Chapter 6 How Effective Was the IMF’s Work on Trade Policy Issues?

International Monetary Fund. Independent Evaluation Office
Published Date:
October 2009
  • ShareShare
Show Summary Details

48. There are no straightforward metrics for assessing the effectiveness of the IMF’s involvement in trade policy issues. Three approaches are therefore taken in this evaluation. At the broadest level, and consonant with some of the academic literature, is there evidence that IMF involvement increased trade volumes? More modestly, was IMF advice, whether in surveillance or conditionality, implemented through lasting policy changes? And even more modestly, did IMF involvement get issues on the table or productively into the public debate?

49. Though subject to serious measurement problems, there is some evidence to support a positive effect of IMF involvement on trade volumes. The methodology used to find such effects (a gravity model) can only be used to detect possible effects of IMF conditionality, because discrete observations on involvement are needed. An earlier study, which found that trade policy conditionality does have a positive effect on trade flows over the medium term, was replicated for this evaluation (with updated data and an effort to address technical problems), and weak evidence of a favorable effect was found.13

50. The reflection of IMF conditionality in lasting changes in trade policy was spotty. For the most part, conditions on trade policy were implemented, even if with delays. But at least partial subsequent reversals were common. Many of the low-income countries examined imposed supplementary duties within a few years of tariff cuts undertaken during IMF arrangements. Many also moved aggressively to PTAs, leaving staff of the view that trade liberalization was off the table for bilateral negotiations. Among the emerging market countries, the record was at least as uneven. In Korea and Indonesia when conditionality essentially accelerated an ongoing program of tariff cuts, changes tended to persist. But other conditions on state monopolies and export taxes (especially in Indonesia and Ukraine where some issues were particularly contentious) tended not to result in lasting changes.

51. Often IMF involvement was not particularly effective even on the third criterion for effectiveness—getting issues into the public debate. Occasionally, press coverage of IMF views on trade policy issues revealed a constructive debate: reports on the work on preference erosion in the context of the Bangladesh program stand out in this respect. But often the IMF seems to have polarized public opinion on an issue of dubious macroeconomic import. In survey results, neither authorities nor staff saw much of a role for the IMF in generating a constructive debate in the context of a lending arrangement.

52. Interpreting the effectiveness of surveillance requires more subtlety. Objectives of surveillance are broader than those of conditionality. While some advice would ideally be implemented immediately, most is put forward for consideration or as a marker for best practice. Often too, trade policy advice is delivered in such general terms (for example, support the conclusion of the Doha Round or pursue multilateral liberalization alongside agreements on PTAs) that implementation or the influence of the IMF could only be defined impressionistically. In the surveillance case studies, policies were not obviously changed in line with IMF views. But did the name-and-shame approach in the U.S. staff reports in the late 1990s help forestall protectionist tendencies? Might some rules of origin have been more complex had the IMF not urged simplicity? Such questions have no definitive answers.

53. Survey results put a fairly favorable light on effectiveness. The IEO survey asked both officials and IMF staff whether IMF advice in four areas (trade in goods, trade in services, PTAs, and customs) enhanced the debate on trade-macro linkages, promoted a change in the right direction, and/ or prevented changes in the wrong direction. Some 50–60 percent of official respondents who had discussed trade policy with an IMF mission saw IMF advice as highly or somewhat effective, though for PTAs the percentage dropped to about 30. Some 80 percent of staff respondents whose mission had provided trade policy advice felt the advice was highly or somewhat effective in all four areas, with PTAs again somewhat lower.

54. Interviews with country officials revealed a mix of views about the IMF’s effectiveness on trade policy issues. Many officials declined interview requests or did not respond to the survey, stating that they had had no interaction with the IMF on trade policy. But those who spoke with the evaluation team had diverse impressions.

  • Trade policy is formulated by segments of the government that typically did not talk to IMF missions and therefore IMF advice was delivered to the wrong audience.

  • IMF advice would need to be packaged more persuasively to significantly help governments make the case for more liberal trade policies against vested interests.

  • The IMF’s expertise on trade policy issues was not as strong as on other macroeconomic issues and therefore its advice lacked credibility.

  • By not focusing actual advice (as opposed to research) on PTAs, the IMF took itself out of a central debate on trade policy.

  • If its trade policy advice—even the generic “stick to multilateral liberalization and cut agricultural subsidies”—ceased, the IMF would be seen as accepting protectionism.

55. Doubts about the trade policy expertise of IMF staff surfaced in interviews, though official and staff survey responses were more reassuring. About 50 percent of officials responding to the survey saw the expertise of IMF missions on trade policy as high or moderate, though another 40 percent indicated no view. About 30 (45) percent of staff respondents felt their mission team had at least one member strongly (moderately) qualified to take positions on trade policy In interviews, several staff members expressed confidence that they could draw on the expertise of economists in the (old) Trade Policy Division should their mission have needed it. Staff interviews pointed to a view that trade policy is an unfunded mandate of the IMF.

The methodology and results are presented in a background paper by Yoon, “Re-evaluating the Effectiveness of Trade Conditions in Fund-Supported Programs” (see

    Other Resources Citing This Publication