Key Messages
The most significant improvements in domestic policy processes include a better diagnosis of the nature of poverty and greater inclusion of various groups. The latter indicates some strengthening of ownership, but in macroeconomic policy areas it remains narrowly based.
Feedback into the broader policy debate in macroeconomic and related structural areas remains limited. However, when a more informed debate and broadening of the policy space did take place on these issues, the policy out-come was generally positive. In contrast, when controversial issues involving difficult trade-offs were avoided during the PRS process, the consequences for “downstream” policies were generally unfavorable.
There has been some strengthening of domestic public accountability—an essential element in the approach—but there is also evidence that undue focus on BWI procedures hampers the development of systems of domestic accountability. The lack of involvement of parliaments is a concern, as is the disconnect in some cases between the “participatory” framework and normal political processes. In most countries, the PRS process is still a long way from providing a strategic “road map” for setting priorities and resolving trade-offs, backed by institutional arrangements for implementation.
BWI measures of the quality of policies and institutions suggest that PRS countries generally started out in a better position than non-PRS countries but did not improve at a faster pace. Important exceptions are public sector management, transparency, and accountability, where some gains have been made, but there is still a long way to go. Trade restrictiveness has also declined more in PRS/PRGF countries than in other low-income countries.
Short-term growth outcomes for PRS/PRGF countries do not show much change from earlier periods, although these countries seem to have weathered the worsening of the external environment in 2000–02 better than other low-income countries.
Evidence on poverty-related outcomes, drawn from the parallel OED evaluation, are still too limited to draw definitive conclusions. The most notable improvements are with respect to various input- and output-related measures (e.g., construction of schools and number of teachers, school enrollment, vaccination rates, and expansion of water facilities). But outcomes such as maternal and infant mortality rates have generally not improved.
Thus far, we have examined the PRS/PRGF process commenting on intermediate-stage inputs and outputs and the role of the IMF. As pointed out in Chapter 1, progress toward the ultimate objectives such as poverty reduction can be credibly assessed only over a much longer time frame. However, it is possible to evaluate what has changed with respect to the two key intermediate outcomes that are likely to be critical to the achievement of longer-term goals: (i) the nature of domestic policy processes; and (ii) actual policies. We also examine the available evidence on aid flows and short-term growth response as well as the initial limited evidence on various poverty indicators. It should be borne in mind that prudence is required in attributing the changes observed to the PRS approach, as the nature of the changes and data limitations make it impossible to test causality in any robust way.
Changes in Nature of Domestic Policy Processes in Macroeconomic and Related Areas
The adoption of the PRS and PRGF frameworks was expected to improve domestic policy processes because of (i) improved diagnostics feeding into a policy debate that considers a wider range of policy options and pays greater attention to poverty and growth effects; (ii) a deeper sense of ownership and national commitment regarding the poverty reduction and growth strategy; (iii) improved public accountability; and (iv) clearer priorities and design of public actions. Many aspects of these intermediate objectives have been discussed in earlier chapters. Here, we summarize a number of core messages, drawing upon additional evidence from the case studies. In particular, we have traced through the effects of the new approach on a number of specific policy issues in each of the case studies, asking whether a broader policy debate took place on each particular issue; whether a consideration of alternative options was triggered (i.e., was the “policy space” broadened); and what was the outcome in terms of policies actually implemented. We also reviewed evidence from the case studies on the extent to which institutional arrangements in three areas—macroeconomic policy formulation, policy imple-mentation (especially links to the budget process), and monitoring and evaluation—are adapting to the PRS/PRGF approach. The country-specific assessments are summarized in Annex 9. The following main messages emerge:
(i) The quality of poverty diagnosis has improved in most cases with more attention to non-income aspects, but there has been less progress in improving other analytical inputs into growth strategies. Significant feedback into the broader policy debate in macroeconomic and related structural areas has been relatively infrequent.1
When a more informed debate and a broadening of the policy space did take place (e.g., the fiscal strategy in Tanzania, petroleum taxation in Mozambique, and tax reform in Nicaragua), the policy out-come was often positive, but these instances were not very frequent. A more common occurrence was for controversial structural issues to be avoided during the PRS process—typically with the result that “downstream” policy discussions were handled in traditional negotiating frameworks, often resulting in failure of policies to address adequately the underly-ing issue.
(ii) There is some limited evidence of strengthened ownership, but it remains more narrowly based in the areas of macroeconomic and related policies and should not be taken to imply “consensus.”
Despite its limitations, the participatory process does seem to have added value in most cases—in-cluding through fostering a greater inclusion of various groups. Once again, this impact was greater on policy issues outside the macroeconomic area but there is some evidence of enhanced ownership of macroeconomic policies within the narrow official circle. The case studies suggest that the potential value added of the PRS approach with regard to this objective varies enormously depending on a coun-try’s circumstances and political structure. The approach as implemented did not allow sufficiently for these differences. In cases (e.g., Vietnam) where there was already a strongly country-owned approach to the development strategy, the potential contribution of a separate PRS approach to developing a “shared vision” was limited; alternative approaches aimed at adapting/influencing the existing domestic process may have achieved similar results more effectively, avoiding the creation of parallel processes.
The most desirable situation is one where the PRS approach strengthens ownership by working through normal political processes. The evidence on how well this has been achieved is mixed. On many issues there does not appear to have been much connection between the “participatory” framework and the regular political process. The debate over Nicaragua’s growth strategy is one obvious example. In this context, the lack of involvement of parliaments is a concern. However, in some cases, issues concerning the overall strategy have begun to be addressed as part of the electoral process (e.g., user fees in several African countries), and electoral and PRS processes seem to have reinforced each other in raising the profile of governance issues in many countries.
(iii) There has been some success in achieving greater public accountability but there is a long way to go.
Individual country experiences vary widely, but overall the PRS approach (and the HIPC Initiative) has generated some momentum toward public accountability through a delineation of concrete expected results and participatory monitoring and evaluation of these results. However, starting conditions were such that deep institutional changes will be required, of which only the first steps are under way even in best performance case.
Significant efforts have been made to put moni-toring and evaluation systems in place, and gradual progress continues to be made over time. However, some efforts have been overelaborate for existing domestic capacity.2 Workable institutional linkages for taking monitoring results and feeding them into the domestic policy (e.g., budgetary) process are crucial. Most countries are still not at this stage, but some of the more mature PRS cases (e.g., Mozam-bique and Tanzania) have made progress. In terms of budgetary accountability, strengthened public expenditure management systems are the key. Here the ev-idence suggests that progress is being made, albeit gradually. Most countries for which assessments are available are still a long way from achieving “good” standards (see Chapter 4).
It is important that the focus should be on building capacity by improving domestic systems and not on generating reports on the PRS for the BWIs and other donors. There are some suggestions in the case studies that concentration on ad hoc efforts to satisfy reporting requirements to the BWIs may hamper improvements in domestic accountability systems, which need to be anchored on improved regular bud-getary processes.
(iv) A mixed picture on improved setting of priorities and design of public actions.
The process was expected to lead, over time, to the establishment of an operationally meaningful growth and poverty reduction strategy composed of (i) strategic guidance, or a “road map,” for setting priorities and resolving trade-offs between compet-ing objectives; and (ii) working institutional arrangements for implementing, monitoring, and updating this road map. Our findings for each of the IEO country case studies are summarized in Table 6.1. The following messages emerge:
To date, no country has both components of such an operational framework. Vietnam comes closest, but the development plan—rather than the PRS per se—provides much of the strategic framework.
Progress is being made over time in some countries (e.g., Mozambique and Tanzania) to put in place the institutional arrangements for imple-menting/updating the strategic road map.
More generally, the three-way linkage between the PRSP, medium-term expenditure frame-work, and budget is typically poor, reflecting some combination of limited costing and priori-tization in the PRSP and the generally poor state of PEM.3 Of these, the PEM weaknesses are probably the most fundamental challenge, requiring comprehensive institutional reforms and capacity improvements without which any en-hanced prioritization in the PRSP will remain wishful thinking.
Assessment of Progress in the Country Cases Against Two Possible Intermediate Objectives1
The focus is on progress in the areas of macroeconomic and related structural issues. In all cases, the assessment is of what is happening with actual policy processes in the country, not on what the PRSP says as a document.
See text for full definition of the objectives.
For countries where they are available (i.e., HIPC cases), this assessment draws, inter alia, upon the results of the PEM benchmarking exercise
Assessment of Progress in the Country Cases Against Two Possible Intermediate Objectives1
| Country | Is there a country-driven strategy and decision-making process that provides an operational road map for achieving growth and poverty reduction and resolving trade-offs?2 | Are working arrangements in place for implementing, monitoring, and updating this road map?2,3 | Is progress toward these two objectives continuing and what should the IMF be doing to help the process? |
|---|---|---|---|
| Guinea | Country-driven strategy includes some macroeconomic policy and related structural reform elements. However, the strategy provides limited strategic guidance to decision making for lack of relevant macroeconomic framework and discussion of trade-offs and contingency plans. The strategy mostly consists of broad objectives, with little detail as to their operationalization or indication of how trade-offs would be handled. | No specific arrangements for implementation; proceeding in piecemeal manner. No comprehensive monitoring arrangements outside of standard budget monitoring; some preexisting outcome indicators monitored. No arrangements for feedback of monitoring into policy formulation or up dating of strategy. |
Process is stalled. IMF could help improve diagnostics base and strengthen capacity in its areas of competence and, on that basis, help operationalize the broad objectives/intentions stated in strategy. However, IMF has limited leverage on degree of institutionalization of the process and willingness of other donors to provide adequate funding. |
| Mozambique | The PARPA (PRSP) provides a country-driven strategy. Key operational decisions are made in the context of the preparation of the annual state budget and the accompanying annual Economic and Social Plan. | Monitoring is geared to the preparation of reports to parliament on budget execution (quarterly) and review of the Economic and Social Plan (annual). The analytical content of the reports to parliament is weak. Capacity constraints are hindering implementation and monitoring. Monitoring by a group of donors providing direct budgetary support is by means of a common performance assessment framework (PAF). |
A multifaceted program of public sector reforms is under way, including with technical assistance in PEM-related areas from the IMF. A Poverty Observatory (established in April 2003) is intended to focus on learning lessons from experience with implementation and providing feedback to policy formulation and implementation. |
| Nicaragua | Initial strategy paper lacked fully developed growth chapter and the macroeconomic framework was superseded by events. Thus, there was little guidance for decision making in these areas. Relevant structural issues were mentioned(e.g., tax reform, and banking sector resolution) but not in a manner that provided strategic guidance. | Arrangements for implementation and monitoring are in the process of being strengthened but remain weak and uncoordinated. Updating of strategy, in particular the growth pillar, was difficult initially but has improved recently. | Modification of strategy is now under way after broadening of consultation in Nicaragua and discussions with donors. IMF could contribute to analysis of growth and macro-micro issues, and informing broader debate. PEM remains a key obstacle to implementation, and collaborative efforts to strengthen this area would be an important contribution. |
| Tanzania | As a document, the PRSP had large gaps and was primarily focused on social sector issues. As a process, however, institutional arrangements have been strengthened. The public expenditure review/medium expenditure framework (PER/MTEF) has in the past five years become the main avenue for public policy debate and strategy formulation. A macro subgroup focuses on macroeconomic issues. The PER/MTEF framework is credited for preserving priority expenditure in the face of sharp shortfalls in donor disbursements. | The updating of the “road map” is implicit in the structure of the PER/MTEF activities. Frequent meetings ensure that discussions remain current, while a national consultative meeting is held once a year to discuss issues and gather views from a broader constituency. A Poverty Reduction Support Group, comprising like-minded donors, monitors developments together with the government and the World Bank on the basis of a performance assessment framework (PAF). However, inputs of civil society on macroeconomic issues are still limited. | Although the PER/MTEF approach is generally deemed to be a success, with the country registering improvement in its expenditure management system (PEM), doubts have been raised on how efficiently PER/MTEF analyses (and thus priorities) are reflected in the budgets. Delays in donor disbursements are still a factor, domestic technical capacities are low, and politics often override these processes. The IMF could participate more in these broader processes—providing guidance on macroeconomic issues and requisite TA to boost capacity. |
| Tajikistan | The PRSP is not sufficiently prioritized to provide an operational road map. | Neither the PRSP nor the public investment program is well integrated into the budget process. Arrangements for monitoring are not yet fully operational. A core set of intermediate indicators for measuring progress in implementation is yet to be established. |
There is ongoing work to strengthen the MTEF and to use it for integrating the public investment program into the budget process. |
| Vietnam | Country-driven strategy for setting broad policy direction remains the regularly updated socioeconomic development plans. However, these plans are not well tied to clear actions or monitorable benchmarks. The PRS process has helped enhance the transparency of the policy debate and implementation of broad policy objectives. Originally, the PRSP was not well aligned with parts of the development plans, including the public investment program. An “infrastructure” chapter was added to the PRSP later. | An extensive system of indicators was developed to monitor implementation of the PRSP. However, the system—while comprehensive—is excessively ambitious relative to implementation capacity. | Increased transparency and monitoring of strategy, as well as expanded participation in deliberations, are helping make the government’s strategy more “operational.” There is significant interest within the government and civil society and among donors for the IMF to help build capacity across the various stakeholder groups to understand and participate in discussions of macroeconomic policy issues. |
The focus is on progress in the areas of macroeconomic and related structural issues. In all cases, the assessment is of what is happening with actual policy processes in the country, not on what the PRSP says as a document.
See text for full definition of the objectives.
For countries where they are available (i.e., HIPC cases), this assessment draws, inter alia, upon the results of the PEM benchmarking exercise
Assessment of Progress in the Country Cases Against Two Possible Intermediate Objectives1
| Country | Is there a country-driven strategy and decision-making process that provides an operational road map for achieving growth and poverty reduction and resolving trade-offs?2 | Are working arrangements in place for implementing, monitoring, and updating this road map?2,3 | Is progress toward these two objectives continuing and what should the IMF be doing to help the process? |
|---|---|---|---|
| Guinea | Country-driven strategy includes some macroeconomic policy and related structural reform elements. However, the strategy provides limited strategic guidance to decision making for lack of relevant macroeconomic framework and discussion of trade-offs and contingency plans. The strategy mostly consists of broad objectives, with little detail as to their operationalization or indication of how trade-offs would be handled. | No specific arrangements for implementation; proceeding in piecemeal manner. No comprehensive monitoring arrangements outside of standard budget monitoring; some preexisting outcome indicators monitored. No arrangements for feedback of monitoring into policy formulation or up dating of strategy. |
Process is stalled. IMF could help improve diagnostics base and strengthen capacity in its areas of competence and, on that basis, help operationalize the broad objectives/intentions stated in strategy. However, IMF has limited leverage on degree of institutionalization of the process and willingness of other donors to provide adequate funding. |
| Mozambique | The PARPA (PRSP) provides a country-driven strategy. Key operational decisions are made in the context of the preparation of the annual state budget and the accompanying annual Economic and Social Plan. | Monitoring is geared to the preparation of reports to parliament on budget execution (quarterly) and review of the Economic and Social Plan (annual). The analytical content of the reports to parliament is weak. Capacity constraints are hindering implementation and monitoring. Monitoring by a group of donors providing direct budgetary support is by means of a common performance assessment framework (PAF). |
A multifaceted program of public sector reforms is under way, including with technical assistance in PEM-related areas from the IMF. A Poverty Observatory (established in April 2003) is intended to focus on learning lessons from experience with implementation and providing feedback to policy formulation and implementation. |
| Nicaragua | Initial strategy paper lacked fully developed growth chapter and the macroeconomic framework was superseded by events. Thus, there was little guidance for decision making in these areas. Relevant structural issues were mentioned(e.g., tax reform, and banking sector resolution) but not in a manner that provided strategic guidance. | Arrangements for implementation and monitoring are in the process of being strengthened but remain weak and uncoordinated. Updating of strategy, in particular the growth pillar, was difficult initially but has improved recently. | Modification of strategy is now under way after broadening of consultation in Nicaragua and discussions with donors. IMF could contribute to analysis of growth and macro-micro issues, and informing broader debate. PEM remains a key obstacle to implementation, and collaborative efforts to strengthen this area would be an important contribution. |
| Tanzania | As a document, the PRSP had large gaps and was primarily focused on social sector issues. As a process, however, institutional arrangements have been strengthened. The public expenditure review/medium expenditure framework (PER/MTEF) has in the past five years become the main avenue for public policy debate and strategy formulation. A macro subgroup focuses on macroeconomic issues. The PER/MTEF framework is credited for preserving priority expenditure in the face of sharp shortfalls in donor disbursements. | The updating of the “road map” is implicit in the structure of the PER/MTEF activities. Frequent meetings ensure that discussions remain current, while a national consultative meeting is held once a year to discuss issues and gather views from a broader constituency. A Poverty Reduction Support Group, comprising like-minded donors, monitors developments together with the government and the World Bank on the basis of a performance assessment framework (PAF). However, inputs of civil society on macroeconomic issues are still limited. | Although the PER/MTEF approach is generally deemed to be a success, with the country registering improvement in its expenditure management system (PEM), doubts have been raised on how efficiently PER/MTEF analyses (and thus priorities) are reflected in the budgets. Delays in donor disbursements are still a factor, domestic technical capacities are low, and politics often override these processes. The IMF could participate more in these broader processes—providing guidance on macroeconomic issues and requisite TA to boost capacity. |
| Tajikistan | The PRSP is not sufficiently prioritized to provide an operational road map. | Neither the PRSP nor the public investment program is well integrated into the budget process. Arrangements for monitoring are not yet fully operational. A core set of intermediate indicators for measuring progress in implementation is yet to be established. |
There is ongoing work to strengthen the MTEF and to use it for integrating the public investment program into the budget process. |
| Vietnam | Country-driven strategy for setting broad policy direction remains the regularly updated socioeconomic development plans. However, these plans are not well tied to clear actions or monitorable benchmarks. The PRS process has helped enhance the transparency of the policy debate and implementation of broad policy objectives. Originally, the PRSP was not well aligned with parts of the development plans, including the public investment program. An “infrastructure” chapter was added to the PRSP later. | An extensive system of indicators was developed to monitor implementation of the PRSP. However, the system—while comprehensive—is excessively ambitious relative to implementation capacity. | Increased transparency and monitoring of strategy, as well as expanded participation in deliberations, are helping make the government’s strategy more “operational.” There is significant interest within the government and civil society and among donors for the IMF to help build capacity across the various stakeholder groups to understand and participate in discussions of macroeconomic policy issues. |
The focus is on progress in the areas of macroeconomic and related structural issues. In all cases, the assessment is of what is happening with actual policy processes in the country, not on what the PRSP says as a document.
See text for full definition of the objectives.
For countries where they are available (i.e., HIPC cases), this assessment draws, inter alia, upon the results of the PEM benchmarking exercise
One would not necessarily expect very rapid progress toward a fully fledged operational road map and implementation framework in, say, the first round of a country’s PRSP. However, it is hard to see how the central elements of the PRS initiative could be achieved without substantial progress toward such a framework.4 The evaluation suggests three important elements that must be part of any way for-ward. First, PRS-specific institutional arrangements operating in parallel with existing domestic processes are unlikely to be able to cope in a sustain-able manner with the breadth and complexity of issues at stake and may well distract from efforts to strengthen existing domestic processes that should be the primary focus of attention. Second, the PRS must include a discussion of key elements of the macroeconomic strategy, and of what priorities will guide policies in the event of unanticipated developments. Third, most countries face tremendous technical capacity challenges in developing such a road map, often spanning several of the following critical areas: setting up databases, analyzing data, policy formulation, policy implementation and monitoring, developing macroeconomic frameworks, and mobilizing adequate external assistance. The BWIs should play a key role in helping countries diagnose the key capacity bottlenecks and mobilize support (including, where appropriate, through direct TA) to alleviate them.
Evidence on What Has Happened to Actual Policies and Final Outcomes
In this section, we give our assessment on three questions. First, is there any evidence that the types of policies implemented in PRS/PRGF countries are different, in aggregate, from those implemented prior to the new approach, or from policies in low-income countries that have not availed themselves of the new initiative?5 Second, what is happening to overall aid flows to countries engaged in the approach? Third, what is the preliminary evidence on growth and poverty reduction outcomes?
Policy outcomes6
One way of assessing policy outcomes is to ask whether policies in PRS countries are moving in the direction that the BWIs themselves judge as appropriate. This clearly cannot be the only benchmark of initiative-wide progress, given the importance of domestic ownership and the tenuous nature of our understanding about country-specific links between policies and growth/poverty reduction, but it provides one useful frame of reference.
The most comprehensive available measure is the World Bank’s Country Policy and Institutional Assessment (CPIA) scores.7 We draw on an OED analysis for groups of PRSP and non-PRSP countries to see whether countries that successfully entered the PRS process had a different experience with respect to the “quality” of policies and institutions as measured by these scores. The results suggest the following:8
Countries that entered the PRS process during 1999–2003 (i.e., 35 countries) already had, at the outset, stronger average policy settings, as measured by the CPIA ratings, than other low-income countries (Table 6.2).
Improvements in the overall policy rating from 1999 to 2003 have been modest for low-income countries as a group, and there was no signifi-cant difference between PRSP and non-PRSP countries.
The only area where PRSP countries improved their CPIA performance more than non-PRSP countries was with regard to public sector man-agement and institutions, even though they al-ready started out with a better policy setting in this area. The largest improvements were in the areas of the quality of budgetary management and transparency, accountability and corruption in the public sector. This suggests that the em-phasis on public expenditure management issues is producing some benefits.
With regard to structural policies, PRSP countries had “better” policies than non-PRSP countries at the outset, but the gap is closing. With regard to external trade, however, a different data set, drawing on IMF measures of trade re-strictiveness and average import tariff rates, sug-gests that trade regimes in countries with PRSPs and/or PRGF-supported programs were liberal-ized at a somewhat faster pace, on average, than in other low-income countries (Table 6.3).9
Since improvements in the quality of policies and institutions take time, one might expect that our sample of 23 “mature” PRSP countries (i.e., with over a year of implementation experience) would have stronger gains than more recent PRSP countries (i.e., those finalized in 2003). However, this was not the case. In the area of “economic management” policies, which are of most direct relevance to the IMF, only countries with later PRSPs showed policy gains.
CPIA Ratings for PRSP and Non-PRSP Groups of Low-Income Countries
Thirty-five countries with PRSPs at end-2003
Covers 28 IDA countries without PRSPs at end-2003; excludes small island states.
A component of the “structural policies” category.
A category of the “public sector management” category.
CPIA Ratings for PRSP and Non-PRSP Groups of Low-Income Countries
| PRSP Countries1 | Non-PRSP IDA Countries2 | ||||||
|---|---|---|---|---|---|---|---|
| 1999 | 2003 | Percent change |
1999 | 2003 | Percent change |
||
| Overall rating | 3.36 | 3.47 | 3.3 | 2.78 | 2.91 | 4.7 | |
| Rating by policy category | |||||||
| Economic management | 3.76 | 3.77 | 0.3 | 2.91 | 2.93 | 0.7 | |
| Structural policies | 3.43 | 3.46 | 0.9 | 2.83 | 3.01 | 6.4 | |
| Social inclusion policies | 3.33 | 3.53 | 6.0 | 2.83 | 3.00 | 6.0 | |
| Public sector management | 2.99 | 3.19 | 6.7 | 2.57 | 2.68 | 4.3 | |
| Memorandum items (selected components of rating categories) | |||||||
| Competitive environment for the private sector1 | 3.54 | 3.60 | 1.7 | 2.84 | 2.96 | 4.2 | |
| Quality of budgetary and financial management4 | 3.12 | 3.46 | 10.9 | 2.54 | 2.73 | 7.5 | |
| Transparency, accountability, and corruption in public sector4 | 2.81 | 3.08 | 9.6 | 2.45 | 2.61 | 6.5 | |
Thirty-five countries with PRSPs at end-2003
Covers 28 IDA countries without PRSPs at end-2003; excludes small island states.
A component of the “structural policies” category.
A category of the “public sector management” category.
CPIA Ratings for PRSP and Non-PRSP Groups of Low-Income Countries
| PRSP Countries1 | Non-PRSP IDA Countries2 | ||||||
|---|---|---|---|---|---|---|---|
| 1999 | 2003 | Percent change |
1999 | 2003 | Percent change |
||
| Overall rating | 3.36 | 3.47 | 3.3 | 2.78 | 2.91 | 4.7 | |
| Rating by policy category | |||||||
| Economic management | 3.76 | 3.77 | 0.3 | 2.91 | 2.93 | 0.7 | |
| Structural policies | 3.43 | 3.46 | 0.9 | 2.83 | 3.01 | 6.4 | |
| Social inclusion policies | 3.33 | 3.53 | 6.0 | 2.83 | 3.00 | 6.0 | |
| Public sector management | 2.99 | 3.19 | 6.7 | 2.57 | 2.68 | 4.3 | |
| Memorandum items (selected components of rating categories) | |||||||
| Competitive environment for the private sector1 | 3.54 | 3.60 | 1.7 | 2.84 | 2.96 | 4.2 | |
| Quality of budgetary and financial management4 | 3.12 | 3.46 | 10.9 | 2.54 | 2.73 | 7.5 | |
| Transparency, accountability, and corruption in public sector4 | 2.81 | 3.08 | 9.6 | 2.45 | 2.61 | 6.5 | |
Thirty-five countries with PRSPs at end-2003
Covers 28 IDA countries without PRSPs at end-2003; excludes small island states.
A component of the “structural policies” category.
A category of the “public sector management” category.
Selected Policy and Outcome Indicators for PRGF-Eligible Countries1
(In percent of GDP and period averages, unless otherwise indicated)
Covers 72 countries for which data are available in WEO.
Covers 46 countries that had ESAF/PRGF arrangements in place during 2000–02.
Covers 23 countries that had full PRSPs at end-2002.
Covers 26 PRGF-eligible countries that did not have a PRGF arrangement in place during 2000–02.
Based on the IMF’s “Trade Restrictiveness Rate.” Countries are rated on a scale of 1 (few restrictions) to 10 (highly restrictive).
Selected Policy and Outcome Indicators for PRGF-Eligible Countries1
(In percent of GDP and period averages, unless otherwise indicated)
| PRGF 2000–02 | ||||||
|---|---|---|---|---|---|---|
| Total2 | Of which PRSP 233 | Other PRGF eligible4 | ||||
| 1997–99 | 2000–02 | 1997–99 | 2000–02 | 1997–99 | 2000–02 | |
| Fiscal balance including grants | –4.8 | –4.7 | –3.6 | –4.9 | –4.1 | –5.4 |
| Government revenues and grants | 20.8 | 22.2 | 20.4 | 21.5 | 25.7 | 26.4 |
| Government expenditure and net lending | 25.5 | 26.9 | 24.0 | 26.5 | 29.8 | 31.8 |
| Current account balance | –8.9 | –8.0 | –8.3 | –6.9 | –8.3 | –6.0 |
| International reserves (months of imports; end of period) | 3.5 | 4.7 | 3.8 | 5.0 | 3.5 | 4.1 |
| External debt (end of period) | 105.8 | 99.7 | 97.9 | 92.2 | 58.6 | 57.9 |
| GDP growth (in percent) | 3.8 | 4.0 | 4.8 | 4.4 | 4.3 | 1.8 |
| Change in terms of trade | 0.7 | –0.6 | 1.4 | –2.0 | 3.0 | 3.1 |
| Trade restrictiveness rating (end of period)5 | 4.0 | 3.1 | 3.7 | 2.1 | 5.5 | 5.0 |
| Average tariff rate (end of period) | 15.2 | 13.0 | 14.0 | 12.6 | 20.2 | 20.7 |
Covers 72 countries for which data are available in WEO.
Covers 46 countries that had ESAF/PRGF arrangements in place during 2000–02.
Covers 23 countries that had full PRSPs at end-2002.
Covers 26 PRGF-eligible countries that did not have a PRGF arrangement in place during 2000–02.
Based on the IMF’s “Trade Restrictiveness Rate.” Countries are rated on a scale of 1 (few restrictions) to 10 (highly restrictive).
Selected Policy and Outcome Indicators for PRGF-Eligible Countries1
(In percent of GDP and period averages, unless otherwise indicated)
| PRGF 2000–02 | ||||||
|---|---|---|---|---|---|---|
| Total2 | Of which PRSP 233 | Other PRGF eligible4 | ||||
| 1997–99 | 2000–02 | 1997–99 | 2000–02 | 1997–99 | 2000–02 | |
| Fiscal balance including grants | –4.8 | –4.7 | –3.6 | –4.9 | –4.1 | –5.4 |
| Government revenues and grants | 20.8 | 22.2 | 20.4 | 21.5 | 25.7 | 26.4 |
| Government expenditure and net lending | 25.5 | 26.9 | 24.0 | 26.5 | 29.8 | 31.8 |
| Current account balance | –8.9 | –8.0 | –8.3 | –6.9 | –8.3 | –6.0 |
| International reserves (months of imports; end of period) | 3.5 | 4.7 | 3.8 | 5.0 | 3.5 | 4.1 |
| External debt (end of period) | 105.8 | 99.7 | 97.9 | 92.2 | 58.6 | 57.9 |
| GDP growth (in percent) | 3.8 | 4.0 | 4.8 | 4.4 | 4.3 | 1.8 |
| Change in terms of trade | 0.7 | –0.6 | 1.4 | –2.0 | 3.0 | 3.1 |
| Trade restrictiveness rating (end of period)5 | 4.0 | 3.1 | 3.7 | 2.1 | 5.5 | 5.0 |
| Average tariff rate (end of period) | 15.2 | 13.0 | 14.0 | 12.6 | 20.2 | 20.7 |
Covers 72 countries for which data are available in WEO.
Covers 46 countries that had ESAF/PRGF arrangements in place during 2000–02.
Covers 23 countries that had full PRSPs at end-2002.
Covers 26 PRGF-eligible countries that did not have a PRGF arrangement in place during 2000–02.
Based on the IMF’s “Trade Restrictiveness Rate.” Countries are rated on a scale of 1 (few restrictions) to 10 (highly restrictive).
These findings call for some caution in attributing to the PRS process credit for institutional and policy improvements across low-income countries.
Aid flows
To examine what has happened to aid flows to countries involved in the PRS/PRGF approach, we di-vided the sample of low-income (i.e., PRGF-eligible) countries into three groups: (i) the 46 countries with PRGF (or converted ESAF) arrangements in place during 2000–02; (ii) the subgroup of 23 countries with full PRSPs at end-2002; and (iii) other countries that had not used the approach.10
On average, the PRSP countries received higher aid flows than non-PRSP countries, and program countries received more aid than non-program countries, but much of this difference reflected starting conditions (Figure 6.1). For example, the PRSP group of countries were already commanding a higher proportion of aid flows on a per country basis at the outset of the initiative.11
Average Net ODA Flows to Groups of Low-Income Countries
(In millions of U.S. dollars)
Source: OECD.Since an analysis using broad groups cannot distinguish the influence of the precise timing of the PRSP, we also looked at the aid paths for individual countries. Only nine countries with full PRSPs had sufficient data to make an assessment of pre- and post-PRSP aid flows.12 Of these, six countries received higher aid after the PRSP (Bolivia, Burkina Faso, Mauritania, Mozambique, Tanzania, and Uganda), while average aid flows declined in three cases (Alba-nia, Honduras, and Nicaragua).
This evidence is inconclusive regarding the impact of the PRSP process so far on the overall selectivity of aid flows. In contrast, evidence discussed in the OED report indicates that World Bank lending to PRSP countries did expand faster than to other IDA countries.
Growth and poverty outcomes
Although it is too early to judge the impact on long-term growth and poverty reduction, there is some evidence on short-term outcomes, especially for growth. However, one has to bear in mind what was happening to the overall external environment. The period immediately preceding the introduction of the PRSP/PRGF was characterized by a favorable external environment, reflected in improving terms of trade (Table 6.3 and Figure 6.2). In contrast, the 2000–02 period was characterized by a deteriorating terms of trade for countries involved in the PRS/PRGF approach. Despite this deterioration, average growth in these countries was largely unchanged. (It improved marginally in the group of countries with PRGF-sup-ported programs and declined marginally in the sub-group of countries with full PRSPs.) In sharp contrast, growth performance deteriorated significantly for countries not participating in the approach, even though they were the only group that did not suffer a deterioration in the external terms of trade. However, while such patterns reveal interesting associations, they cannot prove causation. They suggest that countries involved with the PRS/PRGF approach have done better than other low-income countries in these early stages, but there has not yet been a marked pickup in growth that will be necessary for a substan-tial lasting impact on poverty levels.
Growth and Change in Terms of Trade
(In percent)
Sources: IMF (2002e and 2003d).1 Twenty-three countries with PRSPs at end-2002.2 Forty-six countries with ESAF/PRGF arrangements during 2000–02.3 Twenty-six other PRGF-eligible countries.We relied upon the parallel OED evaluation for assessments of progress toward the various poverty indicators and, more generally, toward the MDGs. We also draw upon the recent Global Monitoring Report (World Bank and IMF, 2004a). An overriding message is that very limited data is available yet for assessing progress vis-à-vis these fundamental objectives. However, the following broad messages put the overall initiative in context:
At the global level, the first MDG goal of halv-ing income poverty between 1990 and 2015 will likely be met. However, this largely reflects improvements in Asia, especially in two countries (China and India) that are not involved in the PRS approach. With current trends, sub-Saharan Africa is seriously off track—with only 8 countries representing only 15 percent of the region’s population projected as likely to achieve the goal (Table 6.4).13
The risks of shortfalls are more widespread with respect to the human development goals. With current trends, several regions will achieve or approach the goal of providing universal pri-mary education, but shortfalls are likely in sub-Saharan Africa and possibly in South Asia, and Middle East and North Africa as well. Prospects are gravest in health; only a small proportion (15–20 percent) of countries is likely to reach the goals for reducing child and maternal mor-tality or for access to safe drinking water and basic sanitation by 2015.
These global and regional trends mask consider-able variation across countries. Among the case study countries, the most spectacular progress was made in Vietnam, which reduced income poverty from 51 percent to 14 percent in little more than a decade.
For the PRSP countries, the OED analysis focused on the set of 12 countries that had issued at least one PRSP progress report. Nine of these country progress reports contained data on quantitative targets linked to poverty reduction. The data indicate notable improvements with re-spect to some inputs and outputs, but not much change with respect to final outcomes. Thus, for example, there have been (i) increases in gross primary school enrolment rates, the construction of schools, and the number of teachers; (ii) some improvements in the supply of drugs to health centers and in vaccinations; and (iii) ex-pansion of water facilities. By contrast, outcomes such as maternal and infant mortality rates have remained stagnant, and in many countries the poor quality of service delivery in both education and health continued.
OED analyzed progress toward the MDGs in the same set of 12 countries. As Table 6.5 indicates, relevant data are available for only about one-third of the 49 specified UN indicators. In criti-cal areas such as extreme poverty and maternal health, data are available only in a few countries or for a few years or for a single indicator. OED concludes that, based on these limited data, there have been modest improvements in liter-acy rates, including those of women, and the control of tuberculosis, but that infant mortality rates appear to have stagnated, a finding echoed by the case studies.
Progress Toward the Income Poverty Goal
Progress Toward the Income Poverty Goal
| Annual Average Growth Rates (2004–15) Scenario |
Population Living Under $1 a day | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Headcount (In percent) |
Number of persons (In millions) |
||||||||
| Per capita GDP |
GDP | 1990 | 2000 | 2015 | 1990 | 2000 | 2015 | ||
| East Asia and Pacific | 6.0 | 6.3 | 29.4 | 14.5 | 2.3 | 470 | 261 | 44 | |
| Europe and Central Asia | 3.9 | 3.8 | 1.4 | 4.2 | 1.3 | 6 | 20 | 6 | |
| Latin America and Caribbean | 2.3 | 3.7 | 11.0 | 10.8 | 7.6 | 48 | 56 | 46 | |
| Middle East and North Africa | 2.3 | 4.1 | 2.1 | 2.8 | 1.2 | 5 | 8 | 4 | |
| South Asia | 4.0 | 5.3 | 41.5 | 31.9 | 16.4 | 466 | 432 | 268 | |
| Sub-Saharan Africa | 1.8 | 3.8 | 47.4 | 49.0 | 42.3 | 241 | 323 | 366 | |
| Total | 3.7 | 4.7 | 28.3 | 21.6 | 12.5 | 1237 | 1100 | 734 | |
| Excluding China | 3.0 | 4.1 | 27.2 | 23.3 | 15.4 | 877 | 896 | 692 | |
Progress Toward the Income Poverty Goal
| Annual Average Growth Rates (2004–15) Scenario |
Population Living Under $1 a day | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Headcount (In percent) |
Number of persons (In millions) |
||||||||
| Per capita GDP |
GDP | 1990 | 2000 | 2015 | 1990 | 2000 | 2015 | ||
| East Asia and Pacific | 6.0 | 6.3 | 29.4 | 14.5 | 2.3 | 470 | 261 | 44 | |
| Europe and Central Asia | 3.9 | 3.8 | 1.4 | 4.2 | 1.3 | 6 | 20 | 6 | |
| Latin America and Caribbean | 2.3 | 3.7 | 11.0 | 10.8 | 7.6 | 48 | 56 | 46 | |
| Middle East and North Africa | 2.3 | 4.1 | 2.1 | 2.8 | 1.2 | 5 | 8 | 4 | |
| South Asia | 4.0 | 5.3 | 41.5 | 31.9 | 16.4 | 466 | 432 | 268 | |
| Sub-Saharan Africa | 1.8 | 3.8 | 47.4 | 49.0 | 42.3 | 241 | 323 | 366 | |
| Total | 3.7 | 4.7 | 28.3 | 21.6 | 12.5 | 1237 | 1100 | 734 | |
| Excluding China | 3.0 | 4.1 | 27.2 | 23.3 | 15.4 | 877 | 896 | 692 | |
Progress Toward Meeting the Millennium Development Goals in 12 Mature PRSP Countries1
The 12 countries are Albania, Burkina Faso, Ethiopia, Honduras, Malawi, Mauritania, Mozambique, Nicaragua, Niger, Tanzania, Uganda, and Vietnam.
Progress Toward Meeting the Millennium Development Goals in 12 Mature PRSP Countries1
| Number of Indicators Specified |
Availability | Improvement Between 1999 and 2003? |
||
|---|---|---|---|---|
| Number of Indicators |
Number of countries |
|||
| 1. Eradicate extreme hunger and poverty | 6 | 1 | 3 | Yes |
| 2. Achieve universal primary education | 4 | 1 | 12 | Yes |
| 3. Promote gender equality and empower women | 4 | 3 | 8 | Yes (2 of 3) |
| 4. Reduce child mortality | 3 | 3 | 12 | No (3 of 3) |
| 5. Improve maternal health | 2 | 0 | 0 | Unknown |
| 6. Combat HIV/AIDS, malaria, and other diseases | 10 | 1 | 8 | Yes |
| 7. Ensure environmental sustainability | 8 | 1 | 12 | Yes |
| 8. Develop a global partnership for development | 17 | 4 | 8 | Yes |
The 12 countries are Albania, Burkina Faso, Ethiopia, Honduras, Malawi, Mauritania, Mozambique, Nicaragua, Niger, Tanzania, Uganda, and Vietnam.
Progress Toward Meeting the Millennium Development Goals in 12 Mature PRSP Countries1
| Number of Indicators Specified |
Availability | Improvement Between 1999 and 2003? |
||
|---|---|---|---|---|
| Number of Indicators |
Number of countries |
|||
| 1. Eradicate extreme hunger and poverty | 6 | 1 | 3 | Yes |
| 2. Achieve universal primary education | 4 | 1 | 12 | Yes |
| 3. Promote gender equality and empower women | 4 | 3 | 8 | Yes (2 of 3) |
| 4. Reduce child mortality | 3 | 3 | 12 | No (3 of 3) |
| 5. Improve maternal health | 2 | 0 | 0 | Unknown |
| 6. Combat HIV/AIDS, malaria, and other diseases | 10 | 1 | 8 | Yes |
| 7. Ensure environmental sustainability | 8 | 1 | 12 | Yes |
| 8. Develop a global partnership for development | 17 | 4 | 8 | Yes |
The 12 countries are Albania, Burkina Faso, Ethiopia, Honduras, Malawi, Mauritania, Mozambique, Nicaragua, Niger, Tanzania, Uganda, and Vietnam.
In areas beyond the IMF’s primary responsibility, the OED evaluation notes that strategies in the health and education sectors are most developed, although some of this predates the PRSP. There has also been some improvement in coordination across sectoral strategies.
The parallel OED evaluation discusses monitoring systems in more detail. It highlights as a common problem attempts to track too many indicators.
This point draws on Bevan (2004).
The Joint Note on Supporting the PRS Process in Africa, pre-pared by the staffs of the European Commission and World Bank comes to a similar conclusion. See European Commission and World Bank (2003). Indeed, they argue that establishment of a common operational framework, including modalities for donor commitments of budgetary support, is essential for moving to-ward a more effective partnership model without which “there is a very real risk that the PRSP could become just another fad, adding work rather than reducing transaction costs.”
Such “before-after” and “control group” comparisons are sub-ject to well-known methodological problems and should not be interpreted as proving causation.
The CPIA gives the World Bank staff’s assessment of the quality of a country’s present policy and institutional framework. It includes 20 equally weighted dimensions in the overall rating of policy, encompassing economic management (comprising management of inflation and macroeconomic imbalances, fiscal policy, management of public debt, and management of the development program); structural policies (broadly covering trade, financial sector, and regulatory policies that determine the enabling climate for the private sector, as well as policies for environmental sustainability); policies for social inclusion/equity (covering policies for human resources development, gender, social protection, and equity of resource use); and public sector management and institutions (comprising public financial management, quality of public administration, control of corruption, and rules-based governance). Each dimension is rated on a six-point scale. The ratings for individual countries are not made public.
We focus on areas within the IMF’s primary responsibility. Such statistical associations do not prove causation, especially given the likelihood of two-way influences (i.e., countries may have PRSPs because they have “better” policies, as judged by the BWIs, rather than vice versa)
This comparison is only to determine what actually happened vis-à-vis these aspects of trade regimes. We make no judgments as to whether the sequencing, and so on was appropriate, which would go beyond the scope of this evaluation.
Data on net official development assistance (ODA) is from the OECD Development Assistance Committee (DAC) database and is only available on a comprehensive basis through 2002. Consequently, we are not yet able to assess systematically what has happened to aid flows for later PRSP cases.
Distinguishing PRSP countries by the maturity of their in-volvement in the process does not appear to alter this result: OED results suggest that the more “mature” PRSP countries were larger aid recipients at the outset but that overall aid to them did not expand faster than to other IDA countries.
We compared average aid flows in the three years before the PRSP to average flows in the years after the PRSP. The comparisons can be distorted by aid associated with natural disasters. This factor tends to bias upward “post-PRSP” aid in Mozambique and bias it downward in Nicaragua and Honduras.
See World Bank and IMF (2004a, Chapter II). These projec-tions are based on long-term economic scenarios prepared by World Bank staff and on specific assumptions about income dis-tribution (i.e., essentially unchanged, as measured by the Lorenz curve, except for China and India). Such projections are in-evitably highly tentative.
