Market efficiency

Market efficiency

 The fair price function of the capital markets provides assurance that
investors can sell stock at the going price and not be taken to the
cleaners. A discussion of fair pricing inevitably leads to the efficient
market hypothesis (EMH), the theory supporting the nation that market
prices are in fact fair. The EMH is probably the single most important
paradigm in finance.
Like technical analysis, market efficiency is a
controversial part of finance. In an efficient market
security prices are based on the available
information so as to offer and expected return
consistent with their level of risk. While most
professors are convinced that the markets are
quite efficient and that free lunches are as scarce
as Ty Cobb baseball cards, some professional
money managers believe otherwise.
Capital market prices are presumed to be fair
because they are the equilibrium result of the
analyses of many people, each of whom is seeking
to increase personal wealth. When a listed stock is
put up for sale, hundreds of people can bid for it.
The markets ensure that the seller trades with the
highest bidder. Conversely, a buyer is confronted
with numerous potential sellers, and the system
ensures that the buyer’s order is matched up with
the best price, which from the buyer’s perspective
is the lowest price. The greater the number of
participants and the more formal the marketplace,
the more an investor is assured of a good (fair)
price.
The efficient market hypothesis

 To speak intelligently about the efficient market hypothesis a person

must under-stand what the hypothesis says and what it does not say.
Efficiency can be categorized by both type and degree.
Types of efficiency

There are two types of efficiencies

1. Operational Efficiency
2. Informational Efficiency
1. Operational efficiency

Operational efficiency is a measure of how well things function in terms of

speed of execution and accuracy. At a stock exchange, operational
efficiency is measured by such factors as the number of orders lost or filled
incorrectly and the elapsed time between the receipt of an order and its
execution. All market participants are concerned with these matters, but
the EMH does not refer to this type of efficiency.
2. Informational Efficiency

Informational efficiency is a measure of how quickly and accurately the

market1 reacts to new information. New data constantly enter the
marketplace via economic reports, company announcements, political
statements, or public opinion surveys, to name a few sources.
 What does all this information mean?
 Is raising unemployment in the United Kingdom good or bad for holders
of U.S. Treasury bonds?
 How about a company’s announcement that it intends to split its stock
five for one?
 Suppose the price of gold jumps $10 an ounce in one day, what effect, if
any, is this event likely to have on stock prices?
• We know security prices adjust rapidly and
accurately to the news without the need to
digest it very long. Sometimes the speed of
adjustment is remarkably fast. For instance, the
author was once sitting in a brokerage firm
punching up his current stock on the Quotron
machine. One of his holding was common stock
in MGM Grand Hotel. The stock was trading at
$101⁄4. At that very moment, across the room,
the bell rang and the red light flashed on the
Dow Jones News Service monitor, indicating hot
news
• The stock was trading at $101⁄4. At that very
moment, across the room, the bell rang and the
red light flashed on the Dow Jones News Service
monitor, indicating hot news. The headline read,
“Fire at the MGM Grand Hotel”. In the seconds it
took the author to walk from the service monitor
back to the Quotron machine, the stock fell to
$71⁄2, which is approximately where it remained
the rest of the day.
• One need not be a Mensa member to realize
that a hotel fire is bad news. In an
informationally efficient market, prices are going
to react fast, just as they did in the MGM
situation. An investor cannot expected to read
about the fire in The Wall Street journal the
following day and think, “Well, I’ll bet that
hammers the stock; I’d better sell,” and then
expect to find that the market is still trying to
sort out the news. Prices would have dropped
long ago.
• Because the market is efficient, the meaning of
the news is discovered quickly, and prices
adjust. Students in an investments course are
sometimes disappointed to learn that simply
taking a stock market course does not ordain
them with the power to read the financial pages
and fluently pick stocks that will double in price
by next week. Things do not work that way.
• Still, the market is not completely efficient. It
still rewards people who process the news better
than the next person. For one thing, not
everyone has access to the same news, nor
does everyone receive the news in a timely
fashion. Because of this discrepancy, market
participants commonly talk about three forms of
the EMH, each of which is based on the
availability of a different level of information.
EMH in Finance

The efficient market hypothesis is one of the most important paradigms in

[Link] efficient market hypothesis deals with informational efficiency,
which is a measure of how quickly and accurately the market digests new
information. It is well established that the market is informationally
efficient.
Degree of informational Efficiency

1. Weak form Efficiency:

The least restrictive form of the EMH is weak form efficiency, which states
that future stock prices cannot be predicted by analyzing price from the
past. In other words, charts are of no use in predicting future prices.
 According to the weak form of the EMH, how a stock arrived at its
current price is irrelevant. It could have followed the route of Stock A, or
it could have behaved like Stock B. The only thing that matters is the
current price. Any information contained in the past price series is
already included in the current price.
 The realization is a difficult pill for most people to swallow. A
survey of a variety of people would reveal that virtually
everyone would identify stock B as clearly a better buy than
stock A. After all, B is “rising” while A is “falling”. Who would
want to buy a declining stock?
 The point that is missed in this logic was made earlier: past prices do
not matter; future ones do. Everyone has access to past price
information2. According to the EMH, so many people are looking at
these same numbers that any “free lunches” have already been
consumed. The current price is a fair one that takes into account any
information contained in the past price data.
 Human nature is prone to extrapolate the past into the future. Business
Week conducted a poll3 in late 1999 asking investors their views on the
stock market. Fifty-eight percent indicated they believed the stock
market was “very” or “somewhat” overpriced, but 52 percent of the
respondents believed that stocks would be higher a year later.
Respondents aged 18 to 24 were most bullish, with 63 percent
predicting the market would be higher in 2000. Oddly, though, 67
percent of this same group predicted a market crash in the coming year.
 Here a stock has penetrated its support level at 0, resulting in
a significant decline to the -5 area. A technical analyst would
call this a breakout on the downside. Chard D shows
congestion in the -2 to +1 range, followed by a sharp break to
a new equilibrium level around -4.
What do these patterns mean? Would an investor be more inclined to buy
one of these stocks than the others? Is one clearly inferior to the others?
Actually, each of these figures was created using the random number
generating function of Lotus 1-2-3. These are four successive Lotus graph;
each graph has a different seed number to start its series. In all four graph,
each observation is either one unit greater than the previous observation or
one unit smaller, and each of the two possible out-comes had a 50 percent
probability of occurring. Are these graphs useful in predicting what Lotus
will select next? Probably not.
Tests of Weak Form Efficiency
Tests if the weak form of market efficiency takes one of two forms:
autocorrelation tests and filter rule [Link] autocorrelation test
investigates whether security returns are related through time. That is, do
patterns develop that provide information about the future?
 Suppose we took 50 pennies lined them up, and turned them such that
20 pennies showed heads (or up stock prices) and 30 showed tails
(down stock prices). We could achieve this result in a variety of ways.
We could have all the heads first, then all the tails. We could also
alternate the heads and tails until the heads run out. Neither of these
patterns would be expected to occur by chance.
 Through the use of a handy nonparametric statistical technique called a
runs test, analysts can test the likelihood that such a series of price
movements occurred by chance. A run in an uninterrupted sequence of
the same observation.
 The runs test calculates the number of ways the observed
number of runs could occur given the relative number of
heads and tails in the sample, and the probability of this
number occurring.
 Probability principles indicate that about 95 percent of the area under
the normal curve lies within 1.96 standard deviations of the mean. The
test indicates a 27.56 percent chance of getting 23 runs when n1 = 20
and n2 = 30. With a Z statistic less than ⎢1.96⎥, we cannot, at a 95
percent confidence level, reject the null hypothesis that the observed
stock price sequence was determined by a random process.
 Numerous autocorrelation tests are reported in the finance literature.
The classic study is an exhaustive one by Eugene Fama. Fama is
credited as the originator of the entire notion of market efficiency. An
occasional study finds marginal statistical significance in return patterns,
but invariably the pattern is economically insignificance, meaning that
when one includes trading fees (commissions), the past data has no
value.
 A filter rule is a trading rule that involves buying shares after they rise in value by x
percent. When they fall x percent from the subsequent high, sell them, go short,
and cover the short when they rise x percent from a subsequent low. Because
anyone can calculate these realized percentages, filter rules should not work if the
markets are weak form efficient.
Semi Strong Form Efficiency

The weak form of the EMH states that security prices fully reflect any
information contained in the past series of stock prices. Semi-strong form
efficiency takes the information set s step further and includes all publicly
available information. The semi- strong form of the EMH states that security
prices fully reflect all relevant publicly available information.
 A plethora of information holds potential interest to investors. In
addition to past stock prices, economic reports, brokerage firm
recommendations, investment advisory letters, and so on all contain a
myriad of details about what affects business performance and stock
value. While no one sees every one of these items, the market does, and
prices move as people make decisions to buy and sell based on what
they learn from the information set available to them.
Tests of Semi-strong Form Efficiency
Extensive academic research supports the semi-strong version of the
efficient market hypothesis. The literature devotes much more attention
to tests of semi-strong form efficiency than to weak form tests. Studies
have investigated the extent to which people can profit by acting on
various corporate announcements such as stock splits, cash dividends,
and stock dividends. While and occasional research paper shows that
small profits could have been made in a particular case, the general
result is consistent: The market reacts to public information efficiency,
and investors will seldom outperform the market averages by analyzing
public news, especially if they must pay commissions to buy and sell.
Strong Form Efficiency

The most extreme version of the EMH is strong form efficiency. This version
states that security prices fully reflect all public and private information. In
other words, even corporate insiders cannot make abnormal profits by
exploiting their private; inside information about their company. Inside
information is formally called material, nonpublic information.
Section 16 of the Securities Exchange Act of 1934 defines an insider as “an
officer or director of a public company, or an individual or entity owning 10
percent or more of any class of a company’s shares.” The law requires
insiders to report their holdings of corporate securities within 10 days of
becoming an insider. They must also report subsequent transactions in
these securities for themselves or a member of their family by the tenth
day of the month following the trade.
The evidence does not support this form of the EMH. Insiders can make a
profit on their knowledge, and every year people go to jail, get fined, or get
suspended from trading for doing so. Inside information gives an unfair
advantage that can be used to extract millions of dollars out of the
marketplace. Where did these millions in profit come from? They came from
the pockets of individual investor who did not have access to the
confidential corporate news. Society does not feel this advantage is fair;
consequently, insider trading is illegal.
For seven issues of the magazine over this period, stocks mentioned in the
article rose an average of 11.54 percent compared to an average rise of
0.12 percent in the Standard & Poor’s 500 stock index. The large Thursday
rise and increased trading volume was compelling evidence that someone
was trading ahead of the public distribution of the magazine. This act was
illegal trading on inside information.
The Semi efficient market hypothesis

The essence of the semi-efficient market hypothesis (SEMH), a cousin to

the EMH, is the notion that some stocks are priced more efficiently than
others. This idea is appealing to many market analysts. Consider two very
different companies such as IBM and a hypothetical start-up firm called
Triple-Scan Video. Everyone has heard of international Business Machines,
which trades on the New York Stock Exchange and many regional
exchanges. Thousands of portfolios contain its shares, and virtually all
security analysts watch it. The likelihood of realizing an unusual gain in the
shares of IBM is extremely small. The stock is priced fairly, and investors
who buy some will likely earn a long-term return consistent with the stock’s
level of risk.
Security prices & random walks

The efficient market hypothesis states that the current stock price fully
reflects relevant news information. While some of the news is expected,
much of it is unexpected. The unexpected portion of the news, by
definition, arrives randomly – the essence of the notion that security prices
follow a random walk because of the random nature of the news. Some
days the news is good, some days it is bad. Specifics of the news cannot be
predicted with great accuracy.
Anomalies

This section reviews several important market anomalies that financial

researchers actively explore. In finance, the term anomaly refers to
unexplained results that deviate from those expected under finance theory,
especially those related to the efficient market hypothesis. Familiar
anomalies include the low PE effect, the small firm effect, the neglected
firm effect, the January effect, and the overreaction effect.