This week we look at trends and how to make them your friend. Whether its stocks, the dollar or interest rates, getting the trend right is a cornerstone of successful investing.

I finish up this week's letter while flying back from DC. On Monday, I was invited to testify before the Congressional sub-committee which is looking into hedge funds and the role they play in the markets. My testimony centered upon my belief that hedge funds should be made available to all investors and my proposal for a new type of investment company which would allow (not require) hedge funds to register with the SEC (or CFTC) in a manner similar to the way mutual funds (which are investment companies) do today. This new type of regulated fund could then offer their fund to the public, with some very specific suitability limitations for a period of time to allow investors to become familiar with the different set of risks involved in hedge funds.

All investments have risk, even (or especially) government bonds. It is not a matter of risk or no risk. Why should 95% of Americans, simply because they have less than $1,000,000 in net worth, be precluded from the same choices as the rich? Why do we assume those with less than $1,000,000 to be sophisticated enough to understand the risks in stocks (which have lost trillions of investor dollars), stock options (the vast majority of which expire worthless), futures (where 95 % of retail investors lose money), mutual funds (80% of which under-perform the market) and a whole host of very high risk investments, yet are deemed to be incapable of understanding the risks in hedge funds? The current system which consigns most investors to second class status is economically wrong, philosophically immoral and politically elitist.

I spent the first part of the week putting together my written testimony, which was lengthy (of course), and then spent Wednesday trying to cut it to three pages for my oral testimony on Thursday, which was limited to five minutes. A lot of beautiful words met the chopping block. I was very pleasantly surprised when Committee Chairman Baker and Ranking Member Congressman Kanjorski engaged me in a very lively public follow-up discussion, even as the hearing ran well over its scheduled time. It was clear they have a certain grasp of many of the issues involved. I think it bodes well for any legislation on hedge funds that would come from this committee.

I will post the entire statement within a few weeks (or less) and I assume it will shortly be on the House website as well, for those interested. I will create a shorter version for a future letter. But for today's letter, let's look at another chapter from my book-in-progress.

The Trend is Your Friend, (Until It Isn't)
Jeremy Grantham is a very highly respected money manager and analyst. His firm, Grantham, Mayo, Van Otterloo (GMO), manages $22 billion. Grantham is a famous deep value investor. He was taking his clients out of stocks in 1998 and 1999 (and even earlier), as value, by his calculations indicated that, traditional stock portfolios simply got out of line. There are many in the investment industry who hold Grantham in almost, if not in fact, guru status. He has earned it.

He told us he lost 40% of his accounts during this period, which is a staggering number, since he manages nearly $20 billion. His large pension fund investors demanded that he keep up with other managers, and he refused, based on his sense of value. Now, these funds wish they had stayed, as Grantham has beaten the socks off his competitors.

How painful must it have been to lose that much business. It is a testimony to his character that he stood by his belief in value even as his income went down. But the clients that stayed also need to be commended, as it is hard to sit out the dotcom bubble as your peers are participating. I remember the articles about how Warren Buffett just didn't get it.

Grantham's basic investment theory is that over time investment classes come back to the average. When asset classes are well above trend he avoids them, and when they are well below trend he buys them. While it can take a long time for some classes to revert to the trend, if you have time and are patient, this style has been successful. Grantham has been very successful at simply investing for the long term using history as his guide.

As a student of history, I like his approach. It let's you get on the right side of long-term trends. You will miss bubble tops and get in too soon on irrational bottoms, but patience and time will see you rewarded. We are going to look at two separate pieces of received wisdom from Mr. Grantham, one from an interview in Barron's in 2001, with great relevance for today, and the other from a debate with Professor Jeremy Siegel, author of Stocks for the Long Run and a major proponent of buy and hold. (We will also dissect Siegel's arguments.)

Let's first look at the Barron's interview (remember, this is 2001), picking it up mid-sentence:

"...And tech and growth will overrun on the downside to become cheaper than normal."

"Q:Then the worst for those stocks is not behind us?

A: The Internet-telecom-tech bubble was the biggest by far in American history. Bigger than the railroads, bigger than anything. To put it in perspective, the S&P peaked at 21 times earnings in 1929. In 1965, in the other great cycle, the post-war cycle, it again peaked at 21 times earnings. Both cycles were built on incredibly strong earnings and productivity gains. In this cycle the index peaked at 33 times earnings, and as we sit here the S&P's P/E is at 26 times earnings. [As we will see in later chapters, P/E ratios are now between 30-35. P/E ratios have actually gotten worse, even as the market has dropped.]

"So, how can you believe that there is going to be a permanent low at a P/E higher than the previous highs? There isn't much hope. My colleague Ben Inker has looked at every bubble for which we have data. His research goes back years and years and includes stocks, bonds, commodities and currencies. We found 28 bubbles. We define a bubble as a 40-year event in which statistics went well beyond the norm, a two-standard- deviation event. Every one of the 28 went back to trend, no exceptions, no new eras, not a single one that we can find in history. The broad U.S. market today is still in bubble territory at 26 times earnings."

[Let me repeat for emphasis: with no exceptions, bubbles and markets will come back to trend.]

"Q:What P/E represents the old trend-line for the S&P?

A: The long-term average is 14. I believe the P/E will come back to 17 1/2 sometime in the next 10 years. A level of 17 1/2 recognizes the world is a better, safer place and therefore we can pay more for it. [This is pre-9/11.] We think the P/E will trend down gracefully. If it happens more quickly, it will be a lot more painful. If it happens in 10 years, there will only be a modest negative return."

[If the P/E trends down gracefully, as Grantham asserts, then that means the market will essentially be where it is today ten years from now. There are clear historical precedents for this.

In fact, that is exactly what Professor Robert Shiller said in his book, Irrational Exuberance. He points out that no stock market at the P/E levels we have seen for the past few years has ever returned anything to buy and hold index investors after ten years. Period.]

"Q:You say we're still in a bubble. Everyone else thinks this is a bear market.

A: The peak was March 2000 and the market has come down a lot, but it has a whole lot further to fall. Great bear markets take their time. In 1929, we started a 17-year bear market, succeeded by a 20-year bull market, followed in 1965 by a 17-year bear market, then an 18-year bull. Now we are going to have a one-year bear market? It doesn't sound very symmetrical. It is going to take years. We think the 10-year return from this point is negative 50 basis points [a basis point is one one-hundredth of a percentage point] after inflation. We take inflation out to make everything consistent."

Q:Your outlook is not pretty. Yet, investors appear to be hanging tough. Do you expect that to continue?

A: When a cycle or bubble breaks it so crushes people's euphoria that they become absolutely prudent for the balance of their careers. I've been talking to older people who went through a wipeout and my best guess is about 95% of the people who have been through a bubble breaking never speculate in that asset class again."

(Ben Stein, actor, lawyer and money maven, tells it another way: "Philip DeMuth, the noted investment psychologist, puts it in a thought-provoking way. As DeMuth sees it, investors who lost big in the tech debacle often cannot bring themselves to sell because that would mean final recognition of their folly in getting in on the wrong side of that bubble. Not only that, but if they sold after colossal losses and the stocks did by some miracle rebound, they would be suicidal. Thus, they refrain from selling because of a combination of fear that they will be wrong again and denial of the finality of the end of the bubble. Through the prisms of fear or just plain self-delusion, investors see hope and keep on buying -- a hold is the same as a buy, as Roy Ash taught me long ago -- and the market stays at startlingly high levels relative to historic norms. Unless the law of reversion to the mean has been repealed -- always a risky bet -- these investors, myself included, are likely to feel more pain.")

A. "....The capital spending cycle is very important to profits, and it is in full-scale retreat. However low interest rates go, who is going to build a plant that they don't need? No one. So capital spending continues down and corporate earnings are still under pressure.

"....The economic recovery will be quite short, two or three quarters, and weak. And then people will get a whiff of the fact that GNP is going to settle back down into a 1.5% range again, because of the capital-spending bust. Finally the negative savings rate will begin to move up, and that will impact top-line growth. The market is no longer in its old game. But this will not destroy the economy. I am not a big bear on the economy at all."

"Q:But what about all the talk of productivity gains?

A: People say productivity justified higher P/Es through higher profits. But I'll give you a simple thought experiment because thought experiments are incredibly useful. Say you come out with a seed corn that is twice as productive -- that is, for every dollar of seed it will grow twice as much corn in an acre. Give it to everybody at the same price as the old seed. Productivity will double. But what will happen to the price of corn and what will happen to the profits of the farmers in the following year? I think it is fairly obvious to everybody that they will be drowning in red ink and there will be corn coming out of every silo. Productivity does not necessarily equate with profit.

However, let us give it only to a farmer in Illinois. What will happen to his profits? They will go through the roof. He will grow twice as much corn per cost as everyone else and he will get rich and famous. Productivity gains are fine if there is a monopoly. If productivity is shared by everybody it flows right through to the consumer. We get fat and happy because the price of semiconductors comes crashing down, the power of the machines goes up, everybody has it, it flows through. It is not a competitive advantage and profits are completely unaffected by it. The whole productivity argument was interesting but it has no relevance to how much money the system makes and how high a P/E you should pay for it."

[For those interested, he believes that there are good opportunities available. He likes bonds, REITS, timber, hedge funds, value stocks and emerging markets.]

The above is consistent with another study by Hussman Econometrics. They calculated that S&P earnings have grown at an average of 5.7% [including inflation] over the last 40 years. Even if the New Era hallucinations were reality - that is, even if the virtual productivity gains became real ones - they suggest there is no reason to think that competition would allow higher rates of profit growth in the future.

What happens next? Rather than another decade of 15% per year growth in stock prices, "a more probable outlook," says Hussman, "is that earnings will grow at a long-term rate of 5.7% annually and that at the end of 10-20 years the price/record earnings multiple of stocks will be about 15...stocks will be about 13% below current levels a decade from now. Add in dividends and you're looking at zero total return over a decade."

It could be worse, they point out. The total return from 1965 to 1982 was minus 20%. "If you are interested in long-term returns," continue the Hussman team, "it is madness to try to squeeze 9% out of a market which is priced to deliver zero."

For the last two years, I have been invited to speak at the National Endowments and Foundations Symposium on the prospects for the economy and the markets for the coming year. Interestingly, when I spoke at this conference in early 2002, I was soundly taken to task by some pension/investment consultants for suggesting that the stock market would drop for a third straight year, even as the economy would Muddle Through.

"You must," I remember one speaker who came after me saying, "have a 22% exposure to large cap growth stocks." I was dismissed as some kind of troglodyte bear brought in to amuse the locals, as this person noted the long list of large clients and solid pedigree of their firm. How could such a company with so many analysts and Ph.Ds be wrong? I note that I was asked back in 2003 and they were not. Perhaps it was that his advice cost his clients another 20%.

Investing By Committee
If you think you have been having investment problems, pity the average endowment and pension fund. Their investments are run by committee, and they hire consultants to give them advice on how to allocate their investments. Typically, these consultants are rear view mirror advisors, with tons of charts and graphs showing how if you simply stick with your stocks for the long run, you will do just fine. They trot out the famous Ibbotson study to prove their point. They show what a "blend" of various stock indexes will do. You must stay in the markets at all times is their basic advice. They tinker with the blends from time to time, but rarely suggest anything but long only strategies. In many respects, they are like the typical broker you know, except they deal with larger numbers.

These endowments are pressed to give up more and more money for their causes, as the need is always great, yet the funds they manage have been shrinking. There were some shell-shocked trustees at this conference, and they are a little more wary of whose advice they take. Not surprisingly, consultants willing to think out of the box are being listened to more and more. There are consultants and advisors who can point to good results for their clients over the past years. (If you are an endowment or pension fund, write me and I will suggest a few.)

Take a Risk, You Get Fired
Before we get to the debate, Mark Yusko from the University of North Carolina made some very interesting points about the consultant problem at the conference. First, he noted that most consultants and managers have a strong incentive to not take risks, where risk is defined as doing something different than the herd of other consultants. If you suggest something different and you are wrong, you lose your job. If you suggest sticking with the standard line, you can blame the market and point out that everybody else had problems as well. You keep your job.

There are some outstanding endowments and pension funds, as well as consultants, that have done well in this environment. They did not stick with the herd, took the risk of being more conservative and today they have been rewarded. They are the exception, however.

The second point Yusko made that I found interesting was that 85% of portfolio performance came from asset allocation and only 15% from actual stock picking prowess. Intuitively this makes sense, but I had not seen any data prior to this.

What he means is that the more important decision for large (and small) investors to make about their portfolio is in which asset class to be positioned. How much in real estate? Gold? Bonds? Stocks? Hedge funds, etc.? Actual stock picking only improves portfolio performance marginally, as large funds must buy larger stocks more representative of the market in general.

While stock picking clearly could make a much bigger impact on smaller individual portfolios, the principle is the same. If you allocate a large portion of your portfolio to stocks, you are subject to the whims of the markets. If you have the bulk of your assets in stocks and they go up, you do well, but if you ride it down, you have the opposite result. Making the decision as to what percentage of your portfolio to devote to particular asset classes has a huge impact on your overall portfolio.

Bull vs. Bear, Siegel vs. Grantham
The highlight of the conference was a debate between the ever bullish Professor Jeremy Siegel of Wharton and the currently bearish on US stocks Jeremy Grantham of GMO Advisors. (Note to self: never follow a Grantham speech again.)

Siegel wrote Stocks for the Long Run. It is the bible for buy and hold investing. Siegel points out that since stocks return 6-7% after inflation (his figures, not mine) over the long run, which is far better than bonds, you must be in stocks, and ignore the ups and downs. It is fair to say he is a proponent of index investing. He does assume you are paying attention to which stocks you own.

To his credit, he did write a very prescient piece in March of 2000 in the Wall Street Journal called "A Sucker's Bet" about the overpriced NASDAQ stocks, pointing out that no investor had ever made money long term on a large cap stock with a P/E ratio of 100, and listed 9 stocks which currently fit that description (Cisco, Yahoo, AOL, etc.).

For Siegel, it is always time to buy stocks. Today you should buy because at current market values he believes stocks are likely to rise 6-7% per year over the decade, thus being back to new highs at the end of that time. Seven percent a year for ten years means the market doubles. He bases this on his study which shows that after the market has dropped 40%, the subsequent five year real returns have averaged 8.6%, and were never negative.

He further points out the market is under-priced based upon his definition of value. You must ignore the current year earnings. To get a true picture, he maintains, you must look at a five year average of reported earnings. This drops the current P/E ratio to only 17.4.

Then he argues that the market will not go back down to the historic average P/E of 14.6. Why? Because markets are more liquid, we are not going to have any economic disasters and investors are smarter than they have been in the past.

The correct P/E ratio that smarter investors will adopt in the future is somewhere in the low 20s.

Then he talks about the new S&P core earnings standard, which will give investors confidence. (See the chapter on earnings for a thorough explanation.) Standard and Poor's tells us that true core trailing earnings for the S&P 500 for the year of 2002 was roughly $23.34 after deducting for options expenses, pension costs and other real world costs, not the reported $45 that companies would have investors believe. That makes a P/E of 37 on trailing earning in late May. At the time of the speech, true core earnings were $20.79.

I sat puzzled for a moment. You can multiply 22 times $20 or $23 and get a market value of the S&P 500 less than half of what it is today. Why buy? "Aaah," he says, "the recent S&P numbers included large write-offs, which will not happen next year. Next year, S&P core earnings will be $40-45 dollars, and thus the market is fairly priced, and will rise over time as earnings rise." Call your broker today. In May of 2002, S&P estimated that core earnings for 2003 would be $36. Siegel was a tad optimistic. Even though we are already at the middle of the year, I think S&P is optimistic as well. A 50% rise in earnings is a big jump in a soft economy. We will see.

Cooking the Data Books
Let's analyze the flaws in Siegel's main propositions. First, why should we use five year operating earnings? Frankly, we now know much of the operating numbers were of the EBIH variety (Earnings Before Interest and Hype). As we will see in later chapters, there is reason to believe that earnings have been over-stated by at least 30% from what more realistic core earnings would have been. Real earnings are what matter, and by Siegel's own admission the market is going to start using S&P core earnings as the gold standard for evaluating profits.

Siegel's data shows P/E ratios of real earnings (not reported earnings), even with a 5 year average, is 22.5. Remember, today's P/E ratio using 2002 S&P core trailing earnings is around 37. No one has yet done the S&P core earnings for the last five years, but if you assume that the ratio would be roughly the same, (a relatively safe bet), you would get the five year average (smoothed) P/E ratio to be about 30, which is comprised of 4 bubble years.

If Siegel is right that investors are now smart enough to know that P/E ratios should be over 20, then that means the market is roughly 1/3 his proposed new trend level, with some serious room on the downside. If markets go back to historical trend, then they are overvalued by 50%. If they drop below trend, as they normally do, then a serious correction is coming.

My second real problem with Siegel's presentation is the assertion that S&P core earnings will be $40-45 in 2003. They were $18.46 in the 12 months ended June, 2002. They improved to $23.75 for all of 2002. I readily admit that earnings are going to grow, probably above average, as companies are aggressively cutting costs (read jobs) and slashing expenditures and the economy is growing, albeit slowly.

Siegel presents no evidence that earnings can grow in such dramatic fashion. I can think of no time in history where this has happened. The S&P 500 is populated with large companies for whom consistent 15-20% growth on the entire index is possible only in go-go growth years. We are in the Muddle Through Economy, and that type of growth is going to be tough.

But let's assume Siegel is right. There won't be as many large write-offs next year. All the really bad news is behind us. Let's assume profits rise by 50%. From $23.75 that takes us to $35.50, which is roughly what S&P projects. At Siegel's smart investor P/E, we are at 781 on the S&P 500, and if it goes back to trend, we are well below 500. By the way, S&P projects core earnings for 2003 to be $36.64, and thus the P/E at the end of this year is still a sky-high 25 if the market could close where it is today.

The only way you can make the assertion that the market is fairly priced on an historical basis today is to manipulate the data to make it give you the results you want to believe.

How to Spot a Market Cheerleader
Why do I bother going through this exercise if I think his data is so worthless? Because it is a prime example of the contortions that market cheerleaders and advocates of a buy and hold philosophy do to justify their beliefs. If the data doesn't justify your belief, then find something that does. If current P/E ratios don't work, then create 5 year smoothed ratios, using earnings levels from previous years that won't be repeated for many years. Don't tell your readers that it usually takes 5-6 years for companies to get back to new earnings highs after a recession.

Now, could the market decide to rise from here and never return? Of course, anything is possible. But History, as Grantham will now tell us, says it is not likely.

Jeremy Grantham took the stage after Siegel spoke, and immediately set the tone with the remark, "Investors are not smart."

Grantham is not a congenital bear. He simply looks for investment classes which are below trend and buys them and sells them when the get above trend. Sometimes he buys too early and sometimes he sells too early, but he believes in the dictum that markets always come back to trend. I have often stated that markets are a mean, lean reversion machine. Grantham is right. Markets always come back to trend.

Let me repeat the concepts from Grantham expressed earlier in this chapter, as they are crucial to your understanding of how to invest in the current market conditions.

Grantham presents us with the benefits of his research into bubbles. He has looked at every bubble for which he could find data. His research goes back years and years and includes stocks, bonds, commodities and currencies. He has found 28 bubbles. He defines a bubble as a 40-year event in which statistics went well beyond the norm, a two-standard- deviation event. Every one of the 28 went back to trend, no exceptions, no new eras, not a single one that we can find in history.

He then argues the broad U.S. market today, as it was in 2001, is still in bubble territory, as it has not come back to trend.

[Let me again repeat for emphasis: with no exceptions, bubbles and markets will come back to trend.]

He notes that many markets and bubbles not only come back to trend, but go down right on past the trend line.

What is trend for the US markets? He gives us four measures. Based upon dividend yield, the market is overvalued by 50%. Based upon Tobin's Q (the market value of a firm's assets divided by their replacement value) the market is too high by 31%. The price of stocks to the 10 year average of real earnings is too high by 31% and as a function of market cap to GNP the market would need to come down by 45% to get back to trend.

The trend line for P/E is slightly under 15. Grantham thinks you could see this trend rise to 17.5 over time, as he does agree the markets are now more liquid and we live in somewhat safer times. But this would mean the market would need to drop substantially to come back to trend. How much depends upon whether you use core earnings, pro forma earnings, and forward or trailing earnings. But they all suggest the need for a substantial correction.

Notice that Grantham uses a 10 year average of real earnings instead of Siegel's 5 years of reported earnings. The longer time frame takes out the effect of the very high four years immediately prior to 2001, and thus does not "curve fit" the data to fit a desired outcome.

It is important for Siegel and other cheerleaders, if they want to maintain any type of credibility, to look for some way to suggest that stocks are fairly valued, as Grantham's next data suggests they are not. If you can't find a reason that stocks are at fair value, then you would logically be forced to acknowledge the effects of trend reversion.

What Will the Stock Market Return Over Ten Years?
Grantham breaks down historic P/E ratios into five levels, or quintiles. Level 1 (or quintile #1) represents the 20% of years with the cheapest values (lowest P/E ratios) in history right on through the fifth quintile which represents the 20% of years with the most expensive.

What kind of returns can you expect ten years later after these periods, on average? Interestingly, the first two quintiles, or cheapest periods, have identical returns: 11%. That means when stocks are cheap, you should get 11% over the next ten years. The last, or most expensive period, sees a return over ten years of only 0.0%. Nada. Zippo.

We are in a period which would easily rank among the most expensive periods.

This basically squares with Professor Robert Shiller's (of Yale) data, written about earlier in this book. It is very public, and all cheerleaders are aware of it. To ignore it, they must show why "this time it's different."

In order for Siegel to predict, as he did, that stocks will return 6-7% over the next ten years, he must show that values lie between the second and third quintiles -- OR that investors will somehow start putting more value on stocks, and thus a floor on the market. He attempts to show current fair value by creating a five year smoothed average, and then argues that this time, investors are smarter and will not do as they have done in the past secular bear markets, which is to take stock valuations well below the historic average.

The problem is that the average over-run of the trend in a secular bear market is 50%, which is why stocks get so undervalued. By that, I mean stock market valuations do not stop at the trend. They tend to drop much lower. For Siegel to be right, we would have to see something which has never happened in history before. Stocks would need to drop to values 25% higher than the long term historical average and no further.

Grantham spent considerable time showing different valuation models at the conference, and all suggest the same thing: there is still more downside.

This is not to suggest he does not find value - he just does not find it in the US stock markets. He likes TIPS or inflation protected bonds, REITS, emerging market debt, market neutral hedge fund strategies, international small cap value and small cap growth stocks, which show historical signs of being undervalued. He is willing to invest in these markets and be patient, believing they will return to the average values and higher, and then he will sell and look elsewhere for value. He is also a big fan of timber, which he shows has done well in every market environment.

One further thing of interest, he shows how a portfolio with the ability to short overvaluations will make anywhere from 2% to 4% more, depending upon the risk taken, than long only portfolios. That squares with my studies and observations. It is why I believe that certain styles of hedge fund investing will become more available over this decade, as the public at large will demand it. When long only portfolios return basically nothing for ten years, and long-short stock portfolios show solid absolute returns which are decent, portfolio envy will overcome the lobbying interests of the mutual fund industry, and hedge funds will be available to the general public.

Living on Tulsa Time
Of course, the first morning plane I could get back to Dallas from DC was cancelled this morning, with no other seats available until late in the evening. As my twins are graduating from high school in Tulsa tomorrow morning, and I need to drive from Fort Worth today, I was a little concerned. Not to worry - the American Airline's staff at the Admirals Club at Reagan with a great deal of extra effort found me a flight out of Dulles.

In fact, on the American Airline's flights I have been on since the employee settlement, I have noticed the staff from top to bottom has been unusually pleasant. This bodes well for American. However, management is very short-sighted to take away the ability of road warriors like myself to give employees a special certificate for services done beyond the call of duty. Frequent AA flyers get a few of these each year and the employees we give them to get bonuses. Bad policy, guys, to take this perk away.

Your wondering what city he's in analyst,