Damodaran - Aswath. .Investment - Fables
Damodaran - Aswath. .Investment - Fables
Damodaran - Aswath. .Investment - Fables
Data from Federal Reserve. Each point represents a year and the stock return in that year is plotted
against the treasury bond rate at the start of the year.
The relationship is murky, at best. In 1981, for instance, the treasury bond rate at the start of
the year was 14% but stocks did very well during the year, earning 15%. In 1961, the
treasury bond rate at the beginning of the year was 2% and stocks dropped 11% during the
year. There is little evidence of a link between the treasury bond rate at the start of a period
and stock returns during that period.
This link between treasury bond rates and stock returns may become stronger if you
consider how much you can earn as a return on stocks. You could define this return
narrowly as the dividend yield (dividends/current stock prices) on the market or use a much
396
broader measure, such as earnings yield, which looks at the overall earnings on the market
as a percent of the current level of the index. The earnings yield is the inverse of the price
earnings ratio and is used widely by market strategists as a measure of how equities are
priced relative to their earnings. Rather than focus on the level of the treasury bond rate,
some market strategists often look at the difference between earnings yields and the treasury
bond rate. They believe that it is best to invest in stocks when earnings yields are high,
relative to the treasury bond rate. To examine this proposition, the difference between the
earnings yield and the T.Bond rate at the end of every year from 1960 to 2000 was
estimated and compared to the returns on the S&P 500 in the following year (see Table
14.2)
Table 14.2: Earnings Yield, T.Bond Rates and Stock Returns: 1960 -2001
Stock Returns
Earnings yield - T.Bond
Rate
Number of
years Average
Standard
Deviation Maximum Minimum
> 2% 8 11.33% 16.89% 31.55% -11.81%
1 -2% 5 -0.38% 20.38% 18.89% -29.72%
0-1% 2 19.71% 0.79% 20.26% 19.15%
-1-0% 6 11.21% 12.93% 27.25% -11.36%
-2-1% 15 9.81% 17.33% 34.11% -17.37%
< -2% 5 3.04% 8.40% 12.40% -10.14%
When the earnings yield exceeds the treasury bond rate by more than 2%, which has
occurred in 8 out of the 41 years, the return on the S& P 500 in the following year has
averaged 11.33%. However, the returns are almost as good when the earnings yield has
lagged the treasury bond rate by zero to 1%. It is true that the annual returns are only 3.04%
in the five years following periods when the earnings yield was lower than the treasury bond
rate by more than 2%, but the annual returns were also negative in the 5 years when the
earnings yield exceeded the treasury bond rate by 1-2%. Thus, there seems to be little
historical support for using earnings yield and treasury bond rates to predict future stock
market movements.
Business Cycles
As with treasury bonds, there is an intuitive link between the level of stock prices
and economic growth. You would expect stocks to do much better in economic booms than
during recessions. What makes this relationship tricky, however, is that market movements
are based upon predictions of changes in economic activity in the future, rather than levels
397
of activity. In other words, you may see stock prices rising in the depths of a recession, if
investors expect the economy to begin recovering in the next few months. Alternatively, you
may see stock prices drop even in the midst of robust economic growth, if the growth does
not measure up to expectations. In Figure 14.7, the returns on the S&P 500 index and real
GDP growth are graphed going back to 1960:
Figure 14.7: GDP Growth and Stock Returns
Data from Federal Reserve. Each point represents a year and the stock return in that year is plotted
against GDP growth during the year
There is a positive relationship between GDP growth during a year and stock returns during
the year, but there is also a lot of noise in the relationship. For instance, the worst single
year of stock returns was 1931, when GDP dropped by about 7%. The very best year of
stock returns was 1954 but GDP declined slightly that year. The same dichotomy exists
during years of positive GDP growth; stock returns dropped in 1941 even though the
economy grew strongly that year but returns in 1995 were very positive as GDP grew about
4% that year. Even if the relationship were strong enough to pass muster, you cannot use it
for market timing unless you can forecast real economic growth. The real question then
becomes whether you can make forecasts of future stock market movements after observing
economic growth in the last year. To examine whether there is any potential payoff to
investing after observing economic growth in the prior year, the relationship between
398
economic growth in a year and stock returns in the following year, using data from 1929 to
2001 is looked at in Table 14.3:
Table 14.3: Real Economic Growth as a predictor of Stock Returns: 1960 - 2001
Stock Returns in Next Year
GDP Annual
Growth
Number of
years
Average
Return
Standard deviation in
returns
Best
Year
Worst
Year
>5% 23 10.84% 21.37% 46.74% -35.34%
3.5%-5% 22 14.60% 16.63% 52.56% -11.85%
2-3.5% 6 12.37% 13.95% 26.64% -8.81%
0-2% 5 19.43% 23.29% 43.72% -10.46%
<0% 16 9.94% 22.68% 49.98% -43.84%
All years 72 12.42% 19.50% 52.56% -43.84%
There seems to be no clearly discernible relationship between returns next year and GDP
growth this year. It is true that the years with negative GDP growth are followed by the
lowest stock returns, but the average stock returns in this scenario are barely higher than the
average returns you would have earned if you had bought after the best economic growth
years (growth exceeds 5%).
If you can forecast future growth in the economy, it can be useful at two levels. One
is in overall market timing, since you will steer more of your funds into stocks prior to
better-than-expected economic growth and away from stocks when you foresee the
economy slowing. You can also use the information to over invest in those sectors that are
most sensitive to the economic cycle - automobile and housing stocks, for instance - if you
believe that robust economic growth is around the corner.
Market Timers
While a variety of ways in which investors try to time markets have been looked at, a
more fundamental question has not been asked: Do those who claim to time markets
actually succeed? In this section, you will consider a broad range of investors who try to
time markets and examine whether they succeed.
Mutual Fund Managers
Most equity mutual funds do not lay claims to market timing, but they do try to time
markets at the margin by shifting their assets in and out of cash. You will begin by looking
at whether they succeed on average. There are some mutual funds that claim market timing
399
as their primary skill and these funds are called tactical asset allocation funds. You will look
at the track records of these funds and pass judgment on whether their claims hold up.
Overall Evidence
How do you know that mutual funds try to time markets? While all equity mutual
funds need to hold some cash - investments in treasuries and commercial paper - to meet
redemption needs and for day-to-day operations, they collectively hold much more cash
than is necessary. In fact, the only explanation for the cash balances that you observe at
equity mutual funds is that mutual funds use them to signal their views of future market
movements - they hold more cash when they are bearish and less cash when they are
bullish. In Figure 14.8 below, the average cash balance at mutual funds is presented, each
year from 1980 to 2001 and the returns on the S&P 500 each year.
Figure 14.8: Mutual Fund Cash Holdings and Stock Returns
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
14.00%
198119821983198419851986198719881989199019911992199319941995199619971998199920002001
Year
-20.00%
-10.00%
0.00%
10.00%
20.00%
30.00%
40.00%
Cash as % of assets at start of year Stock Returns
High cash holdings at mutual funds
In each year, the stock return in that year and the cash holdings at mutual funds at the end of the
year is shown.
Note that the cash balances seem to increase after bad years for the market and decrease
after good years, but there is little predictive power in the level of cash holdings. The
question of whether mutual funds are successful at market timing has been examined widely
in the literature going back four decades.
400
Other studies have looked at whether mutual funds succeed at shifting their money into
higher beta stocks
18
just before equity markets surge and at whether mutual funds earn
higher returns in years in which the market does well but have found little evidence of
market timing prowess on the part of mutual funds.
19
Tactical Asset Allocation and other Market timing Funds
In the aftermath of the crash of 1987, a number of mutual funds sprung up claiming
that they could have saved investors the losses from the crash by steering them out of equity
markets prior to the crash. These funds were called tactical asset allocation funds and made
no attempt to pick stocks. Instead, they argued that they could move funds between stocks,
treasury bonds and treasury bills in advance of major market movements and allow
investors to earn high returns. Since 1987, though, the returns delivered by these funds has
fallen well short of their promises. Figure 14.9 compares the returns on a dozen large
tactical asset allocation funds over 5-year and 10-year periods (1987-97) to both the overall
market and to fixed mixes - 50% in both stocks and bonds, and 75% stocks/25% bonds.
The last two are called couch potato mixes, reflecting the fact that you are making no attempt
to time the market.
18
See Treynor, Jack L., and Kay Mazuy, 1966, Can mutual funds outguess the market? Harvard Business
Review 44, 131-136.. They argued that if mutual funds have market timing skills, they should buy high
beta stocks just before up movements in the stock market, since these stocks should up go up even more.
Their conclusion was that mutual funds did the exact opposite moved into high beta stocks just before
market declines.
19
Henriksson, Roy D., and Robert C. Merton, 1981, On market timing and investment performance. II.
Statistical procedures for evaluating forecasting skills, Journal of Business, v54, 513-533.
401
Figure 14.9: Performance of Unsophisticated Strategies versus Asset Allocation Funds
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
14.00%
16.00%
18.00%
S & P 500 Couch Potato 50/50 Couch Potato 75/25 Asset Allocation
Type of Fund
1989-1998
1994-1998
The couch potato strategies represent fixed allocations (50/50 is always 50% stock and 50%
bonds). The average across asset allocation funds is compared to the couch potato strategies.
One critique of this study may be its focus on a few tactical asset allocation funds. In 1998,
an examination
20
of a much larger sample more than 100 asset allocation funds between
1990 and 1995 also found little evidence of success at market timing at these funds.
Investment Newsletters
There are hundreds of investment newsletters that investors subscribe to for sage
advice on investing. Some of these investment newsletters are centered on suggesting
individual stocks for investors but some are directed towards timing the market. For a few
hundred dollars, you are told, you too can be privy to private signals of market movements.
An analysis
21
of the market timing abilities of investment newsletters examined the
stock/cash mixes recommended in 237 newsletters from 1980 to 1992. If investment
newsletters are good market timers, you should expect to see the proportion allocated to
20
Beckers, C., W. Ferson D. Myers, and M. Schill, 1999, Conditional Market Timing with Benchmark
Investors, Journal of Financial Economics 52, 119-148.
21
Graham, John R., and Campbell R. Harvey, 1996, Market timing ability and volatility implied in
investment newsletters asset allocation recommendations, Journal of Financial Economics 42, 397-421.
402
stocks increase prior to the stock market going up. When the returns earned on the mixes
recommended in these newsletters is compared to a buy and hold strategy, 183 or the 237
newsletters (77%) delivered lower returns than the buy and hold strategy. One measure of
the ineffectuality of the market timing recommendations of these investment newsletters lies
in the fact that while equity weights increased 58% of the time before market upturns, they
also increased by 53% before market downturns. There is some evidence of continuity in
performance, but the evidence is much stronger for negative performance than for positive.
In other words, investment newsletters that give bad advice on market timing are more likely
to continue to give bad advice than are newsletters that gave good advice to continue giving
good advice.
22
The only hopeful evidence on market timing comes from a study of professional
market timers who are investment advisors. These timers provide explicit timing
recommendations only to their clients, who then adjust their portfolios accordingly - shifting
money into stocks if they are bullish and out of stocks if they are bearish. An examination
of the timing calls made by 30 professional market timers who were monitored by
MoniResearch Corporation, a service that monitors the performance of such advisors, finds
some evidence of market timing ability.
23
Note, though, that the timing calls were both short
term and frequent. One market timer had a total of 303 timing signals between 1989 and
1994, and there were, on average, about 15 signals per year across all 30 market timers.
Notwithstanding the high transactions costs associated with following these timing signals,
following their recommendations would have generated excess returns for investors.
24
Market Strategists
The market strategists at major investment banks represent perhaps the most visible
symbols of market timing. Their prognostications about the market are widely disseminated
not only by their investment banks but also by the media. Abby Cohen (Goldman Sachs),
Doug Cliggott (Morgan Chase) and Byron Wien (Morgan Stanley) are all widely known.
While much of what market strategists say about markets cannot be easily categorized as
22
A good market timing newsletter is likely to repeat its success about 50% of the time. A poor market
timing newsletter has a 70% chance of repeating its poor performance.
23
Chance, D. M., and M.L. Hemler, 2001, The performance of professional market timers: Daily
evidence from executed strategies, Journal of Financial Economics, v62, 377-411.
24
The study looked at excess returns after transactions costs but before taxes. By its very nature, this
strategy is likely to generate large tax bills, since almost all of your gains will be taxed at the ordinary tax
rate.
403
bullish or bearish- good market strategists are difficult to pin down when it comes to
explicit forecasts - they also make specific recommendations on preferred asset allocation
mixes that are presented in the Wall Street Journal. Table 14.4 provides the asset allocation
mixes recommended by major investment banks in June 2002.
Table 14.4: Asset Allocation Mixes - Investment Bank Strategists
Firm Strategist Stocks Bonds Cash
A.G. Edwards Mark Keller 65% 20% 15%
Banc of America Tom McManus 55% 40% 5%
Bear Stearns & Co. Liz MacKay 65% 30% 5%
CIBC World Markets Subodh Kumar 75% 20% 2%
Credit Suisse Tom Galvin 70% 20% 10%
Goldman Sach & Co. Abby Joseph Cohen 75% 22% 0%
J.P. Morgan Douglas Cliggott 50% 25% 25%
Legg Mason Richard Cripps 60% 40% 0%
Lehman Brothers Jeffrey Applegate 80% 10% 10%
Merrill Lynch & Co. Richard Bernstein 50% 30% 20%
Morgan Stanley Steve Galbraith 70% 25% 5%
Prudential Edward Yardeni 70% 30% 0%
Raymond James Jeffrey Saut 65% 15% 10%
Salomon Smith John Manley 75% 20% 5%
UBS Warburg Edward Kerschner 80% 20% 0%
Wachovia Rod Smyth 75% 15% 0%
How do these allocation mixes yield market predictions? One way is to look at the
percent allocated to stocks. More bullish market strategists will recommend a larger
proportion of the portfolio be invested in stocks, whereas bearish strategists will overweight
cash and bonds. The other is to look at changes in holdings recommended by the same
strategist from period to period - an increase in the proportion allocated to stocks would
indicate more bullishness. On both dimensions, the market timing skills of strategists are
questionable. The Wall Street Journal, in addition to reporting the asset allocation mixes of
strategists also compares the returns that would have been generated by following each
bank`s allocation advice to the returns you would have made by being fully invested in
stocks over 1-year, 5-year and 10-year periods. To counter the argument that it is unfair to
compare a 100% equity portfolio to a asset allocation mix, the Journal also reports on the
returns on a robot mix - a fixed allocation across stocks, bonds and bills. Figure 14.10
summarizes the returns on all three, as well as the returns you would have earned by
404
following the strategist who had the best mixes over the period and the one with the worst
mixes:
Figure 14.10: Annual Return from Market Strategists' Mixes: 1992-2001
0.00%
2.00%
4.00%
6.00%
8.00%
10.00%
12.00%
14.00%
16.00%
Best Market Strategist Average Market
Strategists
Worst Market Strategist Robot Blend 100% Equity
Best strategist is one with highest
returns to asset allocation advice
Returns to fixed blend
set by Wall Street
Journal
Data from Wall Street Journal. These are the annual returns you would have made between 1992
and 2001 following the asset allocation advice offered by market strategists at major investment
banks.
Note that the returns on the robot mix are higher than the average returns generated by
following the average market strategists. Of the 16 banks that the Wall Street Journal
tracks, only five would have generated returns higher than the robot mix over the period and
even those would have well within a statistical margin for error. Finally, even the best
strategist`s asset mix would have underperformed a strategy of being fully invested in
stocks. Overall, the evidence indicates that the market timing skills of leading market
strategies are vastly overstated.
The Rest of the Story
The evidence on stock market timing is decidedly mixed. While some timing
indicators seem to offer promise in predicting market direction, those who use them do not
earn excess returns. How do you explain this contradiction? In this section, you will look at
the reasons why a unshakeable faith in equity markets in the long term can be dangerous
and why market timing indicators do not pay off for most investors.
405
Stocks are not riskless in the long term
In bear markets, you do not have to spend much time convincing investors that
investing in stocks is risky but a prolonged and strong bull market often leads these same
investors to the conclusion that equity is not risky, at least in the long term. Earlier in the
chapter, you examined some of the evidence, primarily from the U.S. market since 1926,
used to sustain this point of view. In this section, you will evaluate the evidence from other
equity markets in the world to see if it backs up the evidence in the U.S.
Survivor market bias
One of the problems with extrapolating the findings from the U.S. equity market in
the twentieth century is that the United States was perhaps the most successful economy
and market in the world in that century. In other words, you have a selection bias. To
provide an analogy with individual stocks, this would be the equivalent of picking the top ten
companies in the United States, in terms of market capitalization today, and examining
whether your would have made money investing in these companies. The answer, not
surprisingly, will be yes since these companies acquired their large market capitalization
status by being successful over long periods.
To provide some balance, therefore, you have to look at the returns investors in
equities would have earned in other (and less successful) equity markets. The most detailed
look at these returns estimated the returns you would have earned on 14 equity markets
between 1900 and 2001 and compared these returns with those you would have earned
investing in bonds.
25
Figure 14.11 presents the risk premiums - i.e., the additional returns -
earned by investing in equity over treasury bills and bonds over that period in each of the 14
markets:
25
Dimson, E., P. March and M. Staunton, 2002, Triumph of the Optimists, Princeton University Prsss.
406
0%
1%
2%
3%
4%
5%
6%
7%
8%
Country
Figure 14.11: Equity Risk Premiums - By Country
Stocks - Short term Government Return Stocks - Long Term Government Return
Data from Dimson et al. The differences in compounded annual returns between stocks and short
term governments/ long term governments is reported for each country.
While equity returns were higher than what you would have earned investing in government
bonds or bills in each of the countries examined, there are wide differences across countries.
If you had invested in Spain, for instance, you would have earned only 3% over government
bills and 2% over government bonds on an annual basis by investing in equities. In France,
in contrast, the corresponding numbers would have been 7.1% and 4.6%. Looking at 40 or
50 year periods, therefore, it is entirely possible that equity returns can lag bond or bill
returns, at least in some equity markets.
Equity investors therefore have to wonder whether the market they are investing in
currently will be one of the winner markets (like the US in the twentieth century) or a
lagging market (like the Japanese market since 1989). Since there is a probability that every
market (including the US equity market today) can be a lagging market over the next few
decades, you should be cautious about investing too much in equities in any particular
market. You may be able to improve the odds by investing in a global equity fund but even
there, you can be exposed to risk over long time periods.
407
How long term is long term?
Financial experts and advisors who argue that stocks win in the long term are often
ambiguous about what they mean by the long term and investors often define long term in
very different ways- one year may represent long term for an impatient investor whereas 20
years may be long term to a patient investor.
Equities clearly are not riskless over one year, but are they close to riskless if you
have a 20-year time horizon? Not necessarily, for several reasons:
Even long time horizons can be shortened by unanticipated events. For example,
consider the advice given to a 35-year old about her pension fund investments. With 30
years left to retirement, she seems like a perfect candidate for a long-term investment
strategy. That is predicated though on the belief that she will stay healthy and continue
to work for that period. If she has to retire early due to health problems or loses her job,
she may find herself needing to draw on her pension fund savings far sooner.
Investors save over time and they save more in up markets and in their later years:
Assume that you are 35 years old and that you have 30 years until retirement. You will
be saving over time for your retirement and your contributions to your pension fund will
tend to get larger as you get older (and closer to retirement). In effect, this will reduce
the effective time horizon you have on your investments. In addition, you will tend to
save more and invest more in stocks in buoyant stock markets and less in depressed
markets. Given the history of the market, this will imply that you will be over invested in
stocks when they are over valued and under invested when stocks are a good bargain.
Even the most optimistic assessment of the historical data on stock returns can only
lead to the conclusion that while there is a high probability that stocks will earn higher
returns than less risky alternatives over long time periods, there is no guarantee. In
fact, a more realistic evaluation of stock market history, in the U.S. and elsewhere,
suggests that the probability that equities will under perform government bonds over
longer period is too large to be ignored by investors. Even a 5% probability that stocks
will under perform bonds over the long term may be sufficient to induce more risk
averse investors to invest more in bonds and less in stocks.
The perils of investing in equities even with a long time horizon are illustrated when you
look at the Japanese equity market over the last 15 years. An investor who invested his
wealth in the Nikkei in 1989 when the index peaked at close to 40000 would have lost 80%
of his investment by 2003 and is extremely unlikely to recover his losses whole in his
lifetime.
408
Market Timing Indicators and Success
Why do market timers succeed so infrequently if there are so many market timing
indicators that make money, at least on paper? In this section, you will consider some of the
dangers involved with trying to time markets and with following the advice of market gurus.
Hindsight Bias
Market timing always seems simple when you look back in time. After the fact, you
can always find obvious signals of market reversals - bull markets turning to bear markets
or vice versa. Thus, in 2001, there were investors who looked back at 1999 and bemoaned
the fact that they missed getting out of stocks when the market topped at the end of that
year. At that time, though, the signs were not so obvious. There were analysts who argued
that the market was overvalued and indicators that supported that point of view, but there
were just as many analysts, if not more, who saw the market continuing to rise and had
supporting models.
In practice, there is almost never a consensus among investors on whether markets
have hit bottom or peaked at the time that it occurs. It is an interesting fact that optimism
about the future is greatest just as markets top out and the market mood is darkest just as
markets turn around. To succeed at market timing, you cannot wait until a bottom has been
established before buying or for a market top before selling. If you do, you will miss much
of the subsequent payoff.
Timing of Information
If you are considering timing the market using macroeconomic variables such as
inflation or economic growth, you should also take into account the time lag before you will
get this information. Consider, for instance, research that shows that stock prices tend to go
up after quarters of high GDP growth. An obvious strategy would be to buy stocks after a
quarter of high GDP growth and sell after a quarter of negative or low GDP growth. The
problem with the strategy is that the information on GDP growth will not be available to you
until you are two months into the next quarter.
If you use a market variable such as the level of interest rates to make your market
forecasts, you are in better shape since this information should be available to you
contemporaneously with the stock market. In building these models, you should be careful
and ensure that you are not building a model where you will have to forecast interest rates in
order to forecast the stock market. To test for a link between the level of interest rates and
stock market movements, you would look at the relationship between interest rates at the
beginning of each year and stock returns over the year. Since you can observe the former
before you make your investment decision, you would have the basis for a viable strategy if
409
you find a correlation between the two. If you had run the test between the level of interest
rates at the end of each year and stock returns during the year, implementing an investment
strategy even if you find a correlation would be problematic since you would have to
forecast the level of interest rates first.
Noise in Forecast
As the evidence in the last section should make clearly, no market-timing indicator is
perfect or even close to perfect. In fact, the best market timers are right perhaps 60 to 65%
of the time, and even then, only about market direction and not magnitude. In other words, a
specific indicator, be it the returns in January or the level of interest rates, may give you
some indication of whether the market is more likely to go up or down over the rest of the
year but not by how much.
Both of these characteristics of market timing indicators - the significant proportion
of the time that they are wrong in calling market direction and their lack of success at
forecasting the size of the market movement - restrict the investment strategies that you can
use to time markets. Derivatives such as stock index futures and options, which would
generate the highest returns, have to be avoided because the risk of being wrong is too large.
Lack of Consistency
Market timers are the meteors of the investment universe. While they attract a great deal
of attention when they shine, they fade quickly. Looking at the high profile market timers
(Market Gurus) over time, from Jesse Livermore in the early part of this century to Ralph
Acampora, Prudential`s flamboyant market strategist, in the 1990s, you find a diverse
group.
26
Some were chartists, some used fundamentals and some were mysterious about
their methods, but there are three common characteristics that they seem to share:
1. A capacity to see the world in black and white: Market gurus do not prevaricate. Instead,
they make bold statements that seem outrageous when they make them about where the
market will be 6 months or a year from now. Acampora, for instance, made his
reputation with his call that the Dow would hit 7000 when it was at 3500.
2. A correct call on a big market move: All market timers make their reputation by calling
at least one big market move. For Livermore, it was the market crash of 1929 and for
Acampora, it was the bull market of the 1990s.
3. Outside personalities: Market gurus are born show persons, who use the media of their
time as megaphones to publicize not only their market forecasts but the news of their
26
One of the best books on Livermore is the classic Reminiscences of a Stock Market Operator by
Edwin LeFevre, John Wiley and Sons.
410
successes. In fact, part of their success can be attributed to their capacity to make other
investors act on their predictions, making these predictions, at least in the near term, self-
fulfilling prophecies.
So why do great market gurus stumble? The very same factors that contribute to their
success seem to underlie their failures. The absolute conviction they have in their market
timing abilities and their success at timing markets seems to feed into more outrageous calls
that ultimately destroy their reputations. Joe Granville, one of the market gurus of the late
1970s, for instance, spent all of the eighties recommending that people sell stocks and buy
gold and his newsletter was ranked the worst, in terms of performance, for the decade.
The Cost of Market Timing: Transactions and Opportunity Costs
If market timing were costless, you could argue that everyone should try to time
markets, given the huge returns to getting it right. There are, however, significant costs
associated with trying to time markets (and getting it wrong):
In the process of switching from stocks to cash and back, you may miss the best
years of the market. An article titled 'The Folly of Stock Market Timing,
examined the effects of annually switching from stock to cash and back from 1926
to 1982 and concluded that the potential downside vastly exceeds the potential
upside.
27
, In an analysis of market timing, Bill Sharpe suggested that unless you
can tell a good year from a bad year 7 times out of 10, you should not try market
timing.
28
This result is confirmed by Monte Carlo simulations on the Canadian
market, which show that you have to be right 70-80% of the time to break even from
market timing.
29
This research does not consider the additional transactions costs that inevitably flow
from market timing strategies, since you will trade far more extensively if you follow
them. In its most extreme version, a stock/cash switching strategy will mean that you
will have to liquidate your entire equity portfolio if you decide to switch into cash
and start from scratch again the next time you want to be in stocks.
27
Jeffrey, R., 1984, The Folly of Stock Market Timing, Financial Analysts Journal (July-August), 102-
110.
28
Sharpe, W. F., 1975, Are Gains Likely From Market Timing, Financial Analysts Journal, vol. 31, no. 2
(March/April): 60-69.
29
Chua, J. H., R.S. Woodward, and E.C. To. 1987, Potential Gains From Stock Market Timing in
Canada, Financial Analysts Journal (September/October), vol. 43, no. 5, 50-56.
411
A market timing strategy will also increase your potential tax liabilities. To see why,
assume that you have a strategy of selling your stocks after two good years in the
market, based upon the empirical findings that a bad year is more likely to follow.
You will have to pay capital gains taxes when you sell your stocks, and over your
lifetime as an investor, you will pay far more in taxes.
Lessons for Investors
Trying to time markets is a much more daunting task than picking stocks. All
investors try to time markets and very few seem to succeed consistently. If, notwithstanding
this history of failure, you decide to time markets, you should try to do the following:
1. Assess your time horizon: Some market timing indicators such as those based upon
charting patterns and trading volume try to forecast market movements in the short
term, whereas others such as using a normalized PE ratio to predict stock prices are
long term strategies. You need to have a clear sense of your time horizon before you
pick a market timing strategy. In making this judgment, you will need to look not
only at your willingness (or lack thereof) to wait for a payoff but also at how
dependent you are on the cash flows from your portfolio to meet your living needs;
if your job is insecure and your income is volatile, your time horizon will shrink.
2. Examine the evidence: The proponents of every market timing strategy will claim
that the strategy works and present you with empirical evidence of the incredible
returns you could have made from following it. You should consider all the caveats
from the last section when you look at the evidence including:
a. Is the strategy being fit back into the same data from which it was extracted?
You should be suspicious of elaborate trading strategies which seem to have
no economic basis or rationale - buy small cap stocks with price momentum
at 3 pm every Thursday and sell at 1 pm the next day, for instance. Odds are
that thousands of strategies were tested out on a large database and this one
emerged. A good test will look at returns in a different time period (called a
holdout period).
b. Is the strategy realistic? Some strategies look exceptionally good as
constructed but may not be viable since the information that they are based
on would not have been available at the time you would have had to trade.
You may find, for instance, that you can make money (at least on paper) if
you buy stocks at the end of every month where investors put more money
into mutual funds than they take out. The problem, though, is that this
412
information will not be available to you until you are well into the next
month.
c. Have execution costs and problems been considered? Many short-term
market timing strategies require constant trading. The trading costs and tax
liabilities created by this trading will be substantial and the returns, prior to
considering these costs, have to be substantially higher than a buy and hold
strategy for the strategy to make sense.
3. Integrate market timing with security selection: While many investors consider
market timing and security selection to be mutually exclusive, they don`t have to be.
You can and should integrate both into your overall strategy. You can, for instance,
use a volume indicator to decide when and whether to get into equities, and then
invest in stocks with low PE ratios because you believe these stocks are more likely
to be under valued.
Conclusion
If you can time markets, you can make immense returns and it is this potential
payoff that makes all investors into market timers. Some investors explicitly try to time
markets using technical and fundamental indicators, whereas others integrate their market
views into their asset allocation decisions, shifting more money into stocks when they are
bullish on stocks. Looking at the evidence, though, there are no market timing indicators that
deliver consistent and solid returns. In fact, there is little proof that the experts at market
timing - market strategists, mutual funds and investment newsletters, for example - succeed
at the endeavor.
Notwithstanding this depressing evidence, investors will continue to time markets. If
you choose to do so, you should pick a market timing strategy that is consistent with your
time horizon, evaluate the evidence on its success carefully and try to combine it with an
effective stock selection strategy.
413
CHAPTER 15
TEN LESSONS FOR INVESTORS
While the investment stories examined in this book reflect very different investment
philosophies and are designed for a wide range of investors, there are some lessons that can
be drawn by looking across the stories. In this chapter, you will see a number of
propositions about investing that apply across investment strategies. Hopefully, these broad
propositions about investing will stand you in good stead when you are subjected to the
next big investment story by an over eager sales person.
Lesson 1: The more things change, the more they stay the same
Each of the investment stories listed in this book has been around for as long as
there have been financial markets. Notwithstanding this reality, investment advisors
rediscover these stories at regular intervals and present them as their own. To provide a
faade of novelty, they often give these stories new and fancy names (preferably Greek).
Calling a strategy of buying low PE stocks the Omega or the Alpha strategy seems to do
wonders for its curb appeal to investors. In addition, as more and more data on stocks
becomes available to investors, some have become more creative in how they use this data to
find stocks. In fact, the ease with which you can screen stocks for multiple criteria - low PE,
high growth and momentum - has allowed some to create composite screens that they can
then label as unique.
Proposition 1: Be wary of complex investment strategies with fancy names that claim to be
new and different.
Lesson 2: If you want guarantees, don`t invest in stocks
No matter what the proponents of an investment strategy tell you, there are no stock
strategies that can offer guaranteed success. Stocks are volatile and are driven by hundreds
of different variables, some related to the overall economy and some arising as a result of
information that has come out about the firm. Even the most elaborate and best planned
strategies for making money in stocks can be derailed by unexpected events.
Proposition 2: The only predictable thing about stocks is their unpredictability.
Lesson 3: No pain, no gain
It is perhaps the oldest lesson in investments that you cannot expect to earn high
returns without taking risk, but it is a lesson that is often ignored. Every investment strategy
exposes you to risk, and a high return strategy cannot be low risk. If you are an investor
414
who is uncomfortable with large risk exposures, you should avoid any high risk strategy, no
matter how promising it looks on paper. Why are some investors so willing to delude
themselves into thinking that they can earn high returns without taking much risk? One
reason may be that the risk in some strategies is hidden and shows up sporadically. These
strategies succeed most of the time and deliver solid and modest returns when they do, but
create large losses when they fail.
Proposition 3: If you cannot see the risk in a high returns strategy, you just have not
looked hard enough.
Lesson 4: Remember the fundamentals
The value of a business has always been a function of its capacity to generate
cashflows from its assets, to grow these cashflow over time and the uncertainty associated
with these cashflows. In every bull market, investors forget the fundamentals that determine
value - cashflows, expected growth and risk - and look for new paradigms to explain why
stocks are priced they way they are. This was the case in the technology boom of the late
1990s. Faced with stratospheric prices for new economy companies that could not be
explained by conventional approaches, investors turned to dubious models, where growth in
revenues substituted for earnings and cashflows did not matter. In the aftermath of every
bull market, investors discover the truth that the fundamentals do matter and that companies
have to earn money and grow these earnings to be valuable.
Proposition 4: Ignore fundamentals at your own peril.
Lesson 5: Most stocks that look cheap are cheap for a reason
In every investment story in this book, there is a group of companies that are
identified as cheap. Early in this book, for instance, companies were categorized as cheap
because they traded at low multiples of earnings or below book value. At the risk of
sounding like professional naysayers, it should be noted that most of these companies only
looked cheap. There was generally at least one good reason, and in many cases more than
one, why these stocks traded at low prices. You saw, for instance, that many stocks that
traded at below book value did so because of their poor earning power and high risk and
that stocks that traded at low PE ratios did so because of anemic growth prospects.
Proposition 5: Cheap companies are not always good bargains.
Lesson 6: Everything has a price
Investors are constantly on the look out for characteristics that they believe make the
companies they invest in special - superior management, brand name, high earnings growth
415
and a great product all come to mind. Without contesting the fact that these are good
characteristics for a firm to possess, you have to still recognize that markets generally do a
good job of pricing in these advantages. Companies with powerful brand names trade at
high multiples of earnings, as do companies with higher expected growth. Thus, the
question that you have to answer as an investor is not whether having a strong brand name
makes your company more valuable, but whether the price attached to the brand name by the
market is too high or too low.
Proposition 6: Good companies may not be good investments.
Lesson 7: Numbers can be deceptive
For those investors who are tired of anecdotal evidence and investment stories,
numbers offer comfort because they provide the illusion of objectivity. A study that shows
that stocks with high dividends would have earned you 4% more than the market over the
last five years is given more weight than a story about how much money you could have
made investing in one stock five years ago. While it is sensible to test strategies using large
amounts of data over long periods, a couple of caveats are in order:
Studies, no matter how detailed and long term, generate probabilistic rather than
certain conclusions. For instance, you may conclude after looking at high dividend
paying stocks over the last five years that there is a 90% probability that high
dividend stocks generate higher returns than low dividend stocks, but you will not be
able to guarantee this outcome.
Every study also suffers from the problem that markets change over time. No two
periods are exactly identical and it is possible that the next period may deliver
surprises that you have never seen before and that these surprises can cause time-
tested strategies to fall apart.
Proposition 7: Numbers can lie.
Lesson 8: Respect the market
Every investment strategy is a bet against the market. You are not only making a
wager that you are right and the market is wrong but that the market will see the error of its
ways and come around to your way of thinking. Consider, for instance, a strategy of buying
stocks that trade at less than book value. You believe that these stocks are undervalued and
that the market is making a mistake in pricing these stocks. To make money, not only do
you have to be right about this underlying belief but markets have to see and correct their
mistakes. In the process, the prices of these stocks will be pushed up and you as an investor
will make money.
416
While you may be justified in your views about market mistakes, it is prudent to
begin with a healthy respect for markets. While markets do make large mistakes in pricing
stocks and these mistakes draw attention (usually after the fact), they do an extraordinary
job for the most part in bringing together investors with diverse views and information about
stocks and arriving at consensus prices. When you do uncover what looks like a market
mispricing and an investment opportunity, you should begin with the presumption that the
market price is right and that you have missed some critical component in your analysis. It
is only after you have rejected all of the possible alternative explanations for the mispricing
that you should consider trying to take advantage of the mispricing.
Proposition 8: Markets are more often right than wrong.
Lesson 9: Know yourself
No investment strategy, no matter how well thought our and designed it is, will work
for you as an investor, if it is not match your preferences and characteristics. A strategy of
buying stocks that pay high and sustainable dividends may be a wonderful strategy for risk
averse investors with long time horizons who do not pay much in taxes but not for investors
with shorter time horizons who pay high taxes. Before you decide to adopt any investment
strategy, you should consider whether it is the right strategy for you. Once you adopt it, you
should pass it through two tests:
a. The acid test: If you constantly worry about your portfolio and its movements keep
you awake at nights, you should consider it a signal that the strategy that you just
adopted is too risky for you.
b. The patience test: Many investment strategies are marketed as long term strategies.
If you adopt one of these strategies but you find yourself frequently second
guessing yourself and fine tuning your portfolio, your just may be too impatient to
carry this strategy to fruition.
In the long term, not much that is good -either physically or financially - comes out of
these mismatches.
Proposition 9: There is no one best investment strategy that fits all investors.
Lesson 10: Luck overwhelms skill (at least in the short term)
The most depressing lesson of financial markets is that virtues such as hard work,
patience and preparation do not always get rewarded. In the final analysis, whether you
make money or not on your portfolio is only partially under your control and luck can play
a dominant role. The most successful portfolio managers of last year, all too often, are not
the ones with the best investment strategies but those who (by chance) happened to be at the
417
right place at the right time. It is true that the longer you invest, the more likely it is that luck
will start to even out and that your true skills will show through; the most successful
portfolio managers of the last 10 years are less likely to get there because they were lucky.
As an investor, you should take both success and failure with a grain of salt. Neither
is a reflection of your prowess or lack thereof as an investor or the quality of your
underlying investment strategy. While you may not able to manufacture good luck, you
should be ready to take advantage of it when it presents itself.
Proposition 10: It pays to be lucky.
Conclusion
Beating the market is neither easy nor painless. In financial markets, human beings,
with all their frailties, collect and process information and make their best judgments on
what assets are worth. Not surprisingly, they make mistakes and even those who believe that
markets are efficient will concede this reality. The open question, though, is whether you can
take advantage of these mistakes and do better than the average investor. You can but only if
you do your homework, assess the weaknesses of your investment strategies and attempt to
protect yourself against them. If you have a short time horizon, you will also need luck as
an ally.