## 1. Determinants of the Probability of a High Number of Structural Conditions ## Source details **Canonical URL:** [1. Determinants of the Probability of a High Number of Structural Conditions](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2012/_061812.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2012/_061812.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2012/_061812.pdf.json) --- ### Scope and analytical framework - Review period: 2002 to end-September 2011. - Focus: content and application of program conditionality—especially structural conditionality—in relation to the 2002 Conditionality Guidelines, the Staff Statement on Principles Underlying the Guidelines on Conditionality, and subsequent revisions to operational guidance on conditionality. - Core design principles analyzed: - national ownership of programs; - parsimony in program-related conditions (operationalized by the "criticality criterion"); - tailoring to country circumstances; - effective coordination with other multilateral institutions; - clarity in the specification of conditions. - Criticality criterion: program-related conditions should be critical (minimum necessary) to achieve program objectives and goals, critical for monitoring program implementation, or necessary for implementing specific provisions under the Articles of Agreement. ### Data sources and methods - Quantitative base: Monitoring of Fund Arrangements (MONA) database. - Qualitative sources: - 23 Ex Post Assessments of Longer-Term Program Engagement (EPAs) and Ex Post Evaluations of Exceptional Access (EPEs); - 18 case studies (2006–10); - structured interviews with Fund Executive Directors; - surveys of country authorities, in-country donors, Fund mission chiefs and Resident Representatives (countries with programs since 2007); - inputs from headquarters-based donors, civil society organizations, and other stakeholders. - Program coverage: SBAs (including HAPAs), EFF in the GRA, PSI, SCF, ECF in the PRGT, ESF-HAC (upper credit tranche conditionality). Excludes ENDA, EPCA, ESF-RAC, RCF, and staff-monitored programs. ### Key findings (high level) - Conditionality became more focused and generally aligned with program goals. - Tailoring to country and program characteristics, and to initial macroeconomic conditions, improved. - Conditionality became more parsimonious overall. - Number of conditions per review and per year of ongoing program declined in recent years; discontinuation of SPCs supported this. - Structural conditionality increased somewhat in GRA programs in 2011. - Focus shifted increasingly toward Fund core competencies; fiscal measures gained prominence. - Ownership was critical to program success. - Frontloaded conditionality can support implementation but cannot substitute for ownership. - Ownership supported by outreach and communication, program flexibility, focused conditionality, and increased clarity. - Some evidence suggests ownership improved on balance, with possible new issues in 2011. - Coordination with other institutions was generally supportive but posed challenges in some euro-area crisis programs. - Implementation progress stalled in 2011, likely reflecting more challenging economic conditions and adjustment needs. - Clarity of conditionality was uneven; best-practice presentation and linkage to objectives were not uniformly applied. ### Evolution of operational guidance and instruments - Revisions and milestones: - 2003 operational guidance note; subsequent revisions in 2006, 2008, and January 2010 (Operational Guidance Note on Conditionality—December 2009 Revisions). - 2008 revision emphasized criticality and required rigorous justification of conditionality. - March 2009 reforms and early 2010 adjustments discontinued Structural Performance Criteria (SPCs) and shifted to a review-based approach to monitor structural reforms, aiming to: - enhance flexibility; - tailor conditionality to country characteristics; - reduce stigma associated with SPCs and waivers; - preserve safeguards for use of Fund resources. - Review expanded scope relative to prior reviews by systematically studying Prior Actions (PAs) and Indicative Targets (ITs) and incorporating depth-of-conditionality analysis. ### Policy implications and forward-looking considerations - Parsimony could be tested in a more challenging global environment as deeper structural reform needs emerge, notably for the second wave of programs started in 2010. - Coordination with institutional partners will become more essential and more difficult; effectiveness improved if stakeholder engagement is set ahead of crises. - Improved communication and transparency on program goals, strategies, and linkages to conditionality could support ownership and donor coordination; mandating improved clarity in documentation recommended. - Surveillance can be better leveraged: case studies show about half of program structural conditionality was identified as needed reforms during pre-program surveillance. --- ### II. Streamlining and the principle of parsimony - Objective: Help countries solve balance of payments problems and safeguard Fund resources; program-related conditions should be (i) critical to achieve program goals or monitor implementation, or (ii) necessary to implement Articles of Agreement provisions. - Streamlining history: interim guidance note (2000) preceded 2002 Guidelines; parsimony central—conditions should focus on achievable goals and macroeconomic priorities (Guidelines, ¶6). - Streamlining does not preclude substantial conditionality where needed; non-core critical measures require strong justification and, where possible, reliance on other institutions’ expertise. ### A. Program goals, instruments, and conditionality: focus and alignment - MONA sample: 59 LICs and 31 AMs and EMs; period 2005-2011; programs approved after 2005. - Program goals classification: macroeconomic stability, external stability, economic growth, poverty reduction. - Empirical patterns: - All programs identified external and macro-economic stability as overarching goals. - GRA programs centered on external and macro-economic stability. - Share of programs including growth objectives rose from 40 to 60 percent between 2008 and 2011. - Almost all PRGT-supported programs included poverty reduction and economic growth goals. - Fiscal and monetary policies were the most frequently used instruments; GRA programs relied more on financial sector policies; PRGT programs used wider pro-growth and social sector policies. - Alignment findings: - Structural conditionality broadly aligned with program goals and tailored to country circumstances. - Around 80 percent of conditions on average could be linked to specific objectives (GRA and PRGT). - In PRGT countries, about 60 percent of objectives were addressed by structural reforms; in GRA countries, about 70 percent were addressed. - Case study exceptions noted where sequencing, demonstration of macro-criticality, or diagnostics of institutional/political constraints could be improved (examples cited in source). ### B. Focus of structural conditionality - Trend toward core Fund areas, especially fiscal measures. - Sectoral shifts: - Fiscal conditionality dominated structural reform agenda. - In PRGT programs, government-related structural measures were about 58 percent of total structural conditionality during 2008-2011—an increase of more than 7 percentage points compared to 2005-2007. - Financial sector reforms were second most important, notably in GRA programs; financial-sector focus peaked at crisis onset, then fiscal issues increased in prominence (including EU and Iceland programs). ### Determinants of structural conditionality by sector - Breakdown driven by vulnerabilities and fiscal space (Table 1 and Appendix 2). - GRA programs: - General-government/fiscal related conditionality more likely with better economic conditions (lower inflation, higher growth). - Financial sector conditionality more likely with larger exposures (assets and liabilities) and smaller fiscal deficits. - PRGT programs: - More general government conditions with higher debt levels, lower reserves, and smaller fiscal deficits. - Probit models link probability of a large number of conditions (highest quartile) to macroeconomic conditions at program approval; only statistically significant variables summarized in source. ### Public Financial Management (PFM) (Box 3) - PFM conditionality categories: (i) stocktaking and control of arrears; (ii) financial accounting/reporting/auditing; (iii) expenditure controls; (iv) legal/institutional frameworks; (v) advanced PFM reforms. - Sequencing and emphasis: - Arrears: build capacity to identify/verify/report/clear arrears, create prevention frameworks. - Accounting/reporting: improve coverage and quality of fiscal data (classification, timeliness, auditing). - Expenditure controls: budget execution reforms (commitment controls, active cash management), procurement, payroll systems, tracking poverty-related expenditure. - Legal/institutional: revise organic budget/procurement/debt laws; create monitoring units; strengthen audit functions. - Advanced reforms: treasury single accounts, integrated financial management information systems, integration of cash and debt management. ### Macro-social structural conditionality (Box 4) - Macro-social conditionality increased late in review period though remained a small share of total structural conditions. - Program-type differences: - GRA programs had more macro-social conditions (pensions, social sectors, labor markets). - PRGT programs more commonly included civil service reforms; pensions/labor-market reforms outside public employment limited. - Trade-offs: measures lowering fiscal expenditure/raising productivity can help BOP and yield long-term gains but may have short-term employment/income costs; some programs added safety nets (e.g., Latvia, Greece). ### Wage bill ceilings in PRGT programs (Box 5) - 2007 Executive Board request: use wage bill ceilings in LICs only in exceptional cases and with sufficient flexibility. - Observed practice: - Policy well followed: no QPCs set on wage bills starting in 2007; a few ITs monitored wage bills. - Structural conditionality related to wage bill ceilings dropped significantly; last PRGT program with structural conditionality on wage bills: Moldova (SPC in 2006, PA in 2009). ### Quantitative conditionality: focus and evolution - Quantitative conditionality focused on external and fiscal stability. - Time patterns: - Early period: focus on debt. - Global financial crisis: GRA monitoring shifted toward fiscal consolidation. - PRGT mix broadly unchanged with emphasis on external debt. - Monitoring coverage: - Fiscal and monetary stability monitored in almost all programs. - All PRGT programs included targets on external debt. - Use of reserves targets limited where no autonomous exchange rate policy exists. - Indicative targets (ITs): - Used mainly to monitor fiscal performance and social-sector indicators (2006-11 analysis). - Largest share of ITs related to fiscal deficit in PRGT programs. - Shift toward fiscal ITs in GRA programs in latter part of period. - Use of wage ceilings declined in both GRA and PRGT programs. - “Social” ITs (floors for priority spending) monitored performance in fiscal expenditure; poverty-related ITs increased after 2009. ### Volume of conditionality and counting methodology (Section D and Box 6) - Parsimony measured by number of conditions set over a program; primary counting approach = review-year approach (conditions counted in year reviewed). - Counting approaches: - Approval-year approach: counts conditions by year of program approval (less suited for recent/ongoing programs). - Review-year approach: conditions accounted for when reviewed (or expected review year for outstanding conditions); normalized by number of reviews or current programs per year—used in 2011 RoC. - Scope: all conditions set by Executive Board or Management up to September 30, 2011—PAs (Management), QPCs, SBs, ITs, SPCs (now discontinued). Excluded cancelled or modified conditions to avoid double-counting; continuous SBs/SPCs counted only when implementation status recorded in MONA. ### Key statistics and trends (preserved numeric statements) - Number of QPCs stable at around five to six per review (Figure 8). - Inter-quartile range: between four and six QPCs in GRA programs and between five and seven in PRGT programs. - Number of ITs broadly steady overall during 2006-11 (Figure 9). - Discontinuation of SPCs in March 2009 had a sizeable impact on number of structural conditions per review, notably in GRA programs. - Declines in SBs and PAs (both at approval and during reviews) occurred over most of the period. - Distribution across programs of structural conditions per review narrowed in latter part of period; number of structural conditions stabilized after 2009. - Conditions per review would be even lower if arrangements under FCL and PCL were included (no ex-post conditionality under FCL; limited ex-post under PCL). ### Trends in conditionality quantity and frequency - Number of conditions per review remained comparatively low in both GRA and PRGT programs; steady increase in SBs per review noticeable in GRA programs. - Reviews with larger numbers of conditions increased between 2009 and 2011, reflecting need to tackle deeper-rooted issues (competitiveness, large debt). - Total number of conditions per year picked up modestly starting in 2010 due to: - increase in number of conditions per review (GRA programs), and - greater frequency of reviews per year (GRA and PRGT). - More frequent adjustments indicate fine-tuning and a potential need to better leverage surveillance. ### Links between surveillance and conditionality (Box 7) - Sample (programs approved since 2008 with recent Article IV): 34 countries—23 (68 percent) GRA, 11 (32 percent) PRGT. - About 48 percent of program conditions were foreshadowed in previous Article IV consultations. - SBs and SPCs mapped more closely to Article IV recommendations than PAs. - PRGT conditionality generally better related to surveillance recommendations; GRA conditionality showed larger variation due to emergency needs. - Share of foreshadowed conditionality lowest in 2008 for GRA programs and in 2009 for PRGT programs; tended to increase by end of period. ### Depth of structural conditionality (Section E) - Depth = degree and durability of structural change likely if condition implemented (durability, not difficulty or criticality). - 2006-11 programs: increase in depth of structural conditionality. - Variance across country groups: - Advanced market economies programs on average structurally deeper than emerging markets. - GRA programs with European countries showed highest structural depth. - Recent euro-area programs (especially Greece) had structural depth above average. - LIC programs accounted for largest number of conditions but lower average structural depth. ### Trade-off between number and depth of conditions (Box 8) - Regression analysis: negative correlation between number of conditions per review and average depth of conditionality per review in both GRA and PRGT programs (2006-11). - Fixed-effects panel regression (2006-2011; conditions with test dates up to September 30, 2011) result excerpt: - A one standard deviation increase in depth correlated with a reduction in number of conditions by 1.2 and 0.7 in GRA and PRGT programs respectively. - Reported regression outputs (excerpt in source): - Depth coefficient: -2.23*** (GRA), -1.43** (PRGT) - Adjusted R-squared: 0.27 (GRAPRGT combined), 0.21 - Number of obs.: 111 (GRA), 187 (PRGT) - Note: *** and ** denote 1 percent and 5 percent significance levels. ### Ownership and conditionality (Sections III.A, Box 9, Box 10) - Ownership: member countries bear primary responsibility for policy selection, design, and implementation. - Streamlined, tailored, parsimonious conditionality supports ownership; complexity and capacity constraints can undermine ownership. - Empirical indications: - Less use of a large number of PAs at program approval in GRA programs over 2002-2010 suggests improved ownership. - Regression evidence: higher number of PAs at program approval associated with lower implementation ratio of structural conditionality over program life in GRA programs (statistically significant for >2 PAs); not significant in PRGT programs. - Probit marginal effects (selected excerpts in source): - Implementation ratio (t) marginal effect on probability of a PA: -0.0062** (GRA sample), -0.0089** (PRGT sample). - Implementation ratio (t-1): -0.0077** (GRA); 0.0011 (PRGT). - Actual P: 35.55 (GRA combined), 32.75 (PRGT) - Predicted P: 32.91 (GRA), 29.39 (PRGT) - P error: -2.64 (GRA), -3.36 (PRGT) - Note: Table reports marginal effects. P(y=1) denotes a review with at least one prior action. P-values in brackets. * p<0.1, ** p<0.05, *** p<0.01. - Case lessons: - Iceland and Latvia: strong local support for program goals helped overcome crisis; conditionality aligned with authorities’ priorities. - Belarus: weak ownership at top levels led to residual problems; quasi-fiscal activity increased LGP by almost 50 percent, undermining fiscal adjustment and program effectiveness. - Instruments to assess or enhance ownership: - Staff-monitored programs and PAs used as screening devices. - Flexible implementation timelines enhanced ownership; discontinuation of SPCs did not have a noticeable adverse impact on implementation overall (Appendix 4). ### B. Tailoring: capacity, access, and evenhandedness - Design differences reflect spectrum of initial conditions: wide-ranging reform agendas vs. short-term liquidity shocks; large-scale disbursements justify different conditionality. - Capacity implications: - Low-capacity countries may require simpler, phased conditions; high-capacity countries can manage deeper reforms. - Correlation between country capacity (CPIA rating) at approval and total number of conditions per program positive and statistically significant. - Probability of setting at least one high-depth structural condition increases with CPIA: an increase by one in CPIA rating increased likelihood of a deep condition by 20 percent (2005-2009 subset of 54 countries). - Access implications: - PRGT programs: access related to number of PAs for Board approval. - GRA programs: higher access related to number of deep conditions. - Upper-quartile access programs focused more on financial sector reforms; monitoring of higher-access arrangements emphasized fiscal issues. - High exceptional access GRA arrangements (access over 1000 percent of quota) and higher-access PRGT arrangements reflected need to ensure fiscal sustainability. - Debt limits in LIC programs: since December 2009 move toward menu of options for concessionality requirements. ### Survey assessments of tailoring and flexibility - Fund staff, country authorities, and donors generally viewed conditionality as tailored and flexible; staff more convinced of appropriateness and flexibility. - Weak diagnostics or missing follow-up conditionality reduced effectiveness in some cases. ### C. Coordination with partner institutions - Coordination with World Bank and regional institutions essential and generally effective; grew in importance in euro-area programs. - Sub-Saharan Africa: coordination focused on lead-agency principle; Joint-Management Action Plan and JMAP supported coordination. - Euro-area (Troika) programs: Fund, EC, ECB coordination supported program design/monitoring; EC structural agenda less parsimonious and more detailed than Fund’s; overlaps and expanded roles observed (e.g., pension reform specifics). - Survey: >80 percent of respondents noted intensive coordination with donors and effective coordination with World Bank. ### D. Clarity - Guidance calls for clear justification of conditionality and clear linkage between goals, instruments, and structural measures. - Case study assessment: - Most staff reports provided macro-criticality justification and discussed implementation risks but often lacked detailed presentation of policy options and full linkage structure. - Best-practice examples: 2009 Ghana (comprehensive table with measures and macro rationale), 2008 Togo (complete set of past/immediate/envisaged reforms), 2008 Pakistan (box linking structural measures to program objectives). - Non-core conditionality: - Roughly 40 out of 200 measures in 26 programs were non-core. - One third of GRA programs had non-core measures; about two thirds of PRGT programs had non-core conditionality. - Only in about one third of non-core cases was justification detailed; partner institution support rarely documented. - Presentation of policy options and implementation risks: - Nearly 90 percent of staff reports provided broad discussion of implementation risks; detailed discussion of policy options recorded in only about one quarter of case studies—survey suggests under-recording rather than absence. ### Program interruptions, implementation patterns, and determinants - Off-track definition: time between reviews > twice planned interval (more than six months for quarterly, more than one year for semi-annual). - Frequency and duration: - Average share of off-track programs around 15 percent (2002–end-September 2011). - Long-lasting interruptions ~6 percent. - ~60 percent of programs experienced no off-track period. - Interruptions became more frequent since start of global financial crisis; implementation weakened progressively between 2009 and 2011 in PRGT programs; deterioration notable in GRA in 2011. - Average length of interruption declined since 2007; PRGT off-track periods fell as a share of total program periods after 2007. - Implementation metrics: - Implementation measured as weighted average where: not met/waived = 0; partially met/met with delay = 0.5; fully met = 1 (ratio between 0 and 1). - QPCs outperformed structural measures in both GRA and PRGT. - Implementation more even and better in PRGT programs. - SBs implemented better than SPCs historically; discontinuation of SPCs did not worsen structural implementation overall, but implementation deteriorated in 2011. - Flexibility: - About 50 percent of measures reset for a future date were eventually met. - Table 5 status (2002-2011) excerpts: - On initially scheduled test date — Met 74, Not met 15, Pending 11 (percent); - On the rescheduled test date (when rescheduled) — Met 49, Not met 23, Pending 27 (percent). - Determinants: - Larger number of conditions in a review correlated negatively with implementation (panel regression, Table 6). - Implementation appeared unrelated to depth of conditionality, highlighting importance of capacity and ownership. - Combination of low capacity and many conditions linked to worse implementation. - Prior actions more likely when past implementation weak, but more PAs did not necessarily improve overall implementation. ### Institutional classification and sector determinants - Institutional mapping of conditionality categories: - Core Fund expertise: fiscal measures, central bank measures, financial sector measures in advanced economies, exchange rate system measures. - Shared expertise: public expenditure auditing/accounting, civil service/public employment reforms, pensions/social sector reforms, public enterprise reform/pricing, financial sector reforms in LICs/EMs, PRSP development. - Non-core: international trade policy (excluding customs reform), economic statistics, other structural measures (natural resources, agriculture). - Probit findings (summary): - GRA: fiscal conditionality heavier with favorable initial conditions (lower inflation, higher growth); financial sector conditionality heavier with larger vulnerabilities (wider fiscal deficits, larger financial exposure). - PRGT: more public government conditions with higher debt and lower reserves; central bank conditionality more likely with inflation and lower reserves; probability of many conditions responds to macro vulnerabilities and policy space. ### PRGT-specific patterns and surveillance leverage (Appendix summaries) - PRGT conditionality responded to initial macro weaknesses and policy space: more fiscal conditions with larger debt and lower reserves; fiscal focus greater when fiscal deficit policy space existed. - Surveillance-to-program mapping (programs approved since 2008 with recent Article IV): - ~48 percent of conditions at approval foreshadowed in Article IV; share falls slightly to 45 percent up to second review. - SBs and SPCs more closely foreshadowed than PAs (50 percent vs. 37 percent at approval). - Variation by year and program type; policy implication: prioritize surveillance recommendations to improve traction into program design. ### Discontinuation of SPCs and guidance uptake - Discontinuation aimed to reduce stigma and enhance ownership via timeline flexibility. - Evidence: discontinuation did not worsen overall implementation; flexibility in timing supported better implementation (2009–2010). - Survey: <30 percent of mission chiefs/resident representatives (out of 84 respondents) found phasing out SPCs increased ownership or reduced stigma; some felt it reduced Fund leverage. ### Guidance on clarity (Appendix 5) - 2006 Guidance Note: expose criticality, link structural conditions to objectives, convey policy options, detail non-core conditions, separate Fund conditions from authorities’ agenda, discuss implementation concerns, note Bank-Fund collaboration. - 2008 Guidance Note: at approval, staff report should lay out links between goals, strategies, and structural conditionality (separate table); report when expertise not available in other multilateral institutions. - 2009 Guidance Note: staff reports should indicate expectations for subsequent reviews and specify critical future actions as structural benchmarks; reviews should update objectives and strategies for structural reform agenda. *Source: IMF background paper "1. Determinants of the Probability of a High Number of Structural Conditions", covering conditionality in Fund-supported programs (2002–end-September 2011).* ### 1. Determinants of the Probability of a High Number of Structural Conditions .....................19 ### 1. Determinants of the Probability of a High Number of Structural Conditions ### Scope and analytical framework - Review period: 2002 to end-September 2011. - Focus: content and application of program conditionality—especially structural conditionality—in relation to the 2002 Conditionality Guidelines, the Staff Statement on Principles Underlying the Guidelines on Conditionality, and subsequent revisions to operational guidance on conditionality. - Core analytical lens: the five key interrelated principles guiding the design of conditionality: - national ownership of programs; - parsimony in program-related conditions; - tailoring to country circumstances; - effective coordination with other multilateral institutions; - clarity in the specification of conditions. - Parsimony operationalized by the "criticality criterion": program-related conditions should be critical (or the minimum necessary) to achieve program objectives and goals, critical for monitoring program implementation, or necessary for implementing specific provisions under the Articles of Agreement. ### Data sources and methods - Quantitative analysis based on the Monitoring of Fund Arrangements database. - Qualitative assessments drawing on: - 23 recent Ex Post Assessments of Longer-Term Program Engagement (EPAs) and Ex Post Evaluations of Exceptional Access (EPEs); - 18 case studies representative of programs during 2006-10; - structured interviews with Fund Executive Directors; - surveys of country authorities, in-country donors, and Fund mission chiefs and Resident Representatives for countries with programs since 2007; - views from headquarters-based donors, civil society organizations, and other stakeholders. - Program coverage includes SBAs (including HAPAs), EFF in the GRA, PSI, SCF, ECF in the PRGT, and the high access component of the ESF (ESF-HAC), among others, meeting the standard of upper credit tranche (UCT) conditionality. Excludes ENDA, EPCA, ESF-RAC, RCF, and staff-monitored programs. ### Key findings - Conditionality became more focused and generally aligned with program goals. - Conditionality was tailored to country and program characteristics, and to initial macroeconomic conditions. - Conditionality became more parsimonious. - The number of conditions per review and per year of ongoing program declined in recent years, supported by the discontinuation of SPCs. - Structural conditionality, however, increased somewhat in GRA programs in 2011. - Conditionality focused increasingly on areas of Fund core competencies, with fiscal measures taking center stage. - Ownership was critical to program success. - Frontloaded conditionality can support implementation but cannot substitute for ownership. - Ownership can be supported by outreach and communication, program flexibility, focused conditionality, and increased clarity. - Some evidence suggests ownership improved on balance, with some possible new issues in 2011. - Balance between tailoring and evenhandedness was mostly appropriate. - The design of conditionality tended to match country capacity and access granted under the program. - Coordination with other institutions was important and generally supportive. - Coordination helped parsimony and avoided duplicate measures. - Coordination challenges arose with European institutions on some European programs during crisis conditions, reflecting the need for institutional preparedness ahead of crises. - Progress in implementation stalled in 2011. - The parsimony of conditionality and flexible timelines supported execution through 2010. - Difficulties in 2011 likely reflect more challenging economic conditions and adjustment needs. - Clarity of conditionality was uneven. - Best practices on clarity need broader application in program documents. ### Evolution of operational guidance and instruments - Revisions to operational guidance followed prior reviews: - 2003 operational guidance note; subsequent revisions in 2006, 2008, and January 2010 (Operational Guidance Note on Conditionality—December 2009 Revisions). - 2008 revision emphasized criticality and required rigorous justification of conditionality. - March 2009 reforms and early 2010 adjustments led to discontinuation of Structural Performance Criteria (SPCs) in all Fund-supported programs and a shift to a review-based approach to monitor structural reforms. The aim was to: - enhance flexibility of the conditionality framework; - tailor conditionality to countries' varying characteristics; - reduce stigma associated with SPCs and associated waivers; - preserve safeguards for the use of Fund resources. - The review expanded the scope of analysis relative to prior reviews by systematically studying Prior Actions (PAs) and Indicative Targets (ITs), and incorporating analysis of the depth of structural conditionality. ### Policy implications and forward-looking considerations - Focus and parsimony of conditionality may come under pressure in a more challenging global environment. - As structural reform and adjustment needs deepen—notably in the second wave of programs started in 2010—more wide-ranging conditionality may be needed. - It will be important to monitor conditionality closely to ensure measures remain the minimum necessary to address country-specific issues and ensure program success. - Coordination with institutional partners will become more difficult but more essential. - With more institutions and donors involved in program design, increased financing needs, and constrained financing capacities, coordinated policies and conditionality will be critical, especially in co-financed programs. - Coordination is more effective if engagement among potential institutional stakeholders is set ahead of crisis times. - Communication and transparency should be improved. - Enhanced communication on program goals, strategies, and linkages to conditionality could support program ownership and improve donor coordination. - Mandating improved clarity in program documentation and communication could help ensure conditionality remains focused on critical measures. - Surveillance could be leveraged more effectively. - Case-study analysis shows about half of program structural conditionality was identified as needed reforms during pre-program surveillance. *Source: IMF background paper "1. Determinants of the Probability of a High Number of Structural Conditions", covering conditionality in Fund-supported programs (2002–end-September 2011).* ### 8. The remainder of this paper is structured as follows. Section II examines ### _061812 - 8. The remainder of this paper is structured as follows. Section II examines ### II. STREAMLINING AND THE PRINCIPLE OF PARSIMONY - Conditionality in Fund-supported programs aims to help countries solve their balance of payments problems and to safeguard Fund resources. According to the Conditionality Guidelines, program-related conditions should be (i) of critical importance to achieve program goals or to monitor program implementation, or (ii) necessary to implement specific provisions under the Articles of Agreement or policies under them. - Initiatives to streamline program conditionality have been central to Fund efforts since at least 2000, following the issuance of the interim guidance note on streamlining structural conditionality (Streamlining structural conditionality in Fund-supported programs–Interim Guidance Note, 2000). That note preceded the revised conditionality guidelines in 2002. - Parsimony is a key principle of streamlined conditionality: conditions should focus on achievable goals and be directed primarily toward the macroeconomic goals in the Guidelines, ¶6: - (a) solving the member’s balance of payments problem without recourse to measures destructive of national or international prosperity; and - (b) achieving medium-term external viability while fostering sustainable economic growth. - Streamlining does not preclude substantial conditionality where needed. Under streamlining, structural conditionality should be both parsimonious and macro-critical (i.e., conditions should be of critical importance for achieving the goals of the member’s program or for monitoring program implementation, and should be limited to the minimum necessary). - Non-core critical measures require strong justification and a detailed explanation of their criticality; the Fund is expected to draw on other institutions’ expertise, including the World Bank and donors, to the extent possible. ### A. Program Goals, Instruments, and Conditionality: Focus and Alignment - Data and sample coverage: - MONA database entry for program objectives and reform strategies introduced in August 2008. - Sample includes 59 LICs and 31 AMs and EMs, covering the period 2005-2011. - Only programs approved after 2005 are included. - Program goals in MONA are classified into four categories: macroeconomic stability, external stability, economic growth, and poverty reduction. - Program goals and instruments: - All programs identified external and macro-economic stability as overarching goals. - GRA-supported programs centered on external and macro-economic stability. - Share of programs including growth objectives rose from 40 to 60 percent between 2008 and 2011. - Almost all PRGT-supported programs aimed also to achieve poverty reduction and economic growth. - Fiscal and monetary policies were the most frequently used instruments; GRA programs relied more on financial sector policies, PRGT programs used a wider set of pro-growth and social sector policies. - Alignment and tailoring: - Structural conditionality was broadly aligned with program goals and tailored to country circumstances. - PRGT programs included more wide-ranging structural reform agendas; GRA programs reflected capital account challenges and concerns about spillovers/systemic risks in some cases with higher access levels. - Program design accounted for political economy considerations and spillover effects of the global financial crisis. - EPAs and EPEs generally confirmed alignment but noted exceptions where: - Reform agendas could be better aligned with program objectives and sequenced to overcome capacity constraints (example: Sierra Leone). - Macro-criticality of structural measures could have been demonstrated more clearly to ensure focus on core Fund expertise (examples: Benin, Burundi). - Better diagnostics of institutional and political constraints would have enhanced conditionality effectiveness (examples: Belarus, Ukraine, Kyrgyz Republic). - Some programs lacked appropriate follow-up conditionality once initial reforms were completed. - Survey evidence on alignment (Figure 2): - Fund staff and country authorities generally agreed that conditionality had been geared toward critical program objectives; donor responses were more cautious. - Statement assessed: "The structural conditions in the program only covered reforms that were critical for program objectives, design and implementation" — outcome categories: Agree / Disagree / Don't know (percent of respondents). - Box 1 findings (Program Instruments and Structural Conditionality): - Countries classified as having weaker initial conditions if any of six macroeconomic variables were below/above thresholds derived from the sample: real GDP growth, inflation, current account balance, fiscal balance, net international reserves, and medium- and long-term debt. - Programs for countries with weaker initial conditions were more focused; probabilities of finding certain program goals or reform strategies were lower for that subsample. - Choice of reform strategies affected the choice and focus of structural conditionality: presence of a reform strategy for a sector increased probability of structural conditions on that sector (e.g., monetary strategy increased probability of conditions on the central bank; pro-growth and social sector strategies in PRGT programs increased probability of conditions on public enterprise). - Box 2 findings (Structural Conditionality Alignment and Tailoring): - Main conclusions: - (1) Program conditionality was generally well aligned with program goals and reform strategies, although there was room for improvement. - (2) Programs were generally tailored to country-specific circumstances. - Around 80 percent of conditions on average could be linked to specific objectives (both GRA and PRGT). - In PRGT countries, about 60 percent of objectives were addressed by structural reforms; in GRA countries, about 70 percent of objectives were addressed by structural reforms. - Program duration and timing of conditionality were generally in line with country characteristics: PRGT programs tended to have more structural measures but fewer conditions in the first three months (2 compared to 3) and longer time spans with typically semi-annual reviews. - Past performance and political cycles influenced program conditionality design (examples provided: Burkina Faso 2007; Uganda 2006; Gambia 2007). - Program conditionality usually covered multiple policy areas but could be strongly focused when warranted (examples: Armenia 2008 tax reforms; Iceland 2008 banking reforms; Uganda January 2006 domestic arrears). ### B. Focus of Structural Conditionality - Trend toward focus on core Fund areas: - Structural conditionality increasingly focused on core areas of Fund expertise, particularly in recent years and notably for GRA programs. - Guidelines state that conditionality critical to program success "will normally consist of macroeconomic variables and structural measures that are within the Fund’s core area of responsibility" (Guidelines, ¶7 (b)). - The trend reflects an increasing focus on fiscal-related conditionality. - Sectoral composition and shifts: - Fiscal conditionality dominated the structural reform agenda. - In PRGT programs, government-related structural measures accounted for about 58 percent of total structural conditionality during 2008-2011, an increase of more than 7 percentage points compared to 2005-2007. - Financial sector reforms were the second most important area of structural reforms, notably in GRA programs. - As the crisis evolved from a financial crisis to a deeper real-sector crisis, financial sector focus peaked at the onset of the global financial crisis, after which fiscal issues increased in prominence, including in EU and Iceland programs. - Figures and distributions referenced (MONA and staff calculations): - Figure 1: Program Goals and Instruments, 2006-2011 — program goals shares and program instruments shares by GRA and PRGT. - Figure 3: Focus of structural conditionality moving toward core areas of Fund expertise (appendix defines core, non-core, shared). - Figure 4: Distribution of conditions by sector, in percent of total conditions per review (GRA and PRGT, periods 2002-2004, 2005-2007, 2008-2011). - Figure 5: Fiscal and Financial Sector Conditionality in GRA-Supported Programs — share of fiscal and financial sector conditionality (In percent of total structural conditions per review) for 2002–2011 and fiscal/financial sector conditionality in the EU and Iceland and other GRA programs for 2009–2011. - Cross-cutting observation: - Structural conditionality tended to be aligned with program strategies and country circumstances, with fiscal measures increasing in prominence after the peak of financial-sector-focused conditionality at the crisis onset. *Source: Excerpt from IMF staff paper (MONA database analysis, 2005–2011 sample; figures and boxes as cited in the source content).* ### 20. The breakdown of structural conditionality by sector was determined by the ### _061812 - 20. The breakdown of structural conditionality by sector was determined by the ### Determinants of structural conditionality by sector - The breakdown was determined by the extent of vulnerabilities and the availability of fiscal space (Table 1 and Appendix 2). - GRA programs: - General-government or fiscal related conditionality more likely to be higher with better economic conditions (lower inflation, higher growth). - Conditionality in the financial sector more likely to be important when there are larger exposures (total assets and liabilities) and smaller fiscal deficits. - PRGT programs: - More general government conditions were more likely with higher debt levels, lower reserves, and smaller fiscal deficits. - Probit models related probability of setting a large number of conditions (defined as a number of conditions belonging to the highest quartile of the distribution) to macroeconomic conditions at program approval; probit models run separately on main categories of structural conditionality. Only statistically significant variables summarized; detailed results in Appendix 2. ### Public Financial Management (Box 3) - Fund conditionality in PFM grouped into five categories: (i) stocktaking and control of arrears; (ii) financial accounting, reporting, and auditing; (iii) expenditure controls; (iv) legal and institutional frameworks; and (v) more advanced PFM reforms. - Findings and sequencing: - Arrears: conditionality in countries with weak PFM helped build capacity to identify, verify, report, and clear public sector arrears and create frameworks to prevent arrears and strengthen expenditure controls, usually associated with fiscal consolidation needs that triggered programs. - Financial accounting and reporting: focus on improving coverage and quality of fiscal data by strengthening accounting classification, timeliness of fiscal reporting, and auditing procedures. - Expenditure controls: after progress on arrears and fiscal information, emphasis shifted to budget execution systems (commitment controls, active cash management), procurement and payroll systems, tracking poverty-related expenditure, and monitoring fiscal indicators (budget deficit and debt). - Legal/institutional frameworks: conditionality aimed at revisions to organic budget, procurement, and debt laws; PFM-related regulations; revision of roles and responsibilities within ministries of finance; creation of units to control and monitor capital investments, public-private partnerships, sub-national debt, and macroeconomic policy developments; strengthening internal and external audit functions. - Advanced reforms: where basic PFM systems existed, conditionality focused on establishing treasury single accounts, implementing or revamping integrated financial management information systems, and improving integration between cash and debt management functions. ### Macro-social structural conditionality (Box 4) - The focus on macro-social issues in Fund-supported programs increased recently (analysis covers 2002-2011). - Structural conditionality directly related to macro-social issues represented a small share of all structural conditions; macro-social conditionality had been declining since the early 2000s but started to increase again at the end of the review period. - Program-type differences: - GRA programs typically included more macro-social conditions, with greater relative focus on reforms to pensions, social sectors, and labor markets. - PRGT programs more commonly included civil service reforms; conditionality related to pensions or labor markets outside public employment is very limited in PRGTs. - Short- and long-term outcomes: measures aimed at lowering fiscal expenditure in the short term and/or improving productivity through structural reforms can resolve balance-of-payments difficulties and yield longer term gains, but may have short-term costs in employment or income. - Examples: deep cuts in public sector employment, public sector wages (Latvia and Greece), and pensions (Greece); Latvia introduced safety measures such as extension of employment benefits, public works programs, and higher housing benefits. - Sector-specific observations: - Labor markets: many measures involved social partners in decision-making, helping build ownership and supporting program implementation; short-term costs possible, but likely longer-term improvements in labor market functioning and inclusiveness. - Civil service and public employment: some conditions could reduce public employment or wages short-term; productivity-enhancing measures need not have short-term negative effects and have positive longer-term effects; these measures are more common in PRGT programs. - Pensions and social sector reforms: almost half of conditions aimed at improving the social safety net either through higher social spending or better targeting; conditionality in energy-subsidy reforms sometimes helped mitigate impacts on the poor. ### Wage bill ceilings in PRGT programs (Box 5) - In 2007, the Executive Board requested that wage bill ceilings in LICs be used only in exceptional cases and with sufficient flexibility in programs. - Criticisms: wage ceilings could prevent LICs from using donor resources to expand employment in poverty-reducing sectors (health, education); wage ceilings tended to persist, were difficult to monitor, and distracted governments from strengthening institutions to control wage spending. - Observed practice: - An analysis showed the 2007 policy was well followed: there were no QPCs set on wage bills starting in 2007, although a few instances of monitoring through ITs occurred. - Examples of IT monitoring reflected high wage bills in Djibouti and Moldova compared to similar countries and the need to reorient spending towards pro-poor spending in Cote d’Ivoire. - Structural conditionality related to wage bill ceilings dropped significantly—the last PRGT program with structural conditionality on wage bills was Moldova with an SPC in 2006 and a PA in 2009. - Historical listing of programs and types (2002–2011) noted in the source (table of Quantitative PCs or ITs, PAs, SPCs, SBs), and memorandum item showing number of LIC programs approved by year. ### Quantitative conditionality: focus and evolution (Sections C and Figure references) - Quantitative conditionality focused on external and fiscal stability, reflecting key objectives of Fund-supported programs. - Time patterns: - Earlier part of review period: focus on debt. - During the global financial crisis: monitoring in GRA programs shifted toward fiscal consolidation, reflecting growing fiscal consolidation needs. - PRGT programs: mix of QPCs remained broadly unchanged with emphasis on external debt, reflecting Fund involvement in debt-reduction exercises for LICs. - Monitoring coverage: - Fiscal and monetary stability were monitored in almost all programs. - All PRGT programs included targets on external debt. - Use of reserves targets limited in countries with no autonomous exchange rate policy (currency unions and currency boards constitute the totality of programs with no reserves target). - Indicative targets (ITs): - Mostly used to monitor fiscal performance and “social sector” indicators. - ITs complement QPCs (or QACs), often providing guidance on expected values of QPCs later in the program; ITs are normally turned into QPCs with modifications. - ITs can be set on variables that provide additional program-performance information (primary fiscal deficit, base money, change in net credit of banking system to public sector) or on variables critical but difficult to assess precisely as QPCs. - Analysis covers ITs set during 2006-11. - Over the period, the largest share of ITs was related to the fiscal deficit in PRGT programs. - In GRA programs, there was a shift toward fiscal ITs in the latter part of the period, reflecting concerns about fiscal sustainability. - The use of wage ceilings declined in both GRA and PRGT programs. - “Social” ITs—typically floors for priority spending—were used to monitor performance in fiscal expenditure in these areas. - After the 2009 revamping of facilities for LICs which mandated monitoring of poverty-related spending, the use of poverty-related ITs increased in PRGT programs. ### Volume of conditionality and counting methodology (Section D and Box 6) - Parsimony measured by number of conditions set over a program; the paper counts conditionality by the year when conditions were reviewed (review-year approach). - Counting approaches: - Approval-year approach: counts total conditions per arrangement identified by year of program approval; normalized by year or review; comprehensive per program but less suited for recent/ongoing programs and for tracking policy changes during program life. - Review-year approach: conditions accounted for based on year when reviewed (or expected review year for outstanding conditions); normalized by number of reviews per year or by number of current programs per year; allows tracking program developments and policy changes more effectively. This approach underpins the 2011 RoC. - Scope of analysis: - Examined all conditions set by the Executive Board or by Management up to September 30, 2011—whether reviewed or not—and include PAs (set by Management), QPCs, SBs, ITs, and the now discontinued SPCs. - Excluded cancelled conditions and modified conditions to avoid double-counting; continuous SBs and SPCs only counted when implementation status recorded in MONA. ### Key statistics and trends (explicit numeric statements preserved) - The number of QPCs has been stable, at around five to six per review (Figure 8). - Inter-quartile range: between four and six QPCs in GRA programs and between five and seven in PRGT programs. - The number of ITs was broadly steady overall during 2006-11 (Figure 9). - The discontinuation of SPCs in March 2009 had a sizeable impact on the number of structural conditions per review, notably in GRA programs. - Declines in SBs and PAs (both at program approval and during reviews) occurred over most of the period. - The distribution across programs of the number of structural conditions per review narrowed during the latter part of the period, pointing to an effort across the board to further parsimony; the number of structural conditions stabilized after 2009. - Conditions per review would be even lower if arrangements under the FCL and PCL were included, given there is no ex-post conditionality under the FCL and limited ex-post conditionality under the PCL. *Source: Extracted from the IMF background paper covering structural conditionality, including Boxes 3–6 and Sections C–D (2002–2011 analysis).* ### 27. The drive towards parsimony may have reached a plateau at the very end of the ### 27. The drive towards parsimony may have reached a plateau at the very end of the ### Trends in conditionality quantity and frequency - The number of conditions per review remained comparatively low in both GRA and PRGT programs, but in GRA programs a steady increase in the number of SBs per review was noticeable. - The number of reviews with a relatively larger number of conditions tended to increase between 2009 and 2011, likely reflecting the need to tackle deeper-rooted issues, such as competitiveness or large debt burdens. - The shift toward reviews with more conditions was widespread across all types of GRA programs, not limited to a small number of countries. - The total number of conditions per year picked up modestly starting in 2010, reflecting: - an increase in the number of conditions per review (GRA programs), and - a greater frequency of reviews per year (GRA and PRGT). - Conditions per year are defined as the total number of conditions reviewed in a given year, normalized by the number of programs that are ongoing during the same year (on a quarterly basis, but annualized); frequency depended on the initial schedule of reviews (quarterly or semi-annual) and on the implementation record of programs. ### Adjustments and fine-tuning during program implementation - More frequent adjustments in structural conditionality indicate ongoing fine-tuning of programs and a potential need to better leverage surveillance. - Reviews proceed with expected adjustments as new unforeseen circumstances require policy responses; crisis imperatives can crowd out important structural measures in initial program design. - In recent programs with advanced countries (Iceland, Greece and Ireland), adjustments and additions to conditionality were relatively higher during the third and subsequent reviews compared with other cases. - In some cases, lags in fully developing the reform agenda could point to a disconnect between Fund surveillance and program design, suggesting surveillance could be better leveraged to relate program conditionality systematically to vulnerabilities identified during surveillance exercises, including through enhanced prioritization of recommendations during surveillance. ### Links between surveillance and conditionality (Box 7) - Analysis based on programs approved since 2008 with a recent Article IV consultation (less than two years old) included 34 countries: - 23 (or 68 percent) had GRA programs. - 11 (or 32 percent) had PRGT programs. - About 48 percent of program conditions were foreshadowed in previous Article IV consultations, with variation across countries and time. - SBs and SPCs were mapped more closely to Article IV recommendations than PAs. - Conditionality in PRGT programs tended to be generally well related to previous surveillance recommendations; conditionality in GRA programs exhibited much larger variation, reflecting emergency and unforeseen crisis needs. - By the second program review, dispersion across program types becomes more similar. - The share of conditionality foreshadowed in Article IV reports was lowest in 2008 for GRA programs (reflecting unexpected crisis impacts) and lowest in 2009 for PRGT programs; by the end of the period the share tended to increase. ### Depth of structural conditionality (Section E) - Depth is defined as the degree and durability of structural change likely to occur if a condition were implemented; it refers to durability of structural change and does not assess difficulty or criticality of measures. - Analysis of 2006-11 programs pointed to an increase in depth of structural conditionality. - Significant variance existed among countries and regions: - Programs with advanced market economies were, on average, structurally deeper than programs with emerging markets. - GRA programs with European countries demonstrated the highest degree of structural depth. - Structural depth of recent programs with euro area countries (especially Greece) surpassed the average structural depth for other programs. - Programs with LICs accounted for the largest number of conditions, but structural depth in these programs was lower than for the other two country groups. ### Trade-off between number and depth of conditions (Box 8) - Regression analysis indicates a trade-off between the number and the depth of conditions: the number of conditions per review correlated negatively with the average depth of conditionality per review in both GRA and PRGT programs during 2006-11. - Results from a fixed-effects panel regression (period covered 2006-2011; conditions with test dates up to September 30, 2011) indicate: - A one standard deviation increase in the depth of conditionality correlated with a reduction in the number of conditions by 1.2 and .7 in GRA and PRGT programs respectively. - Reported regression outputs: - GRAPRGT - Depth -1 -2.23*** -1.43** - Adjusted R-squared0.27 0.21 - Number of obs.111 187 - Note: *** and ** refer to a 1 percent and 5 percent significance level respectively. ### Ownership and conditionality (Sections III.A and Box 9) - Ownership is crucial; primary responsibility for selection, design, and implementation of economic and financial policies rests with member countries. - Streamlined and tailored programs with parsimonious conditionality support ownership; lack of ownership was sometimes exacerbated by complexity of measures and capacity constraints in public administration. - When slower-than-expected implementation reflects capacity constraints and/or limited ownership, a more prioritized approach in defining reforms may be required. - Fund outreach to authorities and broader stakeholders can help build ownership; extensive outreach combined with focused conditionality and flexible implementation helped achieve higher ownership in some cases. - Recording of outreach activities in program documents was uneven; three out of nine GRA case studies did not report outreach activities though evidence indicated outreach occurred. - Indicators suggest ownership may have improved over 2002-2010 for GRA programs, evidenced by less use of a large number of PAs at program approval. - Simple regression analysis suggests use of PAs at program approval was related to future implementation issues in GRA programs: - A higher number of PAs at program approval was associated with a lower implementation ratio of structural conditionality over the life of the program; this was statistically significant for GRA programs with more than 2 PAs. - In PRGT programs, the relation failed to be significant, suggesting PAs were used regardless of potential implementation issues. - Use of PAs during program implementation underlined their role as implementation tools: - Probit analysis indicated the probability for a review to include a PA was related to implementation record of previous reviews. - In GRA programs, an improvement in program implementation by one percent in the current review reduced the likelihood of having a PA at that review by 0.6 percent; an improvement in the previous review by one percent reduced the likelihood by 0.5 percent. - In PRGT programs, an improvement in the implementation ratio by 1 percent during the current review reduced the likelihood of having a PA at that review by 0.89 percent. - Reported regression/marginal effects excerpts: - Implementation of structural conditionality over the life of the program - All programs GRA PRGT - -0.298 -0.988 0.172 -5.481*** -7.928** -3.156 - Constant - 0.933** 1.441** 0.602 9.183*** 11.05*** 7.353*** - Adjusted R-squared -0.008 0.017 -0.018 0.513 0.624 0.021 - N 873 255 572 829 - Probit marginal effects for prior actions and implementation: - Implementation ratio (t) -0.0062** -0.0089** - Implementation ratio (t-1) -0.0077** 0.0011 - Actual P 35.55 32.75 - Predicted P 32.91 29.39 - P error -2.64 -3.36 - Note: The table reports marginal effects. P(y=1) denotes a review with at least one prior action. P-values in brackets. * p<0.1, ** p<0.05, *** p<0.01. ### Ownership case lessons (Box 10 summary) - Iceland and Latvia: strong local support for program goals helped overcome crisis; conditionality aligned with authorities’ priorities (financial sector restructuring for Iceland; fiscal and financial stabilization while maintaining exchange rate peg for Latvia). - Belarus: a program with weak program ownership at the highest levels left serious residual problems. - Examples highlight the role of ownership in program success and the alignment of conditionality with authorities’ reform plans. *Source: MONA and staff calculations, and IMF staff analysis (programs and reviews during 2002-2011; depth analysis for 2006-2011).* ### 36. In cases of limited ownership, specific instruments can enhance program ### 36. In cases of limited ownership, specific instruments can enhance program implementation ### Ownership and conditionality - Staff-monitored programs can be used to assess the potential for a future Fund-supported program with UCT-quality conditionality, thereby gauging both implementation capacity and the degree of ownership by the authorities. - Prior actions (PAs) at program approval could be used as screening devices for the degree of ownership. - EPAs/EPEs suggested that a flexible approach to program conditionality, with regard to implementation timelines, was found to enhance ownership, while the discontinuation of SPC did not have a noticeable impact (Appendix 4). - Conditionality cannot substitute for ownership: a higher number of PAs did not systematically result in more successful program implementation (Box 9). - Example: In some programs (for example, with Ukraine), program success was reduced by weak ownership despite many PAs (BP4). - Lessons from recent GRA programs showcase the importance of ownership and the difficulty of gauging ownership through simple measures of implementation (Box 10). - Belarus case (program performance and ownership): - Fiscal adjustment was stymied by quasi-fiscal activity (LGP). - Authorities opted to increase LGP by almost 50 percent, compounding demand pressures and exacerbating external imbalances. - Authorities adhered to the letter of the program by circumventing the original conditionality on LGP; conditionality was adapted as LGP growth became apparent. - Lack of comprehensive data on LGP hampered program effectiveness. - The current account deficit widened in 2009 and 2010; overvaluation of the real exchange rate persisted. - A lack of ownership contributed to shortcomings of the program and limited progress at addressing vulnerabilities (see also BP4). - Iceland and Latvia outcomes: - By 2011 Iceland and Latvia were growing again and on their way to overcoming the challenges that necessitated the original programs. - Iceland: value of the krona stabilized; authorities began capital account liberalization under their strategy; significant progress to place public finances on a sustainable path; unemployment remains high; a new, significantly smaller, banking system emerged with work addressing legacy vulnerabilities progressing. - Latvia: largely stabilized banking sector; on course to bring budget deficit down to 4.5 percent of GDP, putting the country within sight of the 3 percent of GDP Maastricht criterion and the program exit strategy of euro adoption; strong program ownership was key; reductions in output and employment accompanied stabilization, with unemployment still exceeding 14 percent. ### B. Tailoring: balancing country circumstances and evenhandedness - Programs need to accommodate a broad spectrum of initial conditions; design differences include: - Countries with wide-ranging structural reform agendas may require deeper structural changes than countries suffering from short-term liquidity shocks. - Large-scale disbursements may justify a different type of conditionality than more limited disbursements. - Fund adopts a liberal view to requests for transactions within the first credit tranche, provided the member itself is making reasonable efforts to solve its problem. - Conditionality should be adapted to countries’ capacity and track record. ### Capacity (designing conditionality to reflect implementation capacity) - In countries with low implementation capacity, program design entails tradeoffs between: - An ambitious structural reform agenda that would help develop capacity, and - A realistic set of conditions, possibly divided into small, achievable modules. - Conditionality could be less complex for low-capacity countries, and more ambitious where stronger institutions allow higher implementation. - Case studies: - 2006 Sierra Leone program: carefully paced structural agenda reflecting post-conflict nature. - 2008 Togo program: took into account capacity constraints after long political crisis and donor disengagement; no measures for the first six months other than two PAs. - Empirical findings: - The correlation between country capacity at program approval and the total number of conditions per program was positive and statistically significant (Figure 15). - Conditions of high structural depth tended to be set more frequently in countries with a higher Country Policy and Institutional Assessment (CPIA) rating. - A probit regression showed that the probability of setting at least one structural condition of high depth in a given review was positively and significantly related to the CPIA rating: an increase by one in the CPIA rating increased by 20 percent the likelihood of a deep condition being set. - Capacity was proxied by the World Bank CPIA rating for macro-economic management; analysis covered 2005-2009 for a subset of 54 countries. ### Access (tailoring conditionality to safeguard Fund resources) - Conditionality was tailored to the access level under an arrangement: - In PRGT programs, access was related to the number of PAs for Board approval. - In GRA programs, access was related to the number of deep conditions. - Interpretation: - In PRGT countries (often lower capacity), focus on measures achievable prior to program approval. - In GRA programs (usually stronger capacity), higher access correlated with deeper structural change as larger BOP needs translated into higher access and greater need for structural reforms. - Focus shifts by access: - GRA and PRGT programs in the upper quartile of the access distribution had structural conditionality more focused than average on financial sector reforms, reflecting more mature economic systems with greater financial difficulties. - Monitoring of arrangements involving higher access focused on fiscal issues. - This applied to high exceptional access GRA-supported arrangements (access over 1000 percent of quota) and PRGT-supported arrangements in the upper half of the access distribution, reflecting the need to ensure fiscal sustainability amid rising external debt. - Debt limits in LIC programs: - LIC programs typically included external debt limits to prevent build-up of unsustainable debt while allowing adequate external financing. - Since December 2009, new guidelines moved away from a single design for concessionality requirements toward a menu of options, better accounting for differences among LICs in debt vulnerability and macroeconomic and public financial management capacity (BP4). ### Survey assessments of tailoring and flexibility - According to survey results, conditionality was generally viewed as both tailored and flexible. - Fund staff were more convinced of the appropriateness of program design and flexibility; country authorities and donors broadly shared the positive assessment (Figure 18). - However, weaknesses in design affected some programs: - Inappropriate diagnostics of underlying problems reduced program effectiveness in some cases (noted in EPAs and EPEs). - In a few cases, critical conditionality or conditionality following-up on initial reforms did not form part of the program. ### C. Coordination (with partner institutions) - Coordination with partner institutions, especially the World Bank, is essential for program success; coordination with regional institutions has grown in importance. - EPAs and EPEs highlighted coordination with the World Bank and other institutions in macro-critical areas of structural reform beyond Fund core areas (e.g., decentralization, energy sector reform). - Examples and findings (BP4): - Sub-Saharan Africa: - Coordination process with multilateral and regional institutions—notably the World Bank—was effective. - Coordination focused on joint or overlapping responsibilities, streamlining Fund conditionality and avoiding duplication. - Responsibilities shared according to the “lead agency” principle, with the Fund focusing on macro-critical conditionality. - Coordination supported by the Joint-Management Action Plan and JMAP; informal and formal coordination were both important. - Some EPAs/EPEs noted occasional challenges in collaboration with the World Bank. - Recent programs with European member countries: - Coordination with the EU supported design and monitoring of conditionality; complementarities of Fund and EU expertise proved useful. - Fund’s expertise operationalized fiscal adjustment path and tackled fiscal risks; EC expertise focused on specific fiscal areas (health care, pension reforms). - In euro area countries, additional stakeholders added complexity to conditionality design, often resulting in extended discussions/decision-making; discussions about private sector involvement affected coordination and program design. - Large and growing number of structural reforms identified by the EU increasingly at odds with the principle of parsimony (Box 11). - Survey results: - Country authorities, Fund staff, and donor agencies viewed coordination as critical to program implementation. - More than 80 percent of respondents noted intensive coordination with donors and effective coordination with the World Bank (Figure 19). - Box 11 summary (Design of Conditionality in Co-financed Euro Area Programs): - EA programs for Greece, Ireland, Portugal supported by Fund, EC, ECB (“Troika”); monitoring via unified macro parameters and structural measures. - Fund conditions for Board reviews based on a standard quarterly framework of QPCs and SBs; EC conditionality based on overall assessment of progress against EC structural agenda and macro targets. - Joint Troika review missions support coincidental disbursements from Fund and EU. - Division of labor generally guided by core expertise; Fund focuses on short-term macro-critical policies, EC on medium-term structural reforms and compliance with EU directives. - Some overlaps exist (e.g., EC/ECB memoranda include fiscal and financial chapters). - Quantitative targets: EC targets designed to be consistent with Fund’s though definitions may differ (e.g., fiscal targets cash basis for Fund, percent of GDP for EC). - EC measures tend to be more precise and detailed than Fund’s; EC structural agenda less parsimonious and expanded over time. - Example: Fund condition for Greece: “a comprehensive pension reform that reduces the projected increase in public spending on pensions over the period 2010-60 to 2½ percent of GDP,” while the EC has a set of precise measures defining the new pension system. - Over time both Fund and EC have ventured into areas initially assigned to the other; EC increasingly focused on fiscal issues while Fund introduced “competitiveness” conditionality in Portugal and Greece. - EC structural conditionality counts: varied across Greece, Ireland, Portugal (Box Figure shows number of measures at Board approval for Greece (May 2010), Ireland (Dec. 2010), Greece (Feb. 2011), Portugal (June 2011)). ### D. Clarity - Clarity in Fund conditionality enables stakeholders to better comprehend key elements of programs. - Operational guidance notes on conditionality call for: - Clear justification of program conditionality. - Clear articulation of the link between program goals, program instruments, and corresponding structural conditionality. - Program documents to discuss possible implementation constraints and provide strong justification and detailed explanation of critical importance of conditions outside Fund core areas of expertise. - Discussions with country authorities on policy options should be conveyed clearly where possible. - Program conditions should be transparently distinguished from other elements of the authorities’ program in program documents and staff reports. - The following assessment of clarity is based on review of country case studies (see BP4 for details), except where otherwise indicated. *Italic: Source: _061812 - 36. In cases of limited ownership, specific instruments can enhance program implementation* ### 51. Country case studies indicated that the implementation of the Guidelines with ### _061812 - 51. Country case studies indicated that the implementation of the Guidelines with ### Clarity in conditionality: presentation, justification, and linkage to objectives - Implementation of the Guidelines on clarity in conditionality has been uneven across case studies. - Most staff reports reviewed: - Provided a clear justification of the criticality of conditionality and presented implementation risks. - Largely ignored specific guidance on the presentation of critical structural measures and on providing a discussion of policy options. - Examples of relatively comprehensive presentation: - 2009 Ghana program contained a comprehensive table with measures and their macroeconomic rationale. - 2008 Togo presented a complete set of past, immediate, and envisaged structural reforms. - 2008 Pakistan program contained a box explaining how structural measures link to the achievement of program objectives. - Guidance uptake after July 2008 revision: - The revision called for clearer explanation of choice of conditionality, in particular its macro-criticality. - Of the 19 case study programs approved in or after 2008, almost all explained macro criticality, usually referring to “macro rationale” in a structural conditionality table. - No staff report followed the guidance note’s full approach of introducing structural conditionality with an explanation of program goals, strategies/instruments, and structural conditionality; only Armenia (2008 and 2009) grouped structural conditionality by strategy. - A few programs succinctly articulated objectives and strategies in executive summaries (2008 Hungary, 2008 Seychelles, 2008 Togo), which, combined with tables linking these to structural conditionality, could qualify as best practice. ### Non-core structural conditionality and justification - Overall finding: - A number of structural measures fell outside the Fund’s core areas of expertise. - Roughly 40 out of 200 measures in 26 programs were non-core. - Distribution: - One third of GRA programs had non-core measures. - About two thirds of PRGT programs had non-core conditionality. - Quality of justification and support: - In one third of cases the justification for the non-core measure was quite detailed; in the rest it was succinct and not different from other measures. - In only three cases was a partner institution mentioned that would provide technical assistance with implementation. - Conclusion: programs could do better in identifying non-core but macro-critical measures and in documenting partner support. ### Presentation and recording of policy options and implementation risks - Discussion of implementation risks: - Nearly 90 percent of staff reports provided a discussion of implementation risks in a broad sense; many provided details about risks for individual measures (e.g., Hungary’s SBA discussed domestic political and external environment risks). - Capacity constraints noted as implementation risks; mitigation examples include slow start up and careful pacing (2008 Togo and 2006 Sierra Leone). - Discussion of policy options: - Reviewed staff reports contained little information on recorded discussions of policy options. - Only a quarter of the case studies contained any record of a discussion between staff and authorities on policy options considered. - Survey results suggested broad satisfaction from country authorities and staff that policy options and implementation plans were being discussed, implying under-recording in program documents rather than absence of discussion. ### Program interruptions and off-track analysis - Definition and classification: - Programs considered off-track when time elapsed between reviews exceeded twice the planned time between reviews (more than six months for quarterly programs; more than one year for semi-annual programs). - An interruption is temporary when an off-track period is followed by a review that brings program back on track. - An interruption is long lasting when a program went off-track for more than 12 months and no review subsequently took place before lapse. - Frequency and duration: - During 2002-end-September 2011, average share of off-track programs hovered around 15 percent. - Long-lasting interruptions accounted for about 6 percent of the sample. - About 60 percent of programs in the sample experienced no off-track period. - After declining earlier in the 2000s, program interruptions became more frequent since the start of the global financial crisis; implementation weakened progressively between 2009 and 2011 in PRGT programs, with deterioration notable for GRA programs in 2011. - Cases of off-track PSIs have been exceptional, with only one case recorded since instrument creation (Uganda). - While interruptions became more frequent, average length of interruption declined since 2007; in PRGT programs off-track periods amounted to about 10 percent of total program periods over 2002-2007, a proportion that fell significantly after 2007. ### Implementation of conditionality: patterns and determinants - Overall trends: - Implementation of conditionality improved over most of the review period. - QPCs outperformed structural measures in both GRA and PRGT programs. - Implementation generally more even and better in PRGT programs. - Structural conditionality implementation was consistently better for SBs than for SPCs. - Discontinuation of SPCs did not seem to affect the implementation record of structural conditionality, which remained broadly unchanged in 2010; however, implementation deteriorated in 2011. - In GRA programs, 2011 worsening reflects increase in programs with at least one non-implemented SB; in PRGT programs the share of programs with at least one non-implemented SB remained stable while average number of SBs not met at review increased. - Measurement: - Overall implementation measured as weighted average of implementation status where: - conditions not met or waived = 0, - partially met or met with delay = 0.5, - fully met = 1. - By design implementation ratio varies between 0 and 1. - Relationship with past performance: - Countries with good track record in previous review tend to have better implementation. - Implementation rate of structural conditionality generally improved at similar pace for the average, the median, and the bottom quartile—indicating across-the-board progress. - Flexible timelines: - Flexibility in timelines enhanced implementation. - About 50 percent of measures that were reset for a future date were eventually met. - Table 5 status of implementation of structural conditions, 2002-2011: On the initially scheduled test date — Met 74, Not met 15, Pending 11 (percent); On the rescheduled test date (when rescheduled) — Met 49, Not met 23, Pending 27 (percent). - Parsimony and depth: - A panel regression showed a larger number of conditions in a given review tended to have negative correlation with overall implementation (Table 6). - Implementation appeared unrelated to the depth of conditionality, pointing to importance of other determinants such as capacity or ownership. - Combination of low capacity and a large number of conditions was linked to worse implementation. - Table 6 panel results (All programs, GRA, PRGT) show negative and statistically significant coefficients on Number of conditions (e.g., -0.018*** in column (1)); Number of conditions × CPIA interaction positive in one specification (0.004*). - Prior actions (PAs): - Programs with lower implementation ratios faced higher probability of having a PA at current or next review. - A higher number of PAs did not necessarily translate into improved overall implementation, suggesting limited effect of PAs on program execution. ### Institutional classification and determinants of structural conditionality (Appendices) - Institutional classification maps structural conditionality to three categories: - Core areas of Fund expertise: (1) fiscal measures (revenue, revenue administration, budget preparation, debt-management, fiscal transparency, inter-governmental relations); (2) central bank measures (operations/reforms, auditing, transparency, financial controls); (3) measures related to financial sector in advanced economies; (4) exchange rate systems and controls. - Areas of shared expertise: (1) public expenditure (auditing, accounting, financial controls); (2) civil service and public employment reforms and wages, non-public sector labor markets; (3) pension and other social sector reforms; (4) public enterprise reform and pricing; (5) financial sector reforms in low-income and emerging markets; (6) PRSP development and implementation. - Non-core structural conditionality: (1) international trade policy (excluding custom reform); (2) economic statistics; (3) other structural measures (e.g., natural resources and agricultural policies). - Trends and determinants: - Reclassification of conditionality goes back to 2002; trends point to further focus of structural conditionality on areas of core Fund expertise. - Appendix analysis shows Fund-supported programs strongly focused with limited dispersion: even at lowest quartile, general government-related structural conditionality accounted for 40 percent of total in PRGT programs and almost 30 percent in GRA programs. - Probit models indicate: - In GRA programs, fiscal conditionality was heavier with favorable initial conditions (lower inflation or higher growth), suggesting focus on fiscal structural reform when policy space exists. - Financial sector conditionality was heavier with larger vulnerabilities (wider fiscal deficit, larger financial exposure through assets and liabilities, larger external exposure through trade). - Focus on public employment was stronger when vulnerabilities (lower growth, larger openness) were more important. *Source: _061812 - 51. Country case studies indicated that the implementation of the Guidelines with* ### 72. In PRGT programs, the focus of conditionality responded to initial ### _061812 - 72. In PRGT programs, the focus of conditionality responded to initial ### PRGT conditionality and determinants - In PRGT programs, the focus of conditionality responded to initial macroeconomic weaknesses as well as to policy space. - More fiscal conditions were likely with larger initial debt and lower reserves, although the focus on fiscal conditions tended to be more important when more policy space in terms of fiscal deficit was available. - Central bank conditionality was also set with regards to vulnerabilities (the probability of setting a large number of conditions increased with inflation, lower reserves) and policy space (more conditions were set when the financial exposure was lower). ### Probit analysis (Appendix 2 summary) - Large number of conditions is defined as a number of conditions within the last quartile. - * and ** denote significance at the 10 and 5 percent significance level, respectively. - (Table columns and coefficients shown in source indicate estimated relationships between sectoral shares of structural conditionality and covariates such as Debt/GDP at t-1, Inflation (t-1), Fiscal balance (t-1), Reserves (t-1), Growth (t-1), Current account/GDP (t-1), Assets and liabilities (t-1), Trade openness index, with Number of observations reported as 31 for PRGT programs and 53 for GRA programs in the table.) ### Leveraging surveillance for conditionality (Appendix 3 findings) - Sample and methodology: - Analysis based on review of programs approved since 2008 which had a recent Article IV consultation (less than two years old) that was not combined with a program review or approval. - Sample includes 34 countries: 23 (or 68 percent) with GRA programs and 11 (or 32 percent) with PRGT programs. - The analysis maps conditionality set between program approval and the second review to staff’s recommendations in the latest available Article IV concluded the year preceding program approval. - Main results: - About 48 percent of conditions set at program approval were foreshadowed in previous Article IV consultations. - The share of foreshadowed conditions set up to the second review declines to 45 percent. - SBs and SPCs were mapped more closely to Article IV recommendations than PAs: - 50 percent of SBs and SPCs were foreshadowed in previous surveillance reports at program approval, and about 45 percent by the second review. - The average was 37 percent for PAs. - Dispersion and program type: - Conditionality in PRGT programs tended to be generally well related to previous surveillance recommendations; conditionality in GRA programs exhibited a much larger variation. - By the time of the second review, the dispersion across program types becomes more similar. - Time variation: - The share of conditionality foreseen in previous Article IV reports was the lowest in 2008 for GRA programs, reflecting the unexpected impact of the crisis on countries such as Iceland or Latvia. - For PRGT-eligible countries, program structural conditionality was the least related to past surveillance in 2009. - By the end of the period, the share of structural conditionality that could be mapped to previous Article IV recommendations tended to increase. - Country examples on mapping and timing: - Hungary: two of three measures had to do with bank resolution, in line with surveillance stressing strengthening bank supervision and regulation. - Costa Rica: need for an improved bank intervention and resolution framework identified in previous Article IV reports and determined most conditionality. - Greece: the 2009 Article IV report laid the basis for the fiscal measures making up the bulk of Fund conditionality. - Cases where stabilization imperatives crowded out surveillance recommendations: - Gambia: seven of fifteen measures related to reporting audited fiscal and monetary data, which had not been anticipated. - Pakistan and Sri Lanka: stabilization imperatives crowded out measures linked to structural surveillance recommendations (e.g., Sri Lanka reforms of state-owned banks were not in the program but recapitalization of a private bank was). - Cases where surveillance recommendations were added during reviews: - Iceland: reform of the Housing Financing Fund and local government fiscal reforms incorporated during reviews. - Greece: reviews added labor market reform to conditionality. - Pakistan: tax administration-related conditionality added during the program. - Hungary: pension reform conditionality added at later stages. - Togo: initial program fiscally oriented, but financial sector reforms were added later. - Policy implication: - Surveillance recommendations could be better prioritized to enhance leverage into program design: - Article IV Staff Appraisals typically cover an array of measures and reforms; some degree of prioritization, taking into account technical and political feasibility, might be helpful in setting benchmarks for policy makers and preparing the ground should a program be requested. - Conversely, program documents could note the surveillance issues to be addressed once the immediate crisis has abated. ### Discontinuation of SPCs and ownership (Appendix 4) - One aim of discontinuing SPCs was to reduce stigma attached to programs to enhance ownership through improved flexibility in timing of structural reforms. - Evidence suggests discontinuing SPCs did not worsen overall implementation of conditionality, which tended to improve in 2009 and 2010; increased flexibility in the timeline supported a better implementation record for structural conditionality. - Fund survey results: - Less than 30 percent of mission chiefs and resident representatives (out of a total of 84 respondents) found that phasing out SPCs had increased ownership or reduced stigma. - Responses confirmed that discontinuation of SPCs had not entailed worse implementation of program conditionality. - Some interlocutors felt discontinuation had reduced the Fund's leverage in supporting structural reforms. ### Guidance on the principle of clarity (Appendix 5) - The 2006 Guidance Note stipulated that: - The criticality of program goals and conditionality should be clearly exposed; - The link between structural conditions and program objectives should be brought out; - Where possible, discussions on policy options with authorities should be conveyed; - Conditions outside Fund core areas of expertise should be discussed in detail; - Fund conditions should be identified (in a table) separate from the authorities’ agenda; - Concerns related to implementation should be discussed; and - Bank-Fund collaboration should be noted. - The 2008 Guidance Note added that: - 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, e.g., by presenting them in a separate text table; - Board documents should transparently report when expertise in a critical reform area is not available within other multilateral institutions; and - The authorities’ reform efforts should be clearly distinguished from conditionality. - The 2009 Guidance Note requested that: - Staff reports indicate what is expected in the subsequent review(s) and specify critical future actions as structural benchmarks; and - Staff reports for program reviews provide an update related to the objectives of the structural reform agenda, as well as the strategies adopted to achieve them. *Source: MONA and staff calculations; text and figures as provided in the source content.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2012/_061812.pdf_