This note provides guidance on use of third-party indicators (TPIs) in Fund reports, in line with the policy endorsed by the Board in November 2017.

Abstract

This note provides guidance on use of third-party indicators (TPIs) in Fund reports, in line with the policy endorsed by the Board in November 2017.

Introduction

1. This note provides operational guidance on how to use Third-Party Indicators (TPIs) in Fund reports. A TPI is an indicator compiled by other organizations (i.e., other than the Fund or member country authorities).1 TPIs are usually presented as time series, a collection of observations measuring the same concept over time. Observations of a TPI can be numerical values (e.g., an index), or categorical values that can be ranked according to a specified scale (e.g., a credit rating). This TPI framework applies to all documents that are subject to the Fund’s Transparency Policy (i.e., Board documents, see Appendix 1 of the Updated Guidance Note on the Fund’s Transparency Policy). The Executive Board has also asked staff to follow similar guidelines for other Fund publications, in particular, Regional Economic Outlooks and Staff Discussion Notes. This note is based on the IMF Policy Paper Use of Third Party Indicators in Fund Reports and incorporates the views expressed by the IMF’s Executive Board during its discussion of that paper.

2. Use of TPIs in Fund reports is widespread and is likely to remain so. Staff’s continued practice of drawing on other institutions’ expertise and estimates is consistent with the Executive Board’s guidance in areas where internal expertise is lacking or limited. Recent experience suggests that some TPIs have been useful in work on macro-critical issues beyond the traditional areas covered by macroeconomic and financial statistics. They provide evidence to help inform analysis or complement assessments, including in areas where official data are not available. Therefore, TPIs play a role in supporting the Fund’s evidence-based approach in priority areas such as risks and spillovers, macro-financial and macro-structural issues, and governance and corruption; the emerging areas of inequality, gender, and climate change; and in program work.

3. The varied qualities of TPIs currently used by staff present challenges and risks to the Fund’s credibility. TPIs used by Fund staff are drawn from numerous compilers, including international institutions, non-governmental organizations, and private sector firms. Some TPIs are based on hard data, while others are based on qualitative assessment by experts, experience, perception, or composites of various underlying data sources. Issues may arise from the opacity of some TPIs’ sources and methodologies, and the quality and reliability of TPIs that are based on perception or value judgment. Lack of transparency in staff’s choice of some TPIs or their components risks undermining the persuasiveness of Fund advice.

4. The Executive Board agreed that a more structured approach to using TPIs would help promote best practice and ensure consistency and evenhandedness. The TPI framework consists of three principles for best practice (transparency, robustness, and reflection of stakeholders’ views); the Third-Party Indicators Digest, a “quality assessment” database for internal use only (as the Indicators Digest is intended to be a database for internal use, staff should not refer to it in external publications); and the review process. Directors attached importance to judicious and evenhanded use of TPIs based on macro-criticality, and relevance (i.e., considerations of country-specific context in interpreting results). Moreover, TPIs should not replace—but rather supplement—an open, candid, robust, and well-documented discussion with the authorities.

5. This guidance note is structured as follows. The next section outlines the principles for best practice in TPI use and provides specific guidance for staff when using and presenting these indicators. The following section outlines resources available to support staff’s use of TPIs, including the Indicators Digest, good practice examples, and examples of practices to avoid.

Principles for Best Practice in TPI Use

A. Motivation Behind the Principles-Based Approach

6. The principles-based approach provides room for staff judgment and flexibility to make progress on surveillance priorities, while avoiding flawed analysis and minimizing reputational risks. This approach recognizes that the universe of available TPIs will continue to evolve, and that there are benefits from staff being able to draw on the most recent and highest quality sources when methodologies and data availability improve over time. Additionally, because there are no international standards for TPIs, it would not be feasible or desirable to establish a positive or negative list of TPIs for use in Fund reports. Thus, there is an important role for staff judgment, stakeholder engagement, and review mechanisms to ensure the quality of end-products.

7. Staff should not view these principles as a prescriptive list, but take a risk-based approach to applying them to keep reports focused and streamlined. Staff should exercise judgment in how to apply the principles for specific indicators and document types to maximize traction and minimize risks associated with TPI use. There are a variety of considerations in determining when a principle is relevant and how to apply it, taking into account how use of a TPI may influence the analysis and the persuasiveness of the policy advice. Box 2 lists issues that staff may need to consider and guidance on how to approach them. Staff can also find detailed information on the most commonly-used indicators in Fund reports in the Indicators Digest to help inform their judgment. Examples provided below illustrate good practices for how the principles can applied, as well as practices to avoid.

B. Principles for Best Practice

Transparency

8. Transparency about the selection of indicators and how they feed into staff’s analysis and policy recommendations is key to promote candor, credibility, and traction. An adequate understanding by staff of the indicators they are working with is critical to ensure proper use in analysis, policy advice, and discussions with the authorities. Additionally, staff may need to highlight some details about the indicator in the report to facilitate interpretation by the reader. Staff should exercise judgment when applying this principle—taking into account the type of indicators used, the subject matter and country context, and the type of report—to keep reports streamlined. Actions that can increase transparency are noted below. As highlighted in Box 2, depending on the context, not all actions need to be detailed in the report.

  • Consider whether and why TPI use is needed. Staff should reflect on what factors are motivating their use of TPIs in specific cases. To the extent possible and where relevant, staff should rely on information the Fund has already obtained in the context of existing activities, including discussions with the authorities.

  • Clear identification of indicators. Explicit identification of TPIs and their compilers helps to distinguish them from data compiled by member countries or the Fund. Staff should be precise when describing the compiler to avoid misattribution.2

  • Justification of indicator choice. To the extent that it’s not self-evident, staff could explain why the indicator was selected—considering strengths of that indicator relative to others— and how the indicator informed the overall assessment. When working with TPIs in the Indicators Digest, staff could draw on the “overall assessment” section and tailor it to country circumstances.

  • Discussion of indicator’s key characteristics. Staff should be clear if the indicator is based on a survey of perceptions, a survey of hard data, and/or expert judgment, as opposed to official data. Staff should acknowledge any other characteristics that could affect interpretation, including timeliness of source data, country and time coverage of the indicator, and representativeness of the sample if it is a survey-based indicator.

  • Acknowledgement of measurement uncertainties. Staff are encouraged to report confidence intervals and/or standard errors (when reported by the compiler) when presenting indicators across time or across countries to avoid drawing attention to differences that may not be statistically significant.

  • Awareness of factors impacting perceptions of objectivity. This includes, for example, the entity’s mission, funding, and governance, and whether the indicator is too blunt a tool to be useful in supporting specific policy advice.

9. The Indicators Digest can serve as a reference for the most commonly used indicators. The Indicators Digest lists the types of source data used in the indicator’s compilation, the compiler’s funding source, a summary of the methodology, and references to research on the indicator. The Indicators Digest also features an overall assessment that should assist users in forming their own judgment on the strengths and weaknesses of the indicator and its appropriateness for the analysis, which staff can draw on and tailor to country circumstances. In that sense, its purpose is to enrich the analysis presented in Fund reports.

10. When incorporating relevant details into the report, staff should aim for efficiency and brevity, and take a tailored approach. Staff should avoid generic disclaimers, as the emphasis should be on highlighting issues that are specific to the TPIs, concept, and/or country being discussed. Staff should adjust presentation to the context, reflecting the risk-based approach and considering the type of report. They could highlight details in the main text, footnotes, appendix, or elsewhere. (See Box 4 for good practice examples, including staff reports on Senegal, Colombia, Comoros, and El Salvador, as well as the EUR Regional Economic Outlook.)

Robustness

11. Assessments and policy advice informed by TPIs should be broad-based, reflecting various quantitative and qualitative information to reduce the risk of flawed analysis. Achieving robustness does not require reliance on multiple indicators (as discussed below), and can be pursued by relying on an individual TPI, official sources, and/or qualitative information. When staff references multiple TPIs from different sources, they should be aware of the potential for correlation among TPIs that may rely on similar inputs.3 Suggested actions to promote robustness include:

  • Supplement with qualitative input and analysis. As noted above, use of indicators should be one of many inputs and a complement to qualitative discussion, informed by knowledge of the country context, staff judgment, dialogue with the authorities, and other sources. In some cases, a discussion of factors that may be influencing a country’s performance over time or relative to other countries can aid interpretation of TPI data.

  • Refer to multiple indicators. Where feasible, staff could draw on multiple indicators from different, independent sources that measure similar concepts and check for common findings or potential discrepancies. In doing so, staff will need to consider the tradeoff between the number of indicators available for this purpose and coverage of the data across countries and reference periods.

12. The Indicators Digest can support staff’s efforts to perform robustness checks. The Indicators Digest groups the most commonly used indicators by the concepts they measure, and presents their relative strengths and weaknesses according to an adapted Data Quality Assessment Framework (DQAF), with summaries presented in charts up front. Other useful information includes each indicator’s coverage across countries and across time.

13. Staff could acknowledge briefly in the report how robustness checks underpin their analysis. Staff should find a way to reflect their conclusions from robustness checks of multiple indicators and/or qualitative information. Staff could do this graphically, in the main text, or in a footnote (see Box 4 for an indicative illustration of approaches used in various document types, including in staff reports on Myanmar and the Review of 1997 Guidance Note on Governance—A Proposed Framework for Enhanced Fund Engagement). When using a familiar TPI that is widely accepted as a good measure of a particular concept, explicit discussion of robustness may not be warranted (see Box 2).

Stakeholders’ Views

14. When TPIs inform analysis, staff should discuss their findings with the authorities and any other key stakeholders. These stakeholders may have concerns about the TPIs, different interpretations of the results, or additional context to provide. Fund reports present the staff’s independent and candid views. However, staff should incorporate or report views of the authorities and any other relevant stakeholders in the staff report. Any modifications to staff reports subject to the Transparency Policy after their issuance to the Board must be consistent with that policy. The process for incorporating stakeholders’ views varies by product:

  • Article IV and UFR reports. Teams should engage stakeholders on or before mission. In Article IV and UFR cases, the most prominent stakeholders are the authorities, although staff could consult other stakeholders, as relevant to the issue and country. Consistent with the Guidance Note for Surveillance under Article IV Consultations, the authorities’ views should be reflected in the report that is issued to the Board (see Box 4 for some examples, including staff reports on Hungary, Somalia, and Comoros).4 Other stakeholders’ views on the indicator should be reflected as relevant. When third countries are mentioned in the report (e.g., text or charts), it is advisable to consult the relevant country team(s) to ensure staff takes these views into account when conducting their analysis and arriving at policy conclusions.

  • Multilateral Policy Issues Documents.5 In these cases, it may not be feasible to incorporate stakeholders’ views in the report that is issued to the Board. Instead, to reflect authorities’ views as needed prior to publication, staff may make modifications to these reports in order to address the comments from the membership expressed at the relevant Executive Board meeting (provided that Management shares those views), as allowed for by the Fund’s Transparency Policy.

  • Policy Papers. Similarly, where a policy paper incorporates cross-country analysis informed by TPIs, it may not be feasible to incorporate stakeholders’ views in the paper that goes to the Board. To reflect authorities’ views prior to publication, staff may revise the paper where they have modified their views in light of a Board discussion (provided that Management shares those views), in line with the Fund’s Transparency Policy.

  • Regional Economic Outlooks (REOs). REOs are not discussed by the Board and, therefore, staff could seek the authorities’ views during the report’s drafting stage by consulting the relevant Executive Director(s). This is particularly relevant for cases where an individual country’s performance—as measured by a TPI—informs some part of the analysis or policy advice in a REO chapter.

  • Staff Discussion Notes and Working Papers. Stakeholders’ views on TPIs can be obtained when the note or paper dealing with a specific country or group of countries is shared with the relevant Executive Director(s).

C. Specific Guidance

15. When using and presenting TPIs, staff should consider their source data. The source data for TPIs fall along a spectrum of four data types, ranging from more objective sources (e.g., official data or surveys focused on eliciting hard data) to more subjective sources (e.g., expert judgment or surveys of perceptions) (see Figure 1). Moreover, some TPIs are composite indicators that rely on a variety of inputs, including other TPIs. The type of source data also has implications for the approach to be taken when using and presenting the TPI (see Figure 2).

Figure 1.
Figure 1.

Types of Source Data

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Adapted from an approach used by the UNDP in its “User’s Guide to Measuring Corruption and Anti-Corruption.”

16. Perception-based TPIs can yield valuable insights but there are also potential pitfalls to avoid. Perceptions could reflect a gap between reality and what’s written in the laws or policies, and could therefore matter for economic decisions. Nevertheless, staff will need to recognize specific challenges when working with perception-based indicators. For example, individual perceptions can change from year to year without fully reflecting changes in the fundamental reality. Further, cross-country comparison or time-trend analysis based on some perception-based indicators may be unreliable, if survey samples are not consistent across countries or change across time. Recognizing measurement uncertainties around the point estimates in a quantitative discussion or presentation will be essential. In some cases, the compilers already provide standard errors or confidence intervals, and staff should incorporate these into their analysis and presentation (see Figure 2 for details).

17. Presenting point scores and acknowledging uncertainty around those scores where relevant—rather than relying on country rankings—could facilitate richer insights and discussion. Rankings reflect relative, rather than absolute performance. Thus, an improvement in a country’s ranking could reflect improvements in its performance, deterioration in other countries’ performance, or some combination of the two. While there may be cases where rankings are relevant (including when a country is benchmarking its own performance in that way), in the case of measuring corruption, staff should not use rankings related to TPIs (see Box 1).

18. Cross-country comparisons may help benchmark performance and provide useful context but should be undertaken carefully. When using TPIs for cross-country analysis in Article IV or UFR reports, it is generally preferable to present a comparison of that country’s performance relative to the average/range of a peer group (based on region, income, or other groupings), rather than an individual country. Staff should identify other specific countries—which are not the subject of the report—in terms of their performance according to TPIs only when this improves the analysis and policy recommendations, with an explicit justification to be included in the staff report where the rationale for doing so is not self-evident. In making cross-country comparisons, an alternative to using point scores or rankings is to cite distance to frontier (DTF) scores, which measure the deviation of a country from the best performer.6

19. Changes in a compiler’s methodology may limit the ability to compare scores across time, before and after the methodological change was implemented. For TPIs in the Indicators Digest, such methodological changes have been flagged, with such information typically included in the methodological notes made available by the compiler.

Specific Guidance on the Use of TPI Rankings on Corruption

Consistent with the April 2018 Board discussion of Review of 1997 Guidance Note on Governance—A Proposed Framework for Enhanced Fund Engagement, the Fund’s assessments of governance vulnerabilities should be holistic, relying on both quantitative and qualitative information. To the extent possible and where relevant, staff should rely on information already obtained by the Fund— including from member authorities, in the context of existing Fund activities—and whenever data gaps exist, they should be specifically acknowledged. Use of TPIs should complement—not displace—the analysis of Fund staff and other international organizations, including the World Bank and regional development banks. The authorities’ views should be adequately reflected in the relevant staff report, they should be informed sufficiently in advance of the intention to discuss these issues, and the discussions should be open and transparent. Staff reports should reflect candid discussions, using clear and direct language.

Indicators that measure the severity of corruption have a long history and are available for a very large number of countries. While some of these indicators have been challenged on the grounds that subjective perceptions may differ from reality, their defenders maintain that they correlate well with more “objective” measures and that—when it comes to investment decisions, tax compliance, capital flight, and brain drain—perceptions actually matter.1

Nevertheless, to avoid the risk of mechanical use and promote a constructive dialogue focused on finding policy solutions, staff will use scores rather than rankings when working with TPIs measuring corruption in all documents circulated to the Board and in published Fund products. In particular:

  • No country rankings. Staff should refrain from using country rankings when working with TPIs on corruption. This includes both ordinal and percentile rankings or any quantile grouping.

    For example: “Country X ranks 129th in the Worldwide Governance Indicators Control of Corruption Index (WGI-CCI)” or “Country Y is in the 3rd quartile of the WGI-CCI” are not acceptable.

  • Use scores. Staff can make use of point scores (which do not carry a ranking) to develop arguments in regard to governance/corruption; they can also consider presenting country scores relative to those of a group of peers (ranges or averages).

    For example: A chart/text comparing Country X’s score from the WGI-CCI to the average of peer countries or a range of scores from peer countries is acceptable. Avoid presentations of point scores that could appear as a ranking (e.g., cross-country “heat maps” that identify countries in quantile groups, or label individual countries according to ranking-related color schemes or bar charts).

  • Acknowledge uncertainty. Staff should be mindful of the degree of uncertainty around reported point estimates and should report confidence intervals and/or standard errors when available. For example, the WGI-CCI publishes a score for Control of Corruption and a standard error for this score. The 90 percent confidence interval around this score can thus be calculated as the score +/-1.645 times the standard error. Below is one example of the provision of confidence intervals.

1/ See Box 3 in Review of 1997 Guidance Note on Governance—A Proposed Framework for Enhanced Fund Engagement.
uA01fig01

Figure X. Country Y: Governance 1/

(Lower scores indicate weaker governance)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Sources: Worldwide Governance Indicators, D. Kaufmann (Natural Resource Governance Institute and Brookings Institution) and A. Kraay (World Bank), 2017. Verisk Maplecroft’s Corruption Risk Index, 2017. Transparency International’s Corruption Perceptions Index, 2017.1/ Use of these indicators should be considered carefully, as they are derived from perceptions-based data. Ranges are for a 90 percent confidence interval. Confidence intervals for peer group averages are negligible.2/ Confidence intervals are not available for this indicator.

D. Role of the Review Process

20. The review process will help promote proper and consistent application of the framework. Authoring departments have primary responsibility for ensuring application of the framework. They should consult STA when questions arise on indicators that are not included in the Indicators Digest (see next section). In its review of staff reports, SPR will verify adherence to the principles and guidance—and will consult with STA as needed—while acknowledging the need to exercise judgment in application to keep reports streamlined.

21. Responsibility rests with area departments to ensure consistency with the TPI framework in other publications. For example, Selected Issues Papers (SIPs) are often informed by data that include TPIs. Additionally, authoring departments should endeavor to ensure that staff follow similar guidelines for other Fund documents and publications that are not subject to the Transparency Policy (e.g., Regional Economic Outlooks, Staff Discussion Notes, and Working Papers).

Risk-Based Application of the Principles: Issues to Consider

This box explains in a series of decision trees how to take a risk-based approach to applying the principles in specific contexts, depending on the indicator’s source data and the concepts measured, as well as its usage. It uses “dark green” to signal that staff should adopt a lighter touch when applying the specific principles referenced in the associated boxes; “light green” to signal that staff could consider adopting a lighter touch for certain principles; and “yellow” to signal that greater rigor may be warranted when applying the principles.

Figure 2.
Figure 2.

Specific Guidance for Staff Analysis and Presentation

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

1/ This matrix reflects two ends of the continuum. Intermediate forms include those in Figure 1, as well as composite indicators blending different kinds of data.2/ Degree of uncertainty of a score is determined by standard errors and/or confidence intervals associated with the score, which are published by many compilers. When confidence intervals overlap it should be assumed that performance is roughly the same.3/ Use of TPIs should serve as a complement to qualitative discussion based on other inputs such as knowledge of the country context, discussions with country authorities and other stakeholders, and other sources.

Resources Available to Staff

A. Third-Party Indicators Digest

22. The Indicators Digest is a central database for internal use only that compiles selected TPIs’ characteristics to inform staff’s judgment. The Indicators Digest was originally prepared in November 2017 and updated in May 2018.7 It provides information on 13 TPIs commonly used by Fund staff in the areas of political risk, governance and corruption, and business climate.8 It highlights the indicators’ statistical characteristics, strengths, and weaknesses. The assessment in the Indicators Digest is not a validation exercise, nor is it intended to present an ex-ante positive or negative list of indicators acceptable for use in staff analysis and Fund products. That is, indicators in the Indicators Digest are not necessarily those endorsed by the Fund. Similarly, indicators that are not in the Digest are not necessarily less reliable. As the Indicators Digest is intended to be a database for internal use, staff should not refer to it in external publications.

23. Although there are no international standards for TPIs, the Indicators Digest uses a structured and transparent method to assess the quality of a TPI. The assessment relies on an adaptation of the Fund’s Data Quality Assessment Framework (DQAF) that was endorsed by the Executive Board in 2003 to provide a comprehensive assessment of quality of official statistics. The adapted DQAF guides the assessment of strengths and weaknesses of TPIs by reviewing the following four relevant dimensions that are broadly related to the transparency, compilation, and dissemination of the data product:

  • Assurances of integrity. Is the TPI produced on an impartial and transparent basis?

  • Methodological soundness. Does the methodology capture all relevant concepts and definitions? Is it well documented?

  • Accuracy and reliability. Are the data sources used for compiling the TPI accurate, reliable, and timely?

  • Accessibility. Do the users of the TPI have easy access to the data, metadata, and relevant contact staff?

24. To assess the quality of a TPI along these four dimensions, the adapted DQAF uses multiple elements that cover transparency, concepts and definitions, source data, statistical techniques, revisions policy, data and metadata accessibility, and assistance to users. Based on information made available by the compiler, a rating is provided on how best practices are observed, partially observed, or not observed for each element of the adapted DQAF. A rating is not given if the element is not applicable, or the assessment is not feasible. The rating of each dimension is generally the average rating of the elements.

25. The Indicators Digest facilitates comparison of the relative strengths of indicators measuring similar concepts along the dimensions assessed. Each spider chart indicates the assessment of the four dimensions mentioned earlier. For each dimension, a rating away from center represents better data quality in one of the four dimensions assessed. However, these assessments do not signal that one indicator is superior to others, since each indicator has its own strengths and weaknesses in terms of country coverage, time series, consistency, and transparency. Instead, as noted above, these assessments can be useful for robustness check.

26. The Indicators Digest will remain a living database with an annual review process led by SPR and STA. The expansion of the Digest will be demand-driven and focused on emerging and macro-critical areas. Nonetheless, it will never be an exhaustive compilation of all possible TPIs staff may need or want to employ in their analysis.

27. To ensure consistency and efficiency of assessment, STA will be solely responsible for applying the adapted DQAF assessment to TPIs. STA will continue to leverage its unique expertise and structured approach to conduct the assessments, benefiting from the experience gained during the initial preparation of the Indicators Digest. Centralization of the assessment will help provide a more evenhanded and streamlined approach and reduce the subjectivity of the ratings. STA will conduct assessments for any new TPIs considered for inclusion in the digest within the annual update process, in close coordination with SPR.

28. Updates of the Indicators Digest will be communicated to Fund departments through outreach events and internal communication. Following past practices, SPR and STA may organize seminars to present the annual updates to the Indicators Digest, including the DQAF assessment of new TPIs. Priorities will be given to departments that use the new TPIs extensively. Minor technical updates to the Indicators Digest will be notified via email. The latest version of the Indicators Digest will be disseminated through the SPR and STA internal websites.

B. How to Approach TPIs not in Indicators Digest

29. STA may provide a preliminary assessment for new TPIs not included in the Digest or advise other Fund staff on how to assess them, as appropriate. Fund staff can request STA to review a TPI currently not included in the Indicators Digest. In consultation with SPR, STA will consider conducting a preliminary assessment depending on whether the indicator is widely used in the Fund and likely to be included in the upcoming updates. Staff are encouraged to refer to the Indicators Digest on how to approach TPIs that are not in the Digest (see Box 3). Staff should not, however, produce their own ratings based on the adapted DQAF as that would deviate from the principles of efficiency, consistency, and evenhandedness.

30. STA will continue to assess TPIs based on publicly available information and direct consultations with the compilers. In most cases, information needed for the ratings are available in methodology notes that are publicly disseminated by the compilers. Interactions with compilers will be aimed at getting a better understanding of the methodology for constructing the indicator, types of data sources, process of data collection and verification, calculation of confidence bands, and any limitations for usage. STA may also consult with other organizations with strong experience in areas where use of new TPIs are common.

How to Approach TPIs that are not in the Indicators Digest

This box provides supplemental guidance on how to approach indicators not included in the Indicators Digest. As the Indicators Digest is not an ex-ante positive or negative list of indicators acceptable for use in staff analysis and Fund products—and it is important to not stifle innovation—staff can use TPIs both inside and outside of the Indicators Digest.

Staff should seek out information about the indicator to inform consideration of whether and how to use it in their analysis. Issues to look at include:

  • Timeliness of the data;

  • Type of source data (i.e., is it an objective or subjective measure of the concept, per Figure 1);

  • Methodological approach used by the compiler;

  • Any structural breaks in the methodology that could impede comparison across time;

  • For survey-based TPIs, any deviations in the size or representativeness of samples that could impede comparison across countries and/or time;

  • For survey-based TPIs, confidence intervals or standard errors if reported by the compiler and their implications for comparisons across countries and/or time; and

  • Characteristics of the compiler and its funding source, if available.

It is not necessary to explicitly detail all this information in the staff report, but staff’s use of the indicator should reflect consideration of these issues.

C. Examples of Good Practices and Practices to Avoid

31. Another resource available to staff are the examples of staff reports and other Fund documents exhibiting good practices in the application of the TPI framework. The following documents have been selected as good practice examples because they apply elements of one or more of the principles or other guidance. While they do not necessarily adhere to the guiding principles in full, they provide a useful illustration of how specific principles may be implemented in practice. Some of these examples provide an indication of the types of characteristics and limitations of a TPI that might be relevant to disclose. However, the extent of discussion about the TPI will vary according to the context—consistent with the risk-based approach (Box 2)—and these examples are not conveying an expected minimum length or level of detail.

32. Additionally, generic examples highlight a sample of practices to avoid. While not a comprehensive list of potentially problematic practices, these examples give staff a sense of some common pitfalls.

Good Practice Examples

Poland Staff Report for the 2017 Article IV Consultation

  • ✓ Light touch applied for TPI that measures economic/financial concept directly, and is frequently used in Fund reports to measure that concept

Ghana Staff Report for the 2017 Article IV Consultation, Fourth Review Under the Extended Credit Facility

  • ✓ Light touch applied for TPI that measures economic/financial concept directly, and is frequently used in Fund reports to measure that concept

Refer to Figure 13

uA01fig03

Inflation Expectations, One Year Ahead

(Year-on-year inflation in percent)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

1/ Expectations for the following year’s average is affected by breaks when the forecast-year changes in January.
uA01fig04

Eurobond Yields to Redemption

(percent)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Source: Datastream

South Africa Staff Report for the 2017 Article IV Consultation

  • ✓ Light touch applied for TPI that measures economic/financial concept directly, and is frequently used in Fund reports to measure that concept

9. Financial conditions remain tight. An index summarizing conditions in the banking sector and capital markets is somewhat above its 10-year average, reflecting relatively high interest rates and credit spreads, as well as uncertainties measured by the implied 1-year volatility of the rand.4 Credit growth moderated, especially for households. Whereas large corporations have adequate liquidity, small- and medium-sized enterprises (SMEs) encounter greater challenges to access financing.

uA01fig05

Financial Conditions Index

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Sources: Bloomberg, LP. and IMF staff calculations.

Senegal—Staff Report for the 2016 Article IV Consultation and Third Review Under the Policy Support Instrument

  • ✓ Elaborates on factors relevant for interpretation of TPI

  • ✓ Analysis reflects country context and authorities’ reform agenda, and links TPI performance to specific bottlenecks and opportunities

30. Progress has been made in improving the business environment, but much work remains to be done. Senegal was in the top ten reformers for two years in a row in the World Bank Doing Business rankings for 2015 and 2016. Under the recent change in methodology introduced in the Doing Business 2017 report, the 2016 ranking was revised to improve Senegal’s ranking from 153rd to 146th, with a loss of one place from this new base to 147th in 20177 This reflects faster reform elsewhere as the index finds continued improvement in Senegal.8 The first phase of the government’s comprehensive structural reform plan to improve the business environment (Programme de réforme de l’environnement des affaires et de la compétitivité – PREAC I 2013–15) implemented around 90 percent of the 56 measures envisioned. The PREAC II 2016–18 includes land and judicial reform, legal measures to promote SMEs, revision of the investment code and simplified tax administration, among other things. Nonetheless, Senegal’s business environment is still considered relatively poor and steadfast implementation of reforms is needed, especially as other countries are moving steadily to improve their business climate. Its low level of FDI relative to peers is evidence that, despite its stability and favorable geographical position in the region, it has been unable to attract investment in line with its potential.

Hungary Staff Report for the 2017 Article IV Consultation

  • ✓ Analysis reflects country context and authorities’ reform agenda

  • ✓ Reflects stakeholders’ views

  • ✓ Considers sub-components of index

26. … Competitiveness. Hungary has lost ranking in competitiveness indicators. The main reported challenges are related to getting electricity, paying taxes, and starting a business. A committee, led by the Minister for National Economy, is already putting together a plan to address these challenges. Perceived corruption and frequent changes in rules and regulations have also been reported as impediments to doing business. It would, therefore, be important to address these issues through enhancing transparency and policy predictability. The authorities noted that the high frequency of introducing new rules and regulations was necessary in the past to meet EU regulatory standards and to deal with the 2008 economic crisis. They explained that compared to the crisis years, the number of laws and regulations that are introduced annually has significantly decreased.

Somalia Staff Report for the 2017 Article IV Consultation and First Review Under the Staff-Monitored Program

  • ✓ Analysis reflects country context and authorities’ reform agenda

  • ✓ Reflects stakeholders’ views

26. …Business environment. The authorities have stepped up efforts to increase the participation of the private sector role in economic activity. In particular, the recent passage of a foreign direct investment law and adoption of a procurement bill will further accelerate the role of the private sector. While the authorities welcome the inclusion of Somalia in the Global Doing Business Survey in 2017, they recommended that some indicators be treated with caution, given that it is the first time the country is in the survey.22

Sri Lanka Staff Report for the 2018 Article IV Consultation and Fourth Review Under the Extended Arrangement

  • ✓ Perception-based TPIs acknowledged

  • ✓ Key characteristics of indicator and its source data discussed

  • ✓ Confidence interval presented in chart

  • ✓ Peer comparators identified as grouping rather than individually

  • ✓ Uses point scores rather than rankings

  • ✓ Robustness checked by presenting related indicators

uA01fig06

Sri Lanka: Corruption Perception and Correlated Indicators

(Index value, EMDEs)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Notes: Transparency International Corruption Perception Index is an indicator that measures perception of corruption. A higher value represents lower perceived levels of corruption. The confidence interval around the corruption perception index is also presented in the chart. The accuracy of the index can be biased by experts’ views (instead of facts on corruption). Non-IMF indicators provide qualitative information about corruption. They do not represent the IMF’s assessment of the level of corruption in Sri Lanka.Sources: World Bank Doing Business Indicators; World Bank Benchmarking Public Procurement Database; PEFA; Transparency International Corruption Perception Index; and IMF staff estimates.

El Salvador Staff Report for the 2018 Article IV Consultation – Selected Issues

  • ✓ Analysis reflects country context and authorities’ reform agenda, and links TPI performance to specific bottlenecks and opportunities

  • ✓ Perception-based TPIs acknowledged

  • ✓ Key characteristics of indicator and its source data discussed

  • ✓ Peer comparators identified as grouping rather than individually

  • ✓ Robustness checked by comparing results of WEF GCI and World Bank Doing Business indicators, and official and empirical data

  • ✓ Global Innovation Index chart uses point scores rather than rankings and considers sub-components of index

Refer to “Reaping the Demographic Dividend in El Salvador

8. El Salvador improved 22 notches in the World Bank’s Doing Business ranking,6 but scope for further progress remains substantial. Reforms impacting positively the ranking covered four categories: dealing with construction permits, getting electricity, paying taxes, and trading across borders. However, El Salvador is still far from the frontier/best performers in the following areas: protecting minority investors, resolving insolvency, enforcing contracts, pointing to weaknesses in the judicial system. It also has a very low new business entry density compared to the LA-5 and other CAPDR economies (chart on New Business Density). According to the WEF,7 the most problematic factors for doing business are crime, corruption, inefficient government bureaucracy, policy instability, and tax rates. In terms of competitiveness and productivity, El Salvador lags behind other CAPDR and LA-5 countries with respect to the institutional and macro environment, labor and goods market efficiency, higher education, and innovation.

uA01fig07

Doing Business Ranking, 2018

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Source: World BankDoine Business Indicators. 2018.
uA01fig08

Global Competitiveness Index Ranking, 2016–17

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Source: World Economic Forum Global Competitiveness Indicators.
uA01fig09

Global Innovation Index, 2015

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Source: Global Innovation Index.

Refer to Footnote 8

The Global Innovation Index gathers data from more than 30 international public and private sources, covering a large spectrum of innovation drivers and results, and privileging hard data over qualitative assessments (57 are hard data, 19 composite indicators, and 5 survey questions).

Colombia Staff Report for the 2018 Article IV Consultation

  • ✓ Perception-based TPIs acknowledged

  • ✓ Elaborates on factors relevant for interpretation of TPI

  • ✓ Use of point scores rather than rankings

  • ✓ Analysis reflects country context and authorities’ reform agenda, and links TPI performance to specific bottlenecks and opportunities

  • ✓ Reflects stakeholders’ views

Refer to Box 2 in staff report, including:

Despite such a strong framework, some indicators suggest still weak perception of corruption. Surveys suggest that corruption has remained in the top 2 of the most problematic factors for doing business in Colombia since 2012; and Colombia’s trust in politicians has worsened somewhat since then reaching a score of 1.7 (out of 7) in the WEF’s 2017–18 report down from 2.3 in 2012–13. Recent analysis suggests Colombia fits well a common transitional process in which perception of corruption may in fact rise even when additional efforts are put in place to investigate and uncovered such activities—the authorities share this view.

Myanmar Staff Report for the 2015 Article IV Consultation – Selected Issues Paper

  • ✓ Analysis reflects country context and links TPI performance to specific bottlenecks and opportunities

  • ✓ Robustness checked by comparing results of WEF GCI and World Bank Doing Business indicators

  • ✓ Acknowledges key limitations of the TPIs

Figure 5.
Figure 5.

Selected Indicators of Impediments to FDI Inflows

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Improve Business Environment to attract FDI

21. It is widely accepted that FDI inflows are conducive to trade due to its financing role in investment and the associated transfer of technology. Although FDI has significantly increased in Myanmar, much of it has flowed into the gas sector. To attract FDI into other sectors, particularly the budding manufacturing sector, it is essential to improve the regulatory and business environment in Myanmar. Currently Myanmar ranks the lowest globally in the criteria of “ease of starting a business” based on the World Bank’s Doing Business 2015. Not surprisingly, its cost of starting a business is among the highest in the region while its regulatory quality is among the lowest.5

Union of the Comoros Staff Report for the 2014 Article IV Consultation

  • ✓ Perception-based TPIs acknowledged

  • ✓ Reflects stakeholders’ views

  • ✓ Considers subcomponents of index

20.…Staff pointed out that survey-based indicators point to competitiveness issues related to poor infrastructure and institutional and governance deficiencies that affect the quality of the business environment. Thus, Comoros ranks low on the list of the World Bank’s Doing Business Indicators (159th of 189 countries). Comoros ranks particularly poorly in the areas of starting a business, enforcing contracts, and government effectiveness. Comoros performs similarly poorly on comparative governance and corruption indicators. The authorities felt that progress in these areas had been made, especially by putting in place updated legislation, but they acknowledged that this had not yet resulted in improvements in the rankings and that it is important to ensure the consistent application of the new laws.

Georgia Request for Extended Arrangement Under the Extended Fund Facility and Cancellation of Stand-by Arrangement

  • ✓ Analysis reflects country context and authorities’ reform agenda, and links TPI performance to specific bottlenecks and opportunities

  • ✓ Charts use Distance to Frontier (DTF) scores and peer comparators

  • ✓ Robustness checked by comparing results of WEF GCI and World Bank Doing Business indicators, and official and empirical data

Refer to Annex I in staff report, including:

uA01fig10

Doing Business, Distance to Frontier, 2010 and 2016

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

uA01fig11

Global Competitveness Indices, 2012and 2016

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Cabo Verde Staff Report for the 2018 Article IV Consultation

  • ✓ Perception-based TPIs acknowledged

  • ✓ Uses point scores rather than rankings

  • ✓ Peer comparators not identified individually

  • ✓ Acknowledges key limitations of the TPI

1. Cabo Verde has made significant progress over the last few decades in economic and social development. Strong economic growth buoyed by a rapid expansion in tourism catapulted the country to middle income status. In the process, the incidence of poverty and extreme poverty, as well as inequality declined.1 Cabo Verde has also made significant progress over the last two decades in improving governance and fighting corruption. However, during the global financial crisis, economic growth fell sharply as tourism receipts declined. Meanwhile, a significant scaling-up of public investment since 2009, and the support of loss-making SOEs brought public debt to 129.5 percent of GDP in 2016, twice the level at end-2009.

Text Figure 1.
Text Figure 1.

Cabo Verde: WGI – Control of Corruption Index for SSA1/

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

1/ The index captures perceptions of the extent to which public power is exercised for private gain. The rankings used are subject to uncertainty a round the point estimate. Rankings reflect the relative, not the absolute performance of the country.Source: World wide Governance indicators (WGI) is dataset summarizing the views on the quality of govemance provided by a large number of enterprise, citizen and expert survey respondents in industrial and developing countries

WHD Regional Economic Outlook: Seizing the Momentum, April 2018, Chapter 2

  • ✓ Perception-based TPIs acknowledged

  • ✓ Key characteristics of indicators and their source data discussed

  • ✓ Uses correct attribution for Worldwide Governance Indicators

  • ✓ Uses point scores rather than rankings

  • ✓ Peer comparators identified as grouping rather than individually

  • ✓ Robustness clarified

Figure 2.20.
Figure 2.20.

Income-Adjusted Corruption Perception Levels, 2016

(Writer values = lower perceived levels of corruption)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Sources: World Bank, World Development Indicators database: Worldwide Governance Indicators by Daniel Kaufmann (Natural Resource Governance Institute and Brooking: Institution) and Aart Kraay (World Bank): and IMF staff calculations.Note: Simple average. Income-adjusted perceived corruption denotes die residual coming from regressions of central of corruption point scores on real GDP per capita. Using other indicators of corruption perceptions produces similar results. The charts show results based on control of corruption given its wide country coverage. As with any perception indicators, point estimates are subject to uncertainty. More details of the results will be a part of a forthcoming book chapter. ADV = advanced economies: Asia = emerging and developing Asia; CIS = Commonwealth of Independent States; EME = emerging and developing Europe; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SSA = sub-Saharan Africa.

EUR Regional Economic Outlook: Europe Hitting Its Stride, November 2017, Chapter 2

  • ✓ Perception-based TPIs acknowledged

  • ✓ Key characteristics of indicators and their source data discussed

  • ✓ Choice of indicator justified

  • ✓ Peer comparators identified as grouping (average) rather than individually

  • ✓ Uses point scores rather than rankings and considers sub-components of index

Figure 2.5.
Figure 2.5.

Estonia: Factors Affecting Institutional Quality1

(Index; 0 = worst, 10 = best)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Sources: Varieties of Democracy Institute (V-Dem) database (version B.2); and IMF staff calculations.Note: EU15 countries are listed in text footnote 1.1 Equal distribution of resources measures poverty and the distribution of goods and services as well as the levels of inequality in these distributions and the proportion of the population ineligible for social services.

To have quantitative indicators over a long period for more countries and dimensions, and following most previous studies, we also employ data from the World Bank’s Worldwide Governance and Doing Business Indicators, the World Economic Forum, the Varieties of Democracy Institute, and other sources (Annex 2.2). Most of these data are perception based and thus more subjective than other economic indicators. Nevertheless, economic decisions are based on agents’ perceptions of many factors, including governance, effectiveness of the judiciary, and property rights protection. CoE 2015 notes that “…other factors, such as public perception, political culture and safeguards against corruption have a clear impact on the ability of courts and judges to command legitimacy and do their job.” The case studies that follow here rely on many sources to understand the context in which judicial reforms took place.

See also:

Annex 2.2 Indicators and Sources

In line with the IMF Board paper on the “Use of Third-Party Indicators (TPIs) in Fund Reports” (IMF 2017c), this annex describes the indicators used and their sources. Specific descriptions of indicators used are described in Annex Table 2.2.1.

Review of 1997 Guidance Note on Governance—A Proposed Framework for Enhanced Fund Engagement

  • ✓ Perception-based TPIs acknowledged

  • ✓ Key characteristics of indicators and their source data discussed

  • ✓ Choice of indicator justified

  • ✓ Robustness checked by comparing results of WGI-CCI and Transparency International GCB

Refer to Annex 3. Corruption/Governance and Inclusive Growth: Empirical Analysis

The baseline estimates use the Control of Corruption Index scores from the Worldwide Governance Indicators Control of Corruption Index (WGI-CCI), which captures perceptions of corruption. The robustness of the results is checked using the Transparency International Global Corruption Barometer (GCB) indicator—which assesses people’s direct experiences with bribery and details their views on the overall prevalence of corruption in the main institutions in their respective country.2

7 The Doing Business 2017 report notes that the three-year update in methodology adds postfiling processes to the paying taxes indicator and a gender component in three indicators, as well as a pilot indicator on selling to the government. Somalia is added to the list of countries for the first time, increasing the total countries to 190,8 The 2017 report notes that “Compared to previous years, there is a lower number of top improvers from Sub-Saharan Africa even though this region accounts for over a quarter of all reforms globally.”22 They indicated that the Global Doing Business Survey provides some insights on where the FGS should focus reforms. They are not surprised by the result and ranking of Somalia as the country is a war-torn fragile state. They, however, have reservations about information related to several indicators, such as corporate income tax female employees on maternity leave; and inspection requirement during construction, as they apply to Somalia.6 Survey-based indicators reflect investors’ perceptions on the business environment.7 The World Economic Forum’s Global Competitiveness Index combines both official data and survey responses from business executives on several dimensions of competitiveness.5 These indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.2 The WGI-CCI indicator captures perceptions of the extent to which power is exercised for private gain, including petty and grand forms of corruption, as well as “capture” of the state by elites and private interests. It is compiled and published by staff from the World Bank, Brookings Institutions, and Natural Resource Governance Institute, aggregating from more than 30 different surveys of enterprise, citizen, and expert perceptions. The GCB draws on survey feedback from more than 100,000 respondents in about 100 countries. It addresses people’s direct experiences with bribery and details their views on the overall prevalence of corruption in the main institutions in their countries. It also captures their perception of the government’s anti-corruption performances, amongst others. One shortcoming of the GCB surveys is that they capture the experiences of the population at large, which typically relates to street-level corruption, not grand corruption by high-level officials. Another limitation is that differences in reported bribery might in part be subject to cultural differences in respondents’ willingness to report it. The WGI-CCI is available annually for 1996–2016 and cover about 200 countries, whereas GCB is available for the period2003–16 and covers less than 100 countries in each year.

Examples of Practices to Avoid

Example 1

  • x Does not use peer groups for cross-country comparisons, instead citing individual countries (identified by name)

  • x Establishes ranking by presenting point scores measuring corruption for individual countries in bar chart

  • x Does not reflect confidence intervals, which are available from compiler

  • x Omits information from original source that is relevant to the interpretation of the data

uA01fig12

Perceptions of Corruption

(Point score)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Example 2

  • x Does not provide additional context when discussing business climate and simply relies on ranking

  • x Does not discuss authorities’ views, despite the authorities having raised several methodological issues and additional context with a bearing on the interpretation of their performance according to the TPI

The business climate in Country X is poor. According to the Doing Business report, the World Bank ranks Country X 120 out of 189 countries. This reflected a deterioration compared to the prior year’s report when Country X ranked 110 out of 189 countries. Country X should prioritize reforms to improve its business climate.

Example 3

  • x Does not acknowledge perception-based TPIs

  • x Omits justification for identifying multiple individual countries in another country’s Article IV report, which is not otherwise self-evident

uA01fig13

Competitiveness

(1 =minimum, 7=maximum)

Citation: Policy Papers 2018, 044; 10.5089/9781498309622.007.A001

Source: World Economic Forum Global Competitiveness Index

References

  • Cerdeiro, Diego A., and Rachel J. Nam, 2018, “A Multidimensional Approach to Trade Policy Indicators,” IMF Working Paper No. 18/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2015, “Union of Comoros: 2014 Article IV Consultation—Staff Report,” IMF Country Report No. 15/4 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2014, Updated Guidance Note on the Fund’s Transparency Policy. (Washington: International Monetary Fund).

  • IMF, 2015, “Myanmar: 2015 Article IV Consultation—Staff Report,” IMF Country Report No. 15/268 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, November 2017, “Europe Hitting Its Stride,” EUR Regional Economic Outlook (Washington: International Monetary Fund).

  • IMF, 2017, “Georgia: Request for Extended Arrangement Under the Extended Fund Facility and Cancellation of Stand-By Arrangement—Staff Report,” IMF Country Report No. 17/97 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2017, “Hungary: 2017 Article IV Consultation—Staff Report,” IMF Country Report No. 17/123 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2017, “Somalia: 2017 Article IV Consultation and First Review Under the Staff-Monitored Program—Staff Report,” IMF Country Report No. 18/7 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2017, “Use of Third-Party Indicators in Fund Reports,” IMF Policy Paper No. 17/272 (Washington: International Monetary Fund).

  • IMF, 2018, “Cabo Verde: 2018 Article IV Consultation—Staff Report,” IMF Country Report No. 18/104 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2018, “Colombia: 2018 Article IV Consultation—Staff Report,” IMF Country Report No. 18/83 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, 2018, “Review of the 1997 Guidance Note on Governance—A Proposed Framework for Enhanced Fund Engagement,” IMF Policy Paper No. 18/55 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • IMF, April 2018, “Seizing the Momentum,” WHD Regional Economic Outlook (Washington: International Monetary Fund).

  • IMF, 2018, “Updated Third-Party Indicators Digest,” (Washington: International Monetary Fund). (Unpublished)

  • Trapnell, Stephanie E., 2015, User’s Guide to Measuring Corruption and Anti-Corruption, ed. by H. Feigenblatt, D. Torres, and A. Timilsina (New York: United Nations Development Program)

    • Search Google Scholar
    • Export Citation
1

This framework is also relevant for indicators developed by Fund staff that rely on TPIs as inputs (e.g., Economic Policy Uncertainty indices and “A Week at the Beach” index). Additionally, TPIs are used across the Fund for various other analytical purposes—for example, as an input in cross-country diagnostics (e.g., in the External Balance Assessment model).

2

For example, the Worldwide Governance Indicators (WGI) are sometimes mistakenly attributed to the World Bank. In fact, they are produced by Daniel Kaufmann—affiliated with the Brookings Institution and the Natural Resource Governance Institute (NRGI)—and Aart Kraay, affiliated with the World Bank. The compilers acknowledge financial support from the Bank, but note that the WGI does not reflect the official views of the Bank, Brookings, or the NRGI.

3

The Worldwide Governance Indicators, for example, are a composite TPI relying on a variety of inputs from other sources, including Transparency International Global Corruption Barometer Survey, Political Risk Services International Country Risk Guide, and Varieties of Democracy Project.

4

In the case of staff reports for program requests and reviews, because the authorities and staff will have agreed on the policy framework, there is no need to reflect the authorities’ views separately.

5

World Economic Outlook, Global Financial Stability Report, and Fiscal Monitor.

6

This approach normalizes the indicator relative to the best performance of all economies over a period. For example, Distance to Frontier (DTF) scores are reported by the World Bank for its Doing Business indicators. For more information on the Bank’s DTF methodology, see Doing Business 2018.

7

The Indicators Digest was originally circulated to the Board together with the policy paper “Use of Third-Party Indicators in Fund Reports” for the November 2017 Board meeting. It was revised in May 2018 to respond to feedback from the Board (e.g., noting each indicator’s type of source data, as well as providing links to source related research) and to incorporate new guidance on the use of TPI rankings on corruption.

8

The initial list of 13 TPIs include the International Country Risk Guide, main political risk indicators of the Economist Intelligence Unit, the Corruption Perception Index by Transparency International, the Worldwide Governance Indicators, the Corruption Index by Verisk Maplecroft, the Index of Economic Freedom by Heritage Foundation, the Global Competitiveness Index by World Economic Forum, the EBRD Transition Indicators, the World Bank Doing Business Indicators, Central Bank Independence and Transparency Indicators, and Consensus Forecasts.

Guidance Note for the Use of Third-Party Indicators in Fund Reports
Author: International Monetary Fund. Strategy, Policy, &, Review Department, and International Monetary Fund. Statistics Dept.