Information about Asia and the Pacific Asia y el Pacífico
IEO Evaluation of Exchange Rate Policy

Staff Response to the Evaluation of the Independent Evaluation Office of IMF Exchange Rate Policy Advice, 1999–2005

International Monetary Fund. Independent Evaluation Office
Published Date:
August 2007
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

Key Points

  • This report provides a wealth of information that is helpful to appraise the Fund’s performance in the critical area of exchange rate policy advice.
  • The IEO’s report, however, disregards much of its own positive evidence on the quality of the Fund’s work, ignores the progress made during the sample period, and is sometimes premised on unrealistic expectations of the Fund’s role. All in all, the report’s consistently negative tone crowds out much valuable information and some useful conclusions.
  • An objective analysis of the evidence gathered by the IEO leads to more comforting conclusions: significant improvement in the treatment of exchange rate issues over time; coverage of several exchange rate issues that is adequate in most cases; and generally good integration between exchange rate and other policy areas. Moreover, country authorities across the membership seem satisfied with the policy dialogue with staff and with the way the Fund plays most of its roles. And the Fund’s work on exchange rate policy decisions has some impact.
  • In other areas, the IEO’s findings confirm earlier staff assessments that further improvements are needed, with priority to be assigned to improving assessments of exchange rate levels, and of spillovers. Data differences should also be addressed.
  • A number of initiatives under the aegis of the Medium-Term Strategy are well under way to address improvement needs and consolidate progress.

1. Exchange rate policy advice is at the core of the Fund’s work, and an area where the Fund has long been aware of special challenges. Indeed, improving exchange rate surveillance was identified as a priority objective by both the 2004 Biennial Surveillance Review (BSR) and the Managing Director’s Medium-Term Strategy (MTS). Considerable efforts are under way in this area.

2. Thus, the Independent Evaluation Office’s (IEO) report provides a welcome opportunity to consider in more depth the quality of the Fund’s work in this area, and to assess whether the initiatives under way are likely to achieve their goals.

3. While it contains a wealth of helpful information, the report unfortunately does not portray accurately either the Fund’s past or its current work in the exchange rate area. First, the report’s conclusions disregard much of its own positive evidence on the quality of the work of staff, management, and the Executive Board, while magnifying any perceived shortcoming. Second, the report is outdated. In part, this is because the sample period (1999–2005) is inevitably some way in the past, but it is unfortunate that the report did not focus on what its own data reveal about trends during the sample period. Moreover, further progress has been made since 2005. Finally, the report is sometimes premised on unrealistic expectations of what the Fund can reasonably achieve, both in terms of its output and its impact. All in all, the report’s consistently negative tone crowds out much valuable information and some useful conclusions.

4. The report’s focus on shortcomings is understandable given the IEO’s brief to help the Fund strengthen its performance, but a broader perspective is needed as a basis for action. There is no doubt that, in spite of significant progress, we need to do more. The challenge, however, is to rank the issues in order of seriousness, as a basis for the Board’s judgments on what more is needed.

The Record

5. We will first try to provide a more balanced presentation of the evidence included in the report, before turning to its recommendations. We focus on: (1) the quality of the Fund’s analysis and advice; (2) its interaction with member countries; and (3) its overall role and impact on policy decisions. Most of the evidence comes from the report itself, including the in-depth review of 30 countries (henceforth “IEO in-depth review”), the review of staff reports for the entire membership (“IEO SR review”), and responses to the IEO questionnaires. Particular attention is paid to assessing quality in more recent work, as the 2004 BSR prompted a wave of new initiatives in the exchange rate area.

Quality of analysis and advice

6. For most of the quality dimensions of exchange rate analysis and advice, results are quite comforting, although there is also room for improvement. We focus broadly on the same quality dimensions identified in the report (see paragraph 19 of the report).1

7. Coverage of exchange rate issues. This seems to have been adequate in the vast majority of cases, particularly in recent years. The IEO in-depth review finds only five cases (out of 30 countries over seven years) where certain exchange rate issues had not been covered for part of the period. Only one of these cases is recent (Saudi Arabia).2 More generally, the amount of analytical work produced by the Fund on exchange rate issues is extensive, and has increased further recently. During 2001–05, the Fund issued annually over 30 Working Papers on exchange rate issues (34 in 2006 and 6 in just the first two months of 2007), typically drawn from selected issues chapters prepared for Article IV consultations.

8. Integration between exchange rate and other policy areas (excluding spillovers, see below). Generally good. IEO questionnaire responses show that “coverage of linkages [between exchange rate issues and other policy areas] in discussions was good overall” (paragraph 22). The IEO in-depth review also finds that integration was good with respect to monetary, fiscal policies, structural policies, and that integration of financial sector and financial stability issues, while “somewhat lacking,” has improved over time (paragraph 11, BD 5).

9. Description of de facto regimes. Generally accurate. The IEO in-depth review finds only three cases (out of 30 countries for seven years) in which the regime description is regarded as inappropriate (staff disagrees on two of these cases, see Annex II). Discrepancies between the description in staff reports and the Monetary and Capital Markets Department’s (MCM) classification—a possible indicator of problems—are also rare (6 percent of cases), and often explained by inevitable lags in the revision of the MCM classification.

10. Intervention policies. The report distinguished several dimensions:

  • Overall coverage. Generally adequate, but there is room for further improvement. The IEO in-depth review finds only five cases (out of 30 economies for seven years) in which the coverage of intervention policies was incomplete (paragraph 24). In three of these cases staff does not agree with this assessment (Annex III). None of the five cases refers to the post-2004 BSR period. This said, the Fund should aim at the highest standard in assessing intervention policies, and staff’s own analysis also points to room for improvement.3
  • Coverage of intervention tactics. Limited, but this should be expected. Most staff reports for surveillance and program work focus on the macroeconomic aspects of intervention, rather than the specifics of how intervention is implemented. This is to be expected given the nature of these reports. More technical aspects are dealt with in the provision of technical assistance to countries that require it. 4
  • Assessments of the adequacy of reserves. A difficult area for the economics profession as a whole, but staff has generally used available techniques appropriately. Staff has often conducted in-depth analyses of the adequacy of reserves (Table 1.1 of the report lists 15 of them), and in cases of extensive reserve accumulation has generally taken an adequate country-by-country approach (Box 3.2 of the report). Various kinds of indicators are routinely used in staff reports to assess the adequacy of reserves. We should be under no illusion that “highly judgmental” assessment can be avoided in this difficult area (paragraph 25), but staff is at the forefront of related research,5 and training in this area is regularly provided to Fund economists.6
  • Staff position on intervention. No evidence it was inappropriate. It is true that staff has generally supported reserve accumulation for precautionary purposes, but not for purposes of maintaining competitiveness. Indeed, intervention for the purpose of maintaining competitiveness may contravene the spirit, and possibly the letter, of Article IV whenever it aims at keeping the exchange rate artificially undervalued, a strong reason why this kind of intervention should not be supported. At the same time, staff has typically been open-minded on the possibility of using sterilized intervention to respond to capital inflows. There is no a priori assumption that sterilized intervention is always ineffective (as argued in the report, paragraph 26), although, the cost of intervention is often regarded as problematic.

11. Data availability. This is an area for improve-ment. The report suggests that in 37 percent of countries data problems impaired the staff’s ability to conduct exchange rate analysis and provide related advice. This partly reflects capacity constraints, which can be addressed through technical assistance. More worrisome are the fairly frequent cases when data shortcomings appear to reflect the authorities’ unwillingness to share them.7

12. Assessment and analysis of the exchange rate levels. Assessments are regularly provided but in several cases the quality of the analysis should be improved. The IEO in-depth review noted that the “sophistication of exchange rate level assessments, as indicated by the use of empirical methods, was good overall” (BD 5, paragraph 20). But it also finds cases in which the treatment was inadequate. The staff’s Stocktaking Paper found that in one-third of its sample the analysis of exchange rate levels was not sufficiently sophisticated. There are, however, signs of recent improvement. This includes, inter alia, the recent broadening to emerging markets of the exchange rate assessments conducted by the Consultative Group on Exchange Rates (CGER) and improved analysis in several reports (for example, China, Colombia, Mexico, Malaysia, Mauritius, CEMAC, and WAEMU).8

13. Assessments of countries’ exchange rate regimes. These are “a standard feature of Article IV reports” (paragraph 31). In addition:

  • Quality of the analysis. Usually adequate—when closely scrutinized. The broader—less in-depth—IEO SR review suggests that the suitability of the regime was not assessed in detail in many cases. However, as the report notes, it is often difficult in practice to separate the assessment of regimes from the assessment of exchange rate levels, when the latter results in findings of misalignments. In those cases, staff’s call for increased flexibility may have been tantamount—admittedly with less than complete candor—to a call for an appreciation. Moreover, an in-depth discussion of pros and cons of a regime change may be less warranted whenever staff is simply calling for the authorities to implement de facto their de jure regime. On this account, only a close review of staff reports can lead to the conclusion that shortcomings were present. It is thus reassuring that the IEO in-depth review regarded the regime analysis as inadequate in only three cases.9 In all these cases more recent reports have addressed these problems.10 The report also notes that staff advice in recent years has leaned toward increased exchange rate flexibility. Conjunctural factors—as well as increased capital mobility—probably played a large role here. First, inflation has been relatively low, so a key reason to peg (the need to lower inflation expectations) has not been present. Second, many countries with inflexible exchange rate policies were facing large balance of payments surpluses: as noted, in these countries the call for more flexibility was often tantamount to a call for an appreciation.
  • Excessive urgency in advocating regime changes. The report cites only two cases.
  • Implementation issues. A sizable minority of members believes that attention to implementation issues could be improved, but this need can be addressed more through technical assistance than better surveillance. As noted above, detailed implementation issues are hard to address in the context of surveillance work, which has a macroeconomic focus. They can, however, be addressed through technical assistance. What is perhaps needed is to promote technical assistance more actively.

14. Spillovers. Consistent with the MTS, this is an area for improvement. The authorities’ responses to the IEO questionnaire show that over 60 percent of the respondents from large emerging markets and over 80 percent of respondents from other country groups perceive that global and regional spillovers affecting their exchange rate developments had been identified in staff reports. Moreover, multilateral surveillance has paid considerable attention to these issues, as shown by the lengthy—and yet incomplete—list of WEO features on exchange-rate-related issues (Table 3.5 of the report). This said, we would agree with the IEO that integration of spillovers in surveillance is still insufficient, as confirmed by the IEO in-depth review.

15. Overall improvement. There is clear evidence of significant improvement in the treatment of exchange rate issues during 1999–2005. Over 55 percent of authorities surveyed saw improvement, with almost no one seeing a deterioration. Results are less favorable for large emerging market countries—perhaps reflecting the fact that, in the most recent period, the Fund often took views on their exchange rate policies that were not fully shared by the authorities. However, even among these countries, over one-third reported improvements, with almost no one signaling a deterioration.

Interaction between the Fund and member countries

16. Policy dialogue. Country authorities seem to be generally satisfied. Interviews revealed some cases of dissatisfaction. But 75–90 percent of the authorities responding to the IEO questionnaire were satisfied with their interactions with staff, and thought discussions: were substantive and two way; their frequency was appropriate; had the right balance between informality, confidentiality, and requirements of reporting to the Executive Board; and were fully reflected in documents subsequently sent to the Fund Executive Board. They also agreed that staff approached the discussions with candor and in a respectful and open-minded way; and were willing to raise politically sensitive issues.

17. Evenhandedness. The IEO in-depth review finds “no clear cut cases of uneven treatment,” nor does the IEO SR review find a pattern that would question evenhandedness.11 On the contrary, the positive results mentioned above regarding policy dialogues emerge across all country groups.

The overall role of the Fund and its impact on policy decisions

18. How members perceive how the Fund plays its various roles. Satisfactory results, in most respects. Based on the responses to the authorities’ questionnaire, the Fund is judged to have played the roles of confidential advisor and sounding board “about right” by about two-thirds of the members, and even in the large emerging market group this proportion exceeds 50 percent (once those answering “don’t know” are excluded). High levels of appreciation are also found for the Fund as provider of credibility, lender, and, albeit to a lesser extent, consensus builder. The results for the roles of “ruthless truth-teller” and “broker for international policy coordination” are not as positive. The role of the Fund with respect to the latter is being enhanced through the introduction of multilateral consultations.

19. Impact. There is evidence of impact, although the authorities’ responses to the IEO questionnaire show that 79 percent of countries who took major exchange rate policy decisions considered the Fund’s assessment instrumental or “helpful at the margin” in shaping their decisions (based on Table A6.1, BD 6). The percentage rises to 90 percent when the few countries that had little or no discussions on these decisions with the Fund are excluded. Even in advanced and large emerging market countries, when there were substantive discussions with the Fund, 74 percent found the Fund’s assessment to be either instrumental or “helpful at the margin.” These numbers are reasonably high, especially considering that absence of impact may simply reflect the absence of an ex ante divergence of views.

Key Findings and Recommendations

20. The above discussion highlights both areas of strength and areas for improvement. Among the priorities seem to be: (1) strengthening further the analytical discussions backing up views on exchange rate levels; and (2) improving the treatment of spillover issues. Improving data availability (e.g., regarding intervention) is also important. Finally, there is also a need to consolidate the progress made in other areas, and reduce remaining shortcomings. Many initiatives—which partly overlap with the report’s recommendations—are already under way and are discussed below (see also Table 1). New initiatives could of course be considered and costed. But, before doing this, it seems appropriate first to assess whether the current efforts have borne fruit. The 2008 Surveillance Review will do this (improving the treatment of exchange rate issues is one of the monitorable objectives to be reviewed).

21. In presenting our conclusions, we will also comment on the recommendations offered by the report. We therefore follow the sequential presentation followed in the latter.

Rules of the game and guidance to staff

22. Reforms to the surveillance framework are critical in consolidating progress and dealing with remaining problems. A revision of the 1977 Surveillance Decision to bring it in line with best practice could also help raise the average practice of surveillance. The proposed revision would underscore the importance of the analysis of exchange rate levels and of spillovers, which the above discussion highlighted as priority areas. While the above discussion did not suggest major problems regarding the focus of exchange rate surveillance and evenhandedness, any remaining weaknesses need to be addressed. A revised Decision would also help consolidate progress and foster focus and evenhandedness in surveillance overall, not just in exchange rate policy analysis. We therefore concur with the recommendation in paragraph 58 on the need for “a revalidation of the fundamental purpose of surveillance.”

23. In contrast, we do not think that the definition of “practical policy guidance” (paragraph 59) is a priority. The current surveillance guidance note does deal with exchange rate issues in fairly general terms, but we regard this as appropriate, at least for the moment. Given the “lack of professional consensus” in this area (page 9), distilling summary prescriptions would be unrealistic. Indeed, such summary prescriptions would risk undermining, rather than promoting, analytically sound advice, given the variety of country-specific features, data availability problems, and the complexity of the issues in question. Thus, staff has preferred to follow an approach based on knowledge dissemination and incentive mechanisms. Close to 400 Fund economists a year participate in seminars on exchange rate issues, and 50–70 in related multiday training (with both numbers growing recently). In addition, new exchange rate workshops have been introduced since 2005 with the goal of motivating performance and better disseminating best practice. Finally, PDR has introduced an internal assessment system to sharpen the internal review of the quality of exchange rate surveillance. 12

24. With respect to the specific initiatives proposed by the IEO in paragraph 59:

  • A review of the stability of the system of exchange rate regimes and exchange rates: we do not see this as a priority at this stage. The WEO regularly assesses the consistency of major countries’ policy mixes, including their de facto exchange rate regimes. However, this issue may be reassessed in light of the description of trends in regimes presented by the forthcoming REARM.
  • A guidance note on analysis of intervention based on information provided by the authorities in Article IV consultations on their reserve goals: we do not see the casting of the current state of knowledge into a guidance note as a priority, but continued research into issues of reserve adequacy is needed. And while discussions of the authorities’ intentions regarding reserve accumulation should remain a feature of Article IV discussions, systematically seeking precise quantitative information from the authorities in this regard would be an undue imposition.

25. The effectiveness of staff’s dialogue with the authorities is crucial. The report does not present sufficient evidence to justify new initiatives in this area (paragraph 60). Judgments on the effectiveness of the dialogue are already a regular part of surveillance reviews and staff performance assessments. Individual cases of lack of forthrightness do point again to the long-acknowledged need for candor in all cases, and we agree that staff, management, the Executive Board, and member countries should work together to ensure that candor is always encouraged and rewarded.

Implementing existing policy guidance

26. The efforts to ensure an adequate description of de facto regimes in Article IV reports are bearing fruit. There is evidence that these descriptions are now adequate in the large majority of countries, and the completion of the work on the REARM should consolidate the progress made and facilitate addressing remaining shortcomings. Thus, new initiatives in this area (paragraph 66) do not seem to be warranted.

27. Providing strong analytical backing to advice on the choice of regimes remains critical, but does not call for new actions. The dissemination of best practice in this area will continue. But it seems excessive to require staff reports to lay out explicitly the forward-looking assumptions on which staff advice on the regime is based (paragraph 67). Regimes do not, and should not, change frequently, and in most cases the staff’s advice will not, and should not, do so either.

28. More analytical discussions of exchange rate levels are needed in many countries (paragraph 68). As noted, through 2005 shortcomings persisted in a sizable number of countries. We have provided examples of further progress since then, but more is needed. We would expect that the initiatives initiated since early 2005 will take some time before they are fully effective. Other initiatives are in progress (including work to assess better the elasticity of trade balances to exchange rate movements and a further broadening in CGER coverage). New steps may be needed and developments will be monitored closely by PDR. In particular, it may be useful to conduct more in-depth work on assessing equilibrium exchange rates for countries in which manufacturing exports are only a small fraction of the commodity (e.g., oil) exports.

29. Data problems continue to deserve attention (paragraph 69). To some extent they reflect capacity constraints in member countries. The report does not offer firm recommendations in this area, noting that there is a need for management and the Executive Board to consider further what lies behind the existing problems. The planned review of data provision to the Fund later this year will constitute an opportunity to do this.

30. Improving the treatment of spillover issues is a goal set by the MTS. As an implementation step, management has recently instructed staff that “Staff reports for systemically important countries…should include a substantive analysis and discussion of spillover issues, drawing the implications of the country’s developments, policies, and vulnerabilities both generally for the international community and specifically for neighboring countries and other affected groups. Whenever relevant, spillover issues should be discussed for other countries.” Staff has also proposed to clarify the importance of spillover issues in a revised Surveillance Decision. We regard these initiatives as adequate at this stage. The recommendation to set up a panel of senior officials who would offer questions to explore in this area (paragraph 70) seems gratuitous, when the Fund already benefits from the advice of capitals through various channels.

Management of work on exchange rates

31. We do not see a clear need to modify the current departmental responsibilities for exchange rate work (paragraph 71(1)). The report notes that these responsibilities are scattered across many departments, but this reflects the fact that exchange rate work permeates many of the Fund’s activities. The Surveillance Committee and the CGER already provide fora for developing Fund-wide perspectives on these issues, and their roles have been strengthened as a result of the MTS.

32. It is well understood that financial sector work needs to be integrated better into surveillance, and it may be worth considering the inclusion in surveillance teams, on limited occasions, of foreign exchange market practitioners (paragraph 72(2)). However, both resource constraints and the need for focus in surveillance dictate that we distinguish carefully between issues that it is essential for surveillance to cover, and issues that belong more to the realm of technical assistance.

Confidentiality and Executive Board oversight

33. We agree that there is need to examine carefully issues related to confidentiality vis-à-vis the Board. The report’s proposals (paragraph 74) involve complex governance issues: since it is the Board that is in charge of conducting surveillance, any arrangements along the proposed lines would need to ensure that the Board had the information necessary for it to fulfill this obligation to conduct surveillance. Similar issues have been discussed by the Ad Hoc Committee on Confidential Information (see FO/DIS/06/138, Revised), and a staff paper reflecting these discussions will be issued for Board consideration shortly.

Facilitating multilateral policy cordination

34. We agree on the importance for the Fund of strengthening opportunities “for potential multilateral concerted action,” based on “rigorous and compelling analysis of scenarios and [involving] a strategic plan to build consensus amongst key players” (paragraph 76). Multilateral consultations are an initiative contained in the MTS aimed at achieving that.

Annex I. Coverage of Exchange Rate Issues in China Reports

35. The report takes the view that in 2001–02 the coverage of exchange rate issues for China was inadequate—more specifically: “In the case of China (in 2001–02), substantive engagement with the authorities was lacking on the specifics of exchange rate regime options identified by IMF staff” (Chapter 3, footnote 3). The report also states that in assessing exchange rate levels, “some traditional indicators of exchange rate misalignment were not brought to bear on the issue through 2005, clouding the overall assessment of renminbi levels” (Chapter 3, footnote 16).

36. However, the 2001 staff report noted in specific terms the policy advice provided to the authorities (including on sequencing) and its rationale, and referred to the previous year’s report for additional details on modalities. The 2002 staff report did contain a short discussion of exchange rate issues, but the recommendations were backed up with a selected issues paper (“Exchange Rate Policy”), and a text box described the functioning of China’s foreign exchange market. All in all, every selected issues volume for China between 2002 and 2006 contained at least one chapter dealing with exchange rate issues (including level assessments, reasons for and modalities of moving to greater flexibility, and spillovers). The 2006 staff report and background papers include a thorough discussion of all relevant exchange rate issues, including the exchange rate policies de facto pursued by the Chinese authorities, the level of the exchange rate (using a wide range of applicable indicators and methodologies), the adequacy of the exchange rate regime, the implications of exchange rate policies for the Chinese economy, as well as the spillovers of those policies for the rest of the world.13 Finally, considerable TA on foreign exchange markets was also provided by the Fund (March and May 2001, June 2002, April and May 2004, and May 2005).

Annex II. Description of Exchange Rate Regimes

37. The report found three cases—not mentioned by name—in which “the staff’s classification of the de facto regime conveyed, at least temporarily, a misleading impression of the regime in place …” (paragraph 23). In staff’s view, the criticism is unjustified in two cases.

Country A

38. The 2003 staff report clearly stated that country A had increased its intervention activity and that part of the motivation was to maintain competitiveness. It then went on to urge the authorities to “maintain their flexible exchange rate policy.” The discussion was accompanied by both bilateral and multilateral exchange rate charts. Thus—although there was some uncertainty at the time as to the exact amounts of intervention—the staff report did give a broadly accurate description of the de facto regime. It is true that at the time the report was issued, the MFD/MCM classification, which had not yet been updated, was “independent float.” This discrepancy should have been avoided, but did not impair the description of the policies provided in the Article IV report.

Country B

39. Both the 2003 and 2004 Article IV staff reports were very clear that large-scale intervention was occurring, and staff recommended that these interventions could be increased and that sterilization should be avoided in order to provide additional liquidity to the economy. Both reports also discussed the likely effect of such intervention on the exchange rate, and that this effect was intended. It can be argued that the Fund relations annex was misleading in that it stated that “country B maintains a floating exchange rate,” although the existing criteria for classification as a floating regime do actually allow for this kind of intervention. In any event, the discussion in the text of the staff report made clear what the authorities were de facto doing.

Annex III. Cases Cited as Involving Incomplete Coverage of Intervention Policies

40. The report (paragraph 24) finds coverage of intervention policies to have been incomplete in five cases. In staff’s view, there is no basis for criticism in three of these cases.

41. Japan. It is not clear which staff reports are regarded as providing an incomplete coverage, and why. The 2003, 2004, and 2005 reports provided thorough assessments of intervention policies in Japan, leading, particularly in 2003, to extensive discussions at the Executive Board. The 2004 Article IV report agreed with the authorities that intervention in 2003 had prevented an undue tightening of monetary conditions and concurred that if such pressures reemerged and threaten to stall the recovery, further intervention could be warranted (paragraph 25 of the staff report). The 2005 report discussed the role that intervention had played in policymaking in Japan and, when assessing its effectiveness, cited the most recent academic research at that time (Box 1). The 2005 report also suggested that the authorities could resume the intervention, if necessary to combat the deflationary spiral (paragraph 28). Earlier reports had also dealt with this issue (e.g., a 2000 Selected Issues chapter assessed the effectiveness of past intervention on the yen-dollar rate).

42. Norway. The only intervention was the accumulation of NFA in the Petroleum Fund. Staff reports did note explicitly that the real exchange rate was affected by this accumulation policy. Indeed, the risk of Dutch disease (i.e., the undesirable effects of rapid real exchange rate appreciation) was discussed in a selected issues paper in 2005 (listed in Table 1).

43. Singapore. The IEO report does not say why coverage of intervention policies was regarded as incomplete. Staff reports do describe the exchange rate and monetary framework. The 2003 and 2004 Article IV reports note that the nominal effective exchange rate (NEER) is targeted by the Monetary Authority of Singapore (MAS) to achieve its inflation target. In the 2003 report, the Fund Relations annex states that the Singapore dollar is “permitted to float” but adds immediately after that “the MAS monitors its value against an undisclosed basket of currencies and intervenes in the market to maintain its value within an undisclosed band.” In the 2004 staff report, the Fund Relations annex states “Singapore’s exchange rate regime is a managed floating regime” and then continues with the same description of intervention policy as the 2003 report. Both reports discuss the path of the NEER in light of this monetary framework centered on it. In this context, it is well understood that intervention in the foreign exchange market is the primary instrument by which the MAS ensures that the target path of the NEER is in line with the inflation objective.

Annex IV. Staff Analysis on Exchange Rates—The Role of CGER

44. The extension of CGER analysis to include several emerging markets and the associated substantial methodological refinements are an important part of the strengthening of exchange rate surveillance out-lined in the Medium-Term Strategy. Yet the main report pays scant attention to CGER. One of the benefits of CGER analysis is its consistency and evenhandedness—assessments for advanced economies and emerging markets are obtained by applying the same analytical framework. If a criticism of the IEO report is that the analytical basis of staff assessments across the membership suffers from arbitrariness, the CGER framework seem to provide an appropriate remedy.

Table 1.Summary of Staff’s Conclusions Based on the Evidence Included in the IEO Report
IssueStaff Reading of IEO EvidenceActions Taken or Ongoing
1. General coverage of exchange rate issuesAdequate in vast majority of cases. Need to avoid isolated cases of lack of coverage of relevant issues.Revision of the 1977 Decision. Various initiatives for better knowledge dissemination (see point 11 below).
2. Integration between exchange rate and other policy areas (excluding spillovers)Generally good.
3. Description of de facto exchange regimesGood, with only a few exceptions. Need to ensure appropriate "labeling" and full consistency with MCM classification in a few cases.The forthcoming Report on Exchange Arrangements, Restrictions, and Markets (REARM) will reassess classification criteria and discuss how to best ensure consistency with staff reports in all cases.
4. Intervention policiesBroadly adequate, with some room for improvement particularly with respect to spillover effects.The upcoming REARM will reassess the issue of data availability regarding intervention and make proposals on how analysis of intervention can be strengthened. See point 8 below on spillover effects.
Staff missions ensure countries are aware of availability of technical assistance (TA) in this area.
5. Data availabilityDeficiencies remain for several countries including for intervention data. Need to continue to support availability of technical assistance and explore extent of cases where authorities seem unwilling to share relevant information.The forthcoming REARM and the Review of Data Provision to the Fund will reassess this issue.
6. Assessment and analysis of the exchange rate levelAlmost always included in staff reports, but in a sizable minority of cases, in spite of recent progress, need to further strengthen the analytical underpinning of assessments.The CGER coverage has been broadened and is expected to be further broadened this year; new analytical work on trade elasticity and equilibrium exchange rates in low-income countries has been completed and more is under way. Various knowledge dissemination initiatives are under way (see point 11).
7. Assessments of exchange rate regimeRegularly included. Quality of analysis generally adequate. Need to strengthen the assessment in some cases. No excessive urgency in advocating change except in rare cases, but need to ensure countries are aware of availability of TA to address implementation issues.Various knowledge dissemination initiatives are under way (see point 11).
8. Integration of multilateral (spillover) issues in bilateral surveillanceA strengthening is needed, in spite of some recent progress.Progress in analyzing spillovers is a monitorable target from the 2004 BSR, and will be reviewed in the 2008 review. Guidance to staff on implementation of the Medium-Term Strategy (MTS) specifically instructed departments to enhance coverage of spillover issues. Staff Briefings to disseminate the findings of the WEO and GFSR across departments have been organized. The role of the Surveillance Committee is being revived.
9. EvenhandednessNo evidence of lack of evenhandedness is found; nevertheless, need to remain vigilant in this critical area.Revision of the 1977 Decision.
10. ImprovementClear evidence of improvement during 1999–2005.To be reassessed by the 2008 Triennial Surveillance Review (TSR).
11. Approaches for consolidating progress achieved and further strengthening exchange rate advice across all areasNeed to clarify general principles for best practice. No clear need for additional detailed guidance notes on all dimensions of exchange rate work. Need to continue information dissemination processes.Enhanced training, best practice workshops, forthcoming book on the best on exchange rate analysis, enhanced review process, 2007 Research Conference.
12. Policy dialogueA large majority of countries are satisfied. Need to ensure remaining shortcomings are addressed.Revision of the 1977 Surveillance Decision. (A revised Surveillance Decision would underscore the criticality of effective policy dialogue.)
13. Overall assessment of how the Fund plays its various rolesAdequate in most dimensions. Need to strengthen the role of the Fund as “broker for international policy coordination.” Some evidence of lack of agreement on how the Fund should play the role of “ruthless truth-teller.”Multilateral consultations have been introduced, strengthening the role of the Fund as broker for international policy coordination. The revision of the 1977 Decision is an opportunity to clarify the expectations of the membership.
14. Impact on policy decisionsEvidence of some impact in the large majority of cases. However, there is scope for improvement.Same as above. Enhanced methodology for assessing the effectiveness of surveillance, to be implemented in the 2008 TSR, will review impact (among other dimensions).

45. Our sense from the outreach staff has done with officials, market participants, and academics is that, while margins of uncertainty remain significant, the methodology is currently at the frontier. Furthermore, the response of area department staff has been encouraging: indeed RES has been approached by desks of several countries not included in the exercise who wished to apply these methodologies to their respective countries. If the CGER results were not useful as an input to staff assessments, it would be necessary to explore which other available methodologies that satisfy the criteria of cross-country consistency are superior, and it would have been helpful had the IEO identified them.

46. Finally, in reference to BD 3, the description of CGER analysis is seriously incomplete. It gives the impression that CGER consists mainly of the macroeconomic balance approach, with just a passing reference to the equilibrium real exchange rate (ERER) approach (a third approach was added more recently). This characterization comes after a (quite weak) discussion of available methodologies for estimating equilibrium real exchange rates, which emphasizes primarily methods analogous to the ERER. Moreover, as recognized in the report, CGER analysis aims at assessing the medium-term consistency of the exchange rate with economic fundamentals. Its assessments should, therefore, be compared to actual exchange rate developments over the course of 3–5 years, and not to actual exchange rate changes at very short horizons. Yet BD 3 refers to CGER estimates having missed the direction of prospective exchange rate changes (paragraph 40).


For brevity, we do not compare point by point our conclusions with those of the IEO, which are essentially fairly negative across the board. Paragraph numbers without further reference refer to the main report; otherwise the number of the corresponding Back-ground Document (BD) is reported.


The report notes that the level of the exchange rate was not reas-sessed in the 2005 Article IV staff report for Saudi Arabia in spite of the terms of trade shock. Such a reassessment was included in the 2006 staff report. The alleged shortcoming identified for China dates back to 2001–02. Indeed, the treatment of exchange rate issues in the 2006 Article IV report is widely considered best practice (Annex I).


“Treatment of Exchange Rate Issues in Bilateral Surveillance—A Stocktaking,” EBS/06/107, August 7, 2006 (henceforth the “Stocktaking Paper”), notes the need to improve the description of the accumulation in reserves of official public sector inflow. Another aspect that should be regarded more closely is the effect of intervention on the intervention currencies (see discussion on spillovers below). The forthcoming Review of Exchange Rate Arrangements, Restrictions, and Markets (REARM) discusses how analysis of intervention can be strengthened.


In 2005–06 at least eight technical assistance reports have dealt in detail with the tactics of intervention.


See, for example, Olivier Jeanne and Romain Ranciere, “The Optimal Level of International Reserves for Emerging Market Countries: Formulas and Applications,” IMF Working Paper No. 06/229; and Joshua Aizenman and Jaewoo Lee, “Financial Versus Monetary Mercantilism: Long-Run View of the Large International Reserves Hoarding,” IMF Working Paper No. 06/280. The WEO has also focused on this issue (see, for example, the September 2003 WEO, Chapter II (“Are Foreign Exchange Reserves in Asia Too High?”)).


Two seminars on this issue were held just this month (“Foreign Exchange Reserves in Emerging Market Countries” and “Accumulation of Official Reserves—Trends and Challenges”).


Assessing whether staff has adequately flagged these shortcomings in staff reports would require more in-depth analysis. Staff reports typically discuss data problems. It is true that the bottom line assessment is usually that data are “adequate for surveillance.” But it is not obvious that, where data problems have “impaired” the ability to conduct exchange rate analysis, the data are necessarily “inadequate” to exercise surveillance.


The IEO staff questionnaire shows that 30 percent of the staff surveyed does not find CGER useful. However, CGER has recently been strengthened, not only in terms of country coverage, but also of methodologies. See Annex IV for a broader discussion of the treatment of CGER in the IEO report.


The Stocktaking Paper had also found that, while calls for exchange rate flexibility were often not backed up by a complete set of pros and cons, only in some cases a more comprehensive discussion was needed.


One of these three cases is Morocco, for which the IEO report mentions the 2005 staff report as an example of good practice (Chapter 3, footnote 25; this footnote lists other examples of recent improvements including Ukraine).


Of the two examples mentioned in the report to illustrate possible perceptions of lack of evenhandedness, one—the United Kingdom—is only one of five cases of alleged lack of forthrightness, but the remaining cases are emerging market countries. The second one—Greece, where staff is faulted for allegedly not challenging the authorities’ unwillingness to share information needed for surveillance—is, like-wise, only one of several such cases (the breakdown by country grouping is undisclosed in this case).


Other activities aimed at sharing best practices include the forthcoming publication of a book collecting the best in exchange rate analysis at the Fund, and the focus of the 2007 Annual Research Conference on Exchange Rate Issues.


See SM/06/248 (7/12/06) (, especially paragraphs 7–9, 23–31 (including Box 3), and 54–59.

    Other Resources Citing This Publication