Prepared by the Policy and Review Department
Approved by Thomas Leddy
March 2, 1998
2. The objective in this note is not to address every issue on which the Fund staff does not see eye to eye with the evaluators: there are many points on which our judgement or even understanding of the facts differs from that of the reviewers, but these are not always essential to the main conclusions drawn. Some of the country-specific issues on which we had views are presented in an annex. Here, we will raise some broad questions on which the Fund staff had comments on important implications from the report.
4. We are concerned, however, that the suggested framework for evaluating the impact of programs on social expenditures may not be feasible. We fully agree that it would be desirable to break down prospective changes in social spending into the four components the evaluators suggest; but this is unlikely to be possible in many, if not most, countries. It is rare to find a classification of wage and non-wage costs in social spending or to find indexes for these separate costs. Even aggregate social spending data are unavailable for many countries. Nonetheless, the proposed breakdown provides a useful conceptual framework and standard to aim for in the future, and should provide further encouragement to the significant improvements in data that would be required.
6. We are, however, puzzled by the suggestion that the overlap between macroeconomic concerns of the Fund and microeconomic concerns of the Bank can be limited to fiscal issues. In fact, the report itself is filled with implicit acknowledgments that the overlap in legitimate Bank and Fund interests goes beyond fiscal issues. For example, the Fund is criticized for focussing solely on the effect of privatization on the budget at the expense of the efficiency implications of various forms of privatization: presumably the Fund is being asked to ensure that microeconomic efficiency considerations are better incorporated into program design. Similarly, the proposal that the Bank provide an assessment of the impact of programs on incomes leads logically to the conclusion that the Fund should concern itself with the implications of macroeconomic policies for social efficiency. In short, macroeconomic and microeconomic concerns are inextricably interactive in areas far beyond narrowly-defined fiscal policy. Thus, while the Fund’s concentration on the former and the Bank’s on the latter should be maintained, collaboration on an array of issues will remain a critical requirement. This needs to be taken into account when considering ways to enhance collaboration between the Bank and the Fund.
8. The report recommends a "taper-in" scheme for Fund support of ESAF countries. In such a scheme, the Fund would gradually increase ("taper-in") its financial support as reform proceeds. As an economy achieves stabilization--by which it seems the evaluators mean low inflation and current account convertibility--the Fund would shift its focus away from advice and toward more financial support. It is hard to see how such a scheme would be consistent with the objectives of the ESAF.
9. Financial support through the ESAF is typically critical during the early stages of adjustment and reform when a major policy effort and large exceptional financing are needed. In later stages, even if inflation is brought down, countries typically still have seriously flawed policies: fiscal deficits may be unsustainable or structural policy weaknesses may have the potential to derail still fragile macroeconomic balances. The ESAF was established precisely because many low-income countries needed a policy framework and support geared not just toward fixing macroeconomic policy shortcomings, but also toward addressing, over the medium term, the structural problems that underpin macroeconomic policy weaknesses. Continuing Fund support could be justified only if the government is committed to concrete measures addressing these serious problems. In other words, it is not easy to envisage circumstances in which the Fund could provide significant financial support to a "post-stabilization" country that aimed at maintaining "a few key features of macroeconomic policy" rather than at improvements in other policies. In the longer term, when a country is in a position simply to maintain policies, bilateral or market financing should be adequate, and Fund involvement should be through annual Article IV consultations and, where relevant, technical assistance.
11. The report questions whether certain aspects of the Fund’s operating style and methods inhibit ownership. Broadly, the report suggests that Fund staff are too secretive, unduly restrictive in their range of contacts, and inflexible in their negotiating positions. These observations are not unfamiliar, and we have been working to deal with them, particularly by increasing contacts of Fund staff with labor unions, civil society and opposition political parties. At the same time, we are cognizant of the fact that the operating style of Fund missions is always a delicate balance. On openness, missions must balance the sensitivities of the government and of the issues at stake against the constructive role the mission can play in interacting with various interest groups. Similarly, while there must be scope for flexibility in negotiations, there are limits dictated by the need to assure that problems are addressed adequately. We must keep these balances under review and seek new ways to deal with them.
12. The report amply demonstrates the tension that can arise between ownership and the need for adequate measures to address economic weaknesses. It is hardly surprising, however, that the report has evoked some frustration within the staff: for example, receiving criticism at once for acquiescing to programs that are "second best" and for inflexibility in negotiating positions defies the difficult balance that every mission faces. For example, the report is critical about several instances of Fund support for poorly sequenced economic reforms that resulted in "second-best" programs. Yet, in at least one instance, it is subsequently stated that the program originated with the government while the role of the Fund in design was limited. In several other instances, poor sequencing can clearly be traced to episodes where the Fund acquiesced to firm resistance of a government to the Fund’s views on the urgency of a particular reform.1 Cases like these lead to two important conclusions: first, actual decisions on providing support to a program often come up against trade offs between the depth of a government’s ownership and what is needed to secure economic objectives; and second, ownership is an important but not sufficient condition for a successful program.
14. The second proposal--ex post conditionality--raises more questions. Seen in one light, the proposal differs little from conditionality as now practiced, particularly when countries commit to undertake significant prior actions or to perform well under a staff-monitored program before drawing from the Fund. If, however, the proposal is to be interpreted as the Fund providing financing in situations where the criteria for support are not well-defined and the path of policies are not specified in advance, the practicality of ex post conditionality would seem more doubtful. Specifically, any commitment of Fund resources needs some provision whereby assistance, once flowing, would be halted in the event of a serious breakdown in policy implementation. Moreover, it is important for countries to know in advance the circumstances under which committed financing will remain available. The specification, ex ante, of conditions on the instruments of policy (as opposed to outcomes) is intended, inter alia, to make clear the actions a country must take to insure the continued flow of Fund financing. Anything less would introduce disruptive uncertainty into the outlook for financing.
16. The report is right to warn that debt ratios should not be used for static analysis only. In Fund work, viability thresholds have been established for use as rules of thumb, but the focus is usually on the evolution of the ratios over time. Moreover, ESAF-supported programs generally include a detailed medium-term scenario in which assumptions about investment and savings decisions several years into the future are presented.
20. Many of these initiatives will place greater demands on staff resources. In particular, aims such as increasing the number and staffing of resident representative offices, allowing more time for missions, providing more technical assistance and coordinating more closely with the Bank, to name a few, are worthwhile but will have significant implications for staffing of the desks and departments working on ESAF countries. Acting on many of the recommendations will involve careful consideration of these issues.
ANNEX
2. In spite of the polarization among the political parties, Bangladesh has made significant progress toward a functioning parliamentary democracy—most recently in 1996, when a constitutional amendment reinforced the framework for fair parliamentary elections. In view of the current government’s parliamentary majority, progress in structural reform is not obstructed so much by a lack of consensus among political parties (the main opposition parties adopted many of the same reforms under the SAF and ESAF while they were in power in the late 1980s and early 1990s) as by strong vested interests that may have undue influence (e.g., civil servants, labor unions, major loan defaulters, and import-competing industrialists). These problems could probably be tackled within the existing parliamentary institutions. Moreover, Bangladesh has a remarkably free press and an abundance of public seminars and workshops where the main policy issues are actively debated. Such institutions should facilitate the task of building a consensus. In this context, the support for reform from Bangladesh’s impressive NGO community and the growing private sector will become increasingly important.
4. We wonder if the evaluators’ perception that recent governments did not make sufficient efforts to seek consensus could reflect a focus in the interviews more on members of the government that took office in August 1997 than on representatives of earlier governments. By not reporting on the significant extent to which successive governments (in particular the one that had left office just before the interviews were conducted) worked within the political system to present feasible reforms, the report missed an opportunity to provide insights into how to promote ownership.
5. Emphasis is placed in the report on the mostly negative image of the Fund which is said to be reflected in sensational treatment in the news media. This is, of course, a subjective issue, but the distinct impression of many staff missions to Bolivia during the past five years is that there has been generally low-key treatment by the press of discussions between the government and the Fund.
7. As regards ownership and the role of the PFP, the report notes that "...the ESAF program...benefited from a reasonable measure of national ownership, especially judged from the apparent commitment of the top levels of political leadership, and from the degree of government involvement in the preparation of policy documents, particularly the Policy Framework Paper (PFP)". This statement accurately describes the process related to Côte d’Ivoire’s most recent ESAF arrangement, the negotiations for which were based on a draft PFP drawn up by the authorities. We were surprised, however, by the seemingly contradictory statement, three pages later, that Fund staff customarily drafted PFPs and that the exceptions were "occasional" and "grudging."
8. A final but important factual point: we were surprised by the claim that the most politically difficult reform issues "...had been largely exhausted by the time the ESAF program began". To use the example cited in the report, even now the authorities are continuing efforts to downsize the civil service and are encountering considerable political resistance.
10. We agree that the timing of the 1994 program, drawn up in the midst of chaotic macroeconomic conditions, was not conducive to the orderly building of consensus. This is often the unfortunate by-product of crisis. Indeed, we are not sure how to react to the statement that "a number of donors....expressed concern that the program had been imposed on an inexperienced and ill-prepared government, without due regard for the critical economic and political conditions it faced." While it would have been ideal to have had time to build a stronger consensus, the situation in 1994/95 clearly indicated that delaying measures (and in the interim increasing their ownership) would also have increased the risks of a major confidence crisis.
11. The observation that the reform effort was spearheaded by one minister is accurate. However, it is important to recognize that Fund staff held extensive discussions on various sectoral issues, with the responsible ministers and other high-ranking officials who implemented the agreed reforms. Evidence of this close involvement was the joint preparation of the policy framework paper (PFP) and the meetings with the Cabinet Committee on the Economy (CCE). The initial draft of the PFP was done by Bank and Fund staff and was distributed to all concerned ministries about a month before it was discussed. In these discussions—which were comprehensive and normally lasted for 12-15 hours—all ministries provided inputs, including sectoral policy proposals and commitments, and exchanged views with their colleagues about the sequencing and consistency of reforms. During meetings with the CCE, staff usually sought the commitment of the committee to the policies described in the Memorandum of Economic and Financial Policies.
13. The report takes the position that Uganda’s program for 1994-97 could have focussed entirely on structural reforms and implies that the fiscal stance under the program could have been eased, because inflation had been sharply reduced. In our view, there would have been serious risks in such a strategy. The 1994-97 program sought to consolidate the stabilization already achieved through appropriate fiscal discipline. The underlying aim was to safeguard a stable economic environment that was credible and attractive to private investors and to support the supply response to further structural reforms. Greater access to bank credit for the private sector was likely to bring about more durable growth than an easing of the fiscal program.
14. Relatedly, the report sees the medium-term fiscal profile under the 1994-97 program as too restrictive and implies that higher aid inflows should have been factored in. This would have allowed more spending on social and capital outlays and/or lower taxes. Why was this not done? First, the program was based on reliable projections of aid inflows--in our view higher estimates would have been unrealistic: this was born out by the experience, as shortfalls in external import support from programmed levels were registered during much of this period. Second, in the longer-term, a strategy based on continued high aid dependency would also be unrealistic. Third, in practice it has been difficult to control outlays on unproductive expenditures at the expense of social and capital outlays.
15. The report rightly criticizes the structure of taxation (a heavy dependence on excise taxes on petroleum and a few other products). However, it was generally accepted that revenues needed to be raised from an unusually low base, and with weaknesses in the administration of direct taxes, the burden fell on excises. Also, we would note that there was a considerable reduction in the reliance on import taxes: maximum import tariff rates were reduced from 50 to 20 percent.
16. We were surprised by two statements: (page 142) that "the exchange rate is seldom even discussed by Fund missions," and (page 145) that "disagreement between the Fund and the government persists, notably the areas of privatization, civil service reforms, and defense expenditure cuts."
18. The report implies that the authorities’ loss of commitment to reform stems at least in part from the process of negotiating annual programs with the Fund. In our view, the situation is more complex. The crux of reforms now needed involves actions that would reduce the extent to which the state controls the economy. These are difficult issues for any government and require significant departures from past approaches to economic policy. We would be surprised if there is any perception of a change in the speed of or flexibility in the negotiations, although the content certainly has shifted to difficult terrain.
19. We were surprised by the statement suggesting difficulties in the working relationship between the Bank and the Fund. Collaboration has been close and effective, especially after 1989/90 when the Fund increased its involvement in Vietnam through technical assistance, the posting of a resident representative, and eventually ESAF support.
21. With respect to the macroeconomic reforms during 1992-93, it is clear that the enormous swings in policy performance resulted in erratic developments in the main macroeconomic variables. However, as the evaluators note, the budget was brought under control in 1993 with the introduction of a cash budget, and inflation dropped sharply within just a few months. The liberalization of the economy was a central element of the program and provided credibility to the adjustment efforts. Thus, the need for liberalizing the economy, which had been stifled by controls and shortages, was so great that the potential short-run costs were probably unavoidable.
22. On the structural reforms, we fully support the view that a much earlier resolution of the problems in the copper parastatal was needed. In the event, the Fund was faced with the dilemma of how to deal with a second best situation: should Fund support be halted over the failure to address these problems or should support continue while allowing flexibility in the timing of highly sensitive parastatal reform?
23. In an economy that had experienced a massive impoverishment over the years, it was unavoidable, given limited resources, that social safety nets in the programs could only partially deal with the enormous problems. Nevertheless, we agree with the evaluators that the failure of the government to control wage costs as called for in the program significantly ate into outlays for nonwage spending, especially in the social sector.
25. First, it is argued that the program failed to take account of the distributional effects of fiscal adjustment and was therefore doomed to fail from the start. The program envisaged a cut in noninterest spending relative to GDP by 7.5 percentage points during 1992-95; the evaluators call this an "astonishing contraction," which would have required that "social and redistributional expenditures be massively curtailed." In fact, no such curtailment in expenditures was envisaged or required. The programmed reduction in spending was to consist of: (i) no further need for the drought-related subsidies of over 4 percent of GDP in the previous year; (ii) reduced net lending to the public enterprises relative to GDP of 3 percentage points, of which 1 percentage point was to stem from privatization proceeds; and (iii) a reduction in the civil service wage bill relative to GDP of 2 percentage points. Other expenditures were programmed to rise relative to GDP. In the event, however, the government failed to bring the financial performance of the public enterprises under control, and spending cuts were made in "softer" areas of the budget. But this certainly was not the intention.
26. Second, the report suggests that Zimbabwe possessed many of the characteristics of a transition economy and that the decline in manufacturing output of 14 percent during 1991-96 should thus have been "broadly predictable" and allowed for in the program. While the Zimbabwean economy was certainly highly protected and subject to a range of controls prior to inception of the program, it was for the most part in private hands (the public enterprise sector probably accounted for no more than 5-10 percent of GDP). Moreover, it is not clear that output would have contracted as it did if the fiscal program had remained on track: fiscal slippages contributed to increases in real interest rates and in the real exchange rate that significantly hurt manufacturing output and real wages. The problem was not liberalization per se, but rather the fact that it was accompanied by widening macroeconomic imbalances that were policy-induced.