1 COVER PAGE OF THE PROJECT WORK TITLE OF THE PROJECT EFFICIENCY HYPOTHESIS OF THE STOCK MARKET CASE STUDY: GHANA STOCK EXCHANGE A PROJECT REPORT UNDER THE GUIDANCE OF Mr Daniel Odei Tetteh SUBMITTET BY CAMARA MOHAMED BINTOU IN FULFILLEMENT OF THE REQUIREMENT FOR THE AWARD OF THE DEGREE OF MBA IN FINANCE SMU SIKKIM MANIPAL UNIVERSITY DIRECTORATE OF DISTANCE LEARNING FEBRUARY 2011
Efficiency Hypothesis of the Stock Exchange Project Presented by MR CAMARA MOHAMED BINTOU
Jul 28, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
COVER PAGE OF THE PROJECT WORK
TITLE OF THE PROJECT
EFFICIENCY HYPOTHESIS OF THE STOCK MARKET
CASE STUDY: GHANA STOCK EXCHANGE
A PROJECT REPORT UNDER THE GUIDANCE OF
Mr Daniel Odei Tetteh
SUBMITTET BY
CAMARA MOHAMED BINTOU
IN FULFILLEMENT OF THE REQUIREMENT FOR THE AWARD OF
THE DEGREE OF MBA
IN
FINANCE
SMU
SIKKIM MANIPAL UNIVERSITY
DIRECTORATE OF DISTANCE LEARNING
FEBRUARY 2011
2
RESEARCH TOPIC
EFFICIENCY HYPOTHESIS OF THE STOCK MARKET
CASE STUDY: GHANA STOCK EXCHANGE
CAMARA MOHAMED BINTOU
MBA-FINANCE IN SIKKIM MANIPAL UNIVERSITY
EMAIL: [email protected]
Tel (00233)541675942
3
ACKNOWLEDGEMENT
First and foremost, I would like to express my deepest sense of gratitude to the
Almighty ALLA H who has given me the opportunity, health and soundness of
mind to write this project work.
I am deeply indebted to my advisor Mr Daniel Odei Tetteh Whose expertise,
understanding, patience, motivation, enthusiasm, immense knowledge and his
constant support, added considerably to my experience. I appreciate his vast
knowledge and skills in many areas and without which this project would not be
possible. I could not have imagined a better advisor and mentor for this project.
I am heartily thankful to Madam Claudia……………. at the Ghana stock
exchange, whose encouragement, guidance and support from the initial to the final
level enable me to develop an understanding of the subject.
I owe my deepest gratitude to all my lecturers at Sikkim Manipal University and
all my colleagues.
Where would I be without my family and friends? To my father Elhadji KABA
CAMARA who put the fundament of my learning character, showing me the joy of
intellectual pursuit. To my mother BINTOU YATTASSAYE who raised me with
so much care and love. To all my brothers, sisters and friends for being supportive
financially and morally.
4
DEDICATION
I dedicate this project to my family and my future wife Miss BINTILY
KONATE
5
Bonafide Certificate:
BONAFIDE CERTIFICATE
Certified that this project report titled ―EFFICIENCY HYPOTHESIS OF THE
STOCK MARKET CASE STUDY GHANA STOCK EXCHANGE” is the
bonafide work of ―CAMARA MOHAMED BINTOU‖ who carried out the project
work under my supervision.
Gajendra Singh Manish Mittal
HEAD OF THE DEPARTMENT FACULTY IN CHARGE
Sikkim Manipal University of Health, Medical and Technological Sciences
Directorate of Distance Education
Ring Road opposite Busy Internet
www.smugh.edu.in
6
EXECUTIVE SUMMARY
The efficient market hypothesis is one of the most controversial market theories in
finance, which was developed in the 1960s by Eugene Fama. According to fama
―an efficient market is defined as a market where there are large numbers of
rational, profit-maximizers actively competing, with each trying to predict future
market values of individual securities, and where important current information is
almost freely available to all participants. In an efficient market competition
among the many intelligent participants leads to a situation where, at any point in
time, actual prices of individual securities already reflect the effects of information
based both on events that have already occurred and on events which, as of now,
the market expects to take in the future.
The purpose of this paper is to determine whether the GSE is weak form efficient
and if it is possible to predict the future of the security prices, using all kind of
information to increase returns on the portfolio and consequently to beat the
market.
As a result of this study, we found that the Ghana stock exchange is weak form
inefficient .Which means that the prices can be predicted and some smart investors
can earn abnormal profits. This inefficiency is due to factors like the volume of
traded stock, the size of the market, lack of adequate knowledge of the market
from the investors.
To resolve this inefficiency, the market has to be large and liquid, the information
must be available in term of accessibility, the cost of the transaction must be cheap
and there should an efficient system of settlement. There should also be a
campaign to sensitize people about the stock exchange. The system of trading
should also be innovative.
7
Tableau of content
TITLE PAGES
Acknowledgement………………………………………………………..III
Dedication………………………………………………………………..IV
Bonafide certificate……………………………………………………….V
Executive summary………………………………………………………V
Abstract ……………………………………………………………………1
Chapter 1…………………………………………………………………..2
Research questions…………………………………………………………2
Motivations………………………………………………………………….2
Justifications…………………………………………………………………3
Limitations of the study………………………………………………………4
Outline of the project…………………………………………………………4
Methodology…………………………………………………………………5
Background…………………………………………………………………..6
Previous studies……………………………………………………………..9
Performance of the GSE…………………………………………………….10
Importance of the stock exchange in the developing countries………………11
8
Political situations in Ghana …………………………………………….13
Chapter 2
History and development of the financial market in Ghana ………………14
Structure of the Ghana stock exchange……………………………………..16
Role of Ghana stock exchange in the Ghanaian economic ………………….17
Challenges facing the Ghana stock exchange…………………………………18
Chapter 3
Trading system…………………………………………………………………21
Clearing and settlement………………………………………………………..27
Legal and regulatory framework……………………………………………….28
Regulations affecting foreign and non-resident investors……………………….28
Listing on the Ghana stock exchange…………………………………………….29
Chapter 4
Literature review…………………………………………………………………33
History of efficient market hypothesis…………………………………………....33
Market conditions consistent to efficient market hypothesis……………………..41
Types of efficiency………………………………………………………………..42
Forms of efficiency………………………………………………………………42
Early Models……………………………………………………………………..49
9
Chapter 5
Technical analysis and efficient market hypothesis………………………………62
Fundamental analysis strategy for predicting the stock returns………………….67
Market anomalies……………………………………………………………….72
Behavioral finance and efficient market hypothesis……………………………82
Chapter 6
Data and research methodology……………………………………………….88
Reasons behind inefficiency……………………………………………………98
Solution for inefficiency………………………………………………………..99
Chapter 7
Conclusion and recommendation……………………………………………..101
10
LISTS OF TABLES
Table 1 TABLE
Table 2 REGRESSION TABLE
Table 3 ANOVA
Table 4 REGRESSION COEFFICIENT TABLE
11
LIST OF ABBREVIATION
EMH efficient market hypothesis
GSE Ghana stock exchange
CSD central securities depository
ATS Automated trading system
GATS General automated trading system
CAT Continuous auction trading system
LDM licensed dealing member
GSE-CI Ghana stock exchange composite index
GSE-FSI Ghana stock exchange financial stock index
OCT over the counter market
FOL First official list
SOL Second official list
EPS Earning per share
BM Book-to-Market value
ROA Return on Asset
GSM Ghana stock market
12
13
There is an old joke among economists which goes like this: “two economists were
walking along the street. One of them saw a note of GHc 10 cedis lying down at
the middle of the road and leans down to retrieve it. The second one said” don’t be
ridiculous because if that were a genuine GHc 10 someone would have picked it up
ABSTRACT
This paper is an attempt to investigate the validity of weak form efficiency market
hypothesis on Ghana stock exchange. A study conducted by frimpong Joseph
Magnus concluded that Ghana stock exchange is weak form inefficient and another
study conducted by Kofi A. Osei also confirmed that GSM is inefficient with
respect to annual earnings information releases by the companies listed in the
GSE.This paper provides a brief description of the Ghana stock exchange and
gives the meaning of efficient market. The regression coefficient is applied on the
Ghana stock exchange all-share index to carry out the test.
If a capital market is efficient the price of the stock should reflect the knowledge
and expectation of the investors and consequently no investor should be able to
receive an abnormal return since there is no way that he could know something
that has not been already reflected in the price of the stock. The main findings of
this project do not support the weak form efficient market hypothesis in the Ghana
stock exchange. The GSE is weak form inefficient which means that some
investors are able to predict the price of the security and get excess returns.
14
Chapter 1
INTRODUCTION
RESEARCH QUESTION
This paper aims at investigating the weak form efficiency of the Ghana stock
exchange and also to determine whether the theoretical ideas are pertaining to the
real life. In this context the Ghana stock exchange will be tested and the results
obtained would be used to answer the following questions:
1- Is Ghana stock exchange weak form efficient?
2- How far does the weak form efficiency hold true in the functioning of the
Ghana stock market?
3- Can the security prices be predicted in the Ghana stock market?
4-can there be investors with information which is not reflected in the security
prices?
5-Is Ghana stock market fully reflecting the information?
6-How is the Ghana stock exchange reacting to new information?
MOTIVATIONS OF THE RESEARCH
As stated earlier, the objective of this study is to test the weak form of efficiency in
Ghana stock exchange which has not been deeply investigated and studies about
the weak form are very few. This study will help the investors to know how the
15
Ghana stock market works, how to invest in the Ghana stock exchange and the
importance of an efficient market such as encouraging share buying, giving correct
signals to managers and allocating the resources. In the objective of testing the
weak form efficiency some theories will be tested and some aspects of the Ghana
stock market will be looked at in order to achieve the stated objective.
JUSTIFICATION OF THE RESEARCH
This project will be added to the body of knowledge by making a special case on
GSE on the subject of efficiency hypothesis of the stock market in its weak form.
This study attempts to investigate the weak form efficiency and ways of improving
the GSE to become a world class exchange.
This study will inform the investors about how to invest in the stock market, the
benefits that they can derived from the stock market and also roles of the
government in improving the efficiency of the stock market.
The Ghana stock market has been operating for some years now. The movement of
the market is measured by the GSE-All share index.
The GSE is an emerging capital market where many of the investors do not have
adequate education about investing in the stock exchange. Investing is not a game
of luck or chance but it requires knowledge about securities, the stock market and
the economic indicators. So there is a need for them to understand how the stock
exchange works and how its responds to the changes in the economic variables
such inflation, GDP, interest rate and others. Also arouse investors‘ awareness
about the opportunities available in the stock market and how to exploit them.
16
LIMITATIONS OF THE STUDY
The aim of this study is to investigate the weak form efficiency of the Ghana stock
exchange. Other forms of the market efficiency have not been taken into account.
The time for this study is limited to the period of January 2007 to December 2008
excluding the holidays and non-working days.
THE OUTLINE OF THE PROJECT
This project work consists of seven chapters. Chapter 1 is the introduction which
deals with abstract, research question, motivation, justification, methodology,
background of the study, previous studies, underlining assumption, performance of
the GSE, importance of the stock market in the developing countries and the
political situation in Ghana.
Chapter2 deals with history and development of the financial market in Ghana,
structure of the GSE, role of the GSE in the Ghanaian economy and challenges
face by the GSE. Chapter 3 deals with the trading system, legal regulatory
framework, the listing requirement and clearing and settlement. Chapter4 deals
with literature review where we define the EMH, talk about the forms and types of
efficient market, market conditions consistent with efficiency and the early models.
Chapter 5 describes other techniques like fundamental and technical analysis,
market anomalies and behavioral finance. Chapter 6 deals data and research
methodology, reasons behind inefficiency and the solutions for efficiency and
chapter 7 deals with conclusion and recommendation for further research.
17
METHODOLOGY
The primary data was collected through interview and questionnaires sent to the
various brokerage houses and interview with the staff and management of the GSE.
The secondary data was collected through the website of the GSE by using
information sources like press releases, half year and annual reports and others.
Other information was collected through the use of the library of the GSE and
other reading materials. In addition to information collected through the internet.
Some information like the Ghana stock exchange all-share index for the period of
January 2007 to December 2008 was collected through the website of the security
and exchange commission in Ghana. In addition to other relevant information
18
BACKGROUND OF THE STUDY
The stock market is an organized public market for the trading of a company stock
and derivatives at an agreed price. The stock market is different from the stock
exchange which is an entity that brings together buyers and sellers of a stock. They
buy and sell securities such as shares and bonds etc... Prices on the stock market
are determined by the forces of the demand (what people are willing to buy) and
supply (what sellers have available for sale).
The market is supervised by officials of the stock exchange as well as the
Securities and Exchange Commission.
On this market, individuals and companies can buy shares and bonds and other
securities of the companies through licensed dealing members or stockbrokers of
the Ghana stock exchange. The buyers include people residing in Ghana as well as
non-residents. It includes Ghanaians and foreigners. Adults (people above 18
years) can buy shares for themselves and also buy in trust for their children.
The Ghana stock exchange is the cornerstone of the Ghanaian capital market.
The idea of establishing a stock market in Ghana lay on the drawing board for
almost two decades prior to its implementation. In February 1989, the issue of
establishing a stock exchange moved a higher gear when a 10 – member National
Committee, under the chairmanship of Dr. G.K Agama, then governor of the Bank
of Ghana, was set up by the government.
The work of the committee was to consolidate all previous work connected to the
stock Exchange project and to fashion out modalities towards the actual
establishment of the Exchange. As a result of the work of the committee, the stock
exchange was established in July 1989 as a private company limited by guarantee
19
under the companies‘ code of 1963. It was given recognition as an authorized stock
exchange under the stock exchange act of 1971 (act 384) in October 1990. The
council of the exchange was inaugurated on November 12, 1990 and trading
commenced on its floor the same day. The exchange changed its status to a public
company limited by guarantee in April 1994.
The exchange was set up with the following objectives:
1- To provide the facilities and framework to the public for the purchase and
sales of bonds, shares and other securities,
2- To control the granting of quotations on the securities market in respect of
bonds, shares and other securities of any company, corporation, government,
municipality, local authority or other body corporate,
3- To regulate the dealings of members with their clients and other members,
4- To co-ordinate the stock dealing activities of members and facilitate the
exchange of information including prices of securities listed for their mutual
advantages and for the benefit of their clients,
5- To co-operate with associations of stockbrokers and stock exchanges in
other countries, and to obtain and make available to members information
and facilities likely to be useful to them or to their clients.
Since its inception, the GSE's listings have been included in the main index, the
GSE All-Share Index. In 1993, the GSE was the sixth best index performing
emerging stock market, with a capital appreciation of 116%. In 1994 it was the
best index performing stock market among all emerging markets, gaining 124.3%
in its index level. 1995's index growth was a disappointing 6.3%, partly because of
high inflation and interest rates. Growth of the index for 1997 was 42%, and at the
end of 1998 it was 868.35 (see the 1998 Review for more information). As of
20
October 2006 the market capitalization of the Ghana Stock Exchange was about
111,500bil cedis($11.5 billion). As of December 31 2007, the GSE's market
capitalization was 131,633.22bil cedis. In 2007, the index appreciated by 31.84%
(see the ―Publications‖ section on the GSE‘s website for more information).The
manufacturing and brewing sectors currently dominates the exchange. A distant
third is the banking sector while other listed companies fall into the insurance,
mining and petroleum sectors. Most of the listed companies on the GSE are
Ghanaian but there are some multinationals. Although non-resident investors can
deal in securities listed on the exchange without obtaining prior exchange control
permission, there are some restrictions on portfolio investors not resident in Ghana.
The current limits on all types of non-resident investor holdings (be they
institutional or individual) are as follows: a single investor (i.e. one who is not a
Ghanaian and who lives outside the country) is allowed to hold up to 10% of every
equity. Secondly, for every equity, foreign investors may hold up to a cumulative
total of 74% (in special circumstances, this limit may be waived). The limits also
exclude trade in Ashanti Goldfields shares.
These restrictions were abolished by the Foreign Exchange Act, 2006
(Act723).There is a 8% withholding tax on dividend income for all investors.
Capital gains on securities listed on the exchange will remain exempt from tax
until 2015. The exemption of capital gains applies to all investors on the exchange.
There are no exchange control regulations on the remittance of original investment
capital, capital gains, dividends, interest payments, returns and other related
earnings. Potential changes at the exchange include the introduction of automated
trading and the listing of some state banks. The Bank of Ghana plans the
development of mutual funds, unit trusts and municipal bonds at a subsequent date.
These changes are aimed at making the exchange more relevant, efficient and
21
effective. The exchange was also involved in preparing the draft law on collective
investment vehicles.
PREVIOUS STUDIES
Some studies have been done in order to test the efficiency of the GSE. First Osei
in 2002 conducted a study on the GSE. The objective was to test the response to
annual earnings announcement on the GSE. The study concluded that the GSE was
inconsistent with efficient market hypothesis with regard to earning announcement.
Frimpong in 2008 test the weak form efficiency on the GSE and rejected the null
hypothesis in concluding that the GSE is weak form inefficient.
As a result of this study, we can say is not a waste of time for an investor to
look for an under-valued or over-valued stock. Some hardworking investors
can reap abnormal profits and also some smarts investors can outperform the
market or consistently beat the market.
This inefficiency has a positive effect on the investors because it can allow
them to increase their returns, to know in which company invests and save
their investment.
The negative effect of this inefficiency is that investors will know about the
under-valued stocks and give low price for them which will discourage the
managers and the shareholders. Sometimes it will lead to the non-
performance of the company which will even discourage other companies to
list in the GSE.
22
UNDERLINING ASSUMPTIONS
This study investigates the weak form efficiency with focus on the Ghana stock
exchange. It explains how the trade, listing, clearance and settlement take place in
the Ghana stock exchange and also how an investor foreign or national can invest
and reap profits from the Ghana stock exchange.
The implicit assumption is that the Ghana stock exchange should be efficient for
investors to make inform decision and not allowing some investors to get excess
returns without accepting extra cost. Hence the topic ―efficiency hypothesis of the
stock market‖
PERFORMANCE OF THE GHANA STOCK EXCHANGE
The benchmark measure of performance of the Ghana stock exchange was the
Ghana stock exchange all-share which is now the GSE-composite index (GSE-CI).
From 1990 to 2009 this index has increased from 77.25 points to 5,572.34 points.
Despite the global financial crisis, the GSE in 2008 has performed well by
recording 10,431.64 points. In 1993, the GSE emerged has the 6th
best performing
emerging stock market with an impressive return of 114%.
In 1994, in a study conducted by Birinyi Associates, a research group in the USA,
the GSE came out as the best performing market among all the emerging markets
with a gain of 124%.
In 2008, the GSE was the world‘s best performing stock exchange despite of the
world global financial crisis with a gain of 58% but in the subsequent year the
performance was poor with a loss of 47%.
23
In February 18, 2011, the stock inched down to 0.86 points for the fourth straight
session to close at 1,048.78 points. The year to now stands at 4.88 percent.
The GSE financial stock (GSE-FSI) which tracks the performance of listed stocks
was also down 0.54 points to 1,036.56 points from 1,037.10 points at the previous
session, with a year-to-date change of 3.66 per cent.
There were three price changes, a gainer and two losers. Standard Chartered Bank,
the sole gainer for the session, was up GH¢0.04 to GH¢46.79. Decliners for the
session were Fan Milk Limited, which dipped GH¢0.02 to GH¢2.50 and Ghana
Commercial Bank fell GH¢0.01 at GH¢2.49.
Market Capitalization closed the session marginally lower at GH¢20,124.29
million from GH¢20,128.54 million. Total shares traded are 210,775 shares, valued
at GHC96, 469.(the Ghanaian journal February 18, 2011).
IMPORTANCE OF THE STOCK EXCHANGE IN THE DEVELOPING
COUNTRIES
One way of achieving economic growth in the developing countries is through an
efficient financial market that will channel resources for productive investments,
mobilize foreign capital and also provide funds to increase the productivity of the
national industries and reduce poverty by creating more jobs.
The financial stock market serves as a veritable tool in the mobilization and
allocation of savings among competing uses which are critical to the growth and
efficiency of the economy (Alile, 1984). Other researchers have showed that the
stock market development positively influences the economic growth (Korajczyk,
1996, Levine and Zervos, 1998).
24
Levine (1991) showed a positive relation between financial stock market and
economic growth by issuing new financial resources to the firms. The financial
stock market facilitates higher investments and the allocation of capital, and
indirectly the economic growth.
An efficient stock market contributes to attract more investment by financing
productive projects that lead to economic growth, mobilize domestic savings,
allocate capital proficiency, reduce risk by diversifying, and facilitate exchange of
goods and services (Mishkin 2001; and Caporale et al, 2004). Further, Levine and
Zervos (1998) found strong statistically significant relationship between stock
market development and economic growth. The result of Filer et al. (1999)‘ studies
show that there is positive causal correlation between stock market development
and economic activity. Many other researchers argue that there is a positive
correlation between financial development and economic growth
(Goldsmith, 1969; Shaw, 1973; McKinnon, 1973 and King & Levine, 1993). They
found that financial development is an important determinant of future economic
growth of a country. Atje and Jovanovic (1989) found also a significant impact of
the level of stock market development and bank development. Joseph
Schumpeter (1912)‘s book was the most important and thorough one of the earlier
contribution on financial development and economic development. For him,
financial development causes economic development – that financial markets
promote economic growth by funding entrepreneurs and in particular by
channeling capital to the entrepreneurs with high return projects. Therefore the
GSE also needs to be efficient so that it can contribute the growth of the Ghanaian
economy.
25
POLITICAL SITUATIONS IN GHANA
One of the major differences between efficient and inefficient market is the
political situation in the country. Political instability sometimes results in stock
market being manipulated by the government which will cause the inefficiency of
the stock market.
When the government is controlling the stock market all the information will not
be provided to the public therefore the decision making process will be difficult
and the decision base on this information will not be accurate.
The prices will reflect neither the true value of the securities nor other information
in the market. This will result in the stock market being inefficient.
Also as a result of this political instability, investors will not be motivated to invest
neither in the Ghana stock exchange nor in the capital market by fear of losing
their investments. Therefore, we need a stable political situation as one of the
conditions of the market being efficient.
26
CHAPTER 2
A-HISTORY AND DEVELOPMENT OF THE FINANCIAL MARKET IN
GHANA
The economy of Ghana underwent economic reforms in 1983 at the time when the
country was collapsing with high inflation around 123%, large exchange rate
swings and negative real interest rate. These reforms were aiming at liberalizing
the Ghanaian economy. So the government of Ghana with the help of the World
Bank had launched the structural adjustment programme to address the
deterioration problems in the financial sector.
As part of the reforms, monetary policy was instituted and market for securities
was created. In 1986 a weekly auction in treasury bills was introduced. Again the
bank of Ghana bills was also introduced in 1988 to take care of excess liquidity in
the system especially in the rural areas and also to provide an avenue of investment
for banks (bank of Ghana consultation paper October 2007).
The Ghana stock exchange was established as part of these reforms. The GSE
though incorporated in July 1989, the normal operations started on November,
1990.
One main objective of establishing the GSE was to enable corporate institutions
and government to raise quick capital to accelerate development in order to reduce
undue reliance on donors.
After establishing the GSE in November 1990 with only 12 listed companies and
one government bond, the market capitalization within two years of operation
increased from GH ¢3 billion in 1991 GH¢4.3 billion in 1992 while the listed
companies increased to 15. With a drop of 46.58% in the GSE All-share index, the
27
GSE ended the year 2009 as the least performing market in Africa. In the previous
year 2008, the gain in the GSE All-share index was 58% and this moved Ghana
ahead of all the African markets.
The market capitalization of the exchange declined by 11% to end the year at
GH¢15.94billion down from GH¢17.90billion in 2008. The decline in the value of
Ghana‘s market was mainly due to price depreciation.
According to Bank of Ghana report, for the fifth consecutive month, headline
inflation went up from 12.8% in January to 13.2% in February and this was due to
prices increased in the first two months of the year. This increased, the report said
was driven by surges in food prices and rising crudes oil prices. (Bank of Ghana
Monetary Policy report: vol 3 no 2: 2008 1).
In this market condition one would confirm the theory put forward by Samuelson
and Mandelbrot followed by Eugene Fama, which is ―Efficient Market
Hypothesis‖: they said that when markets are working properly, then all public
information regarding an asset will be incorporated immediately into the price.
The GSE was adjudged ―the most innovative African stock exchange for 2010‖ at
the African investors prestigious annual index series awards held at the new York
stock exchange on Friday, 17th September 2010.
The African investor index series awards are the only international pan-African
that recognize and reward Africa‘s institutional investors, stock exchanges, best-
performing listed companies, stockbrokers and capital market regulators. The
awards assessed performance between April 2009 and April 2010 out the group of
seven stock exchanges nominated (www.gse.com)
28
In their objective of being efficient and to position the GSE to compete favorably
on the international market as well as attracting more investors in the domestic and
international market, the GSE has adopted the automated trading system(GATS)
which is a computer base trading that automatically submits trades onto the trading
system. The system is designed to match orders placed by the stockbrokers to buy
and sell shares in an electronic order book and these orders are automatically
matched i.e the system determines who buys what at which price and from which
seller
B-THE STRUCTURE OF THE GHANA STOCK EXCHANGE
The exchange is governed by a council with representation from licensed dealing
members, listed companies, the banks, insurance companies, money market and
the general public. The Managing Director of the exchange is an ex-officio
member. The council sets the policies of the exchange and its functions include
preventing fraud and malpractices, trading and settlement and maintaining good
order among members, regulating stock market business and granting approval for
listing, maintaining public confidence in the market and promoting the exchange. It
has the power to suspend or expel any licensed dealer member that contravenes the
rules of the exchange. The council‘s membership includes some of the most
distinguished and competent people in the Ghanaian commerce, industry, finance
and public service.
29
ROLES OF GHANA STOCK EXCHANGE IN THE GHANAIAN ECONOMY
The GSE plays the following roles in the Ghanaian economy:
-Raising capital for businesses
-Major link between companies and governments with capital need and the public
with savings to invest.
-The growth of the economy by mobilizing savings for investment
-Creation of jobs
-Promote savings and investment habit
-Provide liquidity to investors
-Establish a pricing mechanism for financial asset
-Enhance corporate transparency or corporate governance
-Improved share appreciation
-Enhance volumes of securities traded
-Facilitating company growth
-Profit sharing
-Barometer of the economy
-It provides a market place for trading
-Provide a powerful tool for the government and policy makers to manage
economies
-Improve share price appreciation and enhance volumes of securities traded
30
THE CHALLENGES FACE BY THE GHANA STOCK EXCHANGE
The Ghana stock exchange is facing the following problems:
First is the political instability that has been discussed in the earlier section.
The second is the macroeconomic instability or volatility. The Ghana‘s economy
has been characterized by high inflation, high interest rate and large exchange rate
swings as stated by the Bank of Ghana Monetary Policy report: (vol 3 no 2: 2008
1). These variables when they are high, they discourage the investors to invest in a
specific country because they can‘t diversify the risk and also are not sure of
getting their investment back.
Another problem facing the GSE is due to the stock exchange itself. According to
Benimadhu (2003), among the exchange specific issues affecting stock markets in
Africa are low level of liquidity, few listed companies and the small size of the
exchange as well as efficiency. We assume that the stock exchanges in Africa face
the same challenges. Other factors which has stifled the development of the GSE
relates to the absence of a strong and active domestic investor base, led by
institutional investors such as pension funds and insurance companies. Clearing,
Settlement and Trade systems as well as trading infrastructure should be
ameliorated.
The following points need to be noted carefully:
1-Investors confidence
Fluctuation in interest rates, high rate of inflation and instability of the cedi made it
difficult to predict the future of the capital market and because of this, the investors
31
find it difficult to invest or borrow money from the market. According to the bank
of Ghana in his report, inflation went up to 12.8% in January 2008 to 13.2% in
February 2008. The report attributed the price increased to the increased in food
prices and the rising crude oil. (Bank of Ghana monetary policy report: Vol 3 No 2
2008 1)
2-Unpredictability of the market
In terms of instruments traded and the number of participants in the market, the
capital market in Ghana is relatively small as compared to that of the stock
exchange in America or Paris or UK. Fluctuation of the economic indicators and
instability of the cedi have made very difficult for the investors to predict the
Ghanaian capital market. An evaluation of the real return on the Ghana stock
exchange by the Bank of Ghana revealed that ― the total annual real return on the
stock listed on the Ghana stock exchange have followed an undulating pattern
since 1991 falling every two years and rising every two years‖.(Bank of Ghana
report)
3-Price forecast
The capital market in Ghana has been characterized by unstable price. Prices of
goods and services are volatile and the same for inflation, interest rates and
exchange rates. This condition will make it difficult for predicting the share prices
which will confirm the theory of efficient market hypothesis.
Some positive changes
Re-denomination
According to Ghana news agency report in 2008, the re-denomination of the
Ghanaian cedi in july 2007 was predicted as something which will strengthen the
32
financial market in Ghana. This is because it will reduce high transaction costs and
also reduce the inconvenience and high risk in carrying loads of currency for
transaction purposes. It would further ensure compatibility with data processing
software and the strain on payments system, particularly Automated Teller
Machine (ATMs) and above all it would reduce the difficulties in maintaining
financial and statistical records.
Foreign investors
Non-resident foreign investors can now hold more than 10% of any security listed
in the stock market which was not the case few years back. They can trade in
money market instruments and also maintain a foreign currency account with the
local banks which can be credited with transfer in foreign currency from abroad or
other foreign currency account. (www.ghanaweb).
33
Chapter 3
A-TRADING SYSTEM
1-Principles of trading
The purpose of the Ghana stock exchange is the determination of the price of the
securities in a functional and efficient market, where all parties to the market have
simultaneously at their disposal sufficient information as a basis for price
formation.
Procedure that is contrary to good conduct shall not be employed in securities
trading.
A licensed Dealing Member shall maintain confidence in the market through it
activities.
A licensed Dealing Member must not take measures that are designed to distort the
price level or the number of securities, or any measures that do not correspond to
the form or intent of a legal act.
A licensed Dealing Member shall carefully follow the principles and rules
pertaining to trading.
Trading is carried out on the floor of the exchange under the continuous Auction
Trading system (CAT). Over the counter trading is however allowed in Ashanti
Goldfield Company‘s shares. Trading is done in lots of 100 shares. The automated
trading on the Ghana stock exchange is done through three alternative levels of
access, namely
1-on the trading floor of the Ghana stock exchange
2-At the office of the stockbroker through the wide Area Network
34
3-Through the internet
When an investor instructs his stockbroker to buy or sell a particular security, the
broker will enter the order on his terminal using any of the three means and the
order will instantly be routed electronically to the stock exchange‘s ATS
(automated trading system). The system automatically matches the orders resulting
in trades. The GATS records the sale, quantity, buyer and seller and time of trade.
All these are done under the supervision of the Ghana stock exchange.
The Ghana Stock Exchange (GSE) has begun a new arrangement for a "continuous
auction trading" on trial basis for eight equities listed on its counters.
The new arrangement, which started two weeks ago, allows the Exchange to trade
for longer hours and forms part of measures to address difficulties faced in the use
of the current trading system known as the 'Call Over System'.
Equities involved in the new arrangement were chosen at random to give all listed
companies a chance to benefit from the trial run, which will last for three weeks.
If successful, the new arrangement will cover the rest of the equities after an
assessment.
Equities picked for the trial are British-American-Tobacco Company (BAT),
Enterprise Insurance Company (EIC), Fan Milk Limited (FML), Ghana
Commercial Bank (GCB), Guinness Ghana Limited (GGL), Mobil Ghana Limited,
Produce Buying Company (PBC) and Unilever Ghana Limited (UNIL).
The General Manager of GSE, Francis Tweneboa, told the Ghana News Agency
35
(GNA) that the exchange had experienced some difficulties in the current Call over
System.
This system allows offers and bids for only one stock at a time and does not make
room for clients to bid to their advantage.
With the Call over System, names of the stocks are shouted for offers and bids. If a
client wishes to make a transaction after his chance elapses, he or she has to wait
until the next round to make a match.
In the new system, boards will be provided for the quotation or display offers and
brokers can match orders as and when it favours their clients.
Tweneboa said whereas the Call over System lasts for about two hours, the new
arrangement would continue for about six hours from 1,000 hours to 1,600 hours.
The closing time would also give brokers enough room to make settlements.
Yofi Grant, an Executive Director of Databank Brokerage Limited, lauded the new
arrangement, saying it is an attempt to modernise activities of the GSE.
"The Call over System has become obsolete and is a far cry from what exists on
international markets," he said, commending its potential to allow brokers to talk to
clients during trading on the floor.
Grant said brokerage firms agree to the new arrangement, which will serve as a
pre-cursor to electronic trading to enable the Ghana bourse to be part of
36
globalisation.
He said, however, that modernisation of activities on the GSE must be gradual to
allow the smaller of the 12 brokerage firms who trade on the floor to catch up with
the pace in terms of expansion of offices, clients and expansion of the trading floor
OFFICIAL TRADING HOURS
Trading in approved securities, as defined by the listing rules, shall be limited to
the hours when the floor is open for the transaction of business.
There will be no trading in approved securities on the floor or anywhere else
before or after the official dealing hours, as defined by the council from time to
time, unless the council has expressly granted an exemption in which case trading
shall be done strictly within the conditions prescribed by the council.
Dealing hours was as follows:
-Pre-opening period 09.30 GMT 10.00 hrs GMT
-Open AUCTION 10.00 hrs GMT and
-Regular trading 10.00 hrs GMT -12.00 hrs GMT
According to the Business news of Tuesday, 4 January, 2011, the Ghana Stock
Exchange (GSE) has
extended its daily trading hours from three hours to five hours
effective January 4, 2011. The new trading hours shall be between 10.00 hrs and
37
15.00 hrs
GMT instead of from 10.00 hours to 13.00 hours. A statement by the Exchange on
Tuesday said the change in trading hours was to afford dealers increased contact
hours with their clients during the trading day. It is also meant to afford non-
resident investors in time zones different from Ghana, greater opportunity to reach
out to their local brokers, it added. "It is our expectation also that the extended
trading hours will help improve liquidity in the market place," the statement said.
TYPES OF SECURITIES TRADED
Ordinary shares 35
Preference shares 1
Debt securities corporate 2
Government
2 years 56
3 years 36
5 years 3
INDEX
An index is a numerical value used to measure changes in a variable or group of
variables. The index is set of hypothetical or numerical level on the base period or
starting point against which a percentage change can be compared to at any
particular point of time.
38
The stock market index is a specialized tool used essentially to capture the overall
performance of the stock market. It can also be employed to measure how a given
equity or bond portfolio is performing. If an investor owns more than one stock, it
is cumbersome to follow each stock individually to determine the composite
performance of the portfolio.
The stock index will reveal the overall trend in the performing of the market. It
also shows the stock price movements.
The Ghana stock exchange has introduced a new method of calculating closing
prices of equities on the market with effect from January 4, 2011. Closing prices of
listed equities will be calculated using the volume weighted average price of each
equity for every given trading day.
GSE has also published two new indices, namely the GSE composite index (GSE-
CI) and the GSE financial stock index (GSE-FSI)
*GSE composite index (GSE-CI)
The calculation of the GSE composite index will be based on the volume weighted
average closing of all listed stocks. All ordinary shares listed on the GSE are
included in the GSE-CI at total market capitalization. The base date for the GSE-
CI December 31, 2010 and the base index value is 1000.
*GSE financial stock index (GSE-FSI)
This index will have its constituents as listed stocks from the financial sector
including banking and insurance sector stocks. All ordinary shares of the financial
stocks listed on the GSE are included in the GSE-FSI at total market capitalization
except for those stocks which are listed on other markets. The base date of the
GSE-FSI is also December 31, 2010 and the base index value is 1000.
39
The new indices will be published on each trading day beginning January 4, 2010.
B-CLEARING AND SETTLEMENT
Clearing is defined as ―the process of transmitting, reconciling and, in some cases
confirming payment orders or security transfer instructions prior to settlement‖
(BIS, 2003, p.13). Whereas settlement is described as ―the completion of a
transaction, wherein the seller transfers securities or financial instruments to the
buyer and the buyer transfers money to the seller‖ (BIS, 2003, p.45). The objective
for clearing and settlement process is that the system is prompt and reliable as well
as cost-effective and convenient to use. (Guadamillas & Keppler, 2001, p.19)
Clearing and settlement of securities transactions in listed securities shall be
conducted at the premises of the exchange or as may be determined by the central
securities depository (CSD) act or the exchange.
The exchange in conducting the clearing and settlement business shall incur no
other liability or responsibility for the conduct of such business other than seeing to
an orderly conduct of the business until the central securities depository
commences clearing and settlement for listed securities.
The clearing and settlement for a trading session(T) shall take place three business
days after the related trading session which is (T+ 3) at 11.am or within a given
time frame that the exchange shall specify by notice of amendment of this rule.
Settlement is manual but centralized. Currently, the settlement period is T+3
(business days). It is also delivery versus payment. The seller (stockbroker) of a
security is responsible for the validity of all documents delivered. The exchange
has set up a clearing house. All LDM (licensed dealer member) are members of the
40
clearing house. They together with the exchange have accounts with the settlement
bank. On settlement day, the clearing house collects (debits) the bank account of
the LDMs who have bought securities and pays the amount collected into an
account of the clearing house. LDMs who sold securities are next paid or credited
with the amounts due to them move to the buying LDM in the Ghana stock
exchange depository.
C-LEGAL OR REGULATORY FRAMEWORK
Legislation
- The company code, 1963, (act 179)
- The security industry law, 1993 PNDCL 333
- Securities and exchange commission regulations 2003 1728
- Act 1695 GSE regulations/rules
- Stock exchange (Ghana stock exchange)
- Listing regulations 1990 LI 1509
- Stock exchange (Ghana stock exchange)
- Membership regulations 1990 LI1509
- Trading and settlement rules
D-REGULATIONS AFFECTING FOREIGN AND NON-RESIDENT
INVESTORS
Exchange control permission has been given to non-resident Ghanaians and
foreigners to invest through the exchange without any prior approval. However,
one external resident portfolio investor (whether individual or institution) can hold
only up to 10% of any security approved for listing on the exchange. Furthermore,
41
the total holdings of all external residents in one listed security shall not exceed
74%
Ghanaians externally resident and foreigners resident in Ghana though may invest
any limit. The limits also apply in the case of AGC and PBC shares.
There is free and full foreign exchange remittability for the original capital plus all
gains and related earnings.
There is a 10% withholding tax (which is also the final tax on the dividend income)
for all investors both resident and resident. Capital gains on the listed security are
exempt from tax until November 2005.
LISTING ON THE GHANA STOCK EXCHANGE
A company is said to be listed on Ghana stock exchange when its securities are
approved to be bought and sold on the stock exchange. Newly issued shares cannot
be traded on the OCT (over the counter) market before getting listed on the GSE.
In order to get your newly shares listed on the GSE, you need to communicate this
intention early to the GSE and work with the exchange so that your prospectus will
satisfy the exchange‘s listing requirements before the public floatation.
A-Requirements for listing
There are certain requirements that the company must meet in order to qualify for
listing on the Exchange. These requirements are stipulated in GSE rules Book.
Many companies will meet these qualifications.
42
To make it possible for many companies to list on the exchange, there are two lists
with fairly easy requirements:
a- The first official list (FOL)
b- The second official list (SOL)
The approval of an application for the listing securities on the stock exchange is
solely within the discretion of the council of the Ghana stock exchange.
The requirements are as follows:
1-listing requirements for shares
Minimum stated capital
The company must have a stated capital after the public floatation of at least GHc
1million in the case of an application relating to the first official list and GHc
0.5million for the second official list.
Minimum public float
Shares issued to the public must not be less than twenty-five percent (25%) of the
number of shares issued by the company unless reasonable justification is provided
for a smaller percentage.
Payment of shares
Shares must be fully paid for. Except in very exceptional circumstances, the
exchange will refuse listing in respect of partly paid shares
Spread of shares
43
The spread of shareholders existing at the close of an offer should be in the GSE‘s
opinion adequate with at least 100 shareholders after the public offer.
2-listing requirements for debt securities
A company seeking the admission of debt securities to the GSE may be considered
for such admission if the security concerned has a total issue amount of not than
GHc 1million face value or there are at least 50 holders of such security.
In the case of government securities, there is no prescribe minimum in respect of
either amount issue or number of holders to permit admission to the GSE‘s list
Debt securities for which listing are sought shall be created and issued pursuant to
a Trust Deed duly approved by Securities and Exchange Commission.
ADDITIONAL REQUIREMENTS FOR LISTING
The following additional requirements apply to both applications for listing of
equities and debts securities.
Period of existence
For a company‘s securities(whether equity or debt) to be eligible for admission to
the GSE, the company must have published or filed audited accounts in accordance
with the companies code, 1963, (act 179) for at least three years immediately
preceding the date of its application for listing. For the SOL, in respect of shares,
the period is one year.
Profitability
Profitability must have been reasonable through the periods stated above or the
company must have strong potential to be profitable.
44
Conditions relating to Directors and Management of Applicant
There must have been continuity in the management of a company seeking
admission to the GSE.
The character and integrity of the directors and management of the company will
be among the criteria taken into account by the council of the exchange.
At least 50% of the board must be non-executive directors and this number at least
two (2) or 25% shall be independent
45
Chapter 4
Literature review
HISTORY OF EMH
IN GEORGE Gibson‘s book the stock markets of London, Paris and New York,
published in 1899; we see one the first reference to market efficiency. He says, ―In
an open market, when the shares become publically know, the value they obtain
may be considered to reflect the most intelligent appreciation possible”.
Louis jean Baptist Alphonse Bachelier also made reference to market efficiency in
his PhD mathematics dissertation paper to the University of Sorbonne. In his
opening paragraph Bachelier recognizes that “past, present and even discounted
future events are reflected in market price, but often show no apparent relation to
price changes”. And he continue to say that ―if the market, in effect, does not
predict its fluctuations, it does access them as being more or less likely and his
likelihood can be evaluated mathematically‖. So according to this theory the
mathematical performance expectations are useless. Therefore neither buyers nor
sellers can make systematic profit and also the price in such market follows
uncertain fluctuation.
Bachelor‘s work was made known to Paul samuelson through Leonard j. savage
who describes efficiency in terms of winning formula.
From 1932-1953 the trio of cowls, working and kendall stir things up by
challenging the ability of professionals to anticipate stock market prices, since they
essentially would have uncertain progress. We must say that on Wall Street, there
is quite an industry of advice to investors based on the forecasts of specialists.
46
In 1933 Alfred cowls publishes in econometrica ―can stock market forecasters
forecast?‖ he gave a conference before the econometric society where he studied
for 45 professional agencies, the ability of selecting stock or of anticipating the
movement of the markets. They do not have better result than the choice made
randomly.
In 1934 holbrook working publishes ―A random difference series for use in the
analysis of time series‖. In the journal of the American statistical Association. He
establishes that there is no correlation between the successive variations of prices
on the markets.
In 1953 Maurice G. Kendall publishes ―the analytics of Economic times series,
Part 1 : in which he demonstrates that the fluctuation of stock prices on the market
is essentially uncertain. This worried many economists for whom the stock prices
changed in order to reflect the modifications in the forces of offer and demand. We
were going towards the emergence of the hypothesis of market efficiency for
which the uncertain nature of price evolution is a typical symbol of good market
behavior.
In 1963 during a meeting of the Econometric society, Clive W.J Granger and
Oskar Mogenstern present the results of their research on the stock market pricing.
They discovered that the short term movements are uncertain but not long term
ones. They also specify that economic cycles were of little or no importance.
In 1965 Paul A Samuelson publishes ―Proof that properly anticipated prices
fluctuate randomly‖ in the magazine industrial management review.
Samuelson explains that if stock were predictable and not uncertain, it would
therefore be possible to take action in order to generate systematic gains. But these
47
actions made the stocks uncertain. Therefore, the uncertain nature of the stock
markets does not go against the economic forces who to the opposite, work very
well. This is how the market efficiency hypothesis was born which, at first, was
synonym of uncertain markets.
In 1965-1998 in a series of publications, Eugene Fama will become the father of
the market efficiency. But the notion of efficiency covers different aspects and
clarify the debate, james Tobin suggest in 1984 in ―on the efficiency of the
financial system‖, four major forms of efficiency which include, in economics and
finance, the principal meaning of the concept: informational efficiency,
fundamental, total insurance and behavioral efficiency. Informational efficiency is
concerns portfolio management the most.
In 1965 in the journal of Business under the title ―the behavior of stock prices
―Eugene fama publishes his entire doctorate thesis in which he concludes that
stock prices are unpredictable and follow an uncertain market.
In 1970, fama published in the journal of Finance, the beginning of a trilogy
entitled‖ Efficient capital market: A review of the theory and empirical work‖. He
defines informational efficiency as follows: A market is efficient when in each
instant; the prices incorporate all the pertinent and available information. He
suggests three types of informational efficiency: the weak form, semi-strong form
and strong form. He also specifies that the efficiency test includes two problems:
first the cost of the information and the transactions and second the problem of
joint hypothesis.
The joint hypothesis conditions the efficiency tests. We can never know if it is the
market which is inefficient or if it is the model that is false (the model used to
evaluate the financial assets). This hypothesis will seriously annoy the researchers.
48
In 1991 fama published ―Efficient capital market ii‖ in which he suggests an
evolution for the tests of the three types of efficiency. To replace the test on the
weak form, he suggests foresee ability of returns. For the test on the semi-strong
form, he suggests studying events and for the strong form, he suggests tests on
private information.
In 1998 fama published the third part of the trilogy, in which he concludes that the
market efficiency hypothesis has survived all of the challenges, namely by the
studies on the market anomalies and by behavioral finance.
1-The issue
Burton G Malkiel in his paper ―The Efficient Market Hypothesis and its critics‖
says that: At the outset, it is important to make clear what I mean by the term
―efficiency‖. I will use as a definition of efficient financial markets that they do not
allow investors to earn above-average returns without accepting above-average
risks. A well-known story tells of a finance professor and a student who come
across a $100 bill lying on the ground. As the student stops to pick it up, the
professor says, ―Don‘t bother—if it were really a $100 bill, it wouldn‘t be there.‖
The story well illustrates what financial economists usually mean when they say
markets are efficient. Markets can be efficient in this sense even if they sometimes
make errors in valuation, as was certainly true during the 1999-early 2000 internet
bubble. Markets can be efficient even if many market participants are quite
irrational. Markets can be efficient even if stock prices exhibit greater volatility
than can apparently be explained by fundamentals such as earnings and dividends.
Many of us economists who believe in efficiency do so because we view markets
as amazingly successful devices for reflecting new information rapidly and, for the
most part, accurately. Above all, we believe that financial markets are efficient
49
because they don‘t allow investors to earn above-average risk-adjusted returns. In
short, we believe that $100 bills are not lying around for the taking, either by the
professional or the amateur investor.
According to him, Benjamin Graham (1965) was correct in suggesting that while
the stock market in the short run may be a voting mechanism, in the long run it is a
weighing mechanism.
Fama said that market efficiency per see is testable. In 1978, Michael Jensen
declared his belief that "there is no other proposition in economics which has more
solid empirical evidence supporting it."
In 1953 Maurice Kendall published a study3 in which he found that stock price
movements followed no discernible pattern, that is, they exhibited no serial
correlation. Prices were as likely to go up as they were to go down on any given
day, irrespective of their movements in the past. These results lead to the question
of what, exactly, influenced stock prices. Past performance obviously did not. In
fact, had this been the case, investors could have made money easily. Simply
building a model to calculate the probable next price movement would have
enabled market participants to gain large profits without (or with reduced) risk.
50
2-Meaning of efficient market hypothesis
Fama in his classic paper:‖capital market: review of empirical theory‖, define the
efficient market as a market where the security prices ―fully‖ reflect all available
information. According to Malkiel in 1992, a capital market is said to be efficient
if it fully and correctly reflects all relevant information in determining security
prices. Therefore, more formally the market is efficient with respect to some
information set. Moreover efficiency implies that it is impossible to make
economic profits by trading on the basis of the defined information
set.(paper4you.com,2006).
Dyckman and Morse in 1986, state that ―A market is generally defined as efficient
if (1) the price of the security traded in the market act as though they fully reflect
all available information and (2) these prices react instantaneously, or nearly so,
and unbiased fashion to new information‖
As it was said by Malkiel in 1992, if the market is efficient, the company market
value should be an unbiased estimate of the true value. Nevertheless it is important
to stress on the following points:
1-Market efficiency does not require that market price is equal to the true value.
2-There is an equal probability that stocks are over or under valued at any point in
time.
3-And finally, investors should not be able to consistently identify under or
overvalued stocks using any investment strategy.(Damodaran 2006).
51
Nevertheless it is important to stress that markets are not efficient due to their
nature but they are driven to efficiency by the actions of the investors.(efficient
market hypothesis: myth or reality)
Fama (Jan. 1965: ‗The behavior of stock-market prices‘):
‗…an ―efficient‖ market for securities, that is, a market where, given the available
information, actual prices at every point in time represent very good estimates of
intrinsic values.‘
Fama (Sep.–Oct. 1965: ‗Random walks in stock market prices‘):
‗An ―efficient‖ market is defined as a market where there are large numbers of
rational, profit-maximizers actively competing, with each trying to predict future
market values of individual securities, and where important current information is
almost freely available to all participants.‘
Fama et al. (1969):
‗…an ―efficient‖ market, i.e., a market that adjusts rapidly to new information.‘
Fama (1970):
‗A market in which prices always ―fully reflect‖ available information is called
―efficient.‖‘
Jensen (1978):
‗A market is efficient with respect to information set θt if it is impossible to make
economic profits by trading on the basis of information set θt.‘
[‗By economic profits, we mean the risk adjusted returns net of all costs.‘]
Fama (1991):
‗I take the market efficiency hypothesis to be the simple statement that security
52
prices fully reflect all available information. A precondition for this strong version
of the hypothesis is that information and trading costs, the costs of getting prices to
reflect information, are always 0 (Grossman and Stiglitz (1980)). A weaker and
economically more sensible version of the efficiency hypothesis says that prices
reflect information to the point where the marginal benefits of acting on
information (the profits to be made) do not exceed marginal costs (Jensen (1978)).‘
Malkiel (1992):
‗A capital market is said to be efficient if it fully and correctly reflects all relevant
information in determining security prices. Formally, the market is said to be
efficient with respect to some information set, φ, if security prices would be
unaffected by revealing that information to all participants. Moreover, efficiency
with respect to an information set, φ, implies that it is impossible to make
economic profits by trading on the basis of φ.‘
Fama (1998):
‗…market efficiency (the hypothesis that prices fully reflect available
information)...‘
‗…the simple market efficiency story; that is, the expected value of abnormal
returns is zero, but chance generates deviations from zero (anomalies) in both
directions.‘
Timmermann and Granger (2004):
‗A market is efficient with respect to the information set, Ωt, search technologies,
St, and forecasting models, Mt, if it is impossible to make economic profits by
trading on the basis of signals produced from a forecasting model in Mt defined
over predictor variables in the information set Ωt and selected using a search
technology in St.‘
53
B-MARKET CONDITIONS CONSISTENT WITH EFFICIENCY
According to fama there are some conditions that might help or hinder efficient
adjustment of prices to information. These conditions are:
1-A large number of rational, profit-maximizing investors exist, who actively
participate in the market by analyzing, valuing and trading securities. The market
must be competitive meaning no one investor can significantly affect the price of
the security through their own buying or selling.
2-A market in which there are no transaction costs in trading securities
3-A market in which all available information is costlessly available to all market
participants
4-A market in which all participants agree on the implications of current
information for the current price and distributions of future prices of each security.
In such market, the current price of a security obviously ―fully reflect‖ all available
information but a frictionless market in which all information is freely available
and investors agree on its implications is, of course, not descriptive of market met
in practice. Fortunately, these conditions are sufficient for market efficiency but
not necessarily.
A market can be efficient if ―sufficient numbers‖ of investors have ready access to
available information. And disagreement among investors about the implications
of given information does not in itself imply market inefficiency unless there are
investors who can consistently make better evaluations of available information
than are implicit in market prices.
54
C-TYPES OF EFFICIENCY
Efficiency implies that all resources are being used to the potential. We can have
three types namely:
1-Information efficiency implies that the prices of the security reflect all
information relevant to pricing security. A pricing efficient market implies
efficiency in the processing of information i.e. the prices of capital assets at all
times are based on correct evaluation of all available information.
2-Allocation efficiency relates to the ability of the market to direct capital to the
projects with the highest risk-adjusted returns.
3-Operational efficiency is achieved when transactions completed on a timely
basis, accurately and at low cost.
D-FORMS OF EFFICIENCY
The term efficient market was coined by fama in 1970 and he classified it into
three forms namely:
1-WEAK FORM EFFICIENT MARKET HYPOTHESIS
The weak form theory is also known as ―Random Walk‖ says that the current price
of the stocks already fully reflect all the information that is contained in the
historical sequence of prices. In another words, we can say that the price of the
stock already stands adjusted to all the historical information available about it.
Therefore there is no benefit so as far as forecasting the future is concerned and
technical analysis which the use of the past prices to predict the future price is of
no use.
55
This theory can be explained by a simple example: a close look at the stock prices
sometimes reveals that days-of-the week effects (stocks prices may tend to rise on
Monday and fall on Friday) and also; "blue Monday on Wall Street" is a saying
that discourages buying on Friday afternoon and Monday morning because of the
weekend effect, the tendency for prices to be higher on the day before and after the
weekend than for the period of the rest of the week.
Another one is the time-of –the year effects(stock prices may tend to rise in
January), and small firm effect (small firm prices may typically rise more than
large firm prices).
Most of the studies agree support the proposition that price changes are random
and past prices change were not useful in forecasting future price changes
particularly after the transaction costs were taken into account. However, there are
some studies which found the predictability of share price changes for example
fama and French 1988, poterba and summers 1988 in developed market but they
did not reach a conclusion about profitable trading rules.
Hudson, Dempsey and Keasy (1994) found that the technical trading rules have
predictive power but not sufficient to enable excess return in UK market.
Similarly, Nicolaas (1997) also conclude that past returns have predictive power in
Australian market but the degree of predictability of the return is not so high.
The weak form efficiency of the market on the emerging countries presents
controversial result. Most the emerging countries suffer from the problem of thin
trading in addition to that in the smaller markets, it is easier for a larger trader to
manipulate the market. Though it is generally believe that the markets in the
emerging countries are less efficient, the empirical evidence does not always
support this thought. There are two types of groups of finding, the first group finds
56
weak form efficiency in developing and less developed markets are Branes,
1986,(on the Kuala lumpur stock exchange), chan,gup and Pan 1992,(in major
Asian markets); Dicknson and Muragu 1994 (on the Nairobi stock exchange) and
ojah and Karemera 1999,(on the four latin American countries market) despite the
problems of thin trading, on the other hand, the latter group, who evidenced that
the market of developing and less developed markets are not efficient in weak form
are cheung,wong and ho,(1993) on the market of korea and Taiwan in a world
bank study by claessens,Dasgupta and Glen(1995), report significant serial
correlation in equity returns from 19 emerging markets and suggest that stock
prices in emerging ,markets violate weak form; similar findings are reported by
Harvey(1994) for most emerging markets. Nourrendine Kababa (1998) has
examined the behavior of stock price in the Saudi financial market seeking
evidence that for weak form efficiency and find that the market is not weak form
efficient. He explained that the inefficiency might be due to delay in operations
and high transaction cost, thinness of trading and illiquidity in the market Roux
and Gilberson (1978) and Poshakwale S,(1996) find the evidence of non-
randomness stock price behavior and the market inefficiency on the Johannesburg
stock exchange and the Indian market.
57
Statistical Tests of Independence
According to the hypothesis of efficient markets, security returns over time are
independent of one another. When testing for autocorrelation, the significance of
positive and negative correlation in returns is measured over time. Brown (1979)
claimed that it is naive to assume that capital markets are perfect for all securities.
He argued that there is negative auto correlation in monthly stock returns, and he
also found a link between beta and the level of autocorrelation. However, he
admitted that for certain sub-samples of securities, perfect capital markets are
valid, and the predictions of the CAPM are close to reality.
MacKinlay and Ramaswamy (1988), among others, looked at intra-day arbitrage
trading between the futures and spot for the S&P 500 index. They reported some
differences in autocorrelation between returns on the futures and spot at short
intervals but find that most of this goes away at the daily level. MacKinlay (1988)
found significant serial correlation in weekly stock prices, rejecting the random
walk hypothesis. Their result did not, however, support any other patterns and
would have few implications to market efficiency. Jegadeesh (1990) reported
significant negative monthly autocorrelation. The magnitude of abnormal returns
was greatest in January. Furthermore, Boudoukh, Richardson and Whitelaw (1994)
found that autocorrelation on futures was indistinguishable from zero. They
claimed that reports of the death of market efficiency had been premature and
greatly exaggerated.
58
Trading Rules
It is not impossible to have trading rules based on price patterns in the stock
markets. Some of the trading rules will be presented in the following section.
Filter Rules
The filter rule is normally stated as ―purchase stock when it rises by X% from the
previous low, and hold it until it declines by Y% from its subsequent high. At this
point, sell the stock short, or hold cash. Fama and Blume (1966) conducted an
extensive list of filter rules tests. Small filters, typically less than 0.5% seemed to
work the best, and no filters larger than 1.5% produced better results than the buy-
and-hold strategy. The average profits were generally small, but over time, they
substantially outperformed buy-and-hold strategies. Nevertheless, when accounting
for transaction costs, which are substantial in such a transaction-intensive strategy,
all the tested filter strategies became unprofitable. Furthermore, even the most
profitable filter rules imply that cash is never idle, to compete with the buy-and-
hold portfolio. This is very difficult in real life, especially for a world without short
selling, although the aid of modern computing powers might have improved the
performance.
Sweeney (1988) reviewed Fama and Blume‘s tests, and commented that for a
sample from 1970 to 1982, some filter rules seemed to work, provided that
floor traders‘ transaction costs were used. However, Sweeney concluded
that the results were very sensitive to the transaction cost assumptions made
as well as biases in closing price listing
Elton and Gruber (1995) commented that Jennergren and Korsvold (1974) found
some evidence of profitable filter rules in the Norwegian and Swedish markets. We
59
have not succeeded in finding evidence to whether these opportunities still persist,
although one could assume that they are gone now, as the markets have grown
more liquid.
Often, there will be evidence both for and against a theory. A few of the most
serious attacks on market efficiency have come from a few very influential
articles. Elton and Gruber (1995) classified these tests under market
rationality. This is because they belong to a group of seemingly uncorrected
patterns. Under market efficiency, if a pattern is discovered, it should be
traded away. Market rationality deals with patterns that have persisted
through long time periods
Relative Strength Rules
A popular measure of stock performance is relative strength. A general relative
strength rule would measure today‘s stock‘s price pi,t
against its average
price during the past time period pi*. Following the definitions of Levy
(1967), relative strength would be defined as pi,t
/ pi*. A relative strength
strategy would hence imply buying the X% with the highest relative strength
ratio, and invest equal dollar amounts in them. Following this, sell the
securities whose relative strength rank is below a cast-out rank K7
, and use
proceedings to invest in the upper X%.
In his Ph.D. dissertation, described in Jensen and Benington (1969), Levy tested
this strategy on a sample. The results indicated abnormal returns, but lacked
a significance
60
test and proper risk adjustments. The best results were obtained with X=10%,
K=1608
and X=5%, K=1409
with corresponding returns of 20% and 26.1%,
unadjusted for risk. Buy-and-hold returns were given at 13.4%. Jensen and
Benington (1969) looked thoroughly into this investigation. After adjusting
for risk, the most profitable strategy earned 1.4% on the buy-and-hold
strategy, not including transaction costs. A further inclusion of transaction
costs would make the strategy unprofitable. Furthermore, Jensen and
Benington (1969) claimed that Levy‘s results were attributable to selection
bias, as the same data was used for all filters. They finally added that ex
post, filters could always be designed to fit a given data set.
61
EARLY MODELS
The fair game model
Bachelier (1900) presented the fair game approach in his dissertation Theory of
Speculation. In his framework, he distinguished between two sets of probabilities.
The first category he called mathematical probability, which is related to a priori
probabilities, as studied in games of chance. The second category consisted of
probabilities dependent of future events. These latter expectations, Bachelier
(1900) argued, are subjective, cannot be observed and should hence not be part of
the price. Hence, basing the prices on the a priori expectation, the prices would be
fair. In other words, the price should reflect the events known to the market, and
not the events that might occur.
Following this definition, the stock market should be based on security returns
having mathematical expectation equal to zero, using today‘s information in the
market. Consequently, all information should be included in the prices at all times.
Trading on insider information, however, violates this principle; hence the price
would be unfair.
Fama (1970), following what Bachelier once started, used a simple fair game
model to present the efficient market hypothesis. Defining the set Ωtas the
information known at time t, pj,t+1
as the price, and rj,t+1
as return of security j at
time t+1, he presented security prices as follows:
E[pj,t+1
| Ωt] = [1+ E(r
j,t+1| Ω
t)]p
j,t
This is the classical fair game model. A fair game is designed so that the players
know their expected gain or loss to playing the game before playing it. The excess
gain Ωt+1
from the return game can be defined as follows:
Ωt+1= rj,t+1-E[rj,t+1| Ωt] (2)
62
In accordance to Bachelier‘s condition for a game to be fair, the expected excess
gains E[ Ωj,t+1
| Ωt] should be zero. The consequences of this simple statement are
very important to efficient markets. If the market returns follow a fair game model,
then the best return estimate would be the expected return. Hence, there would be
no initiative to speculate, as the markets best estimate would already be reflected in
the prices. This is in harmony with Fama‘s (1970) definitions, as mentioned in
section 2.1. According to these definitions, if returns do not follow a fair game,
then the market is not efficient
Martingale and submartingale model
Let us now consider the price pj,t
of security j at time t. The stochastic process in
the stock prices where
E[pj,t+1
| Ωt] Ω p
j,t (3)
is a martingale, whereas the similar process
E[pj,t+1
| Ωt] Ωp
j,t (4)
is a submartingale. Basically, the submartingale could be considered as a
martingale with an upward drift. This would imply that Cov(rj,t+1
,rj,t
) is positive. As
long as this autocorrelation is included in Ωt, this does not violate the fair game.
We will refer to both as the martingale model, but note that they have some
differences. For the martingale model, we can define xj,t+1
as the excess price
difference, hence
xt+1
= pj,t+1
-E[pj,t+1
| Ωt] (5)
and E[xj,t+1
| Ωt] should be equal to zero. From (1) and (5) it is easy to deduce that if
prices follow a martingale, then returns2
follow the fair game. The opposite is not
necessarily true, however.
63
Tests conducted on the stock market prices have to a large extent been based on the
martingale model, due to its simplicity and testability features. If expected stock
prices do not follow the martingale model, this would mean that the stocks are, on
average, incorrectly priced, and arbitrage opportunities would exist. Under the
efficient market hypothesis, rational investors would exploit this opportunity until
it is gone, hence forcing the price back to the martingale model.
Random walk
The discussions on stock market behavior started with Kendall‘s (1953) discovery
that returns appear to follow random walks. The probability density function ft[ Ωt]
of the returns (r1,t+1
, r2,t+1
,…, rj,t+1
,…, rn,t+1
) follows a random walk if:
ft[r
1,t+1 , r
2,t+1 ,…, r
j,t+1 ,…, r
n,t+1] = f
t[r
1,t+1 , r
2,t+1 ,…, r
j,t+1,…, r
n,t+1| _
t] (6)
The return of security j in period t+1 is independent of the information available at
time t. Consequently, no good estimate based on today‘s information will explain
the future returns. A random walk requires the fair game. Hence, modifying (1), we
obtain
Et[r
j,t+1| Ω
t] = E
t[r
j,t+1] (7)
The extensions (or rather the restrictions) from the martingale model are numerous,
as the random walk is a stronger assumption than the fair game. Firstly, the random
walk implies that returns are identically and independently distributed (i.i.d).
LeRoy (1989) claimed this would make the random walk hypothesis far too
restrictive to be generated within a reasonably broad class of optimization models,
hence martingale models would be more tractable
Secondly, the random walk requires variance and skewness to be unaffected by
information. Poteba and Summers (1986) reported that even though there is weak
serial correlation in volatility, random volatility shocks will occur, and volatilities
64
change over time. LeRoy (1989) added to this discussion, arguing that security
prices are known to have variance patterns of quiet and turbulent periods, not in
agreement with the stable variance assumption. Lo and MacKinlay (1988) put the
evidence in direct relation to the random walk hypothesis. Conducting a variance
bounds test, they found significant serial correlation in stock prices. This would
imply a rejection of the random walk hypothesis.
Finally, the random walk has several other implications, such as price-value
correlation, equilibrium when traded (all information is incorporated), and
normality of price changes. Mandelbrot (1963) found that price change
distributions were not normally distributed, and suggested that these fat tails could
be explained by abandoning a finite variance assumption, and using a more fat-
tailed distribution
Schwert and Seguin (1990) claimed that the conditional distribution of security
returns is much more fat-tailed than the normal distribution, and indicated that the
normality assumption might be violated
65
2-SEMI-STRONG FORM EFFICIENCY OF THE MARKET
The semi-strong form of the market hypothesis suggests that the current price fully
incorporate all publicly available information. Public information includes not only
past price but also data reported in a company‘s financial statements, earnings and
dividend announcements, announced merger plans, the financial situation of
company‘s competitors, expectations regarding macroeconomic factors such as
inflation, interest rate, GDP, unemployment. Sometimes, the public information
does not have to be of a financial nature. For example, for the analysis of
pharmaceutical companies, the relevant public information may include the current
published state of research in pain relieving drugs.
The raison behind the semi-strong is that one should not be able to profit using
something that ‖everybody else knows‖ since the information is public.
Nevertheless, this assumption is far stronger than that of the weak-form. Semi-
strong form efficiency of the markets requires the existence of market analysts who
are not only financial economists able to comprehend implications of financial
information but also macroeconomists, experts adept at understanding processes in
product and input markets. Such skills may be difficult to acquire and it requires a
lot of time and efforts. In addition to that the public information may be relatively
difficult to gather and costly to process.
Some tests on the semi-strong
Quarterly Earnings
Jones and Litzenberger (1970) suggested that trading techniques based on data
from quarterly earnings beat the market and that the market does not adjust
immediately to these types of announcements. This slow adoption to unexpected
66
statements in the quarterly earnings is not consistent with the theory that security
prices adjust rapidly to the release of public information.
A later study by Jones and Latané (1979) drew approximately the same conclusion
as Jones and Litzenberger. Jones and Latané found that the adjustment to
unexpected earnings, published in for instance quarterly earnings, was relatively
slow, and that it extends beyond the earnings report date.
Predicting Cross-Sectional Returns
The cross-sectional approach is simple. Stocks are ranked, and sorted into
number of (usually five or ten) quantiles. For each of these quantiles, the returns
are regressed against risk, estimated by some model, and the results are
compared across the different sections. Hence the name cross-sectional
returns. Regressions are usually done using an approach defined in Fama
and MacBeth (1973). The goal is to determine if any factors can be used to
explain returns, after the adjustment for risk.
Studies gathered under cross-sectional return predictability and market efficiency
use a diversity of publicly available statistics, such as accounting data, to rank
stocks. Typical ranking criteria are accounting ratios such as Price/Earnings,
Book/Market value and size (market capitalization).
The cross-sectional tests involve a joint hypothesis problem, because they test not
only the EMH, but also the model used to estimate risk. The conclusions from tests
concerning these apparent anomalies have given rise to a discussion on the
adequacy of beta as a risk measure. If beta does not adequately describe risk, this
would have implications to the use of the CAPM, and favor APT models. Fama
(1991) pointed out that correlation observed between fundamental variables and
67
returns could be consistent with the EMH, provided they were proxies for omitted
risk factors.
Event studies
Event studies examine how the market responds to certain economic events and if
the stock prices adjust rapidly to these events. The type of economic events can be
either company specific or general for the economic or political situation.
3-STRONG FORM EFFICIENCY OF THE MARKET
The strong form efficiency states that the current prices of the security fully
incorporate or reflect all existing information; both public and private (sometimes
called insider information). So not only the public information which is useless to
the investor but all the information. The differences between the semi-strong form
and the strong form efficiency are that in the latter, nobody should be able to
systematically generate profits even if trading on information not publicly known
at the time. In other words, the strong form states that a company‘s management
(insiders) are not able to systematically gain from inside information by buying
company‘s shares ten minutes after they decided(did not publicly announce) to
pursue what they perceive to be very profitable acquisition. Similarly, the members
of the company‘s research department are not able to profit from the information
about the new revolutionary discovery they completed half an hour ago. The
rationale for strong form market efficiency is that the market anticipates in an
unbiased manner, future development and therefore the stock price may have
68
incorporated the information and evaluated in a much more objective and
informative way than the insiders.
Since the first event studies, numerous papers have demonstrated that early
identification of new information can provide substantial profits. Insiders who
trade on the basis of privileged information can therefore make excess returns,
violating the strong form of the efficient markets hypothesis. Even the earliest
studies by Cowles (1933, 1944), however, make it clear that investment
professionals do not beat the market. While there was evidence on the performance
of security analysts, until the 1960s there was a gap in knowledge about the returns
achieved by professional portfolio managers. With the development of the capital
asset pricing model by Treynor (1961) and Sharpe (1964) it became clear that the
CAPM can provide a benchmark for performance analysis. The first such study
was Treynor's (1965) article in Harvard Business Review on the performance of
mutual funds, closely followed by Sharpe's (1966) rival article. The most
frequently cited article on fund managers' performance was to be the detailed
analysis of115 mutual funds over the period 1955-64 undertaken by Jensen
(1968). On a risk-adjusted basis,he finds that any advantage that the portfolio
managers might have is consumed by fees and expenses. Even if investment
management fees and loads are added back to performance measures, and returns
are measured gross of management expenses (ie, assuming research and other
expenses were obtained free), Jensen concludes that "on average the funds
apparently werenot quite successful enough in their trading activities to recoup
even their brokerage expenses."Fama (1991) summarises a number of subsequent
studies of mutual fund and institutional portfolio managers' performance. Though
some mutual funds have achieved minor abnormal gross returnsbefore expenses,
pension funds have underperformed passive benchmarks on a risk-adjusted basis. It
69
is important to note that the efficient markets hypothesis does not rule out small
abnormal returns, before fees and expenses. Analysts could therefore still have an
incentive to acquire and act on valuable information, though investors would
expect to receive no more than an average ne return. Grossman and Stiglitz
(1980) formalise this idea, showing that a sensible model of equilibrium must leave
some incentive for security analysis.To make sense, the concept of market
efficiency has to admit the possibility of minor market inefficiencies. The evidence
accumulated during the 1960s and 1970s appeared to be broadly consistent with
this view. While it was clear that markets cannot be completely efficient in the
strong form, there was striking support for the weak and semi-strong forms, and
even for versions of strong form efficiency that focus on the performance on
professional investment managers
Some tests on strong form efficiency
Corporate Insiders
Some of the studies of corporate insiders, like Finnerty (1976), have examined
corporate insiders and tried to determine how well they do relative to the market in
general. Other studies, for example Rozeff and Zaman (1988), examined the
returns for the corporate insiders as well as returns for outsiders basing their
decisions on publicly available information concerning insider trades.
The aspect of trades from corporate insiders has been studied since the 1960s.
Rogoff (1964) concluded in his dissertation that insiders have an advantage over
outsiders when they are trading in their own companies. But he also stated that the
variability of the excess returns was too high to conclude that they had a persistent
advantage compared to speculators
70
Jaffe (1974) concluded that insiders did possess special information, but that the
excess profits were substantially affected when including transaction costs. He was
also aware of the limitations of his study, caused by the fact that it only contained
trades that had to be reported. Insiders could always leak information to others or
they could trade in companies that were affected by decisions in their own
company.
Finnerty (1976) published a study that investigated insider trades for the New York
Stock Exchange firms during the time period from January 1969 to December
1972. Based on a methodology developed by Jensen (1968), which described the
risk premium of an individual security above the market return, Finnerty (1976)
evaluated both buy and sell portfolios. For the buy portfolios, results were
evaluated for up to 12 months after the trade. The returns were consistently
positive and significantly different from zero. For the sell portfolios, results were
similar, indicating that securities sold by insiders fell more than the market in the
months following the trade. When comparing the post-trade months, the first
month showed the greatest abnormal return. This might have been caused by the
fact that the insiders‘ trades became public information soon after the trade. The
conclusion from this study was that insiders were able to identify profitable or
unprofitable situations in their own company in the short-run. Although his
methodology was somewhat different, Finnerty‘s (1976) results were aligned with
the results of Jaffe (1974), and both studies refuted the strong-form of the efficient
market hypothesis
Rozeff and Zaman (1988) considered both the potential excess returns for
corporate insiders as well as the possibilities for outsiders to make above average
returns based on publicly available information considering insider trades. A
possible above average return based on publicly available information will violate
71
the semi strong-form of the efficient market hypothesis. This violation is referred
to as the insider-trading anomaly, because it is based on insider trades. Rozeff and
Zaman (1988) also investigated whether positive abnormal returns linked to insider
trades could be explained by other factors such as size or P/E ratios. For the 1973 –
1982 period, an insider-trading anomaly was present. Although the excess returns
for shorter periods disappeared after including transaction costs (2%), the returns
for the 12-month horizon was significantly higher than the market return. When
adjusting the market model for the size and P/E effects, the excess return for the
12-month horizon became insignificant. These results suggested that the abnormal
returns experienced by outside traders, basing their activities on the trades of
corporate insiders, were mainly caused by size and P/E effects. For the insider
trades, the results were positive also after adjusting for both transaction costs and
the size and P/E effects. The study by Rozeff and Zaman (1988) indicated a profit
level of 3.12% per year. Furthermore, they suggested an upper limit of possible
insider trade profits of at most 5% annually.
The suggestion that the size effect is responsible for a great deal of the abnormal
returns is further supported by Gregory et al., 1994). Their findings supported the
conclusion that when you include the size effect, the abnormal returns for insider
trades will become insignificant.
Kara and Denning (1998) based their empirical test of the strong-form efficiency
on insider trades in the period ranging from March 1979 to July 1980. Based on
these data they indicate a rejection of the strong-form hypothesis
72
Stock Exchange Specialists
Because of the monopolistic access to certain information, one would intuitively
expect that stock exchange specialists are able to earn above-average returns.
Niederhoffer and Osborne (1966) found that specialists' monopolistic power over
information provided an edge for them to earn abnormal returns. A study by the
SEC (Securities and Exchange Commission) in the early 1970s also supported this
belief and contradicted the theory of strong-form efficiency. More recent studies
have showed that abnormal returns for specialists are decreasing
Security Analysts
Security analysts have experience and practice in finding undervalued stocks. The
studies of security analysts will therefore give a more meaningful approach to the
strong-form testing, independent of the possession of some specific position. The
evaluation of security analysts preceded the hypothesis of efficient markets.
Cowles (1933) concluded that neither the 16 financial services nor the 24 financial
publications engaged in forecasting the stock market were able to outperform a
random selection of stocks. Even the best records in each of the groups failed to
present results that were significantly better than random performance. Cowles
continued his study and reached the same conclusion in a later paper (Cowles,
1944). The results indicated that leading financial periodicals and services failed to
successfully predict the future course of the stock market.
Holloway (1981) investigated both buy-and-hold strategies and active trading
based on the Value Line ranking. The buy-and-hold strategy accepted the rank-1
stocks from Value Line at the beginning of the year. The portfolio is held
73
unchanged for one year and then adjusted based on the ranking for the next year.
This strategy gave abnormal returns even after the adjustment of transaction costs.
The active strategy version of the Value-Line ranking strategy changed the
portfolio consistently when the stocks weekly moved in or out of the rank-1
category. This strategy was able to earn positive abnormal returns without
transaction costs, but the returns were consumed when adjusting for trading.
Holloway (1981) was also taking into consideration whether the results were
explainable by a failure to adjust for risk. His investigation suggested no such
evidence.
Hunerman and Kandel (1990) tried to reconcile the Value Line record with the
efficient market hypothesis. They claimed that their model showed that Value
Line‘s rankings predicted systematic market-wide factors, and that the alleged
abnormal returns from positions based on Value Line were compensation for the
systematic risk associated with these positions.
Womack (1996) presented some evidence on the ability of analysts to predict or
influence stock prices. He evaluated buy and sell recommendations made by
analysts and concluded that they had a substantial impact on both an immediate
and a long-term basis. He claimed that this was consistent with the expanded view
on the efficient market hypothesis suggested by Grossman and Stiglitz (1980).
They advocated that there had to be returns to costs induced by information search.
74
Chapter 5
TECHNICAL ANALYSIS AND EFFICIENT MARKET HYPOTHESIS
The theory that past prices could be used to guide future investment decisions was
first formalised by Charles Henry Dow (1985-1902), the founder of Wall Street
and Dow Jones Company. His early work gave rise to a number of theories and
techniques which are now employed everywhere by investors in their trading
decisions.
The term technical analysis refers to a myriad of trading techniques. ―Technical
analyst attempt to forecast prices by the study of past prices and a few other related
summary statistics. They believe that shifts in supply and demand can be detected
in charts of market action‖ (Brock et al 1992).
A-PRINCIPLE OF TECHNICAL ANALYSIS
Neely in 1997 describes three principles which guide the behaviour of technical
analysts. The first is that ―market discounts everything‖. In other words, all prices
movement are ―a reflection of the trend in the hopes, fears, knowledge, optimism
and greed‖ [pring, 2002]. As Drew in 1968 noted that the price level is ―never what
they [stocks] are worth, but what people think they are worth‖. As such there is no
need to forecast the fundamental determinants of an asset‘s value. In fact Murphy
in 1986 claimed asset price changes often precede observed changes in
fundamentals.
The second principle states that the ―trend is your friend‖. Technical analysts
maintain that asset prices move in trends and typically do not believe that price
fluctuations are random and unpredictable.
75
Neely provides an example of what markets may be inclined to trend. First we
assume that there is asymmetric information in the market i.e. some participants
will have more information than others. It is likely that the trading behaviour of
these participants will reveal some of their private information to uninformed
participants. Let us assume that some investors have obtained some information
about the fundamental determinants of a company which is likely to push the price
of the share to go up. They will buy the share of the company which will raise the
price of the share as they purchase more.
When uninformed investors notice this behavior, they also will buy the same share
increasing more the price of the share as they also buy. This is known as self-
fulfilling prophecy. Individuals believe that the share price is going to rise, and
acting upon this belief, will cause the share price to rise independent of other
factors. Therefore, the market is predisposed to trends. Some practionners even
appeal to Newton’ law of motion to explain the existence of trend:‖trends in
motion tend to remain in motion unless acted upon by another force.‖
Predictable trends are essentials to a technical analyst. They are what allow traders
to profits by buying assets when the price is rising and selling when the price is
falling.
The third and final principle is that history repeats itself. The assumption is made
that the emotional make-up of investors will not change. As such asset traders will
tend to react the same way to the same market conditions and therefore, the
subsequent price movements are likely to be similar. ―Technical analysts do not
claim their methods are magical; rather that they take advantage of market
psychology‖[Neely, 1997]. The objective is therefore to uncover signals given off
in a current market by examining past market signals.
76
Technical analysis employs two main analyses: charting and Mechanical rules.
A-TECHNICAL ANALYSIS TECHNIQUES: CHARTING
Charting involves graphing the history of prices over some period determined by
the analyst to predict future patterns in the data based upon past patterns. This
method is much more subjective than the mechanical technique. It requires the
analyst to use judgment and skills in finding and interpreting patterns. It is the
process of establishing patterns within a series of data by visual inspection. To
identify trends it is first necessary to find the maximum called peaks and the
minimal also called troughs in the data. Peaks and troughs allow us to establish the
uptrend and the downtrend respectively.
Once technicians have identified the trend line, they will trade as necessary as
possible. Buying at one level and selling at another level.
Detecting the reversal of an established trend is just as important as identifying the
trend itself. If a trend reversal is indicated, a trader can profit by reversing hi/her
position. Peak and trough are also used for this purpose. Local peaks are called
resistance levels and the local troughs are called support levels. A price failing to
break a support level during a downtrend is thought to be an early indication that
the trend may soon reverse.
Another useful patterns identified the future movement of price are head and
shoulders which are considered by the technicians are the most reliable of all chart
pattern.
77
B-TECHNICAL ANALYSIS TECHNIQUES: MECHANICAL RULES
Mechanical trading rule avoids the subjectivity such fear or greed and provides a
consistent and disciplined approach to technical trading. Mechanical rules use
statistical rules in order to obtain buy and sell signals.
We can have three subcategories of mechanical rules namely: Filter rule, Moving
average rules and oscillators.
FILTER RULES
Alexander formulated the filter rules to test the belief held among the
professionals that prices adjust gradually to new information. For him the filter
rules means that ―if the stock price has moved up percent it is likely to move up
more than further before it moves down percent.‖
MOVING AVERAGE RULES
The moving average is a lagging indicator which is easy to construct and it is one
of the most widely used. A moving average, as the name suggests, represent an
average of a certain series of data that moves through times. The most common
way to calculate the moving average is to work from the last 10 days of closing
prices. Each day, the most recent close (day 11) is added to the total and the oldest
close (day 1) is substrated. The new total is then divided by the total number of
days (10) and the resultant average computed. The purpose of the moving average
is to track the progress of a price trend. The moving average is a smoothing device.
By averaging the data, a smoother line is produced, making it easier to view the
underlying trend. A moving average filters out random noise and offers a smoother
perspective of the price action.
78
OSCILLATORS
Oscillators are used as overbought or oversold indicators. A market is said to be
overbought when prices have been trending higher in a relatively steep fashion for
some time, to the extent that the number of market participants long of the market
significantly outweighs those on the sidelines or holding short positions. This
means that there are fewer participants to jump onto the back of the trend. The
oversold condition is just the opposite. The market has been trending lower for
some time and is running out of fuel for further price declines.
Oscillators indicators move within a range, say between zero and 100, and signal
periods where the security is overbought (near 100) or oversold (near zero).
Oscillators are the most common type of technical indicators. The technical
indicators that are not bound within a range also form buy and sell signals and
display strength or weakness in the market, but they can vary in the way they do
this.
79
FUNDAMENTAL ANALYSIS STRATEGY FOR PRDICTING THE STOCK
RETURNS
According the efficient market hypothesis, publicly available information cannot
be used to predict returns and systematically generate abnormal profits. According
to fundamental analysis, stock price reflects the firm‘s ability to generate positive
future earnings therefore to earn abnormal returns using historical accounting data,
researchers suggest that financial statements data should inform on firm‘s future
earnings performance.
OU & Penman in 1989, the pioneers in this research area, document the existence
of significant abnormal returns to a trading strategy that is based on the prediction
of the sign of future changes in annual earnings per share (EPS).
Piotroski 2000 provides evidence supporting the predictive ability of historical
accounting signals with respect to future stock price adjustments for a sample of
value stocks characterized by high Book-to-Market value (BM). They remark that
less than 44% of all high BM firms earn positive return. International Research
Journal of Finance and Economics - Issue 30 (2009).
Market-adjusted returns in two years following portfolio formation and suggest
that investors could benefit by discriminating ex-ante between the eventual strong
and weak companies. Using nine fundamental signals relating to three areas of the
firm‘s financial condition: profitability, financial Leverage, liquidity and operating
efficiency, they classify firms in ten portfolios depending on the signals‘
implication for future prices and profitability. Their results show that the mean
returns earned by a high book to market investor can be increased by at least 7, 5%
annually through the selection of financially strong high BM firms. Moreover, an
investment strategy that buys expected winners and shorts expected losers
80
generates a 23% annual return. Like AB 1997, 1998, their results support the
ability of historical accounting information to predict future firm performance and
the market‘s inability to recognize these predictable patterns.
FUNDAMENTAL SIGNALS
Twelve fundamental signals have been chosen from previous fundamental
research. The selection criteria of these signals have been relied on the predictive
ability with respect to future earnings performance and returns.
Each firm‘s signal realization is classified as either good or bad according to its
expected implication for future earnings and then prices. A binary variable is
created for each signal that is equal to one if the signal‘s realization is good and
zero if it is bad.
1-inventory (INV)
This ratio measures the percentage change in sales relative to the percentage
change in inventory. Inventory increases faster than sales is considered a negative
signal because it suggests difficulties in generating sales or the existence of slow-
moving or obsolete items that will be written off in the future. All these factors
affect negatively future earnings
2-Account Receivable (AR)
This signal represents the percentage change in sales less an analogous measure of
accounts receivable. A disproportionate increase in accounts receivable relative to
sales suggests cash collection difficulties and problems in selling the firm‘s
products. Also, it may likely provoke future earnings decreases as provisions of
future receivables experience an eventual increase. These arguments imply future
81
earnings decreases. So a positive value to this signal is considered good news. It is
attributed one (zero) to this ratio when it is positive (negative)
3-INVESTMENTS (INVES)
This signal measures the percentage change in capital expenditures relative to the
percentage change in sales 2. A negative value of this ratio is considered bad news
by analysts. In fact, capital expenditures are largely discretionary. So an unusual
decrease is generally suspect. This decrease may indicate managers‘ concerns with
the adequacy of current and future cash flows to sustain the previous investment
level. Moreover, analysts associate this decrease with a short-term managerial
orientation. The binary variable associated to this signal takes the value of one
(zero) if this signal‘s realization is positive
(Negative).
4-GROSS MARGINS (GM)
This signal is defined as the percentage change in gross margin less an analogous
measure for sales. A decrease in gross margin relative to sales is considered bad
news. In fact, it indicates a deterioration of the firm‘s terms of trade or lack of
cost‘s production control. According to Lev & Thiagarajan 1993, variations in this
fundamental signal affect long-term performance of the firm and it is therefore
informative with respect to earnings persistence and firm values
5-LABOUR FORCE (LF)
This ratio is defined as the change in sales per employee. It is used by analysts to
appreciate the effects of restructuring decisions. In fact, decisions of labor force
reductions are generally associated with increases in wage related expenses that
82
affect negatively current earnings but bode well for future earnings. That‘s why a
similar signal is generally used to better assess future earnings.
6-7- RETURN ON ASSETS (ROA) AND VARIATIO IN RETURN ON ASSETS
(ΔROA).
ROA is defined as net income before extraordinary items scaled by beginning of
the year total assets and ΔROA is defined as the current year‘s ROA less the prior
year‘s ROA.
8-CASH FLOW
This ratio is defined as cash flow scaled by beginning of the year total assets.
Where cash flow = Earnings before extraordinary items – Accruals, And Accruals
are defined as is common in the earnings‘ management literature (Dechow, Sloan
& Sweeney 1995, Sloan 1996):
Accruals = (ΔCA - Δ Cash) - ΔCL – Dep
Where: ΔCA: change in current assets
Δ Cash: change in cash/cash equivalents
ΔCL: change in current liabilities
Dep: depreciation and amortization expense
According to Piotroski 2000, the three preceding signals provide information about
the firm‘s ability to generate funds. Firm that realizes positive earnings or cash
flow is demonstrating a capacity to generate some funds through operating
activities. Similarly, a positive earnings trend is suggestive of an improvement in
the firm‘s underlying ability to generate positive future cash flows. So, the binary
83
variable associated to each signal equals one if the signal‘s realization is positive
and zero otherwise
9-ACCRUALS (ACC)
This signal is defined as accruals scaled by beginning of the year total assets. This
signal measures earnings persistence. Authors like Sloan [1996], Bradshaw,
Richardson & Sloan [1999] and Collins & Hribar [2000] show that persistence of
current earnings is decreasing in the magnitude of the accruals component of
earnings. That is firms with high accruals are more likely to experience declines in
subsequent earnings performance and others with low accruals are more likely to
experience subsequent earnings increase.
Consistent with the negative relation between accruals and future earnings, I
attribute the value of one to the binary variable if the accruals‘ signal is negative
and zero if it is positive.
10-LEVERAGE:
This signal represents change in the ratio of total long-term debt to average total
assets. Literature review interested on financial leverage‘s impact on firm value
has been controversial. Theoretical studies leaded for example by Harris & Raviv
[1990], Stulz [1990] and Hirshleifer & Thakor [1992] show that leverage is
positively related to firm value. However, Myers & Majluf [1984] and Miller &
Rock [1985] consider that leverage increases can negatively impact firm value.
84
11-LIQUIDITY (ΔLIQUID)
This signal measures the change in the firm‘s current ratio between the current and
prior year. current ratio is defined as the ratio of current assets to current liabilities
at the fiscal year end. Liquidity increases signals the firm‘s ability to service
current debt obligations. So the binary variable attributed to this signal equals one
if ΔLIQUID is positive and zero if it is negative.
12-ASSETS TURNOVER (ΔTURN)
This signal is defined as the firm‘s current year asset turnover ratio less the prior
year‘s asset turnover ratio. The asset turnover ratio is defined as total sales scaled
by beginning of the year total assets. A positive ratio may imply improvement in
asset‘s use that is less assets generating the same levels of sales, or an increase in
firm‘s sale.
MARKET ANOMALIES
A market anomaly is any event or time period that can be used to produce
abnormal profits on stock markets. Stock market anomalies occur across the world.
They do not correspond with existing equilibrium models, where risk is the only
factor which is likely to cause possible variations in stock market excess returns.
The occurrence of pattern in time series of stock market returns, independent of
time varying risk, would indicate that not all relevant information is captured in
stock prices, which is inconsistent with the efficient market hypothesis. Stock
market anomalies exist in every form of the efficient market hypothesis and can be
classified in different categories, like for example firm anomalies, accounting
85
anomalies, events anomalies, weather anomalies and calendar anomalies.(levy and
Post 2005)
1-FIRMS ANOMALIES: are consequences of firm-specific characteristics (levy
and post 2005). One well-known firm anomaly is the SIZE EFFECT, which states
that returns on small firms are higher compared to returns on large firms, even after
risk-adjustment. Banz in 1981 discovered this size effect especially for the smallest
firms in his sample based on total market value of NYSE from the period 1936-
1975. Kein in 1883 presented the same conclusion for NYSE and AMEX firms in
the period 1963-1979. The research of Banz in 1981 and Keim in 1983 in relation
to the JANUARY EFFECT, is discussed in the later section. Another firm
anomaly is the effect that firms which are followed by only a few analysts earn
higher returns. This effect is known as NEGLECTED FIRM EFFECT. Arbel,
carvel and srebel in 11983 looked at 510 firms from the NYSE, the AMEX and the
over-the –counter markets and divided them into three groups of institutional
holding (intensively held, moderately held and institutionally neglected) and three
groups of size (small, medium and large). For the period 1971-1980 they found
that the neglected firms earn significantly higher returns than firms intensively held
by institutional investors for both the small and the medium size firms.
2-Accounting anomalies: accounting anomalies relate to stock price movement
after the release of accounting information. An example of an accounting anomaly
is the earning momentum anomaly, which implies that firms with a rising growth
rate of earning are likely to have stocks that outperform the market. Another
accounting anomaly is that if the market-to-book value (M/B) ratio is low, the
stocks are likely to outperform the market.(levy and Post 2005). This phenomenon
86
is investigated by fama and French In 1992. They divided their total sample of
stocks on the NYSE, AMEX and NASDAQ into ten groups based on M/B ratio
and found that the group with lowest M/B ratio had an average monthly return of
1.65%, while the group with the highest M/B ratio only had an average monthly
return of 0.72%.
3-EVENT ANOMALIES: relate to price movements after an obvious event. This
can be for example the announcement that a firm will be listed on a major stock
exchange. After such an announcement, the price of the stock rises. The
recommendation of an analyst is another example of an event anomaly. Depending
on the type of recommendation, the stock price will rise or fall (levy and post
2005).
4-WEATHER ANOMALIES: relate to price changes during certain weather
conditions. Yuan, Zhen and Zhu in 2006 for example, find a relationship between
stock returns and lunar cycles, looking at stock indices of 48 countries around the
world for the period of January 1973 to July 2001. Their condition is that stock
returns are higher on days around a new moon compared to days around a full
moon. Furthermore, they looked at other explanations for this lunar effect, like
macroeconomic announcements, trading volumes, returns volatility and other
anomalies, but none of them appear to be valid. Another investigation concerning
weather anomalies comes from Saunders in 1993. He explores whether the stock
returns on the Dow-jones industrial average and NYSE/AMEX for the period of
1927-1989 are affected by weather conditions. His results suggest that the weather
does not have significant influence on the stock market returns. This is especially
the case for 100% cloudy days and for sunny days (with 0-20% clouds), where the
mean return for the latter group differs most from the overall mean for all days.
Saunders (1993) states that his results are robust to other anomalies like the
87
January effect, the weekend effect and the size effect. Cao and Wei (2005)
investigate the possible relationship between stock market returns and temperature.
They test whether lower temperatures lead to higher stock market returns due to
aggression and therefore risk-taking and higher temperatures lead to higher or
lower returns depending whether aggression (which causes risk taking) or apathy
(which causes risk-aversion) dictates. Returns on nine stock market indices around
the world between 1962 and 2001 are used. Overall, Cao and Wei (2005) find that
stock returns are significantly negatively correlated to temperature.
The four categories of stock market anomalies discussed above are not further
captured in this research for several reasons. First, in general, there is considerably
less research on these anomalies compared to calendar anomalies like the January
effect, the day-of-the-week effect, turn-of-the-month effect and Halloween effect
and the theoretical foundation for these anomalies is far from solid.
Besides that, for some of the anomalies, like the weather anomalies, the data are
more difficult to obtain and are beyond the scope of this project.
2.3 Calendar anomalies
If a stock market anomaly depends solely on certain periods in a calendar year, it
refers to a calendar anomaly. The followings are calendar namely the January
effect, the day-of-the-week effect, the turn-of the- month effect and the Halloween
effect.
2.3.1 January effect
The January effect is one of the most well-known anomalies. In 1976, Rozeff and
Kinney reported seasonality in stock returns, using monthly rates of return of the
New York Stock Exchange from 1904-1974. The seasonality that they found was
mainly caused by large returns in January, compared to the remaining months of
88
the year. In their research they focused on the existence of seasonality, they did not
test possible explanations for it. Later on, others looked for possible explanations
of the January effect. A selection of those articles is described in this section,
categorized by their main Explanation for the January effect.
The size effect
Banz (1981) discovered that risk-adjusted returns are higher for small firms than
for large firms based on total market value by using return data from stocks on the
NYSE from 1936 – 1975. By defining sub-periods of ten years, he found that this
‗size effect‘ is not a linear function, so this does not mean that the returns increase
when firm size decreases. Nevertheless, for the smallest firms in his sample, the
effect is strongest. Keim (1983) also investigated the negative relation between
firm size measured in total market value of equity and abnormal risk-adjusted
returns. With data of firms on the NYSE and AMEX from 1963 – 1979, he defined
different portfolios based on firm size and found that the smaller the firm size, the
more excess return increased. Furthermore, his results showed that this effect is
much stronger for January than for the remaining months of the year. In his further
research, Keim (1983) found that approximately half of the size effect is caused by
January returns and roughly a quarter is caused by the first five trading days of
January. The findings of Banz (1981) and Keim (1983) suggest that the January
effect found by Rozeff and Kinney (1976) should be more pronounced on small
capitalization indices.
4-DAYS-OF –THE-WEEK EFFECT
The day-of-the-week effect comes down to the difference in returns between a
particular trading day a couple of trading days compared to the rest of the trading
week. The weekend effect focuses on Monday and Friday returns, stating that
89
Monday returns are low and negative and Friday returns are high compared to the
remaining trading days of the week. Extensive research has been done to find
explanations for this anomaly. Here are the most important research that was made
in the last few decades regarding the day-of-the week effect and the weekend
effect.
Cross (1973) was one of the first to report differences in returns on Fridays and
Mondays compared to the rest of the week. With daily return data from 1953 –
1970 on the S&P 500, he found a statistically significant difference between Friday
and Monday returns for almost every year in the sample period in both mean
returns and in percentage of time that the index rose on that day. Moreover, his
results showed that Monday indices following a decline on Friday rose in
approximately 24% of the cases, which is significantly different from the reaction
of the remaining trading days of the week following a decline of the previous
trading day. Subsequent to Roll (1973), French (1980) reported that Monday
returns were negative and lower than returns for other days in the week, using daily
returns from the S&P 500 composite portfolio for the period 1953 – 1977. In his
research, he examined if this is the consequence of the weekend prior to each
Monday or the consequence of every non-trading day (Holiday). His results
illustrated that the return on every weekday on its own (with the exception of
(Tuesday) is higher when the day follows a holiday compared to when the day
follows a trading day.
This is different for Tuesday, because Tuesday is the first trading day after the
weekend when Monday is a holiday. Therefore, French (1980) concluded that the
negative Monday returns are caused by a weekend effect.
Cross (1973) and French (1980) did not look for possible explanations for the
weekend effect, although other authors did. Below a selection is given of the
articles of those authors and their findings regarding the weekend effect.
90
Settlement period
By using mean returns and variances for the S&P 500 and the CRSP value- and
equally-weighted portfolios from July 1962 – December 1978, Gibbons and Hess
(1981) also found negative Monday returns, although no Monday effect in
variances. Moreover, they searched for possible explanations for the Monday
effect. The settlement period explains the more negative Monday returns before
1968 compared to the returns after 1968, because the settlement period was four
business days before February 1968 and five business days after February.
Nonetheless, it does not explain the negative Monday returns from February 1968
– December 1978. Apart from that, the settlement period is nowadays three
business days for stocks.
Release of information
French and Roll (1986) investigated the return variances of weekdays, weekends,
holidays and holiday weekends by means of daily returns on the New York and
American Stock Exchanges from 1963 –1982. They found that the returns are more
volatile during exchange trading hours compared to non trading hours. The three
possible explanations that were given for this are that public information (Which
causes the volatility) is announced more frequently during business days
(weekdays), private information probably influences prices more when the stock
markets are open and the process of trading itself causes volatility. French and Roll
(1986) concluded that their results showed that only a small part of the difference
in variances between trading hours and non-trading hours is caused by mispricing
occurring during trading. The reason for this mainly lies in the difference in
quantity of information announced between trading hours and non-trading hours.
91
The days-of-the-week effect internationally
Jaffe and Westerfield (1985) contributed to the literature by investigating the day-
of-the-week effect internationally. In all countries in their research, namely the
U.K., Japan, Canada and Australia, they found a day-of the-week effect with
significant negative Monday returns and high Friday returns.
Correlations tests suggest that there is a strong correlation between the returns of
the four foreign countries and those of the U.S. Nonetheless, independent of the
day-of-the-week effect in the U.S. Jaffe and Westerfield (1985) found a day-of-the-
week effect in each of the four countries on their own. Similar to Gibbons and
Hess (1981), they looked at settlement periods to explain the day-of-the-week
effect. They only found little evidence for the higher Thursday and Friday returns
in Australia, but for Canada, the U.K. and Japan the settlement period did not
explain the day-of-the-week effect at all. In addition, Jaffe and Westerfield (1985)
investigate the opportunity that measurement errors cause the day-of-the-week
effect. They state that if Monday returns would be influenced by mainly negative
random errors and Friday returns by mainly positive random errors, the correlation
between Monday and Friday returns should be low. Nonetheless, they found a
higher than average correlation between Monday and Friday returns and therefore
conclude that measurement errors cannot explain the day-of the-week effect.
In their exploration of different anomalies on stock market indices in eighteen
countries across the world, Agrawal and Tandon (1994) captured the day-of-the
week effect as well. They found negative returns on Mondays for thirteen countries
(of which seven are statistically significant), but also negative returns on Tuesdays
in twelve countries (of which eight are statistically significant). Furthermore, they
found Tuesday returns to be lower compared to Monday returns in eight countries.
Contrary to these negative Monday and Tuesday returns, they revealed positive
Wednesday and Friday returns in the majority of the countries. After reporting the
92
day by day returns, Agrawal and Tandon (1994) discussed possible explanations
for the negative Tuesday returns. They stated that the time zone hypothesis (which
argues that Tuesday returns are low in some countries due to time-differences that
exceed twelve hours) can explain the negative Tuesday returns in three of the five
countries, but cannot explain the negative Tuesday returns in European countries.
Furthermore, the difference between trading days and non-trading days is not an
explanation for the negative Tuesday returns. After running day-of-the-week
correlation tests and regressions, the null hypothesis that day by day variances are
dependent on the US can be rejected for the majority of the countries. Moreover,
Agrawal and Tandon (1994) argued that the settlement procedure explains a part of
the day-by-day differences (mainly the higher returns on Wednesday, Thursday
and Friday) in returns, but cannot explain the negative Monday and Tuesday
returns for most of the countries. Furthermore, they divided their total sample
period into two sub-periods and found that in the seventies Monday returns are
significantly negative in seven countries and Tuesday returns are significantly
negative in nine countries, while in the eighties the Monday and Tuesday returns
are not significantly negative in the majority of the countries. Finally, they found
that Monday returns are negative in almost all countries if the market declined the
previous week, but mainly positive when the market went up the previous week.
However, this is not found for Tuesday returns.
4-TURN-OF-THE-MONTH EFFECT
The turn-of-the-month effect refers to higher stock market returns for the trading
days surrounding the turn of the month compared to the remaining days of the
month. In the past decades, many researchers discussed the turn-of-the-month
effect, all with their own approach. By looking at the stock returns of the CRSP
93
value-weighted and equally-weighted index for the period (1963 – 1981, Ariël
(1987) concluded that the mean cumulative return of the first half of trading
months is significantly higher than the mean cumulative return of the second half
of trading months. For this, he added the last trading day of the month to the first
half of the following month. In the entire period he examined, the mean return of
the first half of the month was positive and the mean return of the last half of the
month was equal to zero. At five percent significance level, he did not find
unequal variances between the first half of trading months and the last half of
trading months. This effect is not induced by outliers either. After reporting this
effect, Ariël (1987) looked for possible causes of the difference in mean returns
between the two periods. He reported that the effect is not caused by a
concentration of dividend payments, holidays or weekends in the first or last half
of the month. Furthermore, the effect still held after correcting for returns in
January and therefore was not caused by the January effect. Moreover, the effect is
not induced by a few months where the effect is strongest.
5-halloween effect
The Halloween effect refers to lower stock returns during the period May –
October compared to the returns during November – April. It was first presented
by Bouman and Jacobsen (2002). They concluded, using monthly stock returns of
value-weighted market indices of 37 countries for the period January 1970 –
August 1998 that the Halloween effect is present in 36 countries and is particularly
economically significant in many European countries. After using the regression
technique, Bouman and Jacobsen (2002) concluded that there is a statistically
significant Halloween effect at 10-percent level in 20 of the 37 countries. These
94
results, they said, are also robust over time. Besides finding the Halloween effect
in their sample, they looked for possible explanations for it.
After running different tests, they rejected data mining, difference in risk, the
January effect, differences in interest rates and trading volumes as possible
explanations. Furthermore, they did not find any seasonal effect in the news which
could explain the Halloween effect. While Bouman and Jacobsen (2002) did not
find an explanation for the Halloween effect; other articles did present possible
explanations.
BEHAVIORAL FINANCE AND EFFICIENT MARKET HYPOTHESIS
The behavioral finance is a new area of financial research that explores the
psychological factors affecting investment decisions. It attempts to explain market
anomalies and other market activity that is not explained by the efficient market
hypothesis.
The fundamental basis of behavioral finance is that psychological factors, or
behavioral biases, affect investors, which limits and distorts their information and
may cause them to reach incorrect conclusions even if the information is correct.
Behavioral Biases
Behavioral biases can be categorized in many ways, and many of these categories
may be overlapping or indistinguishable. The following factors have been
published as having some impact on the market. Contrarian and momentum
strategies take advantage of some of these psychological factors.
95
Emotional Factors
Emotion often overrules intelligence in decisions and can filter facts, where the
investor may give too much weight to facts that are agreeable and may tend to
ignore or underweight facts that are antithetical to one's predisposition.
There is the fear of regret, or regret avoidance, which causes investors to hold
losing positions too long in the hope that they will become profitable or sell
winners too soon to lock in profits lest they turn into losses.
Overconfidence in one's abilities can lead to higher portfolio turnover and to lower
returns. Despite the fact that few investors or even professional portfolio managers
beat the market over an extended time, there is still a considerable amount of active
portfolio management. One study has shown, on the average, that men have lower
returns than women because they trade more actively, presumably because they
have greater confidence in their abilities.
Conservatism is the hesitation of many investors to act on new information, where
such information should dictate an action or a change to one's strategy. This
hesitation to news, for instance, eventually leads to action at different times for
different investors, leading to momentum as more investors start reacting to the
news.
Belief perseverance is the persistence of one's selection process and investment
strategies even when such strategies are failing or are suboptimal, causing the
investors to ignore new information that may contradict one's decisions.
Loss aversion is the propensity for people to avoid losses even for possible gains.
Hence, people often hang onto losing stocks longer than they should, since selling
96
would actualize the loss; otherwise, it is still just a "paper loss". Loss aversion is
related to the marginal utility of money, where the 1st dollars are more valuable
than additional dollars. For instance, most investors would avoid an investment
where there was a 50% chance of either earning $50,000 on a $100,000
investment, or an equal chance of losing the same amount, because the $50,000
lost would have greater marginal utility, and therefore be more valuable to the
investor, than the $50,000 potentially gained.
Misinformation and Thinking Errors
Misinformation and thinking errors probably account for most market
inefficiencies, since no one knows everything and, even if they did, they may not
make the best decisions based on that information.
Forecasting errors are a typical example, since even stock analysts are frequently
wrong about future earnings and prospects for the companies that they cover.
Representativeness is the extrapolation of future results based on a limited set of
observations or facts; hence, this is sometimes known as small sample neglect.
This is best illustrated by investors seeking a fund in which to invest, and basing
their decision on the fund's most recent performance rather than covering a longer
period of time, especially during bear markets. Often, this is the result of the lack
of due diligence, where the investor relies on brochures or other advertising
materials put out by the company or fund instead of doing independent research.
Narrow framing is the evaluation of few factors that may affect an investment. For
instance, an investor may buy a security because of its past performance without
considering economic factors that may be changing that could change the
performance of the security. Mental accounting is a specific kind of narrow
97
framing in which different investments are mentally segregated, applying different
criteria and due diligence to different investments where the different consideration
may be unwarranted. For example, treating separate investments as if they were the
only investment rather than as part of one's portfolio.
Biased information gathering and thinking is the distortion of personal bias on
facts and thinking to conform with one's current opinions or actions.
Limits to Arbitrage
So if behavioral bias misprices stocks, why don't arbitrageurs take advantage of the
mispricing, by buying and selling until the mispriced stocks are equal to their
intrinsic value? Behavioral advocates have argued that mispricing persists because
there are limits to arbitrage.
First, there is the risk that the mispricing will get worse, and, therefore, present a
risk to arbitraging. The market influence of behavioral bias may be greater than the
influence of arbitrageurs, especially since arbitraging behavioral bias is not risk-
free. And even if prices do trend toward their intrinsic prices, it may take longer
than arbitrageurs think—maybe longer than their investment horizon.
Secondly, financial models may be inaccurate, making arbitrage risky.
Thirdly, much of the arbitrage activity would require short selling, which, with
stocks, is very risky, and many institutional investors are not permitted to sell
short.
98
Herding Behavior
An explanation for momentum strategies is called herding behavior. Banerjee
(1992) defined herding as doing what everybody else is doing, even when evidence
(i.e. private information) implied it might be cleverer not to. His theory was
supported by the momentum patterns discovered in the empirical research of
Jegadeesh and Titman (1993). Grinblatt, Titman and Wermers (1995) investigated
mutual fund behavior, finding that funds did have a tendency of buying past
winners, and, more specifically, engaged in herding. The evidence was not,
however, particularly strong. Since institutional holdings comprise a large
percentage of total equity holdings, at least in the US, this is, at least partially,
compelling evidence for a momentum factor. Herding behavior, as described,
should result in intermediate term momentum, and a long-term reversal pattern,
and its effects are hence, very similar to the theories of noise trading.
Why do people engage in herding? As commented in section 6.2.1, fund managers
might engage in window dressing to take care of agency problems. Furthermore,
tax-loss selling might explain some of the herding behavior. Herding is not,
however, consistent with Markowitz-rationality.
As already mentioned, Black (1986) claimed that true value is not observable.
Hence, it would be difficult to measure whether a stock price really is in
equilibrium. A direct test of herding might therefore be difficult to design.
However, the momentum effects of Jegadeesh and Titman (1993) could be
interpreted as evidence supporting crowd Of course, true fundamental value is not
known (Black 1986).
As already mentioned, Black (1986) claimed that true value is not observable.
Hence, it would be difficult to measure whether a stock price really is in
99
equilibrium. A direct test of herding might therefore be difficult to design.
However, the momentum effects of Jegadeesh and Titman (1993) could be
interpreted as evidence supporting crowd
overreaction. Furthermore, DeBondt and Thaler‘s (1985) results on long-term
reversals might also indicate that there has been an overreaction. In retrospect,
however, it is easy to use herding to explain under- and over-reactions. The counter
argument could be that the theory was fitted to explain empirical results, hence
indicating a data-mining bias. The persistence of the effect is one of the serious
challenges to the EMH.
100
Chapter 6
DATA AND RESEARCH METHODOLOGY
The monthly price indices are examined for the empirical analysis from Ghana
stock market for a period of two years starting January 2007 to December 2008.
The whole sample includes 24 monthly observations. This sample takes in account
all the trading days except the holidays and the weekends.
Some information was collected through direct interview and questionnaires with
brokerage houses which are buying and selling in the Ghana stock market.
Other information was collected through direct interview with the research
department of the Ghana stock exchange and through reading materials in the
library of the Ghana stock exchange.
RESEARCH METHODOLOGY
This paper investigates the weak form efficiency of the Ghana stock market by
employing a statistical test or regression test. Indeed, the hypothesis that stock
market price index follow a random walk is tested. Accordingly to the null and
alternative hypotheses for weak form efficiency test are:
Ho = Ghana stock market‘s price indices are weak form efficient
H1 = Ghana stock market‘s price indices are not weak efficient or are inefficient
101
TABLE 1
Data
MONTH NUMBE
R
GSE ALL
SHARE
INDEX
JANUARY 1 5012.16 MEAN 7559.5762
5
FEBRUAR
Y
2 5044.89 MEDIAN 6664.385
MARCH 3 5092.25 STD
DEVN
2368.7738
92
APRIL 4 5139.65 MAX 10890.8
MAY 5 5224.47 MIN 5012.16
JUNE 6 5294.58 SKEW 0.3708870
63
JULY 7 5341.76 KURTOS
IS
-
1.7249105
55
AUGUST 8 5557.38
SEPTEMB
ER
9 5676.77
OCTOBER 10 5839.62
NOVEMBE
R
11 6387.16
DECEMBE 12 6599.77
102
R
JANUARY 13 6729.00
FEBRUAR
Y
14 7005.29
MARCH 15 7848.14
APRIL 16 9349.59
MAY 17 9815.22
JUNE 18 10346.3
JULY 19 10650.72
AUGUST 20 10790.95
SEPTEMB
ER
21 10890.8
OCTOBER 22 10788.29
NOVEMBE
R
23 10573.43
DECEMBE
R
24 10431.64
103
REGRESSION
1-THE REGRESSION ANALYSIS
Regression analysis is a statistical tool for the investigation of relationship between
variables. Usually the investigator seeks to ascertain the causal effect of on
variable upon another. The investigator also assesses the statistical significance of
the estimated relationships, that is, the degree of confidence that the true
relationship is close to the estimated relationship.
Steps in regression analysis:
1-State the hypothesis
If there is a significant relationship between the variables, the slope will not be
zero
Ho: B1 = 0
Ha: B1≠ 0
The null hypothesis states that the slope is equal to zero and the alternative
hypothesis states that the slope is not equal to zero.
2-Formulate an analysis plan
The analysis plan describes how to use the sample data to accept or reject the null
hypothesis. The plan should specify the following elements.
a- Significance level. Often researchers choose significance levels equal to
0.01, 0.05 or 0.10, but any value between 0 and 1 can be used.
b- Test method. Use a linear regression t-test to determine whether the slope of
the regression line differs significantly from zero.
104
3-Analyze sample data
Using the sample data, find the standard error of the slope, the slope of the
regression line, the degrees of freedom, the test statistics, and the p-value
associated with the test statistic.
a- Standard error. It could be referred to as the standard deviation
b- Slope. Like the standard error, the slope of the regression line will be
provided.
c- Degrees of freedom. The degrees of freedom(DF) is equal to:
DF = n – 2
Where n is the number of observations in the sample.
d- Test statistic. The test statistic is a t-test (t) defined by the following
equation.
t = b1 / stdev
Where b1 is the slope of the sample regression line, and the stdv is the standard
deviation of the slope
e- P-value. The p-value is the probability of observing a sample statistics as
extreme as the test statistics. Since the test statistic is a t-test, use in the
t-distribution to assess the probability associated with the test statistic. Use
the degrees of freedom to compute above
4-Interpret result if the sample findings are unlikely, given the null hypothesis, the
researcher rejects the null hypothesis. Typically, this involves comparing the p-
value to the significance level, and rejecting the null hypothesis when the p-value
is less than the significance level
105
Table 2
REGRESSION TABLES
Regression
Statistics
Multiple R 0.948715909
R Square 0.900061876
Adjusted R
Square
0.895519234
Standard Error 765.6702638
Observations 24
106
Table 3
ANOVA table
ANOVA
df SS MS F Significance
F
Regression 1 116157543.3 116157543.3 198.1362124 1.75272E-
12
Residual 22 12897520.96 586250.9529
Total 23 129055064.3
107
Table 4
INTERPRET REGRESSION COEFFICIENT TABLE
Coefficients Standard
Error
t Stat P-value Lower 95% Upper
95%
Lower
95.0%
Upper
95.0%
Intercept 3586.885109 322.6155376 11.11814123 1.69684E-10 2917.821437 4255.949 2917.821 4255.949
X Variable 1 317.8152913 22.57838429 14.07608654 1.75272E-12 270.9905884 364.64 270.9906 364.64
108
Testing the significance or efficiency
Hypothesis
Ho : r = 0 (There is no correlation between the prices or )
Ha : r ≠ 0 (There is correlation between the prices)
Where : r = 0.900061876
n = 24
t =
t = 9.688002
Degree of freedom (Df) = n-2 =24-2= 22. This is because of two tailed test.
Using a significance level (α) of 5%. Therefore the t critical from a t-table is listed
below:
t α,(n-2) = t 0.05,(24-2) = t0.05,22 = 1.717
T = 9.688002 and the DF = n-2 = 24-2 = 22
109
α = 0.05 and the value in the table is 1.717
RESULTS
The test of regression was performed on two years trading result from the Ghana
stock exchange all-share index to determine whether the market is weak form
efficient or to see whether the prices can be predicted. The t-test statistic for the
slope was significant at 0.05 critical level. The t (24) = 9.688 and the p = 1.717.
Thus we reject the null hypothesis and conclude that the market is inefficient
which means that the prices can be significantly predicted. Also if the value of the
test statistic is greater than 1.717 or less than – 1.717, we reject the null hypothesis
(Ho), that there is no correlation and we accept the alternative hypothesis that the
correlation is significant which means that the prices are predictable and the
market is inefficient in short term. From the result, it concludes that the regression
coefficient is significant; this also indicates that the market is inefficient and the
market is predictable from the above result with margins of error.
110
REASONS BEHIND INEFFICIENCY
One of the main purpose of the main inefficiency in the Ghana stock market is that
most of the investors are not educated enough concerning the pattern of the
market, meaning investors lack adequate knowledge to verify quality shares based
on fundamentals analysis and consequently carrying the transaction or trading
based on speculation or rumor only. The flow of information is not asymmetric to
all the investors at a time resulting into an advantage to a group of investors
acquiring the information early ahead to make abnormal profit.
Another issue is the inequality between the demand and the supply of shares in the
market. First lack of quality shares instigates instability in the market. Second
reason could be the approval of 0% income tax on capital gain encourages
investors to invest in the market which increases the demand where as the supply
remains the same.
Another reason could be the volume of shares traded in the market and also the
size of the market. In addition to the small number of participants and products in
the market, there is also lack of adequate research to improve the efficiency of the
stock market.
There is lack of professionalism in the part of the brokers some of which are only
graduated from the Ghana stock exchange.
111
SOLUTIONS FOR THE MARKET EFFICIENCY
In order to make the market efficient, investors must believe that the market is
inefficient and that it is possible to outperform it or beat the market. Ironically,
investment strategies intended to take advantages of the market inefficiency are the
fuel that keep the market efficient.
The market must be large and liquid and must also have large number of
participants.
In addition to that the information must be available to all the investors in the
market.
Plus the transaction cost must be cheap as compared to the expected returns and
investors must have enough funds at their disposal.
112
Chapter 7
CONCLUSION AND RECOMMENDATIONS
CONCLUSION
Referring to the test that has been conducted, it is concluded that the Ghana stock
exchange is not efficient at its weak form. Therefore there is a need to take actions
to improve the efficiency of the market. First, it is important to ensure symmetric
information among all the investors. In addition to the existing awareness creation
policy it is also important to improve on continuous basis to enlighten the investors
about market structure, trading pattern, financial analysis of the listed companies.
Proper implications of the rules of regulatory commission are also needed to be
ensured so that there will not be any scope to violate the market structure to gain
abnormal profit. It is also important that the regulatory commission enforces
policies to ensure improved quality of the stock market. Finally to make more
efficient it is expected that the authority of the market would introduce
sophisticated means of investment and tools in the near future and above all the
campaigns to sensitize people about the importance of the stock market and how to
invest in the stock market and what are the benefits.
113
RECOMMENDATION FOR FURTHER RESEARCH
As a result of the inefficiency of the Ghana stock exchange, the following points
should be considered for future research.
1-The Ghana stock exchange should be opened to foreign investors.
2-The foreign investors holding in the GSE should be increased
3-Improve on the clearing and settlement system
4-Improve on the volume traded and participants on the market through the listing
requirement.
5-Campaign to sensitize investors about the importance of the stock market.
6-improve professionalism on the part of the brokers through organizing training
7-Strategies to improve the role of information to make the make efficient
8-High cost of listing on the stock exchange
9-Small float of shares
10-Non-performing companies
11-Impact of the macroeconomic variables
Related Documents
