Transcript of Conference Call on 2012 Spillover Report

August 2, 2012

August 2, 2012

Ranjit Teja, Deputy Director, Strategy, Policy and Review Department
Gian Maria Milesi-Ferretti, Deputy Director, Western Hemisphere Department
Petya Koeva Brooks, Advisor European Department

Ms. BHATT: Hi, everyone. Thanks for joining in. I know this has been a really busy week, with a lot of IMF reports out. So, welcome to this conference call on the 2012 Spillover Report. As you know, the Spillover Report examines the external effects of domestic policies in the five systemic economies, which compromise China, the euro area, Japan, the UK, and the U.S.

We have here with us Ranjit Teja, who’s the deputy director of the Strategy and Policy Review Department, and David Robinson, who’s the assistant director of the Strategy and Policy Review Department.

The report was basically prepared by an interdepartmental task force that was headed by Ranjit. So, Ranjit will be giving you brief opening remarks, after which we will take questions.

But we also have with us here other staff from the regional departments, and let me list them. Gian Maria Milesi-Ferretti, who is the assistant director in Western Hem. You probably were in the conference call this morning with him on the U.S. Article IV. Petya Koeva Brooks, who’s the advisor in the European Department. Stephan Danniger, who is the deputy division chief in Asia-Pacific Department. And Ashvin Ahuja, who’s a senior economist in the Asia-Pacific Department, as well. Ranjit, over to you.

MR. TEJA: Thank you, everybody, for joining this call. Perhaps I’ll just say a few words about the spillover reports. They’re still relatively new. This is only the second year in which we’re doing them. And I think it’s important to be clear about what a spillover report is and what it is not.

So, basically the goal of spillover reports is to voice the concerns and the stakes of the Fund’s membership regarding the policies of the Systemic 5 economies, and to bring a measure of quantification to their concerns.

The paper has been timed to coincide with the Article IVs, which many of you have reported on for the systemic economies, including the one for the United States this morning. The spillover report went to the Fund Executive Board alongside the Article IVs, and they were intended to showwhat’s in it for the rest of the world if the S5 heed the Fund’s policy advice or not.

Obviously, spillover reports do shed some light on the nature and scope for global cooperation. But the spillover report is not a grand policy prescription for the systemic economies. So, there’s no attempt to lay out a roadmap for global stability. It’s only to make explicit, to make quantified, what tends to be cast in very general terms in policy discussions.

Let me turn to this year’s report. Based on discussions with the Systemic 5, as well as with selected market countries listed on page 2 of the report -– we had a confidential discussion on their main concerns. None of these authorities are actually identified, so it makes for a more straightforward discussion than might otherwise be the case. And based on that discussion, we picked on a few issues and concerns of theirs.

Obviously, the most important one on everyone’s mind is the euro crisis, and, what might be the spillovers if it were to intensify. Secondly, there is U.S. fiscal prospects and on monetary policy and, more generally, monetary policy from all the advanced economies, not just from the United States. Thirdly, investment in China, it’s been very high, and, in fact, it’s risen, which raises the question: what would be the impact on others if this were, for some reason, to slip and to decline? Who might be affected by that and how much?

Of course, there are other issues. When one thinks of China exchange rate and currency issues always come up. We did not cover it this year because it got extensive coverage last year, so we didn’t repeat that work. And also, there has been a fair bit of China exchange rate coverage in the latest external sector report, which I understand was made public in the last few days.

Fourth, on Japan, we looked at the effect of a potential stress in the market for Japan government bonds, and how that might play out in the world? And more generally, we look at the effects flowing from Japanese foreign direct investment, which could pick up, for a variety of structural and cyclical reasons, in the period ahead.

And finally, on the UK and its financial sector, we had cast it last year as very much a global public good, and this year, we sought to make that hazy notion a little bit less hazy by trying to explain what is meant by that term, what’s the role of the UK fiscal sector, and to show how shocks might propagate further and deeper if they were to originate through the UK.

There’s also a fair bit of analysis of global systemically important banks, both in the UK, as well as in the euro area, and that also receives some coverage in terms of trying to bring some quantification to what might be the impact of the deleveraging from euro area and UK banks, were this to intensify.

I don’t want to repeat the results. It’s easy enough for you to scan. The report is relatively short, and it has a lot of pictures; you can kind of get the gist from those. Let me just give you a few highlights of the results.

With regard to the euro area, the first point that we make and document quite extensively is that the stresses from it are now truly a euro area phenomenon. It’s not just a matter of contagion from Greece or Portugal or whatever. The stresses are much more, now, common to the area as a whole. And as a result, they also are going to have more systemic effects.

We have a number of heat maps that might be of interest to you, simulating what would be the output losses from an intensification of financial stress. And later, I can describe to you what we mean by financial stress, for those of you who are interested in that, and who would be most affected. If you look at those maps, you’ll see that it’s within the euro area that the effect on output is most severe, as well as to the neighbors, most importantly, the United Kingdom, but also Central and Southeastern Europe. Deleveraging, so far, has been modest outside of the euro area by euro area banks and UK banks, but this could turn.

We also sort of look at what might cause the greatest amount of deleveraging. We find that funding shocks would be much more severe than sovereign valuation shocks as far as the deleveraging process is concerned.

Turning to the United States, I think our managing director has put a lot of emphasis on the U.S. as also a source of risk. Its not just the euro area. And obviously, the U.S. fiscal cliff that stands in the path of the U.S. recovery could have quite extensive effects around the world. We try to avoid boxing ourselves into any one econometric approach or model, so we do give the results for a range of models, and you’ll find that these effects range from modest to severe. But regardless of how you look at it, the effects on the neighboring countries, on Canada and Mexico, are always the most severe.

Monetary policy has been a point of considerable controversy. A number of emerging markets have complained about the easing monetary policy in the United States, euro area, and Japan, which is creating pressures on emerging markets in terms of capital inflows, asset price increases, and also raising the price of many commodities.

We find that the impact of recent monetary easing measures in the last year to year and a half is very hard to pin down empirically. Tthe empirical evidence here is not clear-cut. But I do want to emphasize that this does not mean that there is not an effect on emerging market asset prices. On the contrary, major easing episodes such as Q1 and LRTO by the ECB did raise asset prices and commodity prices.

But beyond those two specific episodes, it’s much harder to come up with generalizations that say easing everywhere and always causes--always pushes-- capital out to emerging markets. That’s much harder to say.

When it comes to saving the financial system in advanced countries, when it comes to big episodes such as in 2008, when QE1 was implemented, or at the start of this year, when LTRO was implemented, global financial stability was at stake. The benefit from easing was very clear to the emerging markets. Obviously, this is not the episode that they’re complaining about. Beyond that, however, it’s much harder to attribute capital flows to push from advanced economies. There is also a significant role for capital pull, because these emerging markets have very high growth rates, they have high returns on investment.

Finally, let me just pick up on one more theme, China. I think there is some very interesting work in this report on the impact of a potential decline in Chinese investment, and how that plays out in different countries.

I think the role of China in holding up many economies is not fully appreciated, and you might find some interesting charts there on, for example, the role of Chinese imports in helping to keep up German output. I mean, Germany is an economy that has done relatively well in a neighborhood that has been facing a lot of stresses. The role of China in this is interesting to observe. And, therefore, any change in that situation, if there were to be a decline in investment for any reason, would have important implications for many countries, including important capital exporters, such as Germany, Japan, Korea, and so forth.

I should also mention that there is a large set of background papers that provide the underlying detail and analysis. That will be also published in the coming days. We’re slightly behind on that, but should come out in the next few days.

QUESTIONER: Hi. Yes, you’ve already answered one of the questions, where are the background papers? I just wanted to clarify the way you use percentage here. When, for example, looking at table 6, you say that simulated output losses, for example, for the EU is at roughly 5 percent of GDP loss with feasible monetary responses, are you saying that that is a 5 percentage point difference or that it would be a 5 percent change?

MR. TEJA: No, it’s the difference relative to the baseline.

QUESTIONER: Ten years.

MR. TEJA: At the peak, you’re going to be 5 percent lower than otherwise.

QUESTIONER: Okay. You mean 5 percentage points, so.

MR. TEJA: Five percentage points.

QUESTIONER: It would be a 4 percent contraction, rather than a 1 percent.

MR. TEJA: That’s right, it would be 1 minus 5. Well, for each point, at the peak, whatever you would have been at the peak, you’re going to be 5 below that.

The peak in most of this work occurs roughly two years after the shock.

QUESTIONER: Okay. Secondly, in the part about you assume a 300 basis point rise in yields for some countries, the GIPSIS. And you say half of that has already been factored in. So, in your statement, saying that –- let’s see, where was it -- market access could be lost if the 300 basis point is realized -- increase is realized. Am I correct in the simple deduction that a 150 basis point rise on top of existing yields would make that statement true, that market access could be lost?

MR. TEJA: No, the simulation, because it’s a model, it basically behaves as if there’s no loss of market access in the model. So, the model just sort of works on the assumption that interest rates rise by 300 basis points. That’s part A. Part B assumes result of this increase, you have the same kind of behavior as we had in the fall of 2011, which is to say there is contagion to a number of other economies and also some safe haven flows.

QUESTIONER: I understand. Please forgive me for interrupting you, but in paragraph 9 –-you say the other half of the shock, meaning the remainder of the 150 basis point rise of the 300 basis point rise that you’re assuming, would obviously strain those countries, and possibly cut off market access.

MR. TEJA: Right. So, the reason for mentioning that is that in the normal course, the model doesn’t handle the scenario where market access is cut off as a result of 300 basis points. Nevertheless, it points out that even if you lose market access, the model has something valuable to say about this mere fact that the secondary market price is 300 basis points higher.

And so that’s where we are going with this, and I also want to make clear that we are not 150 basis points away from this scenario materializing. What really makes this scenario materialize is the contagion coefficients that we use. And so far, the contagion that we have seen in the last two months has not been of the same variety that we saw last fall. That’s why I said part B. Part A might happen, but unless part B also happens, and we have the same kind of correlations across the rest of the world’s asset prices, you are not going to see this scenario. So, the linchpin is the contagion to the other asset prices. That’s what’s not yet happened.

QUESTIONER: Okay. And then finally, can you just elaborate on the deleveraging of the EG SIBs?

QUESTIONER: I have two questions. First on the U.S. fiscal cliff, could you elaborate on how the United States can balance the near-term and the medium-term fiscal consolidation? For example, in five years or ten years, what objective should the U.S. Congressmen and the White House try to achieve?

Secondly, in your concluding thoughts, the last paragraph you mention in a case that would intensify euro area stress, the staff finds adjustment used automatic fiscal stabilizer in other systemic advanced economy reduce the output losses by 15 to 25 percent. Could you elaborate a little bit more on that? Let’s take the U.S., for example. Thank you.

MR. TEJA: Perhaps I can ask Mr. Milesi-Ferretti to answer the question, your first question, on what’s the right balance between near-term adjustment and medium-term adjustment for the U.S. deficit, and then I’ll turn to your question on fiscal.

MR. MILESI-FERRETTI: This is Gian Maria Milesi-Ferretti. I’m a Deputy Director in the Western Hemisphere Department, and I’m the Mission Chief for the United States. So we see the need for proceeding slowly on fiscal adjustment in the U.S. because we see a recovery that is still very weak and unemployment that is still high and protracted. Furthermore, there is very little scope for monetary policy to provide a lot of cushion if fiscal policy turns very tight because, as you know, interest rates are already at zero and you can do something more on the quantitative side, but there are limits to the efficacy of that. For these reasons we think you shouldn’t do too much in the short run. But it is absolutely essential to take measures that would make sure that taxes are raised and expenditures gradually reduced over the medium term because the U.S. has a fiscal deficit that is very high and that is rising rapidly.

How can we do that? Well, if you think that a lot of the spending adjustment has to come from entitlement programs from reducing the rate of increasing health spending, from also moderating the growth rate in pension spending, measures to achieve those objectives can be legislated today, but would have effect that will occur gradually over time. And we think this is a much better strategy because it would allow the U.S. economy to strengthen, the housing market to come back, and would sort of hit the U.S. economy in terms of reductions in spending and increases in taxes more heavy at a time when the economy is stronger.

QUESTIONER: So you are not worried that the U.S. politicians will use the economic recovery as excuse of kicking the can down the road?

MR. MILESI-FERRETTI: You point to a real concern, of course. What we are saying is that you should go slowly in the short run, but you should at the same time agree and legislate on a credible program to bring the public finances back over the medium run. And the point that I was making is that you can actually do that credibly in the sense that you can legislate measures today that are going to have some impact today, but a bigger impact in the future. So it is not just promising that you’re going to address the problem in the future. You can actually address it today, but just in a way where the biggest effects do not accrue right away. But we have been pushing in the past consultations with the United States as we’ve been pushing this year for a quick implementation of quick approval, quick agreement, on realistic medium-term plans to bring public finances under control.

MR. TEJA: Turning to your second question, where you asked what do we mean by the other economies using their fiscal stabilizers, automatic stabilizers, to reduce output losses by 15 to 25 percent. So if you look at chart 6 on page 6 you see the whole bar, the blue and the gray part, gives you the effect of the euro area shock on each economy. So euro area falls by about almost 6 percent. U.S.A. by about 2 percent. China by about 1.8 or so, et cetera. That’s the initial shock. And then we say euro area obviously in a crisis will not be able to let automatic stabilizers work. It will be difficult for them because they’ll be under market pressure to show that they’re adjusting and so on. But what if the non-euro area economies were to allow their automatic stabilizers to work and those that have monetary policy space that are quite far from the zero interest rate bound, if they were to reduce interest rates, how much would that help everybody? So that little gray area, which is between 15 to 25 percent of the initial loss, shows you how much you can reduce the loss by having policies respond in this manner in the non-euro area countries.

This model is based on 35 countries, so we’re not taking all the countries of the world into account here, but I think it accounts for more than 80 percent of global output this model. So it gives you a fair indication of what that kind of policy response -- how that would reduce everybody’s loss. Notice that the euro area’s loss also goes down, not because we have them doing something in this scenario, because other people reduce their losses and that helps the euro area’s exports to those countries and, therefore, that diminishes the loss also to the euro area. So that’s the explanation for that sentence and for this chart.

QUESTIONER: Hello. Thanks for doing this. I have a more general question, but maybe you might go into detail as well. I’m wondering if this is the second time that you do this report. What are the major changes you see in terms of spillovers of the yen, the extent of spillovers you see in this year compared to last year? Are there any spillover effects you see different than last year? You see bigger or less important?

MR. TEJA: If I had to summarize last year’s report in one sentence, it would be that spillovers are transmitted mainly through financial market channels. That result has not changed. We still believe that to be the case, but we have done a number of elaborations on that theme. First, the model that we use to show this effect has been improved and sharpened. Last time our model covered about 20 countries. This time it’s 35, so the coverage is more. And many of the somewhat smaller, but still important, emerging markets are now in there. They were not the last time, for example, Thailand. We also have a more sophisticated way of taking into account policy feedback. These improvements are under the radar and behind the scenes.

We’ve not just done modeling work. We’ve also done new and different work. I would draw your attention to the analysis of deleveraging from euro area banks, which uses a completely different conceptual framework. Also the work on China -- China’s investment-- I think is completely different from anything we’ve done before.

I think this year’s spillover report also had many more immediate concerns compared to last year’s. Last year we were still in a world where the world economy was generally healing and yeah, there were some hiccups and there was some slowdown and slow patches and so on. But the intensification of the euro area crisis since then has given this report much more of a tail risk kind of ethos compared to last year, where we looked at more general economic questions. The questions are much more pressing, and that reflects the discussions we had with policymakers because their concerns were much more sharply pointed at risks for the global economy.

QUESTIONER: Hi there. I wonder if you could talk a bit about the range of uncertainty and the effects of a shock in the eurozone and its spillovers. In particular, when we saw the Lehman shock, we saw that there were very, very deep losses of output in other countries, from Germany and Japan, but equally a shock from the eurozone has been very widely anticipated so maybe that would meet its effects. I wondered what the range of outcomes you simulated was. Thank you.

MR. TEJA: Your question is about our uncertainty or how this shock differs from Lehman?

QUESTIONER: Potentially both. The difference maybe contributes to the uncertainty.

MR. TEJA: Well, first let me just say that the range of uncertainty in the results is high because no economic model is going to be able to capture the very nonlinear effects that you have in a real crisis. All these models are essentially linear, and we give it more of a capacity to capture crisis -- to do a better job of crisis simulation--by imposing on them the kind of correlations we see in asset prices during stress times. That is the trick used to make an otherwise pretty linear model behave a little bit more like the real world does. But I would in no way suggest that this does an adequate job of capturing the crisis dynamics that unfolds in a real crisis. So you probably should view these as lower bounds in a real crisis in some aspects.

Your second question is about how might this shock compare to Lehman? This shock is intended to be a shock similar to Lehman, but of a longer duration. It is intended to be a comparable Lehman-like shock. It’s not just the usual increase in yields. It was very carefully calibrated, and the shock is more intense than the one that was explored in the April GFSR, the Global Financial Stability Report. So this is a harder shock so also you see harder effects.

QUESTIONER: Hi, yes, thank you. Going back to the charts, 12 and 13, it says deleveraging due to funding shocks percent of GDP, and I see that it looks like Sweden and France and Spain would all suffer a 25 percent of GDP funding shock. What does that actually mean? Greater than 25 percent of GDP funding shock?

MR. TEJA: I’ll ask Petya to explain. It’s not that the shock is 25 percent of GDP here. This heat map tells you that total change in leverage as a ratio to GDP. So bank balance sheets will shrink by 25 percent of GDP. That’s what the chart would say. Petya will explain what the shock is.

MS. KOEVA-BROOKS: So we have two shocks. The first one is the funding shock, which you see on chart 12. There the underlying assumption is that the unsecured and secured wholesale funding markets are under severe stress and that causes an increase in the counterparty risk. So as a result funding costs in both the euro and the U.S. wholesale markets increase significantly, let’s say by over 100 basis points. Now as a result of this there’s also an assumption about loss of confidence, which prompts a decrease in the value of the derivative market funding. So basically what I’m trying to say is that it’s a pretty sizeable funding shock. What you see there is the deleveraging pressure, which would be exerted on financial systems and economies across the world.

What this chart is not showing you is what the actual impact would be if there were a policy response or if there was a reallocation and adjustment on the side of banks in response to this shock. So this is really more of an indicator of the total pressure that would be exerted on these economies.

And the second chart is just another scenario, which is one of sovereign strength, in which we see a permanent decline in sovereign bond prices in several euro area countries. Again, what I should emphasize here is that the size of the shock is quite large and perhaps more importantly, its duration is long, which is why you see this impact. And the point of those two exercises is also just to show that funding shocks in many ways, the type of funding shocks that we also saw during the Lehman crisis, could in principle have a very large effect.

MR. TEJA: Let me just stress for emphasis one more time the point that Petya made about this heat map being more of a gauge for deleveraging pressure rather than an actual projection of deleveraging because if there is a funding shock, banks are going to do many things. They’re going to raise capital. They’re going to reallocate their assets from high-risk weights to low-risk weights. They’re going to convert their convertible debt and so forth into equity. So all kinds of responses will be endogenously taken and the final deleveraging that you’re going to get would probably be much less. So this is just a “naïve calculation” of what the deleveraging pressure is, but it does give you I think some sense of how bad it could be if it was not possible to raise capital and reduce these pressures.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6220 Phone: 202-623-7100