February 19 , 2003
The academic authors thought they had uncovered a device
to test whether stock markets are efficient.
Instead, they may have uncovered insights into the abuses of stock
analysts during our recent investment bubble.
Investment advisors use several categories to determine
the behavior of equities. Stocks
that show substantial sales and earnings growth over time are classified as
growth stocks (even if they stumble for short periods of time). Stocks that have low stock prices relative to book value are
classified as value stocks.
Since 1927, value stocks have outperformed growth stocks
by an annual rate of 3.7 percent a year.
We will assume that the difference is not caused by measurement errors
(always an issue in an academic study).
However, if markets rapidly absorb all new information
and have it reflected in stock values very soon after it is known (the
efficient market theory), then one class of stocks should not consistently
outperform another class of stock if they have the same characteristics.
Some finance professors have argued that the fact that
value stocks outperform growth stocks reflects the greater risk of value
stocks. For example, some so
called value stocks may be losing price because investors realize that capital
is being used poorly in that company. Indeed,
the appearance of value may be nothing more than a snapshot on the road to
bankruptcy for some companies.
Thus, the value stocks show better price gains over time
because they are riskier according to those professors.
Risk is an undesirable trait that requires higher compensation to
prevent investors from fleeing those stocks.
There is no easy way to determine what risk investors
believe exist in any stocks. However,
if a market already has incorporated all known information into its prices,
then forecasts of superior performance must mean expectations of currently
unknown outcomes. Thus, forecasts
of better stock price performance imply that more risk exists in those issues.
As I believe markets have too much emotion to perform
efficiently in the short run, I am not as willing to make the assumption that
forecasts of stronger price performance must mean more risk.
They may merely intensify the confidence in the management of
Professors Alon Brav of Duke, Roni Michaely of Cornell,
and Reuven Lehavy of the University of Michigan used two different groups of
forecasts to evaluate whether the risks implied in forecasts of future pricing
suggested that value stocks were riskier than growth stocks.
In neither case did they make a scientific sample of all investors.
Instead, they used the forecasts of analysts in the
Thomson First Call database from 1997 through 2001 for one measure.
They then used Value Line price projections over a 15 year period from
1987 to 2001.
In the first study, projections of growth stock
performance far outdistanced value stock forecasts.
The latter study showed little change between the two.
In neither case did analysts expect value stocks to outperform growth
stocks. (Remember, the fact is
that value stocks do outperform. Therefore,
the price difference is not caused by inherent risks between the two classes
The study concluded that as value stocks as a group were
no riskier than growth stocks but they outperformed growth stocks over long
periods of time, value stocks may be bargains.
Frankly, I believe the study shows that we normally
choose the hare and are shocked when the slow and steady tortoise wins.
Instead of reflecting all known information, the stock market reflects
human emotions in the short run. (I
am willing to assume the markets are efficient in the long run, but that is
However, the study suggests something else as well.
The authors believed that stock analysts might have been biased in
their forecasts because they were rewarded by equity banking business, which
flows more rapidly from growth stocks. That
is why they not only used First Call but also sought out Value Line.
The latter analysis is provided on a subscription basis
and is not subject to bias by banking business.
Over their time frame, the First Call analysts projected that strong
growth stocks would outperform high rated value stocks by more than 20
percentage points per year. No
similar discrepancy was observed for the possibly unbiased analysts of Value
This is not a definitive conclusion that banking business
biased analyst opinions, for the analysis was not done over comparable time
periods. However, it strongly
suggests that banking business might have distorted the work of analysts in
houses doing banking business during the formation of the stock market bubble.
In the meantime, do not shun value stocks because past performance might suggest that they are more risky than the growth stocks. Current studies suggest that any presumed higher risk is not justified.