Abnormal Returns and their relation to Brexit
Daan Blees 10815988
Supervisor: Spyridon Terovitis Bsc Economie & Bedrijfskunde
Statement of Originality
This document is written by Daan Blees. I hereby declare to take full responsibility for the contents of the thesis. I declare that the text presented is original and that the only sources used were the sources
I have quoted and mentioned in my reference list. The Faculty of Economics and Business is solely responsible for the supervision and completion of the work, not for its contents.
Abstract
In March 2017, the United Kingdom started its withdrawal process from the European Union after the citizens of the United Kingdom agreed upon a Brexit at the advisory referendum June 23rd of 2016.
This research investigates the effect of this decision on the Cumulative Abnormal Returns of Acquiring companies in the United Kingdom during a takeover announcement. Using an event study, Ordinary Least Squares and a Chow test for structural break, the effect of the Brexit on these Abnormal Returns is investigated. In this research, we find small significant positive abnormal returns for the companies, lying around 1,55% for the different event windows. The Brexit has a negative effect on the
Cumulative Abnormal Returns and is significant in two event windows. The Chow test suggests a structural break exists among the pre- and post-Brexit area. These results could come from the uncertainty and low investors’ confidence in the market after Brexit.
CONTENTS
1. Introduction ... 5
2. Literature Review ... 8
3. Methodology... 10
3.1 Data collection ... 10
3.2 Event study methodology ... 10
3.3 Regression Methodology ... 12
3.4 Hypotheses and Regression ... 14
4.1 Event Study Analysis ... 16
4.2 Regression Analysis ... 17
5. Conclusion ... 21
1. INTRODUCTION
On May 27 2015, the Brexit referendum was announced. The United Kingdom would no longer be part of the European Union. The referendum was held at the 23rd of June 2016. 51,9% of the British voters
decided they wanted to leave the European Union. In March 2017, the United Kingdom started to withdraw themselves from the European Union.
The decision to leave the EU triggered reactions for every part of the economy. The Great British Pound suffered a major depreciation which is still felt. The Pound depreciated against the Euro more than 14% over the last two years. Furthermore, according to the Chartered Institute of
Procurement & Supply (2017) - from now on: CIPS -, one in seven European companies that have UK suppliers removed their business partly or wholly out of the United Kingdom. Another result found in the report of the CIPS (2017) was that one third of the UK suppliers have increased their prices following the major depreciation. This makes sense because a weaker pound makes it cheaper for foreign buyers to buy goods and services in the UK. However, this leads to an increase in prices for the British people too and that is bad for the domestic economy. The economic shocks due to Brexit are felt in the financial markets as well. The Brexit generates a lot of uncertainty. The exit package is still uncertain, and as mentioned above, companies are – partly – removing their businesses out of the UK. All these effects of Brexit contribute to growing uncertainty in the financial markets. In Chart 1, investors’ confidence is given.
As can be seen in Chart 1, the confidence of investors in 2017 is the lowest as of 1995. Will the low confidence of the investors be visible in their reaction to Mergers and Acquisitions? Investors could receive post-Brexit acquisition announcements more negative than before due to the low level of confidence. They could see an acquisition by a domestic company as a last resort for staying
competitive in the international market. Cross-border acquisitions might trigger investors to believe that the confidence in the domestic economy is so low that domestic companies prefer to invest abroad. Conn et al. (2005) find that returns on acquisitions of domestic public targets are negative in the UK.
A lot of research has been done on the announcement effect of acquisitions, but the Brexit consequences regarding this subject remained untouched. That is the reason that I will focus on the relationship of Brexit and Cumulative Abnormal Returns. Similar research on announcement effects during crisis times conducted by di Beltratti & Paladino (2012) suggested that abnormal returns from 2007 to 2010 in the Eurozone were not significant.
The aim of this research is to assess the impact of the Brexit on the Cumulative Abnormal Returns of the acquiring companies’ shareholders. Therefore, the research question is: Does a
takeover announcement cause abnormal returns of acquiring firms in the United Kingdom before and after the Brexit announcement? We will investigate if investors react differently to acquisition
announcements pre- and post-Brexit. The acquisition data will be gathered from Zephyr database. Thereafter, the stock prices for the estimation period and event periods are derived from DataStream database. Firstly, the CARs or four different event windows will be calculated following the mean-adjusted model of De Jong (2007). These CARs thereafter will be analyzed through an event study, where the event is the announcement date, to see whether abnormal returns exist in the last 5 years.
After that, we want to know what the impact of the Brexit is on the abnormal returns. To investigate that effect, a regression is carried out where the CAR are the dependent variables while the explanatory variable is Brexit accompanied some control variables regarding firm size and deal characteristics, they will be explained later. We expect to see a negative effect of the Brexit variable due to the increased uncertainty in the financial markets. The regression will give an insight in the effect of the Brexit and will therefore get us closer to answering our research question. To investigate whether a structural break is present, the methodology of Chow (1960) will be used to test if the two time periods are the same or different.
I find evidence of lower CAR after Brexit. The 4 different event windows and their CAR were all significant. The smaller the event window got, the larger CAR significance got. Furthermore, we find positive CAR. The positive CAR is quite contrary to previous literature, which suggested that the CAR for acquiring companies mostly were negative or close to zero. My regression suggests that the Brexit has a negative and mostly significant effect on the CAR, which is in line with the expectations. Contrary
to previous literature, the method of payment and the difference between a domestic and cross-border acquisition were insignificant. The Chow test gave the confirmation that a structural break exists before and after Brexit.
This research is structured as follows. Section 2 will discuss the previous literature written about the subject, section 3 describes the methodology used in the research, section 4 contains the analysis of the results in relation to economic intuition and section 5 summarizes the conclusions of the research and provides recommendations for further research on the subject.
2. LITERATURE REVIEW
In this chapter, the previous literature regarding the subject will be reviewed. Firstly, we will take a look on the determinants of Mergers and Acquisitions. Subsequently, we will look at literature regarding market uncertainty and what the consequences of the uncertainty are. Because market uncertainty increased and the investors’ confidence decreased, it is important to look at that because we want to know the effect of the Brexit on the CAR. Lastly, the previous research regarding the estimations of abnormal returns will be reviewed and summarized.
M&A determinants
Mergers and Acquisitions take place when two companies combine. There are multiple reasons for companies to merge. According to Jensen (1983), synergies are a major gain from a merger. Gaughan (2002) mentions that companies planning a merger are planning on reducing costs with economies of scale as the major source of synergy. Mergers and Acquisitions are, according to Schoenberg (2006), a popular mean of corporate growth. The popularity can be derived from financial, managerial and strategic perspectives of a takeover. When the number of potential bidders reduces and the number of potential targets increased, mostly during crisis periods, that may allow stock prices to more clearly reflect the advantages for the acquiring company (James and Wier, 1987).
Uncertainty and Consequences
The Brexit led to an increase of uncertainty in the market. The negotiations of the Brexit package are still pending and the British economy sees an outflow of business from the country. Vincent and Bamiro (2013) study the impact of market uncertainty on the significance of stock price fluctuations. In their paper, they claim that an accurate prediction of long- and short-term prices would not be feasible (Vincent & Bamiro, 2013). According to them, information asymmetries between the investors and the information handlers cause the infeasibility of predictions.
In the study of Bloom (2014), Bloom investigates the effect of market uncertainty on economic and financial crises. He found that macro and micro uncertainty rise in recessions. Macro uncertainty contains factors like GDP-growth and volatility where micro uncertainty comes from firm-specific indicators as future growth, cost and sales. Bloom (2014) also argues that bad news events which cause recessions, contribute to an increase in uncertainty. When this uncertainty is high, which is present since Brexit, the expected revenue of investments is low, which induces companies to postpone their investments. Furthermore, the cost of capital will increase due to a higher risk
premium (Bloom, 2014). However, Bloom sees some positive effects of uncertainty. When uncertainty is high, the profits can be really high which might make people more willing to engage in investing.
Ebell et al. (2016) estimate the impact of leaving the EU on the British economy. The depreciation of the currency will increase the exports, whereas it decreases the imports. According to them, this will eventually result in a decrease of consumption, wages and GDD. They conclude that the GDP of the UK will decrease in the long-run by 2,7 % (Ebell et al., 2016). This statement is supported by Ottaviano et al. (2016), they estimate a welfare loss from -1.13% to -3.09%. From the previous research, we conclude that Brexit has a negative effect on the British economy and we will therefore expect a negative Brexit coefficient.
Cumulative Abnormal Returns
Previous research on the announcement effect of a takeover find mixed results. For this research, we will focus on the returns of acquiring companies. In a sample of older studies regarding abnormal returns, Hudgins & Seifert (1996), Pilloff (1996) and Houston & Ryngaert (1994) find negative returns and insignificant negative returns. Contrary to the findings, other research indicates that the acquirer results may not be that negative. James & Wier (1987) and Dubofsky & Fraser (1989) concluded that during the 1970s, acquiring companies gained positive abnormal returns. A comparative study of Andrade et. Al (2001), finds that the returns are almost evenly split between negative and positive returns. The gains and losses were insignificant and do not differ among the different takeover waves.
Previous research mostly agrees upon the fact that the deal characteristics and firm characteristics are important in estimating the CAR for takeover announcements. According to Georgen & Renneboog (2004), equity-financed takeovers provide higher CAR than all-cash deals. Huang and Walkling (1989), Franks and Harris (1989) and Yook (2000) however conclude that cash bids generate higher CAR than equity bids. The general attitude towards the method of payment is that all-cash bids generate higher bidder- and target value. The market value is an important tool in estimating the CAR according to previous research. Moeller et al. (2004) found that large firms have higher Tobin’s q’s and lower Book-to-Market ratio’s than small firms. Dong et al. (2002) believe that these variables cause overvaluation and thus lower Cumulative Abnormal Returns
Because of the uncertainty around the Brexit. It is important to look whether the acquisitions in our sample are domestic or cross-border acquisitions. Prior research suggests on one hand that it does not make a difference whether the takeover is domestic or cross-border (Lowinski et al. (2004). On the other hand, research by Moeller & Schlingemann (2005) and Campa & Hernando (2004) concludes that domestic acquirors gain higher returns for their shareholders. Higher uncertainty and lower confidence might constrain investors’ enthusiasm regarding domestic acquisitions in the United Kingdom.
3. METHODOLOGY
This paragraph describes the data which has been collected for the research and how the research has been built up. To answer the main question: ‘’Does a takeover announcement cause abnormal returns
of acquiring firms in the United Kingdom before- and after the Brexit announcement and what is the effect of Brexit on these abnormal returns’’, an event study will be carried out accompanied by a
regression and a Chow test to see whether there is a structural break. The Brexit gap will be the date when the advisory referendum was held, which was the 23rd of June 2016.
3.1 DATA COLLECTION
The data for the research is split into two parts. Firstly, the data concerning the takeovers has is gathered. This M&A data was retrieved from Zephyr Database. 835 takeover deals were gathered from the database. These deals were made in the last 5 years, had a minimum deal value of 1 million euro’s, were listed on a stock exchange and had an acquiring company based in the United Kingdom. Furthermore, the initial stake in the target company was <50%. The data was adjusted for missing variables, deals that had missing variables were removed from the sample. Deals that were missing data such as ISIN numbers, announcement dates and stock prices were removed from the dataset. The final dataset consisted of 819 takeover deals. Secondly, the stock data for carrying out the event study had to be generated. This data was gathered from DataStream database. The exact steps will be further highlighted in the Event Study Methodology.
3.2 EVENT STUDY METHODOLOGY
To investigate whether abnormal returns are present, an event study will be carried out. According to De Jong (2007), event studies are an important tool in finance and it is interesting to see what
happens from the day that takeover plans become public knowledge and prior to the event date. For that, the event date is the announcement date and we will look at different Cumulative Abnormal Returns (from now on: CAR) and test them for significance to see whether we can conclude abnormal returns exist.
The hypothesis of the event study comes from De Jong (2007), where H0 states that the
expected CAR is zero, and H1 means that the CAR is not equal to zero. When the 0 hypothesis is
rejected, we can conclude that abnormal returns exist. When H0 is not rejected, we can not conclude
The abnormal returns will be calculated using the Mean-adjusted returns model laid out by De Jong (2007). Firstly, the normal returns will be calculated from the [-90 , -30] time period, in which -90 and -30 are days prior to the event date. The normal return is defined as the expected return without conditioning on the event taking place (MacKinlay, 1997, p. 15). According to MacKinlay (1997), the mean-adjusted model assumes that the mean return of the security is constant over time. This will be
an assumption of my event study as well. The
Normal Returns will be calculated using the following the formula:
𝑁𝑁𝑁𝑁𝑖𝑖𝑖𝑖 = 1𝑇𝑇 ∑𝑇𝑇2𝑖𝑖=𝑇𝑇1𝑁𝑁𝑖𝑖𝑖𝑖 (1)
T= T2-T1+1 equals the number of time periods used to calculate the average return, which in our case
is 90 days prior to the event until 30 days prior to the event. Now that we have the normal or expected returns of the securities. The stock data around the event date will be generated. Table 1 describes the different CAR that will be calculated.
Table 1
CAR variables
CAR1010
Cumulative Abnormal Return 10 days prior- and past the eventdate
CAR55
Cumulative Abnormal Return 5 days prior- and past the eventdate
CAR33
Cumulative Abnormal Return 3 days prior- and past the eventdate
CAR11
Cumulative Abnormal Return 1 day prior- and past the eventdate
The stock prices from the days prior- and after the event are generated and then processed into daily returns using the formula: 𝑁𝑁𝑖𝑖𝑖𝑖 𝑃𝑃𝑖𝑖−𝑃𝑃𝑃𝑃𝑡𝑡−1𝑡𝑡−1
(2) Subsequently, the daily returns will be subtracted with the normal returns to generate the daily
Abnormal Returns: 𝐴𝐴𝑁𝑁𝑖𝑖𝑖𝑖 = 𝑁𝑁𝑖𝑖𝑖𝑖− 𝑁𝑁𝑁𝑁𝑖𝑖𝑖𝑖 (3)
The Cumulative Abnormal returns for the windows mentioned above in Table 1 are then calculated using:
The CARs will be tested against the hypothesis stated above with a t-test following from De Jong (2007).
The descriptive statistics for the different CAR are shown below in Table 2.
Variable
Mean
Std. Dev.
Min
Max
CAR1010
0.0154825 0.1085742 -0.60984 0.624778CAR55
0.0157119 0.0762939 -0.41896 0.473881CAR33
0.0166515 0.0645791 -0.30785 0.422669CAR11
0.0150862 0.0512917 -0.27633 0.4039643.3 REGRESSION METHODOLOGY
For further testing of the effect of Brexit, and to answer our research question, a regression is carried out, the regression goes along with an F-test to test the full model, and a Chow test to test whether we have a structural break concerning Brexit. We want to know whether the Brexit has a significant effect on the CAR of the securities. Investors might be more risk-averse in times of uncertainty about the Brexit package and the possibility of an outflow of companies. When investors are more risk-averse, their money will be put into saving accounts or bonds (Riley & Chow, 1992). The regression can explain whether our explanatory variable Brexit has an effect on the CAR. Brexit be accompanied by certain control variables to make the regression and thus the estimation of CAR as unbiased as possible. The variables are given blow in Table 3 and will be discussed later on.
Table 3
Regression variables Dependent
CAR1010 Cumulative Abnormal Return 10 days prior- and past the event date
CAR55 Cumulative Abnormal Return 5 days prior- and past the event date
CAR33 Cumulative Abnormal Return 3 days prior- and past the event date
CAR11 Cumulative Abnormal Return 1 day prior- and past the event date
Independent
Brexit Dummy variable, 1 if the announcement date is past
23/06/2016 and zero otherwise N=263
Domestic Dummy variable, 1 if the acquisition is domestic and zero if
the acquisition is cross-border N=440
MOP Dummy variable, 1 if the method of payment is Cash, and zero
otherwise N=368
LnDealValue The natural logarithm of the deal value in thousands of euro’s
LnMarketValue The natural logarithm of the market value at event date of the
acquiring company in thousands of euro’s
Brexit is chosen as the explanatory variable because we want to examine the effect of the Brexit
announcement on CAR. According to Conn et al. (2005), domestic acquisitions result in higher
announcement returns than border acquisitions, so we have to control for a domestic or a cross-border acquisition. Huang and Walkling (1989), Franks and Harris (1989) and Yook (2000), agree upon the fact that cash bids generate higher bidder and target returns than stock-for-stock acquisitions. According to Gregory (1998) and Loughran & Vijh (1997), hostile acquisitions generate higher value for target- and bidder firms than friendly mergers. The variable hostile was left out of my regression due to a low number of hostile observations, namely 4. The Market Value of the acquiring company has been included because when a company is really large and acquires another company, that will relatively make a smaller impact on share price and thus returns than a smaller company acquiring the same company. Below you will find Table 4 in which descriptive statistics of the independent variables are shown.
Variable Mean
Std. Dev.
Min
Max
brexit
0.320732 0.4670427 0 1mop
0.448781 0.4976732 0 1domestic
0.536585 0.4989641 0 1lnvalueth
10.59173 1.798683 7.084888 17.7437lnmarketvalue
14.13404 1.349794 11.15768 18.442033.4 HYPOTHESES AND REGRESSION
The regression will be carried out using Ordinary Least Squares. The explanatory variable is Brexit, the dependent variable are the 4 CARs and Brexit will be accompanied by the control variables mentioned in section 3.4. The regression equations will look as follows:
CAR1010 = α + 𝛽𝛽1Brexit + 𝛽𝛽2Domestic + 𝛽𝛽3MOP + 𝛽𝛽4LnDealValue + 𝛽𝛽5LnMarketValue| CAR55 = α + 𝛽𝛽1Brexit + 𝛽𝛽2Domestic + 𝛽𝛽3MOP + 𝛽𝛽4LnDealValue + 𝛽𝛽5LnMarketValue CAR33 = α + 𝛽𝛽1Brexit + 𝛽𝛽2Domestic + 𝛽𝛽3MOP + 𝛽𝛽4LnDealValue + 𝛽𝛽5LnMarketValue CAR11 = α + 𝛽𝛽1Brexit + 𝛽𝛽2Domestic + 𝛽𝛽3MOP + 𝛽𝛽4LnDealValue + 𝛽𝛽5LnMarketValue
The independent variables will be regressed on different event windows to see whether shorter or longer event windows create different returns and are different in significance.
The hypothesis for the regression is given below. t-test: H0: 𝛽𝛽 1= 0 H1: 𝛽𝛽1 ≠ 0
When H0 is rejected, we can conclude that the variable Brexit is nonzero and thus has an impact on
the different CARs. If H0 is not rejected, there is not enough statistical evidence that Brexit has en
effect on the different CARs.
The model will also be tested using an F-test of overall significance to test the model against a model with no predictors and to see whether the model itself is significant. The hypothesis is given below: F-test: H0: 𝛽𝛽1= 𝛽𝛽2= 𝛽𝛽3= 𝛽𝛽4= 𝛽𝛽5 = 0 H1: 𝛽𝛽j ≠ 0 for at least one value of j
When H0 is rejected, we can conclude that the model provides a better fit and is thus a better
predictor than the intercept-only model. When H0 is not rejected, we can conclude that the model
Furthermore, a Chow test will be carried out for the event windows [-5,5] and [-3,3] – due to the
Brexit significance in these regressions - to see whether there is a structural break in CAR before and
after the Brexit. According to Chow (1960), a structural break can be tested by testing whether two sets of observations can be regarded as belonging to the same regression model. In the case of Brexit, a structural break seems plausible because the investor’s confidence has decreased and the pound suffered a major depreciation. The F-statistic will be calculated through the following formula:
𝑆𝑆𝑆𝑆𝑆𝑆𝑝𝑝−(𝑆𝑆𝑆𝑆𝑆𝑆1+𝑆𝑆𝑆𝑆𝑆𝑆2)/𝑘𝑘
(𝑆𝑆𝑆𝑆𝑆𝑆1+𝑆𝑆𝑆𝑆𝑆𝑆2)/(𝑛𝑛1+𝑛𝑛2−2𝑘𝑘) ~ 𝐹𝐹4,811
(5)
Since we want to teste whether the slope of the two lines regressions (pre- and post-Brexit), we want to see whether the slope coefficients are the same. The hypothesis for the Chow-test is as follows. H0: 𝛽𝛽1= 𝛽𝛽2 H1: 𝛽𝛽1≠𝛽𝛽2
4.1 EVENT STUDY ANALYSIS
The results of the event study have to be analyzed to see whether abnormal returns exist in the given time period. Four different CARs have been constructed for four different time windows as mentioned above. The results of these test have been processed and summarized in Table 5. Also, an economic intuition for the found results will be provided.
Table 5
Event Window t-value
p-value
Mean
CAR1010
4.05*** 0.000 0.0154825CAR55
5.86*** 0.000 0.0157119CAR33
7.34*** 0.000 0.0166515CAR11
8.40*** 0.000 0.0150862From the table, it follows that the mean of the CARs is around 0.0155, or 1,55%. Which means that the average CAR for the dataset of 818 observations lies around 1,55%. However, because the CARs are calculated for different time windows, the daily returns will vary more.
The mean is quite consistent with prior research, because about half of the sample in Goergen & Renneboog (2002) finds zero or small positive returns. The mean found in this research can be classified as small positive returns. Considering the significance of all t-values following from the 4 different t-tests, we can conclude that our null-hypothesis, which states that CAR are zero, is rejected. From our dataset of companies and stock returns, we can thus conclude that CAR do exist.
The fact that the t-values and thus the significance increases with a decrease in the event window, is quite a logical one because it corresponds to existing literature and has some economic intuition. When a takeover is about to take place, rumors will spread and rumors will become more present when the event date is getting closer. This is an explanation for the increasing significance of shorter event windows. When a takeover will take place, the stock price will fluctuate more heavily than normal. Since fluctuations are a profit possibility for investors, investors will start buying and selling the share which causes the increasing significance of the abnormal returns.
Furthermore, the means are compounded correspondingly to their event windows. When compounded daily, the mean CAR11 returns will be highest and this also supports our theory that abnormal returns around the event window [-1,1] will be the highest. From our event study, we can conclude that abnormal returns exist and the abnormal returns and the significance of them increase when the event date closes in.
4.2 REGRESSION ANALYSIS
To complement our event study, an additional regression is carried out to answer our main question. We want to see whether Brexit has a significant effect on the CAR. In this paragraph, the regression will be pointed out and analyzed to see which influence our variables have on the CAR. The analysis will be accompanied by economic intuition as an explanation for the found results. Below, the results of the regressions, specified in the methodology, are specified. CAR will now again be mentioned as Cumulative Abnormal Returns to prevent confusion considering the variables used.
Table 6
VARIABLES
car1010
car55
car33
car11
brexit
-0.0122
-0.0150*** -0.0107**
-6.88e-05
(0.00800)
(0.00558)
(0.00471)
(0.00374)
mop
0.00122
0.00501
0.000476
0.00402
(0.00762)
(0.00532)
(0.00450)
(0.00357)
domestic
0.000883
-1.06e-05
-0.00498
-0.000313
(0.00805)
(0.00563)
(0.00475)
(0.00377)
lnvalueth
0.00510**
0.00432**
0.00412*** 0.00400***
(0.00259)
(0.00181)
(0.00153)
(0.00121)
lnmarketvalue
-0.0109*** -0.00818*** -0.00892*** -0.00805***
(0.00356)
(0.00249)
(0.00210)
(0.00167)
Constant
0.119***
0.0885***
0.105***
0.0848***
(0.0442)
(0.0309)
(0.0261)
(0.0207)
Observations
817
819
819
819
R-squared
0.016
0.025
0.028
0.032
F-statistic
2,61**
4,14***
4,76***
5,45***
p-value of F
0,0237 0.0010 0,0003 0,0001
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Firstly, the explanatory variable and its significance will be analyzed. The variable Brexit is only significant in two of the four regressions carried out. So, in the CAR55 and CAR33 regressions, we can conclude that Brexit is not zero and thus has an effect on the Cumulative Abnormal Returns of the securities. The Brexit coefficient in all the regressions is negative, which means that the Brexit makes investors’ confidence worse. When we take CAR55, the Cumulative Abnormal Returns are 1,50% lower after Brexit than before the Brexit. The lower confidence in the overall economy after Brexit results in a more negative reaction of investors to a takeover announcement than before the Brexit. Investors might see takeover announcement from the acquiring companies as a last resort to maintain competitive in the international market, which of course is a bad signal. The non-significant effect in
CAR1010 might be explained due to the long event window. As more time goes on, more takeover
specifications and statements will be released to inform the investors about the arguments in favor of the takeover and that might be able to regain the investors’ confidence. In retrospect, CAR1010 might not be the best indicator for abnormal returns.
Furthermore, the control variables MOP and Domestic were not significant in any of the regressions. This is quite strange, because prior research has shown that these two variables were important in estimating the Cumulative Abnormal Returns. The pound depreciated firmly after the Brexit. This will make cross-border acquisitions more expensive and would intuitively have an impact on the reaction of investors. On the other hand, according to the previous literature of Moeller et al. (2004) and Campa & Hernando (2004), domestic acquirors gain better returns than cross-border takeovers. Their samples were not in crisis periods so since this one – partly – is, the effect might be neutralized and insignificant due to these contrary factors.
The control variables regarding size of the deal and market value are in every case significant, mostly under 1% significance. Considering the market value, the outcomes make sense. According to Moeller et al. (2004), large firms have higher Tobin’s q’s and lower Book-to-Market ratio’s than small firms. Dong et al. (2002) believe that these variables cause overvaluation and thus lower Cumulative Abnormal Returns. The control variable lnvalueth is also significant in each regression will be explained in my final version, I think I need the firm value of the target company to analyze that result.
Chow Test
To check whether a structural break in the pre- and post-Brexit area is present, a Chow test is carried out. The test statistic will be calculated following the formula 6. The Chow-test will be carried out for the event windows [-5,5] and [-3,3] because of the significance of Brexit in the regression in that event windows. The values that need to be implemented including the outcome of the Chow test are summarized in Table 7.
(6)
Table 7
Variable Description Event window Values
SSRp33 Pooled Sum of Squared Residuals [-3,3] 3,33352793
SSRp55 Pooled Sum of Squared Residuals [-5,5] 4,68425417
SSR133 Sum of Squared Residuals pre-brexit [-3,3] 2,37264738
SSR233 Sum of Squared Residuals post-brexit [-3,3] 1,08304902
SSR155 Sum of Squared Residuals pre-brexit [-5,5] 3,18002618
SSR255 Sum of Squared Residuals post-brexit [-5,5] 1,66831335
N1 Observations pre-brexit 549 N2 Observations post-brexit 270 CH55 F-value [-5,5] -6,86179393 CH33 F-value [-3,3] -7,16777588 Critical value= F(0.01,4, 811) Critical value [-3,3];[-5,5] 3,35
From the table, we can see that for the event windows [-5,5] and [-3,3] the F-statistic is significant. We can thus conclude that a structural break exists at the Brexit gap, the 23rd of June. This supports the
Table 6 also shows the F-statistics for the overall models.
H0: 𝛽𝛽1= 𝛽𝛽2= 𝛽𝛽3= 𝛽𝛽4= 𝛽𝛽5 = 0 H1: 𝛽𝛽j ≠ 0 for at least one value of j
In the last three models, the null-hypothesis can be rejected at a 1% significance level, for the CAR1010 model, at a 5% significance level. This means that our model provides a better fit and is a better predictor than the intercept-only model.
5. CONCLUSION
This research assesses the impact of the Brexit on the Cumulative Abnormal Returns and investigates whether abnormal returns are present around takeover announcements. In this study, an event study following the mean-adjusted model is used to generate the CAR to see whether abnormal returns exist. Afterwards, these CARs were tested for significance. The CARs of the 4 different time windows were all significant and the significance increased when the event window got smaller. The biggest significance was thus in the [-1,1] event window. From the event study, we can conclude that abnormal returns in the Brexit area exist. Subsequently, these CARs were further tested to see what the impact of the Brexit was. This was done using a regression where CAR was the dependent variable and Brexit the explanatory variable accompanied with some control variables regarding deal- and firm characteristics.
From the regression it follows that Brexit was significant in the event windows 3,3] and [-5,5]. Thus, the Brexit had a significant – negative – impact on the CAR of acquiring companies in the United Kingdom in the last 5 years. Furthermore, the control variables MOP and Domestic were not significant in any of the regressions. This might be due to the fact that a crisis neutralizes this effect which normally is significant. Lastly, a Chow test is carried out to test whether a structural break exists between before- and after the Brexit. The test statistic for the event windows [-3,3] and [-5,5] was significant and we can thus conclude that the CARs behaved differently before- and after the Brexit. To conclude, the Brexit has a significant effect on the CARs of the companies and abnormal returns exist in the Brexit area.
The effect of the Brexit is visible in the short-term which this study was focused on. To fully measure the impact of Brexit on the CAR and possible other factors we have to wait until the Brexit package negotiations come to an end and the long-term effects become clear. Even though the withdrawal process from the European Union was started, there still isn’t a binding exit agreement. In the long-term it will be visible what the exact consequences of the Brexit are and therefore that is an implication to my study. Moreover, my investigation was limited to takeover deals of UK-based companies whereas the effect of the Brexit is felt all over Europe, so the scope of my thesis can be easily expanded to formulate an even more convincing conclusion.
For further research, I recommend the motivation of investors’ reaction to takeover
announcements could be further investigated. As I mention in my results analysis, investors might see a takeover from acquiring companies as a last resort to maintain competitive in the international market. This is something which would be interesting to test. Also, the effect of hostile bids on the Abnormal Returns is an important variable in estimating the CARs, but due to a lack of observations it was not possible in my thesis so this effect can be taken into account for further research.
6. REFERENCE LIST
Andrade, G., Mitchell, M. and Stafford, E. (2001) New Evidence and Perspectives on Mergers. The Journal of Economic Perspectives, 15, 103-120.
Beltratti, A., & Paladino, G. (2013). Is M&A different during a crisis? Evidence from the European banking sector. Journal of Banking & Finance, 37(2), pp. 5394-5405.
Bjorvatn, K. (2004), Economic integration and the profitability of cross-border mergers and acquisitions. European Economic Review, 48, 1211-1226
Bloom, N. (2014). Fluctuations in Uncertainty. Journal of Economic Perspectives, 28(2), 153-76.
Campa, J. M., & Hernando, I. (2004). Shareholder Value Creation in European M&As. European
Financial Management,10(1), 47-81.
Chow, G.C. (1960). Tests of Equality Between Sets of Coefficients in Two Linear Regressions,
Econometrica, 28 (3), pp.591-605
Conn, R. L., Cosh, A., Guest, P. M., & Hughes, A. (2005). The Impact on UK Acquirers of Domestic, Cross-border, Public and Private Acquisitions. Journal of Business Finance Accounting, 32(6), 815-870.
De Jong, F. (2007). Event Studies Methodology. Retrieved from
https://www.coursehero.com/file/29461877/De-Jong-2007pdf/
Dhingra, S., Ottaviano, G., Sampson, t., & Van Reenen, J. (2016). The impact of Brexit on foreign investment in the UK. BREXIT 2016, 24
Dubofsky D. & Fraser D., (1989), Regulatory Change and the market for bank control. In: Gup BE (ed) Bank mergers: current issues and perspectives: Kluwer, Norwell, 21–139.
Ebell, M. & Warren, J. (2016). The Long-term economic impact of leaving the EU. United Kingdom: The National Institute of Economic and Social Research. Retrieved from https://www.niesr.ac.uk/sites/default/files/publications/National%20Institute%20Econ omic%20Review-2016-Ebell-121-38.pdf
Franks, J. R., & Harris, R. S. (1989). Shareholder wealth effects of corporate takeovers: The U.K. experience 1955-1985. Journal of Financial Economics, 23(2), 225-249.
Gaughan, P. A. (2002). Whats the Outlook for M&A in an Uncertain Market? Journal of Corporate Accounting & Finance, 13(5), 51-53.
Goergen, M., & Renneboog, L. (2004). Shareholder Wealth Effects of European Domestic and Cross-border Takeover Bids. European Financial Management, 10(1), 9-45.
Goergen, M., & Renneboog, L. D. R. (2002). Mergers and Acquisitions in Europe. (CentER Discussion Paper; Vol. 2002-6). Tilburg: Finance.
Gregory, A., (1997). An examination of the long run performance of UK acquiring firms. Journal of
Healy, P. M., Palepu, K. G., & Ruback, R. S. Which Takeovers Are Profitable? Strategic or Financial?
Sloan Management Review 38, 45-57.
Houston, J. F., & Ryngaert, M. D. (1994). The overall gains from large bank mergers. Journal of Banking & Finance, 18(6), 1155-1176.
Huang, Y., & Walkling, R. A. (1987). Target Abnormal Returns Associated with Acquisition Announcements: Payment, Acquisition Form, and Managerial Resistance. Journal of
Financial Economics, 19, 329-349.
Hudgins, S. C., & Seifert, B. (1996). Stockholder returns and international acquisitions of financial firms: An emphasis on banking. Journal of Financial Services Research,10(2), 163-180. James, M., & Wier, P. (1987). Returns to Acquirers and Competition in the Acquisition Market: The
Case of Banking. Journal of Political Economy, 95(2), 355-370.
Jensen, M. C., & Ruback, R. S. (1983). The Market for Corporate Control: The Scientific Evidence. Journal of Financial Economics, 11, pp. 5-50.
Loughran, T., & Vijh, A. M. (1997, December 01). Do Long-Term Shareholders Benefit From Corporate Acquisitions? Journal of Finance, 52(5), 1765-1790.
Lubatkin, M., Srinivasan, N. & Merchant, H. (1997) Merger strategies and Shareholder Value
During Times of Relaxed Antitrust Enforcement: The Case of Large Mergers During the 1980s.
Journal of Management, 23(1), 59-81
MacKinlay, A. C. (1997). Event Studies in Economics and Finance. Journal of Economic Literature, 35(1), 13-39.
Martynova, M., & Renneboog, L. D. R. (2008). A Century of Corporate Takeovers: What Have We Learned and Where Do We Stand? Journal of Banking and Finance; ECGI - Finance Working Paper No. 97/2005
Moeller, S.B
.,
Schlingemann, F.P. and Stulz, R.M. (2004). Wealth Destruction on a Massive Scale? A study of Acquiring-Firm Returns in the Recent Merger Wave. (NBER Working Paper no. 10200). Retrieved from: http://www.nber.org/papers/w10200Pilloff, S. J. (1996). Performance Changes and Shareholder Wealth Creation Associated with Mergers of Publicly Traded Banking Institutions. Journal of Money, Credit and Banking, 28(3), pp. 294-310.
Riley, W. B., & Chow, K. V. (1992). Asset Allocation and Individual Risk Aversion. Financial Analysts
Journal,48(6), 32-37.
Schoenberg, R. (2006). Measuring the Performance of Corporate Acquisitions: An Empirical Comparison of Alternative Metrics. British Journal of Management,17(4), 361-370.
The Chartered Institute of Procurement & Supply. (2017). The Brexit storm, how procurement and supply chain professionals are tackling the issues. London, United Kingdom. Retrieved from
Vincent, O.R., Bamiro, K. (2013). Fluctuations in Stock Market Prices: What went wrong, its Implications to Nigerian Economy? International Journal of Computer Applications, 63(19), 13-20
Yook, K. (2000) Larger return to cash acquisitions: signalling effect or leverage effect?. Working paper Johns Hopkins University.
