of providing information on financial soundness. However, there was no agreed set of core indicators that best provided this information, and the Board concluded that more experience would be needed in monitoring financial systems before such a set of indicators could be identified. More importantly, though, it was recognized that present international accounting standards as well as loan classification and provisioning rules were not universally applied, so that data bearing on financial soundness in many countries were of generally poor quality and not amenable to easy interpretation or comparison across countries. Directors agreed, therefore, that modification of the SDDS to include indicators of financial soundness should await the development of standards for the disclosure of macro prudential data and should draw upon the work of other organizations, including the BIS.

20. While work is underway to develop standards and principles necessary for robust financial systems, it is likely to be still some time before suitable standards that could be considered for incorporation into the SDDS are developed. It is not feasible, therefore, to include prudential-type indicators into the SDDS at this stage. Staff propose, however, to provide a paper to the Executive Board on progress in developing macro prudential indicators sometime after the Spring meetings. That paper could consider the advantages and disadvantages of incorporating indicators likely to be available in the near future into the SDDS.

B. The Transition Period and New Subscriptions

21. In establishing the SDDS, the Executive Board made provision for a transition period to provide time for subscribers to make the necessary changes in statistical practices. A substantial volume of work has been undertaken by subscribers to implement their transition plans under the SDDS. The staff assessment as of late-November 1998 is that much has been done but also that much remains to be done. In recent contacts with the staff, representatives of subscribing countries have expressed confidence that they will meet all transition plan commitments. Nevertheless, because the Standard is a new and experimental venture, and reflecting the volume of work that is underway, there is of course a possibility that some transition plans may be implemented late.

22. For the data dimension, the evidence of observance or nonobservance for some categories—mainly those with quarterly periodicity—will not be clear until well into 1999.14 If countries use all the time available to them under the transition period, data that fully met the Standard’s prescriptions would begin to be disseminated at various times in 1999, depending on the periodicity and timeliness prescribed for the data category in question. The following table illustrates the various timeframes by which observance will be tested for different data categories.

Table 1.

The Special Data Dissemination Standard: Reference Periods and Deadlines for the First Release of Data after the End of the Transition Period

article image

The reference periods and release dates shown in the table are the latest permitted after the end of the transition period, unless a subscriber is taking a flexibility option for that data category. In instances where a subscriber disseminates data with a higher periodicity or on a more timely basis than that prescribed by the SDDS, correspondingly earlier deadlines would be observed. Saturdays and Sundays are assumed to be non-working days. However, as a general rule, when a release date falls on a non-working day in a subscribing country, the next working day then becomes the applicable release date.

February 28 falls on a Sunday.

For the General government operations/Public sector operations data category, the relevant fiscal year will end on one of the following dates: September 30 1998, December 31 1998, March 31 1999, or June 30 1999. Data must be published within six months after the end of the fiscal year applicable for each subscriber.

Data as measured on this date (that is, stock data).

The SDDS does not prescribe a timeliness for this data category.

23. Feedback from officials in subscribing countries indicates that bringing statistical practices into conformance with the requirements of the SDDS is an exercise that requires major efforts on many fronts. In some countries, especially those affected by recent economic turbulence, the effort has been hindered by budget adjustments for statistical agencies. Overall, it is likely that the volume of work and the required extent of coordination across agencies are of sufficient complexity that some countries may experience difficulty in completing all of their transition plan commitments on time, despite substantial good faith efforts to do so.

24. An additional complexity is that, for those subscribers that participate in EMU,15 the EMU process raises issues and some potential complications. As EMU unfolds, the focus of macroeconomic analysis may shift over time toward EMU-wide macroeconomic aggregates, away from country specific information. For example, with monetary policy established at the European Central Bank, analysis will inevitably shift toward EMU-wide monetary aggregates. Over time, some data categories may be affected in terms of coverage—initially the categories most likely to be affected are monetary statistics and international reserves Eventually, as markets become more integrated and new settlement arrangements affect estimation of national statistics, other data categories may be affected, in particular, merchandise trade and the balance of payments. The staff is consulting with European counterparts on how to approach these issues and in the near-term is preparing adjustments to DSBB pages for those countries and categories where dissemination practices are likely to change in the near-term.

25. There are some differences between the timeliness prescribed in the SDDS and that called for under EU regulations. For example, the SDDS prescribes monthly industrial production data disseminated 42 calendar days after the end of the reference month, while EU regulations now set a timeliness of one month plus 15 calendar days. Similarly, the Standard prescribes timeliness for dissemination of producer price data of one month following the reference month, compared with one month plus 5 calendar days under EU regulations. Similar issues might arise with respect to financial statistics. In addition, decisions are now being taken by EMU participants that may temporarily affect the timeliness of dissemination of some other data in the coming period.

26. In order to smooth the transition process for all subscribers—while not diluting the Standard—the staff proposes that subscribers be permitted one additional flexibility option, to expire at the close of 1999. A flexibility option allows a subscriber to disseminate data with a lower frequency (periodicity) or with a greater lag (timeliness), or both, than prescribed for the data category under the SDDS. The same limitations would apply as for the current two permanent flexibility options available to subscribers, e.g. these options cannot be used for reserves. This approach is designed to provide some additional temporary margin of flexibility, especially in light of the uncertainties just mentioned, as subscribers bring their practices into accordance with the Standard’s requirements.

New subscriptions

27. New subscriptions will of course continue to be encouraged over time. Indeed, several such subscriptions may emerge in coming months. It is also hoped that participation in the GDDS will provide a number of countries with a vehicle to improve statistical practices in preparation for the more demanding SDDS requirements. The experience indicates that a fairly intensive collaborative effort between the staff and the subscribing country may be necessary before a full determination can be made as to where the country’s practices stand with respect to SDDS prescriptions.

28. The staff proposes that member countries that wish to subscribe to the Standard after the transition period would communicate this informally to the staff and, at that time, undertake to provide the necessary metadata other than the summary methodologies to the staff as soon as possible. The staff would work with the member to determine where its practices stand with respect to the Standard as well as to identify any changes in practices that would be required. If no changes were needed, the member would inform the Secretary of its subscription and the Fund’s public identification of the subscription would be made through the posting of the metadata on the DSBB almost immediately. Of course, a member could make known publicly its intent to improve its data and dissemination practices with the goal of subscription to the SDDS. For members where changes were required, the member would, after the necessary changes had been discussed with the staff and implemented, inform the Secretary of its subscription. Again, the Fund’s public identification of the member’s subscription would be made through the posting of the metadata on the DSBB. The staff also proposes that new subscribers would need to provide to the staff the summary descriptions of methodology called for under the SDDS within three months of the posting of the member’s metadata on the DSBB.

C. Observance and Monitoring

29. This subsection presents proposals for dealing with cases of possible nonobservance of the SDDS that incorporate the views expressed by Executive Directors at the time of the First Review. It also makes some preliminary proposals on monitoring observance.

A framework for dealing with observance

30. At the first review, Directors broadly supported the staffs preliminary proposals to deal with possible nonobservance by a subscriber to the SDDS after the transition period, characterized by a two-track approach to resolving nonobservance cases, distinguishing between deviations from practices prescribed by the SDDS and those relating to the provision of accurate and timely metadata. Directors requested that the staff present a refined proposal for dealing with possible nonobservance, which would reflect their comments and might include indicative timeframes, for consideration at the second review.

31. A framework for dealing with nonobservance should have two key objectives:(i) maintaining the credibility of the metadata that are presented on the DSBB; and, (ii) maintaining a high level of transparency Credibility requires that cases of nonobservance need to be dealt with quickly with the ultimate end of rectifying inaccuracies, identifying subscribers not in observance, and, in extreme cases, removing metadata from the DSBB. While there is obviously a need for a graduated process, with a series of well-defined steps, maintaining the credibility of the DSBB requires that the information it contains be both accurate and up-to-date. Transparency is needed to ensure that users of the DSBB are aware of the steps involved.

32. The staff proposes the following series of steps for dealing with nonobservance. The time lapses described for individual steps would be indicative only. The staff would exercise judgement and, while generally seeking to follow these indicative time periods, would tailor them to the specific situation being addressed:

  • First, a deviation of any kind from the requirements of the SDDS would be brought immediately to the attention of the SDDS country coordinator and in the case of an oversight or technical slippage could presumably be resolved within a few days.

  • Subsequent steps would follow a graduated approach that would have increasingly serious implications and that would differentiate between minor and serious breaches. In the case of possible nonobservance of the practices prescribed for the data dimension or the advance release calendar element of the access dimension of the SDDS, which would be the most likely to arise, the staff would try to resolve the issue with the SDDS country coordinator, and it could be expected that most issues would be resolved at this stage. If a satisfactory solution could not be found within five working days,16 the Executive Director concerned would be notified. If the problem remained unresolved ten working days after notification of the Executive Director, the staff would bring the issue to the attention of management with a proposal that a note be placed on the DSBB transparently describing the problem that had arisen and indicating the authorities’ response and the efforts underway to remedy it. Management would bring the matter to the attention of the subscriber’s Governor for the Fund and the Executive Board would be informed of management’s initiative. If the issue was still not resolved within a further 10 working days, the issue would be referred to the Executive Board. It may take a decision that the subscriber is not in observance of its undertaking under the SDDS, and that a notice to that effect would be posted on the DSBB. The staff believes that it would be preferable to continue to show the subscriber’s metadata on the DSBB but the DSBB should indicate clearly that the Board had determined that the subscriber was no longer in observance of the Standard and the nature of the situation. If, after a reasonable period following the determination by the Executive Board of a subscriber’s nonobservance of its undertakings under the SDDS and the posting of that determination on the DSBB, the subscriber persisted in its failure to observe any of its undertakings under the SDDS, the Board may decide to delete the metadata of that subscriber from the DSBB.

  • Each quarter, subscribers would be required to certify the accuracy of the metadata posted on the DSBB or, if the metadata were inaccurate, to provide the staff with revised metadata on a timely basis.17 The date of last certification of the metadata would be posted prominently on the DSBB. In the event that a subscriber did not meet this certification requirement for two successive certification dates, the staff would enlist the support of the Executive Director in resolving the issue. If the issue remained unresolved after 10 working days, the staff would inform management. The sequence of subsequent steps would follow that described above for the data and access dimensions. Between certification dates, there might be situations where the staff came to learn, including through the Article IV consultation process, of changes in practices that affected the accuracy of metadata posted on the DSBB. In such situations, the staff would, after informing the SDDS coordinator, post a note on the DSBB indicating that the metadata in question were in the process of being updated. While the staff would work with the SDDS coordinator to ensure that the affected metadata were amended expeditiously, subscribers would be required, in any event, to provide the revised metadata at the time of the next quarterly certification. Failure to do so would set in train the steps outlined immediately above.

Monitoring observance

33. At the July discussion of the general topic of international standards and Fund surveillance, Directors concluded that the Fund should monitor compliance with standards in those areas of most direct concern to the Fund and where it had the relevant technical expertise. They acknowledged the practical, resource-related, and conceptual difficulties in monitoring compliance with standards and suggested that the Fund’s approach to monitoring was likely to evolve over time. They noted that it would be easier to monitor compliance in areas where standards provided a clear set of guidelines for actual practice.

34. In discussions leading up to the establishment of the SDDS, it was anticipated that users of the DSBB would likely play a key role in assessing subscribers’ practices vis-a-vis the Standard. However, the extent to which users are monitoring subscribers’ compliance with the Standard seems to be limited. In fact, they may well be relying on the Fund to do so. Against this background, there are compelling reasons for the Fund to take an active role in monitoring observance of the SDDS. Nevertheless, the staff will also continue to seek ways to facilitate and encourage feedback and scrutiny from users of the DSBB.

35. The requirements for the data dimension of the SDDS are objective and unambiguous and therefore amenable to monitoring in a conceptually easy way by the staff. For the access dimension, the element relating to advance release calendars is also amenable to monitoring, albeit with a little more difficulty. The remaining element of the access dimension refers to simultaneous release and prescribes that data be released to all interested parties at the same time. Monitoring this element and the dimensions of integrity and quality—where the issues might not be so clear cut—would be more difficult. Self-declaration by subscribers that practices accord with the requirements of the Standard must be central in these areas.

The mechanics of monitoring

36. The staff could undertake monitoring of observance of these elements in a number of ways:18 (i) subscribers could be required to provide the staff, on an ongoing basis, with documentary evidence, such as press releases, that the Standard was being followed. For the increasing number of cases where dissemination of releases is through the Internet, the staff would need to check the relevant web sites on an ongoing basis; (ii) the staff could develop a network of correspondents in subscribing countries that would continuously report on country practices; or, (iii) if hyperlinks are made a prescribed part of the SDDS—as proposed elsewhere in this paper—they would provide a unique tool to support monitoring of the data dimension. For advance release calendars, the Standard prescribes that subscribers disseminate calendars themselves and also provide the staff with the necessary information for posting on the DSBB. The staff would “monitor” observance of this element by checking the release dates provided for posting on the DSBB to ensure that they were consistent with the prescriptions of the SDDS and with the posted advance release calendars.

37. The first and second options would be time consuming and resource intensive for all parties. The third option—based on the use of prescribed hyperlinks, a step that is desirable in its own right—would have more limited resource requirements. It is expected that staff monitoring of observance through prescribed hyperlinks would require an additional four staff in the Statistics Department on a continuing basis. The staff favors the third option.

A proposal for monitoring

38. On the basis of the above, the staff makes the following proposals on monitoring for Board consideration:

  • The establishment by subscribing countries of national summary data pages on the Internet, to which the DSBB would be hyperlinked, would be prescribed.

  • Current subscribers and those that subscribed before the end of the transition period would, if they had not done so already, need to establish national summary data pages by December 31, 1999. These pages would be updated on each occasion of release of data covered by the SDDS.

  • Members who subscribe in 1999 would need to develop a national summary data page by the end of the year. Members who subscribe from 2000 onwards would need to have developed a national summary data page by the time of subscription.

  • The staff would continuously monitor observance of the data dimension and the advance release calendars element of the access dimension, beginning from end-March 1999, although such monitoring would only become fully effective from end-1999 when hyperlinks were present for all subscribers.

  • Subscribers would be required to certify the accuracy of all of the metadata posted on the DSBB on a regular quarterly basis, and the date of last certification would be posted on the DSBB.19 This would be equivalent to “self-monitoring” of the accuracy of metadata by subscribers and would allow some element of public scrutiny that would pave the way for a more active monitoring role by users. This regular quarterly certification process would begin in the first quarter of 1999. Under this process, subscribers would be required to notify the staff, within three working days of the end of the calendar quarter, that either: (i) all of the metadata posted on the DSBB were fully accurate; or, (ii) certain metadata were incorrect. In the latter case, subscribers would be required to provide corrected metadata within a further five working days.

39. Especially in the initial phase, monitoring would be conducted on a pragmatic and flexible basis. Efficient and continuous monitoring in the fashion sketched here would be greatly facilitated if hyperlinks were in place for all subscribers, slated to be the case by end-1999.

40. In the interests of transparency, it would be important to keep users of the DSBB apprised of each subscriber’s position with respect to the SDDS. This could be done as follows:

  • The latest date on which a subscriber had certified the metadata would be indicated on each of the metadata pages on the DSBB

  • Following a determination by the Executive Board that a subscriber was no longer in observance of the data or the access dimensions of the SDDS, or had persistently failed to provide accurate and timely metadata, this determination would be posted in a prominent location on the DSBB. If, after a reasonable period, a subscriber persisted in its failure to observe any of its undertakings under the SDDS, the Executive Board may decide to delete the metadata of that subscriber from the DSBB.

  • Once each year, the staff would prepare a report for the Board on subscriber status during the year with respect to the practices covered by the SDDS. This report would also be posted on the DSBB.

D. Enhancing the DSBB

41. The use of the DSBB has risen sharply over time and feedback from users has become increasingly positive. The number of hits on the website from unique hosts—that is, excluding repeat hits from users accessing the website on multiple occasions—rose from 2,731 in the initial month (September 1996) to 19,996 by October 1998. The total number of hits, including those from repeat users, rose from 25,343 in October 1996 to 130,006 in September 1998. Total hits over the entire period through October 1998 amount to about million.

42. Hyperlinks to appropriately formatted country websites containing economic and financial data are now in place for 17 subscribers.20 Most of the remaining subscribers plan to introduce such hyperlinks. The staff proposes that such hyperlinks be made a prescribed feature of the SDDS for subscribers, with a transition period of one year, closing at the end of 1999.

43. In September 1998, Directors endorsed the staffs suggestion to engage consultants to advise on ways to enhance further the DSBB in light of rapid prospective changes in technologies and market requirements for data, including the possibility of further developing the DSBB into a user-friendly database system. Accordingly, the staff is in the process of engaging consultants to advise on means and methods for enhancing the DSBB; a preliminary report on various options, costs, and impacts is planned for the Spring of 1999 to be followed by a final report later in the year. Any fundamental changes to the nature of the DSBB would first be presented to the Board for its consideration. The consultants will be asked to advise on three areas: (i) the marketing of the DSBB through the Internet and other means to ensure that it receives maximum exposure to market participants; (ii) the design and implementation of changes to the web site to enhance its usefulness, through simplification of presentation and the use of hyperlinks to background information; and, (iii) the possibility of using the DSBB to develop a user-friendly database system allowing market participants access to large volumes of data—in collaboration, as appropriate, with private sector information services.

E. The Consultation Process

44. The development of the SDDS and subsequently the GDDS has involved a continuing and wide-ranging process of consultation with data producers in countries; statistical experts in international agencies, data users in the financial markets and elsewhere, and private analysts and experts. The collaborative discussion has had to extend beyond the usual counterparts of Fund staff in finance ministries and central banks to reach national experts, outside analysts and researchers, financial market participants and others. The process has inevitably been time consuming and resource intensive, but has been essential to the achievement of worthwhile results. Box 4 reviews the experience.

Consultation with Outside Partners in Work on the SDDS

The work on data dissemination standards began in early 1995. With respect to the SDDS, there have been the following phases of consultation:

  • Initially (April to September 1995), consultation on the design of the SDDS took place via Board papers, generating feedback through Executive Directors (from finance ministries, central banks, and, to a limited extent, statistical agencies), leading to follow-up papers.

  • The second phase involved a round of missions to 24 countries during October to December 1995, for discussions both with producers of economic and financial statistics (typically economics and llnancc ministries, central banks, and statistical agencies) and also with private sector representatives (banks, other financial houses, credit rating agencies, think-tanks, economic analysts, and others). The staff corresponded with another 42 countries on the formulation of the SDDS, on the basis of a detailed questionnaire. The discussions and contact with these parties provided vital information on the needs of data users and the perspectives of data compilers.

  • Phase three (February 1996) was a discussion draft paper released to the media, made available on the Internet and sent for comment to a large number of market participants and other interested parties. This was the Fund’s first experiment with a public discussion paper on a policy matter, seeking outside input before firm decisions were made. The responses were summarized for the Board.

  • The fourth phase comprised a further round of Board papers, followed by approval and implementation of the SDDS in March 1996.

  • Once the SDDS was approved by the Board, letters inviting subscription were sent. 53 countries sent teams to regional seminars that were held in Asia, Europe, and Latin America during June-August 1995. At these seminars, Fund staff guided country practitioners through the requirements of the SDDS, the modalities of subscription, and what would be involved in the preparation of country metadata for inclusion on the Dissemination Standards Bulletin Board (DSBB), that was slaled to open in September 1996.

  • The Executive Board was given a dress rehearsal of the DSBB before it opened to the public, The DSBB opened on the Internet in mid-September 1996 and at that time contained metadata for 19 SDDS subscribers. The opening was publicized with roadshows in London, Singapore, and Tokyo.

  • Once subscription began and the DSBB opened, there were continuing regular contacts between staff and officials involved in SDDS-related activities from the country side, with other international agencies and with DSBB users. A voluntary registration system was established for DSBB users tn June 1997, to provide information on the demographics of users and to facilitate continued dialogue with them.

  • In April 1997, hyperlinks were introduced from the DSBB to country websites containing economic and financial data formatted consistently with SDDS requirements, By December 1998, hyperlinks were in place for 17 subscribers and virtually all of the remaining 30 subscribers have indicated their intention to introduce hyperlinks.

  • In preparation for the first review of the SDDS in December 1997, staff sent a questionnaire to subscribers. The responses provided input for preparation of the review paper and were summarized in that paper.

  • Following developments in the Asian crisis, efforts began to strengthen the SDDS in the areas of reserves and external debt. The issues have been discussed in Board papers, in regular contacts with country officials, and with other official agencies. In addition, in order to solicit feedback from data users and compilers, a consultation paper on SDDS proposals concerning international reserves was placed on the Internet in May 1998. A further consultation paper—on external debt—was placed on the Internet for reaction in October 1998.

III. Modifying the General Data Dissemination System

45. On December 19, 1997, the Executive Board approved the establishment of the GDDS Implementation is proceeding broadly on schedule. In February 1998, the Managing Director wrote to member countries that had not subscribed to the SDDS, providing basic documentation on the GDDS and inviting countries to nominate a coordinator who would work with Fund staff. A preliminary version of the module of the Guide to the Data Dissemination Standards that deals with the GDDS has been used as the primary documentation for regional seminars that have been held in mainland Asia and Africa, as well as for work on the development of GDDS metadata for four piiot countries (Bangladesh, Cameroon, Cote d’Ivoire, and Fiji). On the basis of the work in the regional seminars and in pilot countries, the Guide is being revised to serve the remaining six regional seminars that will be presented in 1999 and to assist the continuing pilot country work.

46. The staff believes that it is useful to review the data coverage of the GDDS in light of the modifications that have been proposed for the SDDS. Although the SDDS and GDDS differ in many respects, there is broad similarity in the data coverage so that events that require changes in the SDDS suggest a need for similar modifications to the GDDS. The principal areas affected by changes in the SDDS are international reserves and external debt. With regard to international reserves, the GDDS that was approved by the Executive Board included as a core indicator gross official reserves denominated in U.S. dollars. Reserve-related liabilities were included as an encouraged component, and no explicit mention was made regarding positions in financial derivatives that may affect reserves. In order to reflect the growing consensus that reserve-related liabilities and derivative positions play a central role in the identification of a country’s reserves position, it is proposed that the GDDS core indicators for international reserves be expanded to include all items in the proposed SDDS template There would be no encouraged components in this data category. As the GDDS emphasizes improvements over time in data quality and data dissemination practices, the impact of this change would be to encourage GDDS participating countries to attach equal weight to the simultaneous improvement of data for gross reserves and reserve-related liabilities and derivatives.

47. External debt data are presently covered in the GDDS under both data categories/indicators and comprehensive frameworks relating to the balance of payments.21 With respect to the former, it is proposed that external debt data for the whole economy become a core indicator and that external debt service remain as an encouraged category. For the comprehensive framework, it is proposed that, for countries where external debt data is of particular analytical or policy importance, balance of payments and disaggregated external debt data be accorded equal priority. The structure of the disaggregated external debt data recommended for the GDDS would follow the recommendations elsewhere in this paper, with the understanding that this structure would be a goal to be achieved over time.

48. A revised version of The General Data Dissemination System, a concise statement of the GDDS which appeared as Annex V to SM/97/275, will be issued as a supplement to this paper. Although there are some changes in terminology aimed at bringing further clarity to the structure of the data dimension, no changes in substance are proposed other than those relating to international reserves and external debt.

IV. Resource Considerations

49. The staff has made a number of proposals with significant resource implications. This section deals with the resource implications for the Statistics Department; implications for other areas of the Fund would need to be assessed on an evolving basis The staff estimates that work relating to the methodological aspects and initial preparatory work for the improvement of external debt data among subscribers (and also the wider Fund membership) would require on average an additional two staff years on an ongoing basis. Once the actual work program of advice and technical assistance to support improvements in member country data systems for external debt came on stream, the resource implications would be much larger; this issue would need to be revisited at a later stage. Annual resource requirements, on an ongoing basis, for staff monitoring of the SDDS are estimated at four staff years, assuming that regularly updated hyperlinks are prescribed. Of course, since this is a new area, the latter estimate cannot be precise. The amount of staff resources will depend on the number of serious cases of SDDS subscribers failing to meet their commitments under the Standard, since the graduated process of staff response in such cases outlined earlier in the paper will be time consuming and resource intensive. The proposed work program for the DSBB might also have significant additional resource implications, which are difficult to quantify now.

50. Redeployment of existing resources and the elimination of low-priority activities, supplemented by limited new resources, have enabled the staff to make substantial progress on the operational implementation of the SDDS since its inception. However, the work program for the SDDS has expanded over time in response to Board feedback and requests and the ability to provide staffing to maintain and refine a credible Standard has been largely exhausted. While some staff resources presently devoted to the SDDS will be freed up by about mid-1999 after issues related to the transition period have been addressed, these resources have been earmarked for work on the implementation of the GDDS. On balance, therefore, an additional requirement of eight staff—excluding those providing technical assistance—on a continuing basis seems to be needed to enable the Statistics Department to implement the work program outlined in this paper. If, as has been the case thus far, the work requirements expand in unanticipated ways, this estimate could well prove to be on the conservative side.

V. Issues for Discussion

51. In discussing this paper, Directors may wish to review the overall progress on the SDDS, to discuss issues related to the closing of the transition period, and to respond to the staff proposals for refinements to the Standard’s specifications concerning reserves and external debt as well as the proposals regarding monitoring/observance and the DSBB.

Overall progress

52. Are Executive Directors broadly satisfied with the way the SDDS is evolving? Do they have comments on its performance as a vehicle for encouraging increased emphasis on data matters and improvements in countries’ statistical systems and practices?

The transition period

53. A substantial volume of activity has been undertaken by countries to implement their transition plans under the SDDS and countries have indicated their confidence that these commitments will be satisfied. The evidence will not be clear until the second half of 1999 for some data categories. The staff will continue to provide timely information on the DSBB concerning the implementation of transition plans and expects to provide an overall assessment in the next SDDS review paper. The staff also proposes that subscribers be allowed to avail themselves of an additional temporary flexibility option through the end of 1999, to smooth the transition without diluting the Standard. Do Directors support the approach and the proposals?

Proposals for refinements

International reserves

54. With respect to international reserves, the staff has proposed a template (Appendix III) that would provide for comprehensive coverage of reserves and potential drains on reserves. Key items would be prescribed with weekly periodicity and timeliness and there would be monthly periodicity and timeliness for the remainder. There would be a one-year transition period, closing at end-1999. Do Executive Directors support the staff proposals in this area?

External debt and the international investment position

55. The staff proposes the specification of a three-year transition period for provision of IIP data as a prescribed data category. Do Executive Directors support this proposal?

56. With respect to external debt, the international community has indicated its need for more frequent and timely data. An effort is underway to improve and coordinate better the external debt information available from existing creditor-side databases, although time lags would remain considerable. Do Executive Directors see this effort as an appropriate and sufficient initiative to quickly improve the availability of information on external debt?

57. Improvements to data on external debt, compiled from country sources, are necessary and important but are likely to take a long time, because in many countries it requires substantial improvements of the statistical infrastructure. The staff proposes a new separate data category for external debt, with three main components (the general government, the monetary authorities and banks, and other), and further breakdowns including by maturity, prescribed with a quarterly periodicity and quarterly timeliness. The staff also proposes to prescribe provision of forward-looking data on debt service, quarterly for the first year ahead and annual for the second year. Subscribers would be encouraged meanwhile to disseminate components as appropriate and feasible. Further consultation will need to be undertaken before the staff can propose transition periods for these refinements. Do Executive Directors support these proposals?

Monitoring observance and procedures for nonobservance

58. The staff has outlined the approach it plans for monitoring observance and the procedures that would be followed in dealing with possible cases of nonobservance. Do Executive Directors support the proposals for dealing with nonobservance? What is their reaction to the suggested approach to monitoring observance?

The DSBB

59. With respect to the DSBB, the staff proposes that hyperlinks to appropriately formatted country data sites be made a prescribed feature of the Standard with a transition period of one year. Work is underway to further refine the DSBB in consultation with private sector experts. Do Executive Directors support making hyperlinks prescribed?

The GDDS

60. Do Executive Directors support the proposed modifications to the General Data Dissemination System (GDDS) in the areas of international reserves and external debt proposed by the staff?

Resource considerations

61. On resources, the staff estimates that an additional eight staff would be needed in the Statistics Department for work on SDDS-related matters in the period ahead, if the proposals presented in this paper are accepted. Would Executive Directors support the inclusion of the proposed new staff resources in the next forthcoming administrative budget round?

Next steps

62. Assuming Board approval of the proposals in this paper, the staff proposes to prepare appropriate documents for transmission to subscribers and potential subscribers in the near future When establishing the SDDS, the Executive Board called for reviews of the Standard at appropriate intervals. The staff proposes to send papers to the Board during 1999, as important issues arise that would benefit from Board feedback. The next overall review of the SDDS would take place by end-1999. Do Executive Directors support this approach?