The Challenges Facing the IMF -- Speech by Agustín Carstens, Deputy Managing Director, IMF

July 27, 2004

Speech by Agustín Carstens
Deputy Managing Director, International Monetary Fund
At the Australian Treasury
Canberra, Australia
July 27, 2004

As Prepared for Delivery

I am very pleased to be here, and to have the opportunity to meet with you at the Australian Treasury. Australia's active participation in shaping IMF policies and in promoting their effectiveness is greatly appreciated, and I view this visit as a valuable opportunity to better understand your views and concerns. I would like to start by giving you my views on some of the challenges currently facing the Fund, and will then welcome a broad exchange of ideas.

The IMF's mandate has not changed over the last 60 years. Our chief goal remains that of promoting global financial stability, and thereby laying the groundwork for sustained growth. What has changed is our approach to achieving this goal, which has been significantly and continuously adapted in response to global developments. The breakdown of the Bretton Woods system, the emergence of Africa as a continent of independent nations, the oil shocks of the 1970s, and the collapse of communism have all brought difficult economic challenges. Most recently, we have been grappling with the growth in size and sophistication of international capital markets. As you know, this growth has provided opportunities, but also increased risks—and the financial crises of the last decade are a painful demonstration of the latter. Looking ahead, we anticipate more challenges. Financial globalization will intensify and cross-border capital flows will increase, especially as populations age in the industrial countries. And emerging markets will likely claim an increasingly larger share of the world economy.

We are working to address the challenges we foresee and to incorporate the lessons we have learned throughout our three main functions—surveillance, the provision of financial support, and technical assistance. These functions are often integrated in practice. But today, I would like to focus my remarks on surveillance, which must remain at the heart of our efforts to prevent crises, promote global financial stability, and foster sustained and high growth. And I know that it is a topic of keen interest in Australia.

Our thinking on surveillance can be informed by considerable recent work. The International Monetary and Financial Committee (IMFC) has paid close attention to this issue over the past two years, and the Fund's Executive Board has conducted several discussions, supported by staff analysis as well reports by the Independent Evaluation Office (IEO) that have had a surveillance dimension. Last week, the Board concluded the Biennial Surveillance Review, which examined the broad range of activities that comprise IMF surveillance, with the aim of making it more effective across the whole membership.

In my view, a general principle underpinning many of the conclusions of this work is that a good relationship between the IMF and its member countries is central to the success of surveillance. Ideally, the Fund is perceived as a trusted advisor and a facilitator of the economic policy debate. I am proud of the relationship between the IMF and Australia—there is frequent contact between the Fund and policymakers here, discussions are frank, and there is open debate on key issues. That Australia's expansion is now entering its 13th year is testimony to your long track record of sound policies. I hope that IMF surveillance has played a positive role in your success, and I would value your views on how such a positive relationship might be replicated more frequently across our membership.

In the remarks that follow, I would like to consider the two main aspects of surveillance. Let me first discuss how we can most accurately identify emerging imbalances and vulnerabilities. And then, I will turn to how we can best motivate countries to take early, effective action to address those imbalances and vulnerabilities, before they cause harm.

Identifying emerging imbalances and vulnerabilities

There are several basic objectives we must meet to ensure that we can accurately and pre-emptively identify emerging problems. First, surveillance must be well-informed. On this dimension, significant progress has been made. The Fund's standards for data dissemination have helped the majority of our members to improve the coverage, periodicity, and timeliness of their data, as well as its quality and availability to the public. We continue to seek improvements in this area, including by increasing the availability of data needed to conduct balance sheet analysis and vulnerability assessments.

And this leads me to the second objective, which is that surveillance must focus on the correct issues. The coverage of surveillance has expanded over the years from a relatively narrow concentration on fiscal, monetary and exchange rate policies to include assessments of external vulnerability, financial sector strength, and structural and institutional issues that have an impact on macroeconomic conditions. While broader coverage can and has enhanced the quality of surveillance analysis, for it to be effective it needs to focus on a few strategic issues. Some issues must be singled out as the key ones, depending on country-specific circumstances and what is of greatest relevance to maintaining global financial stability. We have learned much from recent crises that has helped us to sharpen the analytical tools we use for assessing risks and for helping us to choose the right focus in our discussions. In particular,

· To guide identification of vulnerabilities most in need of attention, we have helped to develop and implement internationally established standards and codes. These encompass a wide range of areas—including policy transparency, financial sector supervision and regulation, corporate governance, and insolvency and creditor rights. I should note that Martin Parkinson was instrumental in developing the Standards and Codes Initiative when he was at the Fund in the late 1990s. So far, 90 countries have published a Report on Standards and Codes (ROSC), summarizing their observance of at least one of these standards. We are continuing to work to strengthen implementation of these initiatives and to refine them.

· Since financial sectors have played a key role in the generation and transmission of vulnerabilities, we are placing a much greater emphasis on financial sector soundness. The Financial Sector Assessment Program (FSAP) remains the key tool for in-depth analysis of weaknesses at the country level, and I am pleased to note Australia's potential interest in participating. Coverage of financial sector issues in Article IV surveillance reports is also being improved.

· We are working to complement more systematically our traditional flow-based analysis with an examination of stocks and structures of assets and liabilities. This includes development of a balance sheet approach, which involves examination of interlinked sectoral balance sheets to gauge exposure to interest rate, exchange rate, and rollover risks. We are also implementing a common framework for more rigorous assessments of public and external debt sustainability.

· IMF teams must also be sensitive to the authorities' priorities, provided that vulnerabilities are not overlooked. More and more frequently we are focusing our surveillance work towards identifying ways to promote higher sustainable growth, in particular in those countries in which macroeconomic stability is not of concern.

A third objective that must be met for effective identification of problems is that surveillance must be conducted evenhandedly across the membership. Clearly, we must be on the watch for countries directly vulnerable to crisis. But increasing financial integration means that surveillance must also focus on the stability of regions and the system as a whole. Surveillance of major industrial countries is therefore essential to ensure global stability—even if a country itself is not at risk, it may contribute to global imbalances that place the system at risk. But precisely because the benefits of industrial country surveillance are, at least in part, a "public good," convincing these countries to take the process seriously is a difficult challenge. And, as Secretary Henry has pointed out1, "why listen to what the bank manager thinks of you if you have no need to apply for a loan?"

A related challenge is how to improve the integration of bilateral, regional, and global surveillance. The latter—global surveillance—has been important to our work from the start, although the vehicles for conducting it have expanded greatly. We are now much more active in monitoring international capital markets. We are an active participant in the Financial Stability Forum, and liaise closely with global regulatory standard setters. Executive Board discussions of the World Economic Outlook (WEO) and the Global Financial Stability Report (GFSR) are supplemented by other, more frequent and less formal Board discussions. Regional surveillance, too, is emerging as an important focus. Surveillance of currency unions is becoming more formal, there is greater coordination of discussions with regional institutions and their members, and the IMF has increased its participation in regional surveillance groups. We are also introducing regional surveillance initiatives, such as the proposed regional financial sector project in Central America. Yet, while good progress is being made in each of these areas, their integration is still weak in some respects. The Biennial Surveillance Review highlighted that there is substantial room to strengthen our analysis of regional and global spillovers—in particular, analyzing how identified global or regional risks may impact individual members.

Improving the impact of the surveillance process

Early identification of imbalances and vulnerabilities is necessary, but not sufficient, for surveillance to be effective. Members must also be motivated to take the actions necessary to address the risks. That, in turn, requires members to be convinced that the advice they receive is appropriate. In addition to being well targeted and based on good information, the advice must take into account country-specific circumstances, especially the constraints that a country may face in implementing the recommended policies.

Central to a good outcome is a frank policy dialogue between IMF staff and the country authorities. Many of the country officials interviewed for the Biennial Surveillance Review noted that this dialogue should be conducted and perceived as a collaborative and mutually beneficial exercise, and not an audit. I agree. Clearly, both parties to the surveillance dialogue can benefit. The IMF can, and should, bring vast cross-country experience and a global perspective, although greater efforts are needed to apply this experience more systematically. The IMF should expect to take away a better understanding of the particular circumstances of each country, as well as lessons that can be passed on to other member countries. Ideally, the process is helpful in flushing out views and policy options on both sides.

The views of both sides must then be presented in a way that leads to constructive discussion within the Executive Board. The ultimate goal of this discussion, of course, is effective peer pressure. At the Board, the international community has the unique opportunity to persuade a member's authorities to act promptly to address emerging imbalances and vulnerabilities. Peer pressure will be most effective if the Board is perceived as being truly representative of the international community. Yet, further progress on the issue of voice and representation will require broader consensus among the Fund's shareholders than currently exists, and I welcome your thoughts of the efforts that would be effective in building agreement in this area.

Board discussions are one important avenue for surveillance to have a positive impact. So too is circulating the findings more widely. Increased transparency of the process—including publication of country documents—is promoting greater accountability and helping markets and donors assess risks more accurately. I should note that about three-quarters of country documents are now published, and voluntary but presumed publication of Article IV staff reports came into effect at the beginning of this month.

Outreach can also be important, and Mr. Callaghan has persuasively made this point to the Executive Board on several occasions. The IMF can play a very useful role in assisting countries with the often difficult task of implementing reforms by helping to convince legislators, interest groups, civil society, and the public at large that certain reforms are necessary. More generally, the Fund can foster a better understanding of the policy process and the nature of its partnership with the member country. In many countries, the Fund has been increasing its contact with national parliaments, is maintaining a more continuous dialogue with civil society, and is seeking out greater opportunities for press coverage.

Another important way that the IMF can help countries to implement reforms is through its provision of technical assistance (TA). This program has become increasingly important in assisting countries to follow up on weaknesses identified through surveillance, including through FSAPs and ROSCs. I would like to thank Australia here for its financial support and provision of experts.

Let me now briefly touch on a couple of additional issues that should influence our thinking on how to improve the positive impact of surveillance. First, we must recognize that while enhanced transparency has been—and should continue to be—an integral part of the Fund's efforts to strengthen the framework for surveillance, there is a tension between transparency and candor. A prime example is the discussion of exchange rate issues. More than once, the Fund has had serious concerns about the sustainability of a country's exchange rate regime. Yet, out of concern for triggering the very sort of crisis it is our mandate to help avoid, we were not forceful in expressing our views. Indeed, the Biennial Surveillance Review found that pointed discussions of exchange rate issues have been both rare and controversial. In the event that a crisis did arise, the Fund was often seen as negligent in its duties, or even complicit in fomenting the crisis. Clearly, this is a tension we need to address more fully.

Another issue is whether, in some cases, the incentives for taking surveillance seriously may need to be sharpened further. Should the Fund say "no" more selectively, assertively, and predictably? One possible way of doing this would be to link access to Fund resources more explicitly to a country's own past actions—its pre-crisis policy efforts, its responsiveness to the surveillance process, or its adherence to standards and codes. This thinking underpinned the IMF's Contingent Credit Line (CCL), although for different reasons it was never used. But we are continuing to pursue the idea in considering the design of precautionary arrangements and contingent access to Fund credit.


In general, I believe we have a good handle on the avenues that should be pursued for improving the Fund's ability to identify emerging problems. What is less certain is the best way forward in convincing countries to address these problems forcefully and in a timely fashion. I would like to conclude by posing some questions to you, which I think get at the most fundamental dilemmas we face in encouraging prompt action.

First, what is the appropriate balance between transparency and candor? In particular, how can the Fund's role as a confidential advisor be reconciled with its role as a provider of signals to markets, creditors, and donors?

Second, can we successfully link access to Fund resources to a country's own actions? Clearly for this approach is to be successful, a number of dilemmas will need to be addressed. How should we deal with negative signals? Will on-off signals be misinterpreted? More fundamentally, would our signals be credible if they are not backed by financial resources?

Third, what more can we do to convince industrial countries to take the surveillance process seriously? Clearly, these countries have considerable in-house technical expertise. It is easy for the domestic benefits of particular policies to be viewed as more important than potential global costs. And access to IMF resources is not a concern.

Last, can a more "strategic" management of the surveillance process help address some of the above questions? That is, is there merit in clearer definitions of objectives, strategies, performance indicators, and resources? I know that you have taken a strong interest in this issue.

I am very interested in hearing your thoughts.

Thank you.

1 Dr. Ken Henry, "Australia and the International Financial Architecture-60 Years On," Sir Leslie Melville Lecture, July 16, 2003.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100