Introduction

1. Overall, the External Advisory Group (EAG) considers the 2014 Triennial Surveillance Review (TSR) to be a comprehensive and balanced account of the state of, and challenges for, Fund surveillance. We agree that the Fund has embarked on a serious effort, particularly since 2011, to deepen and broaden the coverage of surveillance. The Fund has appropriately focused its surveillance on risks and spillovers, begun implementing the Integrated Surveillance Decision, and continued to improve and expand its suite of multilateral surveillance reports. This has helped to plug many of the gaps identified by the crisis. However, it will take time for these seeds to bear fruit, while new challenges are emerging and lessons from the crisis are still being identified. In this regard, further work is needed to refine and deepen analysis, and to provide more tailored advice.

2. In this spirit, we broadly endorse the findings and proposed recommendations. In some areas, we also offer our own perspectives on the challenges and priorities ahead. Before turning to the specific findings and recommendations, we emphasize two cross-cutting issues that are essential for maximizing the impact of Fund surveillance. We encourage the Fund to take a bold approach in both areas.

• With the new initiatives, surveillance messages are too diffuse—the key challenge is to pull these together into a coherent whole that can be digested by member countries and other stakeholders. In this context, the TSR background studies highlight that the array of products has become almost impossible to digest and absorb. In our view, the volume of new products is a symptom of a disconnect between the producers and consumers of surveillance. We strongly recommend streamlining, mainstreaming and integrating the various strands of work.

• Moreover, analytical improvements will be largely irrelevant if they do not reflect the needs of member countries. They risk becoming simply good analysis that’s left on the shelf. The Fund needs to shift gears and think more about how it engages with and influences its members to ensure that surveillance is more focused on their needs and capacities. This means thinking about communication from the start of the surveillance process—as a vehicle for a better policy dialogue, and more relevant analysis and advice—rather than as an afterthought. It also means better communication and collaboration within the Fund, including a sea change in how information and knowledge are shared to help deliver advice grounded in cross-country policy experiences.

Strengthening Risk and Spillover Analysis

3. Given the overarching challenge of pulling it all together, the Fund needs to prioritize as much as possible and call attention to the big risks. The Fund is trying to do too much. By trying to spot every risk “under the sun,” it is in danger of missing the big risks. Admittedly, it is not always possible to tell the difference between the big and the small risks. Nevertheless, the Fund needs to focus on identifying the most egregious fault lines—even those risks that seem remote, but would be consequential if they materialize. Several elements of the TSR help move the Fund in this direction.

• We fully support the recommendation to bolster risk and spillover analysis. The Fund’s spillover analysis should be viewed as a global public good. For spillover analysis to get traction with policymakers in systemic economies, it needs to be as credible and persuasive as possible. One avenue for appealing to members’ self-interests is demonstrating clearly the potential for “spillbacks”—feedback loops from spillover recipients to the source countries. We also feel that improved analysis of tail risks—high impact, low (but not insignificant) probability events—can help to generate traction, both bilaterally and multilaterally.

• The Fund should forge ahead with embedding macro-financial surveillance. This priority element of the surveillance reform agenda remains incomplete. We welcome the recommendation to make it a centerpiece of surveillance. In particular, the Fund should take a leading role in mapping the impact of global financial reform on the real economy. To facilitate this, analysis of macro-financial interconnections at the bilateral level (in Article IV reports and Financial Sector Assessment Programs (FSAPs)) should feed more systematically into multilateral products like the Global Financial Stability Report (GFSR).

• At a more granular level, surveillance needs to distinguish more clearly between gross and net capital flows, and the stocks these create. Focusing on net flows helps identify macroeconomic risks, but the crisis put beyond doubt that analysis of gross flows is also needed to identify financial risk. In the past, surveillance has been overly focused on flows, with less attention paid to stock positions that have been important drivers of major recessions and crises. At the same time, Fund surveillance should also be attuned to changes in stocks and balance sheet interactions that are not just determined by flows, but by changes in asset prices and derivative positions.

• In that context, we strongly agree with the recommendation to revive balance sheet analysis. The Fund needs to analyze national balance sheets and those of private entities at the core of the global financial system. Balance sheet analysis is a critical tool for bringing global interactions and domestic vulnerabilities into full view. This group views the Fund not following through with balance sheet analysis in the wake of the Asian Crisis as a major professional failure. This should be at the top of the priority list from this TSR.

4. Tying this all together, we agree with the call for a more eclectic and flexible approach to risk analysis. The crisis broke the mold for how we think about economic problems and policy solutions. The Fund responded with new analytical and policy approaches. Staff will need to continue to venture outside their comfort zone, engaging in new policy areas without broad, established economic consensus and relying on judgment. In this context:

• Models are no substitute for judgment, particularly in assessing the most extreme or tail risks. With much of the world in uncharted territory, trying to navigate advanced economy exit from unconventional monetary policy, there is an even greater premium for staff to draw on diverse information sources and judgment.

• The Fund needs to show humility, acknowledging the limits of its understanding and avoiding misplaced certainty. No matter what improvements are made to risk and spillover analysis, mistakes will be inevitable. Staff needs to come to terms with the fact that they will get some things wrong, but the feedback this generates from the membership is an important accountability mechanism.

5. Further action is needed to improve data availability to prevent it being the Achilles heel of enhanced risk and spillover analysis. While there is sufficient publicly available data to begin much of the core risk analysis, data availability and data sharing continue to be a problem. The Fund should establish an institutional plan to articulate the priority outstanding data needs, along with the adverse consequences for surveillance if gaps remain. In this respect, balance sheet analysis can help the Fund identify the “known unknowns.” Looking closer to home, the Fund should take steps to deal with the data sharing problems that exist even within the Fund, many of which are highlighted in the TSR background studies.

Tailoring Fund Policy Advice

6. We support the proposal for the Fund to have a stronger role in providing advice on macroprudential regimes and policy. Macroprudential policy is an imperative part of the policy toolkit for addressing the big macro-financial risks, including those identified by balance sheet analysis. This is an emerging policy area, where many jurisdictions do not yet have a fully articulated regime and where policymakers are still finding their way. For example, there is still significant uncertainly around how the various new agencies with macroprudential responsibility (particularly in large advanced markets) will operate and coordinate, and this is likely to be—and should be—a major focus for surveillance in the years to come. In particular, we see the Fund as well positioned to assess whether member regimes and policies meet good practice and are globally compatible. While a global perspective is an area of comparative advantage for the Fund, it will need to build its own capacity and strengthen its guidance on macroprudential policy if it is to meet this challenge.

7. On balance, we agree with the proposed approach for the Fund to be selective in its role in structural policies. Given the growing, or at least more apparent, importance of structural policies for macroeconomic and financial stability and potential growth, it is clear that the Fund should play a role in this area. For instance, most of us feel that some involvement in labor market policies is appropriate. While our views differ on precisely where the Fund should draw the line, we support a principles-based approach to guide Fund involvement. We start from the position that surveillance should cover all structural areas that are macro-critical—those that have a significant bearing on stability and growth—and that this will vary from country to country. With that in mind, it is important to distinguish between: (i) recognizing the macroeconomic implications; and (ii) providing detailed policy advice in specific structural areas. The former should always be an essential component of Fund surveillance, but the latter will not always be necessary or appropriate. The Fund can help to identify the nature and size of a problem, but the structural solution may depend on local circumstances and the expertise required. In many politically sensitive and complex structural areas, misjudged policy advice can be counterproductive. There is also a danger that the Fund may be pulled in too many directions—it cannot be all things to all people.

8. There is room for surveillance to be more attuned to the specific challenges that low-income countries (LICs) face. In particular, the recent wave of natural resource discoveries in many LICs has made intertemporal sustainability a key policy issue. The most important policy decision for the next decade will be deciding the proportion of related revenues that are saved and used to accumulate other assets. The use of future revenue streams as implicit collateral for the issuance of sovereign bonds is likely to have much broader debt sustainability implications. In this context, a public sector accounting of assets and liabilities would help the Fund to better assess government solvency.

9. More broadly, the Fund has within its reach tremendous resources through which it can deliver more value-added surveillance. Member countries have consistently called for more expert and more tailored advice. Delivering this will be a real challenge. The Fund sits atop a deep pool of information and cross-country resources. However, we agree with staff: putting this into practice will depend on a combination of better exploiting the Fund’s existing knowledge and expertise, stronger engagement with other expert agencies, and judiciously building additional capacity. We strongly encourage the Fund to actively pursue the recommendations in this area.

• Big ticket items and a menu of policy options. Bilateral surveillance should focus on two or three priority policy areas, rather than generating the same laundry list of reforms year after year. In this respect, we feel that the Article IV process is at risk of becoming ritualistic, with staff drowning in procedural justice. Given the inevitability of political constraints, we feel that the Fund could shift further toward delivering a menu of potential policy options, without forsaking the preferred advice. While conjunctural surveillance is inherently focused on near-term priorities, macroprudential surveillance is also focused on vulnerabilities that could undermine an economy further out. Where serious faultlines exist, advice should also be about the “long game.” For example, helping to develop or reinforce domestic institutions and mechanisms (e.g., fiscal councils) could, over time, help (particularly in LICs) to strengthen ownership and accountability from within.

• Make greater use of institutional knowledge of cross-country policy experiences. The Fund is uniquely placed to do this and member countries have called for this type of advice year after year. We feel strongly that more needs to be done to strengthen information sharing to more systematically exploit the Fund’s knowledge base and promote internal and member country collaboration in specific areas of research. In particular, we would emphasize the value in identifying lessons and experience in countries facing similar circumstances and policy choices, both within and, importantly, across regions. This is an area where the Fund can learn from other organizations.

• Expand in-house expertise and/or draw on other international agencies’ expertise. This is particularly relevant for the Fund’s involvement in structural issues, which will inevitably test the limits of its expertise. The Fund will need to work out where it should build resources in-house, or draw on expertise in other institutions. Significant analysis and public debate already exist in many areas, so we agree that the Fund should not reinvent the wheel. On the perennial issue of country expertise, the Fund should continue to guard against excessive staff turnover. Longer country assignments—staggered to help dilute capture—can help enhance institutional memory, and the Fund’s ability to provide genuinely expert country analysis and advice.

Communication and Engagement

10. We stand behind the desire to improve communication and engagement. Many of us consider that it is time for a fundamental rethink of how the Fund engages with its members and other important audiences, and how this is integrated into the surveillance process. Communication should not be a last-minute “PR exercise”; instead it is essentially about how to explain the merits of the Fund’s prescriptions to local decision-makers and the communities they serve. We agree with many of the findings and proposed recommendations of the review and would emphasize the following points:

• More member-focused surveillance. Staff is often too focused on internal processes and needs to make more room for listening and being more aware of the constraints on policymakers. This approach is a prerequisite for the Fund to better understand country needs and experiences that can be the basis for more tailored analysis and advice. In this respect, we welcome the proposal for more feedback on the quality of policy dialogue and engagement. The background studies also suggest that the communication challenge begins at home—some parts of the Fund have not yet fully absorbed and embraced the idea of integrated surveillance. We would encourage the Fund to continue to tackle this internal dialogue challenge.

• But also more courageous. The Fund should not shy away from raising key issues, giving difficult messages, or pushing for policy reforms that lie outside the boundaries of a country’s local debate. The Fund should advise members to do what it believes to be the right things, which they may or may not like, and they may or may not follow—that dialogue is the nature of surveillance. In this regard, the Fund has an important role to play in influencing opinion-makers and facilitating broader public debate. We support the call for candid surveillance on both fronts. In particular, the Fund should go out of its way to be candid with systemic economies, including on important tail risks, given the perceptions that it has not always done so.

• And streamline multilateral products. The Fund needs to resolve the multilateral surveillance product overload. Staff recommends integrating the messages across multilateral products, including incorporating messages from the Spillover and Pilot External Sector Reports into the World Economic Outlook (WEO). While we agree with this approach, we believe a more fundamental rationalization and consolidation might be warranted over time. Our views differ on how best to achieve this. Some think that merging the WEO and GFSR would be entirely consistent with the drive to integrate macro-financial surveillance. Others feel this would be going too far, especially as they both have an established reputation and readership. Another option would be to preserve the flagships, but subsume some, or all, of the newer products within them. We urge the Fund to explore in more detail the pros and cons of different streamlining options.

Evenhandedness

11. We recognize the widely held concerns about evenhandedness. This issue has the potential to jeopardize the credibility and legitimacy of surveillance. At the same time, this is not a surveillance-only issue. Issues and perceptions of evenhandedness are driven by much broader factors than surveillance, including quota and governance reform and program conditionality. Thus actions taken in the context of surveillance are only part of the solution.

12. Nevertheless, the different perspectives within this Group reflect the broader lack of consensus on what evenhandedness actually means. In this sense, some of us question whether the term evenhandedness is useful or whether the Fund should focus on providing value-added advice rather than evenhandedness per se. We therefore agree with the proposals to bring clarity and accountability to this issue.

• Reporting mechanisms. As the background study on evenhandedness highlights, there is some sense that pressure to dilute the candor of Article IV reports for systemic economies comes from the high level of attention (including within the Fund) on these countries. A reporting mechanism could be an important tool for tackling these pressures or tendencies towards self-censorship within the Fund, as well as perceptions more broadly. Members also have a responsibility to report any concerns transparently, including through the Executive Board.

• A clearer understanding of how to assess evenhandedness, based on surveillance inputs. We believe that an emphasis on inputs could be helpful as a new framework for thinking about and gauging evenhandedness. The Fund is a learning organization—as it learns from experience, it will, and must, approach similar situations differently over time. Some of us feel that the Fund needs to be better at defending itself by explaining the evolution of its policy advice and combating misperceptions. In that respect, this clarification could be a useful step forward. Most of us think it is not necessary or desirable to have equal surveillance “inputs” and “outputs” for all countries. Indeed, a risk-based approach, focusing on systemic or high-risk countries, should help alleviate evenhandedness issues.

Global Cooperation

13. Effective policy outcomes require global discussion and cooperation, so surveillance must look beyond the bilateral policy dialogue. Given the Fund’s universal membership—as well as the ostensible decline in the effectiveness of other fora, such as the G20—some of us feel it is time for the Fund to reassert itself as the forum for global economic and financial cooperation. Others caution against being overly ambitious and setting the Fund up as the champion of global cooperation—in the first instance, the Fund should focus on getting its analysis and advice right. Looking ahead, we see some merit in the proposal to appoint an expert group to explore the pros and cons of strengthening the Fund’s role in cooperation. This should include the potential implications of updating its structures and mandate over time.

Resources and Prioritization

14. Some of the proposed recommendations in the TSR will have resource implications, raising difficult issues of prioritization. Given our broader concerns that the Fund is trying to do too much, we feel strongly that to create the space to implement the recommendations, Fund management needs to identify areas where the institution will do less. However, doing less is not one of the Fund’s traditional strengths. Difficult decisions need to be made to ensure that the Fund’s attention and resources are not diverted away from analyzing the big risks. The Fund should also seriously consider ways to increase its agility in allocating resources to emerging challenges or new priorities. Although we did not discuss the issue of Fund resources at length, we start from very different positions. Most of us feel that the Fund’s current resources are adequate to meet the challenges ahead and that this should not even be open to debate. Others feel the current resource envelope may not be adequate, particularly given the pre-existing strains and increasing demands on the Fund. We do not wish to prejudge any budgetary discussion by the membership—we firmly believe that it is for them to discuss—and own—any required trade-offs.