IEO Evaluation of Exchange Rate Policy

Background Document 4: Characteristics of IMF Exchange Rate Surveillance: A Full Review of Country Documents

International Monetary Fund. Independent Evaluation Office
Published Date:
August 2007
  • ShareShare
Show Summary Details

1. This background document describes how exchange rate issues were treated in the last two Article IV consultations of the 1999–2005 period.1 For this purpose, the IEO reviewed the following Article IV consultation-related documents:

  • Staff reports; 2

  • Those selected issues papers, finalized between 2001 and mid-2006, that primarily addressed exchange-rate-related issues;

  • Briefing papers and back-to-office reports;

  • In the case of program countries, staff reports, briefing papers, and back-to-office reports for use of Fund resources missions that fall in the same period; and

  • Other documents, including comments received from departments and management on draft papers.

2. In addition to the documents on member countries, the IEO also reviewed the corresponding documents for Aruba—Kingdom of the Netherlands, Hong Kong SAR, Macao SAR, Netherlands Antilles, and West Bank and Gaza, as well as for the Central African Monetary and Economic Union (CEMAC), Eastern Caribbean Currency Union (ECCU), euro area, and West African Economic and Monetary Union (WAEMU). All in all, the assessment included a total of 191 economies.3 The data sets (constructed as the responses to standard questions) compiled from the review are attached as annexes to this background document.

Coverage of Exchange Rate Issues

3. Every Article IV staff report and mission brief mentioned exchange rates, but the extent of coverage varied (Figure A4.1). Consistent with the standard template, the staff reports always described the exchange rate arrangement in a policy discussion section as well as in an appendix; they often devoted one or more paragraphs to the exchange rate in sections describing economic developments and outlook, as well as in the staff appraisal. The number of paragraphs in staff reports referring to exchange rate issues ranged from 3 percent to 50 percent of total. In contrast to the staff reports and mission briefs, the back-to-office summaries (which were sent to management) show an even greater variation in their treatment of exchange rate issues, because they are less template-driven and tend to highlight those points that the mission chief considers particularly important for management. Exchange rate issues were absent from the back-to-office summaries in 86 cases. Finally, IMF staff prepared at least one selected issues paper on exchange rate issues for 132 (out of the 191) economies during 2001 to mid-2006.

Figure A4.1.Exchange Rate Coverage Across Countries

(In percent)

Content of Exchange Rate Coverage

4. In more than 100 cases,4 competitiveness considerations prompted a discussion on exchange rate levels. For most of the IMF membership, the staff exchange rate assessments did not focus on spillover effects, global capital markets, or global imbalances (Figure A4.2). Especially for small countries with limited integration into global capital markets, the discussion of exchange rates limited itself to domestic considerations, such as the implications of the prevailing exchange rate regime for fiscal policy or the impact of exchange rate movements on domestic inflation and competitiveness.

Figure A4.2.Elements in Staff Exchange Rate Assessment

(Number of cases)

5. In a majority of the country cases, the orientation of the exchange rate discussion was predominantly forward looking. For example, such discussion included projections or scenarios of possible exchange rate developments, with associated upside and down-side risks to the economy. Only in two cases was the discussion exclusively backward looking. The discussion was frequently linked with other elements of macroeconomic policy, in many cases exploring either the likely implications of various domestic developments for the exchange rate or pointing out the implications of an exogenous exchange rate development for fiscal and monetary policy.

6. Similar patterns were observed in the selected issues papers. A large majority (about 70 percent) contained conceptual analysis that outlined a framework for thinking about the issue at hand (Figure A4.3). This was often linked to specific country circumstances through applied analysis (about 60 percent), in which staff presented empirical data to illustrate an issue or to support arguments. Formal models were used in about 30 percent of the cases. Literature surveys and historical overviews played supporting rather than central roles. The IEO review found only seven papers that contained neither conceptual nor country analysis in selected issues papers that dealt primarily with exchange rates, and in only 4 percent of the exchange-rate-related selected issues papers (11 papers, as listed in Table A4.1) was there a discussion of global linkages. 5

Figure A4.3.Elements in Selected Issues Papers, 2001–06

(In percent)
Table A4.1.Global Linkages in Selected Issues Papers, 2001–06
EconomyYearPaper TitleDescription of Contents
Canada2005The Effects of U.S. Shocks on the Canadian Economy: Results from a Two-Country ModelMinimalist presentation that models spillovers
China2003Some Implications of China’s Growth and Integration with the Global EconomySees no sign of misalignment, but links integration with global economy to need for flexibility
China2006Implications of Greater Exchange Rate Flexibility in China and Other Asian Emerging Market EconomiesSummary of World Economic Outlook, GEM model for 5 percent appreciation scenario in China and Asian emerging markets (WEO 9/2005)
Euro area2001Why Has the Euro Been So Weak?Explores theories, leans toward shift in portfolio behavior and surge in equity values
Euro area2002Euro Area Trade Flows and the Exchange Rate: How Much Disconnect?Explores why current account did not strengthen during euro depreciation
Euro area2003Exchange-Rate Pass-Through and External Adjustment in the Euro AreaThorough model-based empirical analysis of effects of exchange rate shock in Europe; cross-country comparison, discussion of global imbalances
Euro area2004Global Rebalancing of Current Accounts: A Euro-Area PerspectiveEuro area should boost domestic demand, attract foreign capital, increase competitiveness through structural reform, ease monetary stance
Japan2001Japan and Asia: Policies and ProspectsG-Cubed regional model used to assess transmission of shocks and policies (including quantitative easing and yen depreciation) between Japan and neighbors
Japan2005The Domestic and Global Impact of Japan’s Policies for GrowthExtended version of GEM model used to generate baseline plus two alternative scenarios assessing impact of fiscal, structural reforms on Japanese and world economies
Mexico2003Explanations for the Recent Behavior of the Mexican PesoSurveys global exchange rate links. Cross-country analysis and policy shock of change in reserve management strategy
Thailand2005The Role of Interest Rates in Business Cycle Fluctuations in Emerging Countries:The Case of ThailandUses GEM to argue that more exchange rate flexibility helps economy to better absorb U.S. interest rate shocks

Description of De Facto Exchange Rate Regimes

7. Eight categories of de facto exchange rate regimes were identified in Article IV documents. In at least 12 country cases (some 6 percent of the sample), the IEO found inconsistencies between MFD/MCM’s de facto regime classifications and the descriptions provided in either the body of staff reports or their appendices on IMF relations (Table A4.2). 6 Over the period examined, the review noted changes to de facto exchange rate regimes in 15 countries (or about 8 percent of total).

Table A4.2.Cases of Inconsistent De Facto Exchange Rate Regime Classifications
Classification (2004-05)
EconomyMFD/MCM classification1Description from staff report2Description from appendix on Fund relations2Reference document
ArgentinaManaged floatPegManaged float2005 staff report
AzerbaijanPeg3PegManaged float2004 staff report
EgyptPeg3Not explicitManaged float2005 staff report
EthiopiaManaged floatCrawling pegManaged float2005 staff report
Iran, Islamic Republic ofCrawling peg3Managed floatManaged float2005 staff report
LiberiaManaged float3Not explicitIndependent float2005 staff report
MadagascarManaged float3Managed floatIndependent float2005 staff report
NigeriaManaged floatHorizontal bandManaged float2005 staff report
PakistanPeg3PegManaged float2005 staff report
Papua New GuineaManaged float3Independent floatNot explicit2005 staff report
TanzaniaIndependent floatManaged floatNot explicit2004 staff report
YemenManaged float3Not explicitIndependent float2004 staff report

Exchange Rate Advice

8. In about half of the country cases, the staff appraisal section of Article IV reports suggested that the exchange rate was a live policy issue. In 63 out of the 191 economies (about one-third of the sample), IMF staff provided country authorities with exchange rate advice, overwhelmingly in favor of greater exchange rate flexibility (Table A4.3). In part, advice for greater flexibility appeared to amount to advice for exchange rate adjustment. 7 Such advice was provided in 11 out of 19 cases (net of double counting) where the exchange rate was deemed overvalued, and in 10 out of 15 cases where the exchange rate was deemed undervalued. In about half of these cases, the IMF advice was not accompanied by formal analysis, either of exchange rate regime sustainability or appropriateness of the exchange rate level.

Table A4.3.Exchange Rate Advice and Its Analytical Basis(Number of cases)
Nature of Staff AdviceNumber of Cases With/Without Staff Advice1Regime Sustainability or Suitability Analyzed2Formal Exchange Rate Level Analysis3Currency Deemed Overvalued or Undervalued by Staff
More flexibility5172011 (10)
Less flexibility1100 (0)
Implementation; management of existing regime244112 (6)
No specific advice12814388 (4)

Analytical Basis for Exchange Rate Advice

9. A standard feature of every Article IV staff report is a plot of the real effective exchange rate (REER) index. It is rare, however, to use a more analytical tool of exchange rate level assessment, such as purchasing power parity (PPP), fundamental equilibrium exchange rate (FEER), and behavioral equilibrium exchange rate (BEER) models. In the Article IV staff reports for 191 economies, the IEO review found 17 cases of PPP-based econometric assessments, 12 cases citing the results of the regular CGER exercise,8 14 cases using FEER or BEER models, and 36 cases used other meth ods (Figure A4.4). Multiple methodologies were used in 15 cases, including for Germany, China, Malaysia, and the CEMAC.9

Figure A4.4.Analytical Tools Used to Assess Exchange Rate Levels

Topics in Selected Issues Papers

10. The most frequent topics in the selected issues papers reviewed were exchange rate level and compet-itiveness, followed closely by exchange rate regime; less attention was paid to considerations of exchange rate volatility, and exchange rate pass-through into inflation—possibly reflecting the absence of major inflation episodes in recent years (Figure A4.5). While the total number of selected issues papers on exchange-rate-related issues more than doubled between 2001 and 2005, the distribution of topics discussed did not change markedly.

Figure A4.5.Exchange-Rate-Related Topics in Selected Issues Papers, 2001–05

(Number of cases)

11. Although selected issues papers are not meant to be the main vehicle for delivering policy recom-mendations, they drew policy inferences in about two-thirds of the cases (Figure A4.6). The most common policy inferences concerned structural measures in the context of the need to increase competitiveness. Many of them suggested exchange rate adjustment (devaluation or revaluation), and they generally made a theoretical case for greater exchange rate flexibility. Of the 175 selected issues papers in which staff drew policy inferences, fewer than half spelled out alternative policy options.

Figure A4.6.Policy Inferences in Selected Issues Papers, 2001–06

(In percent)

Reaction of Authorities to IMF Advice

12. Finally, the IEO tried to ascertain from the documents whether national authorities agreed or disagreed with the staff’s exchange rate advice. The review of the documents suggests 39 cases of agreement (out of 191), 26 cases of agreement with qualifications, and 14 cases of disagreement. How to interpret these results, however, is not straightforward. Staff advice on exchange rates cannot be distinguished cleanly from other advice on fiscal, monetary, and structural policy. Even when there is broad agreement on risks, desired reforms, and the overall stance of macroeconomic policy over the medium term, IMF staff and national authorities may disagree over a specific policy action in the short run, and such disagreement is documented with varying degrees of clarity.

Annex A4.1Exchange Rate Surveillance in Article IV Consultations(Full sample of 191 economies for two consultations through 2005)
1.How many paragraphs (percent of total) mention exchange rates?
Article IV staff reports:16 percent
Mission briefing notes18 percent
Back to office13 percent
2.Were there any exchange-rate-related selected issues papers?
Yes (99)No (92)
3.What is the current “de facto” exchange rate regime?
Independent float (30)Crawling band (4)Currency board (9)
Managed float (49)Crawling peg (4)Monetary union (44)
Horizontal band (9)
Other fixed peg (42)
4.What is the monetary framework?
Exchange rate anchor (92)Inflation targeting (22)Other (52)
Monetary targeting (28)
5.Does the staff appraisal in Article IV reports identify the exchange rate as a live policy issue?
No (99)Exchange rate regime issues (46)Volatility issues (14)
Exchange rate level issues (45)Other (22)
6.Is the overall exchange rate discussion forward looking?
Mostly forward looking (105)
Mostly backward looking (39)
Exclusively backward looking (2)
Not clear/can’t tell (45)
7.How intensively does the appraisal discuss exchange rate issues?
Prominently (54)Briefly (121)Not at all (16)
8.Does the prominence of coverage in the Article IV report differ noticeably from coverage in internal documents?
Yes (12)No (175)
9.The staff exchange rate assessment mentioned the following elements:.
Resolution of global imbalances (17)
Global capital markets (14)
Prospective spillovers from trading partners/competitors (31)
Prospective regional spillovers (23)
Domestic considerations (144)
Other (20)
None of the above (21)
10.In the discussions of exchange rate issues, is there explicit reference to:.
Fiscal policy (53)Monetary policy (121)None of the above (48)
Banking/corporate balance sheets (44)
11.In the discussions of which other policy areas is there explicit reference to implications for exchange rates?
Fiscal policy (51)Monetary policy (94)Banking/financial (35)
Structural (43)None (60)Other (18)
12.Did the Fund advise a regime change/adjustment? If so, in what direction?
No (128)More flexibility (51)Less flexibility (1)
Inflation targeting (8)Monetary targeting (5)Other (11)
13.In 2004-05, was the economic environment calm or turbulent?
Calm (122)Turbulent (31)Can’t tell (38)
14.Have there been changes to the de facto regime over 2003-05? What was the old exchange rate regime?
No (176)Managed float (5)Other fixed peg (4)
Crawling band (1)Horizontal band (5)
15.If the answer under 14 is yes, what was the old monetary framework?
Exchange rate anchor (9)Monetary targeting (3)Other (3)
Inflation targeting (0)
16.Is there any formal analysis of regime sustainability/choice? If yes, on what basis?
No (167)Analysis of economic shocks (10)Other (13)
Optimum currency area (5)Cost-benefit analysis (4)
17.At any stage in the process, are there any changes in regime-related issues identified?
No changes across documents (168)Particular issues dropped (11)Other changes (3)
Particular issues added (8)
18.The exchange rate has been explicitly identified in the text as:.
Overvalued (19)Fairly valued (87)Undervalued (15)
Misalignment not assessed (70)
19.At any stage in the process, are there any changes in levels-related issues identified?
No changes across documents (169)Particular issues dropped (8)Other changes (4)
Particular issues added (7)
20.What are the analytical tools used to discuss exchange rate levels?
REER charts (191)PPP estimates (17)Other (36)
FEER/BEER (14)CGER estimates (12)
21.Does Fund analysis point out the limitations/caveats of its tools?
No (159)Yes (32)
22.What prompts the discussion on exchange rate levels?
Current account developments (46)Competitiveness considerations (107)
Capital account developments (22)Appears to be template driven (48)
Fiscal developments (27)Other (38)
23.Is there any reference to the "pointers"/procedures in the 1977 Surveillance Decision?
None (172)
Exchange rate manipulation, protracted interventions (3)
Interventions to counter disorderly conditions (16)
Explicit policy action for balance of payments purposes (0)
24.Was there a discussion of alternative exchange rate policy options?
No (134)Yes (57)
25.Where is any formal analysis on levels/regimes reported?
Article IV reports (19)Pre-mission briefings (5)Other (15)
Issues papers (42)Back-to-office reports (0)N/A (123)1
26.How did the authorities react to any policy advice on exchange rates given?
Agree (39)Agree, but with qualifications (26)Disagree (14)Can’t tell/doesn’t apply (112)
27.Does the country have a Fund-supported program?
Yes (68)No (123)
If yes, are exchange-rate-related issues part of conditionality?
Yes (9)No (59)
28.Was data availability an issue?
Yes (90)No (101)
Annex A4.2Selected Issues Papers on Exchange Rate Issues(Full sample of 256 papers)
1.How long is the paper?
<5 pages (5)2 percent
5-9 pages (61)24 percent
10-14 pages (83)32 percent
15-19 pages (55)21 percent
20+ pages (52)20 percent
2.The paper contains the following elements:.
Literature survey (33)13 percent
Conceptual analysis (183)71 percent
Applied analysis (152)59 percent
REER- or PPP-based analysis (100)39 percent
A formal model (69)27 percent
Formal cross-country comparative analysis (37)14 percent
Discussion of global linkages (10)4 percent
Historical overview (76)30 percent
3.The paper materially deals with:.
Exchange rate regime (69)27 percent
Exchange rate levels (92)36 percent
Exchange rate volatility (25)10 percent
Price level pass-through/inflation (46)18 percent
Competitiveness (78)30 percent
Other (51)20 percent
4.Any policy inferences that are drawn deal materially with:.
Exchange rate regime (52)20 percent
Exchange rate levels (37)14 percent
Exchange rate volatility (6)2 percent
Foreign reserves (22)9 percent
Price level pass-through/inflation (25)10 percent
Structural measures (56)22 percent
Prudential supervision (17)7 percent
Other (24)9 percent
No policy inferences drawn (82)32 percent
5.Are there explicit caveats to the analysis?
Yes (85)33 percent
No (171)67 percent
6.Do es the selected issues paper analysis take account of country-specific factors in its bilateral assessment?
No (37)14 percent
Structure of the economy (112)44 percent
Institutional frameworks (80)31 percent
Political constraints (15)6 percent
Trade patterns (69)27 percent
Extent of integration into global capital markets (29)11 percent
Balance sheet characteristics (29)11 percent
Dollarization (34)13 percent
Policy shocks (37)14 percent
Other (39)15 percent
7.Is there evidence of policy dialogue between the IMF and authorities?
Yes (17)7 percent
No (239)93 percent
8.Do es the paper spell out alternative policy options?
Yes (84)33 percent
No (172)67 percent

Most (but not all) of the missions took place in 2004 or 2005.

The document of reference was the latest version of the Article IV report available (i.e., the published version for those that are published); these were then checked for corrections and/or deletions to confirm whether relevant changes had been made from earlier versions.

The assessment excluded entities with which no Article IV consultation took place (Somalia and West Bank and Gaza), but included those for which only one consultation had been completed in recent years (Iraq and Grenada), as well as countries in the process of redefining their political and economic boundaries (Serbia and Montenegro).

In at least one of the two consultations reviewed.

Several global-linkages-related selected issues papers for major countries, particularly the United States, did not focus directly on exchange rate policy. Such papers are not included in Table A4.1 or Figure A4.3.

Of these 12 cases, 8 cases involve retroactive reclassification by MFD/MCM; in many of these, classification lags resulted from a difference of opinion about the appropriate regime classification between departments or between staff and the authorities.

A similar interpretation was suggested in a recent Board paper prepared by staff. See “Treatment of Exchange Rate Issues in Bilateral Surveillance—A Stocktaking,” EBS/06/107, August 2006, footnote 26.

The Consultative Group on Exchange Rate Issues, an interde-partmental working group within the IMF, produces a semiannual report on the exchange rates of the euro area and 10 industrial countries. For greater details on the CGER exercise, see Background Document 3.

See Background Document 3 for a more detailed analysis, though based on a slightly different coverage of country documents.

    Other Resources Citing This Publication