Transcript of a Press conference on the Analytic Chapters of the Global Financial Stability Report

September 29, 2010


With José Viñals, Financial Counselor of the IMF and Director of the Monetary and Capital Markets (MCM) Department
Laura Kodres, Chief of the Global Financial Stability Division
Jeanne Gobat, senior economist
John Kiff, Senior Financial Sector Expert, MCM Department
Conny Lotze, deputy chief of media relations, External Relations Department.

Wednesday, September 29, 2010
Washington, DC
Webcast of the press conference Webcast

MS. LOTZE: Good day and welcome to this press briefing of the analytical chapters of the Global Financial Stability Report, chapters 2 and 3. Chapter 1 of the report will be released next week on Tuesday, October 5, ahead of the Annual Meeting. On October 4, on Monday, we will be issuing a staff position note on assessing financial sector reform efforts to date and what further reforms are still needed in our view. So stay tuned for that release. You’ve had an opportunity to look at the chapters and the press points on the embargo and to remind everyone the documents and this press briefing are embargoed until 11:00 a.m. Washington time which is 1500 GMT.

Let me introduce the speakers today who will all make some introductory remarks. To my left is Jose Vinals, Financial Counselor of the IMF and Director of the Monetary and Capital Markets Department. Mr. Vinals will have to leave after his remarks, but he will present chapter 1 next week of the GFSR. To my immediate right is Laura Kodres, Chief of the Global Financial Stability Division. Next to her is Jeanne Gobat, Team Leader on chapter 2 on systemic liquidity risks. Next to Jeanne is John Kiff, Team Leader on chapter 3 on solvent credit ratings. Let’s begin with introductory remarks of Mr. Vinals.

MR. VINALS: Good morning to all of you and thank you for attending today’s press conference in connection with the release of the two analytical chapters of the Global Financial Stability Report. These analytical chapters tend to focus on some of the most relevant topics currently facing policy makers as they lay out their financial sector reform agendas to address systemic risks, something which has emerged as very important during this crisis. So let me provide a very brief overview of the main issues tackled these analytical chapters before giving the floor to my colleagues who will provide you with more information on the background and the policy implications and who will also be happy to answer your questions.

Chapter 2 focuses on systemic liquidity risk. One of the defining hallmarks of the recent financial crisis was the inability of multiple financial institutions to rollover or obtain new short-term funding. Financial institutions and regulators failed to account for the rising liquidity risks caused by an increased reliance on short-term wholesale funding. The chapter analyzes the reasons for this development and outlines a comprehensive approach for dealing with systemic liquidity risks moving forward. In particular, it looks at the increased role of money market mutual funds in providing bank funding at the expense of traditional depositors which was, perhaps, insufficiently recognized in the run up to the crisis. Policy makers will need to address liquidity risks both at institution level through high-liquidity buffers, but also at the market level through improvements in market infrastructure and practices in the repurchase market.

Chapter 3 takes a slightly different tact by looking at the role played by credit ratings in the recent crisis, particularly as regards to sovereign ratings. As the chapter explains, sovereign credit ratings have inadvertently contributed to financial instability reflecting the certification role assigned to credit ratings. Over the years credit ratings have become hard-wired in various rules, regulations, and private contract triggers implying that rating downgrades can lead to destabilizing spillover effects as investors rush to sell securities, in particular when downgrades go through the investment quality thresholds. These are the so-called cliff effects. The chapter recommends among other things to move towards reduced reliance on ratings and regulation, in fact, in line with current developments in a number of advanced countries. So I will leave you with these initial remarks and Laura Kodres and my other two colleagues, Jeanne Gobat and John Kiff, the chapters’ main authors, will elaborate further on the main messages. Thank you.

MS. LOTZE: Laura, please?

MS. KODRES: Thank you, Jose, for this introduction. Let me provide just a bit more context for these two chapters and why we have chosen to address these specific issues in this round of the GFSR before I turn it over to my colleagues, Jeanne and John, who will talk about the main findings for these chapters.

Three years after the onset of the global financial crisis, much has been done to reform the financial system but there’s still lots left to accomplish. We need to continue to work hard and examine what went wrong in order to implement corrective policies to make sure that we do not repeat the same mistakes again. Although very different in focus, both of the current chapters zero in on aspects of the financial system that inadvertently contributed to systemic risk and financial instability and, therefore, need to be addressed. We discuss excessive reliance on short-term wholesale funding in Chapter 2 and excessive regulatory reliance on credit ratings and various rules and regulations in Chapter 3.

As you’re well aware, the first round of regulatory reforms is well underway with the first steps focusing on regulations to build up capital and liquidity buffers in financial institutions, mainly in banks. In particular, the agreement reached to implement the quantitative liquidity requirements, as proposed by the Basel Committee on Banking Supervision in September of this year, is a significant step forward toward lowering liquidity risks. The rules will encourage banks to hold higher liquidity buffers and ultimately reduce the mismatch between the cash flows of their assets and the payment obligations on their liabilities. Still, the introduction of one of the new global liquidity standards aimed at ensuring adequately stable funding -- the so-called net stable funding ratio -- will be delayed until January of 2018 making it all the more critical to address systemic liquidity risks, perhaps the defining characteristic of this crisis, at its root which will include policies for funding markets, non-banks, and cross-border issues as well.

Chapter 2 addresses these issues head-on, pointing to the confluence of factors that contributed to the liquidity risk and emphasizing the need for a comprehensive approach to deal with systemic liquidity risk. As pointed out in the chapter, such an approach must also address how funding markets and non-bank institutions interact and how market infrastructure and practices work, including the key repurchase market or “repo” market. The merit of introducing some sort of surcharge or insurance premium to protect against system-wide liquidity shortages also needs to be further investigated.

Chapter 3 examines a different aspect of systemic risk, namely the role of credit ratings and how they contributed to financial instability during the recent crisis. In the aftermath of the crisis, and in particular against the backdrop of massive downgrades of structured credit products, there has been much discussion about the behavior of credit rating agencies and the implications of relying too much on ratings through rules and regulations. A key concern is that when downgrades occur or negative watches or outlooks are issued, those securities that fall below a given threshold force investors to sell their securities, sometimes simultaneously, causing cliff effects especially if that threshold is between the investment and the non-investment grade. Still, while there should be reduced regulatory reliance on credit ratings, the chapter acknowledges the important role played by ratings in reducing the differences in access that the credit rating agencies have to information and that available to investors, particularly for small investors in other countries and even in the United States. Hence, there is a need for a two-pronged approach to this issue. First, enhanced oversight of rating agencies and second, a reduced mandatory reliance in regulation. As well there needs to be measures to reduce the conflicts of interest that arise in the current “issuer pay” business model whereby the issuers of the securities pay for their ratings. Chapter 3 also focuses on the performance of ratings, per se, and how current methodology used by most rating agencies might be contributing to financial instability.

Let me now turn to my colleagues to take you through the main messages of each of the chapters. Thank you.

MS. GOBAT: Thank you, Laura. In Chapter 2 we review the role institutions, regulations, and key markets played in contributing to the systemic liquidity crisis which in some ways is still with us today. The excessive reliance by financial institutions on short-term sources of wholesale funding contributed to the widespread buildup of liquidity risk in the financial systems. We show that a host of factors led to this trend shift in funding structures. These include deregulation and global financial integration, making it easier for financial institutions to access markets including across borders, across currencies, and across sectors. Moreover, regulations under the Basel 2 framework favored repurchase transactions, so-called repos, a form of secured lending over unsecured funding. In the United States, key participants in the shadow banking system, such as money market mutual funds, have played a central role in expanding the short-term wholesale market. We find that they played a bigger role in the United States than elsewhere in the global financial system in part because of favorable regulatory treatment.

We make a number of recommendations to strengthen funding markets and lower systemic liquidity risks. Here we need to bear in mind this area remains work-in-progress and is complex. Solutions must deal not just with institutions but also on how institutions and markets interact and identify factors that contribute to potential liquidity shocks across markets. As to solutions for institutions, the report welcomes the recent prudential liquidity rules proposed by the Basel Committee on Banking Supervision. These are steps in the right direction. They raise liquidity buffers, encourage banks to reduce the mismatch between the cash flow from their assets and the payment obligations on their liabilities, and lower the chances of a systemic liquidity event in the future. It will be the first time that such quantitative rules will be applied at a global level. That said, our report indicates that more analysis must be done to assess the effects these new rules might have on how banks fund themselves and on market liquidity.

In our view, more needs to be done to lower systemic liquidity risk in the global financial system. One of our key messages in the report is that policy makers should consider applying similar liquidity risk guidelines to non-bank financial institutions. Such action would offset any excessive buildup of liquidity risks outside the banking system and avoid regulatory arbitrage sectors. Generally, in our view, institutions that rely excessively on short-term wholesale markets for funding should be required to maintain high-liquidity buffers that can be quickly converted to cash and act as cushion against shocks.

Another key reform area is to strengthen collateral valuation and margin practices in repo markets. A market in the United States, which at its peak in 2007, was estimated at $10 trillion U.S. dollars and which was of similar size in the Euro area. The Committee on the Global Financial System, a forum in which we participate, and the U.S. Triparty Repo Infrastructure Task Force, has put forward a number of reform recommendations to strengthen risk management practices in the repo markets. These recommendations we strongly endorse and include, for instance, more frequent valuation adjustments, validation by supervisors of institutions’ in-house models, more realistic assumptions about liquidation, periods for collateral, and applying a longer term time framework for valuing collateral. Implementation of these recommendations would lower macro prudential risks in repo operations and discourage the procyclical effects that we discuss in the report, including the ability to obtain excessive funding when collateral values are high and margins are low.

We also recommend that more information should be released to market participants to improve pricing and transparency practices, both in the unsecured and secure funding markets. For instance, consideration should be given to having third parties provide on a regular basis data on margins used in collateral for repo transactions as is being done by the U.S. Federal Reserve which is posting repo information monthly on its Web site since earlier this summer.

We also emphasize the benefits of encouraging greater use of central counterparties. As is similar to the case of the OTC derivatives which was discussed in this year’s April GFSR analytical chapter, in our view central counterparties are an efficient market infrastructure solution to lowering systemic risks in repo transactions. They provide common margin requirements. They reduce the risk that a counterparty will not deliver on the collateral and handle collateral efficiently.

Finally, while reforms are underway to strengthen the money market mutual fund industry both in the U.S. and in Europe, we stress that more needs to be done to lower their contribution to systemic liquidity risk. Here we propose that money market funds should overtime have to choose whether to be overstated and regulated as banks with their liabilities treated as deposits with established guarantees or become mutual funds whose net asset value fluctuates. Moving towards floating net asset values should lower the risk of liquidity runs of money market mutual funds because investors would be more aware of their investment risk and thus be more inclined to actively monitor the liquidity risk that their funds assume. Thank you.

MR. KIFF: Thanks, Jeanne. Chapter 3 shows how credit rating downgrades can inadvertently destabilize financial markets. However, it’s not necessarily the ratings per se, but the way that they’re hardwired into various rules, regulations, and triggers. Hence, downgrades can trigger destabilizing knock-on effects and spillovers throughout the financial system. None of this is to say that ratings don’t serve a useful purpose. They aggregate information about the credit quality of various borrowers and their obligations, and that, in turn allows issuers to access global and domestic markets and attract investment funds and adds liquidity to markets that would otherwise be highly illiquid.

Now in previous GFSRs we have covered the structured credit rating debacle quite extensively. In this chapter we focus on sovereign ratings and our empirical analysis shows that rating agencies improved their sovereign rating inaccuracy since the Asian crisis when they cut themselves a rather bad reputation, but they could pay more attention to their short-term debt and contingent liabilities. At the same time, though, sovereigns could do more to provide such information, and we would provide some examples of how that could be done better in the chapter.

Beyond their information role, we show that ratings play a significant certification role. By this we mean that they certify or verify that issuers and securities are of a given minimum credit quality. For example, we found that downgrades from investment to speculative grade led to statistically significant widening of credit spreads of the downgraded sovereigns. This suggests that some sort of decertification takes place. This can lead to adverse feedback loops and ultimately higher borrowing costs for those sovereigns.

We also show that some aspects of rating agency downgrades smoothing policies can inadvertently create procyclical rating cliff effects. The problem is not the way that most rating agencies rate through the cycle; this rating through the cycle is actually a smoothing policy that introduces useful rating stability. The problem is with policies that bottle up potential downgrades that their rating criteria and new information would point to. So if downgrades do finally occur, cliff effects can be extreme. But why the excessive reliance on ratings? Well, basically ratings have come to be seen as objective and easily verifiable third-party opinions. Generally speaking, it became easy both for regulators and private entities to rely on ratings as objective benchmarks, after all, they did a pretty good job until the recent structured credit crisis. Now we find them everywhere. For example, many central banks use credit ratings rather mechanistically in their rules that determine the securities that’s they accept as collateral.

The Basel 2 standardized approach to determining bank capital requirements is also very reliant on ratings.

Ratings play a similar role in private financial dealings, for example, when securities are posted as collateral. Also, private financial contracts often contain rating triggers that end credit availability or accelerate borrowers’ credit obligations if a downgrade occurs. In addition, many investors are obliged to sell securities that are downgraded below certain rating levels; for example, the BBB- level that defines the lower end of the investment grade range.

But the authorities are now taking steps to reduce this mechanistic reliance on credit ratings. This will imply material changes in official sector and market practices and will present challenges for authorities and firms in developing alternative ways of evaluating credit risk. However, it is important to watch out for unintended consequences and recognize that smaller and less sophisticated institutions will continue to use ratings. That’s why we call for rating reliance reduction to pay attention to the size and sophistication of institutions and the instruments concerned. That’s also why we stress the importance of continuing to push rating agencies to improve their procedures, including transparency, governance, and the mitigation of conflicts of interest. In particular, we call for rating agencies whose ratings are used to calculate bank regulatory capital requirements to meet more rigorous validation standards, basically the same ones that banks need to satisfy when they use their own internal models.

However, it seems a little too late to try to force the major rating agencies to switch from the business model where the issuer pays for the rating to one where the user pays, a well known source of conflict of interest. Under the issuer pay model, the issuers can shop for the highest ratings, which incentivizes the rating agencies’ desire to win their business by lowering their standards. Nevertheless, the issuer pay model is more efficient in that the investor pay model would require lining up numerous individual subscribers while ensuring that only those subscribers had access to the ratings in order to eliminate free riders. In any case, just like the music business, locking out free riders in this information society would seem to be a virtually insurmountable challenge. Some have put forward the idea of a public sector rating agency as a way of resolving conflicts of interest and biases and so on, but it would have its own conflicts of interest especially on sovereign ratings. Nonetheless, further exploration and variations on the issuer pay theme is warranted.

The bottom line is that the importance of ratings should be reduced and the current regulatory license given to rating agencies eliminated. Looking ahead, credit ratings should be seen as one of several tools to measure credit risk, not as the sole and dominant one. Thank you.

MS. LOTZE: Thank you very much, John and everybody. Let me remind those watching on line to send questions, if they have any, so we can answer them here. We’ll go to questions now. Are there any questions in the room? Yes?

QUESTION: On page 23 of Chapter 2 you list a number of proposals for this insurance, systemic risk insurance. Are these proposals that the IMF considers -- does the IMF support? There hasn’t been discussion by the IMF itself of these proposals, the merits, so an explanation of that backing of those proposals would be helpful.

MS. LOTZE: Laura?

MS. KODRES: Yes, we list, I think, four such proposals. Some of them have been made by academics and other policy-oriented experts. Basically at this point we don’t want to endorse any one of these proposals. I think that this particular stage of the regulation reform process, where we would measure and apply some type of surcharge or premium or tax or levy to account for systemic liquidity risk, is still not advanced enough to a stage where we know which of the following answers would be the most effective. And so at this point we wanted to put down some of the ideas. We want to make sure that people understand that this is on the table. It’s under discussion. It’s useful to put a number of ideas on the table at this point and then evaluate them. We haven’t evaluated them piecemeal, one by one. What we would need to do is take a close look at what we think the underlying premise or the underlying roles are of systemic liquidity risk. And as we point out in the chapter, this is complicated because it involves not just institutions, but the markets through which those institutions operate. And so to measure systemic liquidity risk will be quite a challenge to start with. Then the next step is to connect that to some type of charge that can be applied to the institutions that provide liquidity, both banks, and we would emphasize, non-banks. And we’re still at the initial stages of this process and, therefore, we don’t want to give a final answer about which approach is likely to be the best. Thank you.

QUESTION: Do you have some sort of timeline that you would think that it would be necessary to implement such a insurance premium?

MS. KODRES: We don’t have a timeframe set out. We are constantly looking at this topic ourselves. You may recall that in the last GFSR in April, we made a proposal or suggestion about a capital surcharge that would be applied in the case of solvency. We felt comfortable that we had gotten far enough along on that dimension to put something on the table also for discussion. In this area we don’t feel that we’re ready yet to put something on the table ourselves. I would hope that within the next six months to a year we come to some notion of how to measure systemic risk well enough to be able to do that. This is one of those highly speculative areas where the problem is extremely difficult. And so I think the approach that firms and regulators have taken so far, which is to raise liquidity buffers and try to reduce maturity mismatches, is the correct initial step and those will go some way to lowering the chances of finding oneself in a liquidity event, a systemic liquidity event. And so those are valid approaches to begin with. The next approach is quite a bit more difficult and that’s why we’re continuing to study it and did not provide any final answer on what the best approach is.

MS. LOTZE: Any other questions?

QUESTION: Yeah, I was just wondering if you could give us kind of a big picture assessment of regulatory reform efforts around the world, not just in the U.S. but also Basel III and what’s going on in Europe in terms of addressing these issues and whether or not you’re sort of satisfied with some of these.

MS. KODRES: Let me just mention to start with that next week we will release a Staff Position Note that will tackle that point head on. Because Basel III is already out there and on the table, let me just express our views about that. You know, we certainly welcome the proposals of the Basel Committee, and they make a substantial contribution to the quality and quantity of capital. And I’ve already spoken about the liquidity measures. The common equity will now represent a much higher proportion of capital than it did before. The definition of capital will contain only a limited amount, up to about 15 percent, of certain intangibles and qualified assets. These will include deferred tax assets, mortgage servicing rights, substantial investments in common shares in financial institutions, and other intangible assets. The phase-in arrangements have been developed to allow banks to move to these higher standards mainly through retention of earnings. And at the global level, that will stabilize and help the world economic recovery. At this point, if the global financial system stabilizes and the global recovery is firmly entrenched, we would propose that we phase out these intangibles completely and scale back the transition period. Those are ideas that could be considered if we have a stronger growth and a more stable financial system.

Certainly moving in this direction will raise the banking sector resiliency to absorb future shocks, and it’s essential that we make progress at least in this dimension to begin with. But I think it is safe to say that putting in this micro-prudential regulation is not sufficient. Appropriate regulation needs to be developed to have what we call a macro-prudential overlay. That overlay should dampen pro-cyclicality and limit the systemic effects of financial institutions, some of which are not banks. And that is our motivation toward a systemic risk-based surcharge of some type, both on the solvency and the liquidity dimensions of financial institutions. I think you should look forward to Monday’s release of the paper to find out more.

MS. LOTZE: Do you have another question?

QUESTION: The other day John Lipsky said that the transition period for the Basel III standards could be accelerated without hurting or crimping the recovery. Can you give some sort of timeline or timeframe in terms of that accelerated framework that he’s or rather the IMF is considering? And secondly, to what extent would you say that the U.S. financial restructuring has encompassed in its parameters the shadow banking system that you are warning about?

MS. KODRES: On the first question, the timeframe for the introduction of the Basel III on capital, I think the studies that Mr. Lipsky was citing are over a four-year time period. And so those studies show that over that time period, economic growth would not be harmed greatly by introducing both the liquidity and the capital requirements. In terms of the –

SPEAKER: I’m sorry. Could I just interrupt you? Do you mean all of them, the ones that are proposed for I think final implementation is something like 2018 or 2019?

MS. KODRES: I think the study took the original Basel proposal before two weeks ago and asked the question, if those were being introduced over a four-year time period, what would be the effect on economic growth?

In terms of whether the U.S. reform efforts have gone far enough with respect to the shadow banking system or the non-banking system, let me just note that there are several aspects that will, in fact, impact the non-banking sector. As Jeanne mentioned, some of the initiatives are about governing money market mutual funds, so there are a number of new policies that will make these funds safer. There are other areas too where initiatives are taking place though not necessarily covered in the Dodd-Frank Bill per se. We mention in the chapter the Tri-party Repo Infrastructure Task Force that also is attempting to improve collateral practices and margining practices in the repo markets. Again, this will go some way to alleviate some issues in the non-banking area. In general, it’s easier to examine banks because we’re comfortable with them. We know what the regulatory environment is. We have in this country several banking regulators. So we think that there is more to be done on the non-banking area. Not all of it has been done. We have made some inroads, but I think there’s still more to do in that area.

MS. LOTZE: Any other follow ups in the room? We don’t have any question online, so if you don’t have any more -- do you want to have another follow up?

QUESTION: On the ratings agencies, I’m having -- forgive me, I’m still learning all of the technicalities here -- I’m having difficulty understanding exactly the depth to which you are calling for separation or government reliance upon the rating agencies. You talk about that separation. Can you explain that a little bit further?

MR. KIFF: Well, among the places that we see this over reliance, for instance, is in central banks. They use ratings to determine not only the collateral they accept, but the margins that they apply to them. When we speak of government reliance, we mean it rather broadly. It would also include the Basel II requirements. Also you’ll see that the securities regulators often embed credit ratings in criteria say for what money market funds can hold, and you find them sprinkled all over the place. And in most cases, they’re in rather mechanistic ways. And, in fact, if you look at the Dodd-Frank Act and legislation that’s coming through the European Parliament, work is being done to remove that reliance. But it can’t be done overnight because you have to have a replacement for the ratings. And so it’s going to be a long process, and we’ve seen instances of unintended consequences along the way. So, again, you have to tap the brakes and be a little bit careful.

QUESTION: So are you talking about, say, the central banks developing in-house expertise? So wouldn’t there be a problem in terms of standards across different central banks or just across the industry itself if you’re having different in-house expertise with all these different standards?

MR. KIFF: That’s an example of the reason why we have to be careful, but it is being done in a very systematic piecemeal basis. I believe the Fed, the New York Fed, has already introduced some of its own analytics into its collateral operations for some of the funkier stuff they’ve been accepting in the course of the crisis and so on -- they’ve widened their collateral, acceptable collateral, in some ways. And in that regard, they didn’t feel comfortable relying on ratings. So it can be done. But we also have to recognize -- and we’ve said this before -- that you have to recognize the level of sophistication and scale of the banks. So a big central bank like the Fed or the Bank of England, they have the scale to be able to do this. Perhaps a smaller central bank in an emerging market country can’t do that so then maybe they will have to require some rating reliance, but it is that mechanistic reliance that we have to be wary of then. There should be some thought process beyond the rating as an acceptability criterion.

MS. KODRES: Let me just emphasize with John that -- these two chapters are all about moderation. And just as your doctor would tell you in your annual physical that consuming food and alcohol in moderation is fine for a healthy lifestyle, similarly we would say the moderate use of short-term wholesale funding and the moderate use of credit rating agencies in the context of investment decision-making are also acceptable for a healthy financial system. And so it’s not that we think credit ratings are a bad thing. No, they’re actually a good thing used in moderation. And similarly, short-term wholesale funding can help banks with short-term issues having to do with their funding processes, and again this is a very healthy situation. It’s the over-reliance on those two elements that has caused difficulty in the crisis, and that’s what we want to emphasize as needing corrective action.

MS. LOTZE: Thank you, Laura, and thank you, John. Do you have any follow ups? We still don’t have any questions online. It seems like all the questions were answered and documents speak for themselves. So we will wrap it up at this point. Thank you very much. Let me remind you of the embargo at 11:00 Washington time, 1500 GMT. Thank you very much for participating and goodbye.

IMF EXTERNAL RELATIONS DEPARTMENT

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