Early Warning Systems and Their Role in Surveillance

Keynote Address by Takatoshi Kato, Deputy Managing Director, International Monetary Fund
At High-Level Seminar, Singapore
February 9, 2010

Thank you, Mr. Ostry, for that introduction and for your opening remarks. Good morning, ladies and gentlemen. It is a great pleasure to be here today among this impressive gathering of experts from so many countries in the Asian region, and from academics, policy makers, and representatives of the private sector from around the globe. Let me begin by expressing my thanks to the Research Department of the IMF for organizing this conference, and to the Singapore Regional Training Institute for their help with the local arrangements.

* * *

At a time when the global economy is recovering from the most severe crisis since the Great Depression, and a decade after the financial crises that roiled many emerging market economies, this seems an opportune occasion to reflect on the lessons from these episodes, and how future crises can best be anticipated and prevented.

While the current crisis had complex and multifaceted causes, perhaps the lesson for the IMF is that surveillance for crisis prevention needs to be much more rigorous, not only in terms of the traditional country-level macroeconomic analysis, but also with a view to better encompassing financial sector and regulatory issues, and to take account of cross-border spillovers and systemic risks. I think it is fair to say that prior to the current crisis, and in the aftermath of the emerging market country capital account crises of the 1990s, IMF surveillance was relatively well geared to spotting vulnerabilities in emerging market economies—for instance, the problems in emerging Europe had been extensively debated, and bilateral surveillance had pretty clearly telegraphed the underlying vulnerabilities and imbalances that made these economies ripe for a crisis.

What was much less well appreciated was the nature of vulnerabilities in advanced economies, especially major financial centers like the United States—and how relatively small problems, like subprime mortgages, could create much broader distress, reflecting amplification from excessive leverage and insufficient capitalization (including by off-balance sheet entities), as well as cross-border linkages that transmitted the bursting of the housing bubble in the United States to other advanced economies and, ultimately, to the rest of the world. An obvious conclusion to draw is that crises in advanced countries, though more rare, are likely to have broader ramifications for the rest of the world and for the system as a whole, than the type of balance of payments crisis we are familiar with from the emerging market country experience.

To strengthen the Fund’s capacity to anticipate such crises, our researchers began, already in early 2008, to adapt analytical tools that had originally been developed in the emerging market country context, and apply them to help pinpoint vulnerabilities in advanced economies. But of course that was just a beginning. What the current crisis showed was the need to improve our understanding of cross-border spillovers, macro-financial linkages within and across countries, and broader systemic risks for the global economy. And it also pointed to gaps in the IMF’s liquidity provision and lending toolkit. Let me take each of these in turn.


Under its Articles of Agreement, the IMF is charged with “oversee[ing] the international monetary system in order to ensure its effective operation.” This is a formidable, and daunting, mandate, and to fulfill it, I think, three elements are required:

Spanning the spectrum of surveillance. The Fund’s traditional modality of surveillance, and the basis of much of its policy dialogue, is bilateral: examining the impact of economic policies on the country itself, albeit through the prism of external stability (and hence taking at least some account of spillovers). At the other end of the spectrum are our flagship multilateral surveillance products—the World Economic Outlook and Global Financial Stability Report—that are concerned with the stability of the global economic and financial system. But in between, there is scope for greater surveillance of policy spillovers, especially those of countries that individually or in aggregate are systemically important. The 2006 Multilateral Consultations are one example of such multilateral surveillance in action, and one of the few attempts in recent years at policy coordination. The G-20 Framework for Strong, Sustained, and Balanced Growth, and the Mutual Assessment Process suggest that, if anything, the recent crisis has rekindled interest in countries taking account of multilateral dimensions when setting policies. The IMF is already undertaking much of the underlying analytical work, and we need to consider how best to support such efforts, including in ways that are representative of the whole IMF membership.

Financial sector oversight. The crisis has reminded us—if we needed reminding—of the critical importance of financial sector supervision at the national and supranational levels for effective crisis prevention. Of course, this is not the purview of the IMF alone, but with the balance sheets of large and complex financial institutions (LCFIs) exceeding the GDPs of many of our member countries (and, in some cases, of their own country), it is clear we cannot ignore them in our surveillance activities. This calls for close collaboration with such bodies as the Financial Stability Board and the Bank for International Settlements, with each institution exploiting its comparative advantage.

Capital flows. Against a backdrop of a trend increase in financial globalization, the current crisis underscored once again the fickleness of capital flows. What was novel about this crisis was that many EMEs faced slowdowns of capital inflows—or outright reversals—not because of any domestic vulnerability, but because of deleveraging by banks in advanced economies. Indeed, there was a very real risk of widespread financial protectionism as politicians sought to ensure that lending continued in their own countries even as banks were trying to restore their capital positions. This was ultimately avoided, but only by dint of concerted multilateral pressure. Now the situation is somewhat the opposite: low interest rates in advanced economies are driving surges of capital flows to EMEs, complicating their macroeconomic management and risking asset bubbles. In this environment, it is natural, depending on country circumstances and the capacity of macroeconomic and prudential policies to deal with the problem, to consider imposing or strengthening controls on capital inflows—but again, it is vital that any such decision take account of multilateral repercussions. I believe the IMF can be of considerable service to the membership if, as part of its surveillance activities, it provides countries with policy advice on when controls on inflows might be appropriate (and when they would not); analytical work on this question is well underway. A multilateral perspective would also underscore that ensuring orderly flows is a joint responsibility, which needs to be taken into account in the policy choices of both capital-receiving and capital-exporting countries.

Liquidity provision and lending

In addition to surveillance, the crisis revealed gaps in our toolkit for rapidly providing liquidity for crisis prevention. I am pleased to say that the IMF moved quickly as it became clear that there would be no “decoupling” by developing and emerging market countries, and that the crisis would have serious repercussions for countries far from its epicenter. First, an SDR allocation equivalent to about US$250 billion provided immediate and unconditional liquidity. Second, we introduced the Flexible Credit Line (FCL), which provides significant amounts of contingent financing with no ex post conditionality. Third, we formalized High Access Precautionary Arrangements for countries for which an FCL would not be appropriate, but which could benefit from contingent financing. Fourth, we doubled normal financing access limits. And fifth, we have sought to ensure that conditionality is strictly limited to those measures critical for the success of a country’s economic program, eliminating structural performance criteria entirely.

These measures, implemented in record time, undoubtedly helped contain the severity of the global financial crisis, and likely helped stave-off some of the worst outcomes. But we must not rest on our laurels. We are looking to further innovations in our lending toolkit, with the ultimate goal of ensuring that we are a reliable source of country insurance to emerging market and developing countries, reducing their need to self-insure by stockpiling costly international reserves. We have also begun a reflection on the structure of the international monetary system, with a view to tracing any inherent instabilities and thinking through possible solutions.

The Early Warning Exercise

So where does the Early Warning Exercise fit into this overall effort at crisis prevention? It is indeed an integral part. To recap briefly, in autumn 2008 G-20 officials called on the IMF and the Financial Stability Board to collaborate in a periodic assessment of systemic risks, which became known as the Early Warning Exercise. The EWE aims to identify vulnerabilities and assess risks of unlikely, but not implausible, downside scenarios that—if realized—would result in policy recommendations different from those underlying the baseline scenarios in the WEO and GFSR. Following discussions at the IMF Executive Board and the FSB Plenary, the key findings of the EWE are presented to senior officials during the Spring and Annual Meetings. The WEO, GFSR, and bilateral consultations with national and regional authorities provide opportunities for follow up.

Since you will hear more about the EWE in the course of today, I will not go into details now. But I did want to mention three features of the EWE that, I think, make clear how it fits into the elements I identified earlier as necessary for effective surveillance:

• First, as you have already heard from Mr. Ostry, and we will discuss in the next session, the analytical tools of the EWE are aimed not only at identifying country-level vulnerabilities, but also how these may translate into systemic risks and cross-border, cross-market spillovers.

• Second, our close collaboration with the FSB help us to pay particular attention to risks that originate in, or could propagate through, the financial sector.

• Third, a vital component of the exercise is examining risks that arise from capital flows and cross-border assets and liabilities.

We believe that these features, together with the dual quantitative-qualitative approach that Mr. Ostry mentioned, provide a sound basis for such an exercise. But of course it is not perfect, and surely there is scope for improvement. I am looking forward to hearing your ideas and suggestions on how to do so.

Thank you.


Public Affairs    Media Relations
E-mail: publicaffairs@imf.org E-mail: media@imf.org
Fax: 202-623-6220 Phone: 202-623-7100