Tuesday, January 26, 2010

On Market Timing

It is a common refrain from mutual fund and investment management marketing material that investors should assume that it is impossible to time the market. In reality, with a fairly simple toolbox small investors can take steps to largely avoid major bear markets while enjoying the bulk of large, multi-year bull markets.
One might wonder, if this is possible, why everyone does not follow such a system. There is a different answer to this question for each major type of investor.
Large investors like pension funds, mutual funds and institutional portfolio managers can not react quickly enough to take advantage of the signals. It takes several weeks, often months, for very large investors to enter and exit stock and bond positions. A rapid exit would have a dramatic effect on the stocks being purchased or sold, and in the case of very large investors, on the markets themselves. For these investors, ‘Buy and Hold’ is the only option, so this is the message they preach in their marketing materials. Further, investors who can not sell do not invest in the development of market timing strategies.
Small investors who own stocks or mutual funds often deal with an Advisor with a large number of clients. Under regulations in Canada and the U.S., most Advisors must call each client for approval to buy or sell securities in their portfolios. An average Advisor with 200 clients to call may take a week or more to contact everybody. Often, these Advisors choose not to act rather than undertake the Herculean effort to reorient their clients’ accounts for changing market conditions. Further, market timing signals usually require quick action. As these Advisors can not act quickly to reallocate all their clients’ funds, they do not invest in the development of market timing systems.

Source: Butler|Philbrick & Associates
Click for larger version.


A small proportion of Advisors are licensed to make changes to all their clients’ portfolios at once without having to call. These Advisors usually have special qualifications, such as a CFA or a CIM, and significant experience such that regulators grant them permission to manage client portfolios with discretion. In Canada, professionals who service individual clients, and who possess a discretionary license, are designated Associate Portfolio Managers or Portfolio Managers. Often these Advisors manage small amounts of capital, usually less than $1 billion. Their discretionary license enables them to act quickly and decisively to protect or deploy their clients’ assets.
Associate Portfolio Managers and Portfolio Managers usually have the training, experience, and regulatory ability to take advantage of high quality market timing systems. However, very few invest the time and money to develop and test trading and timing systems that they can apply confidently to client portfolios. Perhaps they have misplaced faith in the Modern Portfolio Theory they learned in school. Perhaps they haven’t heard about Behavioral Economics, or tested the assumptions of MPT with real data.
Whatever the reason, this is unfortunate, as the payoff to those who take the time to research, develop and rigorously back-test trading and timing systems can be enormous.

Consider a simple timing system for a stock market index with a single signal line based purely on historical stock market index price data. If the market index closes above the signal line, an investor would purchase stocks. If the market index closes below the signal line, an investor would sell his stocks. The following chart illustrates this approach using the Dow Jones Industrial Average from 1966 through 1983. The stock market index is the blue line, and the signal line is red. If the blue line crosses below the red line, sell stocks. If the blue line crosses back above the red line, buy stocks.

Source: Butler|Philbrick & Associates
Click for larger version.

At certain levels of granularity (for example, using monthly closing data rather than daily or weekly), such a simple system yields significant results. By broadening the system to include more asset classes (i.e. commodities, foreign stocks, etc.), and simply allocating an equal portion of portfolios to each asset class, one is able to generate impressive results indeed.
The following chart illustrates a system that discretionary managers can easily apply to client accounts. Since 1973, this system has delivered substantially better results than stocks, and at a fraction of the market risk. The system’s worst year was 2009, with a 0.01% loss on the year. Further, the system has delivered positive returns in 92% of all 12-month periods.

Source: Faber (2009), Butler|Philbrick & Associates
Click for larger version.
A corollary myth in the market timing domain relates to the assertion by many buy-and-hold advocates that an investor must be in the market at all times to enjoy long-term growth. These misinformed ‘experts’ often cite statistics that show returns to a portfolio that was out of the market for the best 10-months out of the last 10, 25, 50, or 100 years. Obviously, the returns to a portfolio that missed the best 10  months of market returns will do substantially worse than a buy-and-hold investor. What the expert fails to mention is that the best months of returns in stocks usually follow or precede the worst months of returns in stocks. Further, these best and worst months usually occur during periods where a market timing model would have parked money in cash. In fact, over the past 138 years, 7 of the worst 10 months, and 8 of the best 10 months occurred during the Great Depression!
The following chart illustrates the returns to four portfolios. The blue line shows returns to a portfolio that managed to miss the 10 worst months. The red line shows returns to a portfolio that missed the 10 best months. The green line shows returns to a portfolio that missed both the best and the worst 10 months, while the purple line shows returns to a buy-and-hold investor.

Source: Butler|Philbrick & Associates
Click for larger version.
Observe that, quite obviously, investors who avoided big losses did better than investors that missed big gains. More interestingly, investors who avoided both large monthly gains and large monthly losses experienced approximately the same return over 138 years as an investor who held stocks from start to finish.
In conclusion, individual investors have an opportunity to commit the necessary time and effort to find a qualified, licensed portfolio manager who has invested in the development and testing of trading and timing systems to best protect and grow wealth. You are likely to pay more for the services of such a team, but the value you receive in return may deliver multiples of your costs.

On Monkey Brains



It is prudent periodically to upset the apple cart in order to see what sort of rot and grime are sitting at the bottom. This series, which will include four or five posts over the next few days, will explore some common myths about investing and the investment industry.

The modern investment industry is highly motivated to conceal the realities that I intend to expose. The fat margins investment firms enjoy are predicated on the assumption that the professionals at these firms possess knowledge and information that is not available to the masses. This may be true in some cases, but large, traditional firms are handicapped in ways that more than offset this value.


The single most important take-away from this series is this: almost no one in the investment industry is motivated to take you out of the market when the risk is high. Because of this, almost no one in the investment industry is motivated to create systems and tools that signal when to get out.


Imagine for example that Fidelity, with $1.57 trillion of assets under managment, instructs their managers to pull their funds entirely out of the market. This would cause quite a dislocation. So what did they do?  As necessity is the mother of invention, large money managers invented 'buy and hold'.

In this context, it makes sense to begin with a short primer on a a relatively new investment theory called ‘Behavioral Economics’ which is rapidly gaining in credibility, even among investment traditionalists. This theory addresses experimentally the many ways that Modern Portfolio Theory, the most widely adopted model in finance, fails to usefully describe reality.

An anecdote from the book SuperFreakonomics (2009) by Steven Levitt and Stephen Dubner illustrates the power of emotional biases to short-circuit rational behavior. The book describes an experiment orchestrated by Dr. Keith Chen where monkeys are conditioned to understand the utility of money as a way to acquire treats.

Once the monkeys learned they could trade a certain number of coins for different treats, the experimenters introduced ‘price shocks’ to test the monkeys’ rational adherence to the basic rules of supply and demand. When researchers ‘charged’ substantially more for one treat over another, monkeys bought less of the more ‘expensive’ treat and more of the less expensive. The demand curve slopes downward for monkeys as it does for humans.

To test for irrational behavior, the experimenters introduced two gambling games. This is where things really get interesting.

In the first game, a monkey was shown one grape and, depending on a coin toss, either received just the one grape, or a ‘bonus’ grape as well. In the second game, the monkey was presented with two grapes to start. When the coin was flipped against him, the researcher took away one grape and the monkey received the other.

Note that in both games the monkeys got the same number of grapes on average. However, in the first game the grape is framed as a potential gain, whereas in the second game it is framed as a potential loss.

How did the monkeys react? From the book:

“Once the monkeys figured out that the two-grape researcher sometimes withheld the second grape and that the one grape researcher sometimes added a bonus grape, the monkeys strongly preferred the one-grape researcher.

A rational monkey would not have cared, but these irrational monkeys suffered from what psychologists call loss aversion. They behaved as if the pain of losing a grape was greater than the pleasure of gaining one.

Up until now, the monkeys appeared to be as rational as humans in their use of money, but surely this last experiment showed the vast gulf that lay between monkey and man.

Or did it?

The fact is that similar experiments with human beings, investors, have found that people make the same kind of irrational decisions and at a nearly identical rate. The data generated by the capuchin monkeys, Chen says, make them statistically indistinguishable from most key market investors. So the parallels between human beings and these tiny-brained food/sex monkeys remain intact.”

Loss aversion is one of the central principles of behavioral economics. The behavioral literature is consistent in observing that people are about twice as sensitive to losses as they are to gains. Investors manifest this behavior by holding on to losing positions for far too long in order to defer realizing a loss. On the flip side, investors sell their winning positions too soon in order to avoid ‘losing’ their gains. 


For evidence that you or your Advisor may be a slave to the loss aversion instinct, look no further than your portfolio. If there are many positions with large losses that have been on the books for several months or years, loss aversion is a likely culprit. Successful investors sell their losing trades quickly while letting their winning trades run.

Loss aversion is just one facet of a broader theory of investor behavior called ‘Prospect Theory’. Behavioral Economics is much broader still, and fills many of the dangerous gaps that are not adequately addressed by Modern Portfolio Theory. Some other important behavioral vulnerabilities are outlined in the diagram below (click image for larger version).



Source: Butler|Philbrick & Associates

The next post in this series will deal with timing the market. It turns out that there are tried and true rules to distinguish between markets that are likely to trend higher and markets that are in jeopardy of steep drops and volatility. It is the development of, confidence in, and above all adherence to these rules that distinguish successful investors from their poorer peers.