Some of our readers have asked us to discuss market timing, that is the portfolio strategy of making wholesale asset shifts between stocks and cash. Market timing might seem to be a good idea. By deftly buying stocks when you think the stock market is going to rise, and selling stocks when you think it is going to fall, it might be possible to turn a stock account into a kind of turbo-charged savings account.
We consider the market relevant mainly as a portfolio performance benchmark. Benjamin Graham (ed. 1973) wrote:
"The investor's primary interest lies in acquiring and holding
suitable securities at suitable prices...He should always remember
that market quotations are there (only) for his convenience..."
Investing is about investing in real companies. The following remarks assume that you are already invested in stocks; here are some of the hurdles that market timers face:
1) The stock market, by its nature, is volatile with an upward bias. According to Ned Davis Research, in the ten years between 1986 and 1995, the stock market returned 14.8% on an annualized basis. These returns would have dropped precipitously if an investor had been out only a few of the best trading days during this period.
Annualized Returns
All 2526 trading days 14.8%
All trading days minus the 10 best days 10.2%
All trading days minus the 20 best days 7.3%
All trading days minus the 30 best days 4.8%
All trading days minus the 40 best days 2.5%
If the investor had missed thirty of the most profitable trading days, about 1% of the total, his stock market returns would have been no better than 4.8%; the return of a savings account. Although the study does not include the effect of losses possibly avoided, it does show that being out of the stock market can be costly as well.
2) Stocks are not for the very conservative investors who expect positive returns at all instances. A company's operating income and its stock price can temporarily decrease even though its long term prospects remain bright; furthermore, the capitalization rate of a company's earnings per share will certainly change if long term interest rates change substantially. What makes financial returns predictable are contracts, loans for banks and short term obligations for investors, which have a prior claim to operating income.
It may be as easy to purchase a stock as a bond, but the two are not equivalent in a company's financial structure. Attempts to provide stocks with all the downside protection of short term bonds are not likely to be profitable in the long run.
3) If you are already invested, there are two apparent reasons to do market timing: a) increasing long range returns; b) avoiding a possible short term loss. The problem of earning good long range returns, we think, is better solved by investing in good companies. You can mitigate the effects of a short term market loss by doing appropriate asset allocation in the first place, by making some adjustments to this allocation appropriate to circumstances, and by investing in the real businesses that you like.
It is usually not necessary to make extreme asset allocation changes if your stock portfolio is appropriately structured.
In this article, we suggested that modulated portfolio adjustments are possible; some of our readers saw only market timing. It is always "rational" for investors to sell stocks, in general, at high prices in order to buy them back at lower ones. The defense that markets have against this is macroeconomic uncertainty, resolved through time. We thought of removing this article; but in the interest of a balanced presentation, have not done so.
"The whole art is to vary the emphasis and center of gravity of one's portfolio according
to circumstances."
John Maynard Keynes
"Economic Articles and Correspondence"
(1938)
Is there a way to consider your stock and bond portfolio as a single asset? Horizon is mainly a long term investor; we tend to be invested in stocks whose earnings are less affected by the economic cycle. We do, however, try to align a portion of our portfolios with likely macroeconomic circumstances. The following discussions have shown that fluctuations in stock prices are determined mainly by expected inflation. So, as a matter of fact, are bond prices. It is therefore possible to view a stock portfolio as a sort of bond portfolio.
The prices of long term bonds are much more volatile to changing interest rates than are those of short term bonds. Bond portfolio managers use the concept of duration to measure how the prices of their portfolios are likely to change when interest rates do. A bond portfolio's duration is somewhat like its average maturity, except it also includes the timing of all cash flows.
This concept can also be applied to equities as well, with the caveat that real stock market behavior is more complex than a single dimension can encompass. A useful formula (Casabona, 1984) to calculate the duration of a stock is:
(1 + r )
f
Duration = _________________
(1+r ) - (1+g)(CE)
f
where: r is the risk free short term interest rate
f
g is the growth in dividends
CE is the certainty equivalent ratio
(between 0 and 1) of the stock; which
compares the return required by
certain earnings (cash flows) with the
higher rate of return required from the
actual investment's risky earnings
Our readers should be aware that financial economics is useful mainly for its qualitative insights. What this equation suggests is simple:
1) When you think interest rates are going to increase, reduce the duration of
your portfolio by increasing the amount of cash held and decrease the certainty
equivalent of your stock investments. That is, substitute cyclical risk for interest
risk by investing in cyclical stocks rather than in bond like interest sensitive stocks.
Be also aware that cyclical risk is just that.
2) When you believe the business cycle is about to turn down, begin to do the opposite.
As econometricians know, it is impossible to predict interest rates exactly. It may be somewhat possible to determine the direction of interest rates given cyclical economic developments. Portfolio management is therefore partly a matter of making duration changes that align the portfolio at or slightly ahead of likely macroeconomic circumstances.
Is there a systematic method for evaluating investment risk? The Gaussian bell shaped curve delineates on a continuous scale all the outcomes that an investor can face and possesses a central tendency, say an annual investment return of 10%, to reflect the fact that a single outcome is more likely than any other. Very conveniently, the central tendency of the Gaussian distribution is the population average. The Gaussian distribution is most appropriately applied to those sciences where there are fixed objects or effects to be measured.
Is the Gaussian curve, particularly its central tendency, applicable to stock market returns? Peter Bernstein (1996) writes, "...all economic data are specific to their own time period...Real time matters more than time in the abstract..." We think that economic developments in fact evolve; they are not like the sampling of marbles from a jar.
Consider, however, a possibly useful but qualitative definition of risk that the Gaussian model provides. In any single year, investment risk is equal to the population variance, that is the dispersion of all years' stock returns:
_
Yearly Investment Risk = Population Variance = f(x(n) - x)
where: x(n) is an observation
x is the population average
More generally, we can say that:
Investment Risk = f(x - E(x))
Risk, here, is a function of the difference between that which actually happens, (x), and that which is expected, E(x). According to this formula, risk includes both that which objectively happens set against the subjective estimate of what is thought most likely to happen.
Risk has an unavoidably subjective component until the present becomes the future. Until that happens, expectations, Palley (1993) writes, are "...bound up with knowledge, learning, and error recognition." The way to reduce risk is to have more accurate expectations, which means to have increased knowledge. We have been suggesting this includes the knowledge of that which is knowable and that which is less so:
1) We have shown that the level of the stock market is described by a component related to Earnings per Share and a more complex component related to the business cycle and Fed policy. It is one of our main suggestions that the best way to achieve good long term portfolio growth is to invest in good companies. The reason for this is the degree of possible knowledge. It is easier to know about a company's markets, management, and strategy than about the Philips curve during any particular economic cycle.
We hold some cash and derivatives in our stock portfolios when the general market is overvalued relative to bonds and when macroeconomic circumstances seem to indicate. Index puts are costly insurance policies, but they can be used to reduce short term volatility. To keep this all in perspective, we do not turn this into a strategy because to do market timing is to rely upon matters that are less certain.
2) It is a good idea to know whether you are primarily an investor or a trader.
3) You might set your goals according to the term of your financial requirements and analyze your risks according to this term. A major principle of financial management is to match long term liabilities (requirements) with long term assets (such as stocks). The following assumes that you are already invested in stocks, that your asset allocation is appropriate to your circumstances and tolerance for volatility, and that the international environment remains benign: If long term interest rates rise appreciably, the price of stocks will drop; but so will the present value of your long term liabilities. In this view, risk is not short term volatility; but the risk of not meeting your long term financial goals.
We have suggested that you can reduce your long term financial risk by choosing companies for long term growth.
"Human nature is...a tree, which requires to grow..."
John Stuart Mill
(1859)
The following suggests a perspective when considering the future and illustrates how economics discusses life.
The philosophical basis of economics is utilitarianism, the doctrine of Bentham and Mill that states the aim of actions is to maximize well-being in the broadest sense of the word. By the following model, a rational person will maximize utility (U) by spending resources on goods (G) until their additional satisfactions are all equal. That is, a person will solve the equations:
Max U (g , g , g ,...g )
1 2 3 n
subject to: budget constraints, or more generally time
constraints and the rules of the game.
(G) refers to the goods that money can buy. It
can also refer to the personal and civic goods
that are included in an enlightened self-interest.
Consider a two period consumption model discussed by Becker and Mulligan, 1994. If your utility does not include equations, please skip to the next paragraph.
Max [ U (c ) + U (c ) x Beta(S) x p ]
0 0 1 1 1
subject to: the budget constraint c + c + S = A
0 _____1___ 0
( 1 + r )
1
where: U (c ) is the utility of present consumption.
0 0
U (c ) is the utility of future consumption.
1 1
Beta(S) is the personal discount factor that people apply to the future.
S, this is a new economic variable, is the level of resources spent on
envisioning the future. S is the amount of time spent on bringing the
future closer to the present (psychologically) by education, planning,
problem solving, or by thoroughly (as possible) understanding the
consequences of what you are doing.
p is the probability that c will happen. There are, in fact, a number
1 1
of states and their associated probabilities.
r is the long term interest rate.
1
People will maximize their present utilities by trading between present and future consumption until their marginal benefits are equal . The principles of utility and optimization produce results that are in accord with evolved common sense (this bear does dance):
1) The longer a person's expected lifetime, the further he should plan into the future.
2) The higher the profitability of investments or a person's net worth, the more patient he should be.
3) When impatience is widespread, that is when long term interest rates are high, investors should be more patient. Value investors please note.
4) Under conditions of uncertainty, people should spend more resources anticipating the most probable states with the highest utilities. This is why you should research your investments.
The Becker article suggests when to be patient.
"Begin...with the things which (are) the simplest
and easiest to understand, and gradually ...reach...
toward more complex knowledge, even treating, as
though ordered, materials which were not
necessarily so."
Descartes
"Discourse on Method, Second Part"
(1637)
We have supplied two major building blocks: market and company analyses. These two building blocks are usually considered independent. Now consider the major unspoken assumption that underlies almost all investment discourse - that of the investment horizon, here defined as the maximum amount of time you are willing to wait before beginning to see the capital appreciation you expect, assuming that you have done your investment homework.
Consider the following scale:
Investment Horizon
Short------------------------------------------------Long
(less than one year) (longer than one year)
Information - Current News and Valuation and Company
Set Used Market Analysis Analysis
Profession - Trader Investor
If you are, by nature, a trader you will be most concerned with predicting the near term course of the stock market, which is influenced by the short term behavior of long term interest rates. If you are an investor, you will be less concerned with financial markets and more concerned with those company fundamentals that determine earnings growth.
Consider another form of our earlier equation of the S&P 500's market behavior:
S&P 500 = EPS * (capacity utilization - inflation )
t+1 t+1
_________________________
long term bond rates
Earnings Per Share are company determined; the terms to the right of that are cyclically determined. Investors will be concerned mainly with company developments that influence long term EPS growth. Traders will be concerned mainly with cyclically determined market variables. During the course of several business cycles, growth in EPS will usually dominate any unfavorable combination of cyclical terms.
Booch, Grady, "Object Oriented Analysis and Design," Menlo Park: Addison-Wesley; 1994 ed. Rationality in a complex world.
Fisher, Philip A., "Common Stocks and Uncommon Profits," New York: John Wiley & Sons; 1958, 1996 ed. What makes growth companies.
Graham, Benjamin, " The Intelligent Investor, 4th ed.," New York: Harper & Row , 1973. Probably the best book on stock market investing ever written. Concerned mainly with company fundamentals and valuation.
Keynes, John Maynard, " The Collected Writings of John Maynard Keynes, Volume XII, Economic Articles and Correspondence, Investment and Editorial," Cambridge: University Press, 1983. The chronicles of Keynes as investor.