Bill Miller, CFA
Chairman, Chief Investment Officer, and Portfolio Manager, Legg Mason Capital Management

Michael Mauboussin
Senior Vice President, Chief Investment Strategist, Legg Mason Capital Management

"Thought Leader Forum Wrap-Up"

Michael Mauboussin
Every year people ask what the key take-aways are from the Thought Leader Forum. There are three key themes that come to my mind. First, our risk attitudes and our decision-making skills are to some degree innate. There is going to be a variation among the population, then, along these lines based on temperament and personality. From our perspective, this is an important theme to keep in mind as we build diversity into our organization. We do not want everyone to be like-minded-we want there to be diversity in our team's risk attitudes and skills.

The second key take-away from this conference is that decision making is heavily influenced by the social context in which the decision is being made. We have learned from psychologists and social scientists that the context in which we make decisions and, perhaps even more importantly, our most recent experiences, will have a very significant influence on the decisions that we make. From our perspective, we want to create an ideal social context to make sure that we are providing a balance of recent experiences-we want to take out the peaks and valleys.

I would like to make a very personal comment about social context. Having been on the sell side for many years, I encountered portfolio managers around the world who were very frustrated. In part, this frustration came from the limits that their organizational pressures put on the good ideas and decisions that they could make. Bill, in particular, has created a very conducive social environment for successful investing at Legg Mason Capital Management with techniques like focusing on the process, calmness, and steadiness of purpose. This is a very important dimension of what we do.

The third point has to do with the ways in which LMCM uses all of the information that has been presented here. We have several initiatives underway to help us make better decisions as an organization. These initiatives are a work in progress; we are not done. First, we want to learn about these decision-making pitfalls. Just because you're aware of these traps does not mean that you can always avoid them, but it is a step in the right direction. Every member of our investment team has gone or will go to the Harvard Behavioral Finance Conference.

The second initiative was to implement investment journals. Dan Kahneman won the Nobel Prize in 2002 for economics, even though he's a psychologist. His advice for investors who want to improve their decision making is to buy a cheap notebook at their local office supply store and write down every decision that they make. Write the date, write the stock price that day, and write down the reasons that you made that decision. It may be useful to write down how you're feeling that day. This kind of journal allows you to document your thinking, and this reduces the influence of hindsight bias-the phenomenon in which once something has happened, we all "knew" it was going to happen in advance. We are now using investment journals at LMCM to introduce some introspection and feedback into our investment process. This is a simple tool that would help any decision maker.

Our third program is focused on our risk attitudes-an effort to understand personalities. Our goal is two-fold: we want each person on our team to understand their own personality profile and their individual areas of strength and weakness, and we also want to understand each others' personalities so that we can work more effectively as a team. We need to understand what kinds of emotions each of us brings to different kinds of situations, not just what kinds of thinkers we are. Again, this is a simple tool that helps us to improve our organization.

Bill Miller
One of the things I found most interesting about Elke's presentation was the distinction between attitudes towards risk and the perception of risk. These two things can easily be conflated. While your attitudes towards risk may be relatively invariant, your perception of risk can vary tremendous based on the facts of the particular situation that you find yourself in. This is why context is so important.

I am often asked how we use the information that is presented at the Thought Leader Forums. It is very clear from all of the studies that have been done that people are very risk averse. People generally need a 2-to-1 payoff before they will take a 50-50 bet. When the problem is fully defined for people, they will still not act rationally. Combine this fact with the ways in which people respond to risky situations with diversity breakdowns and strong recency bias, and when a stock has hit a one-year or three-year low and all of the news is bad, then you can be pretty certain that, in the aggregate, stocks in that situation are mispriced. They are extremely fertile ground for excess returns. Yet most people for both personal psychological reasons and institutional reasons will not buy those stocks.

I looked today at the stocks in our portfolio that were down the most over the last 12 months, and it was not surprising to see that those are the stocks to which we are adding capital every day. The basic reason is that the underlying business values change much more slowly than the stock prices. If they were attractive stocks a year ago at prices 70% higher than today, then maybe today's expected values are slightly lower, but nothing close to the price drop. Your expected returns will be much higher.

There is a new book coming out by Nassim Taleb on rare events. It promises to be really good.

Question: There are two parts to my question. First, what books are you both reading? Second, what is something that you believed historically, but that you have changed your mind about? When you were asked this question last year, you said that in some instances it made sense to buy first and investigate later to catch the upside.

Miller: We were just talking with a hedge fund manager who is a friend of ours. She was telling us about an interesting new opportunity and I asked her if she had bought it yet. She said, "Oh, of course! I've learned about 'buy first, then investigate.'"

Let's talk about the books first.

Mauboussin: Here are some of the books that I have enjoyed this year.

• Donald MacKenzie, An Engine, Not a Camera
  He is an economic sociologist from the University of Edinborough. We often think that our models attempt to describe the economic world around us. In economics, that is to some degree the intent of the models. MacKenzie argues the opposite- that the models we use affect the world that we live in. A simple example would be the Black-Scholes Option Pricing Model. Prior to that model coming into existence, no one knew how to price options effectively. When that model was introduced, options markets became almost instantaneously more efficient. A camera would take snapshots of the economic world. An engine actually affects the world.
• David Warsh, Knowledge and the Wealth of Nations
  This is a 250-year history of endogenous growth theory and the idea of increasing returns. If you are interested in economic history and knowledge economies in particular, this is a very interesting read.
• Eric Beinhocker, The Origin of Wealth
  This is going to be our next book club book. This is a very ambitious title, but the book is up to the task. This book talks about complexity economics, or Santa Fe economics, or non-equilibrium economics. He uses evolutionary models and complexity models to understand economics.
• Judy Harris, No Two Alike
  This book and the next are very personal. It is about the psychology of personality differences between people.
• Michael Lewis, The Blind Side
  I blazed through this book last weekend. I could not put it down. It starts with how the economics of left tackles in the NFL have changed, and tells an amazing story of a young man in Memphis, Tennessee, who grew up in very destitute circumstances and now finds himself as a leading prospect to be an NFL left tackle.

Miller: Michael likes to go first so that he can co-opt the books that I was going to mention. An Engine, Not a Camera is very interesting.

• Dan Gilbert, Stumbling on Happiness
  This is a very well done book. He offers a lot of insight into psychology and how people behave.
• One of our speakers talked about the brain's process of aggregating lots of diverse information and waiting to see what comes out. One member of our team is an expert in Greek and Roman history. I don't know much about the Romans, so I decided to read about the period. I read about six books on the Romans, including Rubicon: The Last Years of the Roman Republic, Caesar: Life of a Colossus, Cicero: The Life and Times of Rome's Greatest Politician, The Battle That Stopped Rome (about the battle of the Teutoburg Forest where they lost four legions in A.D. 9), and The Fall of the Roman Empire (based on archeology). The point of all of this stuff is that it is really interesting to select a particular period in time and come at it from a lot of different points of view. It reminds me of a line out of the preface to Wittgenstein's Philosophical Investigations where he says that the nature of what he was doing was to criss-cross all of these problems from all different angles to see what would come up. It is really interesting to understand how large aggregates of people behaved under those circumstances for long periods of time; how people considered routine certain practices that we would now consider highly unusual. This was a very instructive drill down for me.
• Richard Dawkins, The God Delusion
  Dawkins' new book on religious belief is pretty comprehensive and timely given the current situation in the world today.

Question: What are your projections for the markets over the next twelve months?

Miller: Let's use the Magic Cube! Will the market go up more than 10% over the next 12 months?
[Shakes the Magic Cube and reads the answer] "Cannot foretell now."

We do not build our portfolios around forecasts like this. I would expect the central tendency to be in the low- to mid-teens over the next 12 months. Over the next 18 months, it will probably be in the mid-teen to twenty range. This is in the context of a steepening yield curve.

Question: I have a question about equity duration. On TV these days, we see a lot of ads for dividend-weighted ETFs and a lot of focus on near-term cash flow. No one seems willing to pay for the duration of dividends anymore. Everyone wants current cash flows regardless of growth. How does that affect your investments going forward?

Miller: I think that the market is not willing to look longer term. You can look at the trading volume and market cap for Amazon, for example, and calculate the average investor's time horizon by how quickly it takes the share base to turn over. The average investor in Amazon is looking out about three months. It is fairly clear what will happen in the next three months. That is a very different perspective than the next 12 to 18 months, over which time we feel that things like operating margins are pretty predictable. All of the things that you described make time arbitrage both possible and profitable. The market is not paying up for those things right now. Therefore, that is where the excess returns can be generated, not in trying to surf it near term or in trying to react to information.

Question: Some business magazines are now presenting dividend yields as a great new invention that is the Holy Grail of investing. If you look at sustainable growth stocks over time that trade at the same multiples, would you avoid those with high current cash flows and high dividend yields?

Miller: The work that Michael Goldstein has done argues that what you want are the companies with low payout ratios relative to what they can reasonably sustain. You want high return on capital with relatively low reinvestment opportunities. The market has been paying for safety, so the next opportunity will be in areas in which the yields are going to be rising.

Question: The book club has been very enjoyable for me. It is about clear thinking and clear seeing. Do you have any ideas for books for younger people to help start these habits earlier in life?

Miller: Charlie Munger's Poor Charlie's Almanack is a great book with a lot of wit and wisdom. I can find you titles like this because I have some on my bookshelf. Marilyn Vos Savant has one on the power of logical thinking that is pretty good, but it is not for really young people. There are several recently popular books in popular logic that try to address some of these ideas.

Question: I found Laurence Gonzales's presentation very interesting, but his focus was at a macro level and not directly investment related. He did mention a couple of companies that assumed that the future would look like the past or were able to reinvent themselves. When you are evaluating companies and interacting with senior management, how do you evaluate some of those issues? How important is this approach to thinking about the future? Are there any companies that you find exceptional in this regard?

Miller: My ability to instantly recall things like that is not great, but I will approach the question proactively. An important aspect of our investment process is that we look at companies in a lot of different ways. Our analysts will look at these companies in much more detail than I will. The key thing that I look for when evaluating a business is their long-term economic model. How might that long-term model differ from the model that is currently visible? Can the management articulate that? I want to understand how the capital flows through the business. If you think about a casino company, someone who hasn't looked at these companies in a while might underestimate maintenance capex, for example. If you average it out over a longer period, this will be higher than if the company has just finished a bunch of new properties.

Most importantly, I want to understand how the management thinks about the allocation of capital. Michael has written the definitive piece on share repurchase, which is just a subset of how management allocates capital.

Finally, how do they make decisions within the company? What are the decision-making procedures? One of the best companies in that regard is Amazon, even though it is very difficult to evaluate this from the outside. Randy described Amazon as the "masters of opacity." Amazon tries to hide from you a lot of things, but their decision-making procedures are really good. One of the questions that Dirk Ziff always asks is one that I have adopted myself. He will always ask the CEO, "Tell me a decision that you have made in the last five years that worked out much better than you thought it would, and tell me about another one in which you have made a significant error." That tends to give you some good insights into how they think. I also want to know what the CEO is spending their time doing.

Mauboussin: May I add something? I agree that capital allocation is extremely important. All roads to management evaluation lead to capital allocation. Secondly, one of the models that I have found most useful over the years is Clay Christensen's work on disruptive innovations and technology. This work presents a framework for thinking about innovation and disruption. It is a circumstance-based model, and it provides specific predictions for how companies are likely to react when they are either disrupting or being disrupted. These predictions have held some water over the years and are a useful framework for evaluating whether a company will react intelligently or unintelligently to disruption.

Kodak is a very interesting example of a company that has to manage one business down and another business up simultaneously.

Miller: Homebuilders are an interesting example, as well. The largest builders like Centex, Pulte, Ryland and others are all roughly in the same situation. They are going to be generating a ton of cash; they are managing the balance sheet, not managing for growth; they are all (or will be) buying back a lot of stock; and their free cash flow yields are very high and going up. Only five of the 14 builders at a recent show talked about cash and the balance sheet. The smaller ones were still talking about growth. In this kind of environment, these guys definitely need to be focused on cash and the balance sheet because there is too much capital, and capital needs to be withdrawn from the business. It is instructive that the managements of the larger firms have been through many housing cycles in the past-they understand how to manage through these things. Newer builders may not have the same experience.

Kodak is a little bit different. Kodak has not only had to manage through a technology transition from analog to digital, but also through an enormous business model transition from a very high gross margin business with pricing power and dominant market share to a lower gross margin business with no pricing power and the product turnover characteristic of a technology company. They have also managed through a cultural transition, because very different behaviors are needed to succeed in the latter business than in the former. All of those things have converged to create a lot of turmoil and obscurity from the outside. It is only by spending time with the management and the business heads looking at each business separately that we have come to understand their business. You cannot discern from the GAAP earnings report how the business is changing. You have to get behind the curtain.

Question: What is the most risky industry or overvalued industry right now? What is the most underappreciated or undervalued industry or group of stocks?

Miller: I wish you had asked me that question in May or March. The answer then was easy: the most overpriced sector was commodities. There has been a pretty good correction since then. I think that in the commodity space, there is still a vast distinction between the marginal cost of production and the price of the commodities. There is still a fair amount of risk in these commodities. You can see this in the refinery business. Some of these guys were trading at $3 to $4 in 2001, and now they are at $50 to $60. They have a long way to go if those economics are mean-reverting. I don't think that the economics of refineries are mean-reverting, but there is still some room for the prices to drop. I asked about the marginal cost of production for oil because I had seen $30 and $40 estimates. Ivy made an important distinction between the marginal cost to bring on new production and the marginal cost of existing production. The marginal cost to bring on new production would reasonably include a capital charge, call it 8% to 9%. Once production is on, the capital cost to bring the barrel out of the ground is sunk. If you are not paying attention to this distinction, you can make a very big error about where prices are likely to go. This is what we are now seeing in the natural gas market. The marginal cost of new gas production is reasonably estimated around $7, and the front month is trading around $4.50. The front quarter is trading under $6, I think.

What is interesting about the question is that only about 2% of the time valuations have been less homogeneous than they are right now. So there are not any obvious candidates for significant overvaluation or for really significant undervaluation that look really easy. A lot of them tend to be contingent probability exercises. On a low-risk basis, something like Citigroup or GE are low risk and have a high probability of excess rates of return. If you want a higher-risk profile, then I think that the entire Internet space is really mispriced right now. We have Yahoo! priced in the mid-$40 range. We have Amazon priced in the mid-$50s. Those stocks are way down over the last 12 months, and there has been a lot of pessimism about them.

Question: Thank you very much for this event. It was worth traveling very far to come here. There has been a lot of writing in the mutual fund press about the growth vs. value cycle. Some think that we are entering into a growth cycle. This raises three questions. First, is there such a cycle? If so, are we entering into a growth cycle? Third, do you have any updates to the definitions of growth and value?

Miller: The discussion of growth and value cycles represents a slightly more sophisticated version of the question: "Is there any place I can put my money without thinking to make excess returns?" Can I put it in small cap and feel confident that it will automatically win? Trying to make those kinds of predictions on a timing basis is virtually impossible. What you can do is look at the implied rates of return of various asset classes or segments within an asset class, and you can get a sense of the probability set for where your excess return is likely to be. That certainly favors large cap over mid and small cap. That favors high quality over low quality. This is not just because large beats small and high quality beats low quality; it is because the valuations tell you that you can have a higher expected rate of return in the big companies with a lower risk at the same time. Pretty soon capital will flow there.

I agree with Warren Buffett that "growth" and "value" do not absolutely carve the world at the joints. From our standpoint, the important question is always "Where are the best values?" Are they in so-called value stocks (accounting factor-based low P/E, low price-to-book, low price-to-cash-flow) or are they in growth names? Right now I think that there are greater values in the growthier sorts of names. This would tend to favor the idea of a "growth cycle."

Question: A number of hedge funds and mutual fund groups have been talking about the popularity of value investing, especially since about 2003. So many hedge funds have been piling into value strategies and value stocks that it has become a very competitive space. Have you noticed any money flowing into value strategies, and has this affected LMCM?

Miller: It is very clear that since the end of 1999, the value strategies and the small- and mid-cap strategies have radically outperformed. That has erased just about all of the historic valuation discrepancy between the two. Since 2002 and 2003, the value strategies have worked. The hedge funds pursue a lot of different strategies, but they tend to crowd into the mid- and small-cap space. It remains to be seen if they will migrate out of that range if the strategy stops working. It may be easier to sell to a client an intensive research strategy with small- and mid-cap names than it would be to tell clients that you have a great insight into GE or Microsoft that the rest of the world doesn't understand.

Mauboussin: As of June 30, a Goldman Sachs report said that 59% of hedge fund assets were being put into companies with capitalizations of $10B or less. Mutual funds tend to have about 40% of their assets invested in this group, and 20% of hedge fund assets were in market caps of $20B or more, while about 40% of mutual funds' assets are in this group. Hedge funds have been tilting their assets to small and mid caps, and they have done very well.

It is also interesting to see how small caps have done. If you look at the forward-looking multiple of the S&P 500 vs. the Value Line index (the median of all companies that earn money), at the peak in 2000, the S&P number was 24 or 25 times earnings while the Value Line median multiple was 13. Half the stocks in March 2000 were at 13 or less. Today, the S&P number is at 14, while the Value Line multiple is about 17.5. We have seen a significant shift.

My final point is that it is very hard to be a hot shot hedge fund manager charging two and twenty and tell your clients that you are picking Citigroup and Microsoft, even if that's the right thing to do for the next few years.

Question: This is a philosophical question. Citigroup has bought $15B of its own stock. The management bought most of the stock in order to give themselves more options. When a company buys back its stock, how does this help the average investor since most of the buying goes right back into the pockets of management? After spending billions of dollars to repurchase stock, I don't see any improvement for the average shareholder.

Miller: I will let Michael answer since he is the expert on repurchases. I will say that Citigroup's outstanding shares are falling. We make a big distinction between companies whose number of shares does not decrease after a stock repurchase announcement because they are just issuing options and a company whose number of shares does decrease. There is one psychological factor at Citigroup that is worth noting. The management at Citigroup has the choice of selecting options or restricted stock, and because the stock has been pretty flat since 1999, most of them now choose restricted stock, even though when you do the math, if you assume Citigroup only goes up 6% per year, they would make about 50% more by taking the options. Even they have been conditioned to believe that they do not know what to do to take the stock higher.

Mauboussin: I understand your point. We draw a big distinction between compensation programs-options and restricted stock, which we want to evaluate on their own merits or demerits-and share repurchase programs. Historically, companies like Dell and Microsoft have linked the two, but we like to treat them as separate issues. If they are linking them and making poor buyback decisions, that is obviously relevant. We evaluate each decision on its own merits.

It almost goes without saying, but Citigroup is returning a tremendous amount of capital back to shareholders, a lot of it in the form of dividends. If a company repurchases its stock below fair value or intrinsic value, that will represent an excess rate of return for the ongoing shareholders. We deem that as a positive thing. We think that the stock is undervalued. As a continuing shareholder, you are getting a larger stake in a more valuable business.

Finally, a buyback is a mechanism for returning capital to shareholders. This can be demonstrated mathematically. People have fretted about dividend payouts decreasing, but if you combine dividends with buybacks, that payout ratio has been very stable over the last 20 years. Companies in corporate America have reinvested at roughly the same rate over the last 20 years, but the means to get the money back to the shareholders has changed.

Question: I will disagree with you in one respect. Citigroup has not raised the dividend since Sandy left. It is still $1.96, whereas Bank of America raises their dividend year after year. There has not been a change in terms of returning more of the excess capital back to shareholders.

Miller: I think that is a mistake. I believe that Citigroup has raised its dividend in the double digits over the last three or four years. We will check into this and get you the facts.

I would like to make one clarification. Michael said that a share buyback is a return of capital, which it undoubtedly is. Companies get confused about this stuff, though. At the end of the day, they are allocating capital. The only question they should ask is "Where is the greatest risk-adjusted rate of return on the capital that I have to allocate?" When they return capital to shareholders via a dividend, the rate of return on that capital allocation is, on average, the market rate of return because you pay out a dividend to a widely dispersed group of shareholders, and on average the universe of people out there will not earn above the market rate of return. If you buy the stock back at a price below what it is worth, you earn an excess rate of return for your shareholders. Dividends, therefore, do not offer excess rates of return, while a share buyback at a price lower than the value of the business will offer an excess rate of return.

I thought this concept was widely understood, but we met with a company a few weeks ago that viewed a share repurchase as a failure of imagination because they didn't have anything better to do with the money. At the same time, they said that they were so frustrated with their stock price that they were going to take the company private if it didn't go up. And within a week, insiders were selling stock at the same price that they were so frustrated about. They were confused men!

Question: In all of the discussions about risk today, it strikes me as odd that you do not have a global fund. It seems that people would exhibit even more irrational risk behaviors when investing overseas. Are there more long-term irrationalities in overseas markets than in the US markets? If so, why would you not invest more overseas?

Miller: Yes. The answer is yes. Dean LeBaron won an award 15 or 20 years ago for demonstrating that various techniques-stock evaluation and management techniques-would start here and migrate overseas. So when low P/E stopped working in the US, it would still work in overseas markets because they were less efficiently priced, which might be a different way of saying that they are more irrationally priced.

In general, you will find that overseas markets are less efficiently priced because they are less transparent, there is less capital involved, there are fewer protections for investors, and the accounting standards in many of them are less attractive. On the other hand, overseas markets are also subject to investor exuberances where investors will see that a market is trading at a 20% discount to the US market, and they would think that it is cheap. Of course, maybe that market ought to trade at a 20% discount because it is bad. People forget that.

We only want to compete where we have a competitive advantage. We don't have a competitive advantage in understanding the Argentine market, for example. Although there is an expert in the Zimbabwean market here in the room-he told me last night that I could buy the entire country for $100M.

Question: I would like to pick up on the theme of an abundance of capital. One of the features of the investment landscape today is that there is a massive amount of capital available. So when the music would normally stop on a bubble or an asset class, there is another fool ready to pick up. How do you factor this abundance of portative capital into your investment process? Obviously, there is a valuation vector and an economic vector, but the liquidity vector at the moment is simply overwhelming, and it seems to distort the fundamental picture.

Miller: What is interesting about this question is that it is impossible to maintain that there is simultaneously an abundance of capital and a shortage of savings because those two things are the same. Globally, I agree that there is a savings glut. I would also take some issue with the argument that we are not saving enough in the US. If we saved more in the US and the rest of the world saved the same, then the global economy would be in big trouble.

To your point, we are faced with a situation in which it looks like because of the better use of capital, better management, and the better use of technology, investors are getting higher rates of return on their capital because the knowledge economy is less capital intensive. And because of increasing savings in the developing world, we are getting more capital to invest. This is just a persistent feature of the broader landscape that, combined with the ideas we heard today about decision making and rationality, seems to increase the probability of having greater amplitude of market moves that may diverge from value for longer periods of time.

I was at the 50th anniversary meeting of the CFA in Chicago recently, and I sat in on a presentation by a really smart guy from Pareto Partners talking about commodities. He was making the case for commodities, and he cited several important papers. He was showing how uncorrelated, commodities got almost the same rate of return as equities with lower volatility. Someone in the back of the audience (who obviously didn't see me sitting there) stood up and said, "You know, Bill Miller wrote in March that commodities are a bubble. But you say that everybody should own commodities. Who's right?" Now, I did not actually write that commodities are a bubble, but let's leave that aside for now. The guy answered, "Bill Miller is right. Commodities are a bubble, but you nevertheless should own them." He amplified the point by saying that he believes that the world of commodities has changed so that there will be more of these large moves that no one can predict how big they will be or how long they will last, but that despite the volatility, commodities would generate returns over the long term.

Mauboussin: It is interesting to think about this liquidity question on a central bank level, a corporate level, and then an individual level. The central bank would argue that liquidity is decreasing in the US, Europe and Japan. Individuals, too, would argue that liquidity is decreasing, even in the US. Corporations, however, would probably argue that liquidity is increasing boldly. When we talk about liquidity, we need to specify what kind of liquidity we are focusing on.

David Nelson: I think private-equity people see the opportunity to leverage. Corporations have mountains of cash and they are earning very high rates of return on capital and equity. Rates are pretty low. There is an opportunity, and these guys are taking advantage of it.

Miller: Keynes, in his treatise on money in 1920, wrote about a theory of backwardation and how you earn returns in commodities. It has largely been proven correct. There are three sources of return. There is a return on the collateral that you post. There is a return that you earn from the backwardation-there are more producers with more things they want to hedge than there are people who want to take the other side of the trade. Since commodity prices decline in real terms, producers are worried about losing real money, so they will hedge these things out. You can earn a return because the decline will not be as great as the producers expect. The third source of value comes from the real decline in commodity prices if you hold them. When you net the three sources of value together, it is about equal to the historic deposit on collateral because the hedging and the decline cover each other.

What is interesting in the commodities boom right now is that the entire thing has flipped over. Instead of the market anticipating price declines as it always has historically, the market expects commodities to rise in real terms as far as the eye can see. Therefore, if you are an investor in commodities right now, you get the collateralized rate of return yes, but you get the negative roll as the investor. As prices increase, you get a negative rate of return each month-then you must have a real return to offset the losses that you get on the roll. But with prices way above the marginal cost of production, who wants to make that bet? That is why I would want to wait to invest in commodities. Even if the data that the Pareto guy has is correct, the circumstances that led to commodities being a good investment are totally different now from what they were when those arguments were made.

Mauboussin: We just circulated a paper within our firm by William Bernstein (www.efficientfrontier.com). Bernstein's most recent article on commodities talks about a lot of the issues that Bill just described.

Question: I am curious how some macro situations are affecting your strategy? I understand that Japan is pumping out a lot of liquidity right now. The Economist published an article about debt a few weeks ago. A number of large borrowers have been using debt to magnify returns. Are you worried about a debt crisis or liquidity crunch, specifically in Japan? Next, what are your opinions about the dollar right now?

David Nelson: Up until May, what you said was true. Now Japan's money supply is declining, which is one of the reasons why people speculate that there has been a big correction in emerging markets and commodities. A lot of people were doing just what you were talking about-borrowing at essentially zero short rates in Japan to play the carry trade with higher-yielding assets. The assets of choice were the ones going up: emerging markets, mid cap, small cap and commodities. Now that game has changed. The big question is whether Japan is going to start raising rates. Japan's money supply, which was expanding rather dramatically, is now declining as it is in the US and Europe. There has been a liquidity drain worldwide.

Governmentally, the wind is in our face a little bit as equity investors, but corporations are in very good shape. My opinion is that the consumer is probably in much better shape than people give them credit for because we talk about how much debt they have, but we don't talk about their net worth. The combined net worth of consumers is about $60T, and their debt is only $12T to $15T, so overall the balance sheet looks pretty good. The big problem on the consumer side is that most of that money is in the hands of 5% to 10% of the people. The aggregate balance sheet looks good, but the problem is on the lower end where people are trying to get the American dream, but they don't have the income stream to back it up.

Miller: I would be bullish on the dollar. The Warren Buffett argument against the dollar runs something like this. Imagine you have a farm. You sell your crops every year and you get so much money. But you want to buy more than the crops will pay for, so you borrow money. Then someone else owns an IOU, and when that person wants to collect, you have to sell off a piece of the farm. Eventually, the farm gets smaller and smaller. This sounds superficially plausible, and it is a well-constructed argument.

There is a different and, I think, more accurate way to look at it. If you consult the writings of the new chairman of the Federal Reserve, he seems to go along with this more academically-oriented approach. There are roughly $55T of assets in the US. Assets grow roughly at the rate of nominal GDP, and right now GDP is about 70% driven by consumption. To finance that 70% consumption, we borrow about $800B. In any given year, our debt goes up by $800B, but our GDP (our total production) and productive capacity rises by the rate of nominal GDP, call it 5%, or $2.5T. So we are borrowing $800B to increase our assets by $2.5T, leaving us net $1.7T to the good. From that standpoint, you are sustainable as far as the eye can see on an $800B current account deficit.

This might not work on a $3T current account deficit; you need to do the math to find out. Your risk is not that other framing problem. Your real risk is portfolio preference. The $800B per year is being held all around the world by various banks and parties. If the Chinese or the world decides that it wants to hold fewer dollars in its portfolio and more yen or whatever, that could cause a destabilizing run on the dollar. The risk is really portfolio preference shift, but as my friend Paul McCulley at Pimco is fond of saying, all of those Asian countries are not buying our dollars because they like us. They are buying dollars because it is in their own best interest to do so. It is a complex issue.

The comments, opinions and any forward predictions presented about any particular security, the economy and "the market" are based on the analysis of the speaker. These are not necessarily the opinion of, and should not be construed as a recommendation on the part of Legg Mason Capital Management or any of its affiliates.