Print Email Download Reference This Send to Kindle Reddit This
submit to reddit

Changes In The FTSE 100 Composition Finance Essay

A brief of London stock exchange (LSE) and FTSE 100 index: London stock exchange (LSE) is one of the oldest stock exchanges in the world. It is found in 17th century in a coffee house in London. Then it becomes a famous institution and today it lies at the heart of global financial community. LSE indices are classified in groups which include FTSE 100 index, FTSE 250 index, FTSE 350 index, FTSE All share index and so on, but FTSE 100 index is the most important. It is showed in this FTSE 100 family tree bellow:

Basing on the capitalization market of companies, then it chooses top 100 companies are listed on LSE. In January 1984, FTSE 100 started with 1000 points, now it becomes the most widely used market index in the UK.

AIMS OF TOPIC:

Stock is added or deleted from the FTSE 100 is an important event for investors and companies. This event shows the firm’s performance, financial healthy, perspective and so on. Besides that it helps investors, companies can get helpful information from these events to find consistent with investments, avoid suffering from losses. According to these events, managers can draw lessons to manage their firm profit stably and effectively. Addition, it is important for index trackers who buy or sell the stocks to be sure that the fund mimics the performance in FTSE 100 index. Therefore, get the information around events of added and deleted firms is interesting job. There are some aspects will be analysed:

Firstly, 20 days before effective date to 30 days after effective date that firms are added to and deleted from FTSE 100, changes of price and trading volume will be examined in this paper.

Secondly, determining what extend are industries effected by that event and which company get effect most, which one get benefit.

Finally, according toMase(2007) with term “nearly in – nearly out” firm, we will assess the performance of firms which are potentially added or deleted from FTSE 100.

LITERATURE REVIEW:

According to this topic, there are 4 relevant hypothesises which consistent with the research review:

Firstly, assessment of the stock market index of added and deleted firms give the meaning of the substitutability of financial assets, providing the event which is free of information.

The price pressure hypothesis:

This hypothesis is proposed firstly by Scholes (1972) it posits “temporary effect as a short-run demand increase from investors rebalancing their portfolios moves the stock price above long-run equilibrium. The price pressure hypothesis contrasts with the imperfect substitute hypothesis, which predicts a permanent price effect due to the non-availability of equivalent stocks” (Bryan Mase: 2007).The price effects must be permanent in the permanently downward sloping demand curve. It is suitable with the absence of perfect substitute. The opposite change of price effect after the event is consistent with temporary price pressure which relates to short term downward sloping demand. Harris and Gurel (1986) found a full price reversal after three weeks and it supports the temporary price pressure hypothesis. In 1991, Dhillon and Johnson collected 1983 sample which proved the effect on prices is not reverse in the post and they tentatively attribute to the advent of derivatives trading. They discovered that positive price effect could be due to information, given the related positive price effects on an added firm's bonds.

Secondly, relate to the added and deleted firms, tracking error and the evidence from Empirical Shleifer (1986), Beneish and Whaley (1996), the hypothesis is:

The long term downward sloping demand curve hypothesis:

This hypothesis illustrates the index fund to buy or sell the stock is added or deleted in order to minimize the tracking errors. This removes (releases) a large number of the added (deleted) stock from (to) circulation, thus create an excess demand (supply) that can be mitigated without a price change only if the stocks have perfect substitutes. So the price must increase (decrease) due to imperfect substitutability or downward sloping demand curve.Wurgler and Zhuravskaya (2002) and Denis, McConnell, Ovtchinnikov, and Yu(2003) also showed evidence which matches with permanently downward sloping demand curves for stocks added to the Standard &Poor 500. Furthermore, Denis, McConnell, Ovtchinnikov, and Yusaid that the positive price effects can be clarified by increased monitoring of management and an associated increase in managerial effort. In 2004, Chen, Noronha, and Singal investigated changes in the S&P 500 (period 1961-2000) and they found an asymmetry in the observed price effects. Additions created permanent price effects, whereas the abnormal returns relate to deletions are opposite. The asymmetry explains that downward sloping demand curves do not cause the price effects significantly.

The certification (information) hypothesis:

As A. Gregoriou & C. Ioannidis (2006:347) stated that "A corollary is that you can sell (or buy) large blocks of stock at close to the market price as long as you can convince other investors that you have no private information". This statement relies on assumption that securities are near perfect substitutes for each other. If they are, excess demand for single security will be very elastic and there is no effect on price though investors sale or buy alot of shares.

According to this hypothesis, the additions to (deletions from) the FTSE 100 index composition convey new positive (negative) information about the stocks. Due to the asymmetric information, investors, analysts or institutions cannot distinguish between good and bad stocks. The event that a firms is added (deleted from) FTSE 100 index composition is a clear positive (negative) evidence of the financial health, profitability or prospective of the firm. Thus, the price of added (deleted) firm’s increases (decreases) around the event dates. The findings by Jain (1987), Dhillon and Jonhson (1991) are consistent with this hypothesis.

The liquidity hypothesis

According to this hypothesis, the addition to or deletion from the FTSE 100 index composition draw much attention by the public and analysts leading to the abnormal trading volume. Harris and Gurel (1986) and Mase (2007) found the evidence of increase in trading volume around the event dates. According to A. Gregoriou & C. Ioannidis (2006:348) stated “When a stock is added to the index, this is expected to result in increased volume making the added stock more liquid and the expectations of this benefit accounts for the price increase” The firms want to increase the liquidity because of they will get more attention from investors, analysts and it will reduce the bid-ask spreads. As Beneish and Gardner (1995: 11) pointed out "If inclusion in (exclusion from) the FTSE 100 list is followed by increased (decreased) scrutiny by analysts, investors and institutions, the firm’s information environment is richer (poorer) and the stock will be traded more (less) widely and become more (less) liquid". If changes in information environment relate to changes in FTSE 100 list so the stock price of FTSE 100 list firms will amend to reflect changes in the future levels of available information. Indeed, they found that added (deleted) firms rise (reduce) trading volume, and an increase (decline) in the ‘quantity’ of available information after the FTSE 100 list change.

THE REVIEW OF PREVIOUS STUDIES:

Gregoriou and Ioannidis assessed the changes of the returns and trading volume of firms that were listed on addition and deletion in FTSE 100 index period from 1984 to 2001. The analysis showed that shareholders gain or lose from the added (deleted) firms in FTSE100.

Moreover, added (deleted) firms had grown or declined of trading volume significantly but also increase (decrease) of available information after the changes of FTSE 100 list. Respectively, it proved that investors who hold stocks with more (less) available information, resulting to them have lower (higher) trading cost. Indeed, the findings are consistent with the information cost and liquidity hypothesis whereas it does not help to prove the imperfect substitute hypothesis.

There was a useful research about the performance of 162 added and 160 deleted firms from the FTSE100 by Hamil Et Al (2004) conducted in period 1984 to 2000. According to the 2 indices of firms: Mean Abnormal Return (MARs) from day -5 to day +5 and Cumulative Abnormal Return (CARs) from day -21 to day +60.They found that prior to addition MARs are predominately positive but in the post event period they are negative as well as for deletions. Additions (deletions) experience a sharp price grow (reduce) the day before effective date. Day (-1) experienced a price increase, which is 1.45% for additions and a price decrease which was (-1.11%) for deletions. They stated that this relatively large price change can be accounted for by index funds buying (selling) stock immediately prior to their additions to (deletion from) the FTSE 100 to minimize tracking error. They also found that an anomaly appears on day 0 for additions.

Furthermore, in the period 1992 and 2005 Mase considered the effect of changes in FTSE 100. Especially, he find out short-term impact relate to changes in constituents of FTSE100 index. Specifically, he examined the influence in the period from 10 days before the announcement date to 30 days after the effective date. The positive (negative) cumulative abnormal return (CARs) for addition (deletion) was detected. The effective date saw the start of complete reversal in the CAR, which by t+30 is 0.4% and -0.2%. The reversal is slightly quicker for additions than for deletions. On the other hand, he also found the abnormal level of trading volume and there is liquidity market in this circumstance. The abnormal volume of 6.1% and 4.9% are the highest percent on day prior the event for additions and deletions respectively. The abnormal volume is similar with the respective abnormal returns on day (-1) of 1% and (-1.9%). Besides that, Mase stated that abnormal returns and trading volumes of “nearly in” and “nearly out” firms are less than those gained for additions and deletions. He also pointed that there is statistically significant anticipatory in stocks that just fail to be promoted to the FTSE 100 index composition.

On the other hand, there are some literature which considered the effect of additions and deletion from the Standard and Poor 500 (S&P 500) index. The research from Beneish and Whaley (1996), from stock is added from S&P 500,they pointed out that there are only some partially reversed from positive abnormal return between the announcement and effective date in period 1989 to 1994. Similarly, while assessing impact of the additions and deletions (1990-1995), Lynch and Mendenhall (1997) found significant positive (negative) abnormal return after announcement of firms, and again only partially reversed. Moreover, the trading volume is much higher from announcement date and it is completely large on the day before effective day. Impressively, Harris &Gurel (1986) studied 228 additions from 1973 to 1983 and they tested the presence of price pressure which is believed, is not confound by information problem. They recognized that there is a suggestive of a shift in demand because of there is a large increase in volume on the first trading day after an addition to the S&P 500 is announced. They also delivered that price soared by over 3% promptly following an addition announcement whereas the price raise gradually, subsequently it is reversed next 2 weeks after effective date. From the above results which validated the price pressure hypothesis otherwise it opposes the certification hypothesis.

Incidentally, Lynch and Mendenhall (1997) considered about pre-event performance contrasts for the S&P 500. They found that, their pre-announcement CARs are negative for the added firms but positive for the deleted firms. They discovered extensive and significant positive (negative) abnormal return on the announcement date for additions or deletions. It is very valuable that the pre-announcement CARs in this research are a component of the measure used to establish index constituents in the United Kingdom. The significant abnormal returns with the announcement date are inconsistent with semi-strong form market efficiency, while the price reversal on and after the effective date is equivalent with price pressure hypothesis.

In contrast to Lynch and Mendenhall (1997), the insignificant abnormal return on respective announcement days infer that market is not nonplussed by the announcement and has therefore already included the announcement into returns.

There are several studies examines the price effect relate to the changes in other indices. Beneish and Gardner (1995) analyse the changes incomposition of the Dow Jones Industrial Average (DJIA) which is not tracked by index fund. They found a minor effect on stock price or trading volume of added firms but deleted firms encounter the significant excess return at announcement time and the trading volume absolutely reduces.

DATA:

Firstly, the meetings to review the constituents will be held on the Wednesday after the first Friday in March, June, September and December annually. A firm will be added (deleted) if it rises to 90th position or above (falls to 111th position or below).After review, in the FTSE 100 there are 90 largest stocks, plus 10 of the next 20 largest, i.e., 10 stocks ranked between 91 and 110.Stocks ranked between 91 and 110 of area is called gray area and the stock is not added to index when it's position is below 100 and a stock is not be moved down if it's position is 100 or higher.If there are less addition than deletion, the highest ranking companies will be inserted even they are currently not in the index so as to match the total number of deletions. Companies would not be added to FTSE 100 if it did not pass the liquidity test though those companies are very large. Rules for addition and deletion of stocks are adopted to keep the stability for selecting constituents of FTSE 100 family. All the announcement and effective dates of additions and deletions are issued by FTSE international Limited. The index changes will be reported after the market closes, so the next trading day will be get from the announcement day. The chosen data in this paper is from 2000 to 2010.

Mase (2007) stated that this process contrasts strongly with the process of inclusion and exclusion of the S&P 500 index because it is based purely on the market capitalisation of the respective firms. According to the latter process, a stock is only added to the index when a stock has to be removed and the added stocks are selected on the basic of a number of criteria such as: liquidity, size, industry and financial soundness. In addition, an attempt is made to avoid excessive turnover of stocks within the index. This means that there is likely to be a degree of subjectivity in the newly added firms.

This study will basically associate with previous research given that changes to the FTSE 100 are made by ranking of capitalization market of stock. Indeed, some firms drop out the index then they included again. If investor's understanding and monitoring are important element support positive price for additions, so it would be significant differences between the price effect of new and previous constituents.

There were 8 main industries that will be analyzed, they are from variable fields:

- Basic material

- Services

- Health care

- Industrial goods

- Consumer goods

- Financial

- Technology

- Utilities

The data will be collected from Datastream and Bloomberg. Eview 7 will be used to run regression models. This paper will examine 102 added and 105 deleted firms. The analysis will show what firm is affected most and what firm is benefited the most from the change in FTSE 100 index composition. All of these companies are collected from the website of London stock Exchange (LSE).

Table 1: The classification of the additions and deletions according to industries

Industries Additions Deletions

Consumer Discretionary 11 17

Consumer Staples 12 15

Energy 7 4

Financials 15 12

Industrials 12 9

Information Technology 6 9

Materials 15 10

Support Services 12 11

Telecommunication Services 6 10

Utilities 6 8

Total 102 105

Secondly, the firms which named “nearly in and nearly out” from the list of FTSE100 will be examined. The comparison and contrast between these firms and added (or deleted) firms would be analyzed in the period 2005 to 2010. From the quarterly review of FTSE 100, these firms will be chosen and limited in 3 largest firms remain outside the FTSE 100 and 3 smallest firms inside the FTSE 100 respectively.

The price and trading volume of the additions and deletions are collected from Datastream and Bloomberg.

METHODOLOGY:

The event study methodology is used in this report and it will examine the effect of an event on the value of firm. The main method is that we have to identify the event dates and each event I will divide to several event windows. In this methodology, we need to estimate the abnormal return around the event. Lynch and Mendenhall (1997) and Chen, Noronha, and Singal(2004) measure abnormal returns as the difference between the stock and market return. As we know the formula:

\textrm{Abnormal\ Return} = \textrm{Actual\ Return} - \textrm{Expected\ Return}

According to Mase (2007), this study we will estimate the abnormal returns is considered as prediction errors in the relation with the market from the market model:

(1)

Where and are the rates of return of security i and market on day t

Is a stochastic error term - the abnormal return of security i on day t

The effective date is (t) and the announcement date is (t-7). The common event study approach is calculating the model parameters over periods before the event (Biktimirov 2004). In this study, the period before 20 days and after 30 days the effective date will be examined as they are (t-20 and t+30 respectively). Then, I will divide the research period into some small event windows to consider: pre-announcement window (t - 20 to t – 7); post-announcement window (t - 6 to t – 1); post-event window (t to t + 7 and t + 8 to t + 30).

After running this model, we can get the result of the abnormal return for each day and cumulative abnormal return for each event window around the 2 above event dates. The findings will support us to determine the performance of added and deleted firms. Especially, there are some additions or deletions that occur at the same time, i.e. there is event clustering. According to Mase (2007), this may make the abnormal returns calculated above correlated because the standard event study method is based on an assumption that the event windows do not overlap. Therefore, the statistical significance of the mean cumulative abnormal returns calculated by equation (1) is unable to estimate by the distribution of the individual firms’ CAR. Therefore, to solve this trouble, the methodology by Thompson (1985) was discussed. From this method, for some same event windows of several firms, model parameters and associated variances should be estimated by aggregating the shared-event firm returns into portfolios. For each event cluster, we calculate an equally weighted portfolio return. In this case, there are 22 added portfolios and 19 separate additions, 27 deleted portfolios and 13 separate deletions. The second model by Mase (2007) is therefore presented to examine the return of portfolio:

(2)

Where

Rit and Rmt are the returns to portfolio i and the market on day t

Di1t is a dummy variable with the value 1/14 for the event days t – 20 to t – 7 and 0 otherwise;

Di2t is a dummy variable with the value 1/6 for the event days t – 6 to t – 1 and 0 otherwise;

Di3t is a dummy variable with the value 1/8 for the event days t to t + 7 and 0 otherwise;

Di4t is a dummy variable with the value 1/13 for the event days t + 8 to t + 30 and 0 otherwise.

𝛾i1 is the cumulative abnormal return to portfolio i for the event days t – 20 to t – 7;

𝛾i2 is the cumulative abnormal return to portfolio i for the event days t – 6 to t – 1;

𝛾i3 is the cumulative abnormal return to portfolio i for the event days t to t + 7;

𝛾i4 is the cumulative abnormal return to portfolio i for the event days t + 8 to t + 30;

The model is estimated over the period from t - 20 to t + 210.

Afterwards, the mean cumulative abnormal return is calculated for each event window, denoted M𝛾i1, M𝛾i2, M𝛾i3, M𝛾i4 respectively. Then, t-statistic value is obtained that will test the hypothesis of the M𝛾ij is different from 0:

Where is the cross-sectional standard error of the respective regression coefficients.

The CAR across:

the pre-event window (PRE) (from t – 20 to t - 1) is the sum of 𝛾i1 + 𝛾i2;

the post-event window (PE) (from t to t + 30) is the sum of 𝛾i3 + 𝛾i4;

the post-announcement window (PA) (from t – 20 to t - 1) is the sum of 𝛾i1 + 𝛾i2 + 𝛾i3;

the complete event window (CEW) (from t – 20 to t + 30) is the sum of all 4 dummies.

Additional, to check the robustness, statistic is obtained from the mean of t-statistics of the individual regression model coefficients:

Where:

are the t-statistics from each of the individual model coefficients

N is the number of additions or deletions portfolios.

Secondly, the trading volume response around the announcement date and effective date will be examined. Similar to equation (2), I would like to consider the trading volume according to portfolios for additions and deletions have the same effective date. Cross-sectional means are computed using the estimation technique by Harris and Gurel (1986) and recently applied by Gregoriou and Ioannidis (2003):

𝛴 (3)

Where

Where and are the trading volumes of security i or portfolio i and of the market in event-time period t, respectively,

and are the average trading volumes of the security and of the FTSE All-share in the 8 weeks preceding the announcement week.

The volume ratio,, is a standardized measure of period t trading volume in security i, adjusted for market variation. Its expected value is 1 if there is no change in volume during event-period t relative to the prior 8 weeks.

I would like to calculate the trading volume response in the period from the 8th week preceding the announcement week to the 7th week succeeding the effective week but only focus on discussing findings in the period from week - 4 to week 5 (from t – 20 to t + 30).

Subsequently, I would like to test the hypothesis that whether the mean volume ratio (MVR) is different from 1. The t-statistic is calculated by the following equation:

Where ;

Where

isthestandard error of ;

is the standard deviation of ;

N is the number of additions or deletions.

The findings helps us to determine whether there is abnormal trading volume of added and deleted firms around the event dates so that we can assess the impact of the adding to or deleting from the FTSE 100 on trading volume of those securities. This result may also explain the result obtained from equation (1) for the abnormal return

I apply the same technique for the sample classified according to the above-mentioned industries in order to determine which industries are the most influenced and to what extend that are influenced by the changes in the FTSE 100 index composition. This can be achieved by comparing results among the industries.

The short-term performance of the sample of “nearly in” and “nearly out” firms is examined in the same way. Results will show the price and trading volume response of the ‘nearly’ stocks around the announcement of quarterly reviews. The results are compared with the result of the added and deleted firms in the same period in order to determine the difference between the two samples.

All the results are compared with the above-mentioned hypotheses and previous studies to determine the compatibility among them.

The data is collected from Datastream and Bloomberg. Eview 7 will be used to run regression models.

CHAPTER V: FINDINGS AND DISCUSSION.

V.1 The response of additions to and deletions from the FTSE 100 index composition

V.1.1 The price response

The price response gets the market participants' crucial attention. So it is extremely interesting to examine it. In purpose to estimate the response, equation (1) is applied to the data of the additions and deletions to examine the changes in return in the period from t – 20 to t + 30. The result of the abnormal returns (ARs) on each day in the event window and the cumulative abnormal returns (CARs) over the completed event window is showed by chart below:

Figure 2: Abnormal returns and Cumulative abnormal returns of additions around the announcement and effective date (2000 – 2010)

Figure 3: Abnormal returns and Cumulative abnormal returns of deletions around the announcement and effective date (2000 – 2010)

According to the Figure 2, for the response of additions, in general, the return of added firms considerably and consecutively raise in the period from t – 20 to t – 7 (announcement date), keep going to go up slightly between t – 6 to t - 1, then a little reverse between t and t + 8 and significantly reverse in the period from t + 8 to t + 30. Specifically, Cumulative abnormal return (CAR) is 5.52% on the day t – 7, 6.40% and 4.39% on the days t – 1 and t + 7 respectively and 1.33% over the whole event window (from t - 20 to t + 30). The highest point and lowest points of CAR are on day t – 1 and t + 25 (6.40% and 0.97% respectively). For each small window, the CAR is 5.52% for the window from t – 20 to t – 7, 0.88% from t – 6 to t – 1, -2.00% from t to t + 7 and -3.06% from t + 7 to t + 30. Especially, when considering abnormal returns (ARs) for each day in the event window, I find that the AR is highest on the day before the effective date (t – 1) and lowest on the effective date (1.60% and -1.73% respectively).

The findings of the deletions are nearly opposite with those of the additions. The Figure 3 shows that, for the response of deletions, in general, the price of deleted firms much declines in the period from t – 20 to t + 7 and continue increasing between t - 6 to t + 7 then significantly reverses in the period from t + 8 to t + 30. Especially, the CAR is -6.15% on the day t - 7, -7.92% on the day t + 7 and -2.29% over the whole event window. The CAR is lowest on the day t + 7 (-7.92%) and highest on the day t + 24 (-1.34%). The CAR is -6.15% over the window from t – 20 to t – 7, -0.97% from t – 6 to t - 1, -0.80% from t to t + 7 and 5.63% from t + 8 to t + 20. Interested findings are that the AR reduced by 1.24% on the day before effective date and increases 0.78% on the effective date. The AR is highest on the day t + 17 (1.31%).

As above-mentioned, due to the event clustering, the statistical significance of the mean cumulative abnormal returns calculated by equation (1) cannot be examined by the distribution of the individual firms’ cumulative abnormal returns. So equation (2) is applied for the data and obtains the following results.

As can be seen from the Table 2, the results are similar to those obtained from the equation (1) and plotted in the Figure 2 and 3.

Table 2: Price effects around announcement and effective dates (2000 – 2010)

Additions Deletions

α -0.001 -0.001

(-2.68) [-0.34] (-2.69) [0.07]

β 1.048 1.014

(22.82) [13.39] (22.72) [22.12]

M𝛾1 0.054 -0.061

(7.54) [4.63] (-7.16) [-3.73]

M𝛾2 0.009 -0.009

(1.25) [1.84] (-1.21) [-2.47]

M𝛾3 -0.020 -0.010

(-4.37) [-2.14] (-1.27) [-0.27]

M𝛾4 -0.029 0.055

(-3.72) [-2.60] (3.00) [1.33]

PA -0.040 0.036

(-3.08) (1.81)

PRE 0.063 -0.070

(6.03) (-8.13)

PE -0.049 0.044

(-4.54) (2.32)

CEW 0.014 -0.026

(0.92) (-1.14)

The t-statistics in parentheses are ;

The statistics in brackets are

Furthermore, for addition from the table 2 the mean CAR is significant positive with level 1% in the period from (t-20) to (t-7) and obvious negative between t and t+7, t+8 to t+30. Thought, it is very close to 0 in the period from t to (t-6) to (t-1). Moreover, we can see pre-announcement mean CAR and post- announcement are 5.4% and -4.00% respectively; pre-event mean CAR and post-event mean CAR are 6.3% and -4.9%. The mean CAR over the whole event window is 1.4%. While the PA and PE mean CARs are considerably negative, the PRA (pre-announcement), PRE mean CARs are significantly positive. The CEW mean CAR is insignificantly different from 0.

On the other hand, for the deletions, I found significantly negative mean CAR in the periods from t – 20 to t – 7 and significantly positive mean CAR in the period from t + 8 to t + 30 but it is insignificantly negative in the periods from t - 6 to t -1 and from t to t + 7. The pre-announcement mean CAR and post-announcement mean CAR are -6.1% and 3.6% respectively; pre-event mean CAR and post-event mean CAR are -7.0% and 4.4% respectively. The mean CAR over the whole event window is -2.6%. The PRA and PRE mean CARs are significantly negative while the PE means CARs are significantly positive and the PA mean CAR is insignificantly positive. The CEW mean CAR is also insignificantly different from 0.

V.1.2 The trading volume response:

The trading volume response to the additions and deletions is estimated by the above-mentioned equation (3). As presented in the methodology, I calculate the trading volume response from week -10 to week 7 but I only focus on analysing the response from week – 4 to week 5 (from t – 20 to t + 30). The results are presented in the Table 3 and 4 below.

Table 3: The trading volume response to the additions around the announcement and effective date (2000 – 2010)

Period

MVR

STD

t-statistic

Percent>1

Week - 10

1.16

0.43

3.79

58

Week - 9

1.02

0.23

0.85

57

Week - 8

1.02

0.28

0.84

54

Week - 7

0.92

0.23

-3.65

23

Week - 6

0.98

0.26

-0.85

44

Week - 5

1.00

0.24

-0.19

52

Week - 4

0.98

0.23

-0.68

44

Week - 3

1.03

0.23

1.24

57

Day - 10

1.20

0.53

3.83

64

Day - 9

1.30

0.67

4.49

66

Day - 8

1.34

0.55

6.16

73

Day - 7

1.58

2.25

2.60

69

Day - 6

1.16

0.38

4.22

65

Day - 5

1.26

0.50

5.25

72

Day - 4

1.37

0.79

4.77

63

Day - 3

1.29

0.40

7.29

71

Day - 2

1.42

0.69

6.24

71

Day - 1

4.07

2.14

14.48

99

Day - 0

2.34

1.16

11.74

97

Week 0

1.54

0.55

9.94

90

Week 1

1.29

0.43

6.71

72

Week 2

1.27

0.41

6.76

70

Week 3

1.14

0.33

4.36

67

Week 4

1.09

0.33

2.78

60

Week 5

1.04

0.33

1.29

53

Week 6

1.13

0.45

2.83

63

Week 7

1.12

0.42

2.89

61

Week 0 starts from the effective date;

MRV is the mean volume ratio and indicate how many times the trading volume of the additions increases compared with the daily mean volume over the 8 weeks (week – 3 to week – 10) prior to the announcement.

Table 4: The trading volume response to the deletions around the announcement and effective date (2000 – 2010)

Period

MVR

STD

t-statistic

Percent>1

Week - 10

0.94

0.21

-2.74

34

Week - 9

1.00

0.30

0.12

48

Week - 8

1.01

0.23

0.25

52

Week - 7

0.97

0.20

-1.75

38

Week - 6

0.95

0.23

-2.24

30

Week - 5

1.02

0.20

1.19

54

Week - 4

1.07

0.27

2.47

52

Week - 3

1.13

0.22

6.09

79

Day - 10

1.21

0.66

3.31

54

Day - 9

1.23

0.54

4.33

67

Day - 8

1.25

0.60

4.26

71

Day - 7

1.20

0.46

4.35

63

Day - 6

1.20

0.46

4.53

67

Day - 5

1.13

0.43

3.05

57

Day - 4

1.26

0.60

4.45

64

Day - 3

1.25

0.54

4.68

63

Day - 2

1.42

0.67

6.44

78

Day - 1

3.09

1.80

11.95

97

Day - 0

2.01

1.48

6.98

91

Week 0

1.28

0.37

7.83

88

Week 1

1.17

0.41

4.27

62

Week 2

1.20

0.62

3.33

65

Week 3

1.07

0.42

1.71

40

Week 4

0.97

0.26

-1.19

33

Week 5

0.92

0.30

-2.69

40

Week 6

0.95

0.31

-1.62

37

Week 7

0.98

0.34

-0.46

45

Week 0 starts from the effective date;

MRV is the mean volume ratio and indicate how many times the trading volume of the additions increases compared with the daily mean volume over the 8 weeks (week – 3 to week – 10) prior to the announcement.

For the additions, the Table 3 shows that, compare with the daily mean trading volume over the 8 weeks prior to the announcement, the trading volume declines in week – 4 (from t – 20 to t – 16), progressively increases in the periods from week - 3 to week -1 (from t – 15 to t -1) and then gradually reverse from week 0 (changing week) to week 5 (from t to t + 30). Moreover, we can see that MRVs are significantly higher than 1 between week – 2 to week 4. Especially, the MVR increases substantially on the day before the effective date and on the effective date (respectively 4.07 times and 2.34 times as large as the daily mean trading volume over the 8 weeks prior to the announcement). The percent of additions with the mean volume ratios (MVRs) more than 1 are also highest on these 2 days (99% and 97% respectively).

For the deletions, Table 4 shows that, compare with the daily mean trading volume over the 8 weeks prior to the announcement, the MVRs progressively increase in the period from week – 5 to week - 1 (from t – 20 to t -1), gradually reverse in the periods from week 0 to week 3 (from t to t + 20) and reduce in week 4 and 5 (from t + 21 to t + 30). The MVRs are substantially more than 1 in the period from week -4 to week 1, little higher than 1 in week 2 and 3, significantly less than 1 in week 4 and 5. Notably, I find intensive increases in the trading volume on the day before the effective date and on the effective date (respectively 3.09 times and 2.01 times as large as the daily mean volume over the 8 weeks prior to the announcement). The percent of deletions with the mean volume ratios (MVRs) more than 1 are also highest on these 2 days (97% and 91% respectively).

V.1.3 Discussion of the findings:

For the price response, the above-mentioned results illustrate the effect of the announcement and changes on the price and trading volume. The increases (decreases) in the price and trading volume for the added (deleted) firms prior the announcement and effective date indicate anticipatory trading in these periods. The reversal of the price and trading volume of the added (deleted) firms after the effective date means that the abnormal returns or trading volume is temporary. This supports the price pressure hypothesis. It is possible that the heavy short-term trading by index funds moves prices temporarily away from equilibrium. The results are consistent with those of Mase (2007) who also examined the impact of changes in the FTSE 100 index and, Haris and Gurel (1986), Shleifer (1986) who studied price and volume effects associated with changes in the S&P 500 list. However, these result contrast with those of Lynch and Mendenhall (1997) who found that pre-announcement MCARs are negative for the added firms and positive for deleted firms when they considered the changes in S&P 500 index composition.

In addition, the fact that the abnormal returns increase (decrease) considerably for the added firms (deleted firms) on the day before the effective date, indicates that market participants pay most attention to these stocks on this day. This is consistent with findings of Mase (2007) and Hamill et al (2004). However the findings contrast with those of Lynch and Mendenhall (1997). They documented that abnormal returns are significantly positive (negative) on the announcement date for additions (deletions).

For the trading volume response, the findings exposed an increase in stock liquidity relates to changes in FTSE 100 index composition. The fact that the trading volume of both additions and deletions soar dramatically on the day before the effective date and on the effective date shows that there is short-term buying (selling) pressure and reactions of index fund in order to keep their fund to mimic the performance of the FTSE 100. This is an evidence of the liquidity hypothesis which explains that the addition to or deletion from the FTSE 100 index composition draw much attention by the public and analysts leading to the abnormal trading volume. The result is consistent with the findings of Mase (2007).

Furthermore, the development in the trading volume of the added and deleted stocks the same as with the increase in the price of the stock around the event dates. In particularly both the price and trading volume rise dramatically on the day before effective date of the additions and deletions. Therefore, there is a relationship between the price and trading volume of the additions.

V.2 The response to the changes according to industries

After estimating the price and trading volume of the additions to and deletion to FTSE 100 index composition in general, I would like to classify these changes according to industries in order to determine the impact level to each industry.

V.2.1 The price response rely on industries

I examine mean cumulative abnormal returns (CARs) in 5 windows:

- PRA - Pre-announcement window ( from t - 20 to t – 7)

- PA - Post-announcement window (from t – 6 to t + 30)

- PRE - Pre-event window (from t – 20 to t -1)

- PE - Post-event window (from t to t + 30)

- CEW - Complete event window (from t – 20 to t + 30)

By comparing the results, I find the most and least influenced industry as well as the most and the least profitable industry for both the additions and deletions in each period.

For the additions.

In order to compare the CARs among these industries, I would like to plot them on 5 charts (5 windows respectively) below.

Figure 4: CARs of the added industries in the Complete event window (CEW)

Figure 5: CARs of the added industries

in the Pre-announcement window (PRA)

Figure 7: CARs of the added industries

in the Pre-event window (PRE)

Figure 6:CARs of the added industries

in the Post-announcement (PA)

Figure 8: CARs of the added industries

in the Post-event window (PE)

According to the charts above, the price response of the industries is quite consistent with the general results presented in the Section VI.1.a, i.e the price soars in the pre-announcement and pre-event window and then reverses in the post-announcement and post-event window. The more prices of the added industries increase in the pre-announcement and pre-event window, the more they reverse in the post-announcement and post-event window. The most influenced industry is Telecommunication Services. This means its price is most sensitive to the inclusion in the FTSE 100 index composition. Its CAR increases by 9.92% in the pre-announcement window and then reduce by 15.50% in the post-announcement window. The second and third positions are Information Technology and Industrials industries respectively. Financials and Utilities is the least influenced industries. This mean its price is least sensitive to the inclusion in the FTSE 100 index composition. Their CARs fluctuate around 2%.

Especially, there are 3 industries in 10 added industries have negative CARs in the complete event window. They are telecommunication Service Industries, Consumer discretionary, and Consumer Staples. Industry has the lowest cumulative Abnormal Returns is Telecommunication service Industry by (-5.58%) whereas Information Technology increased the highest by 4.46%. These percent above show that two industries above have the most and the least profitable industries in the complete event window.

Likewise, the price pressure hypothesis among added industries is applied. The abnormal return of 8 industries increases before effective date and reversals from effective date. It is consistent with the study of Malic (2006).

For the deletions

In order to compare the CARs among these industries, I would like to plot them on 5 charts (5 windows respectively) below.

Figure 9: CARs of the deleted industries in the complete event window (CEW)

Figure 10: CARs of the deleted industries

in the Pre-announcement window (PRA)

Figure 12: CARs of the deleted industries in the Post-announcement window (PA)

Figure 11: CARs of the deleted industries in the Pre-event window (PRE)

Figure 13: CARs of the deleted industries in the Post-event window (PE)

The charts above show that, in general, the price of the deleted industries responses in the same direction as presented in the Section VI.1.a. This means the trading volume decreases in the pre-announcement and pre-event window and then reverses in the post-announcement and post-event window. Only Support Services industry increases in the pre-event windows by 2.83%. Similar to the price response of the added industries, the more the prices of the deleted industries decrease in the pre-announcement and pre-event window, the more they reverse in the post-announcement and post-event window, and the most influenced industry is Telecommunication Services with -16.61% CAR in the pre-event window and 36.14% in the post-event window. The second and third positions are Materials and Utilities. The least influenced industry is Support Services. Its CAR deceases by 0.38% in the pre-announcement window and increases 2.35% in the post-announcement window. This indicates that the prices of the Telecommunication Services and Support Services are the most sensitive and least sensitive industries to the exclusion from FTSE 100 index composition.

Moreover, in the Complete event window, Financial, Support Services and Telecommunication Services industries have positive CARs, especially, the CAR of Telecommunication Services is extremely high (19.53%). The Materials industry has the lowest CAR (-15.11%). These mean Telecommunication Services is the most profitable industry while Materials is the least profitable industry over the complete event window.

The result supports the price pressure hypothesis. Out of 10 industries, there are 8 industries with abnormal decreases in return over the pre-event windows and reversals over the post-event windows. This is consistent with the results by Malic (2006). Amazingly, the price of Industrial industry drops in both pre-event and post-event windows.

V.2.2 The trading volume response according to industries:

Consumer Discretionary

Consumer Staples

Energy

Financials

Industrials

Information Technology

Materials

Support Services

Telecom. Services

Utilities

Week -10

1.09

1.05

1.09

1.15

1.01

1.20

1.10

1.22

1.54

1.09

Week -9

0.89

0.97

0.94

0.91

1.08

1.27

1.02

1.01

1.45

1.00

Week -8

1.15

0.91

0.90

1.00

1.08

1.26

0.92

0.92

1.39

0.97

Week -7

1.07

0.96

0.95

0.91

0.99

0.72

0.81

0.89

0.93

0.95

Week -6

0.90

0.99

1.29

1.01

0.99

0.72

1.02

0.98

0.66

1.14

Week -5

1.01

1.01

1.06

0.99

0.79

1.01

1.21

0.98

0.72

1.12

Week - 4

1.01

1.08

0.82

0.97

1.08

0.75

1.02

1.09

0.75

1.00

Week - 3

0.97

1.00

1.07

1.07

1.05

1.32

1.01

0.96

1.14

0.83

Day -10

1.64

1.06

1.48

1.15

1.30

0.99

1.28

1.12

0.74

1.03

Day - 9

1.41

1.03

1.40

1.15

1.86

1.55

1.38

1.19

0.97

0.85

Day - 8

1.24

1.19

1.83

1.23

1.45

1.28

1.55

1.60

0.84

0.82

Day - 7

0.97

2.90

1.11

1.28

2.07

1.75

1.52

1.55

0.95

1.11

Day - 6

1.04

1.01

1.41

1.13

1.29

1.47

1.33

1.23

0.82

0.76

Day - 5

1.31

1.20

1.56

1.32

1.57

1.52

1.21

1.04

0.84

1.04

Day - 4

1.58

1.83

1.63

1.14

1.32

1.72

1.12

1.39

1.14

1.05

Day - 3

1.31

1.41

1.14

1.46

1.14

1.46

1.04

1.69

0.98

1.13

Day - 2

1.46

1.30

1.06

1.43

1.59

1.38

1.66

1.50

1.59

0.86

Day - 1

4.46

3.62

3.08

4.44

3.77

6.34

4.83

4.59

2.05

2.36

Day - 0

2.75

2.00

2.24

2.70

3.08

3.27

1.84

2.06

1.69

1.72

Week 0

1.86

1.39

1.70

1.58

1.51

1.84

1.33

1.59

1.49

1.26

Week 1

1.56

1.25

1.12

1.49

1.15

1.31

1.07

1.35

1.38

1.11

Week 2

1.58

1.21

1.24

1.25

1.40

1.18

1.05

1.30

1.40

1.18

Week 3

1.23

1.10

1.32

1.12

1.14

1.34

1.01

1.09

1.37

1.04

Week 4

1.15

1.02

1.13

1.10

1.04

1.31

0.86

1.18

1.33

1.13

Week 5

1.22

0.98

1.12

0.93

1.36

0.79

0.91

1.06

1.15

0.88

Week 6

1.58

0.96

1.25

1.14

1.10

0.87

1.04

1.05

1.32

1.02

Week 7

1.26

0.86

1.80

1.00

0.99

0.89

1.15

1.23

1.34

0.97

Consumer Discretionary

Consumer Staples

Energy

Financials

Industrials

Information Technology

Materials

Support Services

Telecom. Services

Utilities

Week -10

1.09

1.05

1.09

1.15

1.01

1.20

1.10

1.22

1.54

1.09

Week -9

0.89

0.97

0.94

0.91

1.08

1.27

1.02

1.01

1.45

1.00

Week -8

1.15

0.91

0.90

1.00

1.08

1.26

0.92

0.92

1.39

0.97

Week -7

1.07

0.96

0.95

0.91

0.99

0.72

0.81

0.89

0.93

0.95

Week -6

0.90

0.99

1.29

1.01

0.99

0.72

1.02

0.98

0.66

1.14

Week -5

1.01

1.01

1.06

0.99

0.79

1.01

1.21

0.98

0.72

1.12

Week - 4

1.01

1.08

0.82

0.97

1.08

0.75

1.02

1.09

0.75

1.00

Week - 3

0.97

1.00

1.07

1.07

1.05

1.32

1.01

0.96

1.14

0.83

Day -10

1.64

1.06

1.48

1.15

1.30

0.99

1.28

1.12

0.74

1.03

Day - 9

1.41

1.03

1.40

1.15

1.86

1.55

1.38

1.19

0.97

0.85

Day - 8

1.24

1.19

1.83

1.23

1.45

1.28

1.55

1.60

0.84

0.82

Day - 7

0.97

2.90

1.11

1.28

2.07

1.75

1.52

1.55

0.95

1.11

Day - 6

1.04

1.01

1.41

1.13

1.29

1.47

1.33

1.23

0.82

0.76

Day - 5

1.31

1.20

1.56

1.32

1.57

1.52

1.21

1.04

0.84

1.04

Day - 4

1.58

1.83

1.63

1.14

1.32

1.72

1.12

1.39

1.14

1.05

Day - 3

1.31

1.41

1.14

1.46

1.14

1.46

1.04

1.69

0.98

1.13

Day - 2

1.46

1.30

1.06

1.43

1.59

1.38

1.66

1.50

1.59

0.86

Day - 1

4.46

3.62

3.08

4.44

3.77

6.34

4.83

4.59

2.05

2.36

Day - 0

2.75

2.00

2.24

2.70

3.08

3.27

1.84

2.06

1.69

1.72

Week 0

1.86

1.39

1.70

1.58

1.51

1.84

1.33

1.59

1.49

1.26

Week 1

1.56

1.25

1.12

1.49

1.15

1.31

1.07

1.35

1.38

1.11

Week 2

1.58

1.21

1.24

1.25

1.40

1.18

1.05

1.30

1.40

1.18

Week 3

1.23

1.10

1.32

1.12

1.14

1.34

1.01

1.09

1.37

1.04

Week 4

1.15

1.02

1.13

1.10

1.04

1.31

0.86

1.18

1.33

1.13

Week 5

1.22

0.98

1.12

0.93

1.36

0.79

0.91

1.06

1.15

0.88

Week 6

1.58

0.96

1.25

1.14

1.10

0.87

1.04

1.05

1.32

1.02

Week 7

1.26

0.86

1.80

1.00

0.99

0.89

1.15

1.23

1.34

0.97

Table 7: The trading volume response of the additions according to industries (2000 – 2010)

Consumer Discretionary

Consumer Staples

Energy

Financials

Industrials

Information Technology

Materials

Support Services

Telecom. Services

Utilities

Week -10

1.07

1.02

1.00

0.91

0.82

1.11

0.85

0.85

0.95

0.75

Week -9

0.94

1.14

1.12

1.07

0.94

0.82

0.79

0.86

1.13

1.29

Week -8

0.99

1.25

1.00

1.00

0.97

0.92

0.92

0.93

0.85

1.13

Week -7

1.06

0.94

0.82

0.96

0.90

1.03

0.87

1.00

0.91

1.01

Week -6

0.90

0.93

0.66

0.95

0.86

1.04

1.19

0.86

1.00

0.99

Week -5

1.01

1.03

1.44

1.03

0.92

0.89

1.16

1.03

0.99

0.94

Week - 4

1.08

0.86

1.04

1.05

1.34

1.37

1.14

1.20

0.87

0.74

Week - 3

1.10

1.02

0.99

1.09

1.29

0.85

1.18

1.31

1.20

1.29

Day -10

1.07

1.16

1.12

1.49

0.83

0.80

0.85

1.49

2.10

1.12

Day - 9

1.41

1.25

1.75

1.41

1.03

0.82

1.17

1.25

1.07

1.18

Day - 8

1.19

1.37

1.17

1.71

1.51

0.79

1.21

1.04

1.00

1.34

Day - 7

1.00

1.32

1.07

1.28

1.27

0.66

1.49

1.25

1.08

1.57

Day - 6

1.25

1.56

1.01

1.26

1.17

0.79

1.31

1.06

1.14

1.04

Day - 5

0.91

1.22

1.42

1.23

1.05

0.84

1.08

1.17

0.92

1.77

Day - 4

1.18

1.63

1.04

1.16

1.40

0.62

1.56

1.27

1.27

1.16

Day - 3

1.20

1.54

1.00

1.35

1.30

0.74

1.09

1.64

1.02

1.19

Day - 2

1.43

1.52

1.28

1.48

1.44

0.84

1.93

1.39

1.14

1.57

Day - 1

2.51

4.18

3.02

4.61

3.29

1.96

3.05

2.92

2.06

2.70

Day - 0

1.71

3.01

0.88

2.14

2.10

1.55

1.58

2.91

1.02

2.04

Week 0

1.17

1.51

1.03

1.34

1.26

1.06

1.45

1.28

1.16

1.34

Week 1

1.09

1.51

1.19

1.06

1.13

1.18

1.39

1.07

0.95

1.04

Week 2

0.92

1.28

1.00

1.12

1.05

1.43

2.39

0.92

0.97

0.96

Week 3

0.90

0.93

0.81

1.12

0.90

1.38

1.13

1.12

1.32

1.11

Week 4

0.84

1.02

0.72

1.01

0.94

1.39

0.94

0.93

0.90

0.94

Week 5

0.74

0.87

0.86

0.88

0.98

1.14

1.12

1.17

0.79

0.76

Week 6

0.87

0.90

0.78

0.75

1.16

1.07

1.04

0.98

1.22

0.74

Week 7

0.95

1.17

0.68

0.86

0.99

1.05

1.00

1.05

0.96

0.86

Table 8: The trading volume response of the deletions according to industries (2000 – 2010)

As can be seen from the Table 7 and Table 8 above, we find that, on average, trading volume of both added and deleted industries increase over the pre-event window and reverse over the post-event window. Especially, the trading volume of the day before the effective date and the effective date increase at highest levels. The results are consistent with those obtained in the Section VI.1.b when I examine all added and deleted firms in general. This is consistent with the liquidity hypothesis.

For the added industries, Information Technology and Utilities has the most and the least increase. This means their trading volumes are respectively the most and the least sensitive industries to the inclusions to the FTSE 100 index composition. On the day before the effective date and on the effective date, the trading volume of Information Technology industry also has the highest increasing level (respectively 6.32 times and 3.27 times as large as the daily mean trading volume over the 8 weeks prior to the announcement).

On the other hand, for the deleted industries, Consumer Staples are the most increased industries while Information Technology industry has the lowest level. From this, we can conclude that their trading volumes are the most and the least sensitive industries to exclusions from the FTSE 100 index composition. Consumer Staples is also the industry that has the highest increasing level on the day before the effective date and on the effective date (4.18 times and 3.01 times respectively).

Furthermore, there is an evidence of the relationship between the price and trading volume. When compare the trading volume response with the price response, I find exciting results. For the additions, the most profitable industry is also the industry that has the highest increasing ratio in the trading volume (Information Technology industry). For the deletions, the trading volume of industry that is the least declined industry in price increases at the highest level (Consumer Staples industry).

V.3 The response of ‘nearly in’ and ‘nearly out’ firms

V.3.1 The price response of ‘nearly in’ and ‘nearly out’ firms

The equation (1) will be applied for the data of the additions, deletions and ‘nearly in’, ‘nearly out’ firms in the period from 2005 to 2010. In order to analyse the response and compare with those of the additions and deletions, we will plot the results in the Figures 14 and Figure 15 below:

Figure 14: Cumulative abnormal returns of additions and “nearly in’ stocks around the announcement and effective date (2005 – 2010)

Figure 15: Cumulative abnormal returns of deletions and “nearly out’ around the announcement and effective date (2005 – 2010)

The ‘nearly in’ and ‘nearly out’ firms are termed by Mase (2007). They refer to the firms that just avoid being deleted and just jail to be added to the FTSE 100 index composition. According to Mase (2007: 479), “An examination of trading in nearly out and nearly in stocks reveals the extent to which traders attempt to anticipate the announcement”. He also stated that, in principle, nearly stocks should be unaffected by the announcement because they do not include to or exclude from the index. As above-mentioned, I would like to examine the price and trading response in the period from 2005 to 2010 in order to find the latest trends.

From the Figure 14 and 15, the Cumulative abnormal returns of the ‘nearly in’ (‘nearly out’) firms almost have the same trend as those of the additions (deletions) in the pre-announcement window but they do not have abnormal return after the announcement. This is because they are not added to or deleted from the FTSE 100 index composition. However, the magnitudes of CARs of the ‘nearly’ stocks are much smaller than those of the added or deleted stocks. This may be due to the trading behavior of index funds that buy (sell) the added (deleted) stocks to mimic the performance of the index. They do not make effects on the ‘nearly’ stocks.

In order to estimate the magnitude of the CARs, I would like to apply the equation (2). Because there is no effective date for the ‘nearly’ stocks, I only examine the price effect in the pre-announcement, post-announcement and complete event windows. The results are presented in the Tables 9 and 10 below.

The results in the Tables 7 and 8 below confirm the implication of the CARs shown in the Figures 13 and 14. In all windows, the CARs of the ‘nearly’ stocks are intensively less than those of the additions and deletions. For example, in the pre-announcement window, the mean CAR of the additions (deletions) is 6.4% (-7.1%) while that of the ‘nearly in’ (‘nearly out’) stocks is only 2.2% (-1.2%). However, from this point, we can see that investors are attempting to anticipate these announcements.

Furthermore, at a significant level of 5%, the mean CARs of the ‘nearly in’ stocks are significantly positive in the pre-announcement window and insignificantly different from 0 in the post-announcement and complete event windows while those of the ‘nearly out’ stocks are not significantly different from 0 in all these windows. This means investors may attempt to

Table 9: The price effects of the added and ‘nearly in’ stocks around the announcement and effective date (2005 – 2010)

Additions Nearly in

α -0.000 -0.000

(-0.24) [0.57] (0.20) [0.10]

β 1.103 1.012

(26.11) [13.46] (25.42) [19.91]

M𝛾1 0.064 0.022

(7.74) [5.72] (4.73) [2.97]

M𝛾2 0.025 -0.002

(5.37) [2.98] (-0.38) [-0.08]

M𝛾3 -0.024 -0.006

(-3.95) [-1.77] (-1.43) [-0.84]

M𝛾4 -0.033 -0.014

(-2.66) [-2.16] (-1.66) [-0.87]

PA -0.031 -0.021

(-1.92) (-1.75)

PRE 0.089

(11.76)

PE -0.056

(-3.31)

CEW 0.033 0.001

(1.63) (0.07)

The t-statistics in parentheses are ;

The statistics in brackets are

Table 10: The price effects of the deleted and ‘nearly out’ stocks around the announcement and effective date (2005 – 2010)

Deletions Nearly out

α -0.000 -0.000

(-0.24) [1.00] (-1.96) [-0.88]

β 1.027 1.042

(32.30) [16.71] (32.60) [18.10]

M𝛾1 -0.071 -0.012

(-6.89) [-3.20] (-1.90) [-0.59]

M𝛾2 0.017 0.004

(1.79) [-0.24] (1.01) [1.05]

M𝛾3 -0.012 -0.003

(-1.52) [-1.03] (-0.70) [0.01]

M𝛾4 -0.005 0.006

(-0.20) [-0.20] (0.59) [0.09]

PA -0.000 0.007

(-0.01) (0.49)

PRE -0.054

(-5.03)

PE -0.016

(-0.57)

CEW -0.071 -0.005

(-1.97) (-0.25)

The t-statistics in parentheses are ;

The statistics in brackets are

anticipate additions rather than deletions. This is consistent with findings by Mase (2007) contrasts with findings by Malic (2006) who found that “Investors focus more on deleted stocks than added stocks because they are more worried about losing money on deleted stocks than making money on the added one”.

V.3.2 The trading volume response of nearly in’ and ‘nearly out’ firms

As can be seen from the Table 9 and 10 below, the trading volume of the ‘nearly in’ and ‘nearly out’ stocks are also increase compare with the daily mean volume over the 8 weeks prior to the announcement. This reveals that there is anticipatory trading associated with the ‘nearly’ stocks. However, the increasing levels are substantially less than those of the additions and deletions. Especially, the trading volume of ‘nearly’ stocks decrease slightly on the t – 1 and t while those of the additions and deletions increase at the highest level. Furthermore, after announcements are released (day t – 7), trading volume of additions and deletions dramatically increase while those of ‘nearly’ stock slightly go up. These are because the ‘nearly’ stocks are not added to or deleted from the index, so they are not affected by trading behaviors of index funds. On day t and t – 1, the index fund and investors pay much more attention to the added and deleted stocks than the other stocks.

In addition, when compare the ‘nearly in’ stocks with the ‘nearly out’ stocks, I find that the trading volume response of the former is higher than that of the latter. This may indicate that anticipatory trading occur in the ‘nearly in’ stocks rather than in the ‘nearly out’ stocks. This is similar to the finding of the price response as presented above. This is also consistent with the result by Mase (2007).

Table 11: The trading volume response of the additions and ‘nearly in’ firms around the announcement and effective date (2005 – 2010)

Period

MVR

STD

t-statistic

Percent > 1

Added

Nearly in

Added

Nearly in

Added

Nearly in

Added

Week - 8

1.03

0.99

0.44

0.20

0.51

-0.28

40

Week - 7

0.97

1.03

0.27

0.20

-0.81

1.41

42

Week - 6

1.02

1.01

0.23

0.09

0.52

0.96

58

Week - 5

0.96

0.99

0.20

0.14

-1.24

-0.43

29

Week - 4

1.00

0.95

0.21

0.23

-0.10

-1.89

37

Week - 3

1.05

1.00

0.21

0.20

1.64

-0.13

69

Week - 2

0.99

0.99

0.21

0.21

-0.33

-0.37

54

Week - 1

1.04

1.09

0.26

0.27

1.09

2.54

62

Day - 10

1.27

1.14

0.68

0.42

2.92

2.66

56

Day - 9

1.32

1.08

0.51

0.35

4.55

1.93

75

Day - 8

1.37

1.08

0.50

0.37

5.42

1.80

83

Day - 7

1.29

1.14

0.42

0.38

4.95

2.95

71

Day - 6

1.21

1.06

0.28

0.37

5.32

1.37

75

Day - 5

1.31

1.08

0.26

0.40

8.68

1.55

88

Day - 4

1.35

1.25

0.44

0.81

5.72

2.51

71

Day - 3

1.40

1.23

0.45

0.59

6.42

3.10

65

Day - 2

1.45

1.04

0.60

0.32

5.39

0.97

75

Day - 1

4.57

0.95

2.33

0.31

11.06

-1.25

100

Day - 0

2.21

0.99

1.03

0.41

8.51

-0.19

100

Week 1

1.56

1.09

0.61

0.29

6.64

2.48

92

Week 2

1.33

1.11

0.43

0.28

5.46

3.17

73

Week 3

1.26

1.15

0.46

0.36

4.04

3.41

67

Week 4

1.08

1.09

0.29

0.24

2.05

3.22

62

Week 5

1.01

1.16

0.25

0.41

0.19

3.15

50

Week 6

1.02

1.17

0.29

0.31

0.44

4.37

52

Week 7

1.02

1.23

0.28

0.33

0.58

5.66

54

Week 8

1.13

1.15

0.43

0.21

2.11

5.83

63

MRV is the mean volume ratio and indicate how many times the trading volume of the additions increases compared with the daily mean volume over the 8 weeks (week – 3 to week – 10) prior to the announcement.

Table 12: The trading volume response of the deletions and ‘nearly out’ firms around the announcement and effective date (2005 – 2010)

Period

MVR

STD

t-statistic

Percent > 1

Deleted

Nearly out

Deleted

Nearly out

Deleted

Nearly out

Deleted

Week - 8

0.94

1.01

0.18

0.25

-2.21

0.34

23

56

Week - 7

0.96

0.98

0.28

0.22

-0.91

-0.77

36

48

Week - 6

1.03

1.04

0.19

0.25

0.96

1.22

60

44

Week - 5

1.01

1.04

0.22

0.23

0.39

1.48

47

44

Week - 4

0.92

0.98

0.17

0.19

-3.70

-0.92

23

37

Week - 3

1.01

0.98

0.20

0.21

0.46

-0.74

47

33

Week - 2

1.03

1.06

0.26

0.22

0.71

2.29

49

51

Week - 1

1.18

0.96

0.23

0.18

5.85

-1.81

89

33

Day - 10

1.40

1.06

0.76

0.50

3.80

1.01

75

41

Day - 9

1.38

1.17

0.55

0.49

5.10

2.77

77

59

Day - 8

1.34

1.05

0.51

0.36

4.78

1.13

79

43

Day - 7

1.32

1.13

0.49

0.50

4.76

2.08

72

52

Day - 6

1.30

1.04

0.48

0.39

4.46

0.76

77

32

Day - 5

1.18

1.18

0.35

0.43

3.71

3.30

72

62

Day - 4

1.29

1.09

0.60

0.33

3.53

2.14

70

49

Day - 3

1.34

1.07

0.53

0.38

4.66

1.46

72

46

Day - 2

1.59

1.07

0.73

0.39

5.90

1.37

87

44

Day - 1

3.54

0.90

1.87

0.49

9.91

-1.62

100

21

Day - 0

1.99

1.00

1.55

0.29

4.66

0.09

94

43

Week 1

1.34

1.06

0.32

0.27

7.96

1.68

94

46

Week 2

1.21

1.05

0.33

0.26

4.62

1.46

74

56

Week 3

1.33

1.05

0.69

0.32

3.45

1.26

68

46

Week 4

1.13

1.02

0.36

0.25

2.61

0.63

55

51

Week 5

0.96

0.95

0.30

0.24

-1.03

-1.55

34

37

Week 6

0.97

0.95

0.26

0.25

-0.90

-1.46

47

33

Week 7

0.95

1.00

0.26

0.31

-1.38

-0.11

32

30

Week 8

1.00

0.95

0.28

0.23

-0.06

-1.88

47

41

MRV is the mean volume ratio and indicate how many times the trading volume of the additions increases compared with the daily mean volume over the 8 weeks (week – 3 to week – 10) prior to the announcement.

Chapter VI. CONCLUSION

This research is conducted base on the data of additions and deletions in the period from 2000 – 2010 through the event study methodology. The reason I choose this topic because of stock is added or deleted from the FTSE 100 is an important event for investors and firms. This event shows the firm’s performance, financial healthy, perspective and so on. Moreover, this study of this event will support investors, companies can get helpful information to find suitable investments, avoid suffering from losses. The examination is limited in the event window from 20 days before the effective date and 30 days from the effective date. Otherwise, some previous studies were reviewed in this research to have a more general view; They are studies from Gregoriou and Ioannidis (2003) shows that investors who hold stocks with more (less) available information, resulting to them have lower (higher) trading cost. Mase (2007) pointed out abnormal returns and trading volumes of “nearly in” and “nearly out” firms are less than those gained for additions and deletions. From the research, Hamill et al (2004) found out that this relatively large price change can be accounted for by index funds buying (selling) stock immediately prior to their additions to (deletion from) the FTSE 100 to minimize tracking error. Shleifer (1986) found a significant abnormal post-announcement return of just under 3% for firms added to the S&P 500 index after 1976. There are many other studies about the additions and deletions from FTSE 100 but I have not ever found any studies about the response to changes in the FTSE100 from 2000-2010. Therefore, it is motivate to help me conduct this research in order to contribute analysis from many other resources.

According to the examine above, I found some evidences of the price pressure hypothesis, the long term downward sloping demand curve hypothesis and, the certification (information) hypothesis, the liquidity hypothesis. The price response: The findings of the deletions are nearly opposite with those of the additions. Especially, when considering abnormal returns (ARs) for each day in the event window, I find that the AR is highest on the day before the effective date (t – 1) and lowest on the effective date. Moreover, I indicate that the reversal of the price and trading volume of the added (deleted) firms after the effective date means that the abnormal returns or trading volume is temporary. They are consistent with studies of Mase (2007), Haris and Gurel (1986), Shleifer (1986). However, it is opposite with those of Lynch and Mendenhall (1997) who found that pre-announcement MCARs are negative for the added firms and positive for deleted.

The heavy short-term trading (mainly by index funds) moves price of the additions and deletions temporarily away from equilibrium. The price increases in the pre-event window and reverses in the post-event window. There is also abnormal trading volume around the event dates which supports the liquidity hypothesis. I also find the extremely abnormal returns and trading volume on the day before the effective date and on the effective date. This is an evidence of trading from index fund on these 2 days and relationship between the prices and trading volume. When examine the additions and deletions according to industries, I find that the results from almost industries are consistent with the price pressure hypothesis and liquidity hypothesis. The Telecommunication Services is the most sensitive to the changes. For the additions, the most profitable industry is also the industry that has the highest increasing ratio in the trading volume (Information Technology industry). This confirms the relationship between the price and trading volume.

Finally, I examine stocks that are nearly added to or deleted from the FTSE 100 index composition – termed ‘nearly in’ and ‘nearly out’ stocks respectively by Mase (2007). I find that the magnitude of the price and trading volume response of the ‘nearly’ stocks is clearly less than those of the additions and deletions. The ‘nearly in’ stocks show abnormal returns in the pre-event window while the ‘nearly out’ stocks do not. This indicate that there are anticipatory in the ‘nearly in’ stocks rather than in the ‘nearly out’ stocks. Moreover, there are no evidence of abnormal returns and trading volume of ‘nearly’ stocks in the post-announcement window.

In conclusion, I hope that the results of this research will contribute to the previous studies and give market participants an overview of impacts of the changes so that they can make right and timely investment and management decisions. Moreover, I hope this research may be one of references for someone who wants to continue the study of response to changes in FTSE 100 index in the future. To expand on this research, it is beneficial to examine the long-term (one year or three years) performance of the additions and deletions. This will enable us to estimate the implicit value of being added to or deleted from the FTSE 100 index composition.

Print Email Download Reference This Send to Kindle Reddit This

Share This Essay

To share this essay on Reddit, Facebook, Twitter, or Google+ just click on the buttons below:

Request Removal

If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal:

Request the removal of this essay.


More from UK Essays

Doing your resits? We can help!