Chapter 3: Fiscal Adjustment as Presented in Program Documents and the Internal Review Process
- International Monetary Fund. Independent Evaluation Office
- Published Date:
- November 2003
In this chapter, we turn to program documents submitted to the Board to consider what light they shed on the rationale for and magnitude of the fiscal adjustment proposed in programs. We then examine the internal review process prior to Board approval as well as during program implementation.
Fiscal Adjustment in Program Request Documents
Why would this be an important question? It could be argued that as long as the substantive aspects of program design have been vetted internally and with the country authorities, there is no reason to worry too much about presentational issues in program documentation. In the view of this evaluation, however, such presentation is indeed important. First, it allows the outside world—stakeholders in the country and the IMF constituency at large, as well as critics—to understand better the rationale for the program, assumptions being made, and why certain measures are taken and not others. It will help the institution convey the message that it has a coherent view that is country and program specific. Second, it may in itself improve program design: the more explicitly the program is explained the more careful the process of internal vetting will have to be.
To address these issues, we examined program request documents of 13 SBA and EFF arrangements.1
Do documents clearly discuss the motivation for the program? Do they explain the nature of the balance of payments problem the program is trying to correct?
Does the documentation discuss the program-specific mechanism through which the envisaged fiscal deficit adjustment will assist in solving or preventing the external imbalances described above?
Do documents explain the rationale for the magnitude and pace of the fiscal adjustment and how it is linked to other aspects of the program such as projections for the recovery of growth and private sector activity?
If there are other factors (besides balance of payments considerations) affecting the need, size, and pace of the envisaged fiscal adjustment, do documents discuss these factors with a reasonable degree of analysis and detail?
Finally, the IMF has often been criticized for paying much more attention to the magnitude of the needed fiscal adjustment than to the composition of the adjustment. We ask: do documents discuss specific reasons for the distribution of the fiscal adjustment between revenue and expenditure measures?2
The methodology used to answer these questions is necessarily subjective: we have reviewed program documents presented to the Board and scored them in each category as “highly satisfactory,” “satisfactory,” “marginally satisfactory,” and “unsatisfactory.” Box 3.1 summarizes the criteria used in the scoring (they are further elaborated in Appendix 2, where the code book used in the scoring is presented). This method of evaluating programs depends on subjective judgments of the evaluation team and this is an unavoidable limitation, which must be kept in mind. Nevertheless, we believe the exercise provides useful information to identify some basic patterns. It must also be emphasized that in this analysis we have not focused on judging the appropriateness or otherwise of the fiscal adjustment envisaged, but rather on whether the rationale for the adjustment proposed was clearly explained. To assess the appropriateness of the fiscal adjustment proposed in each case would have required an in-depth case study of each country, including an analysis of the combined effect of fiscal and other policies that is beyond the scope of this study.
Box 3.1.How Well Do Documents Explain the Rationale for Fiscal Adjustment?
The five questions raised in considering how well program documents explain the rationale of fiscal adjustment and the criteria used for rating program documents within these dimensions are as follows:
1. Do documents explain the source of the balance of payments problem the program is trying to correct?
We expected documents to provide a coherent and detailed explanation of the sources of the existing or impending external imbalance that the program aimed to correct or prevent. Balance of payments deficits stem from an excess of domestic absorption over income. However, balance of payments problems arise only when such a gap cannot be financed. The table below classifies the various ways the balance of payments problem can be created. When it is due to increases in excess absorption (rooted in the public or private sector), it gives rise to current account problems while financing problems are reflected in the capital account.
2. The link between the balance of payments problem to be corrected and the need for fiscal adjustment.
Does the documentation clearly explain the program-specific mechanism through which the envisaged fiscal deficit adjustment will assist solving/preventing the external imbalances stemming from the specific sources described above? This explanation becomes particularly important if the origin of the balance of payments problem was not directly fiscal (i.e., it emerged from somewhere other than the public sector quadrants of the table below).
3. The magnitude of the fiscal adjustment.
We expected documents to link the magnitude of the fiscal adjustment to the magnitude of the external adjustment, thus explicitly indicating the portion of the total external adjustment borne by the public sector. What is the basis for the “burden sharing” between the adjustment in the private and public sector? This is particularly important in cases where the program envisages an increase in net external financing (deterioration of the current account balance) while envisaging a reduction in the fiscal deficit. What assumptions are being made about the factors that may induce a reduction in the savings-investment balance of the private sector?
4. Other factors influencing the magnitude of the fiscal adjustment.
Programs may incorporate other factors (outside of the ongoing/impending external imbalance triggering the program) in deciding the envisaged fiscal deficit adjustment. It may include reducing inflation if the original deficit was monetized, reducing the crowding out of the private sector, etc. In these cases, do documents clearly discuss how these factors influence the fiscal adjustment?
5. The composition of the fiscal adjustment.
Do documents thoroughly explain the reasons for the envisaged balance between revenue and public spending changes? Are they related to initial levels and efficiency or equity considerations? Are they influenced by the need for speed and expediency?
|Current Account||Capital Account|
|Shocks to the|
supply of net financing
|Public sector||Example: increased fiscal deficits due to electoral cycle.||Example: terms of trade shock adversely affecting government revenue.||Rollover and default risk or crisis due to perceived unsustainability of debt (both public and total external) because of country-specific developments.||International contagion affecting private and official flows.|
|Private sector||Example: private sector consumption and spending bubble financed by borrowing.||Example: terms of trade shock adversely affecting private income or cost of imports.|
The main results from the analysis are presented in Table 3.1. They can be summarized as follows:
(1) About two-thirds of the programs have been classified as “unsatisfactory” or “marginally satisfactory” in terms of the discussion regarding the justification for IMF involvement or the need for a program in the first place. Most program request documents do not provide either an explanation or a sufficient discussion of the balance of payments problem (current or potential) calling for an IMF-supported program. Where external imbalances do not seem to be the main reason justifying an arrangement, documents often do not present a convincing case of why the arrangement is necessary/recommended on other grounds.3 Most contain a background section with recent economic developments, but the scope and degree of analysis varies greatly across programs. Some Board papers only provide one historical paragraph and take for granted the reasons why the program is needed.
(2) Even if it is accepted that in most cases the motivation for the program was to address external imbalances, a majority of the programs reviewed did not explain the links between the targeted fiscal adjustment and the envisaged improvement in the external situation. In only 40 percent of cases was there an explicit discussion of the program-specific mechanism through which fiscal adjustment could help improve the external imbalance.
(3) Even when the Board papers identify the link between fiscal adjustment and external adjustment, the documentation does not discuss how the specific pace and magnitude of the fiscal adjustment is being set to attain the new balance of payments situation envisaged in the program. This is the area with the worst scores: 9 out of the 13 cases were rated unsatisfactory. Only in one case was it satisfactory.
(4) When there were other factors (other than balance of payments considerations) affecting the envisaged fiscal adjustment, only half of the programs clearly explained how these factors influenced that adjustment. Furthermore, only in three cases (those judged as highly satisfactory in Table 3.1) was there an explanation of how these factors influenced the magnitude of the fiscal adjustment.
(5) Most programs provide a good explanation for the composition of the envisaged fiscal adjustment between revenue increases and spending reductions. In some cases, the analysis is very good indeed, providing not only a sense of why fiscal adjustment should be distributed in the proposed way, but also analyzing intra-revenue and intra-spending changes that can help improve the structure of public finance. However, in about one-third of the cases, programs provide only a long list of measures on both the revenue and spending side without a sufficient sense of rationale or priority.
|1. Does the document clearly discuss the motivation for the program (e.g., the sources of economic disequilibria (balance of payments or other) that the program is expected to address)?||Romania|
|2. Do documents explain the country-specific mechanism by which fiscal adjustment will contribute to address actual or potential balance of payments problems?||Ecuador||Bulgaria|
|3. Do documents discuss how the pace and magnitude of fiscal adjustment is being set in order to address the actual or potential balance of payments problems?||Venezuela||Ukraine|
|4. If there are other major factors affecting the envisaged fiscal deficit adjustment (other than balance of payments considerations), do documents explain clearly how they influence the magnitude of that adjustment?||Romania|
|5. Do documents explain the rationale for the composition of the fiscal deficit adjustment, for example, between revenue increases and spending reductions?||Bulgaria|
Fiscal adjustment and debt sustainability
An area that is conceptually important, but that received less attention than it deserved in the sample of program documents examined earlier, was the linkage between the fiscal deficit and debt sustainability. Table 3.2 provides data on the average ratio of public debt to GDP for 12 of the 13 countries for which IMF-supported program documents were examined in the last section.4 During the preprogram years, debt ratios were relatively low for Ukraine, Romania, and Uruguay but were relatively high in the other countries. Nevertheless, only five programs included a discussion linking the dynamics of public debt to the targeted fiscal adjustment, and even then the linkage to the magnitude of fiscal adjustment needed was weak. This is clearly an area where practice needs to be greatly strengthened and analyses made more explicit. This lacuna has been recognized and more recent practice seems much better.5
T−3 to T−1
T+1 to T+3
|Algeria (SBA 1994)||88.01||81.8||−6.2|
|Bulgaria (EFF 1998)||112.12||78.3||−33.8|
|Costa Rica (SBA 1995)||49.1||50.5||1.4|
|Ecuador (SBA 2000)||76.7||64.93||−11.8|
|Jordan (EFF 1999)||91.7||66.7||−25.0|
|Pakistan (SBA 2000)||93.1||95.94||2.8|
|Peru (EFF 1996)||50.5||41.3||−9.2|
|Philippines (SBA 1998)||108.5||128.0||19.5|
|Romania (SBA 1999)||27.92||28.73||0.8|
|Ukraine (EFF 1998)||30.1||44.7||14.6|
|Uruguay (SBA 2000)||28.1||63.13||35.0|
|Venezuela (SBA 1996)||63.7||30.6||−33.1|
Table 3.2 shows, for purposes of comparison, what happened to the debt ratios. The picture is mixed. Although debt ratios were reduced in half of the programs, in one case, Ecuador, it was largely due to debt-reduction initiatives during the program years.
These short-term changes are not necessarily a good indication of sustainability and robustness of fiscal systems over the medium term because they are influenced by short-term fluctuations in exchange rates, interest rate premia on public debt, and the like. Even if debt ratios rise in the short run, long-term sustainability may be improving if the higher debt levels reflect the short-term costs of reforms that yield benefits over the medium term. Conversely, short-term improvements may not be sustainable if they are achieved through short-term fiscal deficit reduction measures that quickly erode over time (e.g., increasing already high taxes over a low base therefore inducing further tax evasion, or unsustainable cuts in public sector wages without civil service reforms). These problems can be handled satisfactorily only through systematic debt-sustainability analysis which takes account of the different factors that affect debt profiles over time, in particular structural reforms in fiscal systems, an area to which we return later on.
Our evaluation suggests that although there is considerable variation across countries in the direction and size of the fiscal adjustment proposed in IMF-supported programs, there is not enough clarity and transparency in the way fiscal targets are set. Program documents should indicate clearly the extent to which fiscal deficit adjustments proposed are being driven by consideration of burden sharing in reducing aggregate demand, debt-sustainability considerations, or crowding-out concerns.
The Internal Review Process Prior to Board Approval and During Program Implementation
In addition to examining program documents we also attempted to evaluate the extent to which the internal review process for the 15 sample programs focused on key areas of public finance reform. We selected three broad areas: (1) macro and fiscal targets; (2) specific program design issues in the fiscal area; and (3) equity and social spending issues.
Comments from PDR and FAD were reviewed at three stages: (1) the briefs for negotiation; (2) Letter of Intent (LOI) or Memorandum of Economic and Financial Policies (MEFP) and staff reports for the initial program request; and (3) subsequent reviews during program implementation. In the following, a reference to “brief” indicates comments on the brief for the negotiating mission; “LOI/MEFP and Staff Report” deals with comments on the LOI/MEFP or Staff Report for the initial program request and “Reviews” covers remarks on all documents (briefs, LOI/MEFP, and staff reports) related to reviews during program implementation.
Overview of findings
Table 3.3 summarizes the findings by identifying the countries for which each issue listed was raised. Comments on the brief for the initial program request were largely concentrated on the macroeconomic framework and fiscal targets and covered 11 out of 15 programs. Comments on the brief regarding program design were limited to 6 programs and those dealing with equity and social spending extended to 7 programs. Taking all comments at every stage combined, reviewers commented on the macroeconomic framework and fiscal targets in all 15 programs while the other two topics were commented on in only 11 of the 15 programs.6
|Brief for Negotiation of the Arrangement||LOI/MEFP and Staff Report for the Program Request||Documents Relating to Reviews of the Arrangement|
|Macro and fiscal targets|
|Were questions raised on realism of GDP growth projections?||Egypt, Romania||Egypt||Bulgaria, Pakistan, Philippines, Romania, Uruguay|
|Were questions raised on realism of projected private sector investment response?||Costa Rica, Egypt, Jordan||Bulgaria||Algeria, Bulgaria, Egypt|
|Any suggestions to link growth/private sector recovery and fiscal adjustment?||Bulgaria||Peru, Philippines||Bulgaria, Philippines|
|Any suggestions to discuss the factors behind the magnitude and pace of the fiscal adjustment?||Bulgaria, Ecuador, Jordan, Pakistan, Peru, Philippines, Romania, Tanzania, Venezuela||Algeria, Bulgaria, Ecuador, Egypt, Peru, Venezuela||Algeria, Bulgaria, Ecuador, Egypt, Jordan, Peru, Philippines, Romania, Senegal, Tanzania, Venezuela|
|Any discussion of risks to achieving fiscal targets?||Costa Rica, Egypt, Pakistan, Romania||Costa Rica, Ecuador, Jordan, Pakistan, Peru, Philippines, Romania, Tanzania, Venezuela||Algeria, Bulgaria, Ecuador, Jordan, Peru, Philippines, Romania, Tanzania, Uruguay|
|Any suggestions to disentangle causes of fiscal underperformance?||Jordan||Bulgaria, Philippines, Romania||Bulgaria, Philippines|
|Any concerns raised on fiscal sustainability?||Bulgaria, Ecuador, Pakistan, Peru||Algeria, Bulgaria, Ecuador, Jordan, Peru, Romania, Tanzania||Algeria, Bulgaria, Ecuador, Egypt, Jordan, Pakistan, Peru, Philippines, Romania, Tanzania, Ukraine, Uruguay|
|Any comments regarding the need to consider past implementation problems in designing the program?||Bulgaria, Jordan, Pakistan, Peru||Jordan, Pakistan, Peru, Tanzania||Egypt, Jordan, Pakistan, Peru, Philippines, Ukraine, Venezuela|
|Any suggestions for stronger action in areas under the control of the executive that would demonstrate political commitment to the program?||Bulgaria, Ecuador, Jordan, Peru, Tanzania||Bulgaria, Ecuador, Jordan, Pakistan, Philippines, Romania, Tanzania||Bulgaria, Ecuador, Egypt, Jordan, Pakistan, Peru, Philippines, Tanzania|
|Any comments urging a more realistic time frame for reform?||Bulgaria, Jordan, Pakistan||Bulgaria, Pakistan|
|Equity and social spending|
|Any suggestions to deal with issues that would improve the equity of taxation?||Bulgaria, Egypt, Pakistan||Bulgaria, Egypt, Jordan, Peru, Venezuela||Bulgaria, Ecuador, Philippines, Tanzania|
|Any suggestions to deal with issues that would improve the equity of spending?||Egypt, Tanzania||Egypt, Jordan||Bulgaria, Ecuador, Tanzania|
|Did the review process raise the need to analyze trends in social spending?||Algeria, Bulgaria, Jordan||Algeria, Bulgaria, Egypt, Jordan, Pakistan, Tanzania||Algeria, Bulgaria, Ecuador, Egypt, Pakistan, Peru, Tanzania|
|Any proposals for staff to identify how social spending could be protected?||Algeria, Egypt, Jordan, Venezuela||Ecuador, Jordan, Pakistan, Philippines, Venezuela||Algeria, Ecuador, Egypt, Pakistan, Peru, Philippines, Romania, Venezuela|
The area with the most coverage (over the program life cycle) was the discussion of risks to achieving the fiscal targets. This surfaced in only 4 program cases (Costa Rica, Egypt, Pakistan, and Romania), at the time of reviewing the initial brief, but over the life of the program coverage expanded to 13 of the 15 countries, and comments were often extensive. Other areas that were well covered (12 programs) include concerns on fiscal sustainability and suggestions to discuss the factors behind the magnitude and pace of the fiscal adjustment. While initial focus on fiscal sustainability was limited to only 4 programs, reviewers pressed on the rationale for the magnitude and pace of adjustment in 9 of the 15 briefs.
A potentially important feature emerging from the evaluation is the phenomenon of a larger concentration of comments in the later stage reviews when they can only affect mid-course correction rather than initial program design. For all three topics (macroeconomic and fiscal targets, program design, and equity and social spending) there were more comments during the program review phase than at the initial briefing stage. Yet, it is in the earlier stages that comments could have the most influence on the nature of the adjustment process and the biggest impact on the fiscal adjustment strategy.
This finding is particularly surprising since many of the later comments relate to basic design issues, such as risks to achieving the fiscal target, fiscal sustainability, the size of the deficit, optimism of growth projections, and the importance of embedding the adjustment path in a medium-term outlook. However, for many of the programs, these issues are either not raised at all or only lightly touched upon at the stage of the initial negotiation brief.
Half the programs received no substantive comments on strategic elements of design in the initial negotiating brief. For the other half, there was an equal split between those with substantive and detailed comments (Bulgaria, Ecuador, Pakistan, and the Philippines) and those where reviewers touched on design issues more lightly (Egypt, Jordan, Peru, and Venezuela). It may be significant that 3 of the 4 programs with stronger comments on program design is- ministration;sues were follow-up programs (Bulgaria, Pakistan, and the Philippines). Comments dealing with strategic issues at the stage of the negotiating brief concentrated on the need to explain the factors behind the proposed pace, magnitude, and composition of the fiscal adjustment (9 out of 15 programs). Suggestions for stronger action that would demonstrate political commitment to the program receive the next amount of attention at this early stage but were limited to only one-third of programs. Other issues were raised in less than one-third of the programs.
The range of issues covered under program reviews (as the program was being implemented) became much wider. Concerns on fiscal sustainability were expressed in 12 programs. In at least half the programs, comments were made in areas such as pace, magnitude, and composition of adjustment; risks to achieving fiscal targets; actions that demonstrate political commitment; and considering past implementation problems in the design of reforms. For illustrative purposes, we provide some specific examples of the issues raised in these areas.
Test of political commitment
Many specific comments were raised in this area at the review stage. Common themes are the need to roll back exemptions, limit tax incentives, take concrete action against tax evasion, and enforce collection of tax arrears. However, in only three cases (Bulgaria, Jordan, and Peru) did reviewers press for action in these specific items in commenting on the program negotiation brief.
Some of the specific issues raised during the reviews could have been suggested with greater effectiveness at the outset of the program. These include, for example, the removal of tax preferences from the Foreign Investment Act in Bulgaria; concrete suggestions to reduce tax evasion in Ecuador; elimination in Pakistan of exemptions identified by the Committee on Reform of the Income Tax Ordinance; enforcing collection of the largest tax arrears in Peru; bringing forward the rationalization of fiscal incentives in the Philippines; harmonizing Tanzanian income tax relief and investment incentives between the mainland and Zanzibar; and reaching understandings on enforcing bankruptcy laws on major delinquent tax payers in Ukraine. Had these suggestion surfaced at the outset of the program they might have provided an early test of the commitment of the authorities to reform and contributed to a stronger program design.
Over the life cycle of the program reviewers often pressed for unbundling the reasons for poor tax administration; suggested addressing basic issues of sequencing between reducing some taxes and enlarging the tax base; and reforming income taxes to improve equity as well as effectiveness. Most common were suggestions for reforms to improve the equity of taxation. Comments in this area extended to nine programs. However, coming as they did during the review phase, these comments were perhaps less useful.
The review process raised many questions in critical areas also identified as weak by this evaluation (Chapter 7). However, reviewers ended up in a reactive mode, commenting more extensively as programs proceeded, rather than strategically at the outset, when they could have had the biggest impact to improve program design. This is puzzling since later comments are rich and focus on a wide array of critical design issues. Areas where comments could ideally come at the negotiation phase include (1) the need to link the fiscal adjustment with the recovery of private sector activity and growth; (2) specific actions that would demonstrate the commitment of the authorities to the program, particularly in removing exemptions and taking more forceful actions to reduce tax evasion; (3) consideration of past implementation problems in the design of the program; and (4) measures to improve equity and protect social spending.
The tendency for comments to come late in the process can be explained by reference to several important factors:
Reviewing departments seem to provide latitude to mission chiefs in developing the details of the program, particularly when projected outcomes appear ambitious. Subsequent comments reflect the fact that developments are less favorable than originally expected.
Program briefs basically address the immediate requirements of the economy for the early stage of adjustment, while other complementary policies are needed only for later stages. Reviews would then follow up on this sequencing of measures, with the functional departments possibly taking the lead.
There are time constraints associated with the preparation of prenegotiation briefs, which make it difficult to achieve more substantive comments at the outset.
These are important considerations but there is need for a concerted effort to ensure more brainstorming and pooling of potential review resources at an earlier stage when it can have a stronger effect on program design. This could be helped by having the brief articulate more clearly the basis for the fiscal program, including a medium-term fiscal scenario incorporating a basic debt sustainability discussion. This sequence would also be more consistent with the greater emphasis on domestic ownership, since it would involve a fuller exploration of alternative policy options, and potential trade-offs, at the initial design stage. The new guidelines on the review process issued by the First Deputy Managing Director on March 30, 2003 call for early consultations across departments in order to form a common understanding of the issues that are subsequently addressed in the papers. Such earlier consultations would help improve program design.
This analysis excludes two ESAF/PRGF arrangements as the fiscal adjustment in these programs was part of a longer-term development strategy rather than a response to shorter-term balance of payments problems as is more typical in SBA/EFF arrangements. The majority of the 13 SBA/EFF programs had a programmed fiscal adjustment above the average of the large sample.
Other qualitative issues of the fiscal adjustment, such as its impact on the equity, efficiency, and sustainability of public finance, are dealt with in other parts of the report.
Program objectives are often described in very general terms such as “restoring and sustaining a high rate of economic growth,” “alleviating inflationary pressures,” or “re-establishing balance of payments viability over the medium term.” What is usually missing, however, is a discussion of why these issues require the involvement of the IMF and the disbursement of resources in each particular case.
Egypt is not included because of lack of data on domestic public debt during the relevant period.
Staff reports are increasingly assessing debt sustainability following the framework outlined in IMF (2002e).
For 3 out of the 15 cases (Senegal, Ukraine, and Uruguay) there were only a few comments in the fiscal area and these were concentrated on the review phase. There were comments dealing with other areas than those we focus on here. For example, in the case of the Senegal brief, comments focused on energy subsidies, crop credit, and HIPC-related issues. Moreover, FAD agreed with the 1998 country strategy paper and therefore saw no need for major comments on the negotiation brief. For the Ukraine brief, reviewers emphasized the ramifications of implementing measures by decree, the slow progress with overall reform efforts, and the implications of unbudgeted payments to the coal sector. Furthermore, the 1998 Ukraine EFF was negotiated over a long period of time with several comments relating to the negotiation and design of the program having been communicated before the negotiating mission brief was prepared. As a result, there was less need to comment on the initial brief. FAD involvement in the Uruguay program became more important at later stages when the economic crisis intensified and a follow-up SBA was requested.