What type of returns can we expect from the stock market? Do I think we are still in a secular bear market, even as the market tries to make new highs? What causes long term market cycles? Is there a psychological component to them? (Answer: yes.) And can we make any predictions about the future of the stock market? All this and more in this week's letter.

Last week we started a series on the debate between GaveKal and Bill Bonner on whether "It's different this time," centering on whether trade deficits matter. I said I was going to address this topic for the next few weeks. However, that was before major jet lag and time crunches caught up with me in London. If I wrote anything too detailed, I might have to disavow it on Monday. So, we will take up the debate next week, hopefully with a fresher and rested mind.

But in the spirit of that theme, my good friend Ed Easterling of Crestmont Research sent me the following essay on why earnings are likely to disappoint in the coming years. It brought to mind a study that I spoke about to the Value Investing Congress in New York City last Wednesday, which came from my book, Bull's Eye Investing. After Ed's piece, I am going to adapt that part of the book to address what Ed suggests is likely to happen.

And speaking of books, Ed wrote a fabulous chapter in my new book, Just One Thing, called "Risk is Not a Knob." If you like his thinking below, you will really like that chapter. You can learn more about Just One Thing by clicking on this link (www.johnmauldin.com/justonething), or buy it at your local bookstore or buy it direct at www.amazon.com/JustOneThing. And now to why stock market bulls hope it is different this time.

Maybe It's Different This Time

By Ed Easterling

Based upon most forecasts of reported earnings from S&P 500 companies in 2006, the stock market is beginning to look attractive. Earnings per share ("EPS") have increased at double-digit growth rates over the past few years and are expected to tack on another strong leap in 2006. Yet the stock market is down for the first ten months of 2005 while hope abounds that a bull rally is near. Let's explore the lessons of history for insights about reasonable expectations for EPS over the next few years. Since EPS and the overall price earnings ratio ("P/E") can be multiplied to determine the price level of the stock market, a valid estimate of future EPS is essential to develop a realistic view about future market direction and investment strategy.

Earnings Rollercoaster

The average nominal growth rate for EPS since 1950 is approximately 6% per year. The growth rate prior to 1950 is similar, although many economists and financial market analysts prefer to focus on the past five decades or so due to the distortions from The Great Depression and other issues early in the last century. Fifty-five years of modern financial history provides a rich set of experiences to understand the impact of business and economic cycles on EPS.

Average rarely happens: There are significant cycles and volatile swings in earnings that average to a growth rate of 6%. Figure 1 presents the annual increase or decline in EPS for each year since 1950. For more than half of the past 55 years, EPS has changed--up or down--by more than 10%. Therefore, double-digit growth rates or declines are more often the rule than the exception.

Despite the sense by some analysts that a maturing economy should lead to more stable growth rates, economic conditions and accounting practices have caused recent swings to increase in magnitude.

Figure 1 Annual Change in Earnings Per Share

The graph reflects periods of rising and falling earnings, with each cycle containing a series of relatively short spurts and stalls. Rather than the relatively smooth growth rate that many investors expect, earnings grow for a few years at a pace that is well above the average and then slow or decline for a couple of years. The average leg of a cycle runs 2.4 years, with upswings lasting 3.2 years and downdrafts correcting in 1.6 years.

For a smoother view of this jagged series, which has varied between one and five years, we can use a rolling average of the annualized change in EPS. This approach combines the impact of rising and falling periods and can highlight when EPS growth has become extended above the long-term trend line or when it has regressed below it.

As discussed in Unexpected Returns: Understanding Secular Stock Market Cycles , (http://www.amazon.com/exec/obidos/ASIN/1879384620/frontlinethou-20) corporate earnings tend to grow over time at a rate that is similar to gross domestic product ("GDP") growth. Yet the effects of business cycles and other economic factors cause earnings to be more volatile than the economy. Therefore, EPS essentially vacillates around a more stable trend line.

Currently, as reported in the Financial Times on October 22, 2005 in the lead editorial, "Corporate profits are already at a record 10.9 per cent of gross domestic product in the US...it is hard to see how earnings can continue to grow so much faster than gross domestic product." Not only is it hard to see how profit growth will extend its current record levels, but also it is instructive to look at history to understand what may lie ahead. Figure 2 presents the average annualized rate of EPS growth over rolling three-year periods. The graph indicates that EPS has risen so significantly in the past three years that it is vulnerable to a reversal and, potentially, to a series of below-average or negative periods.

Figure 2 Rolling 3-Year EPS Growth

EPS has risen beyond the long-term trend line following a trough during the 2001 recession. As reflected in Figures 1 and 2, this has been a repeating pattern over the past five and one-half decades. As with all volatile series, the average always represents a mathematical average of above-average periods and below-average periods. In these instances, average is only a point around which actual EPS vacillates due to economic conditions and business cycles.

At all times along the cycles, investors are bombarded with messages of hope from bull market pundits. In the troughs, the message calls for a recovery--a "wise" prediction that has played-out many times in the past. Each trough in Figure 2 is followed by a solid rise toward, and then above, the average trend line. However, at the peaks, the message from pundits is likewise hopeful--it's just not wise.

The hopeful, but unrealistic, messages lead investors to expect that the future growth will occur from then current levels. This is the same irrational reasoning that leads some investors to expect average future returns regardless of the level of the market, whether 1469 at the end of 1999 or when it was 777 less than three years later. Likewise, neither bull nor bear projected EPS to grow over the longer-term at the 6% average from its trough in 2002. Extrapolations and predictions using long-term averages are better made from midpoints in the cycle than from its peaks.

Another way to assess EPS history is to break EPS growth into decade-long periods. As reflected in Figure 3, the 1960s, 1980s, and 1990s all reflect similar growth rates; the inflation-infected 1970s had a relatively higher nominal (including inflation) EPS growth rate. Yet, the first six years of the 2000s represent an unexplained level of above-average growth. This means that either economic fundamentals have changed or that we have not had a full balance of rising and falling periods to restore the decade to average. To match the typical historical average, it will take one or more below-average periods.

To illustrate, the first three years of the 2000s started the decade with below-average growth. As a result, in 2002, investors generally could expect that a period of above-average growth would come. Over decade-long periods, the shorter cycles tend to offset each other to produce the average. As expected, a sequence of above-average periods started in 2003. It is now increasingly likely that a recession or below-average period is forthcoming to settle the decade average toward the long-term average.

Figure 3 EPS Growth By Decade

History As A Guide

What is a reasonable scenario based upon the lessons of history? With the average upswing lasting 3.2 years, the optimistic outlook includes hope that next year will again show EPS growth. Although 2005 is the third positive growth year, the optimist could see that one more year of EPS growth is possible--rounding up from two-tenths of a year to include 2006. Since the momentum is strong, let's assume that 2006 EPS grows by another 15%; a growth rate that is the average annual rate in past upswings and very close to the level that many are forecasting for next year.

Then, in 2007, let's include in our forecast a period that is representative of the typical cycle. From our earlier analysis, the lessons of history provide that the average downswing for EPS lasts 1.6 years. During downswings, the average decline in EPS is almost 15% based upon a nine percent annualized average decline. The resulting forecast for reported 2008 EPS is just over $64.

In section four of Unexpected Returns , the reader is guided through a tour that links basic concepts of finance and basic principles of economics to assess a reasonable outlook for EPS into the future. Based upon the relatively consistent historical growth rates in the economy and a strong relationship between economic growth and earnings growth, EPS is estimated for each year over the rest of the decade. This provides another fundamentally-based methodology to assess EPS based upon the long-term trend in economic growth. As presented in Unexpected Returns (pg. 133), the value for 2008 EPS using this second method is $63.30.

Conclusion - Flat and Choppy

If the cyclical patterns from the past fifty years continue, then stock market investors should begin to prepare for another downward leg in EPS. There are many economists and financial analysts that have begun to predict the next recession--and most of them certainly recognize that the current expansion, compared to the nine expansions since 1950, has extended past the mean average number of months and will exceed the median average length during 2006. In this article, we have explored two different methods using lessons of history. Both methods predict S&P 500 EPS near $64 in 2008. If overall stock market P/Es can be sustained in the low 20s--relatively high by historical standards--the stock market in 2008 will be near the same level that it is today. Flat, choppy markets are characteristic of secular bear markets. Of course, the bulls hope that maybe it's different this time .

(Ed Easterling is President of Crestmont Research is the author of Unexpected Returns: Understanding Secular Stock Market Cycles. In addition, he is a member of the adjunct faculty at SMU's Cox School of Business and teaches the course on hedge fund investment management for MBA students. Copyright 2005, Crestmont Research. www.CrestmontResearch.com)

Knowing that the market may be flat and choppy for another two years is one thing, but I suggest that we will see a return of an actual bear market in this period as a slower growth period or an outright recession appears next year. Investors will be disappointed (to say the least!). That disappointment will serve as the catalyst for lower valuations and the recognition that a longer term secular bear market is still in effect. Let's look at a study I wrote about in Bull's Eye Investing:

Evidence for Investor Overreaction

It typically takes years for valuations to fall in bear markets to levels from where a new bull market can begin. Why does it take so long? Why don't we see an almost immediate return to low valuations once the process has begun?

Investors overreact to good news and underreact to bad news on stocks they like, and do just the opposite to stocks that are out of favor. Past perception seems to dictate future performance. And it takes time to change those perceptions.

This is forcefully borne out by a study produced in 2000 by David Dreman (one of the brightest lights in investment analysis) and Eric Lufkin. The work, entitled "Investor Overreaction: Evidence That Its Basis is Psychological," is a well-written analysis of investor behavior that illustrates that perceptions are more important than the fundamentals. Let's look at that study in detail.

In any given year, there are stocks that are in favor, as evidenced by high valuations and rising prices. There are also stocks that are just the opposite. Dreman and Lufkin look at a database for 4,721 companies from 1973 through 1998. Each year, they divide the database up into five parts, or quintiles, based on the companies' perceived market valuations. They separately study price to book value (P/BV), price to cash flow (P/CF), and the traditional price to earnings (P/E). This creates three separate ways to analyze stocks by value for any given year so as to remove the bias that might occur from using just one measure of valuation.

The top and bottom quintiles become stock investment "portfolios" for all three valuation measures. You might think of them as a mutual fund created to buy just these stocks. The researchers then look 10 years back and five years forward for these portfolios. There is enough data to create 85 such portfolios or funds. They first analyze these portfolios as to how they do relative to the market or the average of all the stocks. They then analyze these portfolios in terms of five basic investment fundamentals: cash flow growth, sales growth, earnings growth, return on equity and profit margin. They do this latter test to see if you can discern a fundamental reason for the price action of the stock.

Here's my review of the most relevant parts of their presentation.

First, both the "outperformance" and "underperformance" of these stocks happens in the 10 years leading up to the formation of the portfolio. Almost immediately upon creating the portfolio, the price performance comparisons change, and change dramatically. The in-favor stocks underperform the market for the next five years, and the out-of-favor (value) stocks outperform the market.

I should point out that other studies, which Dreman and Lufkin do not cite, seem to indicate that the actual experience of many investors is more like these static portfolios than one might at first think. That is because investors tend to chase price performance. In fact, the higher the price and more rapid the movement, the more new investors there are who jump in. The 1995 Dalbar study ("Quantitive Analysis of Investor Behavior," www.dalbarinc.com), among many others, shows us that investors do not actually make what the mutual funds make because they chase the hottest funds, buying high and selling low when the funds do not live up to their expectations. The key word, as we will see later, is "expectations." Other studies document that investors tend to chase the latest hot stock and shun those which are lagging in price performance. Thus, forming portfolios of the highest- and lowest performing quintiles is an uncanny mirror to what happens in the real world.

Why does this "chasing the hot stock" happen? Dreman and Lufkin tell us it is because investors become overconfident that the trends of the fundamentals in the first 10 years will repeat forever, "thereby carrying the prices of stocks that appear to have the 'best' and 'worst' prospects. Investors are likely to forecast a future not very different from the recent past, i.e., continuing improving fundamentals for favorites and deteriorating fundamentals for out-of-favor issues. Such forecasts result in favorites being overpriced, while out-of-favor issues are priced at a substantial discount to the real worth. The extrapolation of past results well into the future and the high confidence in the precise forecast is one of the most common errors made in finance."

Some studies have shown the more we learn about a stock, the more we think we are competent to analyze it and the more convinced we are of the correctness of our judgment.

Since you are not looking at the graphs, let me describe them for you. Predictably, the fundamentals improve quite steadily for the first 10 years for the favorite stocks in comparison to the entire universe of stocks. But the price performance rises at very high rates, far faster than the fundamentals, particularly in the latter years. It clearly accelerates. It seems the longer a stock does well, the more confident investors are that it will continue to do well and therefore they award it with higher and higher multiples. The exact opposite is true of the out-of-favor stocks. Even though many of the fundamentals were actually slowly improving, in relation to the market as a whole they were lagging and the market punished them with ever lower relative prices.

At five years prior to the formation of a portfolio, the trends of each group were set in place. The next five years just reinforced these trends. This reinforces the perceptions about these stocks and increases the level of confidence about the future. Again, past (and accumulated and reinforced over time) perception creates future price action.

Never mind that it is impossible for Dell Computer to grow 50 percent a year or General Electric to compound earnings at 15 percent forever. As many times as we say it, investors continue to ignore the old saw "Past performance is not indicative of future results." And that is not to say Dell and GE are not wonderful companies. They are. But their share values, and those of any in-favor stock, eventually rise too high.

How much better did the well-performing stocks do than the poorly performing stocks in the 10 years prior to creating the portfolios? The highest P/BV (price to book value) stocks outperformed the market by 187 percent. The lowest stocks underperformed the market by -79 percent for a differential of 266 percent! If you look at the P/CF (price to cash flow), the differential between the two is 172 percent.

Yet in the next five years, the hot stocks underperformed the market by -26 percent on a P/BV basis and -30 percent on a P/CF basis. The out-of-favor stocks did 33 percent and 22 percent better than the market, respectively. This is a huge reversal of trend.

What happened? Did the trends stop? Did the former outcasts finally get their act together and start to show better fundamentals than the allstars? The answer is a very curious "no."

Dreman and Lufkin find that "there is no reversal in fundamentals to match the reversal in returns. That is, as favored stocks go from outperforming the market, their fundamentals do not deteriorate significantly; in some case they actually improve. . . . The fundamentals of the 'worst' stocks are weaker than both those of the market and of the 'best' stocks in both periods."

In some cases, the trends of the worst stocks actually got worse. Even as the out-of-favor stocks improved in relative performance in the prior five years, their cash flow growth actually fell from 14.6 percent to 6.6 percent. While cash flow growth for the best-performing stocks did drop by 6 percent, it was still almost 2.5 times that of the lower group. Read the following observations of Dreman and Lufkin carefully:

"Thus, while there is a marked transition in the return profiles [share price], with value stocks underperforming growth in the prior period and outperforming growth stocks in the measurement period, this is not true for fundamentals. In nearly every panel [areas in which measurements were made], fundamentals for growth stocks are better than those for value stocks both before and after portfolio formation.

Although there is a major reversal in the returns [prices] to the best and worst stocks, there is no corresponding reversal in the fundamentals."

In fact, in many cases the fundamentals continue to improve for the growth stocks and deteriorate for the value stocks. The data and the graphs clearly show the fundamentals for the growth stocks clearly beat those of the value stocks even for the five years after portfolio formation.

And yet, there is a very stark reversal in price. Why, if not based on the fundamentals?

Dremen and Lufkin go to another research paper, which shows "that even a small earnings surprise can initiate a reversal in returns that lasts many years." They demonstrate that negative surprises on favorite stocks result in significant underperformance of this group not only in the year of the surprise but for at least four years following the initial event. They also show that positive surprises on out-of-favor stocks resulted in significant outperformance in the year of the surprise, and again for at least the four years following the initial event. They attribute these results to major changes in investor expectations following the surprise.

So where was the overreaction? Was it in the years leading up to the surprise that resulted in a very high- or low-priced stock (relative to the fundamentals), or was it in the immediate reaction to the surprise?

Other studies show that analysts (as opposed to investors) are too slow to react to earnings surprises by being too slow to adjust earnings forecasts. Even nine months later, analysts' expectations are too high.

How Secular Cycles Actually Work

It works almost the same for the broad markets. Earnings disappointments produces underperformance. The next disappointment will produce more underperformance. The net recession/slowdown/earnings disappointment is going to erode investor confidence in the stock market. That means we will get lower valuations (P/E). As earnings recover and finally go to new highs (remember, they always do!), the lower valuations actually work to give us a lower overall market price.

If we do not get a serious recession, it may take two recessions to bring us to the final phase of this secular bear market. That would not be unusual, as cycles average about 13 years (8 being the shortest, 17 the longest), and we are only in the fifth year of this current cycle. Nonetheless, the next recession is going to give us an excellent buying opportunity. But right now, caution should be exercised. In the coming weeks (after I finish the current series), I will be writing about a number of indicators that suggest 2006 may see a real slowdown in the US economy. Stay tuned.

London, Thanksgiving and Yeast rolls

I am staying an extra day in London to go to some of the local museums. Carol Tambor tells me there is a fabulous Rubens exhibit at the National Gallery, and some marvelous impressionists at the Tate. Sometimes you have to stop and feed the soul.

I get back, try to catch up on email (it is so hard!) and reading, while getting ready for all my kids (7) and extended family to gather for Thanksgiving. I am going to make a prime roast and my brother will bring the smoked turkey. Last year I asked for recipes on how to make a prime roast and got some fabulous recipes. I have been craving yeast rolls (There are no calories or carbs on Thanksgiving!). I like the light, yeasty, buttery, almost melt in your mouth rolls. If you have a recipe that really stands out, feel free to send it. I am up for trying to make some rolls.

Family, friends and great food. We should have more Thanksgivings. Especially since we have so much to be grateful for. I have been poignantly reminded of late how precious life and family are. Most of us take so much for granted, and it is good to take stock and realize that we can always see the earnings that come from solid relationships grow above trend every year.

Your already tasting those rolls analyst,

John Mauldin

P.S. If you like my letters, you'll love reading Over My Shoulder with serious economic analysis from my global network, at a surprisingly affordable price. Click here to learn more.

• Earnings Per Share

Suggested Reading...

Last call for SIC 2024

Join our online community (it's free!)