COP Strategy
(Class OutPerformance Strategy)
An Investment Policy/ Strategy
Malcolm (Mal) Williams
January 2006
© 2006 Malcolm Williams
Please, no reproduction or distribution without permission.
Contact: 972.732.8887 / copstrat@sbcglobal.net
Rev 2.1
Summary
This paper describes a dynamic asset allocation investment strategy based on the short-term out-performance of ten to fifteen out of 80 asset classes and implemented with various funds invested in the top classes. This is neither a market timing strategy nor one that predicts markets. A survey of salient literature is provided. The strategy is supported by research studies reported herein and by references to literature. The centerpiece of support is a large software research model that contains the monthly performance of 50 different asset classes over the past 17 years. The model has many control variables which allow examining a host of investment strategy alternatives.
The strategy is:
- Asset class selection pays the most for your effort. Studies show that about 90% of a portfolio’s return is due to class selection. Eighty asset classes are used.
- Short term performance, or momentum, of asset classes is used as the basis of selection.
- Selection is based on “outperformance” of asset classes as opposed to absolute performance
- Funds (Mutual and ETF) are available for each asset class and are the preferred investment vehicle to implement a portfolio.
- Monthly review and action plan.
- Recognizing that this strategy does not avoid sudden, serious market down turns, but the latest research using the COP Research Model does offer great help in managing through such events.
- Staying informed in a time/effort efficient manner.
This paper describes why the strategy works, research results, recent actual results from using the strategy and special accommodations for market corrections. Further research studies are underway.
The recent actual results delivered a cumulative 30 percentage points better performance than the S&P500 over four years. That amounts to over 300% of the S&P return. Research studies, discussed herein, covering 10 and 15 years suggest a greater outperformance might be possible. The research studies show returns over twice that of the S&P over the past fifteen years. This outstanding performance is accomplished with less volatility than the S&P.
The paper was written for three purposes. The first is to serve as an investment policy for the author (in accordance with Charles Ellis’ book “Investment Policy”). The second is to serve as a vehicle for generating discussion with others. The paper also contains considerable investment opinion and literature references to serve the third purpose -- to be educational for family and friends.
Contents
Summary
Introduction
Terminology
Disclaimer
Background
Review of the Best Recent Publications on Investing
Further Background
The Strategy
Simply
WHY this Strategy Works
Asset Class Selection
Use of Short Term
Performance or Momentum
Focus on “Outperformance”
Funds are the Preferred
Investment
Monthly Review and
Action
Market Corrections
Staying informed in a Time/Effort Efficient Manner
HOW this Strategy System Works
Strategy Improvement With The Cop Research Model
Strategy Development Chronology
The
Basic Strategy
Handling
Market Corrections
New
Study Results
More Research
Market Strategy Principle
Portfolio Diversification, Why and How Much
Volatility Risk and Modern Portfolio Theory
Market
Timing
“It
is Different This Time”
Gurus
and how to Handle Sea Changes
Taxes
My Personal Experience With This Strategy
To Be Added
Recommended
COP Strategy
An Investment Policy/ Strategy
Mal Williams
January 2006
Introduction
The principal reason for a written investment policy is to guard against ad hoc revisions and hold to the strategy when short term conditions are most distressing and the policy is most in doubt. It also serves as a basis for periodic review to determine if one’s views are in tack or changing. (Ellis p 55)
In my view, a written strategy statement is a confirmation for my own self that my strategy is more than an ad hoc reaction to the environment. This strategy is by its nature an objective one which further helps to remove my emotion from my investing while capitalizing on the market’s inherent psychology, which is manifest in momentum.
Terminology
The investment industry uses a wide variety of words to describe different asset classes, styles, categories, sectors, segments and regions. This paper will use “asset classes” to cover any and all of these. In every case there are one or more indexes that cover each class; therefore, they are all “investable” using Exchange Traded or mutual funds. The investment industry uses the terms “relative performance” and “outperformance” in many ways. Outperformance is a central term in this strategy and will be used here to mean the superior or top performance of an asset class, relative to the performance of all other asset classes. Herein, performance is measured in terms of the short-term, total return performance over the last 12 months. The term “sea-change” will be used to mean a sudden and significant change in the market. As will be seen, it is more in terms of the number of asset classes that change in the leadership of the market than in the market averages. As a result, the number of asset class leadership changes offers a degree of measurement of the degree of the sea-change.
Disclaimer
I have no, none, <0, nada, professional investment credentials. My academic background includes a Master in Science from UCLA. I am an avid reader and an active investor. I conduct my own research. These are my views from my reading and research which you may share at your risk (or not). In the Background, below, I have included quotes, from others, that the reader will find are in direct conflict with this strategy. I do that just so that the reader will understand that this strategy is not at all like those recommended by these popular, competent authors or the investment community. The reviews below provide a framework for understanding and for contrasting this strategy with conventional strategies. If you do not accept full and complete responsibility for your investment decisions and their consequences, read no farther, this is not for you.
This paper is very critical of the investment community and in particular brokers and investment managers. For the most part the critical points come from books that are highly regarded in the academic community. Obviously, there are people in the investment community who are exceptions.
Background
To provide a background for this policy/strategy statement I will first quote, or paraphrase for brevity, from some of the best books, in my opinion, on investments in recent years. It will only cover books and points that are salient to this policy. [Comments in brackets are mine.] {Comments in braces are my summary paraphrases}. These are all amongst the best books on investment strategy. They are all recommended reading, but I do not agree 100% with all or any of them. There are many more books that are intended for people who want to buy individual stocks. These will not be reviewed because, as discussed, that is an undertaking at which very few can succeed - - actually a foolish quest for all but a dedicated stock picking professional, few of whom have good performance. There are also other good publications listed at the end of this paper as recommended reading.
(Interestingly most Wall Street publications, and even the better books, are like country and western songs, that is: “somebody done somebody wrong” and “I can save you.”)
Review of the Best Recent Publications on
Investing
The Four Pillars of
Investing by William Bernstein (2002)
- The competent investor never stops learning.
- The pattern of annual returns is almost totally random and unpredictable. No one, not the large brokerages and certainly not your broker or investment advisor can predict where the market will be tomorrow or next year. [Why do they try? They are paid very highly to try – it sells very well. Did you ever see one give you their track record?]
- The market loses money about one third of the years.
- Expect at least one, and perhaps two severe bear markets in your investing career. [Since we have just gone through a severe market drop, another is not likely for a long time; however, we are likely to not be through with this bear market. Bull and Bear markets beginnings and endings are like the start and end of recessions; they subjective and are not reliably recognized until well after the event. As discussed later, a bear market has just as many years up 10% as does a bull market. History also shows us that market swings are becoming less severe than in the past, and there is usually some asset class doing well somewhere. As a friend of mine says “there are few ill winds that don’t have a good twist for someone.” See Mauldin’s and Easterling’s books recommended below for the best discussion of bull and bear market characteristics.]
- In the long run bonds are at least as risky as stocks. As pointed out by Jeremy Siegel, “stocks outperform bonds only 61% of the years but they outperform in 80% of the 10 year periods.” [This strategy treats bonds as just another asset class -- use them when they are working.]
- The real value of [understanding] the historical record [of financial markets] is as a gauge of risk not return.
- [Bernstein and others go to great lengths to prove that past performance of mutual funds does not predict future performance. Like others he uses past “long term” performance of 5-years and then looks at the results over the next 5-years. Agreed, that’s using ancient history trying to predict the distant future. Much more later]
- David Dreman’s: The Psychology of Stock Market Success painstakingly tracked the opinions of expert market strategists back to 1929. They were wrong 77% of the time. The worst were the market timing newsletters.
- Since you cannot successfully time the market or select individual stocks, asset allocation should be the major focus of your investment strategy, because it is the only factor... of your risk and return that you can control. {Funds are his recommended investment choice.}
- For 20
years, from 1970, the best performing assets were Japanese stocks,
- Paraphrase: {Managed (as opposed to index) mutual funds should never be used in a taxable account because some of them are very tax-inefficient with regard to distributed dividends and gains. The most efficient funds for taxable accounts are ETFs.} [In my opinion “never” is too strong since managed funds seriously outperform indexed funds in a few categories, however, if your tax bracket is high enough “never” may be right, especially in core market holdings were the markets are quite efficient.]
- Your broker is not your buddy. [A long discussion of the seamy underside of the brokerage and investment management business pointing out that brokers, in fact the whole brokerage firm, are sales people and not competent to recommend investments. They thrive or die based on sales fees. Why else is it that the most common short-hand terms to distinguish between the brokerage industry and investors are “sell side” and “buy side”. Guess who gets sold.]
- Neither is Your Mutual Fund [your buddy]. [More of the same—pointing out that fund management incentives are in proportion to the size of the funds they attract (through promotion), not their performance]. [somebody done somebody wrong]
- The popular media is at best worthless and often dangerous. Paraphrase: {The Forbes Honor Roll of funds .... 1974 through 1998....returned 13.6% before loads and taxes ... while the S&P index fund returned 14.3% with no load and very little taxes.} To whom do you listen? ... the market is your best advisor ... [that is the only advisor used in this strategy]
- The only guidance you need is your self discipline and your overall asset allocation.
- Paraphrase: {In earlier chapters Bernstein thoroughly destroyed the Modern Portfolio Theory tenants that are used by virtually all brokers and investment advisors to establish an asset allocation strategy; and then he proposes his own based on a range of asset class characteristics that are no better supported. His strategy, while based on rational analysis, also has no basis for its predictive power or future validity. He says that all you need are large market stocks, small market stocks, large value stocks, small value stocks and REITS.} [If those were all you had to chose from and you think that a given allocation can last you until he writes his next book, then it’s probably as good as you can get. We will be working on a dynamic strategy with about 80 asset classes to choose from -- listed in Appendix A.]
Stocks for the Long Run, Jeremy Siegel, 2002
- Over the long run (since 1801), stocks are by far the best investment, even beating out collectables, real estate, etc. ........ gold is the worst, returning nothing after inflation. From 1982 through 2001, after inflation, gold lost 4.8% while stocks returned 10.5 %, long term bonds returned 8.5% and short term bonds returned 2.8%. The same is generally true of stocks and alternative investments in foreign countries.
- It takes about 20 years to be confident that buy and hold investments will be positive, however, the probability they will be positive is very high at about 10 years.
- There is undeniable evidence that the economy has become more stable over time and this should help stock prices [assuming it holds]. This {and several other factors} means that the average historical P/E of 14-15 is no longer appropriate – the low and even high 20’s are justified as long as inflation stays low and tax policy remains favorable.
- Future [from 2002] long term real returns on stocks should average about 5 to 7%. Inflation should be low. [Therefore, nominal returns should be about 7% to 9%]
- Paraphrase: {Risk is popularly known to be the hand maiden of return – you can’t have one without the other.} [Today’s research has shown that risk, as measured by Beta, is not a predictor of return and that Beta has no predictive value at all.]
- All of this implies that the average investor will do best by diversifying into all stock sectors. [He recommends a relatively fixed asset allocation.]
- The principal motivation for investing in foreign stocks is more complete diversification. [For me it is returns first.]
- In the long run [only] stocks are an extremely good hedge against inflation.
- Business cycle and economic forecasting are notoriously unsuccessful [But it pays so well! -- because the gains are so great if correct]. Despite growing knowledge and an abundance of data, predictions are not getting better. Turning points [especially] are rarely identified.
- Utilize funds as the preferred means of investing. Up to one third of your portfolio should be in small or mid cap funds.
- The evidence in favor of “index” investments in small cap and international stocks is not yet persuasive.
- Small value stocks significantly outperform [long term] small growth and micro-cap growth stocks do the worst of all classes.
- {Very few, if any, individuals have the skills or resources to select individual stocks, especially, over a broad range of asset classes. Funds are much the better investment choice.}
Global Bargain
Hunting, Malkiel and Mei
Investment Policy,
Charles Ellis
- The evidence on investment managers’ success with market timing is impressive – and overwhelmingly negative.
- [This book on the importance of a written policy was targeted at professional investment managers – it can help us all.]
The New Contrarian
Investment Strategy, David Dreman
The average (stock) analyst (between 1982 and 1990) was wrong by 40%.
Contrarian Investment
Strategies, The Next generation, David Dreman, 1998
Rules:
Do not use market-timing or technical analysis
Do not make investment decisions based on correlations [they will change] {all of the major brokerages use correlations for their asset allocation strategy}
There are no ... industries in which you can count on analysts’ forecasts.
Diversify extensively
Volatility
is not risk. Avoid advice based on
volatility.
The (Mis) Behavior of Markets, Benoit Mandelbrot, 2004 (the originator of fractal geometry) (Great for understanding the highly irregular volatility of markets)
{Markets are much more random than believed.}
{Disturbances come in clusters and build on each other.}
{Market disturbances predict more disturbances, but not the direction of the disturbance}
[Does it not follow that lack of disturbances, therefore, predict calm? It is well documented that many of the most severe market disturbances occurred after periods of abnormal calm.]
The Intelligent Asset
Allocator, William Bernstein, 2000
In practical terms it is nearly impossible to find uncorrelated assets. Consequently, we cannot hope for a risk reduction of more than 25% to 30% from diversification.
...the actual correlation of asset returns in severe bear markets is higher than the average correlations... Diversification fails when it is needed most. [This is an extremely important point. See discussion of “down markets” below.]
Unconventional
Success – A Fundamental Approach to Personal Investment, David F. Swensen, 2005
[Investment strategy in brief] ... a well-diversified, equity oriented portfolio using not-for-profit [funds, i.e. ETFs and index funds from Vanguard or equivalent]
[Re: contrarian investment] Investment success requires for more than taking the other side of a market trade.
Another important piece of research on [investing] performance suggests that 100% of returns are derived from asset allocation. [as opposed to timing or individual security selection]
[Dramatically reduce risk of underperformance] by avoiding asset classes that rely on superior active management—{i.e., hedge funds or hot fund managers}.
Diversification provides the free lunch of improved return and risk characteristics, while equity orientation promises the possibility of greater wealth accumulation.
{If you doubt my attacks on the investment community and media (below), read this, very qualified, guy – he makes me seem like milk toast.
Further Background
It is popular with brokers and investment advisors, and their lackey in the media, to disparage mutual funds by telling the public that “very few mutual funds can beat the market” -- that is, do as well or better than the S&P500. What they do not tell you is that very few try. Most are trying to offer alternatives, such as bonds, small stocks and other diversification. Still let’s see, according to Morningstar data (November 2005).
Of all the funds in Morningstar’s data base:
% of %that beat S&P 500
# Total Total Return
3-yr 5yr 10yr
All, Total 21505 100 comparing apples and oranges
All large
cap
All large cap Blend (=S&P) 2183 10 33 31
(Cap = capitalization, a measure of the size of the stocks in the fund)
So we can conclude that only about 31% of all funds are US
equity funds and 78% of these beat the S&P, but, only about 10 percent
really try to compete with the S&P 500.
When you narrow the definition of a fund down to the same as that of the
index, (large cap blend) then the index does beat about 70% of them. Still the
33% is over 600 “large cap blend” funds (or equals) that beat the S&P500. Most important, there are thousands of funds
that beat the S&P500 when their asset class is in favor. Of all
Most asset allocation strategies, that I am aware of, including those in several of the books I recommend, base their allocation on historical correlations between asset classes. The lower the correlation between assets the greater your diversification. Below is a table with some correlations from Morningstar, published in a 1998 article in the Wall Street Journal, for investors to use to select diversified asset classes. The second column has the correlations that Morningstar currently reports. Note that less than 8 years later the change is dramatic.
Class 7-1998
Correlation 12-2005 Correlation
U.S. Large growth to Utilities 39 53
All asset allocation strategies from the investment managers, that I am aware of, also require that you annually “re-balance” your allocation, usually annually, back to your starting allocation percentages. This reduces the amount you hold in what has done really well, because they believe that it is surely about to do poorly, and increases the amount you hold in what has recently done poorly, since they believe they are surely about to do much better. Amongst simple minded fixed strategies this actually works, at least when tax consequences are not considered. Certainly a more dynamic allocation strategy would do better if we had one that works. So, read on.
Think about this industry standard strategy of re-balancing annually and using allocations that are based on long term returns and correlations. Does this sound like a strategy designed for minimum effort on the part of the investment manager – a great strategy for him, or you?
“It ain’t the things that people don’t know that’s the problem. It’s the things they do know that just ain’t so.” --- Will Rogers. “They” just might be the investment industry. For example, (speaking of strategies designed for the broker, not you) according to Rob Arnott (a Nobel Prize level investment researcher in my opinion), the broker’s “classic standard” 60/40 (60% stocks/40% bonds) balanced portfolio is not true diversity. It sounds great – having 40% in bonds for safety. But the dirty little Wall Street secret is that it has a 98-99% correlation with stocks. That means that you get 98-99% of the risk of stocks. Rob’s research tells him that true diversification requires lightly correlated, risky classes, not low risk classes. Pure heresy – brokers can’t sell that. Yet, Rob is highly successful and runs a major research firm.
Almost all independent researchers and investment authors (all of the authors above) will tell you that the average investor (and in fact, other than experts) cannot competently pick stocks or time the market. Their major focus should be on picking an asset allocation and diversification strategy and a strategy to implement the asset allocation in a manner that meets their individual comfort zone with regard to volatility or risk. That is the focus of this strategy.
Mark Carhart, Goldman Sachs Quantitative Research: {A fund performing high one year can be expected to be above average the next year, but only average the third year.}
Mark Hulbert, NYT 7/7/2002: (paraphrased) {Morningstar’s high star rated funds and Louis Rukeyser’s Honor Roll funds, both based on performance of three or more years, have both underperformed the Wilshire 5000 Index by 5.5% and 10% respectively from 1991 through mid 2002.} [I intend to make, and support, the point that short-term performance measurements are superior for judging future performance, but only for the short-term future.]
A CDA Wiesenberger Study (Original 1974 and revised 1992): {The study showed the results of two hypothetical investors--- one who had perfect market timing and one with perfect asset-class selection (using industry sectors) over the periods from 1942 through 1991. Perfect timing outperformed the S&P500 by over 20 times -- wow. Perfect asset-class selection outperformed perfect timing by a factor of 1000 times.}
I get invitations to dinner presentations by investment managers almost weekly. I have attended many. Without exception, they present “canned” strategies based on well outdated principals and analysis. In my opinion, they are not just weak -- they are wrong. It only takes a few technical questions to reveal that there is no in-depth or technical understanding. They are a sales organization that fronts for a correspondent account management firm. They are rewarded by fees. They only know the book and lyrics they have been taught in their training. Unfortunately, to the uninformed, they are very impressive. Furthermore, unless you have several million dollars to invest they can’t afford to do any better – and neither can you -- at finding a competent investment manager.
The Strategy
Simply
This strategy is to select an asset allocation from the top ten to fifteen asset classes that are performing now (last 12 months) from amongst about 80 asset classes (Listed in Appendix A) (Appendix A will be explained later) then to implement the allocation with the currently (again, last 12 months) top-performing funds. This is reviewed monthly and changed if necessary. It uses short-term performance which can be viewed as momentum or inertia. The reasons that this strategy works are provided below. But first, briefly, the strategy is:
- Asset-class selection pays the most for your effort. Eighty, fund-investable, asset classes are used.
- Short-term performance, or momentum, of asset classes is used as the basis of selection of the top-performing funds for investment (e.g., 10 to 15 out of 80).
- Selection is based on “outperformance” of asset classes as opposed to absolute performance. The best performing asset classes, even in a down market, are not that bad, as will be shown below. The focus on short-term performance and monthly reviews will allow reacting much faster in the event of down markets. “Outperformance” is relative to the other asset classes, and the relative performance of asset classes is a much slower moving measurement than the absolute performance of a class. Furthermore, asset classes, and the funds that represent them, are less volatile than the individual stocks that make up the class.
- Funds (Mutual funds and Exchange Traded Funds (ETF)) are the preferred investment vehicle to implement a portfolio. Choose the best funds in the best classes.
- Monthly review and action. (My experience is that you will not trade excessively)
- Recognizing that this strategy does not solve sudden, serious market down-turns.
- Staying informed in a time/effort efficient manner.
Now you can see the basis for the name of the strategy: Class
Out-Performance Strategy or COP Strategy.
WHY This Strategy Works
1) Asset Class
Selection
The fact that asset class selection, or allocation, is the most important part of any investment strategy is extensively and well documented in literature, research and media. As referenced above, most researchers believe that asset selection is responsible for 90% of your return. A newer study by Ibbotson and Kaplan, published in the January/February 2000 issue of Financial Analysts Journal, again found that 90% of the return (good or bad) of investment portfolios and funds is determined by the asset class selection. Only 10% is attributed to selection of investments within an asset class. This was an up date of a 1999 study by Ibbotson and Kaplan and of famous 1986 and 1991 studies by Brimson, Hood and Beebower. With this level of acclaimed academic research it is difficult to understand why so many people “play” at investing in a random selection of stocks. As Bernstein says, “asset allocation should be your major focus”.
2) Use of Short Term Performance or Momentum
On reading the book “Ubiquity” by Mark Buchanan, I find that historians and social scientists believe that (a) maladjustment is a precondition to all sudden and dramatic changes (e.g., economic disruptions, revolutions and wars) in all communities, no matter the character or size of the community, and (b) the distress from the maladjustments must become great enough to overcome (c) “the greatest of all social forces – inertia.” That is a powerful statement “the greatest of all social forces – inertia.” This explains a lot about investments.
Inertia must, therefore, necessarily, be present in financial market communities, which explains why momentum strategies (like this one) work. In fact it seems almost undeniable that momentum works. It is just a question of how well one can manage to benefit.
Long term performance (3, 5, 10, 20 years) is the basis for virtually every asset allocation strategy used in the investment industry. It is also used to select winning funds. Morningstar uses long term performance to assign their star ratings. The fact is, however, that their 5-star rated funds do not show outstanding ongoing performance as has been thoroughly researched and documented by independent sources. Investment managers crow about this to attract clients, then, do their version of the same thing with no independent verification of their ability to select funds. I am not aware of any studies that show that long term momentum is predictive. A word used in science for momentum is inertia. Inertia better describes human behaviors, like herd mentality, than does momentum. If market momentum is a measure of human behavioral inertia, it seems unreasonable, to me, to expect people to have long term inertia. Six to 18 months is a long term for the newest “hot” investment.
So then, what evidence is there that short term performance works to select investments that will perform well in the near future?
Roger Ibbotson and William Goetzmann (professors at
This issue was also studied by three
The No-Load Fund-X Investment Service uses a short term performance selection model to select specific mutual funds for their portfolio. According to the July 2005 Hulbert Financial Digest, out of over 500 investment advisor letters, No-Load Fund-X is the top (or in the top few) for periods of 10, 15, 20 and 25 years on both the total return and risk-adjusted return rankings. They are in the top rankings for all advisor letters, not just mutual fund focused letters. For the last 5 year and 10 year averages the No-Load Fund-X portfolio has beaten the Wilshire 5000 index by 10% and 20.9%, respectively, per year. That is consistent performance that works. No-Load Fund-X does not examine or rank investment classes, just funds directly. Still, their work validates that short term momentum is a valid way of picking near term investments.
The formula for No-Load Fund-X’s fund ranking score uses a simple average of four of each fund’s short-term total-returns, and is:
(12-mo% + 6-mo% + 3-mo% + 1-mo%)/4 plus bonus points
Bonus points are awarded for being in the top 15 in each of the four periods in the formula. The bonus points are, therefore, also based on short-term outperformance, albeit relative to the other funds.
No-Load Fund-X recommends, and follows in their portfolio, an “upgrading strategy” that has investors trading into the new highest ranking funds and selling those that drop below the top. This is both a momentum strategy and the total opposite of a buy-and-hold strategy usually followed by indexing gurus and major brokerages, and is exactly what is recommended in this policy. In a strong sense it is also an outperformance strategy since they flag top performers regardless of market conditions or the asset class that the funds may specialize in.
One of John Mauldin’s (See Recommended Reading) featured guest papers by James Montier (he is often interviewed in Barron’s) features a research study designed to prove that “value” investing is better (and thus, you need them). His study shows that short term price momentum (he used six month return) does improve your return. Their research shows that for value investments, and most others, there is and increase in return of about 7% and for growths investors it is about 10%. Not a bad day’s work for paying attention to short term momentum, as in the COP strategy.
3. Focus on “Outperformance”
An Ibbotson Associates and Morgan Stanley study reported in
Financial Planning, Feb. 1997, provided a month-by-month bar chart showing when
In their January 2006 Quarterly, Fiduciary Trust International produced a chart showing the monthly outperformance of equities versus bonds from 1987 through 2005. In this chart there were many periods of outperformance of one over the other for periods of a year or more with one 5 year period of stocks over bonds and one two year period of bonds over stocks – essentially the same results as Ibbotson Assoc. above.
For the six years, up until mid 2005 small stocks outperformed large stocks. The long term average of periods when small outperforms large is 5.7 years. A total return bar chart of large cap growth, large cap value, small cap growth and small cap value for the years 1990 through 2005, from Morningstar, shows 7 periods in which a given class was in first or second place for two years or more.
While the S&P was down about 40% in 2000, 2001 and 2002 you could have during that entire period broken even in small caps and made about 10% in small-cap value. After that you would have made even more excess over the general market because small cap continued to outperform.
Therefore, we are focusing on a slower moving wave than the market as a whole, giving use some hope of being in the right class and of having a cushion in market downdrafts.
In December 1995, Fisher Investments published a research paper entitled “Beating the Market with Style (read “asset classes”) Rotation Strategy. From the Fisher paper: “historically the market shows defined periods of outperformance and under performance for given styles...” [these periods offer] “semi-predictable style rotation.” “..there is a large spread between [the performances] of winning and losing styles.” Fisher maintains that if one could only stay in the top four and avoid the worst two, of 6 styles studied; one could beat the market by an average of 3% points per year. This paper started my effort on this strategy and the momentum model. I followed their paper with my own research using Morningstar.
A 2002 paper entitled “Multi-style Rotation Strategies” by Ahmed, Penn State University; Lockwood, TCU; and Nanda, T. Rowe Price, came to similar conclusions to mine and Fisher’s as evidenced by their title and sub title of “The Benefits are Considerable.”
In my study using Morningstar data I knew I would be relying on past performance data to select asset classes. So I calculated the return I would receive if in each full year I invested in the two, out of 6, asset classes that were the two top performers in the prior full year. Therefore I was one year late, every year, for a full year – a worst case. I compared this to the average of having invested in all six classes each year on a current basis. The two top, but one year late, classes beat the overall average by several points. I was sold.
I have misplaced my original study that I did in the late 90’s, so I have repeated the study using 14 asset classes and the 10 years through December 2005. This is provided in Appendix B.
The Appendix B study provides still more outstanding support for the returns that might be made by focusing on asset class selection and on short term outperformance. Examine Appendix B, then go back and read the quotes from Fisher, above, and you will see why he made them since they seem to exactly apply to the data in appendix B. Surely we can gain some advantage from monthly reviews of performance.
The period under study in Appendix B is very interesting
because it contains both the booming late 90’s and the bummer early 2000’s; and
most interesting, the dramatic sea change in favored classes that took place
between the boom and the bust. Note that the One-Year-Late top four have great
results in boom of ’97, ’98 and ’99, easily
beating the S&P; and then completely miss the sea change between ’99 and
2000, and have a disastrous 2000. In
2001 and 2002 the strategy ignores the gurus who claim the old growth will
recover and stays with the conservative investments. The net is that the S&P drops 38% over
2000, 2001 and 2002 while the one-year-late top four only drop about 18%. By the end of 2003 the late-four are back
ahead by 17%, whereas the S&P is still down 20%. The strategy completely missed the sea change
from growth to value assets but caught on fast.
Surely we can catch on faster with monthly reviews and not being tied to
year end data. While monthly data is noisier than annual data, the last 12
months should be the same and is available every month. Note that while the one-year-late top classes
missed the sudden sea change in 2000, they performed reasonably well during the
“trending” down markets of 2001 and 2002 even though in 2002 the S&P500 was
down over twice the percentage it was down 2000 (-9.15% versus -22.12%. The reason was that there was not the sea
change in the type of asset classes that were performing at the top. We will examine this more below under Market
Corrections.
The reason that asset classes have sustained periods of out/underperformance is that there appears to be a herd mentality in favoring various classes. In addition, as Fisher points out there are real reasons associated with the economic cycles and market cycles (and I would add currency cycles) why various classes will out/under perform for a sustained period. Regardless of the reason, relative performance is a slow(er) moving waveform.
No one is able to predict in advance what will outperform or if it will make money, but, everybody tries. In fact, most of the investment advice you see in the media and from brokers and gurus is trying to predict by telling you what is “going” to be hot next, especially in individual stocks. Fisher’s paper stated that their analysis would allow them to predict the next top classes. This COP Strategy does not try to predict. I prefer to call it “taking short term bets” even “educated hope” but I do not assume that it can predict anything.
This strategy approach is simple. Ignore all of the media/brokerages forecasts, and only invest in the asset classes that are actually performing in the short-term momentum model. This is a bet that the momentum of whatever is outperforming now will continue just a little longer. Since we do not invest in individual stocks, we can then count on the relatively slow movement in the change in relative performance of asset classes. Classes of assets also move slower than individual stocks. For example if really bad news hits a stock -- say a semiconductor stock -- many technology stocks can drop quickly. Then it takes time for investors to realize that the problem is going to be wide spread and domino, such that the entire class will eventually drop, but on a more gradual basis.
Appendix B results were so impressive I decided to do a more complete study using virtually all asset classes except the small countries. This is shown in Appendix C. The Appendix C study fully confirms the results of Appendix B
I included the standard deviations in this study. It shows, as one would expect, that investing in 15 is less volatile than investing in just the top 10 and the returns are also a little less.
Looking at the Investing in Top 15 and Top 10 lines, one can see that even in the bad markets of 2000, 2001, and 2002, there were always asset classes that could produce positive results.
It is very illustrative to look at the colored highlights on page 3 of App. C. Note that all of the contiguous years in which each asset class stayed in or out of the top 15. There are very few spots on the table that do not have one color or the other, indicating one-year periods of in or out of the top 15 are not the norm.
4) Funds are the
Preferred Investment
Utilizing funds (mutual and exchange traded) as the preferred investment vehicle is almost as extensively and well documented in the literature (especially scholarly literature) as is the need to first focus on asset allocation. While there are a large number of books that purport to tell you how to select stocks and get rich, I am not aware of a single serious book that does not recommend that the public should use funds and not individual stocks. The fund managers are your “hired gun” stock pickers, providing you a “specialist” in each class. You can’t hope to compete with them.
I recommend that you subscribe to Morningstar Principia. Soon you will find that you can sort out the best funds based largely on short-term performance and risk. Don’t listen to the investment managers/brokers, who try to sell their services by claiming that good past performance does not usually repeat-- but theirs does. Morningstar’s “star” selection system does not work as a sole selection criteria, as Morningstar so states. It is based on long term performance, so use it only as a very low, last level of criteria to select funds. I feel it tells you more about the quality of the fund management. Do not ever buy a fund based on a broker’s recommendations. They do not have training in fund selection, only in selling a fund profitable for them -- you can’t blame them. Do your own selection where the sun shines.
The important point is to select the very best funds, but only after you have chosen the best asset classes to get 90% of your return.
5) Monthly review and
Action
Albert Wang (2001
There is absolutely nothing magic about the year end. There is no reason not to review and change any time you are sure you want to. The only issue is how often you can get a good update on market data and how much time you choose to spend. Once each month is reasonable for the former. Each month you may make no change, minor changes or overhauls. This is an iterative strategy that keeps you adding new high performing classes and dropping those going out of favor. It is also a check on your particular fund selection within a class. Clearly you want to be paying a lot closer attention during market sea changes. For this, on a weekly basis, you can follow the data Morningstar publishes on the net for free (Morningstar.com) and on the results in your personal portfolio (Schwab’s At a Glance page is very useful).
This only works if you have a portfolio performance tracking method that provides you with the performance of each fund and your overall portfolio. One should track returns on a monthly, quarterly, 12-month and year to date basis. Track your returns along with indexes for the same periods, so that you can compare to the market and to alternative classes.
Reading this strategy and the research, one could easily conclude that one might be rapidly trading funds. I don’t. I have several funds that I have had for years. As mentioned before, asset classes maintain their relative outperformance for years at a time.
Still, I am prepared to trade every month and this is a lot of work. It is important work for a life time.
6) Market Corrections
As stated, in following this strategy, you are betting that past short term performance will repeat in the near term future. You are bound to lose this bet at times. Here we will examine the conditions that cause this as well as what kind of down market conditions still allow one to make money with this strategy. Below is a chart showing the performance of the Top 15, the Top 15-one-year-late, and SPY from Appendix C. In addition, it shows the number of asset classes that changed each year.
First a couple of observations: The Top 15 handily outperform SPY except in two years, 1997 and 1998. That was due to the fact that SPY was ranked at #2 and #4 respectively in those years. When something as stable as the S&P is performing so well one would probably not bother to hold 15 asset classes. The top 5 or so would probably offer adequate diversity. Similarly, in 2000, 2001 and 2002 when all asset classes are doing so poorly, one would probably not be holding more than the top 5 or so – notice again that in these shaky years it is the very stable bonds that are performing the best. So in both cases one should do better.
Now note that there are two years, 2000 and 2003, in which there were a large number of changes in the asset classes in the Top 15. In both cases the Top 15-one-year-late falls away from the Top 15 by about 35 percentage points; whereas, in years where there are a nominal number of changes, the two stay closer together. In 1997, 2002 and 2004 were the number of asset classes that change are fewest, the Top15 and Top 15-one-year-late are quite close together.
The message is that the number of asset class changes represents the most important measure of sea changes to the COP Strategy, not the markets performance. This should be intuitive based on the fact that this is momentum-based; however, this is something that can be monitored each month and all months instead of only at the end of each calendar year, as is done in the Appendix C Study. Perhaps, this will prove to be somewhat of an indicator here to bet heavier or lighter.
One last observation on down markets: One can see from Appendix C that the top 10 to 15 asset classes produce positive results for the investors in even the worst down markets. Thus, the problem of down markets is not when to get out and back in, but rather, a problem of being in the right asset classes. It is a tough problem, but the data analysis used in the COP Strategy is a big help and it is void of emotion. Who knows how good one might be? In 2002 when the S&P lost -22%, even the one-year-late study made money. Personally, I feel I am improving in both skill and discipline.
See more specific concepts about down markets below under Handling Market Corrections and Volatility and Risk, and Modern Portfolio Theory.
7) Staying Informed
in a Time/Effort Efficient Manner
Bernstein: The popular media is at best worthless and often dangerous. [The term “financial pornography” came up.]
If you are not interested in individual stocks, you have just eliminated a major portion of wasted reading on the part of investors. I personally only read for general education, trends, the state of the economy and the general market. Keep a world wide focus. This reading is to keep your head up to be more cautious or more aggressive. Keep in mind Siegel: Business cycle and economic forecasting are notoriously unsuccessful. Keep a near-term focus and stay tuned in for change.
I only read about 5% of the Wall Street Journal, 5 % of Barron’s, still, I consider them the best. I read John Mauldin’s weekly economy e-letter and new investment books from what appear to be scholarly sources (and that avoid individual stocks). I read what gurus have to say if they are top notch gurus.
A discussion group is wonderful to stimulate your thinking and education. Don’t invite brokers or investment managers – they are well trained in selling strategies that work for them.
Read every good book on the market. Read none that tell you how to select individual stocks, commodities or derivatives or on timing or technical analysis. Be critical in your judgment of these books, especially if the author comes from the investment industry. Seek out and rely more on scholarly authors and publications. These would include professional peer-to-peer journals as well as academic sources. “It’s not so much what you know anymore that counts ... It is how fast you learn.” – Robert Kiyosaki, author “Rich Dad – Poor Dad.”
Follow the top gurus as you would a discussion group, but, do not follow their advice. All of this staying informed is to help you understand why things work as they do and to keep your head up for changes in the environment.
Books on market psychology have been in vogue for the last several years. The field seems to be aiming at helping investors take advantage of the “emotional other” investors. Right now, I think they best serve to help you not be one of the “other.” While this is not Dreman’s only focus he is one of the old masters.
HOW This Strategy System Works
1) Use short-term performance, or momentum to select from the top 10 to 15 (out of 80) asset classes that are outperforming now.
Just after the first of each month, I produce an Asset Class Performance and Momentum Report from Morningstar’s Principia. This is available 5 to 7 days after the end of the month and about 50 of the asset classes are available on line the morning of the 1st of the month. A sample of this report is provided in Appendix A. There is a Major Asset Classes report and an Individual Country report in Appendix A. There is some deliberate overlap in the two for comparison purposes.
The various asset classes are on the left of each report and near the right is the “momentum” performance ranking measurement in the column labeled APA. The report is further explained in Appendix A. The COP Research Model has allowed me to test several momentum calculation methods. The one I have chosen to use from that study is called APA. This formula was suggested by my son several years ago. I have used APA for a couple of years now.
2) Use short-term performance to select top performing mutual funds in each of the categories you have chosen.
I use Principia which reports on both mutual funds and on exchange traded funds (ETF).
3) Create a diversified portfolio from the top-performing funds in the top-performing asset classes.
a) Build as diversified portfolio as you can from those asset classes that are working now.
b) Don’t be afraid to invest in 15 to 20 different funds or none.
c) Create a portfolio based on your risk tolerance
There will be times when there is virtually nothing that is performing and that looks safe enough for you – stay in cash or don’t be afraid to hold only a few asset classes. The fewer you hold, under 4 or 5, the greater your class risk and the greater need for a substantial portion of your portfolio to be in cash. At other times you may find several asset classes are performing quite well and can give you a reasonably diversified portfolio. In which case, do not be afraid to not hold any cash. Also do not be afraid of building a portfolio of funds that do not meet someone else’s idea of diversification. Diversification into assets that are not performing will do you no good. Almost all brokers will tell you to have some assets is a particular class, say bonds, at all times. If bonds are performing at a low or modest level relative to stocks you still might want to have them for comfort. However, you can not just look at the dividends from bonds when the value or the bonds themselves may be dropping. If the “total” return from bonds is not reasonable, you are losing money in absolute terms and certainly after inflation.
Cash is just another asset class. The Ultra-Short Bond class included in Appendix A is considered as a good surrogate for cash in your planning. If it is working it will rise to the top.
What about income? Selling profitable capital assets provides the same kind of dollars as does a dividend. Invest in what works and take your income from profits of any kind. But, everyone says that taking your income from dividends saves reducing your capital when stocks are down. Actually, only when you are in an environment when bonds are working and stocks are not, otherwise, it is not true. Let real performance, not dogma, guide you.
Be patient; be sure momentum is there. There can be a lot of head fakes that can get you in or out too soon. By the way, no system, certainly not a momentum system like this one, is going to work well when there is a sea change. So stay informed to stay alert and cautious at times. Know full well that you will miss the sea changes for a while and suffer for it. You have to decide how long of a while is tolerable or if you would rather bail out. I expect to bail out somewhere in the down 10% range (for the entire portfolio) and take the hit if there is an immediate bounce back. There is no market wisdom here, just a gut feel for my tolerance to losses.
Baron’s, 5/29/06; Steve Leuthold Group Commentary: “Market inflection points are – almost by definition – hostile to trend-following strategies, even very refined ones like ours ----- There is no tool -- fundamental, technical, psychological or otherwise – that helps one to ride the “fat” part of the trend and then to exit gracefully at the top ---- we do not view this sort of setback as a very steep price to pay in exchange for the opportunity to cash in on longer term group trends.”
By now the reader must understand that this strategy produces a very dynamic asset allocation and that this is in direct conflict with the position of highly respected authors and the bulk of the investment community. The reason that the contrary quotes were placed in the background section was to be sure that the reader understands this.
Can you do it yourself? Unless you have greater than $5M you cannot hire a competent personal investment manager. So you do not have much choice, and yes you can if you study hard enough. Again, it can be the most important work of your lifetime.
STRATEGY IMPROVEMENT
WITH THE COP RESEARCH MODEL
Strategy Development Chronology
As mentioned above, I started developing and using the monthly momentum data model for this strategy in early 1996 following my mini research, using Morningstar’s Principia, which was modeled on and followed a 1995 paper by Fisher Investments. I used the momentum data as one of my inputs over the next several years, but cannot claim to have followed the strategy with the slightest rigor. In fact, in late 2001 I wrote a personal investment policy for my own discipline which did not even mention this strategy. Like many people, 2000 and 2001 was not kind to my investments and this was part of a self introspection of my investment skills. In the process, I noted that in 1999, where there were no large market corrections, the strategy was a big help. I also started a re-examination of the basis for the strategy and decided to follow it more closely. As a result my fortunes improved; and with my increased reading, I became more convinced that the strategy worked.
After “falling off the wagon” and suffering a week year in 2004, I again recognized that I would have been better off following the strategy. I resolved to become more disciplined and in 1995 wrote another investment policy for myself, family and friends who were also following the strategy to some degree. To support that effort I went back and re did my 1996 research. The results looked so good that I did a much larger study. That looked great too. All of this research used 10 years of annual data, which was all that was available on my Morningstar Principia subscription. However, in the real world, I made “monthly” investment decisions based on Principia’s monthly report. I wondered how much better the research results would be if I had monthly data or would they be worse because the monthly data is so much more volatile. So, in early 2006 I bit the bullet and acquired 25 years of Morningstar’s monthly return data for 50 different asset classes. I then wrote a large software research model called the COP Research Model. The results were outstanding –even better than with the earlier research. Additionally, the COP Research Model has a wide range of control variables to allow comparing many different investment alternatives. As one example, from a host of research cases that have been run, I have learned a great deal more about how to handle market corrections (see discussion later).
The COP Research Model
As indicated, the model has a data base using Morningstar’s 25 years of monthly-performance data on fifty asset classes. I only excluded classes such as money-market funds and most muni-bond funds. I then developed a huge software model (using Excel and VBA language). The model contains a control panel in which I can set a wide range of parameters to test investment alternatives or cases. Some of the investment alternatives require additional Excel or VBA code changes.
So what was the result of running a one-month late strategy as opposed to a one-year late? What did the monthly volatility do to results? The research cases showed that it was possible to beat the S&P at much lower volatility, or to really trounce the S&P at the same volatility. About the same return as the S&P could be attained at about 60% of the S&P volatility and, at about the same volatility, the annual return over 15 years was 70% higher than the S&P. The COP strategy in the COP model shows only one down year in the last 15 years; and curiously, that is 1994 where it loses a few percent while the S&P gained a little over one percent. That year is to be examined in further research (was there nothing that gained in that year?). In the 2000 through 2002 market the model dropped a little harder in the first few months, but learned fast and actually made money during that period, whereas, the S&P was down about 40%.
The more asset classes one invests in at one time, the lower are both the volatility and the return. To achieve about the same return as the S&P one would invest in about 40 of the 50 classes at one time. To run at about the same volatility as the S&P (with 70% higher return) one would invest in about 8 of the 50 at any one time. Keep in mind that in certain months all 8 of the classes owned might be the highest risk classes and in other periods some of the lowest risk. Still, the model gives you 15-year results including the 2000 through 2002 market drop. Thus, this strategy allows each investor to choose the level of volatility they want