Page 1
VYTAUTAS MAGNUS UNIVERSITY
FACULTY OF ECONOMICS AND MANAGEMENT
Vytautas Strimaitis
LONDON STOCK EXCHANGE AND INDIVIDUAL COMPANIES REACTION
TO “BREXIT” NEWS
Final Master Thesis
Finance study programme, state code 6211LX042
Finance study field
Supervisor: Assoc. prof. dr. Renata Legenzova (degree, academic position, name and last name)
Defended: Dean assoc. prof., dr. Rita Bendaravičienė
Kaunas, 2020
Page 2
2
SANTRAUKA
Baigiamojo darbo autorius: (Vytautas Strimaitis)
Pilnas baigiamojo darbo
pavadinimas:
(Jungtinės Karalystės Finansų Rinkos ir Individualių
Kompanijų Reakcija į „BREXIT“ Naujienas)
Baigiamojo darbo vadovas: (Assoc. prof. dr. Renata Legenzova)
Baigiamojo darbo atlikimo
vieta ir metai:
Vytauto Didžiojo universitetas, Ekonomikos ir vadybos
fakultetas, Kaunas, 2020
Puslapių skaičius: (65)
Lentelių skaičius: (l2)
Paveikslų skaičius: (8)
Priedų skaičius: (8)
Raktiniai žodžiai (International financial markets (G15), Event studies (G14),
Portfolio Choice (G11) )
Tyrimo metu tiriama, kaip “BREXIT” naujienos veikia Jungtinės Karalystės finansų rinką ir
atskiras kompanijas. Darbas apžvelgia kokie veiksniai veikia finansų rinkas ir investuotojų elgseną.
Darbe išskiriamas ir didesnis dėmesys sutelkiamas į politinį nestabilumą ir kaip jis veikia rinkas ir
investuotojus. Apžvelgiami egzistuojantys matematiniai modeliai, kuriais matuojamas politinio
nestabilumo poveikis finansų rinkoms. Darbe buvo pritaikytas “Event-study” metodas nustatant
kaip “BREXIT” naujienos paveikė Jungtinės Karalystės finansų rinkas ir individualias kompanijas,
kuriomis prekiaujams Jungtinės Karalystės akcijų biržoje. Rezultatai parodė, jog iš 17 atrinktų
“BREXIT” įvykių įtakos Jungtinės Karalystės finansų rinkai turėjo tik 5. Individualioms
kompanijoms naujienos susijusios su “BREXIT” turėjo reikšmingą poveikį. Didžiajai daliai
kompanijų įtrauktų į imtį “BREXIT” naujienos turėjo negiamą poveikį dvejuose laiko perioduose: 2
dienos iki ir po įvykio bei 10 dienų iki įvykio. “BREXIT” naujienų poveikis finansų sektoriaus
kompanijoms buvo stipresnis nei kitų sektorių kompanijoms.
Page 3
3
ABSTRACT
Author of Final Master
Thesis:
(Vytautas Strimaitis)
Full title of Final Master
Thesis:
(London Stock Exchange and Individual Companies
Reaction to “BREXIT” News)
Supervisor: Assoc. prof. dr. Renata Legenzova
Presented at: Vytautas Magnus University, Faculty of Economics and
Management, Kaunas, 2020
Number of pages: (65)
Number of tables: (12)
Number of pictures: (8)
Number of annexes: (8)
Key words (International financial markets (G15), Event studies (G14),
Portfolio Choice (G11))
The main problem of the research is how “BREXIT” news affect the UK financial market
and individual companies. Research analyses factors affecting financial markets, individual
companies and investor behaviour. Research concentrates on how political instability affects
financial markets, individual companies and investor behaviour. Furthermore the empirical models
evaluating the effect of political instability are analysed in the work. Research applies “Event-
study” method and estimates how “BREXIT” news affect UK financial market and individual
companies listed on UK financial market. Results determined that “BREXIT” news were
insignificant to the UK financial market because only 5 out 17 events were significant to returns of
it. However analyses of individual companies determined that “BREXIT” news had a significant
effect to them. Larger number of companies of the sample experienced significant negative changes
in their returns due to the “BREXIT” events. Two “event-windows” proved it. “BREXIT” news
effect on financial sector companies was stronger compared to other to other sector companies.
Page 4
CONTENT
Introduction ............................................................................................................................... 5
I. Theoretical Aspects Of Finance Theories, Models And Political Stability And Their
Influence On Financial Markets .............................................................................................. 7
1.1 Comparative Analysis of Traditional and Behavioral Finance Theories ........................... 7
1.2 Traditional and Behavioral Factors Affecting Stock Markets and Prices .......................... 11
1.3 Financial Players’ Behavior Influenced by Political Uncertainty ...................................... 15
1.4 Empirical Models for Events Effect Estimation ................................................................ 17
1.5 Overview of Previous Research on “BREXIT” Effect to Financial Market ...................... 19
II. “BREXIT” News Effect For “London Stock Exchange” And Individual Companies
Methodological Justification ..................................................................................................... 25
2.1 Relevance and Aim of the Research .................................................................................. 25
2.2 Research Hypothesis .......................................................................................................... 27
2.3 Logic and Argumentation of Research Model ................................................................... 30
2.4 Data Sample of the Research ............................................................................................. 34
2.5 Limitations of the Empirical Model ................................................................................... 36
III. Empirical Results On “BREXIT” Effect For London Stock Exchange, Individual
Companies And Various Sectors .............................................................................................. 39
3.1 Descriptive Statistics of London Stock Exchange, Individual Companies and Various UK
sectors ................................................................................................................................. 39
3.2 “BREXIT” News Effect for London Stock Exchange ....................................................... 42
3.3 “BREXIT” News Effect for Individual Companies ........................................................... 45
3.4 “BREXIT” News Effect for Various United Kingdom Economical Sectors ..................... 51
3.5 Discussion of Research Results ......................................................................................... 53
3.6 Limitations of the Empirical Data Findings and Reccomendations for Future Researches .... 57
Conclusions ................................................................................................................................ 59
References................................................................................................................................... 61
Annexes ....................................................................................................................................... 66
Page 5
5
INTRODUCTION
Relevance and topicality. Financial markets are sensitive to factors that occur in the world
and markets and their occurrence have an effect for their performance. Nevertheless, huge role for
price indexes of financial markets has a politics and political news, announcements around the
world and the country the markets trade. One of the major political news that hit the EU and world
was UK’s withdrawal from the European Union announced in 2016 by British referendum. It
affected not only UK financial markets but other countries and members of EU as well. This
agreement to leave European Union got the name of “BREXIT”. For the past years British
referendum announced a lot of news towards “BREXIT” and with further announcement of them
markets reacted differently. Actual day of the agreement to leave EU was delayed and etc. creating
a lot of chaos for the past few years not only in the UK financial markets but also worldwide So it is
important to estimate how strongly “BREXIT” news affects the UK financial market and individual
companies that are listed in UK’s financial market. Previous researches on this topic determined
that “BREXIT” affected negatively United Kingdom financial market (Guedes, Ferreira, Dionisio
and Zebende, 2019). Various economical sectors of United Kingdom also suffered due to the
“BREXIT” (Ramiah, Pham, Moosa, 2017). According to their results financial sector companies
experienced -15.37% loses on 10 days period. Insurance sector companies experienced loss of -
8.18%. Lastly travel and leisure sector companies experienced loses of -3.64% on 10 day period.
Lastly close trade partners of UK also suffered the consequences of “BREXIT” referendum
(Abraham, 2018). NZX 50 index returns decreased by more than 4% on 20 days period until the
referendum vote.
Research object in this research are “London Stock Exchange” and largest individual
companies that are listed on “London Stock Exchange”.
Research problem addressed in this paper is what UK stock market reaction to
“BREXIT” news announcements.
The aim of this research is to analyze what effect has news about “BREXIT” to UK stock
market and individual companies. To reach the aim of the paper the following objectives were set:
1. Define the meaning of traditional and behavior finance and analyze the estimation
models of traditional and behavior finance
2. Name factors affecting stock prices movements
3. Define political uncertainty and its effect on financial markets.
4. To analyze the existing models and results of previous researches of “BREXIT”
effect for financial markets and individual companies
Page 6
6
5. Present the results of the research analyzing how „BREXIT“ news affects UK
financial market and individual companies.
To determine named objectives theoretical and practical analyzes will be applied. For
theoretical analyzes part previous authors research papers will be analyzed and compared. Based on
their results the research aim and problem will be explained. Furthermore the analyzes of previous
researches will contemplate naming research hypothesis and empirical model. Lastly empirical
model based on other authors work will be applied in order to approve or reject the hypothesis of
the research.
Scientific literature for the research were selected from the „Science Direct“, „EBSCO“
scientificic literature websites. Empirical data for the research were selected from „investing.com“
and „londonstockexchange.com“ online websites.
The paper is structured in 3 main parts:
The first part of paper analyses theoretical aspects of the work. In the first part traditional
and behavioural finance theories definitions, models, factors affecting stock prices movements are
analysed. Furthermore analyses of political instability effect to financial markets are presented.
Empirical models estimating the effect of such events to financial markets and previous works on
how “BREXIT” effected financial markets are presented.
The second part of paper develops the method of the research. In the second part of the
work hypothesis based on previous authors works are constructed. Furthermore the relevance and
necessity of the research is reasoned. Lastly the empirical model and data for the research are
presented.
In the third part results of the empirical model are presented. It shows how “BREXIT”
events affected UK financial market, individual companies with largest market capitalization listed
on UK financial market and various economy sectors of UK. Furthermore the hypothesis results are
presented. Lastly limitations of the empirical research and recommendations for future researches
are presented.
Page 7
7
I. THEORETICAL ASPECTS OF FINANCE THEORIES, MODELS
AND POLITICAL STABILITY AND THEIR INFLUENCE ON
FINANCIAL MARKETS
In the first chapter of research theoretical aspects of the work will be analyzed. In the first
part traditional and behavioural finance theories definitions, models, factors affecting stock prices
movements are analysed. Furthermore analyses of political instability effect to financial markets are
presented. Empirical models estimating the effect of such events to financial markets and previous
works on how “BREXIT” effected financial markets are presented.
1.1 Comparative Analysis of Traditional and Behavioral Finance
Theories
In this section the financial theories and their models existing in the financial world will be
analyzed, described and compared. In general there exist two different theories in financial sciences
on stock market prices movements. The earlier one are called “Traditional finance theory” which
started in mid-eighteenth century when John Stuart Mill in 1844 years introduced the concept of the
“Economic man” also called “homo economicus”(Kapoor and Prosad, 2017). Traditional theory
focuses on the concept of “expected utility theory”. Utility in traditional finance theory is
considered as a measure of satisfaction of individuals consuming by good of a service. “Homo
economicus” is the person who tries to maximize his satisfaction with his limited resources and
possibilities. Yoshinaga and Ramalho (2014) supplement the concept of traditional finance theory
and definition of “homo economicus” by adding that economic subjects like “homo economicus”
makes decisions with unlimited rationality, present risk aversion and aim to maximize “expected
utility” at every decision he makes. After John Stuart Mill (1844) introduction of rational investor
concept whose main object is to maximize his utility others authors followed and improved this
theory. Two major theory models were “Markowitz portfolio theory” by Harry Markowitz in 1952
and “Efficient Market Hypothesis” by Eugene Fama in 1970 (Kapoor and Prosad, 2017).
“Markowitz portfolio theory” created the capital asset pricing model which was the most central
assets pricing model in finance until “Efficient Market Hypothesis” creation, after it “Markowitz
portfolio theory” was pushed away because Markowitz portfolio theory could not explain anomalies
in its theory results. Matsumura and Kawamoto (2013) supplement the traditional finance theory
with the explanation for “Efficient market hypothesis”. Authors define it as the fact that stock prices
are affected by the profits of the future profits of companies and information is used efficiently in
anticipation formation. Matsumura and Kawamoto (2013) explain that in efficient markets
information about company is immediately analyzed by investors and reflect the rational decisions
of the investor. Traditional finance theory is based on rational financial player whose purpose is to
Page 8
8
maximize his utility and all information available in traditional finance theory is constructed to fit
the calculations. Traditional theories couldn’t explain the anomalies and disruptions in stock
markets. It couldn’t explain the stock market bubbles, under-reaction, investor sentiment and etc.
That’s when the new field of financial theory started to evolve and it was named “behavioral
finance”. Main purpose of it was to investigate those anomalies that traditional finance couldn’t
explain (Kapoor and Prosad, 2017). The backbone of behavioral finance theory was formatted by
the psychologists when they introduced the concept of “prospect theory” for analysis of decision
making under risk (Kapoor and Prosad, 2017). “Prospect theory” critique the expected utility theory
(traditional finance) as a tool for decision-making in situations involving uncertainty and risk,
adopting as premises the presence of irrationality and the corresponding use of heuristics in
people’s decision-making processes, leading to systematic errors due to biased cognitive processes
(Yoshinaga and Ramalho, 2014). “Prospect theory” explains how people overestimate the changes
involving losses and underestimate the changes involving gains. Even small loses has a higher value
than the same gains, meaning that loss in 2 would be higher than gain in 2. Filbeck et al., (2017) in
their work explain behavioral finance as the field of finance that proposes psychology based
theories to explain stock market anomalies such as severe rises or falls in stock price. They do not
provide such a deep analysis to behavioral finance theory as authors mentioned before but correctly,
strictly and similar to others interpreter the definition and purpose of behavioral finance. Moreover,
Divanoglu and Bagci (2018) defined behavioral finance similarly to other authors but also
implementing that decisions of investors are also affected by sociological factors. They state that
when it comes to investing individual investor can’t act rationally but there exists continues
rationality in traditional financial understanding (Divanoglu and Bagci, 2018). Authors specifically
name what psychological concepts effect the investor decisions such as biases, overconfidence and
etc. Ramiah, Xu and Moosa (2015) analyzes the definition of behavioral finance and they also
describe characteristics of behavioral finance investor. They consider investor not as “rational” as
traditional theory scientists but as how they describe it “normal”. It means that investor can make
mistakes in his valuations (cognitive errors) (Ramiah, Xu and Moosa, 2015).
After analyzing the concept of traditional and behavioral finance theory models of those
two theories will be analyzed. Traditional finance theory as it was mentioned before is based on the
two concepts, risk and return. Financial player always tries to maximize his utility and he is aware
of his risks of loss and always stays rational. In traditional finance theory all the financial decisions
are based on the strict mathematical calculations. Sushma and Rushdi (2018) in his work perfectly
describe and name all existing mathematical theory models for estimation of stock prices. He names
five most important theories in traditional finance: Expected utility theory, Modern portfolio theory,
Capital pricing model, efficient market hypothesis, Arbitrage pricing theory.
Page 9
9
Expected utility theory is based on the decision-making under risk. The utility of a risky
prospect is given by the sum of the utilities u of the alternative possible outcomes of the prospect,
each weighted by the probability that the outcome will occur (Moscati, 2016). Furthermore author
gives an example with lottery ticket and trip to London where the decision maker according to
expected utility theory shall choose the risky prospect or lotteries with the highest expected utility
(Moscati, 2016).
Modern portfolio theory mean-variance analysis is a deployment of the mathematical
model introduced by Markowitz in 1952 (Francisco dos Santos and Siqueira Brandi, 2017). Authors
explain that modern portfolio theory aims to construct a portfolio of assets whose expected return is
maximized with a given level of risk or volatility, defined as variance of the portfolio. Francisco dos
Santos and Siqueira Brandi (2017) explain that investors tolerate risk differently and some investors
accept lower risk with lower returns and some acts on opposite and intends to take bigger risk for a
larger returns. Resulting in trade-offs if the investors risk aversion is not on the level he wants it to
be. In their work Francisco dos Santos and Siqueira Brandi, (2017) explain the formula of “Modern
portfolio theory” where return of portfolio at the time t (𝑅𝑡) is calculated dividing value of asset at
time t (𝑥𝑡) with same asset in previous time t (𝑥𝑡−1). The formula is written below:
𝑅𝑡 =𝑥𝑡
𝑥𝑡−1
(1)
Authors furthermore exclude other factors necessary to calculate expected return of
portfolio, profitability and name factors like Return of asset, Return of portfolio, transpose matrix of
expected returns, weighting of the asset I (proportion of the asset i in the portfolio) and the matrix
of portfolio weights.
Capital asset pricing model was introduced in financial economics by William Sharpe in
1964 and the earliest model was introduced in 1965 when he collaborated with John Lintner.
(O’Sullivan, 2018).Author explains the basic concept of CAPM model and the formula is presented
below:
𝑟𝑎𝑒 = 𝑟𝑓 + 𝛽𝑎𝑒(𝑟𝑚𝑒 − 𝑟𝑓) (2)
Where 𝑟𝑎𝑒 is expected return on asset so that is determined as the price investor willing to
pay for the asset, 𝑟𝑓 is risk free assets return, 𝛽𝑎𝑒 is expected volatility of the price and 𝑟𝑚𝑒 is
average overall expected return of the market as whole (O’Sullivan, 2018). The Capital asset
pricing model received a lot criticism because of market anomalies and due to that reason a new
asset pricing model was introduced with the name of “efficient market hypothesis”.
Efficient market hypothesis is defined as the market where large number of participants
actively trade in order to maximize profits, with each participant striving to anticipate the future
market price of individual securities (Lekovic, 2018). Lekovic (2018) furthermore introduces the
Page 10
10
empirical model with measuring stock market prices for efficient market hypothesis. It calculates
the return of security at period time t. The formula is presented below:
𝑟𝑡 = 𝑎 + 𝑏𝑟𝑡−1−𝑇 + 𝑒𝑡 (3)
Lekovic (2018) name 𝑎 as expected return on securities not affected by past return, 𝑟𝑡−1−𝑡
as return on securities in period t-1-T, 𝑏 as correlation coefficient between return on securities in
period 𝑡 and return on securities in period t-1-T and 𝑒𝑡 as random error (Lekovic, 2018).
Lastly the concept and logic of arbitrage is explained by already mentioned author Sushma
and Rushdi (2018). In his work authors explains Arbitrage pricing theory as greed of investors
where they try to create return by taking benefit of price fluctuation (Sushma and Rushdi, 2018).
Sushma and Rushdi (2018) in their work explain that according to arbitrage model, investor buys an
asset in the market where it is cheaper and tries to sell it in the market where it is costlier.
These financial theories were used a lot in markets in the last 20-30 years but in the past
years they begin to fail explaining the anomalies that occurred more in markets. Due to that reason
new concept and theory was necessary in order to explain anomalies of market and that’s when
behavioral finance started to rapidly evolve. In the next paragraph concept and evolution of
behavioral finance will be analyzed.
Behavioral finance tries to explain psychological, irrational and biased nature of investors.
This new field of finance is guided by different assumptions and factors affecting the stock market
prices. Behavioral finance assumes that psychological factors are the ones to blame for the market
anomalies and its inefficiency. There exists several theories of the behavioral finance and it started
with the Selden who find out in his research in 1912 that movement of stock prices are dependent to
a considerable degree on the mental attitude of the market participant (Ramiah, Xu and Moosa,
2015). After the Selden work new theories of behavioral finance followed. Ramiah, Xu and Moosa
(2015) name all of them in their work from earliest ones until the newest ones: “Realization utility”,
“Prospect theory”, “Risk aversion theory”, “Disposition effect theory” and etc. “Realization theory”
also takes neuroscience in to the researches of financial player behavior which means that
behavioral finance is evolving in to the higher level of science researching financial human
behavior. Sushma and Rushdi (2018) name two more theories of behavioral finance: “Behavioral
Asset Pricing Model” and “Behavioral portfolio theory” (Sushma and Rushdi, 2018). The first
model names investors as informational and noise traders. Informational traders base their decisions
on CAPM models and noise traders do not use any other already mentioned model in their investing
decisions (Sushma and Rushdi, 2018). The behavioral portfolio theory is based on the concept that
investors divide their portfolio into several mental accounts, each one representing a different goal
(Alles Rodrigues and Lleo, 2018). Authors state that behavioral portfolio theory is not fully rational
Page 11
11
and takes in to account some behavioral biases which lead to market anomalies that traditional
finance theory can’t explain (Alles Rodrigues and Lleo, 2018).
To conclude this section it is possible to say that both of the theories are strongly related.
Traditional theory has some major problems because it has a very strict description about investor
in financial markets and assume that all players are the same and not affected by psychological
factors. Secondly traditional theory is very depended on mathematical calculations and can’t
explain anomalies that occur in the market and that’s why behavioral finance as a different field of
finance theory develops and becomes more popular among researchers. Behavioral finance theory
contemplates traditional finance theory with psychological aspects of investor behavior. One of the
main goals of behavioral finance is to explain those anomalies in financial market, through
psychological aspects of financial player. As it was analyzed the behavioral finance considers
investor as normal meaning that he can make valuation mistakes in his assumptions or could base
his decisions on his emotional statement. Furthermore analyzes of models confirms that both of the
theories are related. We see that two main concepts of behavioral models are strongly related to
CAPM model and Markowitz Modern portfolio theory (Alles Rodrigues and Lleo, 2018).
Furthermore already named “Realization theory” takes neuroscience and human brain activity in
particular as factor of our behavior towards financial decisions. This research focuses on behavioral
finance and how political stability influences the market player decision and their investment
portfolio.
1.2 Traditional and Behavioral Factors Affecting Stock Markets and
Prices
In this section factors affecting the price of stocks will be named. They will be named
separately in order to better understand how different theories relate to different factors affecting the
price of the stocks.
There exist many of factors that affect the price of the stock in traditional finance theory.
As already analyzed the main focus of the traditional finance theory is finding the price of the stock
at the particular time. What affects the price mostly according to researchers will be named in this
paragraph.
Irshad (2017) in his research name four macro-economic factors that effects the price of
the stock market. Irshad (2017) names inflation, export, exchange rate and industrial production.
Zhou, Cui, Wu and Wang (2018) also names most commonly used macro-economic factors
affecting the financial markets and also includes three other factors currency supply, interest rate
and government performance. The last one is really important due to the reason that this work will
specifically be focused on this topic of government performance effect on stock markets. Two other
authors that include more macro-economic factors which effects the stock markets are Asteriou and
Page 12
12
Siriopoulos (2000). They also include the growth rate of real GDP, growth of fixed capital
investment and growth of general stock market index. Very similar to the Asteriou and Siriopoulos
(2000) and Irshad (2017) approach to factors that affect the stock markets also provides Lehkonen
and Heimonen (2015) in their work. They also name GDP per capita, industrial production,
inflation. But they also include turnover, domestic credit, narrow and abroad money growth.
More concentrated analysis on the factors that affect stock markets movements were
analyzed by Hillier and Loncan (2019). They name eight factors that change movement of price on
industrial level. Hillier and Loncan (2019) names sales, equity, debt, liquidity, Tobin’s Q,
Debt/assets ratio, foreigner control, dividends announcement. Zhou, Cui, Wu and Wang (2018) as
micro-economic factors name size of the market, number of investors, investor behavior. After these
factors authors also name dividend yield, payout and leverage rates such as factors affecting stock
prices. Divanoglu and Bagci, (2018) in their research as financial factors name cash flow, risk,
liquidity, return ratios and investment duration. Their perspective on factors affecting the prices of
stocks were observed from the investor view. All of the factors affecting stock prices from
traditional finance theory is named in the table below (Table 1). In the next paragraph behavioral
finance theory factors affecting stock market movements will be named.
Table 1
Traditional Factors Affecting Stock Prices
Category Factor
Macro-
economic
level
Inflation, Export, Exchange rate, Industrial production, Currency supply, Interest
rate, Government performance, BVP growth, Investment growth, Turnover,
Domestic credit, Narrow, Abroad money growth
Micro-
economic
level
Sales, Equity, debt, liquidity, Tobins Q, Debt/assets ratio, Foreigner control,
dividends announcement, size of the market, number of investors, dividend yield and
payout, leverage rates, cash flow, risk, return ratios, investment duration
Raheja and Dhiman (2019) in their work names four categories of human emotions
affecting stock market movements. They name conservatism, Overconfidence, Herding, Regret.
Renu and Christie (2018) names seven other behavioral biases that affect our decision making in
stock markets. They name mental accounting, anchoring, Gambler’s fallacy, Availability, Loss
aversion, Regret aversion, Representativeness, overconfidence.
“Conservatism” is related with people inability to adapt and accept changes (Raheja and
Dhiman, 2019).
“Overconfidence” in Raheja and Dhiman (2019) work is described as people’s
overestimation of futures forecast. Overconfidence is also named in other authors work (Renu and
Christie, 2018). Their description is similar to already mentioned Raheja and Dhiman (2019).
Page 13
13
Deeper understanding about “overconfidence” in investment decision making is analyzed by Raut
and Kumar (2018). These authors name overconfidence as the strongest of all named biases in the
finance world. According to them this bias creates the biggest losses in the investment because of
the people overestimation of accuracy of their information, their successes and capabilities.
Overconfident investors, believing that they possess greater precision on security valuations, trade
too much and thereby lower their expected utility (Raut and Kumar, 2018). Overconfidence was
measured by creating a simulation for the investors that were taken to sample (Raheja and Dhiman,
2019). In their work they analyze previous works and show tendency that in a rise of trend investors
become overconfident.
“Herding” in Raheja and Dhiman (2019) work is outlined as person inability to make
decisions on his own and his investing decisions influenced by other investors. Raut and Kumar
(2018) describes it as an influential behavior of financial players whose action shows animal-like
behavior, where they follow blindly other investors decisions ignoring their intuition and having no
confidence in their decisions. Raut and Kumar (2018) states that these decisions are naturally given
by our nature because humans are sociable and want acceptance and recognition from the society
rather than standing alone.
“Regret” is described as people’s cling on the past loss and its huge impact on future’s
forecasts (Isidore and Christie (2018)
“Mental accounting” which means that people keep winning stocks in one account in head
and losing ones in another even though the portfolio is the same (Isidore and Christie, 2018).
“Anchoring” is the bias that makes investor compares prices of stock from a certain point
of its lifetime (Isidore and Christie, 2018). Raut and Kumar (2018) describes “Anchoring” similar
to Isidore and Christie (2018) and states that anchoring is bias of human beings to compare the price
of item from a certain point and in its life spawn. It creates the illusion for investor or buyer that the
prices should rise to that certain point of anchoring. With anchoring people prefer relative thinking
of the price rather than absolute (Raut and Kumar, 2018).
“Gambler’s fallacy” causes the investors to anticipate the change in the trend of the stock
market depending on the number of years of bullish success or bearish failure (Isidore and Christie,
2018).
In Isidore and Christie (2018) work “availability” is explained as the concentration on
easily available information when making investment decisions while ignoring important and
necessary information. Same characteristics but different name for “availability” was given in Raut
and Kumar (2018) work. They explained it as “availability bias” and explained it as investor
tendency to judge the frequency or probability of an event in terms of how easy it is to think of an
example of that event. In this term individuals information taken in to consideration while making
Page 14
14
investment decision could be less valuable in absolute terms and cause losses after investment
decision.
“Loss aversion” makes people sell winning stocks and keep the losing ones in the portfolio
with the intention that the price of it will go up. Deeper look on the loss aversion provides Bian,
Chan, Shi and Zhou (2018). Bian, Chan, Shi and Zhou (2018) explain that people seeing the rise in
stock price tend to sell them sooner and show risk intolerance while if the same person has losing
stocks he tends to keep it and shows risk tolerance. It is very hard to measure all of named
behavioral factors by quantitative methods.
“Representativeness” is the bias that causes investors take past price as the representative
of the future one (Isidore and Christie, 2018). Raut and Kumar (2018) explains
“representativeness” in deeper manner and explains it as bias in which financial player under
circumstances of uncertainty in which financial player makes a judgement about company based on
the similar in essential properties to its parent population and reflects the salient features of the
process by which it is generated. Secondly Raut and Kumar (2018) explain that because of that
financial players select the companies based on their recent returns, popularity, type of management
and etc.
“Endowment bias” is the bias when people tend to have sentimental feelings for a property
and tend to overprice it because of that emotion (Sushma and Rushdi, 2018).
Sushma and Rushdi (2018) in their work not only describes the psychological aspects of
human behavior that could influence the investment decisions and market movements but also
groups all of the already mentioned biases in two categories: cognitive biases and emotional biases.
Cognitive biases are anchoring and adjustment, framing, conservatism, availability, mental
accounting (Sushma and Rushdi, 2018). Emotional biases in his work are endowment bias, loss
aversion, optimism and status quo. This categorization model was firstly introduced by another
author, Pompian in 2011 (Sushma and Rushdi, 2018). Another categorization for these
psychological aspects was introduced earlier in 2000 by scientist Hersh Shefrin (Sushma and
Rushdi, 2018). He also categorizes these biases in two categories, one named “Heuristics”
(overconfidence, anchoring, adjustment, reinforcement learning, excessive optimism and
pessimism) and other “Frame dependent biases” (narrow framing, mental accounting and the
disposition effect) (Sushma and Rushdi, 2018).
Another aspect of psychological factors affecting investor decision making is based on the
cultural differences. As an example could be Zhou, Cui, Wu and Wang (2018) work in which they
tried to evaluate what cultural differences has on investing behavior. In their findings results
showed that countries with smaller cultural difference have similar level of volatility and different
cultural dimensions have different influence on volatility of country financial markets. Other
Page 15
15
research on cultural aspect of decision making for investments were researched by Divanoglu and
Bagci (2018). Authors states that environmental factors have an impact on investors decision
making (Divanoglu and Bagci, 2018). They name environmental factors such as family, friends,
close environment, socio-cultural environment (Divanoglu and Bagci, 2018). Authors explain the
reasons behind it and that could be because of the lack of knowledge in investing, seeking the
approval of others and etc., similar to herding or information cascades (Divanoglu and Bagci,
2018). All of the psychological aspects of human behavior were shown in the table below (Table 2).
Psychological factors in the table 2 are separated in two categories: Cognitive bias and emotional
bias.
Table 2
Psychological Factors Affect Stock Prices
Cognitive bias Emotional bias
Anchoring, Adjustment, Framing,
Conservatism, Availability, mental accounting,
Gamblers Fallacy, Representativeness
Endowment bias, Loss aversion, optimism,
status quo, Overconfidence, Herding, Regret.
Lots of factors affect stock markets and its prices, indexes. When measuring the stock
market movements from behavioral finance theories perspective we see that factors affecting stock
prices movement are based on psychological aspects of human behavior. From analyzes of previous
authors works it is possible to say that many different factors affect human behavior deciding their
investing decisions starting from misinterpretation of information to emotional valuations of stock.
Decision could be influenced by environmental, cultural aspects of specific area the investor is
located. Speaking of factors affecting the decision making and movement of financial markets from
micro-economic level we have firm size, sales, equity, debt, liquidity, Tobin’s Q, Debt/assets ratio,
foreigner control, dividends announcement. In macro-economic level we have factors like inflation,
export, exchange rate, industrial production, grow rate of real GDP, growth of fixed capital
investment and growth of general stock market index. These factors are more easily measurable
then previously mentioned psychological ones and also provides quite reliable information on stock
prices and movements. Nevertheless there could be a lot of more factors that affect stock market
indexes but these already named ones give the perception of how dynamic and easily influenced
stock market is. In the next section political instability and its effect to stock markets will be
analyzed.
1.3 Financial Players’ Behavior Influenced by Political Uncertainty
In this section the effect of political uncertainty, its factors will be analyzed and explained.
Firstly, even though diversification of portfolio with foreign country assets shows larger gains of
portfolio, investors do not tend to invest that much in foreign country assets due to the fact of
Page 16
16
political instability (Smimou, 2014). After deeper understanding and research in to why investors
tend to keep their investment in the home country authors find out that after political instability
other factors follow, that would be taxes system of that country, border controls and political and
social trends (Smimou, 2014). Some of the authors analyze what effect has political instability on a
sector level (Moshi and Mwakatumbula, 2017). Authors analyzed the effect of political stability to
telecommunication market and find out that political instability negatively effects
telecommunication market. Authors provides with the necessity to take a deeper look in political
instability on sector level, because in some sectors it could provide with the benefits (Moshi and
Mwakatumbula, 2017).
Political instability and uncertainty as we can see from already analyzed researches has as
huge impact on investor behavior and their psychological state. Political instability or uncertainty is
researched by the behavioral theory scientists that concentrate on the influence of emotions and
cognitive biases on people judgments and decision making (Soltani, Aloulou and Abbes, 2017).
These authors name political instability as factor that could make financial markets collapse.
Wisniewski (2016) in his work name political uncertainty as the risk that is created by the unstable
political environment which are reverberate in financial markets and diminish shareholder wealth.
Furthermore it is necessary to understand how political uncertainty manifest in nowadays
society or what kind of factors, events from political point of view drives the markets. Mnif (2017)
in his work name political uncertainty factors that drives financial markets. He names events like
presidential elections, terrorism attacks, military invasions/wars and civil overthrows of local
government. Similar to these uncertainties were named by Wang and Boatwright (2019) in their
work authors name it as political shocks or policy changes. Wang and Boatwright (2019) don‘t split
political uncertainty in to the deeper categories and name it as two already mentioned but in their
work the political uncertainty is treated as in Mnif (2017) work. Wisniewski (2016) in his work like
other authors names three already mentioned political uncertainties which are presidential elections,
terrorism attacks, civil wars and wars. Nevertheless, Wisniewski (2016) in his work name fourth
political uncertainty that effect stock markets which is called „political communications“. „Political
communications” is the factor that focuses on political speeches and communiques that could
possibly influence movements of stock market. Irshad (2017) specifies in his work the political
uncertainty factors and names six of them. He names strikes, assassinations, riots, demonstrations,
government longevity and government change. He specifies them more deeply and more
specifically but they all could be grouped in the already named categories that are terrorism, war,
civil war, and elections.
After analyzing other authors work it is possible to conclude that political stability,
certainty is very necessary to the development of countries and stability of their financial systems.
Page 17
17
As we already saw in some of the authors works political stability is one of the key factors affecting
the financial markets and its instability could influence collapse of the financial market.
Furthermore deeper analysis showed that political uncertainty lowers the profits of investors which
lead to larger investment in domestic countries and foreign direct investment loss. Lastly analyzes
of previous authors works helped to group and name key factors which affect the political
uncertainty and stability which are terrorism, civil wars, wars, elections and government
communications. In the next paragraphs models for evaluating event effect for financial markets and
“BREXIT” as the political uncertainty and its effect to financial markets from previous author
works will be analyzed.
1.4 Empirical Models for Events Effect Estimation
In finance abnormal returns of market, volatility can be interpreted as political or financial
uncertainty and after it, it becomes the primary focus for many investment decisions and portfolio
management. Too much of volatility or movement in daily returns is a sign of unstable economy
and high risk of investment while the opposite could be said about low volatility and returns. In
order to measure the volatility and returns of the markets of domestic or even global the various
number of methods are applied.
First and mostly related to original “Event-Study” Method is analyzed by Chen (2004).
“Event-study” model is based on efficient market hypothesis. This hypothesis states that as new
information becomes available, it is fully taken into consideration by investors assessing its current
and future impact. That means that investors immediately reassess individual firms and their ability
to withstand potential event (Chen, 2004). According to researchers if investors react favorably to
event we should expect positive returns and if investors reacts negatively to event we should expect
negative returns (Chen, 2004). “Event-study” model tries to identify the abnormal returns to firms
from a specific event (Chen, 2004). Furthermore, “Event-study” tries to capture the influence of that
event in the specific period of time the event occurred (Bonchev and Pencheva, 2017). In many
research’s event period is selected different and do not have strict rules that should be applied
selecting the period (Bonchev and Pencheva, 2017). Bonchev and Pencheva (2017) in their work
name examples of time-periods, for example (-1, 1), (-3, +3) and etc. The same rules are applied for
“estimation window”. It is up to researchers to select what size of "estimation-window" he selects
but it is important and suggested to select at least 250 days before the major event day.
Some authors try to merge empirical models based on their research logic and develop
modified approach to “Event-study” method. In the present time the “Event-study” method is
becoming more and more complex and Shahzad, Rubbaniy, Lensvelt and Bhatti (2019) represent
the supplemented version of this model. Shahzad, Rubbaniy, Lensvelt and Bhatti (2019) update the
model with more specific control variables in order to better evaluate the effect of events. The
Page 18
18
control variables depend on the hypothesis and questions in the methodology and can variate from
micro to macro level. In Shahzad, Rubbaniy, Lensvelt and Bhatti (2019) case they name “Friday
effect”, “Halloween effect”, “January effect”, “Mark Twain effect”, “Turn of the month effect” and
some control variables on firm level like “Market value”, “labour dependency”, “foreign sales” and
were the company against the “Brexit” on votes. Next work that used different control variables
were done by Davies and Studnicka (2018). Davies and Studnicka (2018) applied “Event-study”
method and estimated effect of “BREXIT” on firm level and were trying to estimate that companies
who were exposed in UK and EU were doing worse. Based on their hypothesis and assumptions
their control variables were used from firm level. Davies and Studnicka (2018) used “industry-
specific exchange rate depreciation” as control variable in their research. Oehler, Horn and Wendt
(2017) same as Shahzad, Rubbaniy, Lensvelt and Bhatti (2019) used Market Capitalization, Foreign
sales as control variables in their research. Oehler, Horn and Wendt (2017) used industrial sector as
additional control variable in their work stating that different industries would be affected different
due to the “BREXIT”. Even though the control variables might differ in different scenarios but the
main focus is still the same to catch the anomalies in the daily, hourly and etc. returns in specific
markets. Control variables are important in this method in future researches because as many
authors work show it is hard to capture whether the observed event has an effect for the financial
markets or other external, unrelated factors might influence the results of indexes. Morales and
Andreosso-O’Callaghan, (2019) in their work explain as one of the limitations of their work the
necessity of larger group of control variables because results might have been affected also by other
factors.
Secondly authors use “Event-study” method to estimate not actual prices and abnormal
returns of the market, but volatility of the market. “Event-studies” whom measure volatility of
markets help us track the negative and positive outcomes of those events. Morales and Andreosso-
O’Callaghan (2019) in their work state that modeling such scenarios help us to understand the
volatility’s persistence and clustering of these kind of events. High volatile markets suggest that
investing in them would be a lot riskier than in low volatility markets and that should be taken in to
consideration when planning your investment portfolio. In econometrics to catch the volatility
Autoregressive Conditional Heteroskedastic (ARCH) and Generalized Autoregressive Conditional
Heteroskedastic (GARCH) models are used. Autoregressive Conditional Heteroskedastic (ARCH)
were introduced and firstly used by Robert F. Engle in 1982 (Morales and Andreosso-O’Callaghan,
2019). Morales and Andreosso-O’Callaghan (2019) explains in their paper that after the appearance
of ARCH method the GARCH were introduced whom were improved by Bollerslev in 1986.
GARCH model overcome the limitations of the ARCH model which was based on past sample
variance (Morales and Andreosso-O’Callaghan, 2019). After “ARCH” and “GARCH” methods
Page 19
19
appearance in finance world other future scientists improved the models so the others methods like
“EARCH”, “GJR (TARCH)”, “GARCH-M” were introduced in future time (Morales and
Andreosso-O’Callaghan, 2019).
Another model for estimation is named “copula functions” (Aristeidis and Elias, 2018).
These models according to Aristeidis and Elias (2018) are usually employed by authors to depict
tail dependence in financial time series. Additionally, the models are capable of detecting non-
normality and fattailedness in stock exchange markets. This model is employed by Aristeidis and
Elias (2018) in their work and they try to evaluate the announcement of “article 50” and negative
effect for the UK and other countries market.
Some methods to estimate the effect of the event takes not a stock market indexes but
foreign exchange markets. Foreign exchange markets and stock prices are closely related and have
been commonly utilized by fundamentalist investors to predict future trends (Bashir et al., 2019).
Given the event uncertainty to market, researchers analyze its effect through Foreign exchange
markets.
Bashir et al. (2019) applies and explains detrended fluctuation analysis (DFA) and
detrended cross-correlation method to analyze the relationship between stock markets and foreign
exchange rates while influenced by the given event effect to them. According to theory analyzed by
Bashir et al. (2019) if the exchange rate is competitive it will affect the trade of the country and its
economy increasing its product profitability and price while on the opposite note depreciation will
decrease the profitability, firm value and their share prices. Which means that changes in currency
will suggest the effect of the event (Bashir et al. 2019). As with other methods this one also takes an
“event-window” and “estimation-period” which is also do not have a strict rule for how long period
should last but also considered to be not as small as few hundred days for major events. To
conclude, this method tries to capture the effect of the events not directly through the markets but
through the exchange rates of countries currency. This method also benefits not only on the
portfolio management and decision making but could also be applied by making and deciding on
financial policies of the country which could benefit efficiently in controlling the situation the
events cause.
1.5 Overview of Previous Research on “BREXIT” Effect to Financial
Market
On the 23rd of June 2016, UK citizens voted to leave European Union also known as EU
and this event was named “BREXIT”. “BREXIT” had an immediate effect on stock markets.
Guedes, Ferreira, Dionisio and Zebende (2019) in their work analyzed how “BREXIT” affected EU
stock markets. Their results showed negative effects in most of the cases except Malta, Bulgaria,
Slovenia. Also Guedes, Ferreira, Dionisio and Zebende (2019) results show that UK indexes are
Page 20
20
less integrated with the other EU countries and that it prevent greater well-being of countries. But
they conclude one plus of it that if some asymmetric shock occurs they could mitigate it through
pound sterling exchange rate.
Other authors work analyzes “BREXIT” effect through the view of UK companies
(Oehler, Horn and Wendt, 2017). Authors analyze “BREXIT” effect on daily returns of companies
depending on the internationalization level. They assume that with higher in internationalization
level companies should have less negative effect to their prices due to the “BREXIT”. Oehler, Horn
and Wendt (2017) in their work accepted the main hypothesis and showed that companies with
more domestic sales experienced larger losses.
The third research about “BREXIT” effect on stock markets of UK was done in two
phases: pre-referendum and post-referendum (Shahzad, Rubbaniy, Lensvelt and Bhatti, 2019). Their
results were similar to previous mentioned authors in this paragraph, they found out that results
were negative at first, but after the referendum their results differ to positive which means that at
some point some investors started to see positive effect of “BREXIT”.
Fourth research work on this topic was also analyzing two periods of “BREXIT” (Bashir et
al., 2019). Their results showed the same patterns in exchange and stock market rates as in
(Shahzad, Rubbaniy, Lensvelt and Bhatti, 2019) work. It showed immediate negative result at first
phase of “BREXIT” or as it called pre-referendum and some positive patterns in post-referendum
phase (Bashir et al., 2019). Authors also analyzed not only UK stock market parameters but also
correlation between other four largest Europe region economies and find out immediate negative
results between UK and Germany, France, Netherlands.
Completely different result in their work got Ramiah, Pham and Moosa (2017). Authors
estimated the effect of “BREXIT” on the industrial level and find out that in total the “BREXIT”
has a negative result for many industries and the effect of “BREXIT” is negative especially for the
UK market. Their results showed a wider perspective for the UK market on firm level and shows
that in longer-period the UK is making losses for its market.
Another author work estimating effect of “BREXIT” to financial markets were written by
Boulton and Bacon (2018). In their work they analyzed the effect of “BREXIT” announcement that
UK leaves EU on 23rd of June in 2016 to ten firms that are exposed in Britain and the European
Union markets. Boulton and Bacon (2018) selected to estimate the period of 180 days before the
event and 30 after. For this estimation they selected previously mentioned event method and their
results showed that 5 days until the announcement of news indexes spiked up because of the
optimistic emotions of investors that UK won’t leave EU. Emotions of investors were positive due
to the pools released on June 20th and answers showed declining chances of “exit” of the UK.
Nevertheless 3 days until “BREXIT” stock prices significantly decreased because pools didn’t work
Page 21
21
and the exit was inevitable. Boulton and Bacon (2018) work showed very similar results to Ramiah,
Pham, and Moosa (2017) work and showed declines in stock prices on the event days and after it.
But as many previous authors work showed the prices spikes up after a few weeks after the
referendum of the “BREXIT” (Shahzad, Rubbaniy, Lensvelt and Bhatti, 2019). The same results are
reflected in Boulton and Bacon (2018) work and the prices rises after 21st day of the referendum.
Table 3
“BREXIT” Effect To Financial Markets
Authors name Results
Industrial – company level researches
Oehler, Horn and
Wendt (2017)
United Kingdom companies with lower internalization level experienced
larger loses than the ones with higher internalization.
Ramiah, Pham, and
Moosa (2017)
Ramiah, Pham, and Moosa (2017) also find out negative results on
industrial level for the UK market.
Boulton and Bacon
(2018)
10 biggest stock indexes were tested in these authors work. Prices
significantly increased 5 days until referendum, huge loses followed 3 days
until referendum and only became positive a few weeks after referendum
Financial market level researches
Shahzad,
Rubbaniy, Lensvelt
and Bhatti, (2019)
Bashir et al.,
(2019)
Loses in financial markets in pre-referendum phase and gains in post-
referendum phase. Also Bashir et al., (2019) found immediate negative
correlation between UK and Germany, France and Netherlands.
Guedes, Ferreira,
Dionisio and
Zebende (2019)
Negative effect for UK and other more related countries to the UK. Results
didn’t change or change insignificant were in Malta, Bulgaria and Slovenia.
Source: compiled by author based on the research referenced in the table
All of the results of previous authors work about “BREXIT” effect to financial markets of
United kingdoms is summarized in the table before (Table 3). All of the researches are categorized
in two categories: industrial and financial level. As we can see from the results on industrial level
United Kingdom experienced loses according to all researchers. The results do not differ much and
in financial level showing that financial markets experienced loses due to the “BREXIT”.
It is also very important to analyze what effect “BREXIT” had on other countries who
were more or less related to UK and were one of the trading partners. Abraham (2018) did a
research what effect “BREXIT” had on New Zealand stock market. Being the close trading partner
to UK, New Zealand should also experience negative results regarding negative news about UK and
Page 22
22
“BREXIT” in particular. Even though analyzes suggests that “BREXIT” could open up new
business opportunities to New Zealand in the long term because New Zealand is not so dependent
on UK as it was in 1970 (Abraham, 2018). Abraham (2018) as many others authors mentioned
earlier did an event study estimation to answer what effect has UK’s announcement to leave EU.
Abraham (2018) analyzed 50 largest, eligible stocks listed on the main board of New Zealand stock
market index (NZX 50 Index). He selected the “event-window” of 40 days. 20 days before the
referendum announcement and 20 days after. Abraham (2018) results showed significant negative
results in period of time of 16 days before the announcement and 16 days after. In this period of
total 32 days the stock market index decreased while after it become positive and started rising.
That means that compared to Boulton and Bacon (2018) results Abraham (2018) results started to
show significant results much earlier than in UK which means that New Zealand reacted more
significantly to financial uncertainty due to the “BREXIT”. New Zealand’s market got better and
positive results faster than UK’s in post referendum phase. That is probably because of the
investment movement from UK to other countries (Abraham, 2018).
Another work analyzing the effect of “BREXIT” to other countries stock market indexes
were calculated by Madhavi and Reddy (2018). These authors were estimating the effect of
“BREXIT” on India’s stock market and its different industry sectors. Authors estimated “BREXIT”
influence for the India’s industry sectors using “event-study” method. Madhavi and Reddy (2018)
used “ARCH” and “GARCH” methods for estimating volatility of those sectors. They were
interested to see which sectors would be affected mostly due to the “BREXIT” referendum. Results
of their research showed significant effect of “BREXIT” to volatility of markets. But the results
were significant in volatility before and after the actual event date. On the event date volatility
didn’t change significantly which means that market reacts before the event and after it. Lastly
authors describe the limitations of their research explaining that longer period is necessary to
estimate the effect of the “BREXIT” because many other external factors affect stock markets and
only longer period could evaluate the effect of “BREXIT” more precisely.
Third research made regarding “BREXIT” effect to other country economies was made by
Morales and Andreosso-O’Callaghan (2019). These authors analyzed how “BREXIT” effected
China’s stock market. Authors estimate the abnormal returns and volatility of mainland China,
Hong Kong and Taiwan’s stock markets. Results showed that these stock markets do not exhibit
any abnormal returns due to the “BREXIT” events. The same goes with the estimation of volatility
to these markets, the effect of “BREXIT” is insignificant to volatility (Morales and Andreosso-
O’Callaghan, 2019). China’s stocks markets do not seem to be panicking and overreacting to recent
major political events that happened in the EU (Morales and Andreosso-O’Callaghan, 2019). Of
course, China should be monitoring major events happening in the world because of the trading
Page 23
23
possibilities in EU or other regions but as the result show political events are not one of them. All
researches findings summarized are presented in the table below (Table 4).
Table 4
“BREXIT” Effect To Other Financial Markets
Authors name Results
Abraham
(2018)
Negative Results in New Zealand’s financial markets 16 days before the
announcement of “BREXIT”. The results got positive only 16 days after the
“BREXIT”
Madhavi and
Reddy (2018)
Authors tested volatility of India’s financial market due to the “BREXIT”.
Results were significant and showed differences in volatility due to the
“BREXIT”. Volatility changed days before the event and after it, but showed no
significant effect on actual date of event.
Morales and
Andreosso-
O’Callaghan
(2019)
Authors analyzed how Chinese stock market were affected due to the “BREXIT”.
No significant effect to Chinese stock market was showed during the period of
“BREXIT”. Authors tested the returns and volatility of the market and didn’t find
and significant results to it.
Source: compiled by author based on the research referenced in the table
To summarize all chapter about theoretical aspects of finance theories, models and political
stability and their effect on financial markets this can be stated. Traditional finance theory is finance
theory based on the logic that investor is rational, risk-averse and aimed to maximize “expected
utility” at every decision he makes. Behavioral finance theory is finance theory that considers
investor not as ration but as “normal” which means that investors can make mistakes in their
calculations. Behavioral finance theory started to develop more rapidly after traditional finance
theory failed to explain financial market anomalies like market bubbles and etc.
Furthermore analyzes of previous researchers determined that stock prices movements
could be influenced by large number of factors from macro-economic to micro-economic levels.
Macro-economic factors such as inflation, export, exchange rate, industrial production, currency
supply, interest rate and etc. were presented in this work. Micro-economic factors affecting prices
of stocks named in this research were sales, equity, debt, liquidity, Tobins Q, debt/assets ratio,
foreigner control, dividends announcement, size of the market and etc. Psychological factors
affecting stock prices named in this work were based on two categories: cognitive bias and
emotional bias. Cognitive bias category consists of anchoring, adjustment, framing, conservatism,
availability, mental accounting, gamblers Fallacy, Representativeness factors. Emotional bias
category consists of endowment bias, loss aversion, optimism, status quo, overconfidence, herding,
regret factors.
Page 24
24
Analyzes of how political stability affects the financial markets it can be stated that
political stability, certainty is very necessary to the development of countries and stability of their
financial systems. Political stability is one of the key factors affecting the financial markets and its
instability could influence collapse of the financial market. Political uncertainty lowers the profits
of investors which lead to larger investment in domestic countries and foreign direct investment
loss.
Lastly to evaluate the effect of such events that increased political instability of country or
etc. and their effect to financial markets models such as “event-study”, GARCH, copula functions
and detrend fluctuation analysis were presented. “Event-study” helps to capture how event affect
the actual prices of markets while GARCH focuses on volatility of the market. Previous works
estimating “BREXIT” event effect for financial markets determined that UK financial market
experienced larger loses in pre-referendum phase and only recovered a few weeks after it (Shahzad,
Rubbaniy, Lensvelt and Bhatti, (2019). Bashir et al., (2019) and Guedes, Ferreira, Dionisio and
Zebende (2019) also find out negative effect between UK and other EU countries. Countries with
strong trade relations with UK also experienced negative effect for the financial markets. In this
case New Zealand and India experienced it. On industrial level companies with lower
internalization level experienced larger loses than the ones with higher internalization. Most of the
UK economy sectors experienced loses due to the “BREXIT” vote.
Page 25
25
II. “BREXIT” NEWS EFFECT FOR “LONDON STOCK
EXCHANGE” AND INDIVIDUAL COMPANIES
METHODOLOGICAL JUSTIFICATION
In the second chapter of the work research methodical part is analysed. In the first part
relevance and aim of the research will be analysed and stated. Furthermore the hypothesis of the
research based on previous works will be constructed. Lastly the empirical model of the research
and data necessary to carry out the research will be presented.
2.1 Relevance and Aim of the Research
In literature review traditional and behavioral finance theories were analyzed. What was
noted from the literature review was that not everything can be estimated using traditional finance
models. In some cases traditional finance models lacks the estimation and can’t get a grasp of
psychological factors affecting the decision making by financial players. For example, traditional
finance models lack evaluations of human emotions such as fear, optimism, overconfidence, mental
accounting, anchoring, Gambler’s fallacy, Availability, Loss aversion, Regret aversion,
representativeness and etc. Emotions affecting the decision making in financial markets previously
analyzed in the literature review should be put in the estimation models in order to better capture
the effect of the event to financial markets.
Secondly, greater amount of the events regarding “BREXIT” is necessary to analyze to
better capture the effect of the “BREXIT” to United Kingdom financial market because all of the
previous researchers were only focused on the “BREXIT” announcement day after the vote on 23rd
of June in 2016. As the previous researches showed “BREXIT” announcement had a significant
effect before and after the referendum. Results showed significant negative effect for most countries
financial markets except for China, Malta, Bulgaria and Slovenia. Markets only recovered a few
weeks after the “BREXIT” announcement. Which indicates that markets estimated the real effect of
the “BREXIT” announcement and recovered to real value of it.
To begin with it is necessary to understand how political instability and uncertainty affect
financial markets. Soltani, Aloulou and Abbes (2017) in their work name political instability as
factor that could make financial markets collapse. Wisniewski (2016) in his work analyze political
uncertainty as the risk that is created by the unstable political environment which are reverberate in
financial markets and diminish shareholder wealth. Mnif (2017) in his work name political
uncertainty factors that drives financial markets. He names events like presidential elections,
terrorism attacks, military invasions/wars and civil overthrows of local government. Other authors
named after describes similarly what could be treated or described as political instability,
Page 26
26
uncertainty. Afterwards Wisniewski (2016) in his work name another political uncertainty that
effect stock markets called „political communications”. Wisniewski (2016) describes political
communications as news made by local government that could influence financial markets. Political
communications are separated from presidential elections, terrorism attacks and others because of
the impact on the financial market. For example terrorism attacks, military invasions and others
could increase political instability in country and have a strong negative impact for financial
markets while political communications could influence it on opposite way and have a positive
effect if the news decreases political instability in country.
The previous paragraph identified different political events that influences financial player
behavior in financial markets. Previously named political events cause fear, pessimism and regret.
Negative emotions could highly decrease total investments in local countries financial markets and
could cause financial crash of the market. Investors do not tend to invest that much in foreign
country assets due to the fact of political instability (Smimou, 2014).
Today the world face a new kind of political instability also known as “BREXIT”. In 2016
British referendum decided to leave European Union and this event were not taken lightly by the
financial markets all over the world and United Kingdom itself. As the previous authors works
showed United Kingdom experienced losses due to the “BREXIT”. Oehler, Horn and Wendt (2017)
in their work states that companies with lower level of internationalization experienced bigger loses
than the ones with higher level of internationalization. Ramiah, Pham and Moosa (2017) analyzed
the effect of “BREXIT” on industrial level and find out that consequences will be much higher in
the future than it was in 2016. “BREXIT” also affected other countries and their economies. For
example close trading partners like New Zealand and India experienced major changes in the
markets due to the “BREXIT”. New Zealand started to experience loses much earlier than UK
which states that New Zealand financial markets are more elastic to uncertainty and instability than
UK. India’s financial markets due to the “BREXIT” experienced large changes in volatility during
the period of pre-referendum and post-referendum, meaning that UK political well-being has a
significant role on India’s financial market.
Various researches makes an insightful conclusions about what effect “BREXIT” will have
on market returns. Additionally, it is relevant to understand and how later news regarding
“BREXIT” affects the financial market. That is why the main purpose of this research will be
capturing how “BREXIT” news affects the United Kingdom markets returns. Analyzing what
financial behavior is influenced by the “BREXIT” could help to local and foreign governments and
investors react to situation more rationally.
After analyzing the effect of “BREXIT” news to United Kingdom financial market it is
necessary to take a deeper look on how it affects individual companies in United Kingdom. Oehler,
Page 27
27
Horn and Wendt (2017) and others analyzed companies with higher and lower levels on
internalization and find out that announcement of “BREXIT” had a larger negative effect for
companies with lower internalization level. Ramiah, Pham and Moosa (2017) in their work
analyzed the effect of “BREXIT” on industrial level and their results showed that the banking,
travel and leisure sectors were experienced largest loses compared to other sectors. Boulton and
Bacon (2018) estimated how “BREXIT” affected ten firms, traded on New York Stock exchange,
that are also heavily linked with the United Kingdom market. Their results showed negative return
on all of them regarding “BREXIT”.
To conclude the reasoning of research it is necessary to understand whether the “BREXIT”
news has an effect and influence to financial markets in a longer time perspective because
according to previous works findings it has a significant effect to financial markets worldwide.
Another reason for this research is that it is also necessary to understand whether “BREXIT” news
announced later after the referendum to leave EU affected the markets negatively like it did at first.
As it was already mentioned “BREXIT” is the event like no others and understanding its effect
could give an insightful notes towards future researches with similar events. The “BREXIT” for this
research is classified as political instability, uncertainty and fear rising event that influences
negative financial markets reaction. In this research effect of “BREXIT” news will be analyzed to
United Kingdom financial markets and individual companies in different industrial sectors listed on
United Kingdom financial market. In existence of negative effect for financial markets in UK
financial markets the measures for future financial policy decisions should be taken because
political instability could have a large negative impact on financial markets and could even cause
financial collapses of it.
The aim of this research is to analyze how “BREXIT” news effects and influences
United Kingdom financial market, individual companies and various sectors.
2.2 Research Hypothesis
In this chapter research hypothesis will be formulated. Formulation of hypothesis is based
on previous research exploring “BREXIT” effect to financial markets. First of all it is important to
assess if later news regarding “BREXIT” had a significant effect for United Kingdom financial
market. Furthermore, what is important to evaluate whether the news regarding “BREXIT” still had
a negative influence to United Kingdom financial market as it did at the time around referendum for
“BREXIT”. In the previous researches authors estimate the effect of one event and do not include
others that happened later on. Ramiah, Pham and Moosa (2017), Oehler, Horn and Wendt (2017),
Boulton and Bacon (2018) and other researchers named in previous chapters only estimate the
announcement of “BREXIT” day which only estimates one event effect for the financial market.
Due to this fact this research will evaluate and focus on the effect of all other key dates of
Page 28
28
“BREXIT” happened later on after the referendum to United Kingdom financial market. As it was
already mentioned in the previous chapters the effect of “BREXIT” announcement for the United
Kingdom financial market was negative. Oehler, Horn and Wendt (2017), Ramiah, Pham and
Moosa (2017), Boulton and Bacon (2018) find out negative results on industrial level and estimated
the effect of “BREXIT” announcement to different sectors. Guedes, Ferreira, Dionisio and Zebende
(2019), Shahzad, Rubbaniy, Lensvelt and Bhatti, (2019), Bashir et al., (2019) in their work analyzed
the effect of that event on a country level to largest stock market indices. Results showed negative
returns of markets in the estimation period of the event. Based on that logic first hypothesis of the
research were formulated:
H1: “BREXIT” news have a significant negative effect on London stock exchange.
The main United Kingdom financial market is London Stock Exchange. In this research
London Stock Exchange index price was taken as the representativeness for the whole countries
financial market. The effect for daily returns due to the “BREXIT” news will be calculated to
London Stock Exchange. Guedes, Ferreira, Dionisio and Zebende (2019), Shahzad, Rubbaniy,
Lensvelt and Bhatti, (2019), Bashir et al., (2019) in their works estimated the effect of “BREXIT”
referendum to price changes of market and due to this reason the changes in prices will be taken as
markets reaction to “BREXIT” events. Hypothesis will be accepted if at least half of the events will
have a significant negative effect in at least one of the estimated “event-windows”.
The second hypothesis is based on Boulton and Bacon (2018) research. Authors estimated
the effect of “BREXIT” announcement to 10 biggest companies traded on the NYSE (New York
Stock Exchange) who is very closely related to the United Kingdom financial market. Authors
estimate how much “BREXIT” affected daily returns on sixty days period around the
announcement day. They calculated the abnormal returns for 30 days before the event and 30 days
after the event. Their results showed growth in returns on the fifth and fourth day until vote
following with large decline in returns on the last three days until the vote. Returns bounced back
only on 21st day after “BREXIT” vote. Based on that logic second hypothesis were formulated:
H2: “BREXIT” news have a significant negative effect on individual companies in United
Kingdom listed on London Stock Exchange.
To test this hypothesis twenty-five companies with largest market capitalization, which
were listed on London stock exchange, were analyzed. Effect for the returns due to the “BREXIT”
news will be calculated for them and after that the hypothesis will be accepted if at least half of the
events will have a significant effect for at least half of the companies in at least one of the estimated
“event-window”.
Furthermore, it is important to analyze which sectors are affected by the “BREXIT” news
more than the others. Ramiah, Pham and Moosa (2017) estimated how “BREXIT” vote affected
Page 29
29
different United Kingdom sectors. Ramiah, Pham and Moosa (2017) based on discussions in
“Financial Times” rise the hypothesis that financial and banking sectors should experience negative
returns due to the “BREXIT” vote. Their results approved this hypothesis and those two sectors
experienced negative returns of 15% during event period. Other sectors affected most were leisure
and travel and Oil and gas producers. Based on these authors work the third hypothesis will be
tested:
H3: “BREXIT” news have a stronger negative effect to financial sector companies
compared to other sector companies.
As it was already mentioned there are twenty-five companies from different sectors with
largest market capitalization listed on London Stock Exchange whose daily returns affected by
“BREXIT” events will be calculated in order to understand whether “BREXIT” events happened
later on had any effect for them. That list contains seven financial sector companies out of twenty-
five. If at least fifty percent of financial sector companies will have a negative abnormal returns in
at least half of the “BREXIT” events the hypothesis will be accepted.
Main focus of this paper to analyze how “BREXIT” events effects returns of UK financial
markets and individual companies. It is known that “BREXIT” raises a lot of questions what will
happen not only to the economy of UK but the whole European Union. “BREXIT” is the event like
no others happened in the last decades which suggests that it is necessary to understand how it will
affect the financial markets so that in future policy changes or suggestion for investors could be
taken in advance of those events.
Nevertheless these kind of events that creates political instability and uncertainty are
known for increasing “Herding” financial behavior in financial markets. Vidanalage, Shantha
(2019) in their work analyze “herding” behavior in the Colombo Stock exchange in Sri Lanka.
Authors analyze the periods of uncertainty in Sri Lanka and presence of “Herding” in the market.
They test how “Herding” behavior changed during different periods of Sri Lanka existence.
Vidanalage, Shantha (2019) state that in 2000-2009 years period the herding should have been
higher and strongly evident due to the fact of political instability, uncertainty and Civil War. Indars,
Savin, Lubloy (2019) in their work states that political turmoil strongly influences the financial
markets because it is strongly related to uncertainty of country. Indars, Savin, Lubloy (2019) also
state that “herding” is highest in the financial uncertainty times and in crises periods, giving the
examples of Asian crises in 1997-1998, subprime crisis in United States of America and etc.
According to them “herding” should be stronger in those periods of time. Which formulates the next
hypothesis for the research:
H4: “BREXIT” news trigger financial herding behavior in London stock exchange
Page 30
30
Hypothesis will be accepted if at “event-window”, changes in individual company stock
volume will be significant at least in fifty percent of the “BREXIT” events for half of the
companies.
After argumentation and formulation of hypothesis that needs to be tested in order to
understand better if “BREXIT” events have an effect for United Kingdom financial market the
empirical model needs to be adapted and constructed.
2.3 Logic and Argumentation of Research Model
In order to test the hypothesis that was named in the previous chapter it is necessary to
analyze the method and logic of the research. First of all for this research the “event-method” will
be used in order to accept/reject hypothesis. The event study methodology applied here is based on
the efficient market hypothesis (Chen, Siems 2004). Efficient market hypothesis assumes that as
new information stemming from an important unpredictable event becomes available, market agents
will take the information into account and will re-evaluate values of individual firms given
economic, environmental, political, social and demographic changes that the exogenous event might
bring about. The power of this methodology is its ability to trace such “abnormal” changes, because
it follows the general valuation of many investors that (re)examine all the available data for the
estimation of the market value of each traded stock (Schwert, 1981). Based on logic that the main
purpose of the research is to test the actual returns of financial markets influenced by „BREXIT“
news „event-study“ fits more rather than other methods because other methods such as GARCH,
EGARCH and others estimate the volatility of the market rather than returns of financial market.
Secondly, „event-study“ model fits the estimation of hypothesis more than the others due to the
reason that research is interested in individual companies and their returns and how „BREXIT“
news affect the different sectors. Different models estimating volatility won‘t be able to answer
whether the effect for some sectors such as financial like in the H3 are more affected than the
others. Lastly GARCH and others models can‘t get a grasp of „herding“ behavior like the „event-
study“ potentially can when estimating the abnormal changes in volume of stocks and indexes.
Boulton and Bacon (2018) in their work also used “event-study” methodology for
estimation the effect of „BREXIT” vote to ten biggest stock indexes. Boulton and Bacon (2018)
estimated period of 60 days for the “event-study”. Estimation contained 30 days before the event
and 30 after event. Abraham (2018) in his research about New Zealand financial market and
„BREXIT” effect to it estimated 40 days period. He estimated 20 days before the event and 20 days
after it. Ramiah, Pham and Moosa (2017) selected period of 10 days for calculating abnormal
returns. Oehler, Horn and Wendt (2017) in their work modify time period for 5 minutes. This
analyzes of “events-window” selection by different authors suggest that there is no strict rules in
selecting period for calculation period.
Page 31
31
Boulton and Bacon (2018) in their work calculates the companies reaction using “event-
study” method a.k.a. risk adjusted return model (RAR) in order to calculate how “BREXIT” vote
day affected 10 largest companies traded on NYSE which are strongly related to United Kingdom
market. Authors calculate percentage changes in the stocks prices in that specific period. Risk
adjusted return model calculates the difference between the actual daily prices and expected which
are calculated by adding the β (the slope) with α (intercept) multiplied with daily returns of the
benchmark. Furthermore the abnormal returns are calculated subtracting actual daily returns of the
stock with expected. Furthermore the cumulative abnormal, average and cumulative average
abnormal returns are calculated in the similar way. Concept of the Boulton and Bacon (2018) model
was used in construction of this research model when calculating the effect for the individual
companies listed on London Stock Exchange.
Oehler, Horn and Wendt (2017) use similar approach as Boulton and Bacon (2018) for
calculating “BREXIT” voting period in their work. The difference between their works is that
Oehler, Horn and Wendt (2017) apply more control variables in their work such as industry type the
company is working, its market capitalization and their domestic sales. Oehler, Horn and Wendt
(2017) also use a lot shorter estimation period and only calculates 5 minutes period for detecting
noise trading.
Morales and Andreosso-O‘Callaghan (2019) introduces their econometrical model for
calculation of China‘s financial market index movements influenced by „BREXIT” and Donald
Trump election. Authors applies different control variables such as economical policy uncertainty
(EPU) index and volatility index (VIX) to better grasp the effect of events.
Based on previous authors empirical models and works the empirical model for this
research was constructed. As it was already have been mentioned the estimation period depends on
what research is trying to prove so event period depends on hypothesis and tests. This research
focuses to grasp the effect for short time period before and after the event to evaluate the power of
the event for the market and the expectations and efficiency of the market. “event-window” selected
for the research is 10 days before the event and 10 days after. Event day in the model is considered
0. Daily data series over the period 3rd of May 2016 to 17th of April 2020 were used for the
research.
Equation for calculating percentage change in returns for London Stock exchange is named
below:
𝑅𝐿𝑜𝑛𝑑𝑜𝑛,𝑡 =𝑃𝐿𝑜𝑛𝑑𝑜𝑛,𝑡 − 𝑃𝐿𝑜𝑛𝑑𝑜𝑛,𝑡−1
𝑃𝐿𝑜𝑛𝑑𝑜𝑛,𝑡−1∗ 100 (4)
Page 32
32
𝑅𝐿𝑜𝑛𝑑𝑜𝑛,𝑡 is return of the London Stock Exchange on the time t, 𝑃𝐿𝑜𝑛𝑑𝑜𝑛,𝑡 is closing price
of the day t, 𝑃𝑠,𝑡−1 is closing price of previous day. The holding period returns of benchmark stock
(S&P 500) was calculated using the equation written below:
𝑅𝑆&𝑃 500,𝑡 =𝑃𝑆&𝑃500,𝑡 − 𝑃𝑆&𝑃500,𝑡−1
𝑃𝑆&𝑃500,𝑡−1∗ 100 (5)
𝑅𝑆&𝑃 500,𝑡 is return of the S&P500 on time t, 𝑃𝑆&𝑃 500,𝑡 is current day closing price of
S&P500 and 𝑃𝑆&𝑃 500,𝑡−1 is closing price of S&P500 on previous day. Using the same formula the
daily returns of the individual companies were calculated:
𝑅𝑠,𝑡 =𝑃𝑠,𝑡 − 𝑃𝑠,𝑡−1
𝑃𝑠,𝑡−1∗ 100 (6)
𝑅𝑠,𝑡 is return of the stocks s on time t, 𝑃𝑠,𝑡 is current day closing price, 𝑃𝑠,𝑡−1 is closing
price of previous day.
This equation is also applied to calculate the daily returns of FTSE 100.
𝑅𝐹𝑇𝑆𝐸 100,𝑡 =𝑃𝐹𝑇𝑆𝐸 100,𝑡 − 𝑃𝐹𝑇𝑆𝐸 100,𝑡−1
𝑃𝐹𝑇𝑆𝐸 100,𝑡−1∗ 100
(7)
A regression analysis was partaken to calculate the α (the intercept) and the β (the slope of
the regression line). This was completed by using the actual daily return of each company
(dependent variable) and the corresponding S&P 500 index (independent variable) over the course
of the pre-event period, -30 days to -10 days.
In this study in order to get normal expected returns, the risk-adjusted return method
(RAR) was used. The expected return for each day of the event period from day -10 to day +10 is
calculated using formula:
𝐸𝑥(𝑅)𝑠,𝑡 = 𝛼 + 𝛽𝑅𝑚,𝑡 (8)
Ex(R) is expected return of the stock and 𝑅𝑚 is return of the market (S&P 500 when
measuring effect for London stock exchange and FTSE 100 when measuring effect for largest
companies in UK). After calculating expected return of the stocks, excess return was calculated
using formula:
𝐴𝑅𝑠,𝑡 = 𝑅𝑠,𝑡 − 𝐸𝑥(𝑅) (9)
Where 𝐴𝑅𝑠,𝑡 is abnormal return for the stock s on time t, 𝑅𝑠,𝑡 is actual return of the stock s
on time t and 𝐸𝑥𝑅 is expected return of the stock. Cumulative excessive returns will be calculated
by adding excess returns of each day from day -10 to day +10.
Cumulative abnormal returns (𝐶𝐴𝑅𝑠,𝑡,𝑖) are calculating using this formula:
𝐶𝐴𝑅𝑠,𝑡,𝑖 = ∑ 𝐴𝑅𝑠,𝑡
𝑖
𝑡
(10)
Page 33
33
Where 𝐶𝐴𝑅𝑠,𝑡,𝑖 is cumulative abnormal returns of stock s in period form t to i. The “event-
window” in research selected from 10 days before the event and 10 days after. Graph of CAR will
be plotted for the event period i.e. day -10 to +10.
Additionally when calculating the abnormal returns of companies the average abnormal
returns will be calculated in order understand whether the companies outperformed the average
return of sample. Average abnormal returns will be calculated using formula:
𝐴𝐴𝑅𝑡 =𝐶𝐴𝑅𝑡
𝑛
(11)
In this equation 𝐴𝐴𝑅𝑡 is average abnormal returns for all stocks on the given day, 𝐶𝐴𝑅𝑡 is
cumulated abnormal returns of all companies listed in the sample and n is the number of companies
in the sample. After calculating the average abnormal returns for all stock on the given day the
cumulated average abnormal returns will be calculated for the event period. The cumulated average
abnormal returns is calculated by this formula:
𝐶𝐴𝐴𝑅𝑠,𝑡,𝑖 = ∑ 𝐶𝐴𝑅𝑠,𝑡,𝑖
𝑖
𝑡
(12)
Where 𝐶𝐴𝐴𝑅𝑠,𝑡,𝑖 is cumulative average abnormal returns of all stocks s in period form t to
i. The “event-window” in research selected from 10 days before the event and 10 days after.
Additional event periods will be added in the research. Abraham (2018) in his work
estimated periods of not only for 20 days before the event ant 20 days after the event but also
additional ones. Abraham (2018) also calculated periods of 10, 5, 2 days before and after the event
to better understand the significance of the event the closer its happening days. In this paperwork
periods of 5 and 2 days before and after the event will be calculated. These additional two periods
will give a better understanding at which period of time the event has a strongest effect for financial
market and companies. Nevertheless, two additional periods of time will be added in this research.
Periods of 10 days before the event until the event day and then 10 days after the event will be
estimated. Those calculations will be made in order to better understand whether the event meets its
expectations. The period until the event day will show the expectations of the financial market and
events expected effect for it and 10 days after will evaluate the real value of event and show the
efficiency of the market. This model will be applied to calculate the abnormal returns of London
Stock Exchange and individual companies listed London stock exchange. Daily returns of each
individual stock will be calculated as well as for benchmark and etc. All of these calculations and
equations named previously will be used to accept or reject hypothesis H1, H2, H3. As it was
identified previously the H1 will be accepted if at least half of the events will have an effect the
returns of London stock exchange. H2 will be accepted if at least half of the events will have an
Page 34
34
effect for at least half of the companies in estimating period and lastly the H3 will be accepted if at
least half of the events will have an effect for the financial sector companies compared to others.
In order to accept or reject H4 similar approach was applied. Cumulated Abnormal
Volume were calculated using “event-study” method to detect the herding behavior. Significant
abnormal changes in volume would signalize about existing herding behavior in financial market.
The herding behavior could be negative or positive meaning that some BREXIT news influence
negative herding and others positive.
2.4 Data Sample of the Research
In order to answer whether the „BREXIT“ events have a significant effect for the United
Kingdom financial market 17 different events were selected. Selected dates includes events from
2016 until 2020. These specific events were selected because of their significance towards United
Kingdom relationship with European Union. All of the events with their actual happening day are
named in the table 5 (starts in page 34, ends in 35).
Some of the events listed in Table 5 are discussed in more detail in order to better
understand what happened on that event. The event on 17th of January in 2017 Theresa May in her
first substantial speak announced that she desired to leave EU without staying in the single market.
“Article 50” event finalized the leaving of United Kingdom from European Union. “Divorce bill” is
the sum of money United Kingdom had to pay for European Union to settle the United Kingdom
share of the financing of all the obligations undertaken while it was a member of the European
Union. Lastly “Chequers agreement” stands for the event happened after the Europeans Union
passed “withdrawal bill” that became a law at the end of June. Theresa May took her cabinet back
to country in order to sign off the collective position for the rest of the “BREXIT” negotiations with
the European Union. “BREXIT” secretary David Davis resigned Theresa‘s May new plan.
Table 5
Key dates of the “BREXIT”
Date News
2016-07-16 Theresa May becomes Prime Minister
2017-01-17 „BREXIT“ means „BREXIT“
2017-03-29 May triggered Article 50 of the Lisbon Treaty
2017-06-08 May lost her parliamentary majority
2017-12-08 UK and EU agrees on „divorce bill“
2018-07-06 “Chequers” agreement” is finalized
2018-11-25 At a special meeting of the European Council.
2019-01-15 The First „meaningful vote“ is held on the Withdrawal Agreement in the UK
„House of Commons“
2019-03-12 The Second „meaningful vote” on the Withdrawal Agreement.
2019-04-12 UK’s deadline for leaving the EU was pushed back to 31 October
2019-06-24 Theresa May set a resignation date of 7 June
Page 35
35
Table 5 continuation
Note: Table has been created by author using data selected from the news portals ‘The week’, ‘BBC news’ and others
Furthermore as it was already mentioned in previous paragraphs 25 different companies
with largest market capitalization listed on London stock exchange is selected for answering the
hypothesis H2 and H3. As it can be seen from the table 6 (starts in page 35, ends in 36) companies
included in to the sample comes from different sectors. Largest number of companies included in
the sample comes from the consumer staples and financial sectors. Sample consists of 7 different
companies for each sector. Furthermore there are 3 companies from industrial sector, 2 from energy
and 2 from healthcare. Last 4 companies from the sample comes from basic materials,
telecommunications, consumer discretionary and utilities sectors.
Table 6
2019-07-24 Boris Johnson wins the Conservative Party leadership race
2019-09-04 After voting to take control of Commons business for the day, members of
parliament backed a bill blocking a 31 October „no-deal BREXIT“.
2019-10-02 Boris Johnson had made a formal proposal to the EU setting out his alternative to
the Irish backstop
2019-10-19 A special Saturday sitting of Parliament was held to debate the revised withdrawal
agreement
2019-12-12 Boris Johnson won general election and broke the Parliamentary deadlock.
2020-01-30 Johnson has passed his withdrawal agreement and paved the way for the UK to
leave the EU
FTSE 100 index largest companies (Market Cap.)
Company Name ICB Industry
Market Cap
(£m)
1 ROYAL DUTCH SHELL PLC Energy 156 941,61
2 HSBC HOLDINGS PLC Financials 112 088,46
3 ASTRAZENECA PLC Health Care 97 413,12
4 BP PLC Energy 92 515,66
5 GLAXOSMITHKLINE PLC Health Care 89 014,40
6 BRITISH AMERICAN TOBACCO PLC Consumer Staples 77 016,86
7 DIAGEO PLC Consumer Staples 70 331,72
8 UNILEVER PLC Consumer Staples 52 962,80
9 RIO TINTO PLC Basic Materials 51 283,21
10 RECKITT BENCKISER GROUP PLC Consumer Staples 44 579,19
11 VODAFONE GROUP PLC Telecommunications 39 969,07
12 LLOYDS BANKING GROUP PLC Financials 39 782,85
13 RELX PLC Consumer Discretionary 38 953,49
14 NATIONAL GRID PLC Utilities 35 286,05
15 PRUDENTIAL PLC Financials 35 167,68
16 BARCLAYS PLC Financials 29 101,06
17 ROYAL BANK OF SCOTLAND
GROUP PLC
Financials 26 401
18 TESCO PLC Consumer Staples 24 180,14
19 EXPERIAN PLC Industrials 23 950,37
20 CRH PLC Industrials 22 510,88
21 ASSOCIATED BRITISH FOODS PLC Consumer Staples 20 789,36
Page 36
36
Table 6 continuation
Note: The table has been created by author using data from londonstockexchange.com
Companies that are listed in the sample will be named differently when presenting the
results of the research. Companies will be listed not in market capitalization order from largest to
smallest like in the table 6 (starts in page 35, ends in 36) but in the order according to the sector
they produce products or services in. Companies will be coded as numbers in the graphs of the
results. The numeric order of them are presented in the table 7 below.
Table 7
Individual Companies Numeric Code In Graphs Presented in Results
No. Company name
1 ROYAL DUTCH SHELL PLC
2 BP PLC
3 HSBC HOLDINGS PLC
4 LLOYDS BANKING GROUP PLC
5 PRUDENTIAL PLC
6 BARCLAYS PLC
7 ROYAL BANK OF SCOTLAND GROUP PLC
8 STANDARD CHARTERED PLC
9 LEGAL & GENERAL GROUP PLC
10 ASTRAZENECA PLC
11 GLAXOSMITHKLINE PLC
12 BRITISH AMERICAN TOBACCO PLC
13 DIAGEO PLC
14 UNILEVER PLC
15 RECKITT BENCKISER GROUP PLC
16 TESCO PLC
17 ASSOCIATED BRITISH FOODS PLC
18 IMPERIAL BRANDS PLC
19 EXPERIAN PLC
20 CRH PLC
21 BAE SYSTEMS PLC
22 RIO TINTO PLC
23 VODAFONE GROUP PLC
24 RELX PLC
25 NATIONAL GRID PLC
2.5 Limitations of the Empirical Model
Nevertheless this empirical model and research has some limitations. The first limitation is
lack of control variables to understand better the effect of the “BREXIT” events whether they
influence the movement of market prices or control factors which are not included in the model. For
22 BAE SYSTEMS PLC Industrials 20 239,80
23 STANDARD CHARTERED PLC Financials 20 168,10
24 IMPERIAL BRANDS PLC Consumer Staples 18 462,53
25 LEGAL & GENERAL GROUP PLC
Financials 18 224,14
Page 37
37
example Oehler, Horn and Wendt (2017) in their work used control variables such as domestic
sales, industrial level and market capitalization in to the model because due to the fact that those
besides „BREXIT“ vote could have had influence to the returns of the FTSE 100 index companies.
In this research similar control variables couldn‘t be applied due to the fact that companies selected
for the research were the ones with largest capitalization, with high domestic sales and from just a
few industries so the results would have been biased. Morales and Andreosso-O‘Callaghan (2019)
in their work where they estimate the effect of Donald Trump election and „BREXIT” vote added
economic policy uncertainty in their work and volatility index (VIX). These notes leaves a space for
future researchers to take in consideration and in to their empirical models some of the already
mentioned control variables to grasp a deeper understanding how „BREXIT“ news affect the United
Kingdom financial market.
Secondly the model isn’t capable of removing the duality of the events that happened in
the same period of time. For example there were 3 events that happened in similar periods of time
and their windows intersect. There was one event happened on 28th of August in 2019 another one
on 4tf of September 2019 and third one on 24th of September 2019. Due to this intersection of
“event-windows” two events had to be excluded from the sample. Events that were treated not so
important were excluded from the sample even though they could have had some effect for the
returns of companies from different sectors and London stock exchange.
Third limitation of the research is evaluation of „herding” behavior in United Kingdom
financial market. Due to the limitations of model itself it could be too hard to grasp the „herding“
behavior in the market because it doesn‘t evaluate investors psychological behavior so well as other
models evaluating „herding“ behavior in financial markets. Vidanalage, Shantha (2019) introduces
two existing models for market-wide herd behavior analyzes. The first one was introduced in 1995
by Christie and Huang and also known as cross-sectional standard deviation model (CSSD) and
second one was introduced in 2000 by Chang and his colleagues known as cross-section absolute
deviation model (CSAD). These two models would have been more fitted to understand whether the
“herding” behavior exists in United Kingdom financial market. Due to this fact the results of model
could be insignificant when evaluating „herding” behavior in the United Kingdom financial market.
To summarize the methodology chapter it is possible to say that political uncertainty
influences negative financial markets reaction and results of many previous works proves it right. If
the political uncertainty is not evaluated correctly by the market it could influence the collision of
one. Researchers and many scientists states that today we face the political uncertainty like no
others which is known as „BREXIT“. That is why it is so important to take a deeper look at political
uncertainties like „BREXIT“ and understand how they influence the financial markets of the local
and other countries. Understanding it could help prevent emotional decision making in investing
Page 38
38
world by the investors, also help to tame the market volatility and understand better what investing
strategy investor should select at that volatile time. As the analyzes of models showed the most
suited empirical model for testing the named three hypothesis is the „event-study” method used by
many authors in their previous researches. „Event-study” method helps trace such “abnormal”
changes, because it follows the general valuation of many investors that (re)examine all the
available data for the estimation of the market value of each traded stock. The research selected 17
different events of „BREXIT“ that happened through 4 year period starting from 2016 and ending
on 2020 to estimate their effect for the London Stock Exchange and individual companies listed on
it. Furthermore to estimate the effect on the company level 25 different companies with highest
market capitalization from different sectors were selected in the research. The companies in the
research comes from financial, industrial, consumer staples, consumer discretions, utilities,
telecommunications, healthcare and energy sectors. Most of the companies selected in the sample
comes from financial and consumer staples sectors. The expected results of the research should
show the negative impact for the returns of selected companies and market itself. Nevertheless the
financial sector should be influenced stronger and more negatively than the others due to the
findings of other researches. Lastly according to previous authors findings „herding” behavior
should be stronger on the days of the event period that is treated as political uncertainty days. That
is why another question for the research is to analyze whether „herding” behavior is present during
„BREXIT” news days. In the next chapter the results of empirical analyzes will be presented.
Page 39
39
III. EMPIRICAL RESULTS ON “BREXIT” EFFECT FOR
LONDON STOCK EXCHANGE, INDIVIDUAL COMPANIES AND
VARIOUS SECTORS
In this part of the Thesis the results of empirical model will be described and analyzed. To
begin with descriptive statistics of London stock exchange and individual companies will be
presented. Furthermore the calculations of how “BREXIT” news affected London Stock Exchange,
individual companies and various United Kingdom economical sectors will be presented. Later on
the discussion of results compared to previous researches will be analyzed. Lastly the limitations of
the empirical data and recommendations for future researches will be presented.
3.1 Descriptive Statistics of London Stock Exchange, Individual
Companies and Various UK sectors
Descriptive statistics will be presented divided by the companies sector. Descriptive
statistics of financial sector and Consumer staples companies will be separated from the others
sectors due to the fact that both of sectors have 7 companies in the sample. Distribution by sectors
the companies produce products and services will help to understand data more clearly. Estimation
period for the research is selected from 3rd of May 2016 to 17th of April 2020.
Table 8
Descriptive statistics of London Stock exchange index (2016-05-03 to 2020-04-17)
Variable Mean Median Max Min SD Skewness Kurtosis
London
Stock
exchange
Index
0,0009648 0,0004954 0,1427 -0,1016 0,015997 0,39310 12,595
Note: Table was created by author using empirical data of London Stock Exchange collected from investing.com
Descriptive statistics of London stock exchange presented in table 8 above shows that
mean is larger than the median meaning that there are major outliers in the end of the distribution.
That means that London stock exchange returns experienced larger gains due to the some events or
reasons in estimation period. Maximum daily returns for the day was 14.27% percent and largest
loses experienced by London stock exchange were -10.16%. Positive skewness indicates that if the
“random” returns would appear on some daily data for some reasons it probably would be positive
in London stock exchange case. Lastly if the Kurtosis index is above 3 it shows that extreme
changes in daily returns appear more frequently.
Page 40
40
Table 9
Descriptive statistics of Financial sector companies’ stock prices (2016-05-03 to 2020-04-17)
Variable Mean Median Max Min SD
HSBC Holding -7,1333e-005 0,00015495 0,063104 -0,10005 0,013209
Lloyds Banking
Group
-0,00075817 -0,00086526 0,11569 -0,23579 0,019730
Prudential -0,00024167 0,00077075 0,14365 -0,18255 0,021096
Barclay’s -0,00059557 0,00017177 0,12544 -0,19454 0,021968
Royal Bank of
Scotlands and Group
-0,00073813 -0,00039093 0,12858 -0,19899 0,022517
Standard Chartered -0,00024633 -0,00029395 0,067710 -0,12991 0,017854
Legal and General
Group
-6,5351e-005 0,00075132 0,15406 -0,22643 0,022317
Note: Table was created by author using empirical data of financial sector companies of FTSE 100 selected from
Investing.com
In the table 9 above results of Financial sector descriptive statistics is submitted. First of all
as it can be seen from the table 9 almost every financial sector company except “Standard
Chartered” has a lower mean than median, meaning that extreme changes in daily returns are on
negative side of the population. That means that financial sector companies experienced much more
extreme loses than gains during the research period. Nevertheless largest gain sections compared
with largest loses section shows that loses were much greater than gains for all of the companies in
the table 9. Lastly compared to London stock exchange standard deviation is larger for most
companies except “HSBC Holdings PLC” meaning that financial sector companies are more
volatile than market.
Analyzing descriptive statistics of consumer staples companies in table 10 (page 41) we
can recon some similarities between these two sectors. In most of the cases except for “Diageo
PLC” and “Tesco PLC” mean is lower than the median meaning that more than a half companies
same as the financial sector experienced higher extreme loses during estimation period. Minimum
and Maximum charts ensures that major outliers on low end of sample were a greater than major
outliers in the high end for most of the companies except for “Unilever PLC”. Standard deviation
for most of the companies are greater than the market meaning that they are more volatile compared
with London Stock Exchange index. The only companies in consumer staples sector with lower
standard deviation are “Diageo PLC” and “Unilever PLC”.
Page 41
41
Table 10
Descriptive statistics of Consumer Staples sector companies prices (2016-05-03 to 2020-04-
17)
Variable Mean Median Max Min SD
British American
Tobacco
-0,00034427 0,00000 0,073238 -0,11227 0,016698
Diageo 0,00037572 0,00019325 0,087220 -0,090933 0,012533
Unilever 0,00028800 0,00039882 0,12600 -0,074282 0,012941
Reckitt Benckiser
Group
-6,7918e-005 0,00000 0,077834 -0,078027 0,013697
Tesco 0,00034063 0,00000 0,093043 -0,094363 0,016533
Associated British
Foods
-0,00044217 0,00000 0,14159 -0,16624 0,017641
Imperial Brands -0,00086978 -0,00046494 0,11590 -0,13859 0,015859
Note: Table was created by author using empirical data of consumer staple sector companies of FTSE 100 selected from
Investing.com
Lastly in the table 11 (starts in page 41, ends in 42) rest of the companies from different
sectors were analyzed. Companies with higher mean than median in the table were “Astrazeneca
PLC”, “Glaxosmithkline PLC”, “Rio Tinto PLC” and “CRH PLC”. That indicates that companies
from “healthcare” sector which were two, didn’t experience more extreme loses than gains.
Nevertheless the minimum and maximum sections show that largest daily loses were higher than
gains for most of the companies except for “BAE systems”. Companies that were less volatile than
the market were “Astrazeneca PLC”, “Glaxosmithkline PLC”, “Relx PLC”, “Vodafone Group
PLC”, “Relx PLC”, “National Grid PLC”, “Experian PLC” and “BAE systems PLC”. Which means
that more than a half of companies in different sectors than financial or consumer staples are not so
volatile compared with the market. Energy sector companies “Royal Dutch Shell PLC” and “BP
PLC” are more volatile in this context and comparison with other companies and market index.
Table 11
Descriptive statistics of sample companies excluding financial and staple sectors prices
(2016-05-03 to 2020-04-17)
Variable Mean Median Max Min SD
Royal Dutch Shell -0,00024702 0,00037927 0,18551 -0,19354 0,018110
BP -0,00019324 0,00010761 0,19544 -0,21671 0,018401
Astrazeneca 0,00071034 0,00053831 0,086414 -0,16737 0,015520
Glaxosmithkline 0,00012066 0,00000 0,069193 -0,080667 0,012472
Rio Tinto 0,00057063 0,0014905 0,13709 -0,12539 0,019574
Vodafone Group -0,00070447 -0,00061877 0,10080 -0,12245 0,015657
Page 42
42
Table 11 continuation
Relx 0,00041942 0,00070697 0,076821 -0,11430 0,012642
National Grid -9,1823e-005 0,00049817 0,093541 -0,10186 0,013628
Experian 0,00062993 0,00067163 0,11625 -0,11730 0,014548
CRH 0,00012891 0,00000 0,11870 -0,15994 0,017950
BAE Systems 0,00015892 0,00033007 0,10008 -0,081762 0,014481
Note: Table was created by author using empirical data of companies of FTSE 100 selected from Investing.com
To summarize analyzes of descriptive statistics it can be said that in estimation period 13
companies were more volatile than market and 12 weren’t. Largest portion of companies that are
more volatile than the market and more sensitively reacts to changes in it were from largest sample
sectors: financial and consumer staples. 6 companies out of 7 were more volatile than the market in
financial sector and 3 out of 7 were more volatile than the market from consumer staple sector.
Secondly results showed in most of the company cases their mean were lower than the median
meaning that companies experienced large loses more times than they experienced large gains in
estimation period due to some events or reasons. That indicates that market reacted negatively to
those news, events or announcements. After analyzes of descriptive statistics and understanding the
results it shows in the next paragraphs the calculations of empirical model will be presented. The
results will be presented in tables, graphs and etc. that will help to accept or reject the hypothesis.
3.2 “BREXIT” News Effect for London Stock Exchange
First of all the results of “BREXIT” events effected United Kingdom market will be
presented. As it was already mentioned the numbers of events taken for estimation were 17.
Estimation periods selected for research as it was already mentioned were 10 days before and after
event, following with 5 days before and after event, 2 days before and after event, 10 days before
event with event day and lastly 10 days after the event.
Results are presented in the Annex 2. All of the periods of events that had a significant
effect were marked in the Annex 2 table. For all of the “event-windows” separate column bar charts
had been drawn to better understand witch and how many of events had a significant effect for
United Kingdom financial market.
Page 43
43
Figure 1 shows the percentage changes of United Kingdom markets returns in event
period during all 17 events in 10 days before and after the event period. The events that are
statistically significant are marked with a star simbol above column in the graph. 5 events out of 17
were statistically significant in the 10 days before and after period. As the figure 1 shows 3 out of
the 5 statistically significant events had positive influence for the market returns and in 21 days
period market index showed significant growth of returns. Strongest grow were calculated on the
12th event which was the election of Boris Johnson as new Prime Minister of UK. The returns of the
market have grown more than 18% on that “event-window”. Furthermore the 16th event when Boris
Johsnon won general election and broke the parliamentary deadlcok of “BREXIT” which slowed
down “BREXIT” implementation. The third event that influenced the growth of London Stock
exchange index was Theresa’s May lost in parliamentary majority and chance to increase her
authority in the parliament. Both of the events increased the returns of the market more than 7%.
That suggests that H1 should be rejected due to the fact that only 29% of “BREXIT” events had a
significant influence for the financial market. But due to the reason that maybe events had a shorter
significance period shorter periods need to be analyzed and calculated.
*
*
*
*
*
-10,50%
-9,00%
-7,50%
-6,00%
-4,50%
-3,00%
-1,50%
0,00%
1,50%
3,00%
4,50%
6,00%
7,50%
9,00%
10,50%
12,00%
13,50%
15,00%
16,50%
18,00%
19,50%
21,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Per
cen
tag
e C
ha
ng
es
"BREXIT" events
Figure 1. London Stock Exchange Cumulated Abnormal Returns CAR (-10:+10)
Page 44
44
Figure 2 above shows how all “BREXIT” events had effected the returns of London Stock
Exchange in other selected event periods. The graph shows all other four selected events periods
and their statistical significance for the returns of London Stock exchange. The results reveals more
detailed information about the significance of the events to UK financial market than in the figure 2.
5 events that were statistically significant in period of 10 days before and after event showed
significance in shorter “event-windows” as well. That is why these 5 events were excluded from
analyzes of shorter periods. What can be said from this data that 3 events showed significant
influence to London Stock Exchange on period of 2 days before and after the event. Theresa’s May
election period, Article 50 and parliament block of 31 October „no-deal Brexit“ had statistically
significant influence for London Stock exchange in that period. Daily returns in that period were
positive in all of the events. 8 event showed significant changes for market returns on period of 10
days before event. On that event period first meaningfull vote was held where UK government were
defeated. Lastly the the speech of Theresa May regarding “BREXIT” and her statements about
future of relations between UK and EU held on 2017-01-17 which is coded as number 2 in this
graph showed significant positive effect for the market in the period of 10 days after the event.
These results show that additional periods for the events show that they were significant in shorter
periods of time. Nevertheles the results of figure 2 do not help to approve of H1.
To summarize the results from these two tables it could be said that “BREXIT” news have
a significant influence for United Kingdom financial market. The difference is that some of the
events have a significant effect in shorter period of time. For example only 5 events out of 17 were
significant in 10 days before and after the event and aditional calculation of other periods increased
* *
*
*
**
*
* *
*
**
*
**
*
*
*
-10,00%
-5,00%
0,00%
5,00%
10,00%
15,00%
20,00%
25,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Per
cen
tag
e C
ha
ng
e
"BREXIT" events
CAR (-5;+5) CAR (-2;+2) CAR (-10;0) CAR (0;+10)
Figure 2. Cumulated Abnormal returns (CAR) of London Stock Exchange
Page 45
45
the sample of significant events for the financial market. The number of significant events increased
from 5 to 10 events out of total 17. 3 events showed significant positive results for daily returns of
London Stock Exchange in 2 days before and after period. 1 event showed significant results on 10
days before the event period and 1 in 10 days after the event. Furthermore the results in 7 cases out
of 10 were positive for the market and increased in the value of daily returns. Nevertheles
distribution of significant periods of events can’t let accept the fact that “BREXIT” events had a
significant effect on UK financial market.
3.3 “BREXIT” News Effect for Individual Companies
Furthermore the effect of “BREXIT” news for the individual companies with largest
market capitalization will be analyzed and presented. All of the calculations for all “event-
windows” were made from annexes starting from 3 to 7. From the tables of annexes from 3 to 7 the
cumulated abnormal average returns of all companies on specific event were calculated and written
in the second row of the table. After that all of the individual companies returns in “event-window”
were compared with that index. At first the “event-window” of 10 days before and after the event
will be analyzed. Results are presented in the figure 3 below. The red arrow in the graph shows the
cumulated average abnormal returns of all companies for all events. In the event period of 10 days
before and after the event it is 0.14%. Companies are coded as numbers from 1 to 25 in the graphs
and their numeric name were presented in methodology part in the table 7. Furthermore statistically
significant changes in daily returns for individual companies regarding “BREXIT” events were
marked with data labels and star symbol meaning that changes in returns were significant in
selected “event-window”. All return data of individual companies presented in the figures
represents the average change in returns during all events for “event-window”.
Figure 3 (page 46) shows that statistically significant changes in average cumulated
abnormal returns for all 17 events were experienced by 9 companies out of 25 in “event-window”
period of 10 days before and after the event. Energy companies “Royal Dutch Shell PLC” and “BP
PLC” experienced significant changes in returns due to the “BREXIT” events and on average their
returns in that period decreased by -2.25% for “Royal Dutch Shell PLC” and -1.83% for “BP PLC”.
4 financial sector companies out of 7 aswell experienced significant changes in returns regarding
“BREXIT” events. “Barclay’s PLC”, “Standard Chartered PLC” and “Legal and General Group
PLC” experienced gains and only “Prudential PLC’ experienced loses due to the “BREXIT” events
when taking average of all returns. On average “Barlcay’s PLC” have grown by 3.21%, “Standard
Chartered PLC” experienced growth by 2.63% and “Legal and General PLC” had largest gains
compared to others by 4.47%. Nevertheless “Prudential PLC” experienced losses of -1.83% on
average regarding “BREXIT” events. “Associated British Foods PLC” experienced losses of -
1.88% on average regarding “BREXIT” events. Other two companies “BAE PLC” and “National
Page 46
46
Grid PLC” experienced gains on average cumulated abnormal returns on estimation period of 10
days before and after the event. “BAE PLC” experienced gains of 1.91% and “National Grid PLC”
returns had grown by 1.49%. This estimation period suggests that “BREXIT” events didn’t have
signficant effect for larger part of individual companies listed on London Stock exchange and H2
should be rejected. Which means that only a few companies of this sample experienced the effect of
“BREXIT” events on period of 21 days. Due to this reason shorter periods of time are calculated
and analyzed.
Figure 4 (page 47) represents cumulated average abnormal returns of individual companies
for all 17 “BREXIT” events on period of 5 days before and after the event. Data labels and star
simbol represents that returns were statistically significant compared to average market returns
during that period of time. The results in figure 4 shows that “BREXIT” events affected more
companies on shorter period of time then on previously analyzed longer period. 12 companies out
of 25 experienced significant changes in returns regarding “BREXIT’ news. The number of
financial sector companies who experienced significant changes in returns increased from 4 to 5.
“Barclay’s PLC”, “Royal Bank of Scotland and Group PLC”, “Standard Chartered PLC”, “Legal
and General PLC” and “HSBC Holdings PLC” experienced significant gains in returns in that
period compared to average market returns of that event period. “HSBC Holdings PLC”
experienced gains of 0.86% on average returns due to the “BREXIT” events. “Barclay’s PLC”
returns increased by 1.95%,“Royal Bank of Scotland and Group PLC” by 1.78% , “Standard
Chartered PLC” by 1.8% and largest gains of returns were calculated for “Legal and General PLC”.
“Legal and General PLC” returns increased by 3.54% on average due to the “BREXIT” events.
Telecomunications company “Vodafone Group PLC” as well as financial sector companies
experienced gains by 1.6% on average in estimated period of time. Consumer staples company
-2,25%*-1,83%* -1,83%
3,21%*
2,63%*
4,47%*
-1,88%*
1,91%*1,49%*
-3,00%
-2,00%
-1,00%
0,00%
1,00%
2,00%
3,00%
4,00%
5,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25Per
cen
tag
e C
ha
ng
e
Companies
Figure 3. Cumulated Average Abnormal Returns of Sample Companies (T= -10:10)
Page 47
47
“British American Tobacco PLC” average returns incresed by 1.4% on estimation period regarding
“BREXIT” news. All other companies experienced loses in returns regarding “BREXIT” news in
estimation period. “Royal Dutch PLC”, “Glaxosmithkline PLC”, “Unilever PLC”, “Associated
British Foods PLC” and “Rio Tinto PLC” experienced loses. Largest loses were calculated for
“Associated British Foods PLC” and it returns decreased by -1.81%. After it folows “Rio Tinto
PLC’ with loses of -1.34%, “Glaxosmithkline PLC” with loses of -1.27% and “Unilever PLC” with
loses of -1.19% on average due to the “BREXIT” news. Lowest loses were experienced by “Royal
Dutch Shell PLC” and it was -0.87% on average in estimation period. Nevertheles the results of
“event-window” states that H2 should be rejected because the effect of “BREXIT” events were
insignificant for larger number of companies than it were significant.
Figure 5 (page 48) represents the calculations of “event-window” for 2 days before and
after the event. The results shows that period of 2 days before and after the event have even stronger
significance to daily returns of individual companies listed on United Kingdom financial market. In
this event period 13 companies experienced significant changes in daily returns due to the
“BREXIT” events. 6 out 7 companies from financial sector experienced significant changes in daily
returns due to the “BREXIT” events. What is more important in this specific period all of the
financial sector companies experienced gains in daily returns. Largest gains were calculated for
“Legal and General PLC”. Their returns increased by 2.48% on average in this period. Furthermore
“Royal Bank of Scotland and Group PLC” experienced gains in returns by 1.76%. Other 3 financial
sector companies experienced similar gains in returns. “Standard Chartered PLC” returns increased
by 1.36%, “Barclay’s PLC” returns increased by 1.34% and “Prudential PLC” experienced gains of
-0,87%*
0,86%*
1,95%*
1,78%*
1,80%*
3,54%*
-1,27%*
1,40%*
-1,19%*
-1,81%*
-1,34%*
1,60%*
-3,00%
-2,00%
-1,00%
0,00%
1,00%
2,00%
3,00%
4,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Per
cen
tag
e C
ha
ng
e
Companies
Figure 4. Cumulated Average Abnormal Returns of Sample Companies (T= -5:5)
Page 48
48
1.29% on average. Lowest growth in returns were calculated for “HSBC Holdings PLC”. Their
returns increased by 0.64% on average. All of the other companies from different sectors
experienced loses in daily returns in this “event-window”. “Royal Dutch Shell” experienced loses of
-0.85% on average for daily returns in this period, “Astrazeneca” experienced loses of -0.61% on
average. 3 out of 7 consumer staples sector companies experienced significant loses in daily returns
in this period and largest loses was experienced by “Reckitt Benckiser Group PLC”. Their loses on
average were -1.14%. After it folows “Diageo PLC” with loses of -0.72% and “Unilever PLC” with
loses of -0.64%. Basic resources company “Rio Tinto PLC” experienced loses on average returns in
this period. Their loses on average were -0.84%. Lastly “Relx PLC” experienced loses of -0.59%
due to the “BREXIT” events. Results of this “event-window” suggests that H2 should be accepted
because the effect of “BREXIT” events were significant for a larger number of companies than it
were insignificant and tha larger number of companies experienced loses due to “BREXIT” events.
Figure 6 (page 49) shows the fourth estimated “event-window” results. It represents the
cumulative average abnormal returns of companies in period of 10 days before the event until the
event day. The significant changes in daily returns were calculated for 15 companies out of 25. It is
the strongest effect for returns of companies calculated so far. The results of this period is similar to
already analyzed 2 days before and after the event period. Most of the financial sector companies
experienced the gains in this period except for “Prudential PLC”. “Prudential PLC” in this period
experienced significant losses of -1.18%. As in the previous estimated “event-window” the largest
gains were experienced by “Legal and General PLC”. Their returns increased by 2.68% in
estimation period on average. Other companies returns do not differ so much from the “Legal and
-0,85%*
0,64%*
1,29%*
1,34%*
1,76%*
1,36%*
2,48%*
-0,61%*
-0,72%*
-0,64%*
-1,14%*
-0,84%*
-0,59%*
-1,50%
-1,00%
-0,50%
0,00%
0,50%
1,00%
1,50%
2,00%
2,50%
3,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Per
cen
tag
e C
ha
ng
e
Companies
Figure 5. Cumulated Average Abnormal Returns of Sample Companies (T= -2:2)
Page 49
49
General PLC”. The second largest gains were experienced by “Standard Chartered PLC”. Their
returns increased by 2.65%. After it folows “Barclay’s PLC” with gains of 2.38% on average in
estimation period. “Royal Bank of Scotland and Group PLC” returns increased by 1.81% and
“Lloyds Banking Group PLC” by 1.7%. “BAE Systems PLC” as well experienced gains in returns
during the estimation period. Their returns increased by 1.7%. All other companies that have had
significant changes in their returns due to the “BREXIT” events experienced negative changes.
Consumer staples sector, Energy sector, Health-care sector, industrial sector and media sector
companies experienced loses in their returns in estimation period. Largest loses were experienced
by “British American Tobacco PLC”. Their average returns decreased by -2.58%. Second largest
loses were experienced by “Astrazeneca PLC”. Their average returns decreased by -1.71%. After it
folows “Associated British Foods PLC” with loses of -1.47%. These three companies experienced
largest loses during this estimation period regarding “BREXIT” events. “Imperial Brands PLC”
experienced loses of -1.13%, “Relx PLC” -0.97% and “Diageo PLC” -0.98%. Energy companies
“Royal Dutch Shell PLC” and “BP PLC” experienced losses of -0.69% for “Royal Dutch Shell
PLC” and -0.83% for “BP PLC”. The results of this “event-window” determines that H2 should be
accepted due to the reason that “BREXIT” events had a significant effect for a larger number of
sample companies and the larger number of companies experienced loses in that “event-window”.
Lastly the “event-window” period of 10 days after the event will be analyzed. The results
are presented in the Figure 7 (page 50). As we can see from the Figure 7 significant changes in
returns were only calculated for 4 companies out of 25. Which means that effect of “BREXIT”
events to the companies met the expectations of the market for most of them before the event day. 2
-0,69%*
-0,83%*
1,70%*
-1,18%*
2,38%*
1,81%*
2,65%* 2,68%*
-1,71%*
-2,58%*
-0,98%*
-1,47%*
-1,13%*
1,70%*
-0,97%*
-3,00%
-2,00%
-1,00%
0,00%
1,00%
2,00%
3,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Per
cen
tag
e C
ha
ng
e
Companies
Figure 6. Cumulated Average Abnormal Returns of Sample Companies (T= -10:0)
Page 50
50
out of 4 companies experienced significant gains in this period. One of them were “Legal and
General Group PLC” and another one were “British American Tobacco”. “Legal and General
Group PLC” returns increased by 1.8% and 1.52% for “British American Tobacco”. “Royal Dutch
Shell PLC” experienced loses of -1.55% on average in this event period. Lastly “Tesco”
experienced loses of -1.33% during estimation period. The results of this “event-window”
determines that H2 is rejected.
To summarize the results of figures it is valid to say that “BREXIT” events and news have
a significant effect for the individual companies listed on United Kingdom financial market. The
strongest effect it has on the periods of 2 days before and after the event and 10 days until the event.
Nevertheles deeper analyzes of results in those two periods showed that “BREXIT” news have a
negative effect for a larger number of companies and their returns. In 2 days before and after the
event period 7 companies experienced loses in their returns and 6 experienced gains. On the period
of 10 days until the event 9 companies experienced decrease in returns regarding “BREXIT” news.
Other 6 whose changes in returns were significant experienced gains. That suggests that companies
do experience negative impact because of the “BREXIT” news. Longer period of 10 days before
and after the event only showed significance in return changes only for 9 companies. The next
period of 5 days before and after the event increased the significant results sample by additional 3
companies resulting in total of 12 but still to accept the significance of the “BREXIT” events for
companies was not enough. Lastly the last period of 10 days after the event showed that changes in
returns were insignificant in most of the cases. Only 4 companies experienced significant changes
-1,55%*
1,80%*
1,52%*
-1,33%*
-2,00%
-1,50%
-1,00%
-0,50%
0,00%
0,50%
1,00%
1,50%
2,00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Per
cen
tag
e C
ha
ng
e
Companies
Figure 7. Cumulated Average Abnormal Returns of Sample Companies (T= 0:+10)
Page 51
51
in returns due to the “BREXIT” events. Which means that expectations of the market were met
before the day of the event.
3.4 “BREXIT” News Effect for Various United Kingdom Economical
Sectors
After analyzing the effect of “BREXIT” news to individual companies with largest market
capitalization listed on United Kingdom financial Market the analyzes of sectoral analyzes will be
presented. The Hypothesis formulated before stated that financial sector companies due to the
“BREXIT” announcement experienced larger loses than other sectors. That is why it is necessary to
evaluate how individual sectors were affected by the “BREXIT” news and whether the financial
sector still suffered more than the others. To analyze that companies from the sample were
structured by sectors and their average abnormal returns were added together. After that statistical
significance of the changes betweens sectors were calculated.
Results presented in figure 8 (page 52) shows that financial sector companies experienced
largest changes in returns compared to other sectors. Actual changes in daily returns of various
sectors are presented in the annex 8. Statistical analyzes contemplates that financial sector
companies experienced larger changes compared to other sectors. On 10 days before and after the
event “event-window” financial sector experienced 11.04% growth in average returns. Other sectors
didn’t experience such a large increase in returns. Industrial sector experienced increase in average
returns by 1.1%, Telecommunications increased their returns by 1.9% and utilities sector returns
have grown by 1.49%. Other sectors experienced losses in returns during 10 days before and after
the event period. Largest loses were calculated in Consumer staples sector and energy sector.
Consumer staples sector returns decreased by -4.39% and energy sectors returns decreased by -
4.07%. Shorter “event-windows” complements first findings about sectors and return changes in
them. Different results only can be seen in estimation window of 10 days after the event. Financial
sector companies experienced 0.34% growth on average returns while other sectors experienced
similar changes in returns in the same period. Only energy sector companies experienced larger
loses compared to other sectors. Energy sectors average returns decreased by -2.55% in that “event-
window”. Nevertheles statistical analyzes confirms that financial sector companies are more
sensitive to “BREXIT” news and experience greater changes in returns because of it. But the
results of this research differs from other authors findings because they find out that “BREXIT” had
a negative effect for financial sector companies while this research suggests differentely. Findings
of this research states that “BREXIT” news has a positive outcome for returns of financial sector
companies. Due to this research results H3 is rejected because “BREXIT” news affected daily
returns of financial sector companies positively and it differs from hypothesis and other authors
findings.
Page 52
52
Lastly as it was mentioned before the “herding” behavior are more present during times of
uncertainty. That is why this research also tried to estimate whether the presence of “herding”
behavior existed in United Kingdom fnancial market during “BREXIT” events. For estimation of
“herding” behavior same model were applied as for estimation of the abnormal returns.
Nevertheless the results of the model didn’t show any significant “herding” behavior during
“BREXIT” events. This could have happened because of the limitations of the model itself.
Furthermore the comparison of this research to other authors works about “Brexit” effect
for financial markets and individual companies will be analyzed and presented. Bashir et al., (2019)
analyzed the effect of “BREXIT” pre and post refferendum phase for financial markets. They tested
how “BREXIT” refferendum affect UK financial market and 4 other European countries
considering named periods of time. They applied detrend fluctuation analyzes in order to analyze
how “BREXIT” affected the UK, France, Germany, Spain and Netherlands financial markets. The
results of their research indentify negative correlation between UK, France, Germany and
Netherlands financial markets meaning that in the pre-refferendum phase UK experienced
depreciation of currency and increase afterwards. According to Bashir et al. (2019) it is due to the
fact that international investors took a more forward approach and investing in stock markets in
order to benefit from the volatile depreciated currencies after the referendum. Guedes, Ferreira,
Dionisio and Zebende (2019) did a similar research to Bashir et al., (2019). Guedes, Ferreira,
Dionisio and Zebende (2019) research applied more EU countries in sample than Bashir et al.,
11,04%*
9,75%*
9,11%*
10,71%*
-6,00%
-4,00%
-2,00%
0,00%
2,00%
4,00%
6,00%
8,00%
10,00%
12,00%
PE
rcen
tag
e C
ha
ng
e
Sectors
CAAR (-10:10) CAAR (-5:5) CAAR (-2:2) CAAR (-10:0) CAAR (0:10)
Figure 8. Cumulated Average Abnormal Returns of Distinct Sectors
Page 53
53
(2019) but they also used the detrend fluctuation analysis in their work. Their results determined
that “BREXIT” referendum affected negatively the relations between EU countries meaning
decreases in returns of the market. The coefficient showing the decrease in relations between UK
and other EU countries didn’t decrease only between UK and Bulgaria, Malta and Slovenia.
Aristeidis and Elias (2018) estimated the effect of referendum for London stock exchange. Authors
determined that London Stock exchange experienced loses of 3% on total in the first day of
announcement meanining that “BREXIT” had a negative impact for the UK financial market. Their
findings also supplemented the findings of Bashir et al., (2019) showing the devalvation of pound
sterling due to the “BREXIT” referendum. Aristeidis and Elias (2018) estimated the largest sample
of countries including not only European Union countries but also from North and South America,
Asia and Africa. Abraham (2018) in his research about “BREXIT” effect to New Zealands financial
markets determined that “BREXIT” negatively influenced New Zealand financial market in pre-
referendum phase and it only recovered a few weeks after it. The results of Abraham (2018)
compared to Bashir et al., (2019) show similar trends in the markets because New Zealand is close
trading partner with UK and turmoil before the vote of “BREXIT” in UK affected New Zealand
while helped to recover after it sooner because international investors invest in New Zealand
companies located in UK for the same reason they invested to UK stocks because of the decreased
value of them due to the “BREXIT” vote. Morales and Andreosso-O’Callaghan (2019) also
analyzed how “BREXIT” news affected the Chinese stock market. Authors used “event-study”
method for their research and their findings determined that Chinese stock market didn’t suffer from
“BREXIT” announcement. Their findings approve of hypothesist that countries with weaker trading
relations are not affected as much as countries with stronger relations. The findings of this research
differs from the results in this research.
3.5 Discussion of Research Results
This research demostrates different results compared to all other authors works. In this
research UK financial market experienced insignificant effect regarding the “BREXIT” news. The
results showed that only 5 events out of 17 were significant in the longest event period of 10 days
before and after the event. Furthermore the estimation of shorter event periods only increased the
significance value of additional 5 events on different periods of “event-window”. That is why it is
imposible to state which event period is significant for UK financial market in the analyzes and to
determine whether the “BREXIT” news and events have a significant effect for UK financial
market. What is more important is that results of the research differ a little bit from the previous
authors work. Bashir et al., (2019), Guedes, Ferreira, Dionisio and Zebende (2019), Aristeidis and
Elias (2018) and Abraham (2018) asses that “BREXIT” news had a negative influence to the
relationships and returns of UK and other countries financial markets in pre-referendum phase and
Page 54
54
only recovering after it. While this research findings suggest that this is not always a true. In some
cases returns of the market are also positive in the pre-event period. 6 events showed increase in
returns of the market before the event meaning that not always the results should be negative until
the event. Theresa May election, Article 50, 1 and 2 meaningful votes, Block of October leaving
and UK leaving day of EU were the events that influenced positive returns to UK financial market
until the event. Nevertheless in most cases the growth in returns can be seen in post event period
suggesting the recovery of market regarding political instability. Aristeidis and Elias (2018) identify
that abnormal returns and volatility are more present in closer dates to event while this research
results didn’t identify the same. Only 6 out of 17 events were significant in short event period.
Furthermore from the company point of view results show some interesting insights. The
research were focused on companies with largest market capitalization and “BREXIT” news and
events effect for them. Results of the research shows that statistically significant “event-windows”
in the research were 2 days before and after the event and 10 days until the event. Both of the
estimation periods show that more companies suffered decrease in returns due to the “BREXIT”
events. In 2 days before and after the event period 6 companies experienced growth of returns while
7 experienced loses and on the other “event-window” 9 companies experienced significant loses in
their returns while only 6 of them experienced growth of returns. Other estimated periods and their
statistical significance were to low to consider it meaningful. “Event-window” of 10 days before
and after the event showed that only 9 companies out of 25 experienced significant changes in their
returns meaning that “BREXIT” events do not have significant impact on longer period of time.
Shorter period of 5 days before and after the event increased the sample of significant changes in
returns to companies but still were to low to consider it meaningful. Lastly the “event-window” of
10 days after the event showed the smallest quantity of significant changes in returns for individual
companies. Only 4 companies out of 25 experienced significant changes in returns due to the
“BREXIT” events. Results suggest that investors evaluate the effect of the event before the event
date and for most companies those evaluations were right. Compared to Boulton and Bacon (2018)
findings this research suplements their findings about “BREXIT” effect to individual companies.
Boulton and Bacon (2018) determined that “BREXIT” referendum had a negative effect for largest
companies listed on New York Stock Exchange which were closely related to UK market.
Furthermore Ramiah, Pham and Moosa (2017) in their research calculated that most of the sectors
experienced negative returns due to the “BREXIT” events. This research suplements their findings
as well showing that for most individual companies from different sectors “BREXIT” news had a
negative impact. Shahzad, Rubbaniy, Lensvelt and Bhatti, (2019) results sumplement strongly with
findings of this research. Authors estimated not only the effect for the whole market but also on
company level. Their results showed that individual companies returns decreased significantly in
Page 55
55
pre-referendum phase and experienced growth only after political stabilization and markets
evaluation of real situation. This research findings supplements their finding showing that most of
the companies suffered losses in the “event-window” of 10 days until the event. Abraham (2018) in
his work measured that New Zealand companies experienced loses because of UK wish to leave
EU. Abraham (2018) aswell as Shahzad, Rubbaniy, Lensvelt and Bhatti, (2019) determined that
strongest negative effect for individual companies were observed on period of time until the
“BREXIT” vote. Furthermore Andrikopoulos, Dassiou and Zheng (2020) in their work analyzed the
effect of “BREXIT” referendum to FTSE 100, IBEX35 and DAX30 non-financial companies. Their
findings identified that FTSE100 25 non-financial companies from different sectors experienced
growth in returns and that referendum were positive for them. Andrikopoulos, Dassiou and Zheng
(2020) results compared to this research findings differs because as it shows individual companies
from non-financial sector experienced significant negative returns due to the “BREXIT” while in
most cases financial sector companies experienced growth in returns. Same interpretation could be
implemented about the other significant results showing “event-window”. Period of 2 days before
and after the event showed that larger number of companies suffered decrease in returns due to the
“BREXIT” events.
To summarize the results of the “BREXIT” effect for individual companies it is possible to
say that this research results suplements other authors findings about negative “BREXIT” effect to
individual companies. In “event-windows” that indicate significant changes for returns of individual
companies larger number of companies suffered loses due to the “BREXIT” events. In “event-
window” of 10 days before the event 9 companies suffered loses while only 6 experienced growth
in returns. On the shorter significant period of 2 days before and after the event 6 companies
experienced growth in returns while 7 had loses.
Ramiah, Pham and Moosa (2017) in their work analyzed how “BREXIT” affected various
sectors of UK economy. They analyzed how the referendum affected them and which sectors
experienced loses and which gained from it. Their findings indentified that financial, travel and
leisure sectors suffered largest loses due to the referendum. Their results determined that chemicals,
oil and gas, beverage, aerospace and defence, tobacco and forestry and papers sectors experienced
gains due to the “BREXIT” referendum. Meaning that only 6 sectors out of 24 gained positive
returns due to it. This research results suplements findings of Ramiah, Pham and Moosa (2017)
work in a certain way that larger portion of sectors included in the sample experienced loses. In this
research 5 out of 9 sectors experienced loses aswell. Nevertheless the main question of this research
whether the “BREXIT” news affect negatively and significantly stronger financial sectors compared
to others. As the results in Figure 8 shows financial sector companies indeed had a stronger effect to
their daily returns due to the “BREXIT” events. The difference between this research findings
Page 56
56
compared to Ramiah, Pham and Moosa (2017) is that financial sector companies experienced gains
due to the “BREXIT” events compared to other sectors. Financial sector companies returns
increased by 11.04% on average due to the “BREXIT” events in period of 10 days before and after
the event. Compared to sector in second place with positive returns, financial sector returns were
larger more than 5 times. Ramiah, Pham and Moosa (2017) in their work calculated loses in
financial sector which were -15.37%. That means that this research findings completelly differs
from Ramiah, Pham and Moosa (2017). Aristeidis and Elias (2019) in their work about how
“BREXIT” referendum affected various countries financial markets also separated banking sector
results from others. They focused on findings about financial sector of United Kingdom. In their
research they estimated that banking sector experienced huge loses due to the uncertainty created by
the “BREXIT” referendum. Their results showed that 5 largest banks of UK suffered decrease in
their stock prices by 21% on average on the next morning after referendum. Additionally a non-UK
banks stocks prices also fell by more than 10% on average. Not all banks fully recovered after it and
some of them prices still remain lower than 10%. Bank of England released 150£ to banks in
lending by reducing the countercyclical capital buffers that banks are required to hold. Nevertheless
the results of Aristeidis and Elias (2019) work shows that financial sector suffered from due to the
“BREXIT” referendum vote and had to take some time to recover. Aristeidis and Elias (2019)
results differ from the findings of this research results.
Table 12
Hypothesis valuation
Hypothesis
No.
Hypothesis Accepted/Rejected
H1 “BREXIT” news have a significant negative effect on
London stock exchange.
Rejected
H2
“BREXIT” news have a significant negative effect on
individual companies in United Kingdom listed on London
Stock Exchange.
Accepted
H3 “BREXIT” news have a stronger negative effect to financial
sector companies compared to other sector companies.
Rejected
H4 “BREXIT” news trigger financial herding behavior in
London stock exchange.
Rejected
Results of the acceptance and rejection of research hypothesis are presented in th table 12.
First hypothesis of the research is rejected. H1 is rejected due to the reason that only 5 events in the
research were statistically significant to London Stock Exchange in 10 days before and after the
event period. Calculations of shorter “event-windows” didn’t increase the statistical significance of
the “BREXIT” events to London Stock exchange. Of course some of the events showed an increase
in their significance in shorter event period but still the number of significant events didn’t increase
to a number that could let accept the H1. Another reason why the H1 can’t be accepted is that larger
Page 57
57
portion of events influenced the growth of the market. 12 out of 17 events influence the growth of
London Stock exchange returns.
H2 is accepted because as the results of the research shows larger number of companies in
the sample experienced loses in their returns due to the “BREXIT’ events. 7 companies out of 13 in
the “event-window” of 2 days before and after the event experienced significant loses and 9
companies out of 15 had loses in “event-window” of 10 days before the event. Other “event-
windows” were too insignificant to identify whether the effect for individual companies were
negative or positive due to the “BREXIT” events.
H3 were based on previous researchers works and stated that financial sector companies
experienced larger loses due to “BREXIT”. Aristeidis and Elias (2019) and Ramiah, Pham and
Moosa (2017) in their works stated that financial sector companies experienced large loses due to
“BREXIT” referendum. This research approves with the point that financial sector companies were
more sensitive to “BREXIT” referendum and their prices were more volatile. However research
results determined other results when analyzing the actual returns of sector. Research results shows
that financial sector companies increased in value and their returns have grown because of
“BREXIT” events and dissaproves with previous authors results. Because of this reason this
hypothesis is rejected.
H4 is rejected aswell because model couldn’t detect any financial herding behavior in the
“event-windows” of “BREXIT” events. Changes in volume for all individual companies and
London Stock Exchange were statistically insignificant. That is why it can’t be said whether the
financial “herding” behavior were present in the UK financial market or not.
3.6 Limitations of the Empirical Data Findings and Reccomendations
for Future Researches
Lastly empirical research has some limatations and they reflect in the work. First limitation
of research is lack of control variables in the work that could intersect with “BREXIT” news. As the
results showed “BREXIT” events had a significant effect for the companies and their returns in
“event-windows”. But this research do not take in to account the influence of different factors like
economic policy index and volatility index VIX that were applied by Morales and Andreosso-
O‘Callaghan (2019) in their research. Those two indexes more or less would have influenced the
final results of the research at least for financial market index of UK. As it was already mentioned it
would be hard to select control variables for individual companies due to the reason that companies
in the research selected were largest based on market capitalization had high domestic sales and just
from a few industries. Nevertheless if future researches would focus on results for larger sample of
companies and “BREXIT” news affect for them those control variables should be applied.
Page 58
58
Second limitation of the research is duality of the events. As it was already have been
mentioned some of the “BREXIT” events and news intersected and due to this reason some of them
couldn’t be estimated and had to be excluded from the research. This logic applies to all news that
could have had influenced UK financial market and individual companies. Some of the events like
dividend announcement, earning reports and etc. could have had an influence to decrease or growth
of returns for individual companies. Other political news as well could have had some influence for
UK financial market and those companies. Future researches should take in to consideration and
analyze every other events, news happening around the “BREXIT” events.
Last limitation as it was already mentioned is that model couldn’t able to detect any
presence of financial “herding” behavior in UK financial market. Political uncertainty effects the
financial markets of countries negatively and increases financial “herding” behavior and that was
analyzed and approved by previous researches. “BREXIT” is new political uncertainty and should
increase financial “herding” behavior as well as others because UK exit would decrease the power
of EU significantly. However the results of the model didn’t show any signs of financial “herding”
behavior in UK financial markets and individual companies meaning that financial “herding”
behavior is not present in the “BREXIT” event periods. Future researches should apply different
and more frequently used empirical models for better detection of financial “herding” behavior in
order to answer whether “BREXIT” affected it.
Page 59
59
CONCLUSIONS
1. Traditional finance theory is finance theory that defines investor as rational, risk-averse and
aimed to maximize “expected utility” at every decision he makes. Behavioral finance theory is
finance theory that considers investor not as rational but as some authors describe it as
“normal”. Behavioral finance theory states that investors can make mistakes in their calculations
(cognitive errors). Behavioral finance started to develop more rapidly after traditional finance
theory failed to explain financial market anomalies like market bubbles and etc. Nevertheless
both theories strongly linked together and could not exist one without another. Traditional
finance theory lacks estimation of psychological factors influencing the market movements
while behavioral finance theory tries to develop a better models for their estimation.
2. Analysis of literature analysis allowed identifying that traditional factors affecting stock price
movements may be divided into micro and macro level factors. Macro-economic factors are
inflation, export, exchange rate, industrial production, currency supply, interest rate,
government performance, BVP growth, investment growth, turnover, domestic credit, narrow,
abroad money growth. Micro-economic factors affecting stock price movements are sales,
equity, debt, liquidity, tobins Q, debt/assets ratio, foreigner control, dividends announcement,
size of the market, number of investors, dividend yield and payout, leverage rates, cash flow,
risk, return ratios, investment duration. Analysis also showed that behavioral factors affecting
stock prices are more and more discussed in scientific papers. Behavioral factors are categorized
in two categories: cognitive bias and emotional bias. Cognitive bias factors are anchoring,
adjustment, framing, conservatism, availability, mental accounting, gamblers fallacy,
representativeness. Emotional bias factors are endowment bias, loss aversion, optimism, status
quo, overconfidence, herding and regret.
3. Political stability, certainty is very necessary to the development of countries and stability of
their financial systems. Political stability is one of the key factors affecting the financial markets
and its instability could influence collapse of the financial market. Political uncertainty lowers
the profits of investors which lead to larger investment in domestic countries and foreign direct
investment loss.
4. To estimate anomalies in market returns exist some empirical models. First and mostly used are
“Event-study” method. “Event-study” model is based on efficient market hypothesis. Authors
adapt this model to estimate how separate events affect the movements of the financial market.
Researchers modify the model by applying more control variables and etc. “Event-study”
method helps to calculate actual returns of the market due to the events. Another adaptation of
the “event-study” method for calculating the anomalies in the market is called GARCH.
Page 60
60
GARCH method calculates the volatility of the market. Furthermore copula functions and
detrended fluctuation analysis models are also applied for estimation of market anomalies due to
some events. Previous works estimating “BREXIT” event effect for financial markets
determined that UK financial market experienced larger loses in pre-referendum phase and only
recovered a few weeks after it (Shahzad, Rubbaniy, Lensvelt and Bhatti, (2019). Bashir et al.,
(2019) and Guedes, Ferreira, Dionisio and Zebende (2019) also find out negative effect between
UK and other EU countries. Countries with strong trade relations with UK also experienced
negative effect for the financial markets. In this case New Zealand and India experienced it. On
industrial level companies with lower internalization level experienced larger loses than the
ones with higher internalization. Most of the UK economy sectors experienced loses due to the
“BREXIT” vote. Companies strongly related with UK market also experienced loses.
5. This research applied “event-study” method to analyze how “BREXIT” events happened in the
future affected UK financial market and individual companies. Furthermore this research tried
to capture financial “herding” behavior in UK financial market due to the fact that “BREXIT”
events creates political instability and it influences financial “herding” behavior. Empirical
model estimated how different 17 events affected London Stock Exchange and individual 25
companies with largest market capitalization listed on UK financial market indexes.
Furthermore the effect of “BREXIT” events for various sectors were analyzed. The main
question of sectoral analyzes was that financial sector companies were more sensitive to
“BREXIT” news and should influence them negatively. Results showed that “BREXIT” events
were statistically insignificant to London Stock exchange due to the fact that only 5 events out
of 17 were significant in event period of 10 days before and after the event. Additional shorter
event-windows increased significance only for a few events. However “BREXIT” events were
statistically significant to individual companies and results approved of hypothesis that
“BREXIT” events has an effect for them. “BREXIT” events were statistically significant in 2
days before and after the event period and 10 days before the event. The results showed that
larger number of companies experienced loses due to the “BREXIT” events. Calculations of
how various sectors were affected by “BREXIT” events showed that financial sector indeed was
more sensitive than others. However the calculations determined that effect for financial sector
due to the “BREXIT” events were positive and increased financial sector returns. Financial
“herding” behavior could not be detected in the research and due to this fact hypothesis that
financial “herding” behavior is present in UK financial market around “BREXIT” events was
rejected.
Page 61
61
REFERENCES
1. Abraham, M. (2018) “Brexit Anouncement Effect on the New Zealand Stock Market:
Theory and Evidence”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=44&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
8040648&db=bth
2. Alles Rodrigues, A. Lleo, S. (2018) “Combining standard and behavioral portfolio theories: a
practical and intuitive approach”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=23&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#db=bth
&AN=129059208
3. Aristeidis, S. and Elias, K. (2018), “Empirical analysis of market reactions to the UK’s
referendum results – How strong will Brexit be?”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S1042443117300604
4. Asteriou, D. and Siriopoulos, C. (2000) “The Role of Political Instability in Stock Market
Development and Economic Growth: The Case of Greece”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=28&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#db=bth
&AN=4373966
5. Bashir, U., Zebende, G.F., Yu, Y., Hussain M., Ali, A. and Abbas G.(2019) “Differential
market reactions to pre and post Brexit referendum”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0378437118313128
6. Bian, J., Chan, K. and Zhou, H. (2018) “Do Behavioral Biases Affect Order
Aggressiveness?”, http://web.a.ebscohost.com/ehost/detail/detail?vid=37&sid=47c16830-
1e47-46e7-9d80-75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=12
9358451&db=bth
7. Charles Moshi, G. and Jacob Mwakatumbula, H. (2017) “Effects of political stability and
sector regulations on investements in African mobile markets”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0308596117302665
Page 62
62
8. Chen, A.H. and Siems, T.F. (2004) “The effects of terrorism on global market capitals”,
https://www-sciencedirect-com.ezproxy.vdu.lt:2443/science/article/pii/S0176268003001022
9. Davies, R.B. and Studnicka, Z. (2018) “The heterogenous impact of Brexit: Early indications
from the FTSE” https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0014292118301387
10. Divanoglu, S.U. and Bagci, H. (2018) “Determining the factors affecting individual investors
behaviors”, http://web.a.ebscohost.com/ehost/detail/detail?vid=11&sid=47c16830-1e47-46e7-
9d80-75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
3952143&db=bsu
11. Filbeck, G., Ricciardi, V., Evensky, H.R., Fan, S.Z., Holzhauer, H.M. and Spieler, A. (2017)
“Behavioral finance: a panel discussion”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S2214635017300503
12. Francisco dos Santos, S. and Brandi, H.S. (2017) “Selecting portfolios for composite indexes:
application of Modern Portfolio Theory to competitiveness”,
https://link.springer.com/article/10.1007/s10098-017-1441-y
13. Guedes, E.F., Ferreira P., Dionisio, A. and Zebende, G.F. (2019) “An econophysics approach
to study the effect of BREXIT referendum on European Union stock markets”, https://www-
sciencedirect-com.ezproxy.vdu.lt:2443/science/article/pii/S0378437119304959
14. Hillier, D. and Loncan, T. (2019) “Political uncertainty and stock returns: Evidence from the
Brazilian political crisis”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0927538X18303822
15. Irshad, H. (2017) “RELATIONSHIP AMONG POLITICAL INSTABILITY, STOCK
MARKET RETURNS AND STOCK MARKET VOLATILITY”
http://web.a.ebscohost.com/ehost/detail/detail?vid=26&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=12
7056156&db=bth
16. Kapoor, S. and Prosad, J.M. (2017) “Behavioral Finance: A Review”, https://www-
sciencedirect-com.ezproxy.vdu.lt:2443/science/article/pii/S1877050917325693
17. Lehkonen, H. and Heimonen, K. (2015) “Democracy, political risks and stock market
performance”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0261560615001023
18. Lekovic, M. (2018) “EVIDENCE FOR AND AGAINST THE VALIDITY OF EFFICIENT
MARKET HYPOTHESIS”,
Page 63
63
http://web.a.ebscohost.com/ehost/detail/detail?vid=21&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
3260403&db=bth
19. Matsumura, K. and Kawamoto, M. (2013), “Derivative trade optimizing model utilizing GP
based on behavioral finance theory”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=6&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=86
170611&db=bth
20. Mnif, A.T. (2017) “Political uncertainty and behavior of Tunisian stock market cycles:
Structural unobserved components time series models”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0275531916301830
21. Moscati, I. (2016) “Retrospectives How Economists Came to Accept Expected Utility
Theory: The Case of Samuelson and Savage”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=15&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=11
5160727&db=bth
22. Oehler, A., Horn, M. and Wendt, S. (2017) “BREXIT: Short-term stock price effects and the
impact of firm-level internalization” https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S154461231630366X
23. O’Sullivan, P. (2018) “The Capital Asset Pricing Model and the Efficient Markets
Hypothesis: The Compelling Fairy Tale of Contemporary Financial Economics”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=19&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
5800463&db=bth
24. Raheja, S. and Dhiman, B. (2019) “Relationship Between Behavioral Biases and Investment
Decisions: The Mediating Role of Risk Tolerance”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=30&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
8681173&db=bth
Page 64
64
25. Ramiah, V., Pham, H.N.A. and Moosa, I. (2017) “The sectoral effects of Brexit on the British
economy: early evidence from the reaction of the stock market”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=42&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=12
1703653&db=bth
26. Ramiah, V., Xu, X., Moosa, I.A. (2015) “Neoclassical finance, behavioral finance and noise
traders: a review and assessment of the literature”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S1057521915001039
27. Renu Isidore, R. and Christie, P. (2018) “The relationship between the income and behavioral
biases”, http://web.a.ebscohost.com/ehost/detail/detail?vid=35&sid=47c16830-1e47-46e7-
9d80-75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=13
6835403&db=bth
28. Shahzad, K., Rubbaniy, G., Lensvelt, M.A.P.E. and Bhatti, T. (2019) “UK’s Stock market
reaction to Brexit process: a tale of two halves” https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0264999317314840
29. Smimou, K. (2014) “International Portfolio Choice and Political Instability risk: A multi-
objective approach” https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S0377221713000544
30. Soltani, H., Aloulou, A. and Abbes, M.B. (2017) “The Impact of Political Instability on
Investor Sentiment and Market Performance: Evidence from Tunisian Revolution“,
http://web.a.ebscohost.com/ehost/detail/detail?vid=39&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#db=bth
&AN=126824341
31. Sushma and Rushdi, N.J. (2018) “A systematic literature review on evolution of Behavioral
finance”, http://web.a.ebscohost.com/ehost/detail/detail?vid=4&sid=13ebbd5f-1c59-48e1-
b5c7-
725861c350c1%40sessionmgr4007&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1
laG9zdC1saXZl#AN=136549311&db=bth
32. Wang, H. and Boatwright, A.L. (2019) “Political uncertainty and financial market reactions:
A new test”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S2110701719301337
Page 65
65
33. Wisniewski, T.P. (2016) “Is there a link between politics and stock returns? a literature
survey”, https://www-sciencedirect-
com.ezproxy.vdu.lt:2443/science/article/pii/S1057521916301107
34. Yoshinaga, E. and Ramalho, B. (2014), “Behavioral finance in Brazil: applying the prospect
theory to potential investors”,
http://web.a.ebscohost.com/ehost/detail/detail?vid=4&sid=47c16830-1e47-46e7-9d80-
75d305b65d0b%40sdc-v-
sessmgr03&bdata=JkF1dGhUeXBlPWlwLHVybCx1aWQmc2l0ZT1laG9zdC1saXZl#AN=12
9236491&db=bth
Page 67
67
Annex 1
Research Paper Concepts
“ARCH” – Abbreviation for “Autoregressive Conditional Heteroskedastic“ empirical model.
“CAPM” – Abbreviation for traditional finance theory model „Capital Asset Pricing Model“
“DAX 30” – Abbreviation for the most important Germany’s stock market index which contains the
largest 30 companies of Germany.
“EARCH” – Abbreviation for modified “Autoregressive Conditional Heteroskedastic” empirical
model.
“EPU” – Stands for “Economic Policy Uncertainty” index that measures the stability of countries
economy and etc.
“EU” – Abbreviation for European Union.
“Event” – Political or non-political happening that could have an influence to financial markets or
individual companies.
“Event-window” – number of days in which the effect of the event is calculated
“FTSE 100” – It is the collective name for the 100 largest companies in United Kingdom that are
traded in United Kingdom financial market
“GARCH” – Abbreviation for “Generalized Autoregressive Conditional Heteroskedastic” empirical
model.
“GARCH-M” – Abbreviation for empirical model “Generalized Autoregressive Conditional
Heteroskedastic in Mean”.
“GJR (TARCH)” – Abbreviation for empirical model “Glosten-Jagannathan-Runkle” for volatility
clustering.
“Herding” – Herding is phenomenon where investors follow other investors investing plan,
behavior rather than making their own calculations and analyzes. In other words investor that
follows herding instinct will more likely buy some investment on the fact that others buy and etc.
“IBEX 35” – Abbreviation for the benchmark stock market index of the Bolsa De Madrid, Spain’s
principal Stock exchange
“NYSE” – Abbreviation for New York Stock Exchange.
Page 68
68
“NZX 50” – Abbreviation for New Zealand Stock Exchange after it changed its name in 2003 from
New Zealand Stock Exchange Limited to New Zealand Exchange Limited.
“RAR” – Abbreviation for “Risk Adjusted Return” empirical model which calculates the abnormal
returns of stocks value influenced by external events.
“S&P 500” – It is a stock market index that tracks the stocks of 500 largest capitalization companies
in the United States of America.
“UK” – Abbreviation for United Kingdom.
“VIX” – Stands for volatility index that was created in Chicago Board Options Exchange. It is the
real-time market index that represents the market's expectation of 30-day forward-looking volatility.
„Brexit“ – United Kingdom withdrawal from European Union. It was announced after the United
Kingdom referendum vote in June 2016.
„CSAD“ – Cross-sectional absolute deviation model. Modified financial „herding“ estimation
model introduced in 2000.
„CSSD“ – Cross-sectional standard deviation model. Model estimates the financial „herding“
behavior in financial markets
Page 69
69
Annex 2
London Stock Exchange Cumulative Abnormal Returns
6,03% 4,63% 5,50% 4,79% 1,23%
0,248856 (0,0997591)* (0,00303925)*** 0,276945 0,650484
6,20% 3,97% 2,37% -0,26% 6,46%
0,265741 0,135116 0,178874 0,928003 0,0117**
4,61% 3,35% 4,29% 1,36% 3,25%
0,16682 0,196941 (0,0132495)*** 0,604892 0,200054
7,45% 5,35% -3,11% -0,48% 7,93%
(0,043659)*** (0,0369865)*** (0,0687845)*** 0,855614 0,0016**
-4,55% -1,39% 0,06% -1,77% -2,78%
0,232512 0,601939 0,9721 0,517533 0,281653
-0,47% 0,06% -0,35% 0,04% -0,51%
0,901262 0,980913 0,843563 0,988844 0,840769
-7,87% -2,27% -0,85% -2,83% -5,04%
(0,0531224)** 0,389928 0,630902 0,311125 0,081*
6,67% 3,42% 2,06% 6,50% 0,17%
0,100469 0,205196 0,247794 0,03** 0,947214
0,12% 0,39% -0,56% 2,77% -2,66%
0,975552 0,878747 0,746602 0,299905 0,309819
4,67% -0,50% 2,26% 0,92% 3,75%
0,92101 0,853961 0,208713 0,744306 0,936478
1,89% 0,51% 0,39% 1,25% 0,64%
0,613934 0,846852 0,823203 0,640969 0,801278
18,55% 14,03% -2,37% -1,95% 20,50%
(1,63E-10)*** (2,13E-13)*** (0,0595598)* 0,359242 4,95E-27
1,94% 2,18% 3,91% 2,17% -0,24%
0,611527 0,407365 (0,0253674)** 0,430207 0,926349
-6,09% -1,78% 0,07% -3,78% -2,32%
0,151336 0,520397 0,970282 0,244602 0,388166
-8,25% -5,84% -1,69% -6,00% -2,24%
(0,0565578)* (0,0347607)** 0,353532 0,05515** 0,442544
7,22% 7,22% 6,69% -2,07% 9,29%
(0,0573369)** (0,0057379)*** (0,000117712)*** 0,451812 0,000278***
4,69% 1,14% 1,57% 1,64% 3,05%
0,230402 0,666978 0,37309 0,565029 0,242777
„London Stock Exchange“ Abnormal Returns
CAR (-10;0) CAR (0;+10)
2018-11-25
Date CAR (-10;+10) CAR (-5;+5) CAR (-2;+2)
2016-07-16
2017-01-17
2017-03-29
2017-06-08
2017-12-08
2018-07-06
2019-10-02
2019-10-19
2019-12-12
2020-01-30
2019-01-15
2019-03-12
2019-04-12
2019-06-24
2019-07-24
2019-09-04
Page 70
70
Annex 3
Individual Companies Cumulated Abnormal Returns of period (T= -10:+10) (Part 1)
Event number 1 2 3 4 5 6
Cumulated Average
Abnormal Returns of
companies on event period
-0,53% -0,24% -0,18% 0,05% -0,28% 0,12%
CAR (-10;+10)
ROYAL DUTCH SHELL
PLC -11,31% -3,40% 2,36% -1,28% 2,04% 3,38%
HSBC HOLDINGS PLC 0,03% 4,13% -1,66% 3,74% 2,24% 0,39%
ASTRAZENECA PLC 9,53% -6,61% -3,78% 6,22% -4,23% 3,79%
BP PLC -8,60% -6,94% 3,59% -3,15% 0,06% -0,70%
GLAXOSMITHKLINE
PLC 4,45% -2,02% -2,83% 5,97% -2,48% -0,80%
BRITISH AMERICAN
TOBACCO PLC -2,67% 6,71% 4,43% 1,69% -4,04% 3,81%
DIAGEO PLC 3,55% 4,20% -0,32% 0,51% 0,58% 3,87%
UNILEVER PLC -0,84% -1,93% 0,30% 2,75% -4,38% 5,47%
RIO TINTO PLC 3,78% 10,11% -6,07% -5,17% -1,79% -3,41%
RECKITT BENCKISER
GROUP PLC -2,32% -0,96% 2,44% 3,64% 4,76% 4,21%
VODAFONE GROUP PLC 0,32% -0,60% 1,67% 2,13% 4,21% -3,68%
LLOYDS BANKING
GROUP PLC -7,53% 5,59% -7,60% -5,86% 0,91% 1,26%
RELX PLC 0,59% -3,13% 1,75% 2,05% -5,52% 3,72%
NATIONAL GRID PLC -0,05% -1,64% 6,54% -6,43% -0,89% 2,15%
PRUDENTIAL PLC -3,30% -6,55% -4,15% 3,05% -4,34% -3,12%
BARCLAYS PLC 7,27% 0,97% -3,76% -6,45% 6,64% -1,58%
ROYAL BANK OF
SCOTLAND GROUP PLC 3,21% 0,93% 1,76% -5,19% 1,93% -6,07%
TESCO PLC -13,98% -6,31% -2,98% -
10,00% 4,22% -0,80%
EXPERIAN PLC 1,92% -3,89% -1,72% -3,93% -0,87% 2,79%
CRH PLC 0,62% -2,12% -4,03% 1,30% -3,03% -4,06%
ASSOCIATED BRITISH
FOODS PLC -1,20%
-
13,55% 0,41% 0,84%
-
10,54%
-
16,65%
BAE SYSTEMS PLC 0,18% -1,49% 1,12% 3,86% 2,12% 6,89%
STANDARD
CHARTERED PLC -0,56% 17,49% -0,44% 3,07% 2,46% -4,77%
IMPERIAL BRANDS PLC -1,88% 6,08% 2,92% 0,42% -0,77% 9,11%
LEGAL & GENERAL
GROUP PLC 5,61% -1,20% 5,54% 7,58% 3,62% -2,22%
Page 71
71
Individual Companies Cumulated Abnormal Returns of period (T= -10:+10) (Part 2)
Event number 7 8 9 10 11 12
Cumulated Average
Abnormal Returns of
companies on event period
-1,81% 1,24% 0,04% 1,04% -0,08% 0,42%
CAAR (-10;+10)
ROYAL DUTCH SHELL
PLC 2,07% -5,17% 0,08% -0,60% -1,37% -4,65%
HSBC HOLDINGS PLC 6,78% -2,27% -1,21% 5,88% 0,29% -1,73%
ASTRAZENECA PLC 0,16% -
11,33% 0,53%
-
16,17% 3,23% 11,73%
BP PLC 4,91% -2,02% 2,34% -1,89% -5,11% -1,61%
GLAXOSMITHKLINE PLC -2,59% -5,33% 1,35% -3,45% 1,37% 4,59%
BRITISH AMERICAN
TOBACCO PLC
-
14,75% -1,96% 6,01% -7,98% -1,38% 5,49%
DIAGEO PLC 5,37% -8,24% 3,65% 0,84% -1,35% 0,26%
UNILEVER PLC 5,79% -3,75% 4,27% 3,85% 0,83% -1,12%
RIO TINTO PLC -2,42% 5,42% -3,86% -2,41% 1,52% -9,66%
RECKITT BENCKISER
GROUP PLC 4,29% -5,32% 8,84% -5,12% -2,54% -4,50%
VODAFONE GROUP PLC 17,36% -
13,55% 3,56% 1,69% 1,91% 15,29%
LLOYDS BANKING
GROUP PLC -8,45% 9,97% 2,49% -0,38% -1,56% -7,48%
RELX PLC 5,03% -1,28% -9,65% 5,28% 0,04% -0,21%
NATIONAL GRID PLC -1,16% 4,62% 3,44% -3,38% 1,95% 2,22%
PRUDENTIAL PLC -6,57% 3,22% -3,85% 8,31% 3,35% -9,81%
BARCLAYS PLC -6,36% 7,88% -2,02% 6,51% 2,81% 1,22%
ROYAL BANK OF
SCOTLAND GROUP PLC
-
15,79% 11,13% -3,43% -3,73% 4,51%
-
12,01%
TESCO PLC -8,41% 12,50% 1,46% 6,53% -0,24% -5,71%
EXPERIAN PLC 5,71% -0,51% -2,78% 4,42% -4,14% 7,49%
CRH PLC -6,94% 3,24% -6,11% 7,44% -0,78% 2,96%
ASSOCIATED BRITISH
FOODS PLC
-
15,42% 13,27% 4,48% 4,63% -7,09% 3,60%
BAE SYSTEMS PLC -
11,22% 8,95% -6,28% 2,69% -0,59% 16,53%
STANDARD CHARTERED
PLC 9,97% 3,12% -6,65% 16,44% 3,33% -9,69%
IMPERIAL BRANDS PLC -8,03% 2,37% 0,53% -7,88% -4,37% 9,86%
LEGAL & GENERAL
GROUP PLC -4,59% 6,05% 3,77% 4,50% 3,28% -2,62%
Page 72
72
Individual Companies Cumulated Abnormal Returns of period (T= -10:+10) (Part 3)
Event number 13 14 15 16 17
Cumulated Average
Abnormal Returns of
companies on event period 0,78% -0,16% 0,49% 0,46% 1,10%
CAAR (-10;+10)
ROYAL DUTCH SHELL
PLC -0,96% -0,89% -4,07% -2,10% -12,32%
HSBC HOLDINGS PLC -0,24% 1,23% -3,45% 0,32% -4,46%
ASTRAZENECA PLC -7,37% -0,86% 3,30% -1,21% -2,22%
BP PLC 0,82% -3,44% -0,71% -5,09% -3,52%
GLAXOSMITHKLINE PLC -2,75% 0,86% 0,55% 0,13% -7,11%
BRITISH AMERICAN
TOBACCO PLC -4,44% -7,46% -3,50% 3,55% -1,59%
DIAGEO PLC -7,03% -2,70% -7,69% -1,61% -5,46%
UNILEVER PLC -4,57% -6,01% -6,79% -6,07% 7,21%
RIO TINTO PLC 1,82% -5,77% 1,69% 2,63% -4,52%
RECKITT BENCKISER
GROUP PLC 0,13% -3,31% -7,30% 0,85% 3,67%
VODAFONE GROUP PLC 3,57% 6,08% 0,34% -8,19% 0,11%
LLOYDS BANKING
GROUP PLC 5,18% 15,97% 8,81% 1,06% 4,49%
RELX PLC -6,48% -4,90% -4,77% -1,69% 3,89%
NATIONAL GRID PLC -3,15% 8,06% 1,77% 2,61% 8,71%
PRUDENTIAL PLC 1,18% -11,62% -5,42% 0,13% 8,30%
BARCLAYS PLC 5,33% 13,90% 15,22% 3,13% 3,75%
ROYAL BANK OF
SCOTLAND GROUP PLC 10,66% 15,16% 9,54% 4,33% -0,19%
TESCO PLC 5,23% 4,19% -1,62% 6,57% 2,56%
EXPERIAN PLC -4,92% -2,90% -9,82% -3,24% 7,18%
CRH PLC 0,34% 2,01% 3,29% -1,97% 3,24%
ASSOCIATED BRITISH
FOODS PLC -0,49% -2,77% 3,47% -0,59% 5,65%
BAE SYSTEMS PLC 4,59% -5,30% 4,04% -3,86% 10,20%
STANDARD CHARTERED
PLC 8,67% -1,08% 9,37% 0,48% -6,44%
IMPERIAL BRANDS PLC 2,37% -11,73% -5,60% 8,67% -4,75%
LEGAL & GENERAL
GROUP PLC 12,07% -0,82% 11,66% 12,72% 11,07%
Page 73
73
Annex 4
Individual Companies Cumulated Abnormal Returns of period (T= -5:+5) (Part 1)
Event 1 2 3 4 5 6
Cumulated Average
Abnormal Returns of
companies on event period
0,12% 0,13% -0,43% 0,49% -0,17% -0,07%
CAAR (-5;5)
ROYAL DUTCH SHELL
PLC -4,02% -0,87% 1,04% 0,95% 0,30% -0,70%
BP PLC -6,23% -3,20% 2,40% 0,43% 0,28% -2,85%
HSBC HOLDINGS PLC 2,60% 2,09% 0,80% 3,29% 0,47% -0,05%
LLOYDS BANKING
GROUP PLC 1,37% 0,34% -4,37% -1,79% -1,22% -0,35%
PRUDENTIAL PLC 6,20% -1,84% -4,53% 3,52% -4,43% -1,78%
BARCLAYS PLC 7,35% 0,09% -3,38% -0,92% 3,24% 0,26%
ROYAL BANK OF
SCOTLAND GROUP PLC 11,58% -1,41% 2,00% 1,46% -1,13% -3,36%
STANDARD CHARTERED
PLC 3,03% 11,25% 3,75% 4,88% 0,31% -3,80%
LEGAL & GENERAL
GROUP PLC 11,01% -0,87% 0,12% 6,61% 0,91% -0,01%
ASTRAZENECA PLC -0,18% -8,76% 0,39% 1,33% -0,18% 4,11%
GLAXOSMITHKLINE PLC -1,23% -3,82% -0,72% -0,06% -0,75% 3,83%
BRITISH AMERICAN
TOBACCO PLC -3,71% 3,86% 3,20% -0,44% 3,58% 0,58%
DIAGEO PLC -2,03% 0,33% -1,34% 0,70% 1,47% 1,78%
UNILEVER PLC -4,56% 0,82% -2,03% -1,22% -0,82% 0,30%
RECKITT BENCKISER
GROUP PLC -3,56% 0,64% -1,95% 0,14% 2,25% 3,89%
TESCO PLC -2,32% -4,07% -2,63% -1,85% 4,90% -1,51%
ASSOCIATED BRITISH
FOODS PLC -5,47% -6,12% -4,30% 0,09% -5,43% -11,15%
IMPERIAL BRANDS PLC -1,49% 2,76% 1,56% 0,03% -0,41% 2,47%
EXPERIAN PLC -0,43% -2,22% -1,55% 0,44% 1,11% 1,02%
CRH PLC 1,89% 3,72% -0,85% 1,60% -4,33% -0,59%
BAE SYSTEMS PLC -2,55% 0,92% -1,03% -0,24% -0,08% 5,43%
RIO TINTO PLC -3,57% 15,55% -1,46% 0,28% 0,04% -4,79%
VODAFONE GROUP PLC 2,19% -2,23% 0,00% -2,44% 1,39% -0,88%
RELX PLC -2,41% -2,92% 2,00% 1,84% -2,61% 4,41%
NATIONAL GRID PLC -0,53% -0,74% 2,07% -6,37% -3,00% 2,09%
Page 74
74
Individual Companies Cumulated Abnormal Returns of period (T= -5:+5) (Part 2)
Event 7 8 9 10 11 12
Cumulated Average
Abnormal Returns of
companies on event period
-0,18% 1,47% 0,36% 0,14% -0,51% -0,03%
CAAR (-5;5)
ROYAL DUTCH SHELL
PLC -0,41% -3,90% -0,15% 1,27% 0,70% -0,09%
BP PLC 0,50% -3,41% 0,60% 1,81% 0,66% 0,22%
HSBC HOLDINGS PLC 3,57% -1,19% -1,45% 1,64% 0,72% -1,97%
LLOYDS BANKING
GROUP PLC 2,17% 6,53% 1,19% 3,27% -2,31% -8,13%
PRUDENTIAL PLC -0,98% 4,99% -3,95% 6,07% 4,55% -4,32%
BARCLAYS PLC -0,72% 4,70% 1,12% 0,54% 1,26% -1,59%
ROYAL BANK OF
SCOTLAND GROUP PLC 1,45% 6,53% -0,02% 1,71% 3,66% -5,11%
STANDARD CHARTERED
PLC 5,79% 3,63% -4,98% 3,52% 3,41% -6,09%
LEGAL & GENERAL
GROUP PLC 1,73% 4,43% 1,62% 3,31% 2,09% 1,04%
ASTRAZENECA PLC -2,03% -9,53% -0,09% -5,47% 1,88% 8,92%
GLAXOSMITHKLINE PLC -4,94% -2,86% -2,62% -3,45% -1,38% 2,07%
BRITISH AMERICAN
TOBACCO PLC 2,14% 0,84% 8,27% -2,94% -2,86% -0,16%
DIAGEO PLC 0,05% -0,06% 1,32% 0,83% -3,09% 0,25%
UNILEVER PLC -1,22% -3,09% 4,58% 1,72% -2,74% -2,96%
RECKITT BENCKISER
GROUP PLC 0,32% 0,70% 3,67% -9,25% -7,22% -4,34%
TESCO PLC -4,11% 7,37% 1,33% 5,55% 1,45% -7,36%
ASSOCIATED BRITISH
FOODS PLC -2,63% 9,51% 1,48% 0,06% -3,18% 4,64%
IMPERIAL BRANDS PLC -8,64% 5,48% 1,99% -3,62% -5,54% 2,89%
EXPERIAN PLC 2,32% 1,40% -1,23% 3,47% -2,89% 3,67%
CRH PLC 2,54% 2,11% -5,60% 1,79% 0,75% 0,13%
BAE SYSTEMS PLC -6,11% 8,13% -0,51% -2,22% 0,00% 5,85%
RIO TINTO PLC -6,90% -1,72% -6,10% -1,58% -0,07% -5,06%
VODAFONE GROUP PLC 9,49% -6,61% 8,74% -2,60% -1,09% 16,52%
RELX PLC 1,67% 0,82% -3,00% 1,52% -1,22% -1,61%
NATIONAL GRID PLC 0,53% 2,03% 2,72% -3,32% -0,29% 1,95%
Page 75
75
Individual Companies Cumulated Abnormal Returns of period (T= -5:+5) (Part 3)
Event 13 14 15 16 17
Cumulated Average
Abnormal Returns of
companies on event period
0,87% -1,39% 0,20% 0,81% 1,15%
CAAR (-5;5)
ROYAL DUTCH SHELL
PLC -1,73% 1,78% 0,30% -1,83% -7,39%
BP PLC 0,26% 1,31% 2,15% -4,77% 0,26%
HSBC HOLDINGS PLC 4,26% 0,89% -3,07% 2,24% -0,22%
LLOYDS BANKING
GROUP PLC 1,91% -3,79% -0,93% 0,27% -1,14%
PRUDENTIAL PLC 4,66% -15,29% 6,18% 1,26% 3,56%
BARCLAYS PLC 5,78% 1,52% 5,94% 4,76% 3,25%
ROYAL BANK OF
SCOTLAND GROUP PLC 6,20% -5,24% 2,23% 7,15% 2,49%
STANDARD CHARTERED
PLC 8,13% -3,40% 4,70% 2,46% -6,06%
LEGAL & GENERAL
GROUP PLC 14,02% -4,06% 4,83% 8,86% 4,60%
ASTRAZENECA PLC -7,81% -1,56% 5,79% -0,23% 0,46%
GLAXOSMITHKLINE PLC -5,21% 2,45% 2,55% 1,18% -6,62%
BRITISH AMERICAN
TOBACCO PLC 3,90% 0,97% -1,06% 4,28% 3,39%
DIAGEO PLC -5,07% 3,79% -4,76% -1,62% -0,39%
UNILEVER PLC -4,84% -0,45% -2,94% -7,98% 7,26%
RECKITT BENCKISER
GROUP PLC -0,25% -0,86% -5,72% -2,87% 7,56%
TESCO PLC 6,80% -4,64% -2,14% 7,44% 5,18%
ASSOCIATED BRITISH
FOODS PLC -3,91% -7,49% 0,79% -0,08% 2,44%
IMPERIAL BRANDS PLC 5,02% -6,36% -4,23% 3,15% -3,83%
EXPERIAN PLC -6,29% 1,05% -4,53% -2,75% 4,73%
CRH PLC -0,53% -0,32% 3,25% -3,15% 3,88%
BAE SYSTEMS PLC 2,98% 0,03% 0,56% -1,69% 2,65%
RIO TINTO PLC 3,70% -1,16% -3,52% -0,07% -6,33%
VODAFONE GROUP PLC 2,68% 3,16% -0,96% -0,74% 0,64%
RELX PLC -7,57% 0,13% -0,97% -0,24% 4,26%
NATIONAL GRID PLC -5,28% 2,79% 0,58% 5,15% 4,21%
Page 76
76
Annex 5
Individual Companies Cumulated Abnormal Returns of period (T= -2:+2) (Part 1)
Event 1 2 3 4 5 6
Cumulated Average
Abnormal Returns of
companies on event period -0,05% 0,04% -0,17% 0,14% 0,19% -0,79%
CAR (-2:+2)
ROYAL DUTCH SHELL
PLC -1,67% -1,79% 0,95% 2,29% -0,51% 1,72%
BP PLC -2,21% -1,67% 1,54% 1,63% 0,74% -0,35%
HSBC HOLDINGS PLC 3,48% 2,18% 0,69% 1,82% 1,26% -0,76%
LLOYDS BANKING
GROUP PLC 2,15% -0,43% -2,48% -0,01% 1,05% -0,41%
PRUDENTIAL PLC 2,27% -1,10% -2,10% 1,74% -2,58% -1,33%
BARCLAYS PLC 3,55% 0,04% -0,15% -2,47% 2,57% -0,56%
ROYAL BANK OF
SCOTLAND GROUP PLC 8,80% 2,10% 1,87% -3,28% 0,82% -2,65%
STANDARD CHARTERED
PLC 2,22% 7,94% 5,40% 3,97% 0,47% -1,00%
LEGAL & GENERAL
GROUP PLC 5,29% -0,26% 0,63% 2,90% -0,14% -0,30%
ASTRAZENECA PLC 0,85% -1,42% -0,51% -1,23% 2,39% 0,25%
GLAXOSMITHKLINE PLC 0,54% 0,04% -0,59% 0,40% 0,78% 0,24%
BRITISH AMERICAN
TOBACCO PLC -1,15% 1,68% 0,69% -1,98% 1,73% -0,67%
DIAGEO PLC -1,37% -0,01% -1,09% -1,10% -0,15% 0,43%
UNILEVER PLC -1,32% -0,29% -1,94% -1,19% -0,31% -1,52%
RECKITT BENCKISER
GROUP PLC -2,10% 1,18% -0,62% -0,66% 3,64% -0,95%
TESCO PLC -5,78% -1,15% -2,99% 0,52% 0,71% -2,52%
ASSOCIATED BRITISH
FOODS PLC -0,29% 1,30% -0,71% -1,43% -2,80% -9,31%
IMPERIAL BRANDS PLC -0,30% 0,24% 1,95% -0,87% 0,59% 0,36%
EXPERIAN PLC -0,70% -3,28% -0,67% 1,63% 1,41% 0,61%
CRH PLC -1,76% -0,22% -0,33% 1,40% -2,39% -0,54%
BAE SYSTEMS PLC -0,83% -1,64% -1,53% -0,56% -1,27% 0,11%
RIO TINTO PLC -6,42% 2,94% -1,46% 4,18% -1,24% 0,22%
VODAFONE GROUP PLC -1,93% -2,44% -1,09% -2,13% 1,34% 0,75%
RELX PLC -1,75% -2,30% 0,04% 0,07% -2,48% -0,06%
NATIONAL GRID PLC -0,91% -0,64% 0,38% -2,07% -0,95% -1,37%
Page 77
77
Individual Companies Cumulated Abnormal Returns of period (T= -2:+2) (Part 2)
Event 7 8 9 10 11 12
Cumulated Average Abnormal
Returns of companies on event
period 0,01% 0,77% 0,12% 0,34% -0,59% 0,07%
CAR (-2:+2)
ROYAL DUTCH SHELL PLC -1,92% -1,87% 0,15% -1,25% 2,40% -0,31%
BP PLC -0,41% -1,01% 1,18% -1,84% 2,10% 0,30%
HSBC HOLDINGS PLC 2,20% -0,30% -0,89% 0,71% -1,21% -0,46%
LLOYDS BANKING GROUP
PLC 0,72% 4,45% 2,12% 3,88% -3,76% -1,85%
PRUDENTIAL PLC 0,26% 1,95% 0,68% 4,13% 0,13% -2,45%
BARCLAYS PLC 1,24% 2,89% 1,70% 3,50% -2,26% 2,75%
ROYAL BANK OF
SCOTLAND GROUP PLC 2,32% 7,35% 2,04% 3,06% -1,70% -0,18%
STANDARD CHARTERED
PLC 3,96% -0,75% -1,89% 3,41% -0,47% -3,74%
LEGAL & GENERAL
GROUP PLC 3,68% 3,45% 4,40% 2,56% 0,49% 2,19%
ASTRAZENECA PLC -1,20% -6,43% 0,23% -3,08% -1,74% 7,06%
GLAXOSMITHKLINE PLC 1,03% -2,42% -0,42% -2,32% -1,31% 1,91%
BRITISH AMERICAN
TOBACCO PLC 1,09% 1,66% -2,98% -0,10% 0,08% -2,94%
DIAGEO PLC 0,22% 0,89% -0,25% -0,97% -0,17% -2,43%
UNILEVER PLC -0,14% -0,87% 1,45% -1,49% -0,91% -2,67%
RECKITT BENCKISER
GROUP PLC -1,49% -2,77% 1,53% -9,19% -4,88% -2,00%
TESCO PLC -4,76% 2,04% -3,01% 6,02% -5,00% -5,86%
ASSOCIATED BRITISH
FOODS PLC -1,86% 5,08% 0,32% 1,69% 0,63% 1,81%
IMPERIAL BRANDS PLC -4,72% 0,88% -2,47% -1,78% -0,31% -0,31%
EXPERIAN PLC 3,34% 0,92% -1,83% 1,29% 0,05% 2,19%
CRH PLC 0,44% -0,33% -1,89% 1,84% 1,07% 1,31%
BAE SYSTEMS PLC -4,80% 2,57% -0,85% 0,55% -0,21% 1,54%
RIO TINTO PLC -7,04% 2,29% -0,18% -1,94% 3,38% -6,14%
VODAFONE GROUP PLC 7,59% -1,77% 3,35% 1,32% -0,89% 11,97%
RELX PLC 1,96% 1,12% -0,11% 0,25% 0,33% -1,12%
NATIONAL GRID PLC -1,41% 0,31% 0,65% -1,83% -0,46% 1,18%
Page 78
78
Individual Companies Cumulated Abnormal Returns of period (T= -2:+2) (Part 3)
Event 13 14 15 16 17
Cumulated Average
Abnormal Returns of
companies on event period 0,61% -0,39% 0,09% 1,13% 0,61%
CAR (-2:+2)
ROYAL DUTCH SHELL
PLC -1,97% 0,96% 0,81% -4,95% -7,39%
BP PLC -1,20% 0,11% 2,34% -5,97% -3,50%
HSBC HOLDINGS PLC 0,70% 0,16% 0,01% 0,70% 0,64%
LLOYDS BANKING
GROUP PLC -0,66% -2,59% -2,07% 3,93% 0,01%
PRUDENTIAL PLC 0,89% 0,37% 12,50% 3,44% 3,05%
BARCLAYS PLC 1,90% -0,97% 0,05% 7,78% 1,21%
ROYAL BANK OF
SCOTLAND GROUP PLC 0,99% -3,38% 1,00% 9,86% 0,90%
STANDARD CHARTERED
PLC 3,26% -1,83% 3,23% 1,54% -2,58%
LEGAL & GENERAL
GROUP PLC 6,91% -1,36% 0,60% 8,50% 2,66%
ASTRAZENECA PLC -1,96% -2,94% 0,28% -0,80% -0,06%
GLAXOSMITHKLINE PLC -0,71% 1,42% 1,69% -1,19% 1,87%
BRITISH AMERICAN
TOBACCO PLC 0,77% -1,65% -0,39% 2,10% 1,86%
DIAGEO PLC -2,14% 2,12% -2,32% -1,58% -2,38%
UNILEVER PLC -0,81% -0,10% -1,88% -0,89% 4,06%
RECKITT BENCKISER
GROUP PLC -0,93% -3,73% -2,15% 1,24% 4,47%
TESCO PLC 5,11% -1,01% -1,46% 3,86% 1,97%
ASSOCIATED BRITISH
FOODS PLC 2,81% -1,29% -1,56% 1,20% 0,87%
IMPERIAL BRANDS PLC 0,01% 5,97% -0,15% 3,33% 2,32%
EXPERIAN PLC 3,02% 0,75% -4,53% -2,19% 1,95%
CRH PLC -1,16% 0,43% 0,27% -2,21% 2,26%
BAE SYSTEMS PLC 3,79% -1,49% -4,60% 1,04% 2,09%
RIO TINTO PLC 0,55% -0,50% 0,80% -1,82% -1,88%
VODAFONE GROUP PLC -0,21% 1,27% 0,46% -0,26% -0,77%
RELX PLC -1,11% -1,59% -2,87% -0,77% 0,39%
NATIONAL GRID PLC -2,61% 1,05% 2,29% 2,32% 1,17%
Page 79
79
Annex 6
Individual Companies Cumulated Abnormal Returns of period (T= -10:0) (Part 1)
Event 1 2 3 4 5 6
Cumulated Average
Abnormal Returns of
companies on event period -0,42% -0,23% 0,41% 0,32% 0,39% 0,22%
CAR (-10:0)
ROYAL DUTCH SHELL
PLC -2,24% -1,33% 1,71% -1,88% 1,32% 2,82%
BP PLC -1,38% -3,07% 0,67% -3,31% -1,25% 2,29%
HSBC HOLDINGS PLC 2,28% 0,77% -2,03% 3,46% 0,79% -0,85%
LLOYDS BANKING
GROUP PLC -3,09% 3,99% -2,41% -1,02% 2,68% 1,45%
PRUDENTIAL PLC -4,16% -5,89% -1,40% 0,97% -2,76% -2,99%
BARCLAYS PLC 4,83% 3,52% 1,12% -2,77% 5,52% -1,80%
ROYAL BANK OF
SCOTLAND GROUP PLC 1,99% -2,01% 3,34% -1,83% 5,28% -3,38%
STANDARD CHARTERED
PLC 2,77% 10,87% 3,28% 4,57% 4,11% -1,86%
LEGAL & GENERAL
GROUP PLC -1,96% -0,27% 1,92% 4,16% 1,44% -2,32%
ASTRAZENECA PLC -1,24% 0,54% 1,16% 2,99% -4,90% -2,25%
GLAXOSMITHKLINE PLC 1,64% -1,73% -0,95% 3,47% -1,31% -0,17%
BRITISH AMERICAN
TOBACCO PLC -3,02% -1,60% 2,79% 1,99% -2,10% 0,82%
DIAGEO PLC 1,51% -0,01% 0,24% -1,52% 0,67% -0,20%
UNILEVER PLC 0,50% 0,05% -0,59% 2,02% -1,25% 2,51%
RECKITT BENCKISER
GROUP PLC -1,16% -2,54% -0,34% 3,40% 4,80% 3,32%
TESCO PLC -6,57% -3,30% 0,41% -1,78% 5,09% 0,62%
ASSOCIATED BRITISH
FOODS PLC 1,31% -7,34% -0,11% 1,12% -6,14% -12,57%
IMPERIAL BRANDS PLC -1,02% 0,07% -0,64% -1,54% -0,69% 8,05%
EXPERIAN PLC 1,50% -1,61% -1,75% -3,84% 0,87% 2,43%
CRH PLC -2,87% -2,89% -1,90% 0,04% -0,55% -1,88%
BAE SYSTEMS PLC 0,88% 1,12% 1,13% 3,30% 2,59% 2,63%
RIO TINTO PLC 3,22% 5,88% -2,51% 1,06% -5,54% -2,32%
VODAFONE GROUP PLC -2,39% 3,65% 3,65% -1,54% 3,04% 3,58%
RELX PLC -1,75% -2,33% 0,38% 1,25% -3,75% 1,51%
NATIONAL GRID PLC 0,00% -0,36% 3,17% -4,76% 1,83% 6,00%
Page 80
80
Individual Companies Cumulated Abnormal Returns of period (T= -10:0) (Part 2)
Event 7 8 9 10 11 12
Cumulated Average
Abnormal Returns of
companies on event period -0,95% 0,56% 0,56% 0,57% -0,94% 0,07%
CAR (-10:0)
ROYAL DUTCH SHELL
PLC -1,15% 0,08% -1,05% 0,19% 1,14% -1,23%
BP PLC 1,62% 0,68% 0,58% -0,15% -1,87% -3,65%
HSBC HOLDINGS PLC 7,41% -3,39% 1,47% 4,04% -1,66% -1,46%
LLOYDS BANKING
GROUP PLC -3,65% 6,10% 4,04% 2,46% -1,54% -0,58%
PRUDENTIAL PLC -1,12% 0,27% -3,41% 4,74% 1,44% -2,25%
BARCLAYS PLC -2,65% 4,45% 3,78% 6,23% -1,84% 3,81%
ROYAL BANK OF
SCOTLAND GROUP PLC -10,12% 6,03% 0,96% 3,69% 0,32% 1,15%
STANDARD CHARTERED
PLC 9,67% -0,23% 0,02% 11,43% -0,44% -6,67%
LEGAL & GENERAL
GROUP PLC -3,41% 0,37% 3,37% 5,12% -0,14% 2,74%
ASTRAZENECA PLC -0,49% -7,73% -0,44% -11,08% 5,11% -2,11%
GLAXOSMITHKLINE PLC 0,83% -2,03% -1,24% -4,90% 0,72% 1,12%
BRITISH AMERICAN
TOBACCO PLC -19,00% -4,88% 6,75% -4,76% -9,24% 2,24%
DIAGEO PLC 2,86% -3,26% 1,53% -2,76% -0,39% -2,89%
UNILEVER PLC 1,76% -2,81% -0,26% -2,70% -0,12% -0,70%
RECKITT BENCKISER
GROUP PLC 7,16% 1,25% 5,12% -9,99% -1,57% 1,04%
TESCO PLC -8,55% 10,97% 2,03% 5,61% -0,27% -2,67%
ASSOCIATED BRITISH
FOODS PLC -5,24% 5,91% -0,02% 2,95% -5,14% 0,90%
IMPERIAL BRANDS PLC -10,26% 1,35% 2,46% -6,09% -11,40% 8,80%
EXPERIAN PLC 3,21% -3,42% -1,70% 3,34% 0,68% 2,31%
CRH PLC -2,68% 2,09% -0,68% 5,50% -2,77% -0,04%
BAE SYSTEMS PLC -2,71% 6,90% -1,60% 5,60% 0,79% 9,19%
RIO TINTO PLC -5,12% 0,16% -5,89% 4,34% 3,12% -4,10%
VODAFONE GROUP PLC 15,51% -5,47% 1,22% -0,43% -3,04% -0,33%
RELX PLC 5,18% -2,07% -6,37% -2,13% 2,20% -1,75%
NATIONAL GRID PLC -2,68% 2,62% 3,35% -6,03% 2,33% -0,99%
Page 81
81
Individual Companies Cumulated Abnormal Returns of period (T= -10:0) (Part 3)
Event 13 14 15 16 17
Cumulated Average
Abnormal Returns of
companies on event period 0,26% -0,51% 1,21% 0,36% 1,44%
CAR (-10:0)
ROYAL DUTCH SHELL
PLC -2,98% 2,30% -3,64% 0,49% -6,33%
BP PLC -0,80% -1,36% -1,09% -1,26% -0,78%
HSBC HOLDINGS PLC -1,33% 1,51% 0,58% 0,86% -1,02%
LLOYDS BANKING
GROUP PLC -2,32% -0,66% 16,80% 2,00% 4,73%
PRUDENTIAL PLC -3,08% -1,01% -2,81% 1,22% 2,11%
BARCLAYS PLC -3,07% 0,67% 15,73% 1,90% 0,96%
ROYAL BANK OF
SCOTLAND GROUP PLC -0,76% -1,59% 22,52% 2,99% 2,21%
STANDARD CHARTERED
PLC 2,55% -1,60% 4,56% 5,26% -3,25%
LEGAL & GENERAL
GROUP PLC 0,72% 0,83% 15,27% 7,21% 10,43%
ASTRAZENECA PLC -0,46% 1,45% -4,64% -2,56% -2,37%
GLAXOSMITHKLINE PLC 3,21% 3,54% -3,76% 0,68% 0,29%
BRITISH AMERICAN
TOBACCO PLC -6,15% 0,55% -5,90% -0,82% -1,52%
DIAGEO PLC 3,19% -0,24% -6,49% -3,01% -5,84%
UNILEVER PLC 3,00% -1,60% -5,53% 0,55% 6,07%
RECKITT BENCKISER
GROUP PLC 3,55% -1,36% -5,60% 1,25% 3,73%
TESCO PLC 2,13% 3,55% 2,90% 3,89% 1,80%
ASSOCIATED BRITISH
FOODS PLC 1,66% -4,64% 0,87% -4,28% 5,84%
IMPERIAL BRANDS PLC 1,14% -12,73% 3,30% 1,14% -1,18%
EXPERIAN PLC 0,33% 1,14% -9,40% -2,86% 4,58%
CRH PLC 2,69% 0,44% 1,63% 1,47% 1,93%
BAE SYSTEMS PLC -0,72% -5,26% -1,90% 0,51% 6,50%
RIO TINTO PLC 0,87% -3,12% -1,39% 5,06% -4,12%
VODAFONE GROUP PLC 3,69% 2,70% 1,96% -7,80% -0,72%
RELX PLC 0,41% -0,91% -7,07% -2,64% 3,38%
NATIONAL GRID PLC -0,98% 4,56% 3,34% -2,13% 8,66%
Page 82
82
Annex 7
Individual Companies Cumulated Abnormal Returns of period (T= 0:+10) (Part 1)
Event 1 2 3 4 5 6
Cumulated Average
Abnormal Returns of
companies on event period -0,11% -0,01% -0,59% -0,27% -0,67% -0,10%
CAR (0:+10
ROYAL DUTCH SHELL
PLC -9,07% -2,07% 0,65% 0,60% 0,72% 0,57%
BP PLC -7,22% -3,87% 2,92% 0,17% 1,31% -2,98%
HSBC HOLDINGS PLC -2,24% 3,36% 0,37% 0,28% 1,45% 1,25%
LLOYDS BANKING
GROUP PLC -4,44% 1,60% -5,20% -4,85% -1,77% -0,18%
PRUDENTIAL PLC 0,86% -0,66% -2,75% 2,08% -1,58% -0,13%
BARCLAYS PLC 2,44% -2,55% -4,88% -3,68% 1,12% 0,23%
ROYAL BANK OF
SCOTLAND GROUP PLC 1,22% 2,94% -1,58% -3,36% -3,34% -2,69%
STANDARD CHARTERED
PLC -3,33% 6,62% -3,72% -1,50% -1,65% -2,91%
LEGAL & GENERAL
GROUP PLC 7,57% -0,93% 3,63% 3,42% 2,18% 0,09%
ASTRAZENECA PLC 10,77% -7,15% -4,95% 3,23% 0,67% 6,04%
GLAXOSMITHKLINE PLC 2,81% -0,29% -1,88% 2,50% -1,17% -0,63%
BRITISH AMERICAN
TOBACCO PLC 0,35% 8,32% 1,63% -0,30% -1,94% 3,00%
DIAGEO PLC 2,04% 4,21% -0,56% 2,03% -0,09% 4,06%
UNILEVER PLC -1,33% -1,97% 0,89% 0,73% -3,12% 2,96%
RECKITT BENCKISER
GROUP PLC -1,16% 1,58% 2,78% 0,24% -0,04% 0,89%
TESCO PLC -7,41% -3,01% -3,39% -8,23% -0,87% -1,42%
ASSOCIATED BRITISH
FOODS PLC -2,51% -6,20% 0,51% -0,28% -4,41% -4,08%
IMPERIAL BRANDS PLC -0,85% 6,00% 3,56% 1,96% -0,08% 1,06%
EXPERIAN PLC 0,42% -2,27% 0,03% -0,09% -1,74% 0,36%
CRH PLC 3,49% 0,78% -2,12% 1,26% -2,48% -2,18%
BAE SYSTEMS PLC -0,70% -2,61% -0,02% 0,56% -0,47% 4,26%
RIO TINTO PLC 0,56% 4,23% -3,56% -6,23% 3,76% -1,10%
VODAFONE GROUP PLC 2,71% -4,24% -1,98% 3,68% 1,17% -7,26%
RELX PLC 2,34% -0,79% 1,37% 0,80% -1,76% 2,21%
NATIONAL GRID PLC -0,06% -1,28% 3,37% -1,68% -2,72% -3,86%
Page 83
83
Individual Companies Cumulated Abnormal Returns of period (T= 0:+10) (Part 2)
Event 7 8 9 10 11 12
Cumulated Average
Abnormal Returns of
companies on event period -0,86% 0,68% -0,52% 0,47% 0,86% 0,34%
CAR (0:+10)
ROYAL DUTCH SHELL
PLC 3,22% -5,26% 1,13% -0,80% -2,52% -3,41%
BP PLC 3,29% -2,70% 1,76% -1,74% -3,24% 2,04%
HSBC HOLDINGS PLC -0,63% 1,13% -2,68% 1,83% 1,94% -0,26%
LLOYDS BANKING
GROUP PLC -4,80% 3,87% -1,56% -2,83% -0,02% -6,90%
PRUDENTIAL PLC -5,45% 2,95% -0,44% 3,57% 1,90% -7,55%
BARCLAYS PLC -3,71% 3,43% -5,80% 0,28% 4,66% -2,59%
ROYAL BANK OF
SCOTLAND GROUP PLC -5,67% 5,10% -4,39% -7,41% 4,19% -13,16%
STANDARD CHARTERED
PLC 0,30% 3,35% -6,67% 5,01% 3,77% -3,01%
LEGAL & GENERAL
GROUP PLC -1,18% 5,68% 0,40% -0,61% 3,41% -5,36%
ASTRAZENECA PLC 0,65% -3,60% 0,97% -5,09% -1,88% 13,84%
GLAXOSMITHKLINE PLC -3,42% -3,29% 2,59% 1,45% 0,65% 3,47%
BRITISH AMERICAN
TOBACCO PLC 4,26% 2,92% -0,74% -3,22% 7,86% 3,25%
DIAGEO PLC 2,51% -4,98% 2,12% 3,60% -0,96% 3,16%
UNILEVER PLC 4,03% -0,94% 4,53% 6,55% 0,94% -0,42%
RECKITT BENCKISER
GROUP PLC -2,87% -6,58% 3,72% 4,88% -0,97% -5,53%
TESCO PLC 0,14% 1,53% -0,57% 0,92% 0,03% -3,03%
ASSOCIATED BRITISH
FOODS PLC -
10,18% 7,36% 4,50% 1,68% -1,95% 2,70%
IMPERIAL BRANDS PLC 2,23% 1,02% -1,93% -1,78% 7,04% 1,06%
EXPERIAN PLC 2,51% 2,91% -1,09% 1,08% -4,82% 5,18%
CRH PLC -4,26% 1,15% -5,42% 1,94% 1,99% 3,01%
BAE SYSTEMS PLC -8,51% 2,05% -4,67% -2,91% -1,37% 7,34%
RIO TINTO PLC 2,70% 5,26% 2,03% -6,75% -1,61% -5,56%
VODAFONE GROUP PLC 1,86% -8,07% 2,34% 2,11% 4,95% 15,63%
RELX PLC -0,15% 0,79% -3,28% 7,40% -2,16% 1,54%
NATIONAL GRID PLC 1,53% 2,00% 0,09% 2,66% -0,38% 3,21%
Page 84
84
Individual Companies Cumulated Abnormal Returns of period (T= 0:+10) (Part 3)
Event 13 14 15 16 17
Cumulated Average
Abnormal Returns of
companies on event period 0,52% 0,35% -0,72% 0,10% -0,35%
CAR (0:+10)
ROYAL DUTCH SHELL
PLC 2,02% -3,19% -0,43% -2,59% -6,00%
BP PLC 1,62% -2,08% 0,38% -3,83% -2,74%
HSBC HOLDINGS PLC 1,10% -0,28% -4,03% -0,54% -3,44%
LLOYDS BANKING
GROUP PLC 7,51% 16,62% -7,99% -0,94% -0,24%
PRUDENTIAL PLC 4,26% -10,61% -2,61% -1,09% 6,18%
BARCLAYS PLC 8,40% 13,23% -0,51% 1,23% 2,79%
ROYAL BANK OF
SCOTLAND GROUP PLC 11,42% 16,75% -12,98% 1,33% -2,41%
STANDARD CHARTERED
PLC 6,12% 0,52% 4,80% -4,78% -3,19%
LEGAL & GENERAL
GROUP PLC 11,35% -1,65% -3,61% 5,51% 0,64%
ASTRAZENECA PLC -6,90% -2,31% 7,94% 1,35% 0,15%
GLAXOSMITHKLINE PLC -5,95% -2,68% 4,31% -0,55% -7,39%
BRITISH AMERICAN
TOBACCO PLC 1,71% -8,01% 2,40% 4,37% -0,07%
DIAGEO PLC -10,22% -2,46% -1,20% 1,40% 0,38%
UNILEVER PLC -7,57% -4,41% -1,25% -6,62% 1,15%
RECKITT BENCKISER
GROUP PLC -3,42% -1,95% -1,70% -0,40% -0,06%
TESCO PLC 3,10% 0,64% -4,53% 2,68% 0,76%
ASSOCIATED BRITISH
FOODS PLC -2,15% 1,87% 2,60% 3,69% -0,19%
IMPERIAL BRANDS PLC 1,23% 1,00% -8,90% 7,53% -3,57%
EXPERIAN PLC -5,25% -4,04% -0,42% -0,38% 2,60%
CRH PLC -2,35% 1,58% 1,66% -3,44% 1,32%
BAE SYSTEMS PLC 5,31% -0,04% 5,94% -4,37% 3,71%
RIO TINTO PLC 0,95% -2,66% 3,08% -2,43% -0,40%
VODAFONE GROUP PLC -0,11% 3,38% -1,63% -0,39% 0,83%
RELX PLC -6,89% -3,99% 2,30% 0,96% 0,51%
NATIONAL GRID PLC -2,18% 3,50% -1,57% 4,74% 0,05%
Page 85
85
Annex 8
Various Sectors Cumulated Abnormal Returns of all periods
CAAR (-10:10)
CAAR
(-5:5)
CAAR
(-2:2)
CAAR
(-10:0)
CAAR
(0:10)
Energy Sector -4,07% -1,43% -1,33% -1,52% -2,55%
Financial Sector 11,04% 9,75% 9,11% 10,71% 0,34%
Health-Care sector -1,49% -2,03% -0,55% -1,74% 0,25%
Consumer Staples
Sector -4,39% -3,01% -3,23% -4,46% 0,08%
Industrial Sector 1,10% 0,93% -0,23% 1,43% -0,33%
Basic Resources -1,07% -1,34% -0,84% -0,61% -0,45%
Telecommunications 1,90% 1,60% 0,97% 1,01% 0,88%
Consumer
Discretionary -0,90% -0,35% -0,59% -0,97% 0,07%
Utilities 1,49% 0,27% -0,17% 1,06% 0,44%