Abstract
The paper reviews the developments in the last 12 years that have influenced the evolution of the IMF's General Data Dissemination System, leading to reforms to enhance its role. The GDDS itself is part of a broader IMF Data Standards Initiative launched in 1996 to help address macroeconomic data deficiencies, which contributed to the emerging economies' financial crisis during the early 1990s. The review takes stock of the experience with statistical technical assistance provided to member countries and the ongoing reforms, within and outside the IMF, to strengthen the GDDS. Such reforms are particularly relevant in the context of the ongoing economic and financial crisis, which once again underscores the role of statistics in guiding policymakers to strengthen defenses against future crises.
This paper benefited from comments from Artak Harutyunyan, Tom Morrison, Louis Venter, and Kim Zieschang. Comments on an earlier draft were received from participants in the International Statistical Conference in Durban, South Africa, during August 16–22, 2009.
I. Introduction
Following the Mexican financial and economic crisis during the early 1990s, the IMF launched a Data Standards Initiative (DSI), in part to help address the data deficiencies that could help prevent a recurrence of similar crises. The first component of the DSI was the establishment of the Special Data Dissemination Standard (SDDS) in March 1996 to guide member countries that had, or that might seek, access to international markets in providing their economic and financial data to the public. In December 1997, the IMF launched the GDDS (as in the case of the SDDS) after consultation with member countries and other international statistical agencies. The GDDS was intended to guide all other Fund member countries in strengthening their national data systems. Both the SDDS and the GDDS promote the dissemination of timely and comprehensive statistics and thereby contribute to the pursuit of sound macroeconomic policies and improved functioning of financial markets in support of the IMF’s mandate. There are currently 66 SDDS country subscribers and 97 GDDS participants—together comprising a total of about 88 percent of the IMF membership.
The GDDS is a framework for developing statistical systems comprehensively, emphasizing collaboration among the data-producing and disseminating agencies. Since its inception, a number of complementary international and regional developmental and statistical initiatives have been launched and have significantly expanded global awareness and the scope of statistics, with implications for the role of the GDDS. Moreover, through its various technical assistance (TA) programs in statistics, including several regional GDDS projects, the IMF has garnered useful hands-on experience to inform a comprehensive review of the GDDS, and statistical development more generally, after the first 10 years.
The recommendations of the GDDS review were approved by the IMF Executive Board in November 2008 and had been endorsed, via consultations, by a large number of member countries. The key recommendations comprise (a) greater emphasis on data dissemination and somewhat less focus on metadata; (b) more effective use of the plans for improvement to prioritize TA needs; and (c) better alignment of the GDDS with the SDDS, among other things, to make it easier for countries, where relevant, to graduate to the SDDS to secure better access to international capital markets. These, and other changes that have now been incorporated into the GDDS, should better align the framework with the subsequent international initiatives to enhance the streamlining of statistics into the poverty-reducing strategies with better monitoring of the MDGs and other objectives.
This paper provides a review of the developments in the last 12 years that have influenced the evolution of the GDDS, leading also to the reforms that have been introduced to enhance its role. These reforms are considered to be particularly relevant in the context of the ongoing economic and financial crisis, which once again underscores the role of statistics in guiding policy makers to strengthen defenses against future crises.
II. The GDDS
The GDDS uses a framework compatible with the SDDS to help countries that may not meet SDDS data coverage, periodicity, and timeliness requirements, but are focusing on building statistical capacity. Accordingly, rather than prescribing the monitorable data coverage, periodicity, and timeliness requirements of the SDDS, the GDDS provides various objectives that countries can adopt according to national statistical priorities. The GDDS emphasizes active interagency statistical collaboration and a central role by the national GDDS coordinator to address the weak links in the system and improve it over time (for details see IMF, 1978). Participation in the GDDS is voluntary but also requires, among other things, that a country appoint a GDDS coordinator—an official in one of the key national statistical agencies (usually from among the national statistical agency, the Ministry of Finance, or the central bank) to coordinate GDDS obligations and encouraged activities. GDDS members are expected to review and update the metadata (description of current practices) on their main macroeconomic and sociodemographic statistics on either a “best-effort” or “when-merited” basis. GDDS participants are, however, expected to update their plans for improvement on an annual basis. The metadata has to be posted on the IMF’s Dissemination Standards Bulletin Board (DSBB).
The GDDS, as a framework for statistical development, complements the broader IMF objective of assisting member countries to eventually graduate from the use of Fund resources and achieve better access to international capital markets.1 Greater access to capital markets is contingent on improved data dissemination (e.g., as evidenced by subscription to the SDDS) to facilitate investor decision making.2 In this regard, the GDDS was also envisaged as a vehicle for gradual progress toward subscription to the SDDS for those countries that had an interest in doing so. In practice, notwithstanding some expectations that some GDDS participants would rapidly become SDDS subscribers, only eight countries to date have actually done so.3
The GDDS, together with the SDDS, have recommended and prescribed practices, respectively, with regard to coverage, periodicity, and timeliness for the dissemination of various macroeconomic data categories. As such the GDDS/SDDS formed the first phase of the IMF’s response to the economic and financial crises of the early 1990s. Following the mid-1990s emerging economies financial crisis, The IMF collaborated with other international agencies and member countries in a more comprehensive effort to further develop international best practices.4 The objective was to strengthen the global financial architecture, among other things, to provide for broader data-quality best practices that go beyond the dissemination of data. Such endeavors led to the development of the Data Quality Assessment Framework (DQAF) and the data module of the Reports on the Observance of Standard and Codes (ROSC) process, through which the IMF undertakes comprehensive assessments of member countries’ statistical systems in producing quality macroeconomic data consistent with international standards. The GDDS/SDDS dissemination benchmarks (as relevant) are used in conjunction with the DQAF when undertaking the ROSC missions to member countries.
A. Promoting the GDDS
The IMF has provided considerable TA to member countries and collaborated, among others, with the UK Department of International Development (DFID), Japan, and other international organizations, including the World Bank, to promote the GDDS. In the early stages, the TA entailed promoting awareness and training through workshops for country officials and the preparation of metadata and plans for improvement for several countries. A number of these efforts were undertaken via regional GDDS projects5 and workshops, mostly funded by DFID and the Japanese TA subaccount at the IMF. Starting in May 2000, the first metadata, including plans for improvements, of countries that fulfilled the GDDS requirements were posted on the IMF’s DSBB.
A successful outcome of the IMF’s TA efforts was the rapid growth in the GDDS participation by member countries from all regions of the world (see Figure 1). Notable with this outcome was the rapid increase in the participation by countries in Africa (AFR), largely reflecting the rapid increase (especially during 2001 through 2003) in GDDS participants from Anglophone African countries, which benefited from support from the DFID-funded GDDS project for these countries. Secondly, notwithstanding relatively rapid growth of GDDS participants in the Middle East and Central Asia Department (MCD) countries, progress lagged other world regions. In response, since 2005, the IMF has made special efforts to step up assistance in promoting the GDDS in the region (see below).
Regional and DFID-Funded Project Countries’ Representation in the GDDS and SDDS (2001-2007)
(In percent of IMF member countries in each region)1
Citation: IMF Working Papers 2009, 278; 10.5089/9781451874235.001.A001
Source: IMF Statistics Department.1 Africa (AFR), Asia and Pacific (APD), Europe (EUR), Middle East and Central Asia (MCD), and Western Hemisphere (WHD). DFID project countries as a percentage of the total in AFR.Regional and DFID-Funded Project Countries’ Representation in the GDDS and SDDS (2001-2007)
(In percent of IMF member countries in each region)1
Citation: IMF Working Papers 2009, 278; 10.5089/9781451874235.001.A001
Source: IMF Statistics Department.1 Africa (AFR), Asia and Pacific (APD), Europe (EUR), Middle East and Central Asia (MCD), and Western Hemisphere (WHD). DFID project countries as a percentage of the total in AFR.Regional and DFID-Funded Project Countries’ Representation in the GDDS and SDDS (2001-2007)
(In percent of IMF member countries in each region)1
Citation: IMF Working Papers 2009, 278; 10.5089/9781451874235.001.A001
Source: IMF Statistics Department.1 Africa (AFR), Asia and Pacific (APD), Europe (EUR), Middle East and Central Asia (MCD), and Western Hemisphere (WHD). DFID project countries as a percentage of the total in AFR.The experience with the provision of TA via the DFID-funded GDDS for Anglophone African countries during 2001-06 (Phase I) has yielded important lessons.6 First the project, which was executed jointly with the World Bank, provides useful hands-on collaboration between the two Bretton Woods Institutions and a donor in providing concerted and targeted TA using the GDDS framework. In this context, DFID was able to leverage its financial support in working closely with the Fund and the World Bank to take advantage of their extensive knowledge of and operations within the target countries. The GDDS’s comprehensive approach (covering both macroeconomic and socio-demographic statistics)7 and its focus on collaborative country and development partner efforts to facilitate the statistical development process facilitated synergies of the TA provided by both institutions. Second, a number of shortfalls in achieving some of the objectives of the project could be traced to inadequate awareness and support for statistics, including lack of provision of adequate national budget and donor resources. Third, a number of best practices within the project led to the conclusion that there was scope to enhance data dissemination both to better highlight the successes of the project and also provide a better link between project inputs and final outputs—an increasingly important part of evidence-based monitoring and evaluation.
Phase II of the DFID-funded GDDS Project (2006-09) was adapted to address some of the outstanding issues from the Phase I project. In particular, TA delivery was refocused into a modular approach to target groups of five to six countries with similar capacity-building needs and encourage increased ownership and accountability by country officials. Two such modules (the GDDS and SDDS modules)8 have already assisted some seven countries to improve data dissemination by posting them on national websites, entailing improved coordination among the national statistical agencies. A number of countries have also incorporated their statistical plans for improvement/statistical strategies into their poverty reduction strategy papers (PRSP) and associated medium-term expenditure frameworks to secure improved funding for reforms.
For the countries in the Middle East and Central Asia, the IMF increased its efforts to promote their participation in the GDDS to comparable levels to those in other parts of the world. Stepped up TA and training workshops have yielded improved GDDS participation (see Box 1).
B. Other Relevant International Initiatives
Since the GDDS was launched in 1997, several international initiatives have been introduced to enhance the scope and effectiveness of support for economic development, more specifically focused on reducing poverty while promoting evidence-based monitoring and evaluation. These initiative comprised (a) the establishment of the Partners in Statistics for the 21st Century (PARIS 21) as an advocacy group to promote statistical development; (b) the Monterrey Consensus and other regional efforts to significantly increase resource flows to developing countries; (c) the introduction of nationally-owned PRSPs as operational frameworks to identify and fund development priorities; (d) several high-level fora on aid effectiveness [Rome (2003), Paris (2005), and Accra 2008)] to elaborate on measures to enhance aid effectiveness with concrete actions to be implemented by both developing and partner countries; (e) several international roundtables on results [Washington (2002), Marrakesh (2004)9, and Hanoi (2007)] to flesh out some specific development agendas and (f) a High Level Consortium Meeting in Dakar (2009) reviewing ten years of work of PARIS 21, and agreeing the Dakar Declaration for the Development of Statistics (DDDS).
Recent IMF Efforts to Increase GDDS Participation and SDDS Subscription
Participation in the GDDS
Middle East and Central Asia region
The IMF Statistics Department’s (STA) active engagement in the Middle East and Central Asia (MCD) region in the past four years through outreach activities and technical assistance missions resulted in an increase in the regional participation in the GDDS. Since 2005, eight countries from the region and West Bank and Gaza have started participation in the GDDS, the most recent four being Algeria, Bahrain, Saudi Arabia, and the U.A.E. With the latter three GDDS participants, all the Gulf Cooperation Council (GCC) countries are currently participating in the GDDS. The combined membership of countries in the region in the GDDS and the SDDS increased significantly to 78 percent from about 44 percent of the total at the beginning of 2005.
The 2005 regional GDDS outreach workshop in the United Arab Emirates (U.A.E.) jump-started the process, and a recent GDDS workshop in Bahrain, focused on changes in the GDDS framework stemming from the Seventh Review of the Fund’s Data Standards Initiatives (Seventh Review), laid ground for future improvements. The Middle East Regional Technical Assistance Center (METAC) has also been instrumental in achieving this progress.
Other regions
STA has conducted five outreach workshops covering the GDDS in other regions of the world during the past four years. Assistance in preparing for GDDS participation has also been provided to several countries. Since 2005, a total of 11 countries from other regions of the world started participation in the GDDS.
The most recent GDDS workshops focused on changes in the GDDS framework stemming from the Seventh Review of the IMF’s Data Standards Initiative, emphasizing the need to improve data dissemination practices and to develop advance release calendars (ARC) and national summary data pages (NSDP). Progress in these areas, which are part of the SDDS requirements, is expected to assist countries in preparing for the SDDS subscription.
Increased SDDS subscription
STA has conducted six SDDS outreach workshops and 16 SDDS assessment/subscription missions during the past five years. As a result of these efforts, eleven countries have subscribed to the SDDS since 2004, bringing the total subscription to 66 countries, a number of countries are close to subscription and are expected to subscribe to the SDDS in the next two years.
To assist the MCD countries in preparing for SDDS subscription, STA organized a regional SDDS workshop in Bahrain in March 2009. Officials from all the GCC countries, Jordan, and West Bank and Gaza attended the workshop during which, action plans for subscription to the SDDS were developed.
There were also a number of regional initiatives, for example those sponsored by the regional UN Agencies and development banks, which have sought to focus on the special needs of each region.
Together, these initiatives have broadly sought to better integrate the priorities of developing countries and development partners by emphasizing the strengthening of capacity building in the recipient countries, avoiding duplication, and broadly promoting the mainstreaming of statistics and monitoring into national development strategies. The greater emphasis on more specific and time-bound targets associated with poverty reduction such as the millennium development goals (MDGs) have also forced developing countries and partners to focus much more on elaborating benchmarks and data to permit improved monitoring and evaluation. The greater accountability imparted on developing countries by some of these initiatives has given rise to sizable new demand for macroeconomic and socio-demographic data even as the national capacities to fill the data needs have remained largely inadequate.
The National Strategy for the Development of Statistics (NSDS)—an important output of the Second Roundtable’s Marrakech Action Plan for Statistics—is an especially good example of the complementariness with the GDDS.10 The NSDS incorporated the GDDS as its core as it was designed to expand on the key elements of the GDDS. As its name suggests, the GDDS is output oriented, focusing on data and its dissemination while the NSDS encompasses a broader continuum of statistical development issues, with more emphasis on the requisite inputs. These inputs include infrastructure (e.g., office buildings, information technology systems, including software and other equipment), surveys and census, administrative data and information, long-term training, and data demand issues. Such inputs fit in more closely with projects supported by the World Bank, the regional development banks, and donors. Their increased commitment to funding these activities is yielding substantial support for downstream TA in support of the GDDS data and their dissemination. In particular, improved funding for surveys, censuses, and administrative data systems generate source data for macroeconomic statistics in direct support of the IMF TA mandate.
C. Review of the GDDS
In 2007, the IMF undertook to review the GDDS during its first 10 years.11 This review took place against the background of the evolving global context, in part influenced by the experience with the subsequent other international initiatives.
The review entailed a desk evaluation of relevant documents, including TA mission reports. It also included consultations with member countries through regional GDDS assessment workshops.12 The review was broadly endorsed by the IMF Executive Board but the main recommendations to enhance the GDDS were approved in the subsequent Board discussion of the Seventh Review of the Data Standards Initiative in November 2008.13 The key changes to the GDDS comprise:
Strengthening the emphasis on data dissemination: the proposal shifts the focus from updating metadata description of existing statistical practices (an intermediate output) to the final output, namely data. GDDS participants are expected to review and update the metadata on either a “best-effort” or “when-merited” basis. This approach embodies the changes since the inception of the GDDS with a wider recognition that data dissemination creates its own demand and is likely to raise the profile and visibility of and, hopefully, support for statistical agencies and reforms;
Realigning the data categories of the GDDS: the aim is to simplify and reformulate the data categories of the GDDS to more closely approximate those of the SDDS, albeit with less demanding requirements for data periodicity and timelines. The changes should assist GDDS countries with better access to capital markets (since many of them are already borrowing in such markets) and makes it easier, for those interested, to graduate to SDDS subscription;
Strengthening plans for improvement: this would be via more comprehensive and regular update schedules (at least on an annual basis) that would better serve the coordination among national statistical intragency partners (and with developing partners) in prioritizing TA and training;
Strengthening the emphasis on graduation to the SDDS: this is important as it helps to set more concrete outcomes for TA for those countries wishing to graduate to the SDDS. Better alignment of the GDDS data categories with those of the SDDS and the introduction of national data summary pages (via a national website) and adoption of advance release calendars for GDDS participants should create a clearer path to SDDS subscription;
Achieving better regional balance: the GDDS should be promoted as a source of guidance for all IMF member countries, and this requires extra efforts to spread its benefits in a more balanced manner to all regions.
III. Challenges For The Future
Judging from the key recommendations of the GDDS review and the entrenched challenge of recurring economic and financial crises, the stage has been set for renewed vigor to pursue the objectives of the enhanced GDDS. The emphasis on data dissemination, more focused TA (based on improved plans for improvement) and enhanced regional drives to encourage graduation to the SDDS should enable more emerging and developing economies to eventually attain self-sustaining national statistical systems. Admittedly, not all Fund member countries have the capacity or resources to attain the SDDS subscription status even in the medium-term. However, for the well targeted number of countries which can make the leap, the benefits are substantial. Such progress would not only help support better informed macroeconomic policies to prevent or address economic/financial crises but would also help sustain the relevant countries’ greater participation in the global economy and attract and retain capital inflows, especially foreign direct investment. Moreover, the demonstration effect of successful countries and peer pressure on the rest should further benefit international statistical efforts.
More importantly, further international statistical progress is highly dependent on resource mobilization and enhanced public awareness and support for further statistical development. Another important issue in this regard is the need to build on the momentum of various international initiatives to strive for more focused and enhanced collaboration among international providers of financial and TA resources. Together these developments should significantly improve the environment to foster greater effectiveness and success of TA.
IV. Conclusions
The GDDS and the SDDS were introduced in response to the emerging economies’ crisis in the early 1990s to address a demonstrated need to significantly enhance data dissemination. Over the years with recurring crises in emerging economies and the current global financial crisis, the urgency for further significant improvement in data and their dissemination remain important issues, which need enhanced corrective measures.
In practice, the implementation of the GDDS in the first 10 years tended to focus on the developmental aspect of the framework, emphasizing enhanced awareness and preparation and update of metadata—an intermediate output—that was seen as critical to identifying weaknesses in the national data systems in order to address them. The 10-year experience with the GDDS, including complementary assessment of Fund TA based on the framework, suggests a need to introduce some reforms. With the benefit of already reasonably well-established metadata, it seems reasonable to complement this work with a focus on data dissemination, an effort that would respond to the continued urgent need for generating the “final product” of statistical development, and–for those countries interested–facilitating the transition towards subscription into the SDDS, with its attendant benefits as countries integrate into the wider international economic community.
Subsequent to the launching of the GDDS, a number of international initiatives have complemented and enhanced the role the GDDS even as they have created conditions for its refinement. The GDDS (and the SDDS) were complemented by the DQAF to provide a more comprehensive framework for assessing national macroeconomic statistical systems. Various international efforts to enhance aid effectiveness and evidence-based monitoring and evaluation of time-bound MDGs have further enhanced the role of the GDDS. These developments, together with the experience of hands-on TA to various IMF member countries, have helped to identify a number of areas in which the GDDS need to be refocused. The review of the GDDS after 10 years, the complementary Seventh Review of the IMF’s Data Standards Initiatives, and consultations with countries authorities provided an agreed set of recommendations to be implemented to improve the GDDS.
Such reforms are designed to reposition the GDDS into a more appropriate framework to assist member countries’ endeavors to further strengthen their statistical capacities. Further refinement in the quality of data and their dissemination are particularly critical to underpin enhanced policies and help countries have better access to international capital markets in an increasingly global economy.
References
Cady, John, and Anthony Pellechio, 2006 Sovereign Borrowing Cost and the IMF’s Data Standards Initiatives: IMF Working Paper 06/78.
International Monetary Fund, Statistics Department, 1997, The General Data Dissemination System.
International Monetary Fund, Statistics Department, 2007b, The General Data Dissemination System: Guide for Participants and Users.
International Monetary Fund, Statistics Department, 2008, Evaluation of the UK DFID-Financed Technical Assistance GDDS Project for Selected Anglophone African Countries (2001-06). Available via Internet: www.imf.org/external/np/pp/eng/2008/032108a.pdf
International Monetary Fund, Statistics Department, 2008, Assessing the General Data Dissemination System (GDDS)—What Has Been Accomplished After Ten Years, and Where Do We Go From Here? Available via Internet: http://dsbb.imf.org/Applications/web/dsbbhome/
International Monetary Fund, Statistics Department, 2008, Seventh Review of the Fund’s Data Standards Initiatives. Available via Internet: www.imf.org/external/np/pp/eng/2008/111908.pdf
Executive Board Reviews Data Standard Initiatives, 2008, Public Information Notice (PIN) No. 08/147 (Washington: International Monetary Fund). Available via Internet: www.imf.org/external/np/sec/pn/2008/pn08147.htm
Kester, Anne Y., 2006, IMF Data Standards Initiatives: A Consultative Approach to Enhancing Global Data Transparency, IMF Working Paper 06/102.
Kibuka, Robin D., 2007, Mainstreaming Statistics in the Poverty Reduction Strategy Approach to Provide for More Effective Technical Assistance: Some Experience at the IMF, IMF Working Paper 07/255.
Morrison, Thomas K., ed., 2005, Statistical Capacity Building: Case Studies and Lessons Learned (Washington, International Monetary Fund).
UK. Department of International Development, 2008, Evaluation of DFID Support to Phase One of the GDDS Project (2002-06) and to the World Bank Trust Fund for Statistical Capacity Building: Strategic and Management Overview.
World Bank and PARIS21, 2007, Scaling Up Investment in National Statistical Systems, (Washington).
For details see http://www.imf.org/external/np/exr/facts/
There is some empirical evidence suggesting that country borrowing costs in the international capital markets can be reduced by participation in the GDDS and even more so with subscription to the SDDS. For details see Cady and Pellechio, 2007.
They comprise Armenia, Bulgaria, Kazakhstan, Kyrgyz Republic, Moldova, and Romania. Cyprus and Malta subscribed on 30 November, 2009.
The good practices were incorporated in standards and codes. The IMF and the World Bank have recognized international standards in 12 areas related to their work. In assessing countries’ observance of these standards, and helping them implement reforms where needed, the IMF and the World Bank aim to improve the functioning of the relevant economies as well as the basis for investors’ decisions, and ultimately to promote greater financial stability and prevent financial crises. For details, see http://www.imf.org/external/standards/.
For details see http://dsbb.imf.org/Applications/web/getpage/?pagename=gddsprojectnews#PI
For details see DFID 2008 and IMF 2008a.
The IMF’s mandate is closely linked with macroeconomic statistics while those of the World Bank and Regional Development Banks are more aligned with socio-demographic statistics.
The GDDS module’s objective is to assist with mainstreaming statistics into the poverty reduction strategy papers to enhance national priority and medium-term funding for broader statistical reforms. Moreover, the module, together with the SDDS module, also assists participating countries to enhance data dissemination by developing national summary data pages and the use of advanced release calendars—an approach that is already a key feature of the SDDS subscription.
An important outcome of the Second Roundtable was the Marrakech Action Plan for Statistics. For details see http://www.mfdr.org/2ndRoundtable.html
For details see Kibuka, 2007.
For details see IMF, 2008b.
The regional workshops took place during 2007 in South Africa and Thailand (April), Mexico (June), and Senegal (November) and were funded by DFID and the Japanese TA sub account at the IMF.
For details see IMF, 2008c and 2008d.
