Background Document 7: Assessing the Effectiveness and Impact of IMF Exchange Rate Surveillance
- International Monetary Fund. Independent Evaluation Office
- Published Date:
- August 2007
1. This background document presents a framework for assessing the effectiveness and impact of IMF exchange rate surveillance and related IMF advice and provides the basis for some of the conclusions drawn in the main report. For this purpose, it is helpful to think of the process of IMF surveillance in terms of a results chain that connects inputs to outcomes (Figure A7.1.). A critical element of surveillance concerns how IMF staff performs analysis, forms a view, and communicates the resulting advice to national authorities.
Figure A7.1.How IMF Views Connect to Outcomes: A Closer Look
Effectiveness at Each Stage
2. Although Figure A7.1 does not fully represent the complex nature of what the IMF does in practice, it is sufficient to show that the impact of IMF surveillance on member countries’ policies depends on effectiveness at different stages of the results chain. At the “inputs” stage, for example, effectiveness is determined by such factors as the accuracy, timeliness, and comprehensiveness of data, the adequacy of staff expertise and analytical tools, and how well the internal review process works.
3. In this connection, the evaluation finds that a data deficiency of one type or another has been a factor in 90 of the 191 economies for which the documents for the last two Article IV consultations of the 2002–05 period were reviewed. Specifics of the data problems vary across countries, but some could certainly weaken the effectiveness of exchange rate surveillance. Indeed, some 40 percent of the staff surveyed by the IEO identified better availability of data as an area where significant improvement could be made to raise the overall quality of exchange rate analysis (see also Background Document 6).
4. How to address data deficiencies is a difficult issue. When the authorities themselves do not have the data, which is found to be the case in several countries, technical assistance may be a solution. In other cases, as documented in Background Document 6, the authorities have the data but are unwilling to share them, often on grounds of market or political sensitivity. Such data may include details of official foreign exchange market intervention, composition of foreign exchange reserves, foreign exchange transactions of government-affiliated entities, and the like. Although access to these data may not always be necessary for high-quality surveillance, the IEO has found several cases where IMF staff appeared to be conducting discussions with national authorities from a position of substantial informational disadvantage.
5. At the “operations and communications” stage, effectiveness might be influenced by such attributes of staff views and analysis as accuracy, feasibility, and candor. A regular criticism that the IEO has heard from country officials is that the analysis underlying staff advice is not transparent or the advice is not detailed enough for implementation (see also Background Document 6). It appears that authorities often judge the usefulness of advice, not just in terms of accuracy, but also on the basis of how the advice is derived and how practical and feasible it is in the specific country context.
6. How to treat market or politically sensitive material would also condition the IMF’s potential influence on countries’ policies. Sometimes sensitive material is “sanitized” in an attempt not to offend the authorities or upset the markets, but such practice may compromise effectiveness by diluting the sharpness and clarity of the message. The evaluation finds that sensitive issues are sometimes not discussed at all. About 20 percent of the senior staff surveyed noted that the authorities’ reluctance to discuss certain topics was an impediment to clear and candid treatment of exchange rate issues (see Background Document 6). When such issues are discussed, they may not be reported to the Executive Board (see below). Although the desk review found very few cases of explicit deletions from staff reports, IEO interviews with mission chiefs suggest that self-censorship may be at play in some cases.
Impact at the Final Stage
7. IMF surveillance may help influence member countries’ policies through four broad channels: (1) direct communication to authorities of IMF Executive Board views; (2) peer pressure exerted by other members, occasioned by IMF Executive Board views; (3) pressure from the public or markets; and (4) advice delivered by IMF staff in policy discussions with the authorities. A series of interviews have suggested that, of the four channels, national authorities and IMF mission chiefs seem to value advice delivered by IMF staff as particularly effective. In part, this reflects the confidential nature of many of the topics that are bound to come up in the area of exchange rate policy. IMF mission chiefs have indicated to the IEO incidences where sensitive exchange rate issues were discussed on the understanding that the substance of the discussions would remain confidential.
8. Impact through the policy discussions channel may be indirect and may not necessarily show up as a policy decision. A number of officials, including from major advanced countries, have told the IEO that even when discussions do not change their prior views, they nonetheless appreciate the opportunity to interact with IMF staff, as such interactions allow their views to be tested and verified.
9. In contrast, Executive Board deliberations appear to have had only a limited influence. However, some mission chiefs indicated to the IEO that prospective support from the Executive Board could strengthen their position in policy discussions with authorities. In other words, the existence of the Executive Board as a review body could lend credibility to a view expressed by staff, to the extent that it also represents the Board’s view.
10. As to the IMF’s potential influence through public policy debate, some officials have expressed misgivings about the IMF actively using this channel. The IMF’s influence is limited in any case if the authorities do not agree to the publication of an Article IV consultation report. On the other hand, when the IMF’s assessment of domestic policies is positive, authorities are typically eager to publicize such endorsement. First, endorsement by the IMF of a country’s exchange rate regime or economic policies adds to the authorities’ credibility, which may in turn facilitate access to capital markets. Second, endorsement by the IMF of certain policies can support decision making within the country by helping to overcome differences of view among different branches of government. IEO interviews of country authorities suggest that these channels of influence occasionally operate.
11. The impact the IMF has on the markets is difficult to assess. Indeed, the IEO did not find a convergence of views among the large number of market participants it interviewed.1 Perception seems to matter. In countries with poor transparency, for example, IMF views may affect markets to the extent that market participants perceive such views to be based on superior information. Likewise, in countries pursuing IMF-supported programs, market participants take IMF views seriously because they perceive the IMF to have influence on economic policy decisions. In countries for which information is readily available, on the other hand, almost all market participants took the view that published IMF assessments could not provoke significant market reaction, except in the very short run or if supported by market positioning.
12. Officials indicated to the IEO two opposing views about the potential impact of IMF statements on the level of exchange rates. Some said that, given the well-known margin of error in any exchange rate level assessment, IMF views would be treated as just another view and would not trigger market reaction; others argued that, given its well-established reputation as a neutral international body, the IMF’s views would be treated differently and would therefore impact the markets.