Transcript

Transcript of an IMF Economic Forum: The 'Links' that Bind

January 31, 2003

Ashoka Mody (Moderator), Chief, Financial Studies Division, Research Department, IMF
Stefano Cavaglia, Executive Director, Equity Strategy, UBS Asset Management
Vincent Reinhart, Director of Monetary Affairs, Board of Governors, U.S. Federal Reserve System
Anoop Singh, Director, Western Hemisphere Department, IMF
Randy Kroszner, Council of Economic Advisers, U.S. White House

MR. STARRELS: Well, good afternoon. On behalf of the IMF's Research and External Relations Departments, welcome to the IMF and to this year's second Economic Forum. This event today is the concluding act of a two-day conference that has explored the relative importance of linkages involving trade and capital flows and their relevance for investors, policymakers, and international financial institutions.

After opening statements by members of the panel, you will be welcomed to join in the discussions with your questions. If you do, kindly remember to use the microphone attached to each of your seats and to identify yourselves.

To get us moving in the right direction, I take pleasure in turning the floor over to Ashoka Mody of the IMF's Research Department and an organizer of today's event.

MR. MODY: Thank you, John.

I'm actually going to be extremely brief because we have a very distinguished panel. I have five slides, and what I'm going to with these five slides is to give you a very quick overview of why we had this conference and what are some of the results from it, just to give a little bit of a context to the policy discussion. So these are research results, and what we hope from the panel we will get is a policy perspective on these questions.

The first slide, to jump right into it, is in some sense what motivates these discussions, and it shows that there have been very large and increasing co-movements in financial markets. In the left chart, you see that co-movement for the G-7 countries; in the right chart, you see it for the emerging markets. The extent of correlation in financial markets for emerging markets is less that for the G-7, but it is trended up.

In contrast, for real output, i.e., for GDP, you see lots of ups and downs. You don't necessarily see a trend, and as the conference suggested, there's some controversy where the real co-movements increased or decreased over the period. And if you do the chart the way we have done it, it seems to just go sideways. You will notice that in the first chart the real and financial co-movements for the G-7 countries tend to go up and down together. So there seems to be some connection between the real and financial co-movement in the G-7 countries, but less so in the emerging market economies.

Now, underlying these co-movements--and this is when people talk about globalization, this is what people have in mind--you see that the trade between countries, the yellow line on the top chart, has steadily increased. It has not increased a lot, but it has increased steadily. In contrast, financial openness has increased enormously, as

can be seen from the red line. And another measure of financial openness is in the bottom chart, which shows the cross-border holdings of financial assets.

So these charts are consistent with the previous set of charts which show that financial openness has increased tremendously during this period and so have financial co-movements. In contrast, while trade has increased somewhat, you don't see a strong increase in real co-movements.

So what are some of the analytical findings of the conference? I'll just very briefly touch on them. A few papers in the conference find that financial and real co-movement are indeed linked to each other, and these papers, for example, claim that when companies become more multinational, the fact that they trade and invest abroad also leads to co-movements in financial assets. This is particularly so in the second half of the 1990s, which connects up to the fact that the financial co-movements increased tremendously in that period.

The second question the conference asked was: Why do the financial linkages themselves increase in certain time periods? And we had a number of different suggestions, including the fact that there may be certain links between regional financial centers and global financial centers, and that there is some possibility of misinformation which then causes different types of herding behavior among investors.

Finally, a set of papers looked at much longer-term trends in terms of why co-movements are increasing, and we refer to them as structural or institutional features. And in that set of factors are things like the degree of specialization and the institutional changes such as norms and standards by which financial and other markets are regulated.

So the bottom line of this was that, yes, there is some slow-moving change through institutional and regulatory policies that is bringing countries together. At the same time, some short-run factors bring countries together through mechanisms that operate essentially in the financial market. And there is some evidence that the real and financial go together, although it's not very strong evidence.

Okay. So just very briefly, then, the last two slides ask the question: Why might the financial and the real linkages go together? And as the conference discussed quite extensively, the theory is ambiguous on this. The theory says that indeed they may go in opposite directions for very good reasons, and the possibility that they go together suggests that there are some global factors, or what economists call "global shocks," that lead to the possibility that they both co-move.

And so then the question is: In the 1990s, what was that global shock? Did it have a real foundation such as some productivity gains, institutional changes? Or was it temporary? Was it really essentially a bubble? And this last slide just gives a sense of what those global shocks might be. There is an increasing role of financial centers. There might be trade integration which is bringing countries together. There may be coordinated policies across countries. And there may be other institutional factors. But the same factors that cause greater co-movements also cause crisis to spread across countries.

What you see, in conclusion, is perhaps more important, namely that global shocks can have a beneficial influence in bringing countries together, though they might also have the effect of making crises spread more broadly.

With that, I'm going to conclude and invite Steven to make his presentation.

MR. CAVAGLIA: Thank you for inviting us, our organization, to this distinguished panel. What I'm going to try and do is offer an asset management perspective to the problem, and essentially let me ask--what I'll do is address some--start with some questions.

Why do we care about stock prices, both in the short run and the long run? What are these factors that affect stock prices? And what are these factors that we uncover, what do they tell us about integration or globalization? Are these factors related to fundamentals? And what do we as asset managers and what are plan sponsors thinking about and perhaps what should policymakers think about if some of the insights that we have are along the right direction?

So what I want to do is I want to first dispel the view that asset managers are just coming in and out of markets or taking bets on securities on an inter-daily basis and that sort of thing. That's not what we're all about. We're very much concerned and we're very well aware that strategic allocations, long-term allocations, are very important to investor welfare. And, indeed, we spend a lot of time thinking about these issues.

The one in particular that we worry a lot about is home bias. Think about a Finnish investor that has a very high home bias. What that means is that he's very heavily invested in Nokia. It doesn't take a lot of econometrics to show you or to demonstrate that that is not likely to be a highly optimal investment.

Now, we've been talking about globalization for the last several days. There's extensive research. So in a sense, the more global the world is, the less this is a problem, right? Whether I hold Finnish stocks or U.S. stocks or European stocks, it shouldn't matter. But we're not that far along yet. There may be globalization, but there are sources of diversification that investors may want to exploit. And, indeed, what we have found and what I will discuss is that the sector dimension becomes critical in thinking about risk. And the sector dimension may also be a critical insight for policy purposes.

If you think about the wealth that investors have, what we're urging some of our investors to do is not just to think about their financial wealth but their labor income. So, again, think about the Finnish investor, the investor who happens to work for Nokia. What that investor may want to do is diversify not only across countries but diversify across his income stream from his labor as well as his financial wealth, therefore, being away from the tech sector.

Indeed, asset managers are not just thinking about day-to-day best but long-term strategic bets, and we spend a lot of time thinking about that. And that's what led four or five years ago to some of the work that has actually caused a restructuring of our entire organization and which also has been partly--very much in line with some of the work that's been presented here.

So what we asked ourselves is: What are the factors that determine stock prices? Is it country factors? Is it global industry factors? And what we found is that for developed markets, global industry factors dominate country factors.

Let me show you a slide, and this is not meant to be confrontational to research by the Fund. Indeed, we have been swapping notes over the last several days, and I know that the Fund has taken a different view. We have looked at essentially a set of security prices across 22 developed markets, and what we tried to do is measure the dispersion in security price returns. How much of that dispersion is captured by industry differences and how much of it is captured by country differences?

So we as asset managers are very much interested in dispersion of security prices because that determines mispricings and it determines allocations. So what we've done is we've measured the relative dispersion--if you've got lots of dispersion in security prices, how much of it is because of industries, how much is it because of countries? And then we just take a ratio of the two.

What we find is that in the last several years, the dispersion of security prices accounted for by industry factors is about two and a half times as large as country factors. Very much in theme with this globalization literature, but the insight is to say that the source of that globalization is globalization in the industry factors themselves.

Now, there has been some criticism of this published work which says, well, is it just a large cap effect? Are we just picking up the large companies? So we do this for large companies, and we find the same pattern. We do it for small companies; we find the same pattern.

We removed the tech sector, so there's been this criticism that somehow this is only determined by a tech bubble. So we removed media, technology, and telecoms. We essentially said they never existed, and we redid our analysis. And, again, we come up with the conclusion, not as strong but still the conclusion that industry dispersion is much more important than country dispersion and, thus, may be a source of diversification.

Now, for emerging markets, clearly it is still the fact that country factors matter. There has been continuous reference to the pioneering work of Bruno Salnick (ph), and indeed across the conference over the last several days, people have cited his book saying that country factors dominate industry factors.

Indeed, in the next edition--and we are privy to the copies of the next edition--he will cite this work and endorse it and state quite clearly that it is now the case that global industry factors appear to dominate country factors. Now, clearly, this debate will go on for years, but our firm has taken a strong stance.

Now, once you start thinking along this dimension, you may take a more micro perspective to economic integration or to globalization. And what you want to start perhaps thinking about is integration of markets or integration of securities by looking at industry factors. So one of the issues that we were asked to address was how do emerging markets fit into all of this.

In this next slide, we test your ability to read from afar, and what we've done is we've looked at the extent to which industry factors in emerging markets correlate or are related to industry factors in developed markets. So, indeed, it is the case that for emerging market countries, you know, what the policymakers do, what the country factors are, it determines a lot of the volatility. But guess what? If you really look at the data over time, the industry factors are becoming more closely correlated. So tech in the developed world is behaving very much like tech in the emerging markets world. And this shows the pattern over two sample periods, from '95 to present and from '88 to '95. On average, that correlation has risen quite markedly from about 0.07 to 0.35.

Now, clearly, there are some sectors, like utilities, which are highly local in nature, which haven't co-moved.

One of the things that we've done as a firm is we've said that perhaps there are leading--there are companies that lead this integration effort. So what we have is essentially a sample of about 100 companies that are in the emerging markets universe that we treat as developed securities because we believe that they behave very much like developed securities. And to provide hard evidence for that, on average for an emerging market stock, the way that stock co-moves with industry factors in the developed world is about 0.27. For this universe of what we call leaders, just as a label, we find that that correlation is much higher.

Now, is all of this somehow--this is statistical analysis of asset prices. Is all of this related to fundamentals? And I won't be as elegant as a lot of the research that has been presented over the last few days, but I'll just try and provide you a feel for some numbers and for some data that we have looked at which convinced our firm to really change the way we were organized.

So let me ask the first question: Is the capital stock of corporations becoming more global? So what we have done is we have looked at cross-border M&A activity, and we looked at how much of it is intra-sector and how much of it is inter-sector. And if you look historically, in the late 1980s, early 1990s, it was pretty much evenly divided. Now, if you looked at the press over the last several years, you had this anecdotal evidence that suggested that companies were thinking in terms of focusing on their core activities and expanding globally.

Well, indeed, this is very much supported by the data. If you look at most of the--you know, about 70 percent of all M&A activity now is banks' buying up banks across borders. So that suggests that there's something about bank stocks that behaves very different from utilities when shocks to the system hit. And we as asset managers have chosen to really think about sector differences as distinguishing characteristics for security prices, but as also a tool for diversifying for our investors.

Now, this is very much akin to the work that was shown yesterday. Is it the case that if Microsoft has a lot of sales activities in Europe it is very much affected by what happens in the tech sector in Europe? And so all this graph says is the more a firm like Microsoft has sales in Europe, the more it is exposed to those European tech factors, and so on and so forth for other industries.

Asset managers also look at what we call risk factors, and these are common risk factors, like size or value growth. You often hear these, and in financial terms, you know, essentially what value, for instance, captures is these distressed securities that are mispriced by the market. And you are compensated for buying distressed securities.

Now, it is well known and there's much empirical evidence that says if you buy distressed securities, you will earn a premium on the market, and you'll earn a premium of about 270 basis points per annum. Now, much of this work is done in this country paradigm fashion. If you look at the U.S., you say, well, if I invested in value stocks and sold growth stocks, I would earn a premium of about 3 percent. You do that in Italy, so on and so on and so forth. And you aggregate, you get your world from just a collection of countries, and you will have a compensation for buying distressed securities of about 3 percent.

Now, what happens if you do this on an industry basis? Who says that countries are the right way of grouping securities or of thinking about investments? So now think about your investments on an industry basis, which is more akin to the productive capital which drives the security prices, and what do you find? You find that your premium for bearing that risk is on the order of 420 basis points per annum.

This is to suggest that markets are pricing in risk factors in a different fashion and in a fashion that, from an industry perspective, is quite interesting.

Now, we've done this for not just value, not just size, but for a number of other factors, and what this graph tries to demonstrate is that the country factors have declined in importance and the global industry factors which are highlighted by the orange chart, the orange portion, have increased in importance in terms of, again, risk factors. All of this to try and suggest to you that the fundamentals are very much in line with the prices that we see in the marketplace.

So what have we done as a firm? Our firm used to be very much organized in a fashion similar to the Fund. We used to invest--our global portfolios were done by taking country positions, and then we had these country directors with teams of analysts that would analyze companies within each country.

Four years ago, we decided that the market trends suggested that industry factors would become more important, and, therefore, we have realigned all of our analysts and all of our organization on an industry basis. It was a bold, strategic move which to date has paid off and which to date, you know, has paid off for our investors.

Now, what does all this mean in terms of, again, what I started out with, this notion of long-term asset allocations? What we're doing with a number of select plan sponsors is we're trying to redesign benchmarks, the benchmarks against which asset managers are measured. And we're going to back to square one and we're saying: Is the productive capital of various sectors telling us something about the way funds should be allocated globally?

Now, at the same time, that's one inference that we have as asset managers. I would suggest that if those asset prices provide information about the productive capital, they may also provide information about employment and unemployment and, thus, that sector share price indices in the move towards globalization would be a very good leading indicator for sectoral unemployment, which then policymakers could act upon.

So hopefully I've convinced you of the merits of a sector approach to globalization. Thank you.

MR. MODY: Thank you very much.

MR. REINHART: Well, thank you for this opportunity, Ashoka. I appreciate the invitation. My name is Vincent Reinhart. I'm the Director of the Division of Monetary Affairs at the Federal Reserve Board, and I'm also the Secretary of the Federal Market Committee, which means I have to give the disclaimer that what I say here represents my own views.

I actually have to give an additional disclaimer. What I say here doesn't necessarily represent the views of my own family, down from my wife, who's the Deputy Director of Research at the Fund, to our Basset hound.

Now, I'm going to work on the basic principle that if the organizers ask you more questions than they allot you minutes, you're allowed to cherry pick. And I'm going to take a perspective somewhat different than most of the time spent in this conference; that is, I'm going to take a macroeconomist perspective and ask what I should learn from the work done on mostly finance over the last two days.

But I'm also going to suggest that I'd like to be an emissary from the Planet Macro to the people working in finance, saying there are some lessons from where we sit that you could also learn. Since this is the IMF, I would like to reassure the EXR people that the Planet Macro is not in the Gamma quadrant that Ken Rogoff has identified as a real threat. That's a real inside joke for you people. And I also in the end would offer a little bit of perspective from my current job working at a central bank.

I'm going to focus on three issues: financial openness and the volatility of real GDP. That's really only going to set the stage to say it really is important to understand how financial linkages work, and it may have very direct consequences, that is, in terms of ability of economies to grow and the ability of economies to grow stably.

Second, Ashoka and Robin asked some questions about corporate governance, and I'm going to offer a different perspective on the way I think about corporate governance problems and what macro consequence that has.

And then, third, I'd like to talk a little bit about what increasing financial linkages do to the monetary policymakers in the core country as opposed to the periphery.

The first thing to note in terms of financial openness and the volatility of real GDP is I'd like to start with a macroeconomic observation which owes to my colleagues at the New York Federal Reserve, Meg McConnell et al. If you just plot U.S. real GDP in the post-war period, it's evidence that the volatility of GDP growth has come down sometime starting in 1982. It may lose a little in this picture of annual growth rates, but if you look at the statistics on U.S. real GDP growth, you will see that there's been an impressive decline--if you look at the statistics, you'll see that there's a statistically significant decline in the volatility of real GDP growth.

Now, macroeconomists being macroeconomists are offering real side factors. Geez, I did my Gamma quadrant joke already, Ken. I'm sorry. Macroeconomists tend to look for real macro phenomena that could explain the decline in volatility and real GDP growth. They talk about different behavior in investment, different behavior in durables, different cycles in the auto industry. But what's actually striking to an international economist is that the decline in real GDP growth is actually even more evident in world GDP; that is, if you look at the Madison data from 1950 and you take out the U.S. real GDP, it's the case that the standard deviation of real GDP growth has dropped from 1.4 percentage points down to 0.9 percentage points in the last 15 years.

So the stories that you spin to explain the U.S. phenomena, which are very U.S.-centric, probably can't help to explain this phenomenon. Financial economists might speculate, though, that part of this must go to improvements in existing financial markets, the addition of new instruments, and that has in turn allowed various risks to be traded to those who are more willing and able to bear them.

If that's the case, that allows more effective smoothing of production and consumption, and actually there's a benefit to the opening up of financial markets. And that sends a relatively positive story.

Looking for global financial linkages is an important endeavor. Why? Because it is possible that the deepening of financial markets has expressed itself on the real macro economy and in terms of lower volatility of real GDP growth.

That sounds like a positive story, but it's not quite. If you actually look at the underlying data across countries in the Madison data set, you'll see that not all countries have shown improvements. For instance, if you look at Latin America, total volatility of real GDP has fallen in three countries--Chile, Colombia, and Uruguay--but it has risen over that same period in Argentina, Brazil, Mexico, Peru, and Venezuela.

So if there's an overall story to be told, there's also a specific story to be told. Why do some countries exhibit declines in volatilities and others not? Well, if you're going to take a financial economist perspective, switching back again, it might matter--indicate that sequence matters. That is, a mature country with developed financial infrastructure may be better poised to take advantage of new opportunities to trade, to fit new instruments into the existing legal structure, and to use existing instruments better with new opportunities to hedge.

If that's the case, an emerging financial market may not have the depth to take advantage of those benefits, and it may be more prone to financial crisis, unless it happens to have a more stable macro climate or some obstacles to the trade of assets.

So that's the issue, and that's a question I pose back to the financial economists. Is there something about the structure of the markets that might indicate that some countries can take advantage to reap macro benefits of tighter financial linkages and some don't?

The second question I was going to address is look at differences in corporate governance internationally. And, again, there was a certain poignancy in the questions Ashoka sent in that they seemed to suggest that: Does the United States benefit from having such good standards and corporate governance? Two years ago, you could ask that question. It's a little harder today.

But from a macroeconomist standpoint, the trade in capital must be a good thing. Just think of it in very simple terms. On the left side is the core--an economy in the core of the global system, financial system. It faces a declining marginal product to capital; that is, as it adds capital, the marginal product to capital declines. It is open to the world financial market, and so it equates the marginal product of the capital in its country to the world cost of funds.

Consider a country in the periphery, that is, the periphery of the global financial market, that faces a local cost of funds that is actually much higher than that world cost of funds. Firms in that country will equate the marginal product to capital to that higher hurdle rate, and they will invest less capital. If you open up an economy to the global financial system, you go from the local cost of funds, the story would run, to the world cost of funds, and the capital stock increases. Tighter financial linkages should make the country better off.

The problem is it doesn't actually happen that way because it's the--which is representative of the Lucas paradox; that is, capital doesn't really flow from the core to the periphery all that often. And that's an open question.

Perhaps part of the resolution is something about corporate governance. Suppose that it just so happens to be the case that managers misuse borrowed funds to some extent. They shirk. In a developed economy, that is, in the core, there's probably a big enough pool of potential managers that shirking is a constant-scale problem; that is, there's enough potential MBAs that no one of them would skim a little bit more off the top if you gave them two units of capital instead of one.

But it's quite possible at the periphery, with a much more limited pool of potential managers, shirking increases with scale. So if you give them two units of capital, they don't just take twice as much; they take even more.

Shirking adds to the cost of total capital. And what happens? If you open up or you unify the core and periphery, what will happen? The all-in cost of funds in the core will be a little bit higher because of their shirking; that is, because of the efforts of accounting firms and the like to take a little bit off the top. That will be true. The capital stock will be lower as a result. But it's also the case that the periphery doesn't benefit as much from the opening of trade. Why? As you give them more capital, more shirking results.

So most of the benefits of trade in capital would be captured by managers at the periphery rather than the owners of that capital.

That suggests perhaps that the trade in disembodied financial capital may not accrue as substantial a benefit as you might think without--without what? Without foreign ownership so that core management is introduced into the periphery. Without foreign listing of periphery equity to enforce core governance standards. And perhaps we have to recognize that there should be a sequencing of liberalization so that infrastructure at the periphery is established; that is, does it really make sense for asset managers to allocate capital to countries that don't have good internal corporate governance rules, that don't have unified accounting standards?

Do we miss something if we focus on the financial flows and not the infrastructure of the management that uses those financial flows? If that's the case, then that also puts a reason for international codes and standards to establish best practices to flatten out that shirking curve in economies on the periphery of global financial markets.

Okay. I said I'd talk about three topics, and the third would be monetary policy at the core, that is, a very specific question, and then I'll generalize it a bit. Does U.S. monetary policy lose its effectiveness as global financial linkages increase? That seems to be a question at least some of us in this room would care about.

The problem is it's really hard to predict the effect on international monetary arrangements as financial markets become more linked. Why? It might be the case that an increase in the menu of financial instruments--deeper markets, tighter arbitrage across those markets, and more sophistication on the part of market participants--imply that there are increased hedging opportunities, and among the risks you could hedge are exchange rate risks, which means from the economy's perspective, floating exchange rate regimes might look more attractive because they don't particularly impose a burden of transactions costs on the firms and households in your economy.

But it's also possible that an increased menu of financial instruments--deeper markets, tighter arbitrage across those markets, and more sophisticated market participants--might introduce enough other relative prices, that is, across individual firms, risk spreads, equity prices across countries that exchange rates are not relied upon as much to allocate global demand; that is, national borders don't matter as much when investors own the equity of many different firms located in many different countries.

That might make fixed exchange rate regimes more attractive. In either case, what would that matter for U.S. monetary policy? Well, according to the Mundell-Fleming model of floating exchange rates, you've got an IS curve and you've got a monetary policy function. I called it a Taylor rule there. A U.S. monetary policy ease would encourage depreciation that improves the U.S. net export position, that is, the Taylor rule curve shifted out, the exchange rate tended to depreciate, the IS curve then shifts out.

The U.S. benefits because aggregate demand in the U.S. increases. The world benefits because the global money stock increases from the increase owing solely to the United States. In other words, the U.S. gets a bigger share of a slightly bigger world pie in terms of overall income. So there's no reason in this floating example that if the monetary arrangements tend more to the open economy side--tend more to floating exchange rates that there's any loss of effectiveness of U.S. monetary policy as a result.

Now, suppose more countries choose to smooth their bilateral exchange rates with the United States. I'd contend that's not a return to a fixed rate exchange regime. Why? Because it's a fixed--it's a regime in which N minus 1 countries respond to the monetary policy in the end; that is, U.S. monetary policy will still be set according to U.S. interests. Rather, foreign central banks will have to vary their policies to shadow Federal Reserve policy; that is, if you want to keep your exchange rate relatively smooth, then you've got to move your domestic interest rate whenever the Federal Reserve moves its domestic interest rate so you don't open up an interest rate arbitrage opportunity.

But if that's the case, there's actually a magnification effect of U.S. monetary policy. Why? In the Mundell-Fleming model, again, if monetary policy eases, that's got to encourage the increase in the money supply in other countries. Why? Because if they don't, their exchange rate would appreciate against the U.S. dollar, thwarting their desire to keep their bilateral exchange rates smooth.

That automatically benefits income in the United States. Why? Not through a relative price effect, but because the money supplies in all foreign economies will have increased, implying world income expands and U.S. trading opportunities expand; that is, the U.S. gets the same share of the even bigger global pie. The global money supply actually increases by more.

So why did I go through that? Why I went through that, it's not obvious how tighter global financial linkages will affect international monetary arrangements, and in the bipolar choice of international monetary arrangements, either a pure float or smoothing to the U.S. dollar, U.S. monetary policy will remain effective.

Now, of course, as financial markets become more tightly linked, whether it's under fixed exchange rates or under floating exchange rates, U.S. economic conditions are more likely to be influenced by foreign factors. Why? Because the U.S. will have significant positions in foreign assets. There will be foreign ownership potentially of U.S. intermediaries. There will be more correlated movements in asset prices, and there will be the potential for financial crises to spill over to the United States. You don't have to think more than six years back in total and look at the experience since 1997 that those are definitely potential issues. And what will U.S. monetary policy have to do? It will have to gauge the effects of those influences on U.S. activity and inflation and base policy on the forecast. It's unlikely--or given our statutory obligation, the Federal Reserve would not suddenly gear its policy to international considerations; rather, the Federal Reserve will gear its policy for the implications of those international effects on the U.S. economy.

The sad fact is we have a very strong interest in deepening markets abroad. Why? We have a very large current account deficit, and the counterpart of that current account deficit is a capital account surplus. We want it to be easy for foreign investors to acquire U.S. assets. Why? Because if they were suddenly to stop being willing to acquire U.S. foreign assets, that would force a correction in the U.S. current account, with obvious implications for aggregate demand.

Now, just very briefly, as to monetary policy or economies at the periphery, given tighter financial linkages, or what will Federal Reserve policy mean for economies in emerging markets, I think the short answer is U.S. monetary policy has always been an important influence. For instance, if you look at net private capital flows, the green bars represent periods in which the Federal Reserve was easing. The yellow bars represent periods in which the Federal Reserve was tightening from 1970 to the year 2000. The plain fact is when U.S. interest rates are falling in an easing episode, it's generally the case--the last two years might not be the best example, but it would be generally the case that capital flows more readily to emerging markets. When the Federal Reserve is tightening, it's less likely to be the case.

So I think from both the U.S. perspective and the emerging market perspective, increasing global financial linkages provide no reason for a major change in the conduct of policy because those influences have always been important. Rather, it may change relative mixes of the importance of various effects.

[Inaudible comment off microphone.]

MR. SINGH: Ashoka, thank you very much. I can just make a few remarks from here because I don't have a presentation to put on the computer.

I want to talk briefly about global trade and capital linkages as I am seeing them from the perspective of Latin America. The balance between trade and capital integration has implications, in my view, for the vulnerability and crisis-proneness of economies. And as I have begun to look at Latin America in the last year, this balance-between trade and capital integration-is quite different from what we observe in some other regions, certainly compared to many Asian economies.

Generally, we have made the case for trade liberalization with reference to its own effects on efficiency and growth. Similarly, the case for capital account and capital market opening has also generally been made-or not made-with reference to its impact on efficiency and growth and so on.

But as I have looked at Latin America, I have become convinced that there is another important aspect, and that is the balance between trade market integration and capital market integration.

I will go into this issue in a little more detail but, before I do that, let us step back a bit and review the consensus in the policy prescription for making economies more crisis-resistant. Briefly, there is a now a lot of consensus that exchange rates need to be more flexible, therefore allowing a more automatic adjustment process to take place in response to shocks. Second, public debt probably needs to be at a much safer level than all of us previously thought was the case. And, of course, we all now stress the importance of financial sector soundness. However, this prescription may be more difficult to implement in the presence of an imbalance between trade and capital integration of countries.

Turning now back to Latin America, you find that their trade openness is quite low, certainly much less than in other emerging market countries. But, as a group, Latin America probably has much stronger capital market integration than other countries.

Let us review briefly recent trends. Even after the liberalization of trade that did take place in the 1990s, we find that the share of trade as a whole in GDP in Latin America increased just to about 30 percent in the last year or so, by the end of the decade, and this is far less than the trade-to-GDP ratio we've seen in some other parts of the world. And as we look at the exports, you find that the gap with other regions is even greater. The export ratio to GDP in Latin America has increased just to about 18 percent, certainly under 20 percent, and is about half that of many other regions.

In addition, in many economies in Latin America, exports have remained somewhat concentrated, and in many economies there are distortions in the trade structure that make some of these economies quite protectionist. There is a further point that, as you look at the components of the increase in trade within Latin America, you find that regional trade accounted for a relatively important component of the increase. Such intra-regional trade can contribute positively to growth, provided, of course, that regional trade arrangements have benefits that exceed costs, and are broadly consistent with multilateral opening, and making economies internationally competitive. In Latin America, Mercosur has been an important example of a regional trade arrangement, and we need to assess carefully its economic costs and benefits, to ensure that the benefits exceed the costs. Equally, regional trade arrangements should be designed to ensure that they do not become an obstacle to the adjustment process within economies.

Let us return now to some of the implications of an imbalance between trade and capital integration for the vulnerability of economies. Let us look at some indicators of vulnerability. You find that if you look at the debt-to-GDP ratios in Latin America, on the face of it they're not out of line with other parts of the world. But that certainly changes if you look at the ratios of debt to exports or the ratios of public debt to exports, which are certainly very much higher than in other parts of the world. You also have another kind of vulnerability in that capital market integration in Latin America over the last 10, 15 years tended to increase the dollarization of financial systems.

Let us now look at some policy implications. Certainly compared to Asia, it is very much more difficult for Latin America to generate the kind of trade surpluses in response to depreciations of the exchange rates that we saw in Asia. And as we look at what's going on in Argentina and in Brazil, yes, we are seeing far-reaching shifts of the trade account. We're seeing large trade surpluses, and as we saw in other crises, the shift of these trade accounts is probably proving to be stronger than anyone expected.

However, if you look at the amounts involved of this shift in Latin America, it is still relatively small relative to the capital accounts, the debt, and the capital flows that have characterized countries in Latin America.

Another implication stems from the effects of exchange rate changes on the public finances. Depending upon the structure of the public finances, and the extent of linkages of the public debt, fiscal problems can be exacerbated by exchange rates. This makes it more difficult to achieve the fiscal adjustment in response to external shocks.

And so, with a substantial imbalance between trade and capital integration, as exchange rates adjust, it is more difficult to have an export-led kind of recovery from crisis. It is also more difficult to manage fiscal sustainability because of the effect the exchange rate changes might have on revenues and spending in government budgets.

So where do we go from here? I think it is clear--and we all accept this--that there needs to be more focus on trade liberalization for its own sake because of its own effects on efficiency and growth. However, economies cannot typically change their extent of trade openness overnight. We should reflect on the factors that have made economies in Latin America not very open--and assess whether there has been domestic resistance from vested interests, and if so how best a consensus can be developed to overcome them.

Regarding capital integration, it is, of course, problematic to consider approaches that would scale back capital account integration already achieved. What this means is that we may have to live with this imbalance for some time, and this is going to make it possibly more difficult, more prolonged for some countries in Latin America to recover fully from their current difficulties.

And there is also the issue of access to industrial markets. In addition to the usual effects, I would imagine that income inequalities in producing countries are also adversely affected when products in agriculture are blocked from industrial country markets. In concluding, it would be useful to get some reactions from the audience on the points I have raised, in particular, whether they agree that imbalance between trade and capital market integration makes it more difficult for economies to resist crises, and how quickly this situation can change. Ashoka. Thank you.

MR. MODY: Randy, you have the last word.

MR. KROSZNER: People have been leaving watches here, and, of course, the Fed doesn't put a lot of trust in the White House certainly to run monetary policy or to leave their watch or timing. So I'm Randy Kroszner, a member of the Council of Economic Advisers, on leave from the University of Chicago, so you'll be getting a perspective from me that is obviously an administration perspective, but it's still through the prism of the University of Chicago.

Now, some people would say that's a very distorting prism. Hopefully during the discussion we can see how you think about that. I wouldn't think so.

So I wanted to pick up on a number of the themes that were raised here. Anoop had ended up with some issues on politics and politics and finance, some issues that others had raised also about monetary policy as well as about corporate governance. And also to give a little bit of a plug for the Economic Report of the President, this year actually we're doing something unusual, having a chapter on development economics. Most times the chapters in the Economic Report of the President, particularly the international ones, are focused on how policies are helping the United States. Here what we're doing is focusing on how policies are helping other countries, and particularly the Millennium Challenge Account Initiative that the President has put forward of increasing development aid by $5 billion and trying to give it out on a more sensible basis and hopefully a more effective basis than in the past, and also an emphasis on the role of the World Trade Organization and world trade in promoting growth outside of the U.S. And, of course, that also redounds to benefit in the United States.

I just quickly want to put up some numbers that probably people have seen already. That seems a little blurry.

Well, the pattern is fairly clear. It's going up.

[Laughter.]

MR. KROSZNER: World merchandise trade volume has gone up, nearly quadrupled since the mid-1970s, and if I were to normalize this by world GDP, we'd also see a dramatic increase from about 15 percent of GDP to about 25 percent of GDP. So it's a clear increase over time.

Now, some people have talked about much longer-term trends, and some of you know that I've done some historical work looking much further back into the past. So some people argue, well, is there really that much of a change from the early part of the 20th century around World War I when the world was quite open and we became less open during the Great Depression and never really fully recovered from that or are only close to approximately where we were 80 or 100 years ago.

To some extent that's true, but one has to look at different pieces of what's going on, and I think some of this was brought up in one of the papers that had some of the longer-term trends. What we actually see is a difference in the way people invested.

In the old days, when the markets were seen as very open and there was a lot of cross-border investment, almost all that investment was very narrowly focused on just a subset of sectors: agriculture, mining, and railroads, and in particular, mining. That's where a lot of international foreign direct investment was going.

Today, of course, it's much more diverse, and I think this emphasis on sectors that were in a number of the papers and a number of the previous presenters have mentioned I think is very important to emphasize because that's something that, when you look at the aggregated numbers, you don't really see. And that's a real benefit that I think is part of the trade and capital flows that we have today that we didn't have if you look back a hundred years, just improvements in--also, the numbers tend to ignore services or poorly measure services. Obviously there are a lot more cross-border service flows, and broadly the information economy, which just is not well captured in some of these statistics.

So I think that even though this is just one piece of things--and I had a longer-term chart, you'd see that height being similar back a hundred years earlier--this is only one piece of the puzzle. There's a lot else that is going on, and I think we are much more integrated than we once were.

But why do we care about all this? Ultimately, I think that the reason we emphasize this and why there should be any conferences at all at the Bank or the Fund or any organization is thinking about growth, because we care about the linkages because they may have some sort of effect on growth, either causing instability which would harm growth or providing greater integration that could promote growth.

So to preview one of the pieces of the Economic Report of the President that we'll be putting out next Friday--this is just a little bit clearer--what we have here are openness and growth, and the dark blocks are the non-globalizers living in the darkness of the past, and the globalizers that are more open, a little bit more transparent. And so you can see the relative growth rates between the two.

In particular, in the most recent decade, the 1990s, you see a dramatic difference between the globalizers and the non-globalizers. There seems to be much more benefit in more recent decades, in the 1980s and 1990s, than there had been before to globalization and opening your economy up.

Now, why might that be? I want to talk briefly about some of the reasons for that. Certainly this is not going to be an exhaustive list, and it's really just a suggestive list, not one where I have sort of systematic, empirical evidence to provide, but hopefully some suggestive ideas that kind of bring together some of the more systematic papers that have been presented throughout the conference.

Well, one of those things I think is the link that some people like Nina Passick (ph) and others have shown is the link between trade and increase in productivity through work on Chile and I think this work--and there are other countries that people have looked at, where you see a lot more trade, a lot more international competition in a particular sector, you tend to see much higher productivity growth in the emerging market where they're facing more trade. And so there's sort of a very natural link there between greater trade, greater productivity, and ultimately greater growth.

And I think to some extent this helps to provide some micro foundation for the results in the Kosatrok (ph) and Litnen (ph) paper, saying that in the globalization period, the most recent period, which was--I believe they had sort of mid-1980s and through 2001, there's a lot that's left unexplained in the model and could be explained by productivity shocks, and maybe this is one of the means by which we could provide the micro foundation to explain why we're seeing that difference in the globalization period in the results that they have, that there is this greater linkage between trade, productivity, and growth.

Second--and this is picking up on a theme that Vince had talked about--the increase in capital flows, and associated with that you have to have good governance, you have to have good rule of law, you have to have good property right enforcement because, otherwise, of course, the capital is not going to flow to you. Just as you're not going to invest in a firm where you don't have an ability to get your money back from that firm, where you don't have good contract enforcement or good contracts, it's precisely the same as thinking about going to different countries. If you don't think you can get your money back, if you think it's going to be difficult to enforce contracts, then, of course, you're going to be loath to invest there.

I also think that to some extent what we're seeing is--and so promoting best practices on these issues is important. But to some extent, I think we're seeing an endogenous policy response. Precisely because there's been more capital flows or more potential for capital flows, that actually gives more of an incentive for good policy to be followed, because if there really wasn't much capital to be offered, then not having the good governance, not having the good property rights enforcement, it didn't make as much difference because there wasn't as much capital, there wasn't as much flow, so not as much benefit.

So I think to some extent we're actually seeing some endogenous policy response, because if you think back to the old days of a century ago or more than a century ago, the contract enforcement and property right enforcement basically in many emerging markets was done through the colonies. They were basically extending the rule of law from the colonizer's country to the colony. Then particular during the 1930 and 1940s and 1950s, countries pulled out from their colonial rules, leaving more self-rule. But at the time there still wasn't as much global capital that was flowing, and so there wasn't as much benefit perhaps to adopting good policies to be able to attract that capital.

In the more recent period, where there was more capital to be available, I think we've seen a stronger trend towards democracy, towards rule of law in a number of countries. Now, some of that, of course, may be due to regression analyses that people have done at the Bank and the Fund, like, for example, some of the work that [inaudible] and others, looking at the rule of law and income per capita and seeing a very strong, positive relationship between those, and so that in some sense we get the strong correlation here.

But what's driving what? Is it the rule of law that's driving income per capita or, to some extent, is it income per capita that's driving the law? So we have to be very careful about that. We have to be very concerned about drawing causal inferences from this. But I think there is a simultaneous determination that's going on, having an interaction between the politics and the economics.

So, to some extent, it's partially endogenizing the story that Vince was telling. Since the story that I was going to tell was very similar to Vince's, I have to add a little bit of a twist. There's sort of an endogenous positive policy response to the increased openness, increase in capital flows, and more of an incentive for openness.

Finally, just very briefly, again, on one of the issues that was raised by Vince on thinking about monetary policy in various countries. Something that I've been struck by is if you look at outbreaks of hyperinflation or countries with high inflation, it's really come down quite a bit over the last decade. So a lot of people were very concerns about Argentina having explosive inflation. A few years ago, when Brazil broke their link to the dollar, there was concern about explosive inflation. When Russia broke its link to the dollar in '98, again, a concern about explosive inflation.

In none of these countries have we seen that. Certainly the price level has risen significantly in Argentina, but it's nothing like a classic Argentine or Latin American hyperinflation. Brazil has been able to keep inflation under control. Russia certainly has.

And so I think one of the reasons for this is much, much greater currency competition than there had been, again, part of this due to the globalization that we see. It's much more difficult to have high inflation and generate a lot of government revenue from that, because, of course, that's one of the main motivations for very high inflation, is the seigneurage. It's certainly disruptive to economic relations, typically does not bring you a lot of favor with the public when a politician does this, but sometimes when you need money and you need money fast, the best way to do it is to try to run the printing press quickly.

But if there's much more international competition on the currency side--and I think this has to do with technological innovations that make it much easier to switch large sums of money between currencies and between countries--as well as the dollarization that Anoop had mentioned in many of the formerly high-inflation countries, so that even on just a day-to-day basis, on a transaction-to-transaction basis, you can switch at almost no cost to just using dollars or using deutschemarks or something else. A lot of money also flowed into--a lot of actual cash flowed into Russia during the 1900s. And so it just becomes much more difficult to use this lever as a way to generate a lot of revenues because sometimes the elasticity is changed. People aren't stuck trading in the local currency. They can switch more easily. And I think that helps to put a lid on high inflations.

If you look at the top decile of inflations around the world, that has come down just dramatically over the last few years. Really, the last bout was when the former Soviet Union broke up into the different pieces. You initially had high inflations, and now you have much lower inflations there and throughout the world.

And so I think in some sense what Vince was saying is correct, that U.S. monetary policy is having much more of an effect worldwide because it's become more and more difficult to deviate too far from that, that is, to have a very high inflation. Now, that doesn't mean that there can't be high inflations, there won't be short bouts of these high inflations or occasional hyperinflations. But if you just look around the world, it's really very different, and a lot of the concerns that we have in many emerging markets that without a peg, without some sort of particular linkage they would have hyperinflation, we don't seem to be seeing that. And I think, again, that's coming from this openness, from greater linkages.

So to conclude, we want to think about the interaction between politics and finance, that we don't want to just think about these relations as purely exogenous, that there's a close endogenous link between the two. But ultimately I think there's something very important that we can do to push for greater trade, greater trade openness, because I think that sets into motion a virtuous circle in terms of the politics as well as the economics. And so anything that we can do to promote trade I think not only has the traditional benefits of improving people's welfare by greater diversity of choice and such, but I think they have greater productivity, greater growth, and also more pressures for good governance and good rule of law.

And just one final note on politics and finance, particularly since it's difficult to be the last speaker on the last panel on the last day. When I came to the Council of Economic Advisers 15 years ago as a graduate student, I arrived in September of 1987, and we all know what happened in October of 1987. Well, I arrived at the Council of Economic Advisers as a member to be confirmed in the fall of 2001, and we know what's happened to the stock market since then.

I'm on leave from the University of Chicago and will eventually go back there. So if you ever see my name associated with coming back to Washington, sell.

[Laughter.]

MR. KROSZNER: So that's the important link between politics and finance. Thanks.

[Applause.]

MR. MODY: Thank you very much.

We have some few minutes for some discussion. So if you will identify yourself and keep your questions and comments pithy, I would really appreciate that. I'd invite any questions or comments. Yes?

MS. : In the recently concluded U.S. and Chile FTA negotiation, the United States allowed Chile to reserve the right of capital control. I heard this is the first exception for any bilateral trade agreements of the United States with other countries. Does this reflect a policy change of the United States Government?

MR. KROSZNER: That's directed to me.

[Laughter.]

MR. KROSZNER: No, it doesn't at all. Actually, it emphasizes a greater role for capital market openness than we previously have had. Typically, these issues concerning capital market openness were done through separate treaties, the so-called Bilateral Investment Treaties, the BITs, rather than being integrated in as a fundamental part of a free trade agreement. And so what we've been doing is trying to emphasize this even further, the important role of capital market openness. And it is really only under very, very extreme circumstances that some sort of temporary deviation is permitted.

So actually I think it's quite the opposite, that it shows a much greater commitment on the part of the administration to make sure that there is capital market openness by bringing this into the free trade agreement, where typically it was done separately.

MR. : Yes, I have one question to the last speaker. Your point was that the explosion of inflation did not develop because countries are aware that they cannot collect as much in inflation tax bills as the (?)-tution. But the argument could go the other way around because for the same monetary expansion to finance the deficit, one should see now an explosion of inflation so that your argument would be the other way around, because Argentina to finance the same deficit and the same amount of monetization, inflation will have an explosion because you have much less demand for the local currency. So it would work the other way around.

My argument usually is that why we have not seen the explosion of inflation is that the country has invested a lot in changing the regime to inflation targeting where this system has much more credibility. So inflation does not go out of the roof because people trust that the government is not going to allow it. Because your argument is the other way around, your agreement inflation should be an explosion rather than being [inaudible] because there is not enough of the currency.

MS. : This is also related to the issue of high inflation. The opera ain't over 'til the fat lady sings, and the Argentine default was in 1982, and its hyperinflation came in '89-90. Similar span of years between the Brazilian default and its hyperinflation in the 1980s. So it's good to be optimistic, but it has its limits.

MR. TYSON: Jim Tyson from Bloomberg News. A question for Mr. Singh and Mr. Kroszner. Given that open trade provides such clear benefits, I wonder if you'd both briefly comment on the concrete plans for the administration and for the IMF in encouraging greater trade openness.

MR. SINGH: Well, the IMF doesn't trade, so I guess it should be up to a member country to answer that.

[Laughter.]

MR. SINGH: But what I would say is that in our reports, the World Economic Outlook and other works, we are giving a lot more attention in being very specific as to which products, which countries, and what are the costs and the benefits. So I think the best we can do is to develop the agenda and publicize it, and we are doing that.

MR. KROSZNER: Well, as you know, the President put an enormous amount of emphasis on getting trade promotion authority. That was one of the key legislative goals of the President's early time in office. That was something that under the Clinton years had been lost from the President. And so he put an enormous amount of emphasis in getting that, and we were able to achieve that.

Having achieved that, we have now very expeditiously concluded free trade agreements with Chile, Singapore, have negotiations going on with a number of other countries. Whenever any other country comes through, this is one of the key things that we talk about both--well, actually throughout the administration. It's not just in the White House and the U.S. Trade Administration, but the State Department, every department really emphasizes this and the importance of trade, and we're trying to move as quickly as possible to pull down trade barriers as much as possible. I think Ambassador Zoellick has been very aggressive on this and will continue to be so. And it is an extremely high priority for the President.

May I touch on the other questions on inflation?

MR. MODY: Sure.

MR. KROSZNER: I actually don't think we're that much in disagreement. It all comes down to the expectations about what's going to happen. Certainly if there's an expectation that the local authorities are going to try to use monetary policy to finance themselves, then, of course, people will just fly immediately away and you'll have very explosive inflation.

I think it's precisely because they understand that that they adopt a variety of targets of independent central banks, inflation targeting, et cetera, to try to convince people that that's not what they're going to do because they realize that's not a very efficient means to finance themselves. It was never an efficient means, but it was a means that I believe has become less valuable and more costly to do.

So I don't think we're all that far apart. I think it really depends on the expectations. And the optimism--I did try to put a clear caveat in that this doesn't mean that we'll never have a high inflation or hyperinflation once again and there can be delays. But I do think that we've seen more restraint than I think most people had expected ex ante. And, also, if you just generally look around the world for--unfortunately, I neglected to bring the chart with me for the top decile of countries having inflation. It's dramatically lower than it was, and also the median is--both the mean and the median are much lower than they were.

And so I don't want to say that it's never going to happen again, but I think there are strong forces making it less likely that it will occur, but I don't want to say that it's impossible. Certainly it's possible, and as Vince well knows, anything with monetary policy is possible.

MS. : Georgia San-(?) , with USAID. I'd just like to make a general comment on the corporate governance issue. We've been approaching corporate governance a little bit more distinct from rule of law in that we've been working with institutional development and counterpart organizations such as securities commissions, listing standards of stock exchanges, and also through the banks, where we're trying to provide access to credit to small enterprises. And I'm just wondering if you feel that given the limitations of the foreign aid program in the financial sector technical assistance, do you think it's better for us to work in regional fora solely rather than working on a bilateral basis with countries, given some of your findings?

MR. KROSZNER: Well, I think one has to work on all levels. I don't think that the findings suggest that one should only work bilaterally versus work regionally. I think one has to use all the instruments available to try to promote good governance and good rule of law. And this is really one of the things that is emphasized in the President's Millennium Challenge Account. One of the key aspects of it is good governance. And so money will be given out as a reward to countries that actually have good governance, good rule of law, and good economic freedoms and such.

So it's really something that can be encouraged at all levels, and what we'll be doing through the Millennium Challenge Account is encouraging it on a country-by-country basis, basically by running a contest amongst the countries; those that have the best governance structures are the ones that are most likely to be rewarded. And I think the technical assistance that USAID and many others can provide could be very, very valuable in ensuring improvements in corporate governance, and that can be done both bilaterally as well as regionally, sectorally, et cetera.

MR. MODY: Any last comments from the panel?

[No response.]

MR. MODY: Well, I think we have reached exactly 3 o'clock, 2:59, so I thank you all for coming. Thank you very much to the panel.

[Applause.]

[Whereupon, the meeting was concluded.]