1. The 2000-02 comprehensive review of the Fund’s conditionality culminated in the adoption of a set of conditionality guidelines by the Executive Board on September 25, 2002 (http://www.imf.org/External/np/pdr/cond/2002/eng/guid/092302.htm).¹ The 2002 Conditionality Guidelines (the “guidelines”) replace the 1979 Conditionality Guidelines and the Interim Guidance Note on Streamlining Structural Conditionality of September 18, 2000. A review of experience with the guidelines, looking at how they have been implemented and at their impact, is conducted periodically; summary statistics on conditionality have been prepared annually since 2008.

2. As discussed in the staff statement attached to the guidelines, the key principles that should guide the Fund in designing and setting conditionality are: (i) national ownership of reform programs, (ii) parsimony in program-related conditions; (iii) tailoring of programs to a member’s circumstances; (iv) effective coordination with other multilateral institutions; and (v) clarity in the specification of conditions. Further, the guidelines require that program-related conditions will generally be established only on variables and measures that are critical to achieve the program goals, critical for monitoring program implementation, or necessary for implementing specific provisions under the Articles of Agreement.

3. This note elaborates on the operational implications of these principles and the guidelines.² It is not intended as a comprehensive stand-alone guidance note, but rather as an aid to the implementation of the guidelines. The operational issues covered are grouped into three areas: (i) the process of negotiation and program design; (ii) the design of conditionality; and (iii) the presentation of conditionality in Board papers for the use of Fund resources.

4. Since the guidance note was first issued in 2003, it has been revised on three occasions. A revision undertaken following the 2005 review of the guidelines underlined the importance of ownership in the success of Fund-supported programs, and provided enhanced guidance on the design of conditionality. The 2008 revision implemented the Board’s recommendation to strengthen efforts to achieve parsimony by emphasizing criticality as well as requiring rigorous justification of conditionality. Finally, the 2009 revision reflects the reform of the Fund’s conditionality framework in March 2009 to discontinue performance criteria for structural measures in all Fund arrangements effective on May 1, 2009, including those under facilities designed for low-income countries, and to place greater reliance on a review-based approach to monitor structural reforms in Fund-supported programs. ³

I. Negotiation and Program Design Process

5. The guidelines indicate that: (i) authorities have the lead role in designing their programs; (ii) the Fund should encourage members to build broad support for sound policies; and (iii) Fund-supported programs and conditionality should be coordinated with other international institutions in countries where those institutions are also involved.

6. The guidelines establish the principle that the primary responsibility for the design of the program lies with a member’s authorities. The intent is to promote ownership by ensuring an active dialogue at all stages in the process of developing, implementing, and monitoring a Fund-supported program. This approach is intended to foster greater flexibility in program design, and hence greater ownership, without weakening conditionality or the quality of Fund-supported programs, for which the Fund retains responsibility. There is no requirement that country authorities actually draft the Letter of Intent (LOI) and Memorandum of Economic and Financial Policy (MEFP), but staff should be responsive to the authorities’ desired role in the drafting of these documents. As with other aspects of the guidelines, this may require allowing more time for program formulation. The guidelines implicitly recognize that experience in this regard will vary, and that the authorities might have limited capacity or inclination to draft program documents. Drafting technical memoranda will continue to require a high degree of collaboration between staff and country authorities.

7. In helping the authorities to design a Fund-supported program, staff should (i) seek early on the views of country authorities and make every effort to accommodate their preferences while ensuring that the Fund-supported program goals are achieved and (ii) follow a program preparation process that can facilitate reflection on the links between program goals, strategies, and conditionality. In particular:

• To develop the program goals of the Fund arrangement, staff could build on issues identified in previous discussions with the authorities, including Article IV consultations, technical assistance missions, staff visits, ex-post assessments, reports under the Financial Sector Assessment Program and bilateral meetings. These goals are understood to represent a subset of the authorities’ broader goals, or intermediate steps toward a longer term goal, and should be identified as specifically as possible. It is expected that goals that extend beyond the program period would be addressed in the context of a potential successor arrangement or in some other dialogue with the authorities, such as in the context of surveillance. Further, reform strategies to be pursued should be identified early on, including any specific structural measures that are critical to the achievement of program goals. When possible, staff should also consider appropriate reform sequencing drawing on analytical work on program design and reform complementarities. Modifications to the program design, including on reform strategies and related conditionality, may be required as new information becomes available during negotiations and the authorities’ views are firmed up; indeed, the guidelines emphasize the importance of flexibility to allow considerations of ownership, implementation capacity, and members’ circumstances. When time pressures on negotiations are high, the ability to provide for such flexibility will necessarily be constrained.

• The preparation process for program approval should be used to help ensure that the conditionality is parsimonious and well matched to tightly specified program goals. In this process (including in Policy Consultation Meetings), staff should sketch out the reform strategy that is viewed as critical to achieve the goals supported by the Fund arrangement and, to the extent possible, anticipated structural conditionality should be indicated. Policy Notes (PNs) for negotiating missions should, when possible, outline the authorities’ initial proposals and the proposed staff response, including the options staff will invite the authorities to consider and the rationale for criticality of proposed conditionality. When the staff is not in a position to report the authorities’ views, this should be made clear in the PN. The purpose of the above program preparation process is to define as early as possible the contours of the program, including in terms of the links between program goals, reform strategies, and supporting conditionality. As noted above, flexibility in modifying the program design during negotiations will remain essential.

8. Staff should assist members in broadening support for sound policies. Participatory processes are a requirement in the formulation of Poverty Reduction Strategy Papers (PRSPs), most Fund arrangements under the PRGT, and have been adopted in many other program countries. Staff should support the authorities’ efforts to engage in a transparent participatory process in developing a policy framework, and should continue to be prepared to assist the authorities in this process including by giving seminars, meeting with various interest or political groups (parliamentary committees, business groups, and civil society organizations) and by being available to the media. In doing so, staff should be aware of the authorities’ views on staff contact with domestic groups and, if necessary, seek their prior agreement. Resident representatives could be particularly useful in providing this assistance. Broadening political support for policies may require allowing more time for program formulation.

9. Explicit assessments of potential implementation difficulties should be included in PNs. Assessments should reflect specific challenges to broader ownership in a country as well as other country-specific issues such as constitutional and political structures. Such assessments take on particular importance when implementation of key measures hinges on legislative approval or on actions by lower levels of government (e.g., fiscal adjustment by sub-national governments or reforms in state-owned enterprises with significant autonomy), rather than on executive actions alone. In some instances, where implementation requires cooperation by several ministries, the positions of these ministries are relevant to the assessments. To help ascertain prospects for successful implementation, the approval of specific measures by the entity responsible for the decision (e.g., the cabinet, the Parliament, sub-national governmental bodies) may be specified as prior actions. In some cases, where there are doubts as to whether a particular critical measure would be implemented after approval, it should be made a prior action (see also paragraph 19).

10. PNs should include an assessment of technical capacity, including capacity building in the program. PNs and staff reports should explicitly identify key capacity weaknesses, explain how those weaknesses may affect the ultimate design of the program, and highlight the measures under the program to build capacity in those areas.

11. Collaboration with other multilateral institutions. The staff should consult, as needed, with the staffs of other multilateral institutions involved in the country early on in the program design process. PNs should provide details of coordination in the areas of overlap, including a clear delineation of analytical responsibilities (in line with the lead agency principle). Continued close communication will be essential to adapt coordination as the Fund-supported program is being implemented (see also paragraph 28 for a discussion on Bank-Fund collaboration).

II. Designing Conditionality

12. Program-related conditions generally must be either critical to the achievement of program goals, to monitoring implementation, or necessary to the implementation of specific provisions under the Articles of Agreement. The 2002 Guidelines replace the somewhat weaker standard of macro-relevance in the earlier Interim Guidance Note on Streamlining Structural Conditionality (September 2000). Leaving aside conditions necessary for implementation of provisions under the Articles, a judgment that a condition is of critical importance means that if it was not implemented, it is expected that the goals would not be achieved or that program monitoring would not be possible (and, conversely, all critical measures generally must have conditionality associated with them, because the Fund needs to be able to interrupt purchases or disbursements if the program is off track or if sufficient information is not being provided to adequately monitor program implementation). In judging that a condition is critical, staff affirms that, other things being equal, the expectation is that the condition should not be eliminated later. In particular, staff must avoid setting conditions on measures that may be desirable, but that are not critical for achieving the program goals or for monitoring implementation of those goals.

13. The criticality criterion applies to all measures—whether they are in the Fund’s core areas or outside, or whether they are or are not covered by another agency’s (e.g., the World Bank’s) conditionality. Recognizing that conditions that are critical to the success of Fund-supported programs are more likely to be in the Fund’s core areas than outside, the focus of structural conditionality should be primarily on core Fund competencies. For conditions outside these core areas, the guidelines require more detailed explanation of their criticality along with a strong justification. Moreover, for non-core but critical measures, the Fund will, to the extent possible, draw on the advice of other multilateral institutions, particularly the World Bank, or of bilateral donors that can provide the expertise. When the expertise in a critical area of reform is not available within the Fund, or other multilateral institutions and bilateral donors, the authorities would need to obtain the required expert input from others with the necessary expertise. In cases where the necessary expertise from these other sources is either unavailable or judged inadequate, approving a Fund-supported program would expose the Fund to reputational and financial risk. In such circumstances, the Fund would be faced with choosing between these risks or not supporting the authorities’ program.

14. Staff should apply the principles of parsimony and criticality to all program-related conditions. Staff should refrain from introducing conditionality that is requested by donors if it is not critical for program success. To the extent that the addition of a donor-driven reform measure as a program-related condition is indispensable to ensuring program financing and thus program success, such measures should be clearly identified in the staff report and their criticality justified accordingly. As for conditions requested by the authorities, program ownership is clearly of utmost importance for program success. Nevertheless, reform measures that are not critical for achieving the program goals or for monitoring implementation but that the authorities wish to highlight could be addressed in the LOI/MEFP. In particular, a clear distinction should be drawn between conditionality upon which Fund financing is contingent and other elements of the authorities’ reform agenda noted in the LOI/MEFP.

15. Level of detail of conditions. Conditions should be formulated in ways that provide the authorities with clarity as to the measures on which access to Fund resources depends. Conditions formulated at a high level of detail may provide helpful guideposts to the authorities, but run the risk of being seen as micromanagement. These considerations must be balanced on a case-by-case basis.

16. The scope of reviews. Reviews have backward- and forward-looking components. The policy areas on which the backward looking component of the review is to focus should be specified, to the extent possible, in the form of quantitative performance criteria, indicative targets and structural benchmarks. Reviews are also forward-looking, providing the opportunity to assess prospects for achievement of program objectives in light of a range of available information. This may entail modifying program objectives and measures as new information becomes available; for example, if additional adjustment is needed owing to changes in economic conditions. To limit the possibility that reviews will be used to add conditionality unless clearly justified by changes in circumstances (e.g., the existence of a new PRSP or economic developments that require a reassessment of priorities), the original program documents should give, to the extent possible, a broad sense of the uncertainties that could require additional or modified conditionality during reviews. Conditionality beyond the first year of a multi-year program will normally be set in the context of reviews. Conditions added during reviews should be anchored on the initiating program document and those added in new reform areas would require a clear justification as to why the related measures had become critical to achieving the program goals. Finally, while the last program review may not always be the most opportune time to assess the outcome of a Fund-supported program, staff could use such reviews for stocktaking and, where possible, provide a preliminary assessment of stated program goals and their achievement and of the experience with program implementation.

17. Review-based approach to structural conditionality. Since May 1, 2009, monitoring of performance on structural reforms under Fund-supported programs is conducted through reviews, with all structural conditions now to be associated with a review and to take the form of structural benchmarks or prior actions; structural performance criteria are no longer used. Reviews provide the primary tool for monitoring performance on the structural elements of all Fund arrangements. The discontinuation of structural performance criteria seeks to ensure that an excessively narrow or rigid focus on specific criteria is avoided. Under the review-based approach, all Fund-supported programs should include a specified agenda for critical structural reforms, if relevant.

18. Use of structural benchmarks under the review-based approach: Structural benchmarks should be critical for the achievement of program objectives, either by themselves or because they represent key components of a broader reform measure that is judged to be critical (e.g., steps in a tax reform or privatization program). A member’s failure to meet a structural benchmark does not by itself automatically interrupt a purchase or a disbursement under an arrangement. Rather, deviations serve as indicators that the Fund-supported program may be off track. Completion of the review would then require a judgment by the Board that there are factors giving confidence that program objectives are being achieved. In case of substantial deviations from structural commitments and weak policy commitments to correct slippages, staff and management could decide not to propose completion of a review. Informal country matter sessions provide a forum to inform the Board about the status of discussions. Once new understandings are reached, the review could then be completed, allowing the member to make the associated purchase or disbursement.

19. Use of prior actions. A critical measure may be specified as a prior action to be completed before the scheduled date of a Board discussion to approve an arrangement, complete a review, or grant a waiver, when it needs to be implemented immediately in order to achieve program goals or monitor implementation, including when there are significant doubts that the measure would be implemented at a later date.⁴ Conditions set for the completion of future reviews are not prior actions and should not be referred to as such; rather, they are either quantitative performance criteria or structural benchmarks. As with other forms of conditionality, prior actions are to be applied parsimoniously and must be justified in terms of their criticality to program objectives. Prior actions are often used prominently in cases of countries with weak track records of implementation, and can still play a useful role by ensuring that conditions are met and that programs meet a minimum implementation standard. However, staff needs to be mindful of the possibility that prior actions may be implemented without genuine ownership of the program, and that implementation may be nominal or temporary as a result.

20. Waivers of quantitative performance criteria. A waiver may be granted for nonobservance of a quantitative performance criterion if the Fund is satisfied that the program will nevertheless be successfully implemented—i.e., that it will achieve its goals—either because of the minor or temporary nature of the nonobservance or because of corrective actions taken by the authorities. While the language on waivers in the guidelines does not explicitly refer to cases where the Fund considers that a performance criterion is no longer critical (whether because circumstances have changed or because it judges that it previously erred in the design of conditionality), these could be cases where the Fund remains satisfied that the program will be successfully implemented and the deviation in such cases would be considered “minor” in terms of its impact on program goals.

21. Outcomes-based conditionality. The guidelines explicitly note that program-related conditions may be set on targets and objectives as well as on actions. However, outcomes chosen as conditions must still be reasonably under the control of the authorities, so that exogenous shocks do not force undue use of waivers and thereby lessen assurances to the member of the availability of Fund resources. Moreover, outcomes-based conditionality should be designed to minimize the risk of situations in which targeted outcomes are achieved by means of policies that would undermine the achievement of program goals, such as reaching an intermediate outcome-based target of higher revenue by increasing import tariffs. Even when conditions cannot be applied to outcomes, there may be scope for formulating conditions in such a manner that they may be met by implementing various measures.

22. Implementation timetables should be realistic but still appropriately ambitious. While the authorities should be free to set demanding timetables where they consider them helpful in driving their agenda forward, the staff should not press for overly ambitious timetables. Staff often faces constraints in forming an accurate judgment on implementation capacity and political priorities. Overly demanding timetables have also been a major reason for frequent requests for waivers.

23. Floating tranches. The use of floating tranches is also explicitly allowed for in the guidelines as another possible device available to staff in the design of conditionality. The availability of a floating tranche is contingent on implementation of a specific measure or measures, which has no fixed date attached, while the other purchases or disbursements are, as is normally the case, governed by conditions tied to test dates. Floating tranches may enhance ownership by giving authorities greater flexibility in choosing the timetable on which reforms, particularly structural reforms, are implemented, especially where there is some unavoidable uncertainty about the timing of the measure or where an upfront commitment on specific timing is for some reason not desirable. The internal logic of floating tranches suggests that the measures to which they are attached should satisfy two criteria: first, they must strengthen the external position over the medium term, so as to warrant the release of additional Fund resources; but they must also increase the balance of payments need in the short term—otherwise, the program would be either underfinanced without the relevant measure, or overfinanced with it. Examples of measures that could satisfy these criteria include trade liberalization and debt restructuring needs. Structural reforms that might be appropriate for floating tranche conditionality are not expected to occur in very many cases, but it is therefore all the more important that the option of using them be considered when the above criteria are met.

III. Presentation in Board Papers

24. Discussion of program design issues in Board papers should provide a clear explanation of the choice of conditionality, in particular the judgments of criticality supporting the inclusion of particular structural measures.

• Since the judgment of criticality depends upon program goals and the strategies adopted to achieve them, these should be set out as clearly as possible. At the time of approval of a new arrangement, the staff report should lay out the links between the program goals and reform strategies, and the corresponding structural conditionality. One option is for these links to be presented in a separate text table in the staff report.⁵ The program goals should be specified as clearly as possible. The staff report should avoid conflating the objectives of the Fund-supported program with the authorities’ broader goals, and should clearly identify and explain the strategies, or broad areas of policy action, embodied in the program. Further, all future program modifications should be anchored in the originating program document and deviations from this document should be justified. Where possible, staff reports should convey the discussion between staff and authorities on policy options.

• All staff reports need to indicate as clearly as possible what is expected in the subsequent review(s) and specify critical future actions as structural benchmarks. This presentation is essential to ensure that the scope of the review is as predictable and transparent as possible, providing assurances to members regarding access to Fund resources.

• Staff reports for program reviews should provide an update related to the objectives of the structural reform agenda, as well as the strategies adopted to achieve them. The staff report should also assess if the structural elements of the program have been implemented successfully, i.e., that the program is on track for achieving the associated objectives.

25. More detailed explanations are needed for conditions outside Fund core areas. Conditions may be established on measures outside the Fund’s core areas of expertise, but such conditions would require detailed explanations of their criticality along with strong justification. This reflects the general presumption that measures in non-core areas are less likely to pass the test. Further, staff reports should provide the necessary information as to which institution or donor provided expert advice. When the expertise in a critical reform area is not available within the Fund or other multilateral institutions and bilateral donors, and staff relies on the authorities to obtain the required expert input from other sources, the risks noted in paragraph 13 should be transparently reported in staff reports.

26. Staff’s assessment of potential implementation difficulties should be discussed. A candid discussion on concerns about ownership will be expected in staff reports (as noted in paragraph 9 for PNs). It will be important to acknowledge concerns about ownership which would increase risks to program implementation from the outset to enable the Board to make informed judgments. In these cases, staff will need to convey the rationale for going ahead, including on the basis of the implementation of key measures necessary to improve prospects for program success. Care will have to be taken to ensure that a discussion of ownership does not undermine confidence in the program.

27. Program conditions should be clearly identified and distinguished from other measures in program documents and staff reports. Tables for quantitative performance criteria and indicative targets are already required and are a standard part of Fund practice. All structural conditionality is also routinely identified in all program staff reports. An effective means of doing so is to include tables in the LOI/MEFP listing all quantitative performance criteria, indicative targets, prior actions and structural benchmarks. If the authorities wish to list measures that are not conditionality, these should be listed separately and identified as part of the authorities’ reform agenda. These can be addressed in the text of the LOI/MEFP or in a text table attached to the LOI/MEFP, thus offering the authorities a venue for describing their broader goals, but should be clearly distinguished from conditionality on which Fund financing depends. The status of all past performance criteria, indicative targets, prior actions, and structural benchmarks should also be reported, and discussed in more detail for unmet or delayed conditions. This can also be done in tabular form. Staff reports should include clear justifications for waivers of performance criteria.

28. Bank–Fund collaboration. A discussion of collaboration with other multilaterals, in particular the World Bank, has also become a requirement of program staff reports, where applicable. While much of this information should be contained in the annex on Bank-Fund collaboration, particularly relevant information should be brought into the main staff report. Staff reports should also discuss how the Bank- and Fund-supported programs have been coordinated – for example, where the Fund macro framework has been modified to accommodate Bank-supported reform programs, where reforms under a Bank-supported program have been chosen and sequenced to contribute to Fund-supported program goals, etc. Monitoring responsibilities of the multilateral institutions should be clearly delineated, while bearing in mind that the Fund bears the ultimate responsibility for establishing and monitoring its conditionality. (See the framework provided in the Joint Management Action Plan for strengthening Bank-Fund collaboration in areas of overlap. ⁶