1
Effects of index reviewing on liquidity: An event studies on the AEX-index.
Abstract
This thesis studies the effects of the reviewing of the AEX index on liquidity based on a
sample of index additions from 2003 to 2015. The addition of a stock to a large index can
affect liquidity in different ways. Using an event study, three different periods are tested to
be able to separate these different effects. Evidence is found for an announcement effect
with increased liquidity for five days after the announcement. Furthermore I find a large but
temporary increase in liquidity around the effective addition. Lastly, I find a permanent
increase of liquidity over a 50 day period after the addition to the Index.
JEL Classification: G10
29-06-2016
Economics and Finance
Olivier Lieshout, 10406638
Olivier.Lieshout@student.uva.nl
R. Sperna Weiland, Supervisor
2
Index
1. Introduction ... 3
2. Theorethical Framework ... 4
2.1 Theoretical effects and empirical evidence ... 4
2.2 Predictions and hypothesis ... 7
3. Methodology and Data ... 8
4. Results ... 10
5. Conclusion ... 13
6. Bibliography ... 14
Statement of Originality
This document is written by Student Olivier Lieshout who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
1. Introduction
Large and well known indices like the S&P 500 and the Dow Jones are reviewed over time and this results in companies being replaced or inserted. The addition of a stock to an index could influence the interest of investors and analysts in the stock. The effect of an addition to a large index on the price has been studied extensively, these studies have found that a newly added stock has excess return compared to the market.
One of the hypotheses on what causes this price effect is the liquidity hypothesis (Amihud & Mendelson, 1986). If stocks get more liquid when added to an index this should have a positive effect on the price of a stock. The effects of an addition on liquidity have been researched and evidence for a positive effect on liquidity has been found. However, most of this research focuses on a short period around the addition since in this period the returns are studied. There is no consensus whether the excess liquidity is permanent or temporary, due to the lack of research performed. Beneish and Whaley (1996), for example, find no evidence for a permanent increase in liquidity for additions to the S&P 500. Hedge and McDermott (2003) on the other hand find that there is a significant permanent increase in liquidity.
In this research I use an event study approach to research three different effects of addition to a large index. First the effect of the announcement of the addition, second the effect of the effective addition and third a permanent effect over a longer term. Additions to the AEX-index from 2003 to 2015 are used to study these effects. The AEX-index is an index consisting of the 25 largest listed companies in the Netherlands. This index is an interesting index for this research due to the difference in announcement and reviewing policies compared with other index such as the S&P 500. Firstly I find that liquidity of a stock, as measured by the Amihud measure, is positively affected by the announcement of addition to an index. Secondly, I find that the effective addition creates a large temporary increase in liquidity. Thirdly, I find evidence for a permanent increase in liquidity by finding a significant increase in liquidity for a longer period after addition to an index.
The remainder of this thesis is organised as follows; In section 2.1, I discuss the relevant literature on index inclusion. Based on these previous studies, I present my hypotheses on the effects of index inclusion on liquidity in section 2.2. Section 3 explains the methodology and data used to study the effects on liquidity. In section 4 the results of the tests are discussed. The conclusions and recommendations for further research are presented in section 5.
4
2. Theoretical Framework
The term liquidity is used for many different aspects of financial markets. In this research the focus is on market liquidity, which can be seen as the ease with which an asset can be bought or sold (Brunnermeier & Pedersen, 2009). Market liquidity has a time, size and price dimension (Lo, Petrov & Wierbicki, 2003). These aspects of liquidity can be important to investors. Liquidity is high if a stock can be traded quickly, in large volumes and without an impact on the price (de Jong & de Roon, 2011). These different aspects of liquidity are strongly linked. If no counterparty is available at the time of the trade, liquidity is low. To make sure the trade is executed quickly an investor will have to lower his price. De Jong and de Roon (2011) state that therefore the price effect of liquidity combines the different dimensions. The order type used by an investor shows this intuition (Lynch and Mendenhall, 1996). A market order, an order that is executed immediately at the best available price, minimises the time aspect of liquidity. This order type has the possibility to be executed at a higher price than the last traded price because no better price was available at that moment. Conversely a limit order, an order which is executed at a set price, will only be executed as soon as this there is a counterparty for this price. This lowers the liquidity in the time dimension. A Limit order supplies liquidity on a market and a market order consumes the liquidity (Goyenko, Holden & Trzcina, 2009). If a limit order is used other investors are able to buy or sell against that price and therefore it creates liquidity.
2.1 Theoretical effects and empirical evidence
Extensive research has been done on the effect of a stock’s addition to an index on the returns of a stock. A positive effect of an addition to an index is well documented in the literature (Dhillon & Johnson, 1991; Beneish & Whaley, 1996; Hedge and McDermott, 2003). Various theories are used to explain this effect.
Harris and Gurel (1986) explain this effect with the price pressure hypothesis. They argue that the effect is temporary and caused by index funds who have to rebalance their portfolio. Sheiffer (1986) on the other hand finds that the price effect is permanent and attributes this to the imperfect substitution hypothesis. He argues that the demand curves of stocks are downward-sloping, which can explain a permanent price increase if more investors are interested in a stock after addition. On the other hand Dhillon and Johnson (1991) and Jain (1987) find evidence that the addition to the S&P 500 includes new information, which can be seen as positive signalling. The information hypothesis is contradicted by the evidence presented by Harris and Gurel (1986) and Sheiffer (1986). Amihud & Mendelson (1986) argue in favour of the liquidity hypothesis, which states that the addition of a
5
stock to a big index has a positive effect on the liquidity of a stock. Investors will therefore value the stock at a higher price.
The liquidity of the stock might be affected by different forces since liquidity has multiple dimensions. First of all, being part of a well known index will influence the amount of index investors (Hrazdil, 2009). Index investors are investors that have as main goal to track an index closely. Most of these investors are investing in index funds which track the index for them (Hrazdil, 2009). When a stock is added to a new index these index funds will need to rebalance their portfolio. They have to include the new stock and exclude the stock that gets delisted from the index. To minimize their tracking error, these trades will all be done close to the effective date of the change (Lynch and Mendenhall, 1996). This excess trading in the newly added stock lowers the time needed to find a counterparty which yields a positive effect on liquidity (Hrazdil, 2009; Lynch and Mendenhall, 1996). Lynch and Mendenhall (1996) find evidence that there is a permanent excess trading volume after the announcement day. However the liquidity only increased temporary, and lasted up to ten days. This shows that a higher trading volume does not immediately imply more liquidity. They argue that this temporary increase in liquidity can be a result of a high amount of limit orders used by index funds. They also argue that market makers are temporary lowering their bid-ask spread because of the high volumes traded around the effective date. Beneish and Whaley (1996) likewise find a permanent increase of volume for S&P 500 additions from 1986 to 1994. Liquidity on the other hand just increases for a couple of days after the addition. They conclude that this is because in 10 days after adding index funds are done rebalancing their portfolio, which will then lower liquidity again. Hedge and McDermott (2003) believe a change in ownership structure can cause a permanent increase of a stock’s liquidity. They argue that addition to a bigger index permanently attracts more non-informed investors. These ‘non-informed investors’ do not try to beat the market by speculating on news and other developments and mainly use the stock for diversification. Hedge and McDermott (2003) believe this group improves the liquidity of a stock, since they will not buy the stocks based on news events. They state that non-informed investors are likely to use limit orders instead of market orders and therefore increase liquidity. If a market order is used, the trade could become more expensive. Hedge and McDermott (2003) study additions to the S&P 500 from 1993 to 1998 to see if there is an increase in liquidity. In their studies on a permanent effect of liquidity they do not include the 10 days after and before the effective date to make sure the result was not biased due to index funds’ rebalancing. They find that not only the volume changes permanently but also the liquidity. The standardised trading volume increased by 27% on average and the average effective quoted bid-ask spread dropped by 16.19%.
6
An increase of available information can also have a positive impact on liquidity. One of the reasons more information is available is the increase in analysts following the company (Easley, O’Hara and Paperman, 1998). These analysts lower asymmetric information, which enable investors to more accurately estimate the value of a stock. More precise information enlarges the liquidity via the price dimension, since with less asymmetric information the bid-ask spread will be lower (Brennan and Subrahmanyan, 1995). Roulstone (2010) finds that more coverage by analysts has a positive effect on the liquidity of a stock. Lower information asymmetry can also be caused by more arbitrageurs watching the index. Kumar, Sarin and Shastri (1998) find that stocks being part of a big index on which option trading is possible will have a higher liquidity on average, because stock prices will be a better estimation of the value of a company.
In this thesis I will use data on the AEX-index additions to study the effects of index inclusion. Since most studies have been done using data from the additions to the S&P 500 it is important to analyse in what way the S&P 500 and the AEX differ and what kind of effects this might have on the hypotheses.
Both the S&P 500 and the AEX base their index on the size, measured in free flow market capitalisation of a stock (Euronext, 2016; Standard and Poor’s, 2016). The two also have the same goal as an index, namely tracking the biggest companies in their country of interest. What also makes them comparable is that they have a limited size. The AEX is limited to 25 companies whereas the S&P 500 always consists of 500 companies.
Nevertheless their reviewing policies differ. The S&P 500 is updated on an as-needed basis (Standard & Poor’s, 2016). When a new stock gets a place in the index it will be announced five trading days before the actual change. Euronext on the other hand uses set dates to update the index (Euronext, 2016). They normally only change the index on the third Friday of March, June, September or December. The third Friday of March is the only moment a stock from the AMX can be added to replace another stock. The other dates are used to refill the index if there is an empty spot caused by a merger or a bankruptcy. In these quarterly updates there is also a possibility for newly listed companies to take place in the AEX if they are big enough.
Furthermore the announcement policies are different. A review of S&P 500 is always announced 5 days before the effective date (Standard and Poor’s, 2016). Euronext announces the addition of a stock earlier. Euronext has no set announcement date, but always announces the addition between 10 to 20 trading days before the addition (Euronext, 2014).
7
The set dates of the addition and the clear reviewing policy used by Euronext make it possible for risk arbitrageurs to enter. Beneish and Whaley (1996) state that most index funds will wait to a moment close to the effective date to rebalance their portfolio. They do this to minimise their index tracking error. Risk arbitrageurs will buy the stocks after the announcement date and wait for an increase in price caused by index funds’ rebalancing. Beneish and Whaley (1996) find that the five days after the announcement day the liquidity is higher. They argue that this is mainly because of risk arbitrageurs entering the market. Hedge and McDermott (2003) also studied the effects the announcement on liquidity. They also find that for the four days after announcement the liquidity was higher than before announcement.
An important difference between the two indices is the predictability of an addition to the index. Since the S&P 500 does not use set reviewing days it will be hard for arbitrageurs to enter before the announcement day (Hedge & McDermott, 2003). Chan and Howard (2002) study the announcement effect on the AIO, the Australian index which tracks the biggest companies. Additions to the AIO are more predictable than the additions to the S&P 500 since they use clearly described rules for addition (Chan and Howard, 2002). They find a significant announcement effect for the AIO, even though there might be a better possibility to predict an addition. Chan and Howard (2002) also study long term effect volume for additions to the AIO. They find permanent increase in trading volume after the addition.
Brealey (2000) studies the effect of an addition to the FTSE 100 and the FTSE All-Share. He measured the excess returns on newly added stock up to 10 days before and after the effective date. He found a positive effect on price but not statistically and economically significant. Chan and Howard (2002) argue that this is because these additions were predictable and the number of days measured was too low to find significant results for this market.
2.2 Predictions and hypothesis
The effects on liquidity described in the theory have different timeframes in which they could affect liquidity. I will make a distinction between three timeframes, namely the days just after announcement, the days just before and just after the effective day of addition and a longer period after the addition.
For the first period, the days just after the announcement, a positive effect on liquidity has been found (Beneish and Whelay, 1996; Hedge & McDermott, 2003). If arbitrageurs act in a similar way as has been found for the S&P 500 one would expect a similar effect for the AEX. Nevertheless this is interesting since the predictability of a stock’s addition to an index is higher for indices with a similar
8
reviewing policy to the AEX (Chan and Howard, 2002). If risk arbitrageurs enter the market after the announcement day this should have a positive effect on liquidity. Based on the findings by Chan and Howard (2002) who distinguished a significant effect after the day of addition I expect to find a similar effect on the liquidity.
The days just before and just after the addition are the second period of interest. Around this time the index funds will start rebalancing their portfolios. The increased interest in the stock, combined with arbitrageurs selling their positions has been shown to significantly increase liquidity (Lynch and Mendenhall, 1996; Hrazdil 2009). Also Hedge and McDermott (2003) find that in this period the liquidity is higher than before the announcement. This increased liquidity is mainly caused by the rebalancing of index funds has a temporary impact on the liquidity (Lynch and Mendenhall, 1996). The AEX in that matter should not differ too much from the S&P 500. Index funds try to minimise their tracking error and therefore will do this close to the effective date of addition (Lynch and Mendenhall, 1996). Therefore I expect that for the AEX a similar movement can be seen. In the days just around the moment of addition the liquidity should be higher than before the announcement day.
The last period of interest is the period after the rebalancing by index funds. The short term shock of rebalancing should not be effecting liquidity anymore. The literature describes three significant long term effects liquidity. First an increase in analysts and public interest in the company will lower the information asymmetry. Second a change in ownership structure with a bigger group of non-informed investors who buy the stock for diversification. Third the increase of index arbitrageurs watching the stock which also lowers information asymmetry. Hedge and McDermott (2003) find a significant increase in liquidity measured after the addition. They left out the first 10 days after addition to make their sample not biased because of index funds rebalancing their portfolios. Since these three influences should also be applicable to the AEX I expect to find an increase in liquidity compared to the period before announcement.
3. Methodology and Data
Since liquidity has different dimensions there are many different ways to measure it. One of the most common measurements to use is the bid-ask spread as a measurement for liquidity. This measurement is often used because it directly measures the price impact of an immediate transaction. When the bid-ask spread is larger this will cause the price impact to be higher and therefore it can be used to measure liquidity (Goyenko, Holden & Trzcina, 2009). Limit orders supply
9
liquidity since these orders give prices for which they are willing to buy. The other way around a market order widens the bid-ask spread and therefore lowers liquidity.
Goyenko et al. (2009) find that using the bid-ask spread is a good measurement for liquidity but it also has some downsides. First of all the measurement requires a lot of data that is not widely available. To use this way of looking at liquidity one needs all the data from the exchanges on which orders were sent to the exchange, and at what price every trade was executed. This data is not often supplied and therefore only some were able to use this measurement.
Goyenko et al. (2009) study the ability to use more widely available data to measure liquidity. One of their main findings was that the price impact of liquidity can be measured using the liquidity measurement introduced by Amihud (2002).
Amiit=
Where Retit is the stock i’s return on day t and Volit is the stock i’s volume, measured in the currency the stock is trading, on day t.
The Amihud liquidity measure uses data which is more available to researchers, and will be able to work as a good measurement for liquidity. The price impact of a trade can summarise the time, size and price dimension of liquidity (Goyenko et al., 2009). A value of the Amihud measure shows the change in return per unit traded of the currency the stock is listed in. If this value is low the effect of selling or buying a stock on the price are low. So if the stock’s liquidity increases when added to the AEX the value of the Amihud measure should decrease.
To study if the addition to the AEX-index has a direct impact on liquidity I will follow the methodology used by Hedge and McDermott (2003). The methodology is based on an event studies and the events used are the day of announcement and the day of addition.
Three different timeframes will be used to test the three hypothesis. To test whether the announcement of an addition to the AEX increases liquidity the five trading days after the announcement will be used. This period is used because within this period index arbitrageurs will be active (Beneish and Whaley, 1996). To measure if there is a significant short term impact on liquidity I will use a sample containing the 4 days before the addition, the day of addition and the 4 days after addition. This timeframe is used because within this period index funds will be rebalancing their portfolios (Lynch and Mendenhall, 1996). A sample consisting of 50 trading days starting after the
10
10th day after addition is used to measure whether an addition to the AEX yields a lasting positive effect on liquidity.
To be able to compare the liquidity after the event the average liquidity over 50 trading days before announcement is used. The five trading days before announcement are omitted to reduce the possibility of biased outcomes. This value is used to compute a ratio that shows the liquidity on a day compared to the pre-announcement liquidity. This ratio is the level of liquidity on a day divided by the average liquidity before announcement. For each company in the samples used daily ratio’s are computed to test the liquidity after the events. Following Hedge and McDermott (2003) this is done to level out the differences in liquidity level for each company. Otherwise a stock with a higher liquidity can bias the findings. The ratios of all days and all companies are combined. The mean of all these ratio’s all together should not be different from 1 if liquidity stays the same after addition, since the pre-announcement liquidity is used to compute the ratio. To test whether the sample mean differs from 1, a T-test for one sample is done. If the liquidity did not change significantly during the period of interest the mean of the ratio is expected to be 1.
In the period from 2003-2015 25 companies have been added to the AEX index. Not all of these are useful for this research and two of them are omitted. The addition of Unibail-Rodamco to the AEX in 2007 is omitted because the day of addition was on the same day as the IPO. Therefore no earlier data was available and no comparison was possible. The addition of Tele Atlas is omitted because shortly after the addition TomTom made a takeover bid and later on acquired Tele Atlas. A takeover can bias the data and should be therefore left out (Hedge and McDermott, 2003). Data on the opening price, the closing price and the trading volume on these stocks is retrieved via Datastream. This is the data needed to compute the Amihud measure.
4. Results
The results of the tests on the three different periods are presented in Table 1. The three columns show the results of the three different timeframes tested. The first column shows the results of the test on announcement effects. The mean of the ratio dropped significantly for this period. This period has the least observations of the three periods of interest, because the sample consists of just the five days after announcement for the 23 companies. The mean of the ratio for this period is .7986, which shows that the Amihud liquidity measure has decreased. A decrease in the Amihud measure shows that the stock’s liquidity increased. Although this increase in liquidity could be expected based on the theory based on the S&P 500 it is an interesting finding. The announcements
11
for the AEX are more predictable than for the S&P 500 as described earlier. This increase in liquidity shows that at least not all investors speculate on the addition of stocks. Risk arbitrage is the most likely reason for this increase in liquidity. These investors will try to get positive returns after the effective day of addition and therefore trading increases after the announcement. The confidence interval for this period is rather wide, caused by a relatively high standard error. The standard error for the other periods is lower because these samples consist of more days. Even with limited observations caused by the limited amount of companies added to the index the effect of announcement significant.
Table 1 – Results of the three different periods
Period [AD, AD+5] [ED -4,ED +4] [ED+10, ED+60]
Mean of Ratio .7986*** .6258*** .9260**
Std. Error Mean .06317 .04296 .03307
T-Value -3.188 -8.711 -2.238
N (Total amount of ratios) 115 207 1150
Companies with Ratio>1 17 out of 23 20 out of 23 20 out of 23
% Change in Amihud -20.14% -38.42% -7.40%
95% Confidence interval of mean
Lower Upper Lower Upper Lower Upper
.6734 .9237 .5414 .7195 .8611 .9090
In this table the three different periods are presented in the three different columns. The Mean of Ratio is the average ratio compared to the Pre-announcement period [AD-5, AD-55]. The differences in N are caused by the days in the sample tested. All samples consist of the same 23 companies. The confidence interval is computed on the base of the Mean of ratio and the standard error of the Mean. The significance of the tested means are shown as follows:
*** Significant at 1% ** Significant at 2% * Significant at 5%
For the second period of interest literature suggests an increase in liquidity mainly based on the portfolio rebalancing by index funds. The liquidity compared to the pre announcement period has increased significantly. The mean of the ratio over this sample is .6258 and this is highly significant. Compared to the findings of earlier research this increase of liquidity is rather large. Hedge and McDermott (2003) for example find a drop in ratio of 5.2% over this period, whereas based on this sample the ratio drops by 38.4%. The large difference between these outcomes is likely to be caused by two big differences in the research done by Hedge and McDermott and this research. Firstly an important reason is difference in timeframe. Since 1970 the amount of index-funds has increased from no significance on the market to a being a large part of the overall market (Greenwood & Scharfstein, 2012). Hedge and McDermott (2003) used a data from additions to the S&P 500 from 1993 to 1998. In this period index-funds were not as large as they are in the period used in this thesis. Since the liquidity effect around the addition day is caused mainly by these index funds their
12
enormous growth could enlarge this effect. Secondly, the difference in announcement day could be a reason the results differ. The announcement of an addition to the S&P is done five trading days in advance whereas an addition to the AEX is done 10 to 20 trading days in advance. Since the announcement day is further away from the addition more time is given to other investors to study the stock and maybe speculate on a positive price impact after addition.
For the last period of interest, the 50 day period starting 10 days after addition, the results are shown in the column on the right. Also for this period I find a significant increase in liquidity. This increase in liquidity is attributable to a drop in price uncertainty, caused by less asymmetric information, and an increase of limit orders used by un-informed investors (Hedge & McDerrmott, 2003). The results show that this effect is also visible on for the additions to the AEX. The mean of the ratio over this sample is .9260. This result is significant at 2%, which is mainly caused by the smaller effect on liquidity. The standard error of mean is lower than for the other two periods which makes the confidence interval more accurate. The increase in liquidity is interesting as it has not been found for all research on the S&P 500. Beneish and Whaley (1996) did not find a significant permanent increase of liquidity but on the other hand Hedge and McDermott (2003) did find evidence for a permanent effect on liquidity. One of the main differences between these two studies is the difference in approach. Hedge and McDermott (2003) left out the period around the effective date of addition since it could cause bias. Following their methodology these results also are based on a sample omitting the days around the effective day of addition.
Since the three periods are all compared to the same pre-announcement sample it is possible to compare the different timeframes. The comparison between the announcement effect and the addition effect is interesting. A T-test for two means is done to test whether the effect around the effective day is significantly larger than the effect of the announcement. The results of this test are shown in table 2. The mean of the of the addition period is smaller than the mean of the announcement period and this is significant at 2%. This shows that the increase of liquidity caused by risk arbitrageurs trying to make a profit by holding the shares up to the announcement day is smaller than the effect caused by the index funds rebalancing. This is similar to the findings of Hedge and McDermott (2003).
13
Table 2 – T-test on difference in mean announcement effect and addition effect
Period [AD, AD+5] [ED-4, ED+4]
Mean 0.798592 0.625811
Variance 0.458953 0.381965
Observations 115 207
T-value of difference 2.261693
P value 0.012351
In the table the results of the T-test are presented. The mean is the mean ratio over the sample period. The test is done with a hypnotised difference in mean of 0. The T-value and the P-value show the significance of the results.
The results in Table 1 also show that after the period of addition the liquidity drops. In the last period the liquidity is higher than before the addition but significantly lower than around the addition. This shows that the effect around the effective date of addition is indeed a temporary shock.
5. Conclusion
Although a lot of research on the price-effect of an addition to a well known index is done, not much research has been done on the direct effects on liquidity. This research adds to the studies on liquidity effects of additions to indices. Based on mainly the liquidity hypothesis the literature suggests that the addition of a stock should yield a positive effect on liquidity. In this research I find a positive effect on liquidity for three different timeframes. Firstly, I find that the announcement of the addition to an index has a positive influence on liquidity. This increase in liquidity is mainly caused by risk arbitrageurs trying to make an excess return. Secondly, I find a large increase in liquidity around the effective date of addition. The portfolio rebalancing by index investors close to this effective day of addition enlarges the liquidity in this timeframe. Thirdly, I find that the liquidity is permanently larger after the inclusion, which is attributable to the smaller information asymmetry.
The effect on liquidity around the effective date of addition is not permanent. The temporary shock in liquidity that I find is larger than found in earlier studies based on the S&P 500. This could be attributed the difference in timeframes since index funds grew rapidly over the last 20 years.
This research shows that the theory based on the S&P 500 also holds for the AEX-index, even with differences in the announcement and reviewing policies. Besides that it also contributes to the research on the effects of adding a stock to an index on the stocks liquidity.
14
Further research should be done on the effect of growing index funds on the increase in liquidity around the addition. This could help explain the large increase found based on this sample AEX sample. Furthermore it would be interesting to know whether other reviewing policies and announcement policies find similar results. Therefore research on different indices should be done.
6. Bibliography
Aitken, M., & Comerton-Forde, C. (2003). How should liquidity be measured? Pacific-Basin Finance Journal, 11(1), 45-59.
Amihud, Y. (2002). Illiquidity and stock returns: cross section and time series effect. Journal of Financial Markets, 5, 31-56.
Amihud, Y., & Mendelson, H. (1986). Liquidity and Stock Returns. Financial Analysts Journal, 42(3), 43-48.
Avera, W. E., Groth, J. C., & Kerry Cooper, S. (1985). Liquidity, exchange listing, and common stock performance. Journal of Economics and Business, 37(1), 19-33.
Beneish, M. D., & Whaley, R. E. (1996). An Anatomy of the "S & P Game": The Effects of Changing the Rules. The Journal of Finance, 51(5), 1909-1930.
Brealey, R. A. (2000). Stock Prices, Stock Indexes and Index Funds, (Bank of England Quarterly Bulletin). available on: http://ssrn.com/abstract=764145
Brennan, M., & Subrahmanyan, A. (1995). Investment analysis and price formation in securities markets. Journal of Financial Economics, 38, 361-381.
Brunnermeier, M. K., & Pedersen, L. H. (2009). Market Liquidity and Funding Liquidity. Review of Financial studies, 22(6), 2201-2238.
Chan, H. W. H., & Howard, P. F. (2005). Additions to and Deletions from an Open-Ended Market Index: Evidence from the Australian All Ordinaries. Australian Journal of Management, 30, 23-55.
Dhillon, U., & Johnson, H. (1991). Changes in the Standard and Poor's 500 List. The Journal of Business, 64(1), 75-85.
Easley, D., O’Hara, M., & Paperman, J. P. (1998). Financial analysts and information-based trade. Journal of Financial Markets, 1, 175-202.
Euronext. (2016). Index rule book AEX family. Geraadpleegd van file:///C:/Users/Acer/Downloads/aex_family_rules_version_16-01_jun_2016%20(2).pdf Goyenko, R. Y., Holden, C. W., & Trzcina, C. A. (2009). Do liquidity measures measure liquidity?
15
Harris, L., & Gurel, E. (1986). Price and volume effects associated with changes in the S&P 500 list: New evidence for the existence of price pressures. The Journal of Finance, 41(4), 815-829. Hedge, S. P., & McDermott, J. B. (2000). The Liquidity Effects of Additions to the S&P 500 Index.
Geraadpleegd van http://papers.ssrn.com/sol3/papers.cfm?abstract_id=236189
Hedge, S. P., & McDermott, J. B. (2003). The liquidity effects of revisions to the S&P 500 index: an empirical analysis. Journal of Financial Markets, 6(3), 413-459.
Hrazdil, K. (2009). The price, liquidity and information asymmetry changes associated with new S&P 500 additions. Managerial Finance, 35(7), 579-605.
Jain, P. C. (1987). The Effect on Stock Price of Inclusion in or Exclusion from the S&P 500. Financial Analysts Journal, 43(1), 58-65.
Jong, F. de, & Roon, F. de. (2011). Illiquidity: implications for investors and pension funds (Panel Paper 23). Geraadpleegd van http://arno.uvt.nl/show.cgi?fid=114742
Kumar, R., Sarin, A., & Shastri, K. (1998). The Impact of Options Trading on the Market Quality of the Underlying Security: An Empirical Analysis. The Journal of Finance, 53(2), 717-732.
Lo Martin, A. W., Petrov, C., & Wierzbicki, M. (2003). It's 11 PM—Do you know where your liquidity is? The mean-variance-liquidity frontier. Journal of Investment Management, 1(1), 55-93. Lo, A. W., Petrov, C., & Wierzbicki, M. (2003). It's 11 PM—Do you know where your liquidity is? The
mean-variance-liquidity frontier. Journal of Investment Management, 1(1), 55-93.
Lynch, A. W., & Mendenhall, R. R. (1996). New Evidence on Stock Price Effects Associated with
Charges in the S&P 500 Index. available
on:http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1298790
Shleifer, A. (1986). Do Demand Curves for Stocks Slope Down? The Journal of Finance, 41(3), 579-590. Standard & Poor's. (2016). S&P U.S. Indices Methodology. available on
