Transcript

Mr. Ranjit Teja, Deputy Director, IMF Strategy, Policy and Review Department
Gita Bhatt,External Relations Department
Friday, September 2, 2011

MS. BHATT: Good afternoon. Thanks for joining this conference call on this report that we're going to be publishing today, the “Consolidated Spillover Report, Implications on the Analysis of the Systemic Five.” You will have probably already seen the report as well as the IMF survey article on the OMBC. Without further ado, I will hand this over to Ranjit Teja who is the deputy director of the IMF's Strategy Policy and Review Department. Ranjit will make a few opening remarks and then we will take questions.

MR. TEJA: Thank you all for joining this call. I don't want to take too much time with introductory remarks because you have the background material, including the report itself, which at 9 pages is less dense and less long than is usually the case with our reports. I think the messages that are in it should be quite straightforward.

I want to give a few points by way of background to this report. Probably the most important would be that this whole spillover exercise which we did for the five major countries was an effort to step up our global surveillance of the core economies of the world and to highlight the various channels through which policies spill over to the rest of the world. You may not consider many of the messages in the report to be surprising. For example, that financial channels are the most important channels through which policies affect others is not news to most people. But I think it is one thing to say it and it's quite another to show it analytically. In fact, much of the work that the economists do using large models does not adequately take into account all these financial channels.

So we have tried as part of this effort to bring a lot of new modeling work to see how exactly these channels work and what kind of differential effects a certain policy will have; whether the U.S. does something or the U.K. or China, how exactly does it impact others. We've tried to bring a lot more specificity and a lot less generality to our surveillance work and this is what this exercise has been geared to--to bring about a more detailed and specific kind of discussion in IMF surveillance than is usually the case. So that's just by way of background.

I won't go again too much into the main policy messages. If I had to summarize it in just one or two sentences, I would say that the crucial message is the fact that the financial channel of transmission is far more important than any direct effect that monetary or fiscal policy may have. It gets to the heart of matters, especially in the kind of situation where we are today, where you need to target policies at ameliorating the stresses in the financial sector. Whether we're talking about fiscal policy or monetary, you need to make sure that these are squarely targeted at easing tensions in the financial markets so as to have a positive effect on the rest of the world. This is what matters for the rest of the world: what will European policy do to financial market conditions, what will U.S. fiscal policy do to financial market conditions and so forth. So that is very much the ethos and our way of thinking in these spillover reports. Let me stop there and I'd be happy to take any questions you may have.

QUESTION: In your discussion about the euro area financial risks of transmission, you say in point 8, "Were financial stresses from sovereign debt sustainability in the periphery to spread to core euro area banks, the hit to banks across the world in terms of risk premia would in many cases rival that from the Lehman event of 2008." I think you may have said that technically. I couldn't find that simple directness in the actual euro area spillover report and I'm wondering if I'm wrong about that. Are you putting that on a finer point?

MR. TEJA: It's exactly the same point that was made in the euro area spillover report so there's no difference there. In fact, if you look at Figure 6 of the report, you see the bars of what happens to U.S. banks, Japanese banks, et cetera. They're quite close to the diamond which is the height of the probability of distress that occurred during Lehman. I think it's exactly the same as in paragraph 13 of the euro area report.

QUESTION: You said it was especially important today to understand that financial transmission is the most important priority. I'm wondering if you can explain why that's especially important today.

MR. TEJA: It's as important today frankly as it was when we wrote this report 2 months ago, so that has not changed. The reason why we highlighted it, this is not just an argument being adduced to support -- there's been a lot of news lately that the IMF has been urging Europe to do more to shore up its banking system. We didn't just make up this policy message. The relationship between stresses in the euro area and the stresses that are felt by banks in the rest of the world is something that we have extracted from the data in financial markets. It's not just a policy assertion, it's really what our technical work is saying. We also have a policy point to make which is I think broader. But what we were doing here was bringing the technical analysis that has been underlying this policy line that we have been making.

QUESTION: I have three questions if I may. One, Mr. Teja, what surprised as to the outcome of these reports and analyzing them? Is there anything in these spillover effects that you hadn't thought before or that you were really surprised of? The second question I have is at one point in your report you say that real effects or transmission effects concerning movements of yields and risk premia are different in reality than they are in standard large macro models. Does that mean that the results that you present by being based on standard large macro models are not adequate to capture the spillovers? The third question I have concerns figure 3 in your report, and I'm really trying hard to understand. I don't know what you are showing me and I would like for you to explain what are you showing in figure 3?

MR. TEJA: First regarding what surprised us, as I said at the beginning, it is one thing to have a certain prior which we have as economists. It's quite another to be able to demonstrate it using empirical data and what kind of results we generate on what happens to financial conditions from any policy. I think that's a point that is not sufficiently appreciated although it sounds obvious. Let me give you an example and it comes to this issue you raised about large standard macro models versus not so standard models. What we have tried to use is not-so-standard models. We've tried to change the models that are being used. I won't be technical about it, but let me give you an intuitive reason of what's wrong with large standard macro models. If you think how much countries trade with each other, what is the share of Europe's exports to the United States? It's maybe equal to 2 percent of European GDP, 2 or 3 percent of European GDP. So if Europe's GDP goes down by 1 percent, what will be the effect on the U.S.? It will be tiny. It will be like 0.2 or not even that. We're taking a small percent of a small percent so that whenever you use that kind of logic, what models tell you is the effect from one country to another, and it's naturally going to give you relatively small answers. What really happens in the real world is that when output goes down in Europe or anywhere else, asset prices in Europe are affected by that. And when asset prices are affected in Europe, in reality we know that they're also affected in the U.S. and that's what pulls down the U.S., not just the fact that European growth was less by 1 or 2 percent.

So the whole point of our analytical work was to build in realistic correlations between asset prices--between what's the effect of a policy or of a shock in Europe to European asset prices, how does that affect the U.S., the same story vis-à-vis other country pairs. What really matters is how much prices are set in a certain market, how much they transmit to other markets. So what U.S. fiscal policy or monetary policy does to U.S. asset prices is more important for Brazilians and Indians than what actually it does directly to U.S. GDP at least in the near term of 1 to 2 years, so that's sort of the message from this.

On your question about figure 3, figure 3 is simply showing what is the effect of planned fiscal adjustment using one of the models that we developed for this exercise on the rest of the world. What we show there is that those effects are relatively modest on the rest of the world. The planned fiscal consolidation has relatively modest effects on the others. What you really need to do is contrast figure 3 with figure 4 which highlights what is the effect of a loss of confidence, an increase in interest rates and so on, and that has a much more negative effect on the rest of the world. Therefore, when you put these together, the message is that you must have medium-term credibility and, from the rest of the world's point of view, that's very important and that's why we have been emphasizing not just here but in other contexts that in order to support the recovery now, you must have medium-term fiscal consolidation and confidence of the financial market underpinning that.

QUESTION: My question is about spillover effects of monetary policy in diverse economies. I wonder, what would be the effects of new expansionary action in the US on emerging markets.

MR. TEJA: You’ve raised a very interesting point and I think a point that can create confusion, so I’m very glad you raised this question.

A summary of this is in paragraph 14 of this paper, but the underlying description of our analysis of U.S. monetary policy is contained in the U.S. spillover report. So I would like to, use this occasion to clarify what has been the message of the spillover report on U.S. monetary policy.

I think the first message is that U.S. monetary policy in normal times has very large spillovers -- has large and important spillover effects on the rest of the world. That’s point number one.

Point number two is that in times of stress, as in after Lehmann in 2008, and when QE1, for example, was implemented, this effect gets magnified. It’s even larger. That’s point number two.

Point number three relates to the effect of QE2 on the rest of the world’s financial markets and output. And when we looked at this question using the methodology that we had developed of event studies, we found that QE2 had a more modest effect, a much more modest effect, than QE1. So, we’re not saying that U.S. monetary policy has no spillover effects or small spillover effects. On the contrary, it has very large spillover effects. But since the specific question we were looking at was QE2, on that specific episode we found that the results did not seem to suggest a huge spillover on other countries. And we also can see, in the underlying report, that this may be due to technical and empirical reasons. I won’t go into that because U.S. monetary policy had many effects and QE2 had many effects and not all of them can be captured by the model that we use. We just reported it in a neutral way and I think a lot of people have misread that to imply that we think that U.S. monetary policy does not have spillover effects on capital flows, emerging market bond yields, et cetera. That is not true. We do believe it has important effects, but just for QE2, the evidence was not compelling.

QUESTION: Can we expect a decrease in the effect, I mean, if we have a QE3, for instance?

MR. TEJA: I really don’t know the answer to that. It’s a completely speculative question. It depends how big it is and so forth, but if I had to draw a lesson from the spillover report, the answer would hinge on, well, what is that going to do to financial markets? How will they receive it? Is it done in response to easing tensions there or not? And also, all these considerations will bear on what kind of effect it will have.

QUESTION: I have a question - a very technical question on the timing of the report. We get the report on September 2nd and the date is July 11th, so I would like to know why there is this almost two-month delay, and especially because there are some messages that could have been tweaked if you had had further data that came out since you wrote this report. For example, on point 14, when you assess the effect of QE2 you say that there was a concurrent improvement in global conditions, but, for example, U.S. growth figures have been sharply revised down and maybe it wasn’t such an improvement in global conditions. So, why is the report coming so late and why was it not published in July?

MR. TEJA: I can give you the straightforward answer on this. Normally a report like this is circulated two, three weeks in advance of the Executive Board discussion and we normally publish it right after that, so the usual lag is about three to four weeks from the time that it’s issued to the Executive Board. So, why did we not publish it on the 1st of August or 2nd of August, which would have been the normal time? It’s mainly because everyone goes off on holidays on August 1 and we wanted this report to get the attention that we believe it deserves in terms of the messages and the analysis we had. So, it had just to do with that.

Relatedly, you had a question of whether we would have changed some of the analysis, revised some data, and so on. I’m not aware of anything that I would have changed, for example, when they’re talking about the improvement in conditions during QE2, I think the data for that period, August to November of 2010 has not been revised. So, we were just referring to that period with regard to the empirical estimation, which was done for that period. So, that data has not changed. I don’t see that anything would have changed as a result of, the passage of four extra weeks.

QUESTION: If I can make a follow-up. The effect of QE2, you seem to imply it was mostly in Q3 of 2010, but QE2 was launched formally in November 2010 and it was hinted only in August 2010, so it seems that communication by the Fed had more effect than its actions.

MR.TEJA: You’ve raised an interesting point. The methodology that we used to assess QE2 was what was the instant effect--within a week or so upon the announcement of QE2, and we have given much more stress to announcement than to actual implementation. I think it’s an important point because financial markets are forward-looking, they discount information, so once they hear that there’s going to be a QE2 or QE3 or QE anything, the effects in financial markets are very rapid. I mean, they occur in a matter of days and when it actually comes and how it unfolds and so on has relatively less effect because it’s expected. So, that’s why our whole methodology for assessing QE2 really was to see what was the effect it had on announcement, and QE1, by the way, had huge effects instantly, long before a single dollar was shelled out.

QUESTION: Can you say what is the next step in the process? What do you expect the implications of this report will be for policymakers in the G-20 in particular or what you hope the effect will be?

For example, you mention monetary policymakers might have to take into account the impact of policy stimulus or policy decisions beyond our national borders. Do you expect that this will form part of the discussion at the G-20 or how do you -- you know, what do you think the impact will be or the next step should be?

MR. TEJA: In terms of the spillover report, our goal was not simply to have a spillover report and then hope that someone takes it into account next time they meet internationally. It’s not just that. What we really wanted to do was to put in place a process, to change the way we are doing IMF surveillance where we go from just discussing a country’s policies (whether it’s good for that country or not) to looking at how its policies affect the rest of the world. I think that’s a question our surveillance has not tackled enough. We have been quite up front in acknowledging this as a shortcoming that needs to be remedied, and hence we really see the spillover reports as the start of a process that will be baked into the way the Fund does surveillance in the future.

On your second point, I would say, we do expect to incorporate some of this work in our other reports, such as the World Economic Outlook. We’re also doing this year a new consolidated report for the IMFC, which is when the ministers of the IMF gather here next month. So the analysis and the messages will be integrated into our other work. That is how we see this process. It’s not that they will discuss this specific report, but I think the messages from it will be built into our work and will be in the background.

QUESTION: Well, just a quick follow-up then. Do you think the importance of the spillover analysis, is more important in times of stress than in other times?

MR. TEJA: Yes. Absolutely. I think it’s important for us to be doing it all the time. But in terms of with what urgency and attention is it received, clearly, people will pay far more attention in times of stress. But I would still stress that this is something we need to do all the time because vulnerabilities build up in peacetime, and we need to continuously look at the implications of how these countries react or don’t react.

QUESTION: I’m having difficulty, please forgive me, in understanding -- and I sense from the other questions -- the real importance of this. I mean, the spillover reports individually came out, they were covered, they said some important things, but the overarching themes that you’re trying to develop, you’ve clearly tried to avoid sort of any policy implications. And it was my understanding that when the spillover reports, you know, as a pilot project were initiated, there was a great fanfare as to their potential application to developing multilateral policies, either at the G-20 or through the MAP process. And really what I read or what I get, and I may be reading it wrongly, is that really this is don’t take too much from this report; that, yes, we have some data showing that financial channels are the most important priority, that really don’t develop any sort of important policy implications from this. Am I missing something here?

MR. TEJA: In fact, I think we’re quite up front in this, especially if you read paras 1 and 2 of this report—which explain what spillover reports are (para 1) and what spillover reports are not (para 2). It says they are not a recipe for global coordination; they’re not a back door multilateral consultation; they’re not a substitute for the G-20 MAP.

So, I take your point that you may have taken the spillover initiative to be more than what it was, in terms of doing the spillover analysis and then saying, all right, here’s what we think what each country needs to do, and so on. We see that as something that’s a broader part of IMF surveillance. It’s not just for spillover reports to do.