What effect did the BREXIT have on the abnormal stock
returns of REITs and how do different firm characteristics
affect this relationship
Bachelor thesis at the University of Amsterdam
Thesis supervisor: Mario Bersem
Bachelor student: Koen Spelde Student number: 10819606 BsC Economie & Bedrijfskunde Field: Finance & Economics
Statement of Originality This document is written by Koen Martijn Spelde who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used creating it. The Faculty of Economics and Business responsible solely for the supervision of completion of the work, not for the contents.
Contents:
1. Introduction 5 2. Literature overview 7 3. Methodology 11 3.1 Event study methodology 11 3.2 Regression analysis methodolgy 15 4. Data 18 5. Results 19 5.1 Event study results 19 5.2 Regression analysis results 23 6. Conclusion 24 7. Appendix 25 8. References 32Abstract: The Brexit outcome affected the stock market on almost every sector both domestically as internationally. In this paper the effect of the Brexit on the real estate sector of the UK is examined. This is done by conducting an event study to test how UK REITS performed during the Brexit. The model used is the market adjusted model and found a negative overall effect of cumulative abnormal returns for the REIT sector and negative significant cumulative abnormal returns for 35 of the 39 REITS. Thereafter a regression analysis is provided to test the relation between the cumulative abnormal return and the market capitalization of the REITS. Besides the market capitalization, the leverage ratio, book to market ratio and the dummy variable for the FTSE100 and FTSE250 are included as control variables. The regression analysis found a negative relationship between the CAR and the market capitalization. In increase in market capitalization resulted in a further decrease of the negative cumulative abnormal return.
1. Introduction: On June 24 in 2016, the United Kingdom voted to exit the European Union in a referendum. Of all participants, 51,9% voted in favor to leave the EU (The Guardian, 2016). This outcome was unexpected because the bookmakers expected a 90 percent chance that the UK would remain within the EU (Bloomberg, 2016). This outcome, the so-called Brexit, had many consequences on both domestic and international levels and caused many concerns on the political, social and economical level. Dhingra, Ottaviano, Samson & van Rheenen et al. (2016) found that in case the Brexit would happen, stock prices on the London stock Exchange would be negatively affected. The Brexit affected the returns in different sector types on both domestic and international stock markets. The day after the outcome the FTSE250 fell almost 12 percent in dollar terms. Moreover, an economic consequence is that 63% of UK export goods and services go to EU membership states. When the UK is no longer part of the EU, the UK will face trade tariffs because they cannot oblige to the EU trade agreement anymore. Dhingra et al. (2017) concluded that of all scenarios, the Brexit will reduce the welfare of the average UK citizen. In addition, a political consequence is that the position of EU residents who are living and working in the U.K. became uncertain. Monfared & Pavlov (2017) found prove that many EU residents left the London area due to the in 4 months’ time after the outcome of the referendum. Subsequently, real estate prices dropped between a range of 1,9% to 3%. Real estate and stocks are major components in valuing assets. Zeckhauser and Silverman (1983) found at least 25 percent of the value of corporate firms is associated with real estate. Gyourko and Keim (1992) published a paper and found prove that the real estate market is integrated in the stock market. Therefore, the stock market contains relevant and appropriate knowledge of the real estate market. Both stocks and real estate prices are subject to changes in the economy. Real estate and stocks are important investment vehicles of both individual and large institutional investors. A decrease in stock prices will result in a decrease in personal wealth, but research varies on how the Brexit influenced stock markets and different sectors in the UK. Most of the research was done for multiple sectors at once. Research
regarding the Brexit and the performance of real estate investment trusts (REITs) is only shortly mentioned in the paper of Ramiah, Pham & Moosa, (2016). Ramiah et al. (2016) researched the effect of the Brexit respecting different sectors of the UK. This was done calculating the abnormal returns of a 10-day event window regarding the Brexit. However, they found consistent results according to the paper of Dhingra et al. (2016). Moreover, market efficiency showed quick price adjustments of stock prices in 3 days after the outcome Oehler et al, (2017). Therefore, central in this thesis is to re-examine the abnormal returns of real estate stocks which formulate the following research question: What effect did the BREXIT have on the abnormal stock returns of REITs and how do different firm characteristics affect this relationship In this thesis, both the interest upon UK REITs and their performance after the Brexit outcome are considered. This is done by conducting an event study. The purpose of this thesis is to indicate abnormal returns for the REITs sector using a smaller event window than the research of Mackinlay, (1997). The aim of this thesis is to provide investors with insights, as they often consider real estate to diversify their portfolio. In addition, real estate is considered to be a relatively stable investment in someone’s overall portfolio. Therefore, it is interesting to consider what effect the Brexit had on these relatively stable stocks. The event study of this thesis calculates the abnormal returns, which will be based on the market-adjusted model. After this, the average t-test statistic is formulated to calculate the overall effect of het performance of UK REITS. Thereafter, a t-test is conducted for every REIT separately to establish their performance compared with the market return. In line with other research respecting abnormal returns and the Brexit, the results of the abnormal returns show a negative relation with the Brexit. After this event study was carried out, multiple regressions were performed. This was done with the use of STATA. Here, the relationship between the cumulative abnormal return and the market capitalization of the REITs was examined. In section 2, the literature overview will be described. The relation between the stock market and the Brexit will be discussed as well as what research was already
conducted with respect to the real estate market and the stock market. Thereafter, the first hypotheses will be formed. Moreover, the relation between stock returns and market capitalization will be described. After that, the firm characteristics which were chosen for the regression are explained. This is followed by the second hypothesis. In section 3, the event methodology is described. Then, the assumptions of a multiple regression are defined. In section 4, the collected data will be analyzed, which will be followed by some descriptive statistics. In the last section the results will be discussed, together with a conclusion based on these results. 2. Literature review
Brexit and stock markets
Research varies on how the Brexit affected the stock returns. Alkhatib & Harasheh, (2018) published an article on how global financial markets experienced this shock. They found significant positive abnormal returns from US T-bonds on the event date. Moreover, Ramiah et al. (2016) wrote a paper where they conducted research on what impact the Brexit referendum outcome had on different sectors of the British Economy. By measuring abnormal returns, an event study over the period June-July in 2016 was conducted and the results showed that the banking sector and leisure sector were affected the most. The short-term market risk was adjusted by the Bank of England, which resulted in negative abnormal returns and higher risk for this sector. Besides that, the Financial Times wrote that banks will leave the UK when the outcome results in leaving the EU, which would eventually lead to depreciation of the pound. In addition, for UK residents, holidays became more expensive which resulted in the negative abnormal return of the leisure sector. This is consistent with the research of Bonchev and Penchava (2017), who focused on the abnormal returns of bank stocks in Europe and found significant negative abnormal returns. They also concluded that bank size matters in the level of negative abnormal return. Oehler, Horn and Wendt (2016) researched short term price effects of different sectors, and analyzed how the FTSE100 reacted after the Brexit referendum. This was done by taking the abnormal returns of the first 3 days after the FTSE100 opened. The results (figure 1) show that large and medium cap-prices and returns were very volatile
ranged between 5798 – 6117 points. This research provided equally weighted mean values from the market capitalization and the standard deviation from the firms of the FTSE100. The statistical t – test of the abnormal and cumulative abnormal returns were found to have significant results at a 1 percent level. The abnormal returns of firms with a higher domestic operating degree were found to have a larger negatively affected return, in comparison to the companies that operate more internationally. Figure 1: The FTSE100 index from 23/6/2016 – 30/6/2016 Definition of REITs: A REIT is a company or corporation which use the combined investments of all types of investors to purchase real estate property, Before REITs were created this was only available for large and wealthy individuals. A REIT is a company that owns, operates or finances income-achieved activities. REITs can be classified if at least 75 percent of the REIT’s profits are related from rental properties. Secondly, REITs are required to pay out 90 percent of their income. In addition, it is not possible to have shareholders who have a larger stake of 10% equity shares. When this is obliged, the REITs are not subject to corporate tax. The payout is done by paying dividends to their shareholders and investors. It is not rare the dividend yield is between 5 and 15%. Relationship stocks and real estate: The public real estate market has developed into an asset class, which provides the possibility of extending the scope of commercial real estate without acquiring property (Ling and Naranjo, 2002). According to research, investors tend to invest in real estate
because they have a low correlation with other assets. REITs are originally seen as a mix between exposure and risk. Investors are likely to invest in REITs to diversify their risky portfolio. This implies that the REITs return and the volatility of their stocks are also related to the same macro-economic variables that affects bonds (Clayton & MacKinnon, 2003). The relationship and direction between the stock market and real estate stock is widely tested. Likewise, there is research done in previous papers about the causality for trends in stock markets and real estate. Okunev, Wilson & Zurbruegg, (2000) performed their research by conducting both linear and nonlinear causality tests of the relationship between trends in the stock market and real estate. They conduct their research of using REITs and the S&P500. They found a relationship between their observations and data, and between stock markets and real estate prices between 1972 and 1998. Therefore, there is a linear relationship between the two assets but there is a lag of time before equity REITs return are affected by the S&P return. Moreover, research done by Gyourko and Keim (1992) found evidence that the stock market contains relevant and appropriate knowledge of the real estate market. They regressed US equity REITs against the S&P500 and found results that the S&P returns have predictionary power in predicting Equity REIT returns. This supports the research of Andersson (2014), which is about the causality between the stock market and the housing market. Previous research provided from Oehler et al, (2016),Ramiah et al. (2016) and Bonchev et al. (2016), caused the researcher of the current paper to expect that the cumulative abnormal returns of the REIT sector are negatively affected through the Brexit. Therefore, the first hypothesis is proposed: 𝐻!: 𝑡ℎ𝑒 𝑎𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑟𝑒𝑡𝑢𝑟𝑛𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑖𝑡𝑠 𝑎𝑟𝑒 𝑛𝑜𝑡 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒𝑙𝑦 𝑎𝑓𝑓𝑒𝑐𝑡𝑒𝑑 𝑡ℎ𝑟𝑜𝑢𝑔ℎ 𝑡ℎ𝑒 𝐵𝑟𝑒𝑥𝑖𝑡 𝐻": 𝑡ℎ𝑒 𝑎𝑏𝑛𝑜𝑟𝑚𝑎𝑙 𝑟𝑒𝑡𝑢𝑟𝑛𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑖𝑡𝑠 𝑎𝑟𝑒 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑙𝑒𝑦 𝑎𝑓𝑓𝑒𝑐𝑡𝑒𝑑 𝑡ℎ𝑟𝑜𝑢𝑔ℎ 𝑡ℎ𝑒 𝐵𝑟𝑒𝑥𝑖𝑡. In order to find what could influence the cumulative abnormal returns of the REITs sector, three firm characteristics are explained which have a relationship with the stock market. These firm characteristics are market capitalization, leverage ratio and book to market ratio.
Firm characteristics: In this paper, it is important to examine the relationship between stock returns and market capitalization. Market capitalization is measured by number of shares outstanding times the market price. Previous research showed a strong relationship of stock returns and market capitalization. For instance, Van Dijk (2011) aggregated previous research conducted on this relationship. Van Dijk combined the research of Banz (1981), Keim (1983) Lamoureux and Sanger (1989) and concluded that the papers had in common that a correlation exists between these two variables. In addition, Oehler et al. (2016) stated that abnormal returns of stocks which have a larger domestic degree are affected more negatively through the Brexit. The research of Bonchev and Pencheva, (2017) found that higher market value of the banks resulted in a larger negatively cumulated abnormal return regarding the Brexit. Therefore, it is expected that the relation between these two variables negatively affect each other. Moreover, it is expected that a relatively high market capitalization will increase the negatively cumulative abnormal return even further. Secondly, Leverage ratio is determined by total debt divided by total assets. This ratio is a measure of the financial risk a company has. Opler and Titman (1994) provided evidence on how financial leverage affects corporate performance. Their research found prove that highly leveraged firms will perform worse in an economic downturn. This could be explained by the fact that highly leveraged firms will often have a greater difficulty to pay interest payments. Therefore, it is expected that this variable will show a negative relation with the cumulative abnormal returns. Lastly, book to market ratio is mentioned in this paper. According to Fama and French (1992), stock returns are positively related to the book-to-market ratio. A book-to-market ratio higher than 1 implies that the NPV is higher than the value investors pay. This implies that a high book-to-market ratio results in a higher stock return. Moreover, the coefficient of this variable is expected to be positive for explaining the CARs. As previous mentioned, important drivers of returns are market capitalization, leverage ratio and book-to-market ratio. After controlling for the last two variables, it is expected
that market capitalization negatively affects the cumulative abnormal return. Therefore, the second hypothesis is stated as follows: 𝐻!: 𝑡ℎ𝑒 𝑚𝑎𝑟𝑘𝑒𝑡 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 ℎ𝑎𝑠 𝑛𝑜 𝑒𝑓𝑓𝑒𝑐𝑡 𝑖𝑛 𝑒𝑥𝑝𝑙𝑎𝑖𝑛𝑖𝑛𝑔 𝑡ℎ𝑒 𝐶𝐴𝑅𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑈𝐾 𝑅𝐸𝐼𝑇𝑠 𝐻": 𝑡ℎ𝑒 𝑚𝑎𝑟𝑘𝑒𝑡 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 ℎ𝑎𝑠 𝑎 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑒𝑓𝑓𝑒𝑐𝑡 𝑖𝑛 𝑒𝑥𝑝𝑙𝑎𝑖𝑛𝑖𝑛𝑔 𝑡ℎ𝑒 𝐶𝐴𝑅𝑠 𝑈𝐾 𝑅𝐸𝐼𝑇𝑠 3. Methodology: In this section, the methodology of an event study is described and discussed. Thereafter, the regression analysis is described as well as the assumptions of an OLS-regression. 3.1 Event study methodology: To analyze the reaction on the core real estate market of the Brexit, an event study is conducted. An event study measures the impact of a specific event on the value of the firm (Mackinlay, 1997). The event study of this paper is based on a one-day-event. Therefore, it a relatively smaller event window when compared to the event study literature from Mackinlay, (1997). However, the paper by Ederington, Guan and Yang (2015) show examples of event studies which are similar to the event study of this study. Conforming to the literature of event studies (Mackinlay, 1997), an estimation window is important to define. The paper of Mackinlay states that the average estimation window lies between 210 and 120 days. This is the period before the event without considering the event date. In addition, literature from Brown and Warner, (1985) conducted event studies using an estimation window of 240 days. Literature has shown (e.g. Mackinlay, 1997, Ahern, 2008) that the sampling error fades away by using a large estimation window. Different papers have shown that the chosen estimation window would be aligned between 90 days and 240 days prior the event. The estimation window is used to determine the benchmark of the index behavior. Therefore, in this paper the estimation window will be 180 days.
Figure 1: source De_Jong_2007 In this thesis, the return of the UK REITs listed on the London Stock exchange will be analyzed. This is done by using financial market data, which is the best way to explain event studies because the event influences the stock price almost immediately. The result of the Brexit vote came after the close of the British stock market. Therefore, the impact of the event was observable after the day after the referendum. Hence, the date of the event will be considered as June 24, 2016. Conforming to the literature of Ederington et al. (2015), the event window of this paper will be 2 days prior the Brexit and 2 days after the Brexit. This is often used for a small window, because it reduces the type 2 error. Evidence has shown a bias exists when aggregating abnormal returns. The event window is the time prior to the event and at least 1 day after the ‘event date’, to capture the changes in stock returns caused by the event (Ederington et al., 2015). In addition, the estimation window of 180 days is quite long. Therefore, to add more robustness for the estimation window, a shorter estimation window was chosen. Hence, the other estimation window is 90 days instead of 180 days, which is consistent with the literature of Mackinlay (1997). In the appendix there is a table which show the abnormal return calculated with the estimation window of 90 days. Models: Mean-adjusted model: A variety of models have been developed that can be used to calculate abnormal returns of event studies. The mean-adjusted model calculates the expected return of the average of the values T1 and T2. Where T = T2 – T1 + 1 equals the number of daily datapoints chosen to calculate the normal return. This model uses the normal returns as
benchmark from the chosen estimation window. This is obtained by the following formula: 𝑁𝑅#$ = 1 𝑇H 𝑅#$ %& %" A disadvantage of the mean-adjusted model is the average of the chosen benchmark given a certain estimation window. Following different papers this period is rather arbitrary. The mean adjusted model ignores wide stock price fluctuations related to benchmark period returns. This is especially the case with the same type of companies where the shares of those companies behave the same when the whole market goes up or down during the event period. This results in biased abnormal returns, which is not caused by the event but rather by price fluctuations between markets. To correct for this, the market-adjusted model is used to calculate the expected return of the different REITs Market-adjusted model: According to an event study, the daily return of the securities stock and the market index return will be matched separately for each day of the estimation window. This is done by running a single index model regression to obtain each security’s alpha and beta. Alpha is the intercept term and beta the slope of the regression of the firm return against the corresponding market return. This is obtained with the following formula: 𝐸(𝑅#$) = 𝛼# + 𝛽#𝑅'$( + 𝜀 #$ Where: 𝐸(𝑅#$) = the expected return on per stock’s observation on day 𝑡. 𝑅'$ ( = the actual return from the chosen market index for the estimation window on day 𝑡. This includes daily returns. 𝜀#$= is the mean disturbance term with an expected value of zero and a variance of 𝜎*$&. 𝜀#$ = uncorrelated with the market return and the firm return. 𝛼# & 𝛽# are parameters of the market model.
Central to an event study is the measurement of abnormal returns. An abnormal return is the actual ex post return minus the expected return retrieved with the market adjusted model over the event window (Mackinlay, 1997). Abnormal returns determine the difference between the security performance in comparison with the benchmark return. This is obtained with the following formula: 𝐴𝑅#$ = 𝑅#$ − 𝐸(𝑅#$) Here, 𝑅#$ is the return of the stock of observation 𝑖 on day 𝑡. Cumulative abnormal return: The accuracy of an event study is less when there is only one observation used. So, the abnormal returns around the event window must be aggregated which helps to make a good interpretation of the effect on stock market returns around the event. As such, the cumulative abnormal return is the sum of the individual abnormal returns for the chosen event window. The CAR is the short-term effect on the returns of investors who invested in the assets class which is central in this study. The CAR of the different securities will be interpreted to evaluate the impact of the Brexit. Moreover, the cumulative averaged abnormal return (CAAR) will be interpreted. This will be examined with the following formula: 𝐶𝐴𝐴𝑅 =+"∑+ 𝐶𝐴𝑅# #," 𝐶𝐴𝑅($",$&) = ∑$&$ , $"𝐴𝑅#$ Here, N is the number of events. N equals 1, because the Brexit is a single event. The CAR of each firm will be calculated over the event window (-2,2). The second formula is the CAAR this is the overall performance of all the REITs together. It is important to examine significant results of the return of each security after the event date. By doing so, it might indicate whether the abnormal return of each company differ from the benchmark return. To test whether the calculated results are significant, the CAR t-test will be done. Here, the following formulas will be used: 𝑇0112, √+!""# $!""# 𝑇012 = 012(&',&)) 45(6(+',+)) = 012(&',&)) 7(+',+))
Where: 𝜎 is the standard deviation of the estimation window. 𝑠 is the standard deviation of the average of the CARs The first hypothesis as defined in the literature overview is to test whether the Brexit had a negative effect on the cumulative abnormal returns. If the current researcher’s expectations are true, it would be in line with existing literature. H!: the difference between the CAR and the Benchmark return = 0 H": the difference between the CAR and the Benchmark return < 0 3.2 Regression analysis methodology: Followed from the literature review, the objective of the regression analysis is to examine the relationship between the CARs and the market capitalization. To test this the following model is used: 𝐶𝐴𝑅8&,&= 𝛽!+ 𝛽"∗ 𝑀𝑎𝑟𝑘𝑒𝑡𝑐𝑎𝑝(𝑚) + 𝛽& ∗ 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑟𝑎𝑡𝑖𝑜 + 𝛽9∗ 𝐵𝑜𝑜𝑘𝑚𝑎𝑟𝑘𝑒𝑡𝑅𝑎𝑡𝑖𝑜 + 𝛽: ∗ 𝐹𝑇𝑆𝐸100_250 + 𝑒 The second hypothesis as stated in the literature overview will conduct to test this relationship: the expectation of the coefficient of the market is expected to be negative, H!: There is a relationship between the CARs and the market capitalization H": There is a negative relationship between the CARs and the market capitalization To examine the relation between the dependent and the independent variables, the following is expected. As stated in the literature overview the expectation of the coefficient β" will show a negative sign which indicates a negative relationship between
In addition, with the direction of the coefficient leverage ratio, it is expected that the REITs that have a higher leverage ratio will negatively affect the negative cumulative abnormal return. Moreover, the coefficient is expected to be positive, so the higher the book market ratio the more positive the cumulative abnormal return. The dummy variable is included because the different REITs retrieved for this sample are along different indexes. The dummy variable will show the variable 1 if the REIT is along the FTSE 100 or FTSE250 and the value 0 if not. The coefficient is expected to be negative, which is consistent with the idea of market capitalization. REITs with a bigger market cap will be more affected by a shock relative to small cap stocks. Because of the small sample, no other variables were included to the multiple regression. Adding more variables does not mean the regression will offer better predictions. Due to the 39 observations, outliers are likely to occur. Therefore, before running the regression, the highest and lowest value are replaced by the next value aggregating inwards the extremes. This increased the R-squared and thus the validity of the multiple regression model. Secondly, a regression must hold the OLS-assumption to run a valid multiple regression. Therefore, a test of linearity, normality, and multicollinearity were conducted. The multiple regression was carried out with robust standard errors. This corrects for heteroskedasticity. OLS assumptions: In order to have unbiased estimates, the data must satisfy a number of assumptions: Linearity The linearity assumption requires that the dependent variable has a linear relation with the independent variables and the residuals. Before doing an OLS-regressions, one should test for normality, heteroskedasticity and multicollinearity, otherwise the results might be biased and the BLUE conditions are not satisfied. Test for normality To test whether the residuals are normally distributed, the Skewness/kurtosis test were conducted. The residuals should be normally distributed otherwise the predicted values are inaccurate. As such, the hypothesis is stated as follows:
H!: The residuals are normally distributed
H": the residuals are not normally distributed
Skewness/Kurtosis tests for Normality --- joint ---
Variable Obs Pr(Skewness
) Pr(Kurtosis) adj_chi2(2) Prob>chi2
Residuals 39 0.402 0.904 0.750 0.688 Table(1) For the number of observations, the p-value of skewness is greater than >0,05 for the event window, which implies that skewness is asymptotically normally distributed. Similarly, Pr (kurtosis >0,05) is asymptotically normally distributed. Lastly, for both values, the chi2 is greater than 0,05 which implies its significance a at a 5 percent level. Therefore, we can reject the null hypothesis. Following, the residuals show a normal distribution according to the skewness test for normality. In the appendix a graph is illustrated to show the residuals normality. Test for homoskedasticity Homoskedasticity examines if the error of the regression has the same variance. In case of heteroskedasticity the regression model is more likely to give wrong estimates. To test on heteroskedasticity the Breusch-Pagan test is conducted. H!: the error terms have the same variance (homoscedastic) H!: the error terms have not the same variance (heteroskedastic)
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: Constant variance
Variables: fitted values of rsq chi2(1) = 1.11 Prob > chi2 = 0.2922 table(2) Since the p-value is larger than 0,05 the null hypothesis cannot be rejected. Therefore, the residuals have the same variance and are homoscedastic. Test of multicollinearity
Generally, multicollinearity occurs when two or more independent are highly correlated. This means that the explanatory power of the independent variables affects the other independent variables. Therefore, it might become difficult to detach the relationship between the independent variable and the dependent variable. The outcome of multicollinearity is given in the table below:
Variance inflation factor
VIF 1/VIF Bookmarket 1.21 .827 Marketcapm 1.171 .854 FTSE100 250 1.111 .9 Leverageratio 1.015 .985 Mean VIF 1.127 . Table (3) When VIF is higher than 10 there will be multicollinearity between the estimates. Because this is not the case the multicollinearity assumption holds. 4. Data In this part of the research, the data used to conduct an event study is discussed and explained. Thereafter, the descriptive statistics of the regression variables are shown. For an event study, it is important to determine the market index, which will serve as the benchmark of the normal return. The market index is the relevant market index of the companies. According to Mackinlay, (1997) the market index should be a large stock-based index. Therefore, in this paper, the market index used is the FTSE All Share. This index represents the 900 biggest companies in the UK, including the daily prices of the 39 REITs. The REITs included are from different indices along the FTSE according to their market capitalization. The daily data of the REITs stock returns are retrieved from Yahoo Finance. After calculating the cumulative abnormal return of each REIT, a regression analysis is provided to explain these different CARs. Moreover, the dependent variable will be the CARs of the event window (-2,2). There will be 4 independent variables including 1 dummy variable. There are 3 variables retrieved from Compustat, namely the market capitalization, the leverage ratio and the book-to-market ratio.
The market capitalization is retrieved of each different REIT. This is done by determining the price of the stock times the shares outstanding at the end of June 24, 2016. Secondly, the leverage ratio on June 24, 2016 was calculated by taking the long-term debt plus the debt in current liabilities divided by stockholder’s equity. REITs use a lot of borrowed capital to finance their investments. Lastly, the Book to market ratio is included in the regression. The book to market ratio is retrieved by common shareholders equity divided by the market cap this was also retrieved for the same date. The dummy variable will show the value 1 if the REIT is along the FTSE100 or FTSE250 and the value 0 when the REIT is from the FTSE All Share. This dummy variable will show what effect the REITs have when it is along the FTSE100 and FTSE250 on the cumulative abnormal return. Table (4) present the descriptive statistics of the variables used for the regression. This includes the mean, standard deviation, minimum and maximum. Descriptive Statistics
Variable Obs Mean Std.Dev. Min Max
CAR22 39 -.088 .068 -.252 .009 Marketcap(in million) 39 1315.897 1720.023 108 7950 Leverageratio 39 .573 .36 .099 1.676 Bookmarket 39 .905 .195 .561 1.46 FTSE100_250 39 .385 .493 0 1 Table (4) 5. Results: In this section the results of the event study will be discussed and illustrated. Thereafter, the regression analysis will be interpreted as well as the effect of the cumulative abnormal returns. 5.1 Event study results In this thesis, the event study was from 2 days prior the event to 2 days after the event. The results of all the 39 REITs stocks returns are shown in the following graph:
Figure (2) Abnormal return calculated with estimation window (-191, -11 prior the event date) When interpreting the graph, there is a sharp drop for almost all of the REITs returns form after the event took place. This already illustrates that the Brexit event could have an influence on the returns of the securities. The result of the CAAR is given below to explain if there is a significant effect of the Brexit event of the average CAR of the REIT sector. T-test T-statistic CAAR -8.1647*** When referring to the first hypothesis, the t-statistic shows a negative significant result. Therefore, the null hypothesis is rejected at a 1 percent significance level. This implies that the Brexit had a significant negative cumulative return for the REIT sector in the UK. The abnormal return as well as the cumulative abnormal return of each REIT for the event window and the outcome of the t-test are provided in the table (5). Day -2 -1 0 1 2
Company name Index AR AR AR AR AR
1 AEW UK Reit plc FTSE-Allshare 0,52% 1,14% -2,65% -0,55% -0,03% -1,57% -4,97*** 2 Assura plc FTSE 250 0,46 % 2,61% -10,17% -4,93% 4,26% -7,78% -4,21*** 3 Tritax Big Box REIT plc FTSE 250 -3,00 % 0,74% -4,54% -7,93% 6,89% -7,84% -10,70*** 4 British Land Company Plc FTSE 100 0,23 % 0,10% -15,96% -7,00% 2,49% -20,14% -18,94*** 5 BMO Real estate
Invest. Lim. FTSE-AllShare -0,05 % 1,18% -2,06% -9,10% 4,78% -5,24% -7,36*** 6 Big Yellow Group PLC FTSE 250 -0,22 % -0,23% -8,14% -4,05% 1,10% -11,55% -7,73*** 7 Capital & regional PLC FTSE All Share 1,58% -0,60% -9,80% -5,54% -2,82% -17,19% -14,05*** 8 Capital & counties Properties PLC FTSE 250 -0,25 % 0,73% -13,69% -8,50% 3,46% -18,25% -13,82*** 9 Custodian Reit PLC FTSE All Share -0,02 % -0,52% -1,03% -0,59% -1,55% -3,70% -4,33*** 10 GCP Student Living plc FTSE 250 0,19% 0,26% -1,65% -0,08% -0,07% -1,35% -3,84*** 11 Derwent London Plc FTSE 250 2,03% -0,03% -21,34% -9,95% 4,10% -25,20% -25,42*** 12 Ediston Property Investment Company plc FTSE All Share -0,04 % 0,13% -4,19% -1,31% -0,46% -5,85% -10,42*** 13 Empiric Student
Property plc FTSE All Share 1,08% 2,33% -3,52% -6,04% 6,30% 0,16% 2,09** 14 Great Portland
Estates Plc FTSE 250 -0,17 %
1,13% -19,43% -8,19% 5,96% -20,71% -20,73*** 15 Ground Rents
Income Fund PLC FTSE All Share -0,04 % 0,16% -0,84% -0,32% -0,38% -1,43% -5,05*** 16 Hibernia REIT Plc FTSE All Share 0,23 % 1,29% -1,65% 0,64% -0,27% 0,24% -0,84 17 Hammerson FTSE 250 -0,09 % 0,36% -9,73% -5,78% 3,50% -11,74% -12,26*** 18 Intu properties plc FTSE All Share 0,65% 0,50% -8,88% -4,45% 1,38% -10,79% -10,74*** 19 Land Securities Group plc FTSE 100 -0,19 % -0,16% -12,00% -6,37% 6,19% -12,53% -10,83*** 20 LondonMetric Property Plc FTSE 250 0,73% 0,74% -6,87% -8,71% 4,88% -9,23% -11,03*** 21 McKay Securities Plc FTSE All Share -0,07 % 6,29% -9,81% -18,08% 2,97% -18,70% -10,23*** 22 NewRiver REIT plc FTSE All Share -0,71 % 0,66% -4,55% -4,97% -0,25% -9,82% -10,07*** 23 Picton Property
Income Limited FTSE All Share 0,97% 0,81% -5,73% -11,61% 10,33% -5,23% -5,96*** 24 Primary Health Properties Plc FTSE 250 0,04 % 1,69% -1,37% -1,60% 0,06% -1,18% -1,06 25 RDI REIT P.L.C. FTSE All Share -0,23 % -0,44% -2,19% -6,11% 1,80% -7,17% -5,58*** 26 Regional REIT
Limited FTSE All Share 0,14% 0,04% -2,29% -5,48% 0,88% -6,71% -5,54*** 27 Real Estate credit
investments FTSE All Share 0,01% -0,01% -4,69% -3,26% -0,92% -8,86% -11,22*** 28 Safestore Holdings plc FTSE 250 1,63 % 0,26% -12,44% -6,94% 0,87% -16,62% -12,90*** 29 Schroder
European Reit Plc FTSE All Share 0,00% -0,05% -3,62% -3,69% 1,92% -5,43% -1,75*
30 Segro Plc FTSE 100
-2,00 %
0,05% -8,87% -4,22% 5,17% -9,86% -11,01*** 31 Shaftesbury PLC FTSE 250 0,43 0,18% -4,77% -5,54% 1,30% -8,39% -9,73***
Table (5) the results of the event study. *** significance at 1%, *significance at 10% The last column has a significant outcome of 35 of the 39 different CARs of the REITs. This outcome means that, compared with the market return, the 35 REIT returns perform significantly worse than the market. This shows that there were abnormal returns during the Brexit event. Moreover, 4 of the REIT stocks show no significant result. This means that the difference between the market and the stock returns does not deviate significantly, and these stocks were not performing worse than the market. In the appendix the tables 10 and 11 of the cumulative abnormal return of a shorter estimation window. According to the literature of Mackinlay (1997), Brown and Warner, (1985), the estimation window should be aligned between 90 and 240 days. The results in the appendix show that the estimation window of 90 instead of 180 days are not the same (Table (11). Nevertheless, this gives no other result when computing the significance of the overall CAAR. Therefore, the same can be interpreted either by using 90 or 180 days as the estimation window. A possible explanation for the small difference in cumulative abnormal return is that the beta is not the same for the 2 estimation windows, which is due to fewer observations in the shorter estimation window. Due to this the variance and thus the standard deviation can be larger. This resulted in another beta compared with the longer estimation window. This is consistent with the literature of Mackinlay, (1997). 5.2 Regression results: 32 Urban Logistics
REIT plc FTSE All Share 0,31% 0,52% -0,84% -0,71% 0,91% 0,19% 0,08 33 Secure Income REIT Plc FTSE All Share -0,67 % -0,58% -2,68% -4,87% 0,19% -8,61% -6,77*** 34 Schroder Reit Limited FTSE All Share -0,02 % -0,01% -0,06% -0,05% 0,00% -0,14% -0,59 35 Target Healthcare REIT Limited FTSE All Share 0,92 % 0,41% -1,17% 0,18% 0,58% 0,93% 1,07 36 Town Centre
Securities PLC FTSE All Share -0,04 % -1,41% -5,73% -4,73% 2,78% -9,14% -5,29*** 37 UK Commercial Property REIT Limited FTSE 250 0,28 % 2,00% -5,79% -11,04% 6,97% -7,57% -8,97*** 38 The Unite Group plc FTSE 250 -2,16 % 0,28% -4,16% -5,51% 1,14% -10,42% -9,35*** 39 Workspace Group plc FTSE 250 0,97% 2,86% -13,21% -12,79% 2,34% -19,82% -14,47***
Multiple linear regression: Since all assumptions are satisfied, the multiple linear regression can be conducted. A linear regression estimates whether the independent variables explains the dependent variable. The focus of the regression is to answer the hypothesis and prove if there is a relationship between the cumulative abnormal returns and the market capitalization. Linear regression
CAR22_w Coef. St.Err. t-value p-value [95% Conf Interval] Sig
Marketcapm 0.000 0.000 -2.36 0.024 0.000 0.000 **
Leverageratio -0.041 0.028 -1.46 0.153 -0.097 0.016
Bookmarketratio 0.080 0.055 1.44 0.158 -0.033 0.193
FTSE100_250 -0.049 0.019 -2.63 0.013 -0.087 -0.011 **
Constant -0.101 0.063 -1.61 0.116 -0.228 0.026
Mean dependent var -0.087 SD dependent var 0.064
R-squared 0.375 Number of obs 39.000
F-test 6.820 Prob > F 0.000
Akaike crit. (AIC) -113.482 Bayesian crit. (BIC) -105.164
*** p<0.01, ** p<0.05, * p<0.1 Table (6) At first, the market capitalization and its significance will be analyzed. the market capitalization is significant at a level of 5 percent. Furthermore, the t-statistic is negative, which means that when the market cap increases, this negatively affects the cumulative abnormal returns. This is consistent with the expectation of this coefficient. The second significant coefficient is the dummy variable. When the REIT is along the FTSE100 or FTSE250 the results show that the CAR will be negatively larger. This is consistent with the paper of Oehler et al. (2016). Stocks which dominate the stock market are relatively more affected by shocks on the stock market. Furthermore, the control variable leverage ratio shows a negative relation with the CAR but is not significant. This was expected in the literature overview, but when it is not significant there is no proof of this. The second control variable book-to-market ratio show a positive relation so the higher the book market ratio the better the abnormal return. This was expected but not with no significant level we cannot prove it. 6. Conclusion
The purpose of this thesis was to conduct an event study and interpret this for a specific sector of the London Stock Exchange. The main goal of this thesis is to answer the research question: “what effect did the BREXIT have on the abnormal stock returns of REITs and how do different firm characteristics affect this relationship?” Previous research from Ramiah et al. (2016) found significant results of multiple abnormal return in different sectors in the regarding the Brexit event. This is consistent with the research of Oehler et al. (2016). Research about one specific sector was not provided yet. Therefore, the real estate sector was taken into account. Research proved that the real estate market is integrated in the stock market (Gyourko & Keim, 1992). Hence, there stocks show the same macro-economic volatility as bond stocks Clayton & Mackinnon, (2003). In this research an event study is conducted to measure the abnormal returns of the REIT’s sector of the UK regarding the Brexit. This is established with the market-adjusted model and the FTSE All Share as benchmark return. The event study showed an overall significant negative cumulative abnormal return. This is consistent with the first hypothesis. After testing each REIT separately, the outcome establishes that 4 REITs which did not have a significant CAR. These firms did perform worse than their normal performance relative to the market. Thereafter a multiple regression analysis was conducted to test the relationship between the cumulative abnormal returns and the market capitalization of the REITs. The results show a negative significant effect at a 5 percent significance level that an increase in market capitalization would decrease the CAR further. This relationship is consistent with the second hypothesis. For further research, this thesis is not really representative. The biggest limitation is the small number of REITs observations. This thesis focused on only one specific sector in on specific country selecting one shock. Moreover, is it difficult to compare this to other countries. Therefore, the reliability is not be considered high, and it would be hard to generalize the results. In addition, not all firm characteristics were significant. Nevertheless, there are more variables which can explain the stock market returns but, in this thesis, only 3 variables where provided in the regression. In future research it would useful to test the same relationship but across different countries.
7. Appendix: Abnormal returns & cumulative abnormal return of the event window (-2,2) provided with the T-test: Day -2 -1 0 1 2
Company name Index AR AR AR AR AR
CAR (-2,2) CAR T-Test 1 AEW UK Reit plc FTSE-Allshare 0,52% 1,14% -2,65% -0,55% -0,03% -1,57% -4,97*** 2 Assura plc FTSE 250 0,46 % 2,61% -10,17% -4,93% 4,26% -7,78% -4,21*** 3 Tritax Big Box REIT plc FTSE 250 -3,00 % 0,74% -4,54% -7,93% 6,89% -7,84% -10,70*** 4 British Land Company Plc FTSE 100 0,23 % 0,10% -15,96% -7,00% 2,49% -20,14% -18,94*** 5 BMO Real estate
Invest. Lim. FTSE-AllShare -0,05 % 1,18% -2,06% -9,10% 4,78% -5,24% -7,36*** 6 Big Yellow Group PLC FTSE 250 -0,22 % -0,23% -8,14% -4,05% 1,10% -11,55% -7,73*** 7 Capital & regional PLC FTSE All Share 1,58% -0,60% -9,80% -5,54% -2,82% -17,19% -14,05*** 8 Capital & counties Properties PLC FTSE 250 -0,25 % 0,73% -13,69% -8,50% 3,46% -18,25% -13,82*** 9 Custodian Reit PLC FTSE All Share -0,02 % -0,52% -1,03% -0,59% -1,55% -3,70% -4,33*** 10 GCP Student Living plc FTSE 250 0,19 % 0,26% -1,65% -0,08% -0,07% -1,35% -3,84*** 11 Derwent London Plc FTSE 250 2,03% -0,03% -21,34% -9,95% 4,10% -25,20% -25,42*** 12 Ediston Property Investment Company plc FTSE All Share -0,04 % 0,13% -4,19% -1,31% -0,46% -5,85% -10,42*** 13 Empiric Student
Property plc FTSE All Share 1,08% 2,33% -3,52% -6,04% 6,30% 0,16% 2,09** 14 Great Portland
Estates Plc FTSE 250 -0,17 %
1,13% -19,43% -8,19% 5,96% -20,71% -20,73*** 15 Ground Rents
Income Fund PLC FTSE All Share -0,04 % 0,16% -0,84% -0,32% -0,38% -1,43% -5,05*** 16 Hibernia REIT Plc FTSE All Share 0,23% 1,29% -1,65% 0,64% -0,27% 0,24% -0,84 17 Hammerson FTSE 250 -0,09 % 0,36% -9,73% -5,78% 3,50% -11,74% -12,26*** 18 Intu properties plc FTSE All Share 0,65% 0,50% -8,88% -4,45% 1,38% -10,79% -10,74*** 19 Land Securities Group plc FTSE 100 -0,19 % -0,16% -12,00% -6,37% 6,19% -12,53% -10,83*** 20 LondonMetric Property Plc FTSE 250 0,73% 0,74% -6,87% -8,71% 4,88% -9,23% -11,03*** 21 McKay Securities Plc FTSE All Share -0,07 % 6,29% -9,81% -18,08% 2,97% -18,70% -10,23*** 22 NewRiver REIT plc FTSE All Share -0,71 % 0,66% -4,55% -4,97% -0,25% -9,82% -10,07*** 23 Picton Property
Income Limited FTSE All Share 0,97% 0,81% -5,73% -11,61% 10,33% -5,23% -5,96*** 24 Primary Health Properties Plc FTSE 250 0,04 % 1,69% -1,37% -1,60% 0,06% -1,18% -1,06 25 RDI REIT P.L.C. FTSE All - -0,44% -2,19% -6,11% 1,80% -7,17% -5,58***
Table (7) the results of the event study. *** significance at 1%, *significance at 10% Abnormal returns & cumulative abnormal return with 190 days as estimation window and event window (-1,1) 26 Regional REIT
Limited FTSE All Share 0,14% 0,04% -2,29% -5,48% 0,88% -6,71% -5,54*** 27 Real Estate credit investments FTSE All Share 0,01 % -0,01% -4,69% -3,26% -0,92% -8,86% -11,22*** 28 Safestore Holdings plc FTSE 250 1,63% 0,26% -12,44% -6,94% 0,87% -16,62% -12,90*** 29 Schroder
European Reit Plc FTSE All Share 0,00% -0,05% -3,62% -3,69% 1,92% -5,43% -1,75*
30 Segro Plc FTSE 100 -2,00 % 0,05% -8,87% -4,22% 5,17% -9,86% -11,01*** 31 Shaftesbury PLC FTSE 250 0,43 % 0,18% -4,77% -5,54% 1,30% -8,39% -9,73*** 32 Urban Logistics
REIT plc FTSE All Share 0,31% 0,52% -0,84% -0,71% 0,91% 0,19% 0,08 33 Secure Income
REIT Plc FTSE All Share -0,67 %
-0,58% -2,68% -4,87% 0,19% -8,61% -6,77*** 34 Schroder Reit
Limited FTSE All Share -0,02 %
-0,01% -0,06% -0,05% 0,00% -0,14% -0,59 35 Target Healthcare
REIT Limited FTSE All Share 0,92% 0,41% -1,17% 0,18% 0,58% 0,93% 1,07 36 Town Centre
Securities PLC FTSE All Share -0,04 % -1,41% -5,73% -4,73% 2,78% -9,14% -5,29*** 37 UK Commercial Property REIT Limited FTSE 250 0,28 % 2,00% -5,79% -11,04% 6,97% -7,57% -8,97*** 38 The Unite Group plc FTSE 250 -2,16 % 0,28% -4,16% -5,51% 1,14% -10,42% -9,35*** 39 Workspace Group plc FTSE 250 0,97% 2,86% -13,21% -12,79% 2,34% -19,82% -14,47*** Day -1 0 1
Company name Index AR AR AR CAR (-1,1)
1 AEW UK Reit plc FTSE-Allshar e 1,14% -2,65% -0,55% 2,06% 2 Assura plc FTSE 250 2,61% -10,17% -4,93% -12,49% 3 Tritax Big Box REIT plc FTSE 250 0,74% -4,54% -7,93% -11,73 % 4 British Land Company Plc FTSE 100 0,10% -15,96% -7,00% -22,86 % 5 BMO Real estate
Invest. Lim. FTSE-AllShar e 1,18% -2,06% -9,10% -9,98% 6 Big Yellow Group PLC FTSE 250 -0,23% -8,14% -4,05% -12,42% 7 Capital & regional PLC FTSE All Share -0,60% -9,80% -5,54% -15,94% 8 Capital & counties Properties PLC FTSE 250 0,73% -13,69% -8,50% -21,46% 9 Custodian Reit PLC FTSE All Share -0,52% -1,03% -0,59% -2,14% 10 GCP Student Living plc FTSE 250 0,26% -1,65% -0,08% -1,47% 11 Derwent London Plc FTSE 250 -0,03% -21,34% -9,95% -31,32% 12 Ediston Property Investment Company plc FTSE All Share 0,13% -4,19% -1,31% -5,37% 13 Empiric Student
Property plc FTSE All Share
Table (8) Cumulative abnormal return for event window (-1,1) 14 Great Portland Estates Plc FTSE 250 1,13% -19,43% -8,19% -26,49% 15 Ground Rents Income Fund PLC FTSE All Share 0,16% -0,84% -0,32% -1,00% 16 Hibernia REIT Plc FTSE All Share 1,29% -1,65% 0,64% 0,28% 17 Hammerson FTSE 250 0,36% -9,73% -5,78% -15,15% 18 Intu properties plc FTSE All Share 0,50% -8,88% -4,45% -12,83% 19 Land Securities Group plc FTSE 100 -0,16% -12,00% -6,37% -18,53% 20 LondonMetric Property Plc FTSE 250 0,74% -6,87% -8,71% -14,84% 21 McKay Securities Plc FTSE All Share 6,29% -9,81% -18,08% -21,60% 22 NewRiver REIT plc FTSE All Share 0,66% -4,55% -4,97% -8,86% 23 Picton Property Income Limited FTSE All Share 0,81% -5,73% -11,61% -16,53% 24 Primary Health Properties Plc FTSE 250 1,69% -1,37% -1,60% -1,28% 25 RDI REIT P.L.C. FTSE All Share -0,44% -2,19% -6,11% -8,74% 26 Regional REIT
Limited FTSE All Share
0,04% -2,29% -5,48% -7,73% 27 Real Estate credit
investments FTSE All Share -0,01% -4,69% -3,26% -7,96% 28 Safestore Holdings plc FTSE 250 0,26% -12,44% -6,94% -19,12% 29 Schroder European Reit Plc FTSE All Share -0,05% -3,62% -3,69% -7.36% 30 Segro Plc FTSE 100 0,05% -8,87% -4,22% -13,04% 31 Shaftesbury PLC FTSE 250 0,18% -4,77% -5,54% -10,13% 32 Urban Logistics
REIT plc FTSE All Share
0,52% -0,84% -0,71% -1,03% 33 Secure Income
REIT Plc FTSE All Share
-0,58% -2,68% -4,87% -8,13% 34 Schroder Reit
Limited FTSE All Share
-0,01% -0,06% -0,05% -0,12% 35 Target Healthcare
REIT Limited FTSE All Share
0,41% -1,17% 0,18% -1,21% 36 Town Centre
Securities PLC FTSE All Share -1,41% -5,73% -4,73% -11,87% 37 UK Commercial Property REIT Limited FTSE 250 2,00% -5,79% -11,04% -14,83% 38 The Unite Group plc FTSE 250 0,28% -4,16% -5,51% -9,39% 39 Workspace Group plc FTSE 250 2,86% -13,21% -12,79% -23,14%
The REIT’s beta along with their index from each REIT calculated with the market-adjusted model Table (9) Abnormal returns & cumulative abnormal return with 90 days as estimation window and event window (-2,2)
Company name Index Beta (-191, -11)
1 AEW UK Reit plc FTSE-Allshare 0,022 2 Assura plc FTSE 250 0,344 3 Tritax Big Box REIT plc FTSE 250 0,264 4 British Land Company Plc FTSE 100 0,925 5 BMO Real estate Invest. Lim. FTSE-AllShare 0,156 6 Big Yellow Group PLC FTSE 250 0,606 7 Capital & regional PLC FTSE All Share 0,132 8 Capital & counties Properties PLC FTSE 250 0,933 9 Custodian Reit PLC FTSE All Share 0,068 10 GCP Student Living plc FTSE 250 0,029 11 Derwent London Plc FTSE 250 0,833 12 Ediston Property Investment Company plc FTSE All Share 0,077 13 Empiric Student Property plc FTSE All Share 0,121 14 Great Portland Estates Plc FTSE 250 0,706 15 Ground Rents Income Fund PLC FTSE All Share 0,009 16 Hibernia REIT Plc FTSE All Share 0,319 17 Hammerson FTSE 250 0,862 18 Intu properties plc FTSE All Share 0,996 19 Land Securities Group plc FTSE 100 0,964 20 LondonMetric Property Plc FTSE 250 0,454 21 McKay Securities Plc FTSE All Share 0,226 22 NewRiver REIT plc FTSE All Share 0,302 23 Picton Property Income Limited FTSE All Share 0,19 24 Primary Health Properties Plc FTSE 250 0,324 25 RDI REIT P.L.C. FTSE All Share 0,573 26 Regional REIT Limited FTSE All Share 0,126 27 Real Estate credit investments FTSE All Share -0,157 28 Safestore Holdings plc FTSE 250 0,593 29 Schroder European Reit Plc FTSE All Share 0,114 30 Segro Plc FTSE 100 0,724 31 Shaftesbury PLC FTSE 250 0,583 32 Urban Logistics REIT plc FTSE All Share -0,267 33 Secure Income REIT Plc FTSE All Share 0,115 34 Schroder Reit Limited FTSE All Share -0,009 35 Target Healthcare REIT Limited FTSE All Share 0,049 36 Town Centre Securities PLC FTSE All Share 0,072 37 UK Commercial Property REIT Limited FTSE 250 0,276 38 The Unite Group plc FTSE 250 0,627 39 Workspace Group plc FTSE 250 0,616
Company name Index AR -2 AR -1 AR 0 AR 1 AR 2 CAR (-2,2)
1 AEW UK Reit plc FTSE-Allshare 0,503% 1,117% -2,667% -0,568% -0,053% -1,67% 2 Assura plc FTSE 250 0,47% 2,78% -11,06% -5,72% 4,73% -8,80% 3 Tritax Big Box REIT plc FTSE 250 -2,96% 0,87% -4,96% -8,30% 7,18% -8,17% 4 British Land Company Plc FTSE 100 0,17% 0,01% -15,83% -6,89% 2,34% -20,20% 5 BMO Real estate
Invest. Lim. FTSE-AllShare -0,05% 1,21% -2,18% -9,21% 4,86% -5,37% 6 Big Yellow Group PLC FTSE 250 -0,25% -0,26% -8,13% -4,04% 1,06% -11,62% 7 Capital & regional PLC FTSE All Share 1,61% -0,64% -9,35% -5,14% -3,00% -16,53% 8 Capital & counties Properties PLC FTSE 250 -0,36% 0,55% -13,36% -8,23% 3,14% -18,26% 9 Custodian Reit PLC FTSE All Share 0,02% -0,45% -1,18% -0,71% -1,42% -3,74% 10 GCP Student Living plc FTSE 250 0,20% 0,27% -1,62% -0,05% -0,06% -1,26% 11 Derwent London Plc FTSE 250 1,96% -0,11% -21,38% -9,99% 4,02% -25,51%
Table(10) estimation window of 90 days. (-101,-11) The Reits cumulative abnormal return of both estimation windows. Company name CAR (-2,2) Estimation window (-191,-11) CAR (-2,2) Estimation window (-101,-11) AEW UK Reit plc -1,57% -1,67% Assura plc -7,78% -8,80% Tritax Big Box REIT plc -7,84% -8,17% British Land Company Plc -20,14% -20,20% BMO Real estate Invest. Lim. -5,24% -5,37% Big Yellow Group PLC -11,55% -11,62% Capital & regional 12 Ediston Property Investment Company plc FTSE All Share -0,07% 0,11% -4,26% -1,38% -0,47% -6,07% 13 Empiric Student
Property plc FTSE All Share 1,15% 2,46% -3,81% -6,29% 6,55% 0,06% 14 Great Portland
Estates Plc
FTSE 250 -0,24% 1,00% -19,16% -7,95% 5,72% -20,64%
15 Ground Rents
Income Fund PLC FTSE All Share -0,01% 0,23% -0,96% -0,43% -0,27% -1,44% 16 Hibernia REIT Plc FTSE All Share 0,28% 1,39% -1,91% 0,42% -0,07% 0,10% 17 Hammerson FTSE 250 -0,11% 0,32% -9,73% -5,78% 3,45% -11,84% 18 Intu properties plc FTSE All Share 0,59% 0,42% -8,81% -4,39% 1,25% -10,94% 19 Land Securities Group plc FTSE 100 -0,30% -0,30% -11,92% -6,31% 5,98% -12,85% 20 LondonMetric Property Plc FTSE 250 0,76% 0,78% -6,95% -8,77% 4,96% -9,22% 21 McKay Securities Plc FTSE All Share 0,02% 6,41% -9,87% -18,13% 3,14% -18,43% 22 NewRiver REIT plc FTSE All Share -0,70% 0,61% -4,20% -4,66% -0,41% -9,36% 23 Picton Property
Income Limited FTSE All Share 0,90% 0,90% -6,09% -11,93% 10,54% -5,58% 24 Primary Health Properties Plc FTSE 250 0,15% 1,94% -2,02% -2,16% 0,56% -1,53% 25 RDI REIT P.L.C. FTSE All Share -0,30% -0,54% -2,06% -6,01% 1,64% -7,28% 26 Regional REIT
Limited FTSE All Share 0,16% 0,13% -2,62% -5,77% 1,08% -7,01%
27 Real Estate credit
investments FTSE All Share 0,03% -0,05% -3,94% -3.68% -0,89% -8.31% 28 Safestore Holdings plc FTSE 250 1,67% 0,29% -12,29% -6,80% 0,87% -16,26% 29 Schroder European Reit Plc FTSE All Share 0,02% -0,01% -4,30% -4,15% 2,14% -6,31% 30 Segro Plc FTSE 100 -1,94% 0,20% -9,33% -4,63% 5,50% -10,19% 31 Shaftesbury PLC FTSE 250 0,13% 0,13% -4,97% -5,72% 1,29% -8,91% 32 Urban Logistics REIT plc FTSE All Share 0,62% 1,07% -1,88% -1,59% 1,89% 0,11% 33 Secure Income REIT Plc FTSE All Share -0,65% -0,55% -2,76% -4,94% 0,26% -8,65% 34 Schroder Reit
Limited FTSE All Share 0,00% 0,00% 0,01% 0,01% 0,00% 0,02%
35 Target Healthcare
REIT Limited FTSE All Share 0,92% 0,40% -1,16% 0,19% 0,56% 0,92% 36 Town Centre Securities PLC FTSE All Share -0,08% -1,50% -5,45% -4,49% 2,59% -8,93% 37 UK Commercial Property REIT Limited FTSE 250 0,32% 2,06% -5,89% -11,13% 7,08% -7,56% 38 The Unite Group plc FTSE 250 -2,15% 0,35% -4,46% -5,78% 1,32% -10,72% 39 Workspace Group plc FTSE 250 0,88% 2,80% -13,48% -13,04% 2,33% -20,51%
Capital & counties Properties PLC -18,25% -18,26% Custodian Reit PLC -3,70% -3,74% GCP Student Living plc -1,35% -1,26% Derwent London Plc -25,20% -25,51% Ediston Property Investment Company plc -5,85% -6,07% Empiric Student Property plc 0,16% 0,06% Great Portland Estates Plc -20,71% -20,64% Ground Rents Income Fund PLC -1,43% -1,44% Hibernia REIT Plc 0,24% 0,10% Hammerson -11,74% -11,84% Intu properties plc -10,79% -10,94% Land Securities Group plc -12,53% -12,85% LondonMetric Property Plc -9,23% -9,22% McKay Securities Plc -18,70% -18,43% NewRiver REIT plc -9,82% -9,36% Picton Property Income Limited -5,23% -5,58% Primary Health Properties Plc -1,18% -1,53% RDI REIT P.L.C. -7,17% -7,28% Regional REIT Limited -6,71% -7,01% Real Estate credit investments -8,86% -8.53% Safestore Holdings plc -16,62% -16,26% Schroder European Reit Plc -5,43% -6,31% Segro Plc -9,86% -10,19% Shaftesbury PLC -8,39% -8,91% Urban Logistics REIT plc 0,19% 0,11% Secure Income REIT Plc -8,61% -8,65% Schroder Reit Limited -0,14% 0,02% Target Healthcare REIT Limited 0,93% 0,92% Town Centre Securities PLC -9,14% -8,93% UK Commercial Property REIT Limited -7,57% -7,56% The Unite Group plc -10,42% -10,72% Workspace Group plc -19,82% -20,51% Table(11)
Graph of the normally distributes residuals of the regression 0 2 4 6 8 D e n si ty -.2 -.1 0 .1 .2 Residuals Kernel density estimate Normal density kernel = epanechnikov, bandwidth = 0.0218
8. References: Ahern, K. R. (2009). Sample selection and event study estimation. Journal of Empirical Finance, 16(3), 466–482. h Alkhatib, A., & Harasheh, M. (2018). Performance of Exchange Traded Funds during the Brexit Referendum: An Event Study. International Journal of Financial Studies, 6(3), 64. Banz, R. W. (1981). The relationship between return and market value of common stocks. Journal of Financial Economics, 9(1), 3–18. Brown, S. J., & Warner, J. B. (1985). Using daily stock returns. Journal of Financial Economics, 14(1), 3–31. Bonchev, L., & Pencheva, M. (2017). The impact of the Brexit vote on stock returns: an event study on European bank industry. Retrieved from: http://lup.lub.lu.se/luur/download?func=downloadFile&recordOId=8925454&fileOId= 8925455 Clayton, J., & Mackinnon, G. (2003). He Relative Importance of Stock, Bond and Real Estate Factors in Explaining REIT Returns. The Journal of Real Estate Finance and Economics, 27(1), 39–60. Dhingra, S., Huang, H., Ottaviano, G., Paulo Pessoa, J., Sampson, T., & Van Reenen, J. (2017). The costs and benefits of leaving the EU: trade effects. Economic Policy, 32(92), 651–705. Ederington, L., Guan, W., & Yang, L. (Zongfei). (2015). Bond market event study methods. Journal of Banking & Finance, 58, 281–293. Fama, E. F., & French, K. R. (1992). The Cross-Section of Expected Stock Returns. The Journal of Finance, 47(2), 427–465.
Gyourko, J., & Keim, D. B. (1992). What Does the Stock Market Tell Us About Real Estate Returns? Real Estate Economics, 20(3), 457–485. Keim, D. B. (1983). Size-related anomalies and stock return seasonality. Journal of Financial Economics, 12(1), 13–32. Lamoureux, C. G., & Sanger, G. C. (1989). Firm Size and Turn-of-the-Year Effects in the OTC/NASDAQ Market. The Journal of Finance, 44(5), 1219–1245. Mackinlay, A. C. (1997). Event Studies in Economics and Finance. Journal of Economic Literature, 35(2), 13–39. Retrieved form: https://web-b-ebscohost- com.proxy.uba.uva.nl:2443/ehost/detail/detail?vid=0&sid=13b11889-ce86-4c1f-8df4-112b1b83bfe1%40sessionmgr103&bdata=JnNpdGU9ZWhvc3QtbGl2ZSZzY29wZT1zaXR l#AN=9704153582&db=buh Monfared, S., & Pavlov, A. (2017). Political Risk Affects Real Estate Markets. The Journal of Real Estate Finance and Economics, 58(1), 1–20. Oehler, A., Horn, M., & Wendt, S. (2017). Brexit: Short-term stock price effects and the impact of firm-level internationalization. Finance Research Letters, 22, 175–181. Okunev, J., Wilson, P., & Zurbruegg, R. (2000). The causal relationship between real estate and stock markets. The Journal of Real Estate Finance and Economics, 21(3), 251– 261. Pain, N., & Young, G. (2004). The macroeconomic impact of UK withdrawal from the EU. Economic Modelling, 21(3), 387–408. Ramiah, V., Pham, H. N. A., & Moosa, I. (2016). The sectoral effects of Brexit on the British economy: early evidence from the reaction of the stock market. Applied Economics, 49(26), 2508–2514. Van Dijk, M. A. (2011). Is size dead? A review of the size effect in equity returns. Journal
Zeckhauser, S., & Silverman, R. (1983). Rediscover your company’s real estate. Harvard Business Review, 61(1), 111–117. Retrieved from: https://web-a-ebscohost- com.proxy.uba.uva.nl:2443/ehost/detail/detail?vid=0&sid=ae2106c7-a705-435b-a3a1-376a333d8b5e%40sessionmgr4007&bdata=JnNpdGU9ZWhvc3QtbGl2ZSZzY29wZT1za XRl#AN=3868118&db=buh
