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

Background Document 6: IEO Survey of the Authorities and Staff

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

1. This background document summarizes the results from the IEO survey of country authorities and IMF staff. The first section sets out the methodology. The second section summarizes the main survey find-ings. The third section presents some analysis of the survey results. The fourth section presents a further analysis of the survey responses for large emerging market economies. Annex A6.1 provides additional detail on the methodology. Annex A6.2 lists the economies surveyed. Finally, Annexes A6.3 and A6.4 reproduce the survey questions for national authorities and for IMF staff, respectively.

Survey Methodology

2. The survey was conducted in late 2006 and early 2007. The survey module for the authorities was managed by an outside company,1 while the module for IMF staff was managed by the IMF’s Technology and General Services Department. All survey recipients were advised that their responses would be treated as confidential and that individual countries and staff members would not be identified.

Populations surveyed

3. A single survey questionnaire was sent to all member countries through the offices of IMF Executive Directors. Also surveyed were the central banks of four regional monetary unions 2 and the European Commission. In total, 192 questionnaires were sent out. For reporting purposes, the economies were classified into four categories: “Major advanced;” “Other advanced;” “Large emerging market economies” (“Large EMEs”); and “Other emerging market economies/developing countries” (“Other EMEs/DCs”).3 See Table A6.1 for details.

Table A6.1.Authorities’ Views of IMF Impact on Major Exchange Rate Policy Decisions(Number of economies)
Extent of ImpactAdvancedLarge EMEsOther EMEs/DCsTotal
Instrumental133337
Marginal881834
No impact7018
Little/no discussion with IMF63211
Total22145490

4. A single questionnaire was also sent to current IMF staff at the senior economist level and above in all area departments and in selected functional departments. The staff members at these levels were targeted because of their longer experience in dealing with exchange rate issues and their greater access to confidential discussions and material. Some 791 staff members received the survey.

5. The response rates from both populations (59 percent and 47 percent, respectively) were high enough to yield statistically meaningful inferences (see Annex A6.1 for details).4

Main features of the survey questionnaires

6. The two survey modules were structured according to the main questions guiding the evaluation, as presented in the Issues Paper. Particular emphasis was placed on certain aspects of the process of surveillance, such as the nature of the dialogue between staff and authorities and the impact of IMF activities on member countries’ policies, where the survey (along with the direct interviews of authorities and staff) would be the main source of evidence. In addition, for triangulation purposes, questions in both surveys explicitly incorporated hypotheses regarding the quality of Fund advice and analysis developed on the basis of the IEO’s two desk reviews.

7. Each survey module contained 15 sets of questions.5 Many questions were the same for both populations, although a few questions (such as those concerning the internal review process of the IMF) were specific to one survey module. Several of the questions asked respondents to rate, on a five-point scale, their level of agreement with particular statements.6 Other questions asked respondents to select from a range of options the one that best described their views.

8. Respondents to the staff survey were given the option of selecting either a particular country or “no country” at the outset of the survey, based on their level of interaction with country authorities on exchange rate matters during the evaluation period. If a country was selected, it was used as a reference point for subsequent country-specific (but not general) questions. Staff who did not select a country (as in the case of staff involved largely in a reviewing capacity), received only a subset of the subsequent questions. Staff were also given the option to select “No involvement” with exchange rate matters, in which case the survey was not pursued further with them.

Major Findings

Areas of exchange rate policy for discussion

9. The survey first asked both the authorities and IMF staff a series of five questions about areas of exchange rate policy selected for discussion. For convenience, the survey considered five exchange rate policy areas: regime choice, regime management, exchange rate level, cross-border effects of policies (spillovers), and financial stability considerations.

10. As a starting point, the survey asked the authorities to identify the areas of focus in internal policy discussions (Figure A6.1).7 The results show that the five areas of exchange rate policy received varying degrees of focus in internal discussions during the 1999–2005 period. There were important differences across the four country groups: for “Major advanced” countries, spillover effects received the most attention, with over half of the respondents identifying the topic; for “Large EMEs,” management of the exchange rate regime was the most prevalent topic in internal policy discussions, being identified as a focus by nearly all respondents.

Figure A6.1.Survey of Authorities: Relevance of Issues for Internal Discussions

(In percent)

11. The survey then asked the authorities how they viewed the importance given to the different exchange rate areas in their discussions with IMF staff (Figure A6.2). Here, we observe some interesting differences both across country groups and from the previous survey question. For example, 40 percent of the respondents from “Major advanced” countries thought that spillover effects were a focus of discussions with IMF staff (whereas in the previous survey question some 60 percent had identified the topic as a focus in internal discussions). And for financial stability concerns, the authorities of “Large EMEs” saw these as far more important in internal policy discussions (about 60 percent) than in their dialogue with the IMF (about 30 percent).

Figure A6.2.Survey of Authorities: Coverage of Discussions with IMF

(In percent)

12. Being asked the identical question, the staff respondents saw different areas of emphasis (Figure A6.3). The contrast is particularly pronounced for both “Major advanced” and “Other advanced” economies. For example, among staff respondents who worked on “Major advanced” countries, almost 60 percent saw financial stability considerations as an area of focus, whereas no country respondents did so. Likewise, some 80 percent of the staff identified management of the exchange rate regime and spillover issues as areas of focus (whereas only 30 percent and 40 percent of the country respondents did so for these two issues, respectively).8

Figure A6.3.Staff Survey: Coverage of Discussions with Authorities

(In percent)

13. The survey asked both the authorities and staff for their views of how the authorities agreed with the staff’s analysis and assessments of exchange rate issues (Figure A6.4). A large majority of the surveyed authorities (more than 70 percent for most country groups) said they were in broad agreement with the staff analysis and assessments, though the percentage was somewhat lower (less than 60 percent) for “Large EMEs” (left-hand panel). For each country group, a smaller percentage of the staff respondents saw broad agreement (right-hand panel). For example, about 70 percent of the staff respondents who worked on “Other advanced” countries saw broad agreement, compared with about 80 percent of the authorities; for “Large EMEs,” the share of the staff seeing agreement was only 30 percent, compared with the authorities’ 60 percent.

Figure A6.4.Staff’s Analysis and Assessment

(In percent)

1Staff’s perceptions on the extent to which the authorities agreed with staff’s analysis and assessment.

14. Figure A6.5 indicates where disagreement lies. About 25–35 percent of both populations indicated disagreement in the areas of regime choice, management of the regime, and level assessment, and virtually no one identified disagreement in the other two policy areas.

Figure A6.5.Staff Analysis and Assessment: Major Areas of Disagreement

(In percent)

Quality of the dialogue

15. The second part of the survey (consisting of six questions) probed aspects of the dialogue on exchange rate issues and the role of the IMF Executive Board, by asking both the authorities and staff to rate their level of agreement with a series of statements.

16. First, in terms of the broad nature of the dia-logue, the findings are positive on many dimensions, with similar responses across the two populations (Figure A6.6). Of the country respondents, 70–90 percent appear satisfied with many process-related aspects of their interaction with IMF staff. In particular, close to 75 percent agreed that discussions with staff were substantive and two-way, but close to 30 percent of the respondents from “Other advanced” economies disagreed. On other issues—such as the appropriate frequency of discussions, balance among informal-ity, confidentiality, and the requirements to report to the Executive Board, candidness and willingness to raise politically sensitive issues, and respectfulness and open-mindedness—80 percent or more were positive about staff performance. The authorities and staff share the perception that Board documents generally do a good job in reflecting the authorities’ policy discussions with staff (about 80 percent of each population).

Figure A6.6.On the Nature of the Dialogue: Interactions Between Authorities and Staff

(In percent)

17. Second, the next question asked the authorities and staff how they perceived the handling of such sensitive policy matters as foreign exchange market intervention and exchange rate regime selection (Figure A6.7). There is a major difference between the authorities and the staff, especially for “Other advanced” countries and “Large EMEs.” For “Other advanced” countries, 40 percent of the staff respondents thought that the authorities restricted discussion of certain sensitive matters, whereas about 20 percent of the country respondents thought that they did. For “Large EMEs,” on the other hand, sensitivity is recognized more by the authorities (about 30 percent of total) than the staff (about 20 percent). About a third of the 19 countries that reported excluding sensitive policy issues from the discussion said they did so based on concerns about leaks.9

Figure A6.7.On the Nature of the Dialogue: Off-the-Table Issues and Leaks

(In percent)

18. Third, the survey asked the authorities and staff about the sources of external exchange-rate-policy-related analysis and the contribution of the Executive Board to the development of policy (Figure A6.8). Across both populations, the percentage of those who think that the IMF is the main source of external analysis rises as the country concerned is smaller and less advanced. For example, no respondents from “Major advanced” countries said that the IMF was the main source of external advice, whereas more than half of those from “Other EMEs/DCs” said so. For each country group, moreover, a larger share of the staff respondents than of the country respondents saw the IMF as the main source of external analysis. As to the contribution of the IMF’s Executive Board, except in “Other EMEs/DCs,” fewer than 30 percent of the authorities or staff respondents thought that the Board was consequential in the development of country policy. The large share of respondents from the “Other EMEs/DCs” who thought otherwise may partly reflect the fact that this group includes a number of countries with IMF-supported programs; in these countries, close to 70 percent saw the Board as consequential in the development of country policy.

Figure A6.8.On the Nature of the Dialogue: Sources of Analysis and the Executive Board Contribution

(In percent)

19. The last three questions sought to identify which aspects of the IMF’s staff work should be strengthened to help improve the substance and usefulness of the dia-logue. For presentational purposes, the three questions address the following three sets of issues:

  • Analytics. The analytical or empirical underpinnings of staff assessments and advice; multilateral and regional perspectives informing the work; integration of exchange rate policy advice with advice in other policy areas; and the sectoral balance sheet implications of exchange rate movements.
  • Implementability. Whether country-specific factors and constraints are taken into account when formulating advice; whether general advice is developed into more concrete implementation issues; the cost and benefits of policy options; and the sense of urgency given to certain policy actions.
  • Delivery. The modalities of interaction between authorities and staff; the clarity and nuance of views; the documentation provided to the authorities; and the timeliness of the analysis.

20. Across the three sets of issues, sizable minorities of the authorities saw room for improvement (Figures A6.9, A6.10, and A6.11). Moreover, the share of those who saw room for improvement is considerably larger among the country respondents (about 30–40 percent for analytics and implementability) than for the staff respondents (about 10–20 percent). The gap in perception seems to be particularly large overall for all aspects of implementability, and to a lesser extent analytics.10 Although this same pattern is also observed for the third area (delivery), a smaller share of the authorities identified room for improvement (20–30 percent).

Figure A6.9.Improving the Substance and Usefulness of Discussions: Analytics

(In percent)

Figure A6.10.Improving the Substance and Usefulness of Discussions: Implementability

(In percent)

Figure A6.11.Improving the Substance and Usefulness of Discussions: Delivery

(In percent)

Impact of IMF work

21. The survey tried to ascertain the impact of IMF work on the exchange rate policies pursued by member countries during the 1999–2005 period (Figure A6.12). An interesting pattern of responses emerged. Except for “Other EMEs/DCs,” the IMF’s impact, as perceived by both populations, diminishes as the country in question is larger and more advanced. For example, none of the respondents from “Major advanced” economies considered the IMF as having been instrumental in major exchange rate policy decisions. The staff respondents, however, tended to regard the IMF’s impact as somewhat greater than the authorities did: in particular, for the “Large EMEs,” close to half of the staff thinks that the IMF affected these countries’ major decisions, which is double the authorities’ percentage. As with a number of previous questions, however, the views of the authorities and the staff converge for “Other EMEs/DCs.”

Figure A6.12.Opinions About IMF Impact on Countries’ Major Decisions on Exchange Rate Policy

(In percent)

Role of the IMF

22. The next set of questions addresses the role of the IMF by asking the authorities and IMF staff to rate the extent to which the IMF had played different roles in relation to exchange rate matters—both in the context of country-specific discussions and for the IMF membership as a whole. For the purpose of these questions, the middle point of the scale was considered to represent “about right,” while the end points represented “missed opportunities” (or “underplayed”) and “overplayed.” 11

23. Concerning the role of the IMF vis-à-vis individual countries, more than two-thirds of the surveyed authorities agreed that the IMF was getting it “about right” in its roles as “confidential advisor” and “sounding board” and just above half agreed that that was also the case in its role as a “consensus builder” among domestic policymakers (Figure A6.13).12 There are differences across country groups, however, with respondents from the “Large EMEs” most likely to see “missed opportunities.”

Figure A6.I3.Survey of Authorities and Staff: Views on the Role of the IMF1

(In percent)

1 Responses indicating that the roles were played “about right” are not displayed.

24. As to the IMF’s role vis-à-vis the entire member-ship, roughly two-thirds of the country respondents saw the IMF as getting it “about right” on its roles as “provider of credibility” (through assessment of national policies) and as contingency “lender” (Figure A6.14) On the other hand, fewer than half of the surveyed authorities gave the same rating for the IMF’s roles as “ruthless truth-teller” to the international community and “broker” for international policy coordination, with a significant number seeing scope for the IMF to be more proactive. About 40 percent of the staff shared the latter view. Some country differences are striking. For example, while more than 40 percent of the country respondents from “Large EMEs” saw missed opportunities for the IMF’s role as lender, a similar percentage of those from “Major advanced” countries thought that the IMF had overplayed this role.13

Figure A6.14.Survey of Authorities and Staff: Views on the Role of the IMF1

(In percent)

1 Responses indicating that the roles were played “about right” are not displayed.

Content of IMF analysis

25. The survey asked the authorities and the staff to identify how they viewed the various dimensions of the content of IMF analysis: (1) the tailoring of advice to country-specific circumstances; (2) bias in exchange rate regime selection; (3) attention given to intermediate regimes; and (4) attention given to structural (as opposed to cyclical) factors in exchange rate regime selection (Figure A6.15). For the most part, only the minority of both populations expressed concerns about the approach the IMF uses to assess regime choice and suitability. The most prevalent concern of the authorities related to bias in regime selection, with 30–60 percent of the respondents raising this concern, depending on the country group (even 40 percent of the staff respondents identified the concern). The proportion of those expressing concerns is by far the highest for “Large EMEs.”

Figure A6.15.Authorities’ Concerns on the Approach the IMF Used to Assess Regime Choice or Suitability

(In percent)

Authorities’ views

26. The next five survey questions asked the authorities for their views of selected aspects of IMF analy-sis.14 First, as to the IMF’s methods to assess the level of exchange rates, about 20 percent of the respondents found them adequate and useful (Figure A6.16). An additional 45 percent thought that more could have been done, while still finding the analysis useful. The country respondents from “Large EMEs” tended to see more problems: about half of them saw the methodologies as either impractical or inadequate.

Figure A6.16.Authorities’ Views on the IMF’s Methodologies in the Assessment of the Equilibrium Level of Exchange Rates

(In percent)

27. Second, as to the methods used by the IMF to assess external competitiveness, a similar pattern emerged (Figure A6.17). More than 75 percent of the country respondents saw the methodologies as adequate and useful, or indicated that, while more could have been done, what was done was useful. Once again, about half of the respondents from “Large EMEs” find these methods to be either inadequate or of little practical use.

Figure A6.17.Authorities’ Views on the IMF’s Methodologies in the Assessment of External Competitiveness

(In percent)

28. Third, the survey asked the authorities how they view the usefulness of IMF analysis drawn from the experience of other countries (Figure A6.18). About three quarters of the respondents said that the IMF’s analysis was at least somewhat useful for decision making. The majority of the respondents from “Other advanced” and “Large EMEs” identified a problem, reporting that cross-country analysis was either not provided by IMF staff or not useful for decision making when provided.

Figure A6.18.Authorities’ Views on the Usefulness of the IMF’s Analysis Based on the Experience of Other Countries

(In percent)

29. Fourth, the authorities were asked to select one statement that best characterized the staff’s analysis of global and regional spillovers, global imbalances, and the interaction between bilateral and multilateral analysis (Figure A6.19). On average, about half of the country respondents considered that the staff had done a reasonable job of integrating the analysis of spillover effects impacting their economies, but this assessment is much more pronounced among the authorities from “Major advanced” countries than for those from “Large EMEs.”

Figure A6.19.Survey of Authorities: Global or Regional Spillovers Affecting Your Country’s Exchange-Rate-Related Developments

(In percent)

30. Finally, the authorities were asked whether the regional and global impact of their policies was considered (Figure A6.20). Overall, about 40 percent of the country respondents said that this was not a relevant issue for their countries. Among those who said it was relevant, the assessments varied quite drastically: in the “Major advanced” group, all of them found the issues well integrated into the analysis, but in the “Large EMEs” about 90 percent of the respondents indicated that the staff either rarely identified the issue or that it was not well integrated.

Figure A6.20.Survey of Authorities: Global or Regional Impact of Your Country’s Exchange Rate Policy

(In percent)

Staff views

31. The next six questions asked the staff about aspects of IMF analysis. First, the survey sought to understand why, as identified by the Executive Board, clear and candid treatment of exchange rate issues in IMF reports remains a challenge, by suggesting nine possible reasons (Figure A6.21). More than 40 percent of the staff respondents pointed to the expectation that their analysis would be published and the need to preserve close relationships with country authorities as important factors; almost as many respondents, however, said that these were not the issue. A somewhat smaller share (about 37 percent) of the respondents considered the interlinkage of exchange rate policy issues with other macroeconomic policy areas as posing difficulty, and almost 30 percent pointed to the IMF’s internal review process as a significant influence on clear and candid treatment (presumably by creating more cautious language and diluted coverage).

Figure A6.21.Staff Survey: Impediments to Clear and Candid Treatment of Exchange Rate Issues

(In percent)

32. Second, staff were asked to identify the scope for improvement in the analysis of eight exchange-raterelated policy areas (Figure A6.22). Debt sustainability analysis is the only area that was seen by the majority of the staff respondents (more than 50 percent) as offering little or no scope for improvement. The largest share of the staff respondents (more than 40 percent), on the other hand, saw room for improvement in the analysis of policy spillovers and the integration of regional and multilateral perspectives.

Figure A6.22.Staff Survey: Quality of IMF Analysis

(In percent)

33. Third, staff were asked to identify the areas of significant room for improvement that would raise the overall quality of analysis (Figure 6.23.). Consistent with the previous answers by staff, more than half of the respondents recognized that improving the analysis of policy spillovers would help in that regard. The majority also recognized exchange rate level assessment as a promising area for improvement (this is somewhat surprising given that in answering the previous question, only 35 percent had identified room for improvement there). Around 40 percent of the staff respondents identified “regime choice and sustainability” and “better availability of data” as the next two most promising areas.

Figure A6.23.Staff Survey: Areas of Significant Room for Improvement

(In percent)

34. Fourth, as a followup question, those who identified “better availability of data” as a promising area for improvement (about 40 percent of the staff respondents) were asked to rate their level of agreement with a series of data-related statements (Figure 6.24.). The staff had a split view on whether data availability problems had impaired analysis, but about 65 percent saw the IMF as having an informational advantage over the public. Almost 40 percent said that the authorities were technically unable to provide data, while more than 30 percent considered them unwilling to share some critical data.

Figure A6.24.Staff Survey: Agreement with Selected Statements on Data Quality

(In percent)

35. Fifth, the staff respondents were asked to rate the extent of help they received from various internal sources on exchange rate policy matters (Figure 6.25.). Seventy percent of them saw area departments’ analysis as helpful, and 45 percent cited significant help from the Research Department. More than 70 percent of the respondents did not regard as significantly helpful the 1977 Surveillance Decision.15

Figure A6.25.Staff Survey: Internal Sources of Help for Staff’s Work on Exchange Rate Issues

(In percent)

36. Finally, the last of the staff-directed questions asked those respondents who were involved in work on CGER-covered countries to rate the usefulness of the CGER methodological approach (Figure 6.26.).16 On average, 40 percent of this group saw CGER work as very useful, including 60 percent of those working on “Major advanced” economies.

Figure A6.26.Staff Survey: Overall Usefulness of the CGER Exercise

(In percent)

Did the quality of IMF analysis improve?

37. The survey concluded by asking both the authorities and staff whether the overall quality of the staff’s exchange rate analysis and assessment had changed, and to what extent, from about 1999 to the end of the evaluation period (Figure 6.27.). Overall, a slight majority of authorities saw improvements in the overall quality—albeit with sizable percentages of the respondents from “Large EMEs” thinking that it was unchanged and those from “Other advanced” and “Other EMEs/DCs” thinking that it had improved, and in some cases significantly. On the issue of quality improvement, there was a significant gap in the perceptions of the authorities from “Major advanced” economies and “Large EMEs” relative to those of staff, while perceptions for the other country groups were about the same.

Figure A6.27.Perceived Change in the Overall Quality of IMF Staff’s Analysis and Assessment over the Evaluation Period

(In percent)

Analysis of Survey Findings

38. This section takes a deeper look at (1) the authorities’ perceptions of IMF impact on their major exchange rate decisions and (2) the authorities’ assessment of how the overall quality of IMF staff analysis and advice has changed over time.17 First, as to the IMF’s impact, we examined the 90 economies in which, in the views of the authorities, major decisions had been taken during the 1999–2005 period (Table A6.1). As noted earlier (see Figure A6.12), the impact was more often seen as instrumental in other emerging and developing economies than in advanced and large emerging market economies.

39. As highlighted above, the level of attention given to a topic in internal discussions did not always correspond to the attention given in the authorities’ discussions with the IMF. In 37 cases, mainly smaller emerging and developing economies, the authorities saw the IMF as instrumental in helping shape major policy decisions. By contrast, in 11 cases, mostly advanced economies, the authorities reported that, to the extent that major decisions were taken related to exchange rate policy in 1999–2005, there was little or no discussion with the IMF. In most of these latter cases, given the country’s own technical capacity, no followup support was requested or received.

40. There was some evidence of a correlation between the extent to which the authorities took sensitive policy issues off the table, and the impact the IMF had on their policy decisions (Table A6.2 y). For example, almost 70 percent of the country respondents who said that the IMF’s impact was instrumental did not exclude or restrict issues for discussion, whereas the share was down to about 40 to 50 percent for those who saw no impact or had little or no discussion with the IMF. The evidence was mixed with regard to the links between the authorities’ agreement/disagreement with staff analysis and IMF impact. A large share (more than 80 percent) of those who said that the IMF was instrumental agreed with the findings of staff analysis, but the same was true for those who said the IMF had no impact (Table A6.3).

Table A6.2.Impact and Restrictions on Issues for Discussions(In percent)
Extent of ImpactSensitive Issues Excluded or RestrictedNeutralNo Exclusion or Restrictions
Instrumental191169
Marginal221959
No impact6337
Little/no discussion with IMF4555
Table A6.3.Impact and Agreement with Staff Analysis(In percent, unless otherwise noted)
Extent of ImpactAgreementDisagreementArea of Greatest Disagreement1
Instrumental8119Regime choice/level assessment
Marginal5941Level assessment
No impact8813Regime choice
Little/no discussion with IMF6436Regime choice

The area that was selected by the highest number of respondents among those who indicated that they disagreed with staff’s analysis.

The area that was selected by the highest number of respondents among those who indicated that they disagreed with staff’s analysis.

41. Impact also appeared to be correlated with the perceived quality of the dialogue, the importance of the IMF as a source of external advice, and the significance of Executive Board contributions (Table A6.4). Among the country respondents who said that the IMF was instrumental, only 8 percent disagreed with the statement that their discussions with IMF staff had been substantive and two-way; only 22 percent disagreed with the statement that the IMF was the principal source of external advice; and no one disagreed with the statement that the Executive Board made a significant contribution to policy development. The percentage of respondents that disagreed with those statements in general rises with a fall in impact.

Table A6.4.Impact and Aspects of the Authorities’ Interactions with the IMF(In percent)
Disagreement That:
Extent of ImpactDiscussions were substantive, two-wayIMF was principal source of external analysisExecutive Board provided an important input
Instrumental8220
Marginal93630
No impact508888
Little/no discussion with IMF455590

42. Analysis of the survey data also indicates a correlation between the respondents who said that the IMF was instrumental and those who said that the quality of IMF analysis had improved (Table A6.5). Indeed, 70 percent of the country respondents who said that the IMF was instrumental also said that the overall quality of IMF analysis had improved, whereas half or less of those from the other groups saw improvement.

Table A6.5.Impact and Views of How the Quality of Analysis Has Changed(In percent)
Extent of ImpactBetterAbout the SameWorse
Instrumental7030
Marginal44506
No impact5050
Little/no discussion with IMF453618

43. The fact that the overall quality of analysis has remained the same would not be a cause for concern if the staff’s work had been perceived as satisfactory to begin with. This, however, does not appear to have been the case along various dimensions. For example, many of the country respondents who saw quality remaining the same agreed that staff could have done more in the area of analytical and empirical underpinnings for their assessments and advice (Table A6.6) and could have better integrated country-specific factors and constraints into their analysis (Table A6.7). Moreover, even among those country respondents who saw the overall quality as improving, the evidence suggests that about half felt that more could have been done, in particular with regard to the integration of regional and multilateral perspectives and the sense of urgency given to the policy advice. The authorities did not equate statements that quality had improved or remained the same from 1999–2005 with the view that no further improvements could have been made. In many cases, the authorities saw room for improvement.

Table A6.6.Views on Overall Quality Change and Staff Analysis1(In percent unless otherwise noted)
Analytical UnderpinningsMultilateralIntegration with Policy AreasBalance Sheet
Overall Quality Has Become:AgreeNeutralDisagreeAgreeNeutralDisagreeAgreeNeutralDisagreeAgreeNeutralDisagree
Better363331443422351847342639
About the same492229292448351947332048
Memorandum item Worse (number of economies)102202102111

Respondents were asked to rate the extent of agreement with the statement that IMF staff could have done more to improve the usefulness and substance of discussions in the respective areas.

Respondents were asked to rate the extent of agreement with the statement that IMF staff could have done more to improve the usefulness and substance of discussions in the respective areas.

Table A6.7.Views on Overall Quality Change and Implementability1(In percent unless otherwise noted)
Country-SpecificImplementationUndue UrgencyCost-Benefit
Overall Quality Has Become:AgreeNeutralDisagreeAgreeNeutralDisagreeAgreeNeutralDisagreeAgreeNeutralDisagree
Better433028442531502228373330
About the same451936313138231860331850
Memorandum item Worse (number of economies)202102111211

Respondents were asked to rate the extent of agreement with the statement that IMF staff could have done more to improve the usefulness and substance of discussions in the respective areas.

Respondents were asked to rate the extent of agreement with the statement that IMF staff could have done more to improve the usefulness and substance of discussions in the respective areas.

44. The survey of authorities suggests that the level of attention given to exchange-rate-related topics in authorities’ discussions with staff does not always correspond to the attention given internally (Figures A6.1 and A6.2 above). At the same time, the IMF’s impact on major policy decisions taken by the authorities was found to differ substantially across country groupings (Figure 6.12.). Information from all three of these survey questions can be combined to assess the relationship between “gaps in coverage,” as suggested by the relative emphasis placed on different exchange rate issues internally and in discussions with IMF staff, and IMF impact, as perceived by the authorities. Figure A6.28 shows a positive correlation between these gaps and the extent of impact: the larger the number of countries with perceived “gaps in coverage,” 18 the smaller the perceived IMF contribution to the authorities’ policy decisions across countries.19

Figure A6.28.IMF Impact on Major Exchange Rate Policy Decisions and Differences in Emphasis in Policy Discussions1

(In percent)

1 Authorities’ responses (in percent of economies reporting a given degree of impact) regarding the relative emphasis given to exchange rate issues in internal discussions compared to those with IMF staff.

Further Analysis of Survey Responses for Large EMEs

45. This section provides a detailed analysis of the survey data for large EMEs by comparing, for each economy, the survey responses from the authorities with the “matched” answers from the corresponding staff respondents. When there are more than one staff respondent for a country, the staff responses were averaged. Specifically, for rating questions, the staff response was the simple average of all responses; for questions that ask for the selection of an option, the staff response is given by a simple majority view (in the case of a tie between two or more options, we choose the one that is judged to reflect the majority view more closely on the basis of how all the respondents have selected the available options).20 The numbers reported in the figures below refer to the large EMEs for which matched responses are avail-able; these matched results support the findings of the overall surveys.

On the focus of discussions

46. The staff and the authorities differ significantly in their perception of the focus of their exchange-rate-related discussions (Figure 6.29.). In terms of the five policy areas identified in the surveys, there is a high level of agreement (in 13 out of 15 countries) on the area of exchange rate regime management, whereas the level of agreement is low for policy spillovers (4 out of 15 countries—counting diagonally from the north-west corner) and for financial vulnerability assessment (5 out of 14 countries). The average or representative staff responses for large EMEs are similar to the overall survey results, except on the issue of regime choice. The “representative” staff identified regime choice as a focus of discussion in only 4 out of 15 countries (26 percent), whereas 56 percent of the staff respondents in the overall survey made that same assessment.

Figure A6.29.Focus of Discussion with the IMF1

(Number of economies)

1 1+2 = A focus; 3 = Neutral; and 4+5 = Not a focus.

2 One authority respondent chose the “don’t know” option.

On staff analysis

47. The authorities’ agreement with staff analysis coincided with a similar perception from staff in only 6 out of 15 economies (Figure 6.30.).21 This supports the findings from the overall survey, namely, staff have a somewhat guarded view of the authorities’ agreement with their analysis. For example, staff saw the authorities disagreeing on important details of their analysis in 5 out of 15 cases where the authorities said they agreed.

Figure A6.30.Authorities’ Agreement with Staff Analysis1

(Number of economies)

1 b+c = Broadly agree; d = Disagreements on important details; a+e = No dialogue or not clear/inconsistent; and f+g = Major disagreements.

On restricting discussions

48. Staff’s perception of the extent to which the authorities restricted the discussion of sensitive issues differed considerably from the authorities’ own assessment (Figure 6.31.). Staff consider that the authorities restricted discussions in only one of the seven cases where the authorities reported having done so. There are only six countries where, according to the authorities, the discussion of issues was unrestricted. Relative to the findings from the overall survey, the countrymatched data indicate a greater perception gap of the staff from the authorities’ own assessment.

Figure A6.31.Off-the-Table Issues1

(In percent)

1 1+2 = Restricted; 3 = Neutral; and 4+5 = Unrestricted.

On impact on major decisions

49. Concerning the impact of IMF advice on major exchange rate policy decisions, there is agreement between the staff’s perception and the authorities’ stated assessment in 7 out of 15 countries (Figure 6.32.). Also, staff tended to overestimate (relative to the authorities’ assessment) their impact by a large margin: 6 countries are found below the diagonal elements of the matrix, with only 2 countries found above. Relative to the overall survey findings, however, the “representative” staff working on a large EME tends to have a less sanguine assessment of their impact. For example, they see the IMF’s impact as instrumental in 27 percent of the cases (5 out of 15 countries), whereas 39 percent of the surveyed staff in the overall survey considered their impact to be instrumental for large EMEs. From the authorities’ side, major decisions were taken in more than 90 percent of the cases (14 out of 15), and among them 60 percent (8 out of 14) saw the IMF’s impact as marginal.

Figure A6.32.The IMF’s Impact on Major Exchange Rate Decisions1

(In percent)

1 a = Instrumental; b = Marginal; c = No impact; d = Little/no discussions; and f = No major decisions taken.

On the role of the IMF

50. In terms of three roles the IMF can play vis-à-vis member countries, staff’s perception again differs widely from the authorities’ assessment (Figure A6.33.). Both agreed on the extent to which the IMF plays the respective roles in less than half the cases. In particular, they agreed that the IMF plays these roles “about right” only in 4 out of 13 countries for the “confidential advisor” role; 7 out of 15 countries for the “sounding board” role; and 4 out of 10 countries for the “consensus builder” role.

Figure A6.33.The Role of the IMF1

(Number of economies)

1 1+2 = Too little; 3 = About right; and 4+5 = Too much. Two country respondents and three groups of staff chose the “don’t know” option.

2 Two country respondents chose the “don’t know” option.

3 Two country respondents and three groups of staff chose the “don’t know” option.

On overall quality changes

51. Regarding the change in the overall quality of IMF analysis and assessment, the perceptions of staff and the authorities matched in 40 percent of the cases over the 1999–2005 period. The country-matched survey data reinforce the overall survey finding that, for large EMEs, staff see improvement in more cases than the authorities (13 out of 15 versus 4 out of 15; Figure A6.34).

Figure A6.34.Perceived Change in the Overall Quality of IMF Analysis1

(Number of economies)

1 1+2 = Better; 3 = About the same; and 4+5 = Worse.

Annex A6.1.
Features of the Survey Responses

As noted in the text, the response rates from the two populations were high enough to yield statistically meaningful inferences. Specifically, about 60 percent of the authorities responded to the survey. Across the four country groups, the response rate ranged from just below 50 percent to close to 90 percent (Table A6.1a). All four regional central banks also responded to the survey.

Table A6.1a.Survey of Authorities’ Responses(Number of economies unless otherwise noted)
Economy TypeEconomies SurveyedRespondentsResponse Rate1
Major advanced29889
Other advanced211886
Large EMEs201680
Other EMEs/DCs1427251
Total19211459

In percent

Includes the European Central Bank as well as the European Commission

In percent

Includes the European Central Bank as well as the European Commission

The response rate by staff was also significant, but somewhat lower at 47 percent, with about half of the staff who were targeted in area departments taking the survey. Fewer than 10 percent of the respondents did not select a country (Table A6.1b).

Table A6.1b.Staff Survey Responses(Number of staff unless otherwise noted)
By Departmental Affiliation
Staff surveyedRespondedResponse rate1By Economy Selection
Area43623053Major advanced12
Functional35514039Other advanced16
Total79137047Large EMEs81
Other EMEs/DCs187
No country identified22
No involvement52
Total370

In percent.

In percent.

About 90 percent of the country response submissions came from central banks, although this ratio was smaller for “Major advanced” economies. A significant fraction (two-thirds) of the respondents were above the level of senior advisor or department head, with a significant majority also indicating they were directly involved in policy discussion, both internally and with IMF staff. Close to 70 percent of all respondents had been involved in discussions throughout the period. Except for institutional affiliation, the respondents’ characteristics were broadly similar across country groups.

Among the respondents to the staff survey, more than 90 percent selected a country, and more than a third of these respondents reported having led missions. The staff respondents tended to have recent involvement in the selected countries, typically for about two to three years.

In total, the staff respondents selected 115 countries, of which 80 had corresponding survey responses from the authorities (Table A6.1c). This overlap serves as a quality control device in drawing inferences from some questions, particularly for the group of “Large EMEs” where the overlap was particularly significant.

Table A6.1c.Authorities’ Response and Staff’s Country Selections(Number of economies)
Staff
SelectedNot selectedTotal
Authorities
Responded8034114
Of which:
Major advanced448
Other advanced9918
Large EMEs15116
Other EMEs/DCs522072
No response54378
Of which:
Major advanced101
Other advanced043
Large EMEs404
Other EMEs/DCs303970
Total11577192
Annex A6.2List of Economies Surveyed
Large AdvancedOther AdvancedLarge EmergingOther Emerging and Developing
CanadaAustraliaArgentinaAfghanistan,
Euro areaAustriaBangladeshAlbania
FranceBelgiumBrazilAlgeria
GermanyCyprusChinaAngola
ItalyDenmarkColombia
JapanFinlandEgyptArmenia
United KingdomGreeceIndiaAruba
United StatesHong Kong SARIndonesiaAzerbaijan
IcelandIran, Islamic Republic ofBahamas, The
IrelandMalaysiaBahrain
IsraelMexicoBarbados
KoreaPakistanBelarus
LuxembourgPhilippinesBelize
NetherlandsPolandBenin
New ZealandRussiaBhutan
NorwaySaudi ArabiaBolivia
PortugalSouth AfricaBosnia and Herzeg
SingaporeThailandBotswana
SpainTurkeyBrunei Darussalam
SwedenUkraineBulgaria
SwitzerlandBurkina Faso
Burundi
Cambodia
Cameroon
Cape Verde
Central African Economic and Monetary Community
Central African Republic
Chad
Chile
Comoros
Congo, Democratic Republic of the
Congo, Republic of
Costa Rica
Cote d’Ivoire
Croatia
Czech Republic
Djibouti
Dominica
Dominican Republic
Eastern Caribbean Currency Union
Ecuador
El Salvador
Equatorial Guinea
Eritrea
Estonia
Ethiopia
Fiji
Gabon
Gambia, The
Georgia
Ghana
Grenada
Guatemala
Guinea
Guinea-Bissau
Guyana
Haiti
Honduras
Hungary
Iraq
Jamaica
Jordan
Kazakhstan
Kenya
Kiribati
Kuwait
Kyrgyz Republic
Lao People’s Democratic Republic
Latvia
Lebanon
Lesotho
Liberia
Libyan Arab Jamahiriya
Lithuania
Macao SAR
Macedonia, FYR
Madagascar
Malawi
Maldives
Mali
Malta
Marshall Islands
Mauritania
Mauritius
Micronesia, Federated States of
Moldova
Mongolia
Morocco
Mozambique
Myanmar
Namibia
Nepal
Netherlands Antilles
Nicaragua
Niger
Nigeria
Oman
Palau
Panama
Papua New Guinea
Paraguay
Peru
Qatar
Romania
Rwanda
Samoa
São Tomé and Príncipe
Senegal
Serbia
Seychelles
Sierra Leone
Slovak Republic
Slovenia
Solomon Islands
Sri Lanka
St. Kitts and Nevis
St. Lucia
St. Vincent and the Grenadines
Sudan
Suriname
Swaziland
Syrian Arab Republic
Tajikistan
Tanzania
Timor-Leste
Togo
Tonga
Trinidad and Tobago
Tunisia
Turkmenistan
Uganda
United Arab Emirates
Uruguay
Uzbekistan
Vanuatu
Venezuela, República Bolivariana de
Vietnam
West African Economic and Monetary Union
Yemen, Republic of
Zambia
Zimbabwe
Annex A6.3Survey Questions for National Authorities
Q1.Please rate the degree to which the following exchange-rate-related issues have been a focus of internal policy attention in your country/economy at some point during 1999–2005.
1—2—3—4—5—0
1—“Significant focus,” 5—“Not at all a focus,” 0—“Don’t know”
(a) Exchange rate regime choice (e.g., peg, monetary union, floating, choice of nominal anchor), including contingency planning
(b) Management of the regime(s) in place over the period (e.g., inflation targeting, intervention, international reserves levels, currency auction systems)
(c) Exchange rate level (e.g., competitiveness, sustainability, impact on inflation)
(d) Effects of other countries’ policies on your country; or of your country’s policies on others (e.g., regional spillovers, global imbalances, contagion)
(e) Financial stability considerations (e.g., currency mismatches and other potential balance sheet vulnerabilities)
Q2.For the same set of exchange-rate-related issues asked in the previous question, please rate the level of focus on these issues in discussions with IMF staff during 1999–2005.
1—2—3—4—5—0
1—“Significant focus,” 5—“Not at all a focus,” 0—“Don’t know”
(a) Exchange rate regime choice (e.g., peg, monetary union, floating, choice of nominal anchor), including contingency planning
(b) Management of the regime(s) in place over the period (e.g., inflation targeting, intervention, international reserves levels, currency auction systems)
(c) Exchange rate level (e.g., competitiveness, sustainability, impact on inflation)
(d) Effects of other countries’ policies on your country; or of your country’s policies on others (e.g., regional spillovers, global imbalances, contagion)
(e) Financial stability considerations (e.g., currency mismatches and other potential balance sheet vulnerabilities)
Q2a.If you answered “1” or “2” to any of the statements listed in the previous question, please indicate who took the initiative to raise the respective issue (please select one):
1—“Raised by staff,” 2—“Jointly raised,” 3—“Raised by the authorities,” 0—“Don’t know”
(a) Exchange rate regime choice
(b) Management of the regime(s) in place
(c) Exchange rate level
(d) Effects of other countries’ policies on your country; or of your country’s policies on others
(e) Financial stability considerations
Q3.Please indicate the authorities’ overall level of agreement with the IMF staff’s analysis and assessment in the policy area(s) that have been a focus of internal policy attention in your country/economy at some point during 1999–2005 (as you identified in Q1). Please select the ONE statement that best describes your opinion.
The authorities in my country …
(a)… did not have any dialogue of substance on exchange-rate-related matters during the period
(b) … agreed with all aspects of the IMF staff’s analysis and assessment;
(c) … broadly agreed, except for minor differences in emphasis, detail, or timing
(d) … agreed on some aspects, but disagreed on important details (e.g., emphasis, timing, or political feasibility)
(e) … found the IMF staff’s analysis and assessment not clear enough or too inconsistent to form a clear view
(f) … disagreed on important aspects of analysis and assessment and did not follow the staff’s recommendations
(g) … disagreed on important aspects of analysis and assessment, but felt obliged to follow the staff’s recommendations
If answer under Q3 was d, e, f or g, please answer Q3a; otherwise skip to Q4.
Q3a.You indicated in the previous question that there was a lack of agreement or clarity with the IMF analysis. In the question below, please indicate in which areas there was lack of agreement or clarity (check all that apply):
(a) Exchange rate regime choice
(b) Management of the regime(s) in place
(c) Exchange rate level
(d) Effects of other countries’ policies on your country; or of your country’s policies on others
(e) Financial stability considerations
(f) None of the above
Q4.Please rate your level of agreement with each of the following statements about the nature of your interaction with the IMF on exchange-rate-related policy issues:
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know”
(a) IMF staff conducted substantive, two-way policy discussions with the authorities in your country
(b) Discussions with IMF staff were held with appropriate frequency
(c) Discussions with IMF staff had the right balance between informality, confidentiality, and requirements of reporting to the Executive Board
(d) The IMF was the principal source of external analysis and assessment
(e) The authorities at times excluded certain sensitive policy issues (e.g., foreign exchange market intervention, choice of exchange rate regime from substantive discussions with IMF staff
(f) The authorities excluded or restrained consideration of certain issues because of concerns about possible dissemination of information, including to the IMF Executive Board
(g) The IMF mission teams approached discussions with candor and were willing to raise politically sensitive issues
(h) The IMF mission teams approached discussions in a respectful and open-minded way
(i) Policy discussions with IMF staff were fully reflected in documents subsequently sent to the IMF Executive Board
(j) Considerations at the level of the IMF’s Executive Board provided an important input into the development of policy
Q5.Please rate your level of agreement with each of the following statements about the substance and usefulness of the IMF’s exchange-rate-related discussions with your country:
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know”
IMF staff could usefully have …
(a)… provided its analysis and assessment in a more timely fashion
(b)… focused more on country-specific factors and constraints
(c)… provided a better analytical/empirical underpinning for its views on exchange rate issues
(d) … developed broad points of view more concretely into issues of implementation
(e) … been more careful in advocating certain policy actions as urgent
(f)… better considered the costs and benefits of policy options
(g)… brought in more multilateral or regional perspectives
(h)… better integrated exchange-rate-related advice with advice in other policy areas (e.g., monetary policy, financial stability)
(i) … explored more fully the sectoral balance sheet implications of exchange rate changes
(j) … engaged in more informal interaction with the authorities (e.g. through workshops)
(k)… formulated its views in a clearer and less nuanced fashion
(l)… provided written documentation of its assessments (e.g., for the authorities to study and share with others)
Q6.To the extent that the authorities in your country have taken major decisions related to exchange rate policy in 1999–2005, please identify how the contribution made by the Fund can be best characterized. Please select the ONE option that best reflects your views.
(a) Fund assessments were instrumental in helping shape policy decisions
(b) Discussions with Fund staff were helpful at the margin
(c) Fund assessments had no impact on the decision taken
(d) Policy decisions were taken with little or no discussion/interaction with the Fund
(e) Discussions with Fund were unhelpful or counterproductive
(f) No major decisions were taken during this period
(g) Don’t know
Q7.To the extent that a major policy decision was being considered or taken in 1999–2005, please identify the extent to which the Fund provided adequate followup support (e.g., through the provision of technical assistance). Please select the ONE option that best reflects your views.
(a) Fund support was instrumental for implementation
(b) Practical assistance by the Fund was helpful at the margin
(c) Fund support had no influence on implementation
(d) Fund support for implementation was unhelpful or counterproductive (e.g., insufficient understanding of operational realities)
(e) Given my country’s own technical capacity, no follow-up support was requested or received
(f) No major decisions were implemented during this period
(g) Don’t know
Q8.The IMF is sometimes characterized as having various roles. Please rate the degree to which, in your opinion, the IMF has played the roles specified below on matters related to members’ exchange rate policies:
1—2—3—4—5—0
1—“Too little (missed opportunities),” 3—“About right,” 5—“Too much (role is overplayed),” 0—“Don’t know”
In the context of your own country:.
(a)“Confidential advisor” to the authorities in your country
(b)“Sounding board”/intellectual partner for discussing the authorities’ policy views
(c)“Consensus-builder” among domestic policymakers
For the Fund membership as a whole:.
(d)“Ruthless truth-teller” to the international community
(e)“Broker” for international policy coordination (e.g., resolution of global imbalances)
(f) “Provider of credibility” (e.g., in capital markets, to the donor community) through assessment of national policies
(g)“Lender” in the event of possible adverse contingencies
Q9.Please rate the level of your agreement with each of the following options regarding the approach IMF staff used to assess the choice or suitability of the exchange rate regime in your country:
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know”
(a) Tailored to country-specific circumstances (e.g., extent of financial integration in global/regional capital markets, degree of openness)
(b) Did not favor any particular exchange rate regime (or combination of exchange rate and monetary regimes) over others
(c) Sufficient attention was paid to intermediate regimes (i.e., it did not present the options available as only a flexible exchange rate regime or a hard peg)
(d) Sufficient attention was given to structural considerations (as opposed to cyclical ones, e.g., impact of any exchange rate misalignments on inflation)
Q10.Please indicate which of the following options best describes your view on the methodologies the IMF used to assess the equilibrium level of exchange rates (e.g., macroeconomic balance, fundamental or behavioral exchange rate equilibria). (Please select only one.)
(a) Adequate and useful
(b) Could have done more, but still useful
(c) Interesting, but not of much practical use
(d) Inadequate
(e) Other (please specify:———————)
(f) Don’t know
Q11.Please indicate which of the following options best describes your view on the methodologies the IMF used to assess external competitiveness (e.g., real effective exchange rate indices, export market shares). (Please select only one.)
(a) Adequate and useful
(b) Could have done more, but still useful
(c) Interesting, but not of much practical use
(d) Inadequate
(e) Other (please specify:———————)
(f) Don’t know
For Questions 12–14, please indicate your opinion on particular aspects of the Fund’s coverage of exchange rate issues by selecting the ONE statement that best characterizes your view, or check “Don’t know” or “Doesn’t apply.”
Q12.In the discussion of policy options for your country, analysis based on the experience of other countries was…(please select only one).
(a)… very useful for decision making
(b) … somewhat useful for decision making
(c) … not useful for decision making
(d) … not provided by the IMF staff
(e) Doesn’t apply
(f) Don’t know
Q13.Global or regional spillovers affecting your country’s exchange-rate-related developments were…(please select only one).
(a)… identified by the IMF staff and very well integrated into policy discussions
(b) … identified by the IMF staff and reasonably well integrated into policy discussions
(c) … identified by the IMF staff but only partially integrated into policy discussions
(d)… rarely identified by the IMF staff
(e) Doesn’t apply
(f) Don’t know
Q14.The global and regional impact of your country’s exchange rate policy was…(please select only one).
(a)… identified by the IMF staff and very well integrated into policy discussions
(b) … identified by the IMF staff and reasonably well integrated into policy discussions
(c) … identified by the IMF staff but only partially integrated into policy discussions
(d)… rarely identified by the IMF staff
(e) Doesn’t apply
(f) Don’t know
Q15.In summary, how do you rate the overall quality of IMF staff’s analysis and assessment at the end of the evaluation period (in 2005) in comparison to a few years ago (from about 1999)?
(a) Significantly better
(b) Better
(c) About the same
(d) Worse
(e) Significantly worse
(f) Don’t know
Annex A6.4Survey Questions for Staff
Q1.Please rate the degree to which the following exchange-rate-related issues have been a focus of attention in staff’s discussions with the authorities in [country name, as specified] at some point during your involvement.
1—2—3—4—5—0
1—“Significant focus,” 5—“Not at all a focus,” 0—“Don’t know”
For [country name, as specified …] the following issues have been a focus:
(a) Exchange rate regime choice (e.g., peg, monetary union, floating, choice of nominal anchor), including contingency planning;
(b) Management of the regime(s) in place over the period (e.g., inflation targeting, intervention, international reserves levels, currency auction systems);
(c) Exchange rate level (e.g., competitiveness, sustainability, impact on inflation);
(d) Effects of other countries’ policies on the country/economy under consideration; or of that country’s policies on others (e.g., regional spillovers, global imbalances, contagion);
(e) Financial stability considerations (e.g., currency mismatches and other potential balance sheet vulnerabilities).
Q1a.If answer under any of the options in Q1 is “1” or “2,” please indicate who took the initiative to raise the respective issue:
1—“Raised by staff,” 2—“Jointly raised,” 3—“Raised by the authorities,” 0—“Difficult to say/Don’t know”
(a) Exchange rate regime choice;
(b) Management of the regime(s) in place;
(c) Exchange rate level;
(d) Effects of other countries’ policies on the country/economy under consideration; or of that country’s policies on others;
(e) Financial stability considerations.
Q2.Please indicate the authorities’ overall level of agreement with the IMF staff’s analysis and assessment in the policy area(s) that have been a focus of policy attention at some point during your involvement (as identified under Q1). Please select the statement that best describes your opinion for the country/economy selected above.
The authorities in [country name, as specified …]
(a)… did not have any dialogue of substance on exchange-rate-related matters with IMF staff during the period;
(b) … agreed with all aspects of the IMF staff’s analysis and assessment;
(c) … broadly agreed, except for minor differences in emphasis, detail or timing;
(d) … agreed on some aspects, but disagreed on important details (e.g., emphasis, timing or political feasibility);
(e) …disagreed with the IMF staff’s analysis and assessment, but their own position was not clear enough or too inconsistent to judge why;
(f)… disagreed on important aspects of analysis and assessment and did not follow the staff’s recommendations;
(g) … disagreed on important aspects of analysis and assessment, but nonetheless followed the staff’s recommendations.
Q2a.If answer under Q2 is (d), (e), (f) or (g), please indicate in which areas there was lack of agreement or clarity (check all that apply):
(a) Exchange rate regime choice;
(b) Management of the regime(s) in place;
(c) Exchange rate level;
(d) Effects of other countries’ policies on your country; or of your country’s policies on others;
(e) Financial stability considerations.
Q3.Please rate your level of agreement with each of the following statements about the nature of staff interaction with the authorities in [country name, as specified] on exchange-rate-related policy issues:
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know/Doesn’t apply”
In [country name, as specified…]
(a)… IMF staff conducted substantive, two-way policy discussions with the authorities;
(b) … discussions with the authorities were held with appropriate frequency;
(c) … discussions with the authorities had the right balance between informality, confidentiality, and requirements of reporting to the Executive Board;
(d) … the IMF was the principal source of external analysis and assessment for the authorities;
(e) … the authorities at times excluded certain sensitive policy issues (e.g., foreign exchange market intervention, choice of exchange rate regime) from substantive discussions with IMF staff;
(f)… the authorities excluded or restrained consideration of certain issues because of concerns about possible dissemination of information, including to the IMF Executive Board;
(g) … the IMF mission team approached discussions with candor and was willing to raise politically sensitive issues;
(h) … the IMF mission team approached discussions in a respectful and open-minded way;
(i) … policy discussions between the authorities and staff were fully reflected in documents subsequently sent to the IMF Executive Board;
(j)… considerations at the level of the IMF’s Executive Board provided an important input into the development of policy advice (e.g., changes to the policy line or to the urgency of proposed actions at Board level).
Q4.Please rate your level of agreement with each of the following statements on how the substance and usefulness of IMF staff’s exchange- rate-related discussions with the authorities in [country name, as specified] could realistically have been improved?
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know”
IMF staff could have …
(a)… provided its analysis and assessment in a more timely fashion;
(b)… focused more on country-specific factors and constraints;
(c)… provided a better analytical/empirical underpinning for its views on exchange rate issues;
(d) … developed broad points of view more concretely into issues of implementation;
(e) … been more careful in advocating certain policy actions as urgent;
(f)… better considered the costs and benefits of policy options;
(g)… brought in more multilateral or regional perspectives (e.g., regional spillovers, global imbalances, contagion);
(h) … better integrated exchange-rate-related advice with advice in other policy areas (e.g., monetary policy, financial stability);
(i) … explored more fully the sectoral balance sheet implications of exchange rate changes;
(j) … engaged in more informal interaction with the authorities (e.g., through workshops);
(k)… formulated its views in a clearer and less nuanced fashion;
(l)… provided better written documentation of its assessments (e.g., for the authorities to study and share with others).
[If answer under option (g) was “1” or “2,” please continue with Q4a; otherwise skip to Q5]
Q4a.Why was it that staff’s treatment of multilateral or regional perspectives in the context of [country name, as specified] left room for improvement? Please select all that apply:
(a) There is always room for improvement;
(b) Coverage of these issues was pushed back by the authorities;
(c) Staff did not raise these issues with the authorities;
(d) The analytical framework for analysis of these issues was lacking;
(e) Multilateral surveillance tools (such as the WEO and GFSR) didn’t provide relevant inputs.
Q5.To the extent that the authorities in [country name, as specified] have taken major decisions related to exchange rate policy during the period of your involvement, please identify how the contribution made by the Fund can be best characterized. Please select the option that best reflects your views.
For [country name, as specified …]:
(a) Fund assessments were instrumental in helping shape policy decisions;
(b) Discussions between Fund staff and the authorities were helpful at the margin;
(c) Fund assessments had no impact on the decision taken by the authorities;
(d) Policy decisions were taken with little or no discussion/interaction between Fund staff and the authorities;
(e) Discussions between Fund staff and the authorities appear to have been unhelpful or counterproductive;
(f) No major decisions were taken during this period;
(g) Don’t know.
Q6.To the extent that a major policy decision was being considered or taken by the authorities in [country name, as specified] during your period of involvement, please identify the extent to which the Fund provided adequate follow-up support (e.g., through the provision of technical assistance). Please select the option that best reflects your views.
(a) Fund support was instrumental for implementation;
(b) Practical assistance by the Fund was helpful at the margin;
(c) Fund support had no influence on implementation;
(d) Fund support for implementation turned out to be unhelpful or counterproductive (e.g., insufficient understanding of operational realities);
(e) Given the country’s own technical capacity, no follow-up support was requested or received;
(f) No major decisions were implemented during this period;
(g) Don’t know.
Q7.Do you believe that IMF staff advice was not acted upon by the authorities in [country name, as specified …]? If so, what accounts for this lack of responsiveness by the authorities? Please check all that apply:
(a) Fund “leverage” outside of program relationships is too limited;
(b) Political realities didn’t allow for implementation;
(c) Fund advice wasn’t time sensitive (e.g., no immediate need to respond);
(d) Fund staff didn’t have access to policymakers at the appropriate level;
(e) The staff work was restricted by management or other departments;
(f) The analysis provided by staff failed to convince the authorities;
(g) The authorities’ own capabilities made it difficult for staff to add value;
(h) With hindsight, the authorities’ judgment appears to have been validated;
(i) Other [please specify: …];
(j) Don’t know;
(k) Doesn’t apply (i.e., the authorities did act on the advice).
Q8.The IMF is sometimes characterized as having various roles. Please rate the degree to which, in your opinion, the IMF has played the roles specified below on matters related to members’ exchange rate policies:
1—2—3—4—5—0
1—“Too little (missed opportunities),” 3—“About right,” 5—“Too much (role is overplayed),” 0—“Don’t know”
In the context of the countries you have worked on [or country name, as specified]:
(a)“Confidential advisor” to the authorities;
(b)“Sounding board”/intellectual partner for discussing the authorities’ policy views;
(c)“Consensus-builder” among domestic policymakers; For the Fund membership as a whole:
(d)“Ruthless truth-teller” to the international community;
(e)“Broker” for international policy coordination (e.g., resolution of global imbalances);
(f) “Provider of credibility” (e.g., in capital markets, to the donor community) through assessment of national policies;
(g)“Lender” in the event of possible adverse contingencies.
Q9.Clear and candid treatment of exchange rate issues is a challenge. To what extent have each of the following possible explanations been factors in making this so?
1—2—3—4—5—0
1—“Very significant,” 5—“Insignificant,” 0—“Don’t know”
(a) It was difficult to separate exchange-rate-related issues from countries’ overall macroeconomic policies and developments;
(b) Time and resources constrained the amount of attention paid to exchange-rate-related issues;
(c) There was a lack of adequate and readily available analytical tools;
(d) The internal review process tended to produce more cautious language and/or diluted coverage in staff reports;
(e) The need to preserve close relationships with country authorities tended to produce more cautious language and/or diluted coverage in staff reports;
(f) The expectation of publication of staff reports tended to produce more cautious language and/or diluted coverage in staff reports (e.g., because of possible adverse market reactions);
(g) For confidentiality reasons, certain information could not be passed on to the Executive Board;
(h) Management limited the range of issues that the staff could freely discuss with national authorities;
(i) National authorities took certain issues “off the table.”
Q10.On the basis of your experience between 1999–2005, please indicate your opinion on the quality of Fund analysis in the following areas:
1—2—3—4—5—0
1—“No scope for further improvement,” 5—“Significant room for improvement,” 0—“Don’t know”
(a) Assessment of exchange rate regime choice and suitability;
(b) Management of the regime(s) in place;
(c) Assessment of exchange rate levels and competitiveness;
(d) Assessment of policy spillovers;
(e) Financial stability considerations;
(f) Debt sustainability analysis;
(g) Capital account developments;
(h) Integration of bilateral and regional/multilateral surveillance.
Q11.On the basis of your experience between 1999–2005, in which of the following areas do you see significant room for improvement insofar as it would raise the overall quality of exchange-rate-related analysis (including subsequent reporting to the Executive Board)? Please check all that apply:
(a) Analytical methods for assessing exchange rate levels and misalignments;
(b) Analytical methods for assessing regime choice and suitability;
(c) Analytical methods for assessing spillovers across countries, balance sheet effects and related vulnerabilities;
(d) Better use of information from other sources (e.g., BIS, OECD);
(e) Better availability of data (e.g., on intervention activities);
(f) Greater backing for staff to raise contentious or confidential issues with the authorities;
(g) Greater backing for staff to share information and analysis at the Board level;
(h) Other [please specify: …];
(i) Don’t know.
[If answer includes option (e) and respondent has selected an individual country at the start of the survey, follow up with Q11a; otherwise skip to Q12].
Q11a.Please rate the level of your agreement with the following statements about the availability of data for IMF exchange rate analysis in [country name, as specified].
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know”
For [country name, as specified], …
(a)…, the availability and quality of data has impaired staff’s ability to conduct exchange rate analysis and provide related advice;
(b) …, Fund staff tended to have greater access to critical data on exchange rate matters than the public;
(c)…, the authorities were technically not able to provide staff with some critical data needed for exchange rate analysis and related advice;
(d) …, the authorities were unwilling to share some critical data/information needed for exchange rate analysis and related advice;
(e) …, staff had to regularly cross-check the authorities’ data with information from other sources.
Q12.Please rate the degree to which each of the following has been a source of help in your work on exchange rate policy issues:
1—2—3—4—5—0
1—“Significant source of help,” 5—“Not at all a source of help,” 0—“Don’t know/Doesn’t apply”
(a)1977 Surveillance Decision;
(b) Board decisions/discussions on exchange rate issues;
(c) Board discussions on individual countries;
(d) Internal surveillance guidance notes;
(e) Direct guidance by management;
(f) Analysis/research provided by (own or other) area department(s);
(g) Analysis/research provided by RES;
(h) Analysis/research provided by INS;
(i) Analysis/research provided by STA;
(j) Analysis/research provided by PDR;
(k) Analysis/research provided by MFD/ICM;
(l) Analysis provided by the WEO/GFSR.
Q13.Please rate the level of your agreement with each of the following statements regarding the approach IMF staff used to assess the choice or suitability of exchange rate regimes:
1—2—3—4—5—0
1—“Strongly agree,” 5—“Strongly disagree,” 0—“Don’t know/Doesn’t apply”
(a) Tailored to country-specific circumstances (e.g., extent of financial integration in global/regional capital markets, degree of openness);
(b) Did not favor any particular exchange rate regime (or combination of exchange rate and monetary regimes) over others;
(c) Sufficient attention was paid to intermediate regimes (i.e., staff did not present the options available as only a flexible exchange rate regime or a hard peg);
(d) Sufficient attention was given to structural considerations (as opposed to cyclical ones, e.g., impact of any exchange rate misalignments on inflation).
Q14.For those countries that are covered by the CGER exercise, how do you rate the overall usefulness of this exercise for staff’s country work. Please use the following scale and choose “Don’t know/Doesn’t apply” in case you have never worked on countries covered by CGER:
1—2—3—4—5—0
1—“Very useful,” 5—“Not at all useful,” 0—“Don’t know/Doesn’t apply”
Q15.In summary, how do you rate the (a) overall quality and (b) impact/traction of IMF staff’s analysis and assessment at the end of the evaluation period (in 2005) in comparison to a few years ago (from about 1999)?
1—2—3—4—5—0
1—“Significantly better,” 3—“About the same,” 5—“Significantly worse,” 0—“Don’t know”
(a) The overall quality of staff’s analysis and assessment in 2005 was…;
(b) The overall impact/traction of staff’s analysis and assessment in 2005 was….
1Fusion Analytics L.L.C., Washington, D.C.
2The European Central Bank, the Eastern Caribbean Central Bank, the Central Bank of West African States, and the Central Bank of Central African States. The survey responses of countries belonging to a regional monetary union and that of the central bank were separately included in the results and tabulations presented below.
3“Major advanced” and “Other advanced” follow the World Economic Outlook classification. All other countries are grouped as “Large EMEs” or “Other EMEs/DCs” on the basis of PPP-adjusted GDP for 2004, with $250 billion as the cut off.
4The response rate for the authorities’ survey was high across all four country groups (at more than 80 percent among the advanced and large emerging market economies, and 50 percent for other emerging market and developing countries), arguing against any significant self-selection bias. Responses in both surveys were checked for plausibility by combining answers across questions, yielding mutually consistent results (see Tables A6.2 through A6.5 and Figures A6.29 through A6.34 for examples).
5See Annexes A6.3 and A6.4 for the list of survey questions.
6For example, a “1” corresponds to “strongly agree,” while a “5” corresponds to a “strongly disagree.” A “don’t know/does not apply” option is also given.
7Unless otherwise noted, the figures derived from rating questions depict the percentage share of those selecting “1” or “2,” typically corresponding to “agreed” or “strongly agreed.” Also unless otherwise noted, the percentage shares in the figures exclude those who selected the “don’t know/does not apply” option.
8In a follow-up question, both authorities and staff were asked to identify who had taken the initiative to raise issues in the five policy areas. Staff respondents generally saw themselves as the more proactive of the two in the dialogue on exchange rate issues.
9In total, there were 13 countries that reported concerns on leaks.
10In a follow-up question, the staff respondents identified the lack of an analytical framework and of relevant inputs from the multilateral surveillance products (e.g., the World Economic Outlook and the Global Financial Stability Report) as important reasons for the poor integration of multilateral perspectives into exchange-rate-related analysis and policy advice at the country level.
11The percentages reported in Figures A6.13 and A6.14 include the answers of those respondents who selected the “don’t know” option.
12Concerning the role of the IMF as consensus builder, 12 percent of the country respondents and 11 percent of the staff respondents selected the “don’t know” option.
13Overall, between 4 percent and 8 percent of the authorities selected the “don’t know” option when it came to the IMF’s roles vis-à-vis the entire membership, percentages that rise to 7 percent and 17 percent (“broker” for international policy coordination) for staff.
14Except for the final question, the remainder of the survey proceeds separately for the authorities and staff
15The share falls from 70 percent to 60 percent if we include those who selected the “don’t know” option (who amount to about 20 percent of the total).
16For details on the CGER methodology, see Background Document 3.
17The seniority of the respondent did not appear to affect the conclusions drawn in this section, either for the authorities or for IMF staff.
18For each exchange rate topic, this expression is meant to capture the difference in emphasis given to that topic by the authorities in their own internal discussions to their discussions with the IMF, as reported by country respondents.
19Regression analysis suggests evidence of a relationship between “gaps in coverage” and IMF impact for three of the five policy areas (regime choice, regime management, and spillovers) covered in Figure A6.34, when controlling for whether or not policy issues have been taken “off the table” by the authorities (Figure 6.7.). Given the nature of the control variable, causality may therefore run from the “gaps in coverage” to impact, suggesting that efforts to address these gaps are likely to increase the impact of IMF advice.
20When this failed to break the tie, we took as the staff response the option that most closely matched the authorities’ response.
21Again counting diagonally from the north-west corner.

    Other Resources Citing This Publication