Time to put on our thinking caps. This is an important and seminal e-letter, as it will help form the basis and rationale for future investment suggestions. Rather than focus on where the market is going in the next two weeks, today we are going to try to get an idea over where the market is headed in the next ten years. Armed with that information, we can more logically structure our portfolios to take advantage of the prevailing winds.
Warning: This is not a letter to read casually. I have spent almost three times the normal amount writing this letter. This letter reviews what will be some very new concepts to some of you. It may take some time. I think that if you take the time, you will be a much better investor in the future. The time reading this will pay off in your investment portfolio for a long time.
I have spent some time in the past reviewing Dr. Robert Shiller's foundational work, Irrational Exuberance. In brief, he shows that mainstream index investors have never made a profit over a period of ten years when P/E ratios have gotten to the high valuations they have today. It is just one clue to the puzzle of where the investment winds will be blowing, but it is an important clue.
We can find more clues in a ground-breaking book by Michael Alexander called "Stock Cycles". I am going to review his book at length because it will help us understand the fundamental causes of stock market cycles. Armed with this information, we will all be better investors. I will tell you how to get the book later. By the way, this book was written over January to March of 2000. His theory has accurately described the markets since then.
Let's jump to the conclusion first: Alexander's work shows that using past stock market cycles to predict the performance of the stock market one year from now is pretty much a random chance. Statistically, from almost any starting point, you have about a 50/50 chance of the market going up or down, using price movements alone to make your prediction. But there are certain long term cycles which are not random, and the probabilities of those repeating are very high. As you would expect, the patterns and techniques of successful investing changes somewhat dramatically from pattern to pattern and cycle to cycle. The trick, of course, is to figure out where you are in the cycle.
I have long been suspicious of stock market cycle theory, especially Long Wave theory. Long Wave (or Kondratieff Wave) theory says the economy and markets repeat every 56 or 60 years, with discernable periods marking the changing cycles. I readily concede that there are seemingly repeatable past patterns, but there are not enough data points to satisfy my need for any type of statistical certainty. It is an interesting theory that tells you where you have been, and tells you where you are going, but does not tell you where you are or when you will get there with any certainty.
I remember, as do many of my readers, how Long Wave theory predicted the end of the economic world in the late 80's. How many of you remember the direct mail flooding our mail boxes, not to mention the books, screaming gloom and doom? Obviously, they were wrong.
The reason is that too many analysts try to make Long Wave theory a precise predictive model. They do not look at the underlying fundamentals which cause the cycles.
It is like watching two men seemingly walking the same way in a large city. Maybe they are friends and are walking together. They could be total strangers either going to the same location, or getting ready to part ways on the next block. Until you know who the men are and where they are going, using their past travels to predict future events is simply guessing.
It is one thing to use the stars, as the ancients did, to construct a calendar to predict seasons, planting times and weather patterns. It is another to use the stars to predict personal fortunes. One has a basis in fundamentals, the other simply notices patterns which (like much stock market analysis) may have no connection or can be manipulated for personal benefit.
Alexander provides, at least for me, the missing link between the patterns in stock cycles and the underlying economic fundamentals. I now see a logical connection between the position of the stars and the seasons.
Alexander does not contend these cycles are as precisely predictable as the Spring Equinox. Rather, he suggests that when the underlying fundamental conditions occur, we can look for Spring-like conditions. Just as you plant certain types of plants in spring and certain types in winter, there are some investments which do better in their respective parts of the stock cycle. Carrying the analogy further, it is easier to grow your portfolio in economic spring than in economic winter. You have a much wider variety of "plants" from which to choose in spring.
You can plant spring crops during the winter, but you're going to have to wait until Spring to see them come up. It can be a long cold winter in the meantime.
To help us see what part of the cycle we are in, he first describes several types of stock cycles and then he looks at why these cycles may occur. We will keep to his pattern in this letter.
First, he takes a purely statistical view of the stock market, looking for repeating patterns. For his purposes, a period where the stock market out-performs money market funds is good and where it under-performs is bad. Is there any pattern?
It turns out the only statistically valid non-random cycle he can find is a 13 year cycle. Since 1800, there have been 15 alternating good and bad cycles of 13 years, from stocks being undervalued to being overvalued and back again. There was one period where the pattern instead of reversing, continued for an additional (and exact) 13 years. 2000 was a 13 year peak in his model. There is a probability of only 3.9% that this pattern is random.
Looking at the data, it would suggest that index investors have little hope for capital gains over the next thirteen years. Buy and hold investors will probably be better off in money market funds, just as they were in 1966 and 1929.
Simply based on this statistical model, Alexander concludes that there is a 75% chance of a negative capital gains return for index fund investors over the next 20 years. However, returns in any one year period are essentially random. Even in "over-valued" markets, the odds are essentially even that an index fund will outperform a money market fund for a 12 month period.
"Given today's low dividends and high valuations, a money market fund is, on average, a better investment over the next 5-20 years than the S&P 500 Index.... In the case of over-valued markets (like today), holding for longer time periods, even up to 20 years, does not increase your odds of success." He wrote that in early 2000, prior to the first crash.
Let me stop here and say that Alexander is not saying to avoid the stock market. He is simply pointing out, consistent with my long term theme, that buy-and-hold index investing will not work in this next cycle. Simply picking any old mutual fund and expecting a rising tide to raise your boat will only have a random chance of success in the next economic cycle. You have to change your investment strategy if you want to succeed.
In chapter three, Alexander looks at the historical cycle of bull and bear markets. First, he points out that stocks have returned about 6.8% per year in real returns (adjusted for inflation) over the last 200 years, but about 4.6% or two-thirds have come from dividends. The remainder corresponds to the real annual growth in GDP over that time. We analyzed a National Bureau of Economic Research study a few weeks ago which demonstrated this very point. The stock market does not grow faster than the economy. If it goes too high or too low, it always comes back to trend.
But stock prices fluctuate dramatically. There have been 7 secular bear markets and 7 secular bull markets since 1802. These are periods of at least 8 and up to 20 years where stocks are either generally rising or falling over the entire period. There are, of course, bear market rallies and bull market corrections, but the long-term trend is still either up or down.
If you were in the stock market during the 95 years of the bear market cycles, you only achieved a 0.3% annual average rate of return. If you picked the 105 years of the bull market cycles, you made a 13.2% rate of return. Your actual returns for any one ten year period would be totally dependent upon when you made your initial investment. The cycle length from peak to peak is 28 years on average.
Is there some model we can use to look at the overall cycle to help us determine why the dramatic price movements? Here Alexander provides a new way to look at price fluctuations.
He looks at a ratio he calls P/R, or Price to Resources. "Resources are simply the things (plant, equipment, technical knowledge, employee skills, market position, etc.) available to the business owner to produce a profit. R is essentially retained earnings, or that portion of profits used to invest in the business to grow the business.
While P/R (like P/E or Price to Earnings) is not particularly useful for predicting individual company or industry performance, when looking at the market as a whole, a clear pattern develops. P/R peaks at bull market tops and rebounds at bull market bottoms.
But the fluctuations do not appear to be as volatile. That is because while earnings may swing wildly from one year to the next, retained earnings are not subject to such wild swings.
Management continues to use current resources and invest in new resources in an effort to increase the business, even in recessions. Plus, resources tend to accumulate over time. Companies with large resources can weather tough economic conditions better and can come back more quickly.
There is a direct relationship between earnings and resources. As the resources of a company or nation accumulate and are put to work, the company or nation becomes more prosperous, and earnings increase. If a nation (or its businesses) fails to increase its resources, the ability of those resources to produce a profit will decrease over time. That means earnings will decrease.
The collective P/R ratio is the estimate of the value investors put on the ability of an economy to produce earnings. Now it gets interesting, at least for me.
Understanding Stock Market Behavior
Earnings, we are told, are what drive the price of a stock. But earnings growth for the period 1965-1982 was roughly the same as for 1982-1999. Yet we all know that the S&P 500 had significantly different results. The first period was one of no stock price growth, and the latter saw growth of over 1000%.
What was the difference? Clearly, it was how investors perceived the relative value of the earnings. In a period of high inflation, earnings growth of 6-7% is not all that impressive. In today's low inflation environment it is.
"Since the Civil War cycle there have been two effects of inflation. First, inflation reduces the value the market places on earnings, resulting in a flat trend, rather than a rising trend in the index. Secondly, the effect of the cheapening dollar makes the real value of the index fall even further. As a result, P/R falls to extremely low levels during inflationary bear markets."
When inflation ends, you get the benefit of the old earnings growth and new growth, giving the market a double boost. Investors become very optimistic about earnings growth and adjust their future value of stocks accordingly. But as I have often asserted, trees cannot grow to the sky. For 200 years, the overall market has not grown much faster than the growth in GDP. Remember the study we examined a few weeks ago which demonstrated that very fact?
Now we enter a period where the expectations of earnings growth cannot match reality. The stock market must come back to trend, which can be a painful adjustment for some investors. Alexander notes, "The situation is very similar to 1929. The effect of both the monetary conditions and a very optimistic assessment of the earnings growth still to come are priced into the index. This is shown by the extraordinary high level of P/R. We should expect the current monetary cycle to be followed by a "real" cycle [More later]. It should start with a secular bear market in which lower earnings growth will be the problem, not inflation.
The goal of every business is to grow its income and to grow its income at a faster rate over time. The income you get for the money you invested, or the profits you generate from a given level of resources is called the Rate of Return or ROR.
However, there appear to be very real upper limits on both the absolute value of and the growth of the ROR that can be achieved for a given level of resources. This ROR fluctuates over time, just as P/E and P/R do. Why wouldn't ROR be constant, as many firms try to do? Why can't ROR just grow every year, as market cheerleaders on TV constantly predict?
What appears to happen over time is that firms, in a moment of optimism, either build too much capacity or resource (R) and the ROR drops as capacity utilization drops; or, firms invest too little and thus the growth of ROR is self-limiting.
Managers simply cannot know the exact amount of future resource needed. They can do their best to make very intelligent guesses, but in the end there is usually either too much or too little resource.
It is a difficult job. Too much resource and you don't get a reasonable return. You use resources which cannot be easily re-allocated to some more productive use. Too little and you invite competition or give up market share. Further, there is that nasty thing called competition which makes it possible for a lot of businesses to build capacity for the same market, all hoping to increase their business and market share. Then you end up with too much capacity and no ability to raise prices. Computers, oil, soybeans, ships, etc. The list is endless. Supply and demand works. The business cycle is real.
In the telecommunications industry, management decided the world needed large amounts of fiber optics cable. We now use less than 5% of the capacity of that new cable. Clearly, the industry overbuilt. But all the firms which supplied equipment for that expansion also assumed that the future would look like the past and built large factories capable of building massive amounts of fiber optic cable equipment. The over-capacity went right down the food chain.
The 90's were characterized by the growth of capacity in almost every industry, including "mature" industries like agriculture, shipping, mining, retailing, etc. We now have a new level of total "R" or resources available to US businesses and the world. But since economic growth and profits do not grow faster than GDP, whatever growth we do have will be spread over a larger amount of Resources.
This means the rate of return of "R" will be smaller than it has for the last ten years. It follows that the growth of earnings will be smaller as well.
Expansions and Expectations
One of the great charts in Stock Cycles shows the relationship between the length of economic expansions and the expectations investors have for the stock market. The longer we think economic expansions will last, the more we are willing to pay for earnings which will compound at 15% forever. Every time we come to a period like the one we are in toady, we are told that this time it is different.
If earnings truly could compound at 15% forever, a P/E ratio of 25 would not be illogical. But earnings cannot grow faster than GDP. Period. Earnings will come back to trend.
Repeat: this is because we build (or invest in) too much resource for a given market or technology. The potential profit is spread over a greater amount of resource, and earnings growth suffers.
Long Waves Explained (Finally)
Alexander then jumps to the Long Wave cycle. Greatly simplifying, the theory says that there are two sets of stock market cycles in each economic Long Wave. You have a bull and bear market which are mostly influenced by monetary policy and events and are followed by a bull and bear market cycle which is mostly influenced by "real" events, such as earnings and economic performance.
The theory then says:
"The extraordinary gains in recent years results from investors discounting future earnings growth over longer periods of time. This makes the market extraordinarily leveraged to the economy....The average length of economic expansions was shorter during the 1970's than they were either before or since. The [coming cycle] could also be characterized by short business cycles like in 1883-96 rather than a lengthy slump like in the Depression. Shortened expansions would gradually shift the market from a future-oriented to a present-oriented valuation scheme, resulting in a contraction in P/E. The result would be a secular bear market as the valuations slowly adjust, even though economic growth might be fairly good. This, of course, is what is predicted to be imminent by P/R."
Alexander shares my concern, which I mentioned previously, about the lack of connection between the Long Wave theory and the actual economy. But he has, in my opinion, found a connection which not only provides the missing link, but when taken to its logical extension, offers some very exciting prospects for future investments.
The economists Schumpeter and Mensch both tried to establish a theoretical base for the Long Wave based upon bursts of innovation. More recently, Harry Dent (the Roaring 2000's) has expanded upon their work. Alexander uses Dent's terminology to put forth his own new thought.
The importance of this process is straight-forward. If you agree with Alexander's logic, then you will have "two, largely independent, periodic phenomenon that we can use to characterize the changing economic environment that brings about the stock cycle."
Dent sees the innovation cycle being comprised of four periods: the innovation period, the growth boom, the shakeout and the maturity boom. Alexander calls the end of the maturity boom the economic peak, which is the time when the economic impact of the new innovation has been completely played out.
Basically, a new process or technology is invented such as the cotton gin, telephone, electricity, airplanes, computers, etc. Following a period of innovation, there is a rapid growth of the "New Economy". Not surprisingly, there is too much capacity built and a number of companies falter.
During the shakeout, there is another process going on. We see a second innovation phase of the mature technology. Companies which come up with new innovations now see a second growth boom prior to the final "maturing" seen in the economic peak.
Now we come to the best part of Alexander's work. He goes to a number of sources and derives 9 different innovation cycles beginning in the early 1500's. While this or similar efforts have been done before, what Alexander does that is new is to relate these cycles to their importance to the overall economy: What proportion of the GDP did these innovations contribute to growth?
Over time, as the innovation becomes mature and new innovations come on the scene, the talk is of the "New Economy" changing the world and replacing the "Old Economy." But eventually even the "New, New Thing" becomes mature and plays a less significant part of the economy as even newer innovations appear. It is a repetitive cycle. It is no different than what we see today. The cycles and phases are eerily the same.
Basically there is a connection between the Long Wave and the innovation cycle that seems to have "worked" well enough for the last cycles or about 500 years. Alexander notes that the Information Economy seems to have come about 17 years later than the average 53 years. Thus, rather than being mature in the 1980's, it was just beginning. If nothing else, that explains why the Long Wave theorists were wrong.
There is nothing magic about a Long Wave of 53 or 56 years. What is important is the Innovation Cycle. It is the latter which influences the economy. Analysts who used the Kondratieff or Long Wave as a time prediction tool were wrong. The usefulness of the Long Wave is to help us analyze what is the basic nature of the underlying economy and how the Innovation Cycle is affecting the Economy.
Finally, Alexander writes of Harry Dent's projection that the long boom will last until 2007, which corresponds to the Baby Boom generation: "Dent's alignment of generations and the spending wave with his phases of the innovation wave seems to break down after going back more than one cycle."
The book is only $14.95. The last chapters on the innovation cycle alone are worth the price of admission. There is much more in the book than I can hope to comment on here. I suggest you read it. You can get the book here or from Amazon.com. If you buy the book directly from Alexander's publisher (iUniverse.com) he gets more money. He deserves it. I cannot recommend it highly enough.
Stock Cycles is just one more clue that we have examined this year which leads me to the conclusion that we are entering a different phase of the stock market. It probably began in 2000.
The "winners" will be investments which focus on absolute return strategies (as in income mutual funds and certain types of hedge funds) and value based investing (such as that espoused by Warren Buffet and Graham. As always, new innovations will always bring rewards to investors. In the last decade, a rising tide lifted all boats. This cycle, stock picking will be critical.
Investment styles which depend upon 15% compound earnings growth will be frustrated. There will invariably be large rallies which could last for months as investors yearn for the profits of "yesteryear", but as earnings growth fail to catch up with stock price growth, these will be set aside. Astute traders will be able to take advantage of these moves. This will be an environment in which "dynamic asset allocation" should be useful to investors who want a more aggressive growth potential. Hedge funds which specialize in opportunistic trading and Long-Short styles should also do well.
At the beginning of this cycle, bonds should do well. Although the theory and historical analysis also seems to suggest that inflation will come back, we should have some more time in our bond funds.
Finally, Alexander's work on innovation cycles is very exciting to me. Catching the next growth cycle, no matter when it begins, will be of great value. What will spur the next round of huge growth? Bio-tech? Nano-tech? Or some new tech yet to appear?
There is no reason to sit around this next cycle and miss out on the fun. We just need to wait until the Three Amigos tell us when to get back in, invest conservatively and watch the Yield curve. That is what I will try and help you do in my regular e-letter.
New Orleans, New York and Costa Rica
For business reasons, I have decided to jump down to the annual New Orleans investment conference this Saturday, Dec. 1. I know some of you will be there, and would be delighted to meet with you.
I will be in New York on December 10-12 speaking at a hedge fund conference in Midtown Manhattan, and will have time to meet with clients and potential clients.
My bride and I will be taking a week-long (and deserved, I think) vacation to Costa Rica just prior to Christmas. I will be in San Jose for one night and then on to a Pacific Coast jungle resort. I would be interested if any of you have ideas on what we should see in San Jose. I will use the time to clear my head so I can write this year's annual 2002 prediction issue the week I get back. I have promised the Love of my Life I would keep business thinking to a minimum, which she deserves for putting up with me the rest of the time.
Finally, I hope you do not get the idea I am some Gloomy Gus predicting the end of the world. The world will not end; it will just be different than it has been. Our goal is to invest realistically, in tandem with the cycles, so that we can relax and enjoy life. Our investments are just a means to an end, not the end in themselves.
God has blessed us. It seems to me that to spend our time worrying about the future and not enjoy our blessings and sharing them misses His point entirely. Let's be good stewards, make good plans and then enjoy our lives.
Your having more fun than the law allows analyst,