The announcement effect of appointing a former politician to the board of directors on stock returns : US Evidence

The announcement effect of appointing a former politician to the

board of directors on stock returns

US Evidence

Pascal Mertens

Msc Business Economics: Finance Master Thesis Finance

Student Number: 10872000 Supervisor: dhr. dr. J.E. Ligterink Date: 15-08-2016

Statement of Originality

This document is written by Student Pascal Mertens who declares to take full responsibil-ity 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

creat-ing it.

The Faculty of Economics and Business is responsible solely for the supervision of com-pletion of the work, not for the contents.

Abstract

This study examines the announcement effect of appointing a former politician to the board of directors on stock returns over the period of 2000 through 2004. An event study with interaction variable is conducted using a sample of 777 appointed board members, where 16 percent of the appointed board members have a political background. The re-sults show a small significant negative effect for appointing a former politician to the board of directors on stock returns. In contrast, the appointment of a politician in a high regulated industry shows a small positive effect on stock returns. Consequently, investors consider former politicians on the board of directors an added valuable to a company whenever the industry is highly regulated, because of the lobbying power of the former politician.

Table of Contents

1. Introduction

Jolanda Sap is a former Dutch politician who has been appointed as a member of the board of directors of KPN, a large telecom company in the Netherlands (Maaiveld, 2015). In November 2014 KPN appointed Jan Kees de Jager, a former Dutch minister of fi-nance, as chief financial officer to the board of management (KPN, 2014). A reason for a company to add a former politician to the board of directors is that politicians have lob-bying power. Loblob-bying power gives a person or a group the power to influence or put pressure on the government (Goldman, Rocholl, & So, 2008). Appointing a former politi-cian could give a company more lobbying power, which favors the company by influenc-ing the legislators and therefore gaininfluenc-ing legislative advantages. A second argument in favor of hiring a former politician is the familiarity of the politician. To the public a poli-tician is associated with a familiar face, which increases the publicity of a company. More publicity creates more opportunities, which could lead to a growth of the company. This leads to a higher stock return, because investors notice that the company is growing. Although a familiar face increases publicity, there is also a possibility that a politician has a negative reputation because of poor decision-making in the past or the higher risk of corruption. In this case, hiring a former politician could lead to a negative effect on the growth of a company (Hillman, 2005).

By performing this study, it will be examined what the announcement effect of appointing a former politician to the board of directors on the stock return is and whether there is a difference between industries. In previous studies (Cebula & Rossi, 2015; Hillman, 2005) a positive causal effect was found for the announcement of appointing a board member on stock returns and a positive correlation was found between the number of politicians on a board and company value, but in none of these studies an analysis has been performed regarding the effect of the announcement of a politician on the stock re-turn. This announcement effect explains the possible influence of politicians on the stock market. In a broader perspective, it shows whether investors consider politicians as an added value for the company. This added value can be a reason for directors of a compa-ny to adjust their strategy by appointing a politician.

In addition, in the literature review section three hypotheses will be formulated to eventually test the announcement effect of appointing a former politician to the board of directors on the stock return in a high regulated industry. It is assumed that former politi-cians have more lobbying power in comparison to other appointed board members that do not have a political background. To test the hypotheses an event study will be used, where the cumulative abnormal return is the dependent variable of the analysis. The cu-mulative abnormal return will be estimated by respectively, the market adjusted mean return model and the market model. These are two different models to estimate the ab-normal return.

The outline of this study is as follows. First, in section two the resource depend-ency theory, PEP-theory, and related studies will be discussed to come up with three hy-potheses at the end. In the methodology section the market adjusted return model, market model and interaction regression model will be elaborated. Finally, the interaction regres-sion model will be used to test the hypotheses. In section 4 the data is gathered for all variables of the three models. Further, in section 5 the results will be discussed based on the regressions. Finally, in section 6 a general conclusion will be given regarding the study and some limitations will be discussed.

2. Literature Review

In the existing literature only a few studies have been performed to examine the effect of adding a former politician to the board of directors on stock returns (Fan, Wong, & Zhang, 2007; Hillman, 2005; Warner & Watts, 1988). However, viewed from a broader perspective in terms of corporate governance, in several studies (Carter, Simpkins, & Simpson, 2003; Giroud & Mueller, 2010; Larcker, Ormazabal, & Taylor, 2011) effect of a change in corporate governance on company performances or stock returns was exam-ined. In the following paragraphs of this section the resource dependency theory, politi-cally exposed person (PEP), and studies of the change in corporate governance related to stock returns will be dealt with to formulate the hypotheses. The resource dependency theory and PEP theory are of importance for the study, since both can explain the reason-ing for appointreason-ing a politician.

2.1. Resource dependency theory

Mid 1900 corporate governance theories started to play a more important role when it comes to analysing company performances and the cause of changes regarding these per-formances. One of these theories is the resource dependency theory, which states that a company is in need of an external resource to lower risks or costs (Alhaji & Yusoff, 2012). In this this study the resource dependency theory is of relevance for companies that are interested to gain more influence in lowering costs or risks in a certain area or industry. For example, a lower excise tax on cigarettes is more interesting for a tobacco company than for a timber company. To reduce the costs of a cigarette, a tobacco compa-ny could appoint a (former) politician to invest in more lobbying power to prevent or reduce coming tax increase. This politician could be able to convince the sitting power to implement a law or subsidy that is beneficial for the company. Consequently, by this form of lobbying power the costs or risks of company diminishes, also known as rent-seeking (Edlin & Stiglitz, 1995; Kreuger, 1974). According to the resource dependency theory, adding a former politician to the board should have a positive effect for compa-nies and investors because of the rent-seeking. In the following paragraphs of this section empirical evidence will be shown in line with the theory.

2.2. PEP (Politically exposed person)

From a different point of view, adding a former politician to the board could also have a negative effect on a company, like corruption, fraud or bribery (FATF, 2012b; World-Check, 2008). According to the Financial Action Task Force (2012a), an international governmental institution that fights against money laundering, bribery, corruption and financing of terrorism, adding a politician to the company leads to more risk for a com-pany. In this theory a politician is called a Politically Exposed Person (PEP) and is de-fined as an individual who had recently been in a prominent public function (FATF, 2012a; World-Check, 2008). The reason that the presence of a former politician in a company can be a risk is because of the influence and strong position in a negotiation. A former prominent politician is trustworthy for a financial institution. Therefore, riskier deals will be closed despite the absence of an extra assessment that shows whether a pro-ject or investment is profitable and legal. It seems like these type of deals are positive for the company in the first place, but the opposite is the true. Eventually, the institution fig-ures out that the project is not profitable but actually an illegal and negative investment. This leads to less confidence in the institution and therefore stock prices will drop. An example of the negative influence by a PEP is the scandal of the Riggs Bank and former Chilean dictator Augusto Pinochet in 2004 (Carrington & Johnston, 2006). Riggs Bank was one of the largest banks of the United States in the 1980s. In 2004 three scandals were publicly disclosed, one of the three was the scandal with Augusto Pinochet. The bank created several illegal shell-, trust-, and offshore companies for Pinochet that were beneficial for both the bank and Pinochet. The Office of the Comptroller of the Currency (OCC), which supervises and regulates the national banks in the United States (OCC, 2016), fined Riggs Bank for non-transparency and missing documents for around $25 million dollars. In addition, to the fine given by the OCC, Riggs bank had an agreement with PNC to be acquired for $24.25 per share, before the scandals were publicly dis-closed. Afterwards, Riggs Bank was acquired by PNC for $20.00 per share. Moreover, the market value in this short period of time dropped from $22.44 to $19.43 per share

that adding a PEP to the company leads to higher risk in terms of corruption, fraud or bribery, and lowers the value of the company.

2.3. Corporate governance and firm value

One of the founders of the effect of corporate governance change on company value are Warner and Watts (1988). In their study, Warner and Watts (1988) attempt to find a rela-tion between an actual change in management and stock prices. They examined this by using a logit regression and event study, where the logit regression shows a relation be-tween the probability of a management change and the stock return with a binary out-come for the stock return. The outout-come of the study led to a non-significant difference in stock return based on an announcement of board changes. Although Warner and Watts (1988) did not find a significant result, they introduced a certain relation between corpo-rate governance and stock returns.

In several studies (Bennedsen, Kongstedb, & Nielsen, 2008; de Jong, Dejong, Mertens, & Wasley, 2005; Eisenberg, Sundgren, & Wells, 1998; Kiel & Nicholson, 2003; Yermack, 1996) it was examined whether the size of the board of directors has an effect on firm value. According to Hermalin and Weisbach (2003) larger board size leads to free-riding problems of the members. Hence, the larger a board is the less effective is the decision making. Eisenberg et al. (1998) found a negative correlation between board size and firm performance in Finland, where Bennedsen et al. (2008) found a negative causal effect of board size on firm performance for small and medium size companies in Den-mark. Moreover, Yermack (1996) found a negative correlation between a large board size of U.S. stock listed firms and market value.In general, the majority of the studies support the finding of a negative effect of larger board size on firm performance. However, de Jong et al. (2005) did not found an effect at all by analysing the Dutch market, where Kiel and Nicholson (2003) found a negative effect for the Australian market.

In line with Warner and Watts’s (1988) are Farrell and Hersch (2003), and Carter, Simpkins, and Simpson (2003), Farrell and Hersch (2003) investigated the change in gender of board members on stock returns. More specifically, appointing a woman as a member of the board. Even though the results were insignificant, a positive abnormal

stock return was found. In addition to the appointment of a woman to the board, Carter, Simpkins, and Simpson (2003) also examined the level of diversity of the board of direc-tors on firm value. In their study a cross-sectional analysis is used, with “diversity” and “woman” as the independent variables and firm value as a dependent variable. The meas-urement of the firm value is analysed by the Tobin’s Q. Both studies found that appoint-ing a woman to the board has a positive effect on stock returns. At the same time Carter, Simpkins, and Simpson (2003) show that a high level of diversity is positive for stock returns of the investor.

Furthermore, Fan, Wong, and Zhang (2007), investigated the effect of the board of directors of firms with political connections on stock returns after an initial public of-fering in China. The main difference in their study with respect to the other studies is the institutional background, the privatization of state-owned companies of the 1980s in Chi-na. A large number of the privatized companies were controlled by bureaucrats, therefore a distinction was made between privatized companies with and without bureaucrats as management. According to Fan et al. (2007), boards without political connections experi-ence higher stock returns after initial public offering. One of the reasons for this higher return can be explained by the grabbing hand argument; bureaucrats maximizing their own benefits at the expense of the profits of the controlled companies. Fan et al. (2007) show that the institutional background is essential when it comes to analysing the effect of politicians and firm value. In 2013 the Chinese government prohibited bureaucrats for holding a position in the board of the directors of public firms, following many resigna-tions of bureaucrats. Chan, Du, Lin, Peng, and Tang (2016) examined that stock returns were significantly lower after the resignation of bureaucrats. Comparing these two studies results in a contradiction of the value of a politician, namely the lack of knowledge and the resource of allocation.

Moreover, Goldman, Rocholl, and So (2008) explored the difference in return of democratic and republican related companies and the announcement of a new political connected board member. Their analysis was based on cross-sectional data after 2000, just after the republicans won the election. As a result, the stock price of republican relat-ed companies had increasrelat-ed and the stock price of democratic relatrelat-ed companies had

de-At last, the most closely related studies are that of Hillman (2005), Huang, Hsu, Khan, and Yu (2008), and Cebula and Rossi (2015). First, Hillman (2005) analysed the effect of the numbers of former politicians on the board of directors on the stock return of a company. Her findings show that the firm value is positively correlated to a larger number of politicians on the board of directors. Also, in heavy regulated markets there were significantly more former politicians on the board of directors than in less regulated markets, therefore both results support the resource dependency theory. Secondly, using data of Taiwan, Huang et al. (2008) examined the effect of adding a board member to the board of directors on the stock marke. A cumulative abnormal return analysis was used to gather the information on stock returns, resulting in a positive effect of the announcement of appointing a board member to the board on stock returns. The reason for this positive effect, according to Huang et al. (2008), is that a new outside director brings more exper-tise or new abilities to the board of directors. Cebula and Rossi (2015) observed the same outcome by using data of Italy. These findings assume that appointing a director is a sig-nal for the investors that companies will change their strategy, otherwise known as the signalling effect.

To summarize, in several studies the effect of corporate governance changes on stock return or firm value was investigated. From Warner and Watts (1988), that a change in management effects firm performance, to Cebula and Rossi (2015) and Huang et al. (2008) that the announcement of appointment of board members affect stock returns. This study combines the studies of Cebula and Rossi (2015), Hillman (2005), and Huang et al. (2008) to analyse the announcement effect of a former politician to the board of directors on the stock return. Therefore, a contribution to the existing literature will be an event study to discover a possible causality of the announcement of appointment of a former politician to the board of directors on stock returns. For controllers of a firm it is valuable knowledge to know whether investors think that political connected board members form a surplus or a risk for a firm. To be more specific, do investors add more value to political connected board member or to a non political connected board mem-bers? Previous studies suggest that the stock return will be higher by appointing a former politician to the board. In the following section the hypotheses regarding this question will be developed.

2.4. Hypotheses

In earlier studies (Cebula & Rossi, 2015; Huang et al., 2008) a positive causal effect of appointing a member to the board of directors on stock returns for the countries Taiwan and Italy is found. A similar analysis will be performed for this study to verify the out-come of the effect for adding a board member to the board of directors on the stock re-turn, hence the first hypothesis is:

H1: The announcement of appointing a board member leads to a higher stock return.

The second hypothesis will provide the information whether there is a significant differ-ence by adding a former politician to the board. The studies of Goldman et al. (2008) and Hillman (2005) discover that politicians have a positive effect on firm value, therefore the hypothesis is:

H2: The announcement of appointing a former politician as board member leads to a higher stock return.

The third hypothesis will be based on the that companies in higher regulated industries appoint more politicians as board members, because of the necessity of lobbying power of the politician by the company (Goldman et al., 2008). Consequently, the hypothesis is:

3. Methodology

In order to examine the hypotheses an event study will be used for finding the effect on the stock return by the announcement of appointment of a former politician or non politi-cal related person to the board of directors of a company. An interaction regression mod-el, where the cumulative abnormal return (CAR) is the dependent variable, will be used to analyse this effect. Before clarifying the regression model, the computation of the stock return, and cumulative abnormal return will be described. The abnormal return for the CAR will be estimated by the Market adjusted return and the market model for more precision;

Stock return:

!",$=

'",$− '",$)*

'",$)* (1)

where P is the stock price, R is the stock return, and the subscripts t and i are respective-ly, the time (in days) and a S&P 500 listed company. The stock return is calculated by the stock price of today minus the stock price of yesterday and then divided by the stock price of yesterday. The dividend is excluded from the calculation because it gives a pre-mium to a random stock price whenever a company pays out their dividends. Moreover, the dividend payment is unrelated to a stock market changes and including dividends into the stock return calculation yields to a higher stock expected return in specific periods (Fama & French, 1988).

Market adjusted return model:

.!",$ = !",$− !/$ (2.1)

where AR is the abnormal return, R is the stock return, Rm is the equally weighted index of the NYSE, and the subscripts t and i are respectively, the time (in days) and a S&P 500 listed company. The abnormal return is determined by the stock return subtracted by the equally weighted index of the NYSE, which is the excess return of a specific stock return related to the index return. The measurement of abnormal return is an alternative of the

mean adjusted return, namely the equally weighted index is substituted by the expected stock return, based on previous stock returns of the specific company (Brown & Warner, 1985).

Market Model:

.!",$ = !",$− 2"+ 4"!/$ (2.2)

where AR is the abnormal return, R is the stock return, Rm is the equally weighted index of the NYSE, alpha and beta are the intercept and the slope, and the subscripts t and i are respectively, the time (in days) and a S&P 500 listed company. The abnormal return is determined by the stock return subtracted by the alpha and the equally weighted index of the NYSE times the beta. The alpha and beta are estimated by the regression of the stock return against the market return. An estimation window of 185 days before till one day before the event window is used to estimate the alpha and beta (Brown & Warner, 1985).

Cumulative abnormal return:

5.!_",()6,76)= .!_",$

6

$8)6

(3)

where CAR is the cumulative abnormal return, AR is the abnormal return, and the sub-scripts t and i are respectively, the time (in days) and a S&P 500 listed company. The cumulative abnormal return is determined by the sum of abnormal returns, whereby the event window is set to eleven days. Considering the fact that the main interest of the study is on the effect of the announcement day, the event window is set five days before and five days after the announcement with the announcement on day zero.

Interaction Regression model:

5.!",()6,76) = 2 + 4* ':;",$ + 4< =>=",$ + 4? '!@",$ + 4A ':BC",$ + 46 5;.",$ + 4D @EF"

where CAR is the cumulative abnormal return and the dependent variable in the regres-sion, POL is a dummy variable of the announcement of appointment of a former politi-cian to the board of directors, MEM is a dummy variable of the announcement of ap-pointing a non-political related person to the board of directors or an appointment of a board member, PRI is a dummy variable for the period before the announcement, POST is a dummy variable for the period after the announcement, CLA is a dummy variable for the independency of a director, IND is a dummy variable for high regulated industries, BOARD is the number of board members at the company, AGE is the average age of the board of directors, SIZE is the market capitalization, beta ten and eleven are interactive variables of adding a politician as board member and respectively, market capitalization and regulated industries, α, ß, O and ε are the coefficients, and a fixed effect variable and the regression error. The subscripts t and i are respectively, the time (in days) and a S&P 500 listed company.

Furthermore, ß1 estimates the effect of the announcement of appointing a former politician to the board on the stock return. A positive (negative) beta means that there is a causal positive (negative) relationship between the announcement of appointing a former politician to the board and the stock return. However, there is no significant relationship between both of the variables if the beta is zero or close to zero. On the basis of the hy-potheses a positive beta is expected. Following is the coefficient ß2, estimating the effect of the announcement of appointing a board member on the stock return. The implication of beta two is equivalent to beta one, and therefore a positive beta two will support the hypothesis. Third, the coefficient ß3 estimates the stock return of the first five-day aver-age prior to the event. There is no abnormal return expected as the event did not occurred yet, and for that reason the beta is supposing to be zero or close to zero. Consequently, the expectation is that the values of coefficients ß2 and ß3 are different because of the announcement effect of the appointment of a former politician to the board for ß3. In ad-dition, coefficient ß4 estimates the stock return five-days average after the event. A posi-tive abnormal return is foreseen, because not all the effect is incorporated at the first day of the announcement.

Equally important are the coefficients ß5 up till ß9, which are the control varia-bles of the regression (4). Firstly, ß5 is the estimator of the independency of the director

related to the company. Multiple studies (Kiel & Nicholson, 2003; OCC, 2016) describe a positive effect on stock returns of appointing an independent board of director instead of appointing a related person to board of directors. Following their findings, a positive rela-tion is expected between independency and stock return. Next, ß6 is the estimator of highly regulated industries on stock returns, based on previous studies which found that stocks of companies operating in a regulated sector outperform companies of non-regulated sector (Fan, Wong, & Zhang, 2007). Hence, the prediction is a positive beta for companies of highly regulated industries. Thirdly, ß7 estimates the effect of the number of board members on stock return. As the effect is clarified before by Bennedsen et al., (2008) and Hermalin and Weisbach (2001), amongst other studies, it is assumed that a larger size of the board of directors leads to lower stock return. Therefore, a negative beta is expected for the board size on stock return. Further, ß8 is the estimator for the average age on stock returns. Giroud and Mueller (2010) and Goldman et al., (2008) found that a negative effect for the age of the board, because a younger board of directors is willing to take more a risk and are more open for changes in the company. For this matter a negative beta is expected for the variable age. Finally, the coefficient ß9 is the estimator of the market capitalization of a company on the stock return. A high market capitaliza-tion, in general, earns a normal stock return and contains less idiosyncratic risk. In con-trast, stocks with smaller market capitalization are riskier, but the possibility of an excess returns is higher (Lewellen, 2004). In brief, a negative beta is expected since a high (small) market capitalization leads to lower (higher) excess returns.

4. Data

This section entails the various data sources and describes and clarifies the variables us-ing descriptive statistics. For the analysis daily data with a timespan of five years is used, from 2000 till 2004. The reason for this period is that it contains all the available data to perform the analysis and the fact that this period is not inflicted with the credit crisis of 2008.

The variables of interest for the analysis are the announcement of appointing a board member with and without political background. These variables are hand-collected by using the databases of ISS retrieved from WRDS (WRDS, n.d.) and LexisNexis Aca-demic (LexisNexis, n.d.). The ISS database contains the year when a board member is appointed, since the announcement date is needed for the analysis the database of Lex-isNexis Academic is used. A specific search string is implemented to find the dates of announcement of appointing a member, since the LexisNexis Academic is a database of news articles. A different search string was used to identify whether the new board mem-ber is a former politician. The definition in this study that defines whether a person has a political background is based on the fact that the person has worked as a bureaucrat, dip-lomat or politician. Furthermore, announcements are excluded from the analysis when announcements fell on the same date with other events, for instance an acquisition of a company, announcement of dividend payments or a stock split. This hand collected in-formation is a big part of the study, because gathering the data of these two variables was time consuming. Table A.1 of Appendix A shows the directors and companies which are gathered by hand and used for the analysis. By exclusion of the directors that were ap-pointed before 2000 or after 2004, on the same date as other appointments by the compa-ny, or other stock related events, 777 board directors remain for the analysis. The reason for the exclusion of these directors is for the distinction whether a director has a political background. When Otherwise no distinction could be made if for example one out of five appointments has a political background. Consequently, only unique announcements of appointing a director are included in the analysis. Column three and four of Table A.1 show the announcement date and shows whether the appointed director has political background. In addition, the scatter plot of Figure A.1 of Appendix A shows that there is no correlation between the announcement date and unique directors. In other words, the announcement of a director appointment is spread out over time.

Moreover, the data for the daily stock returns and the equally-weighted returns of the NYSE of the companies on the S&P 500 from January 1999 till January 2005 is col-lected by the database of CRSP retrieved from Wharton Research Data Services (WRDS, n.d.). The reason that the starting date is different from the analysis time period is be-cause of the estimation window of the market model, which estimates the expected stock

return. The names of the companies on the S&P 500 from 2000 up till 2004 are obtained by Compustat. Furthermore, the control variable regulated industry is constructed by the data of Compustat and Fan et al., (2007). The database of Compustat contains the North American Industry Classification System (NAICS) code, which is a classification code for business in North-America (WRDS, n.d.). On the basis of the classification code a distinction is made between highly regulated and non or less regulated industries. In line with Fan et al., (2007), the following industries are defined as highly regulated: finance and real estate, natural gas, telecommunications, electricity, transport and water industry, natural resources and public utilities. Other control variables for the analysis are gender, independency of a new director with respect to the company, market capitalization, size of the board and the age of director. Gender, independency of a new director, age, and size of the board are gathered from the database of ISS, whereas market capitalization is obtained from the database of CRSP. The market capitalization is computed by multiply-ing the absolute stock price by the shares outstandmultiply-ing and is included because it controls the size effect. For the analysis the natural logarithm of the market capitalization will be used to discover a linearity between the stock return and market capitalization.

Table 4.1 shows the aforementioned variables and their descriptive statistics. The daily stock return and the price, for calculating the market capitalization, are winsorized at 0.01 and 0.99 to cut the outliers. The average daily stock return is 0.00% and varies 3.00%, shown by the mean and standard deviation. The average daily stock return of the NYSE index is the same but is less volatile because of the standard deviation of 1.00% shown in Table 4.1. In addition, the variable Ann. Director (pol.) is the dummy variable for the announcement of appointing a board member with political background. The sample size shows that there are 777 board members appointed, whereof 101 (16%) with a political background.

Furthermore, the first variable of the table is the natural logarithm of the market capitalization (Market cap(ln)). A natural logarithm is used to perform the analysis be-cause it gives a linearity of the market capitalization, therefore the effect of a change in market capitalization shows what the effect is on stock return. Moreover, the market capi-talization is winsorized to reduce the effect of the outliers. Second, the regulated industry

regulated in the sample. In addition, the size of board (Board) is between 5 and 25 mem-bers, and on average there are eleven members on the board of directors. The table also shows that 17% of the appointed members are women (female) and that the average age of the board of directors approximately is 55, with the youngest being 32 years-old and the oldest 75 years-old. The last variable Independency of dir. is a dummy variable with respect to the independency of the appointed director. WRDS (n.d.) distinct directors into three categories: independent, employees and linked to the company before. This study combines the categories, employees and linked to the company before, to form the classi-fications independent and affiliated to the company. The table shows that 80% out of 777 appointed directors are independent of the company. The following section will discuss the results of the performed analysis and the gathered data.

Table 4.1: Descriptive statistics

This table shows the variables that are used for the event study. Market cap is the natural logarithm of the market capitalization, Regulated industry is a dummy variable of regulated industries, Board size is the number of board of directors for a company, Gender is the dummy variable for the number of women or man on a board, Ann. Director (pol.) is a dummy variable whether the appointment of a new member is a politician. Age is the age of a new appointed director, Independency of dir. is a dummy variable that classi-fies whether the new member was related to the company, Stock return is the daily stock return of the S&P companies, and ewretx is the return of the NYSE index. The variables LogMV and stockreturn are winso-rized at 0.01 and 0.99. Furthermore, N is the sample size, Mean is the average, and S.D. is the standard deviation.

N Mean S.D Quantiles

Min 0.25 Med 0.75 Max

Market cap(ln) 8547 15.89 1.27 12.18 15.04 15.80 16.58 20.10

Regulated industry 777 0.29 0.45 0.00 0.00 0.00 1.00 1.00

Board size 777 10.75 2.48 5.00 9.00 11.00 12.00 25.00

Gender 777 0.17 0.37 0.00 0.00 0.00 0.00 1.00

Ann. Director (pol.) 777 0.13 0.34 0.00 0.00 0.00 0.00 1.00

Age 777 54.72 6.84 32.00 50.00 55.00 59.00 75.00

Independency of dir. 777 0.80 0.40 0.00 1.00 1.00 1.00 1.00

Stock return 8547 0.00 0.03 -0.08 -0.01 0.00 0.01 0.10

5. Results

This section will test the formulated hypotheses based on several regressions. However, a correlation table will be analysed and the cumulative abnormal returns will be tested for the level of significance, for giving a better insight into the variables and better under-standing of the regressions.

5.1. Correlation and (Cumulative) abnormal returns

Table 5.1 provides information about the correlation between the independent variables in the regression. Companies in regulated industries appoint slightly

Table 5.1:Correlation table

This table shows the correlation between variables used in the following correlation matrix. It provides possible correlations between the Moreover, the largest correlation is that of the size of the board and mar-ket capitalization, thereafter the size of the board and regulate industries.

Ann. Director (pol.) Regulated industry Market

Cap (ln) Gender Age

Independ-ency of dir. Board size Ann. Director(pol.) 1.000 Regulated industry 0.0504 1.000 Market Cap (ln) 0.0752 0.1774 1.000 Gender 0.0190 -0.0476 0.0021 1.000 Age -0.0299 0.0992 -0.0213 -0.1901 1.000 Independency dir. 0.0474 0.0115 -0.1206 0.1055 0.1810 1.000 Board size -0.0038 0.2539 0.4460 0.0011 0.0497 -0.0650 1.000

more politicians than less regulated industries or directors have preference to join a com-pany that is more regulated. This is in line with the resource theory and Hillman (2005), that companies in the regulated industries appoint more board members with political background and therefore lobbying power. Also, companies in regulated industries have a larger board than less regulated industries. A reason for the larger board is that the com-pany has to follow more laws, subsequently more specialisation or lobbying power is

pendent director and market capitalization is shown in the table. This means that a higher market capitalization leads to the appointment of directors that are related to the company instead of an independent directors of the company.

Table 5.2: Significance of the cumulative abnormal return

This table shows the event window (11 days) and the cumulative abnormal returns. The CAR is based on the abnormal return estimation of the market adjusted return or market model. The significant test is based on event window across time from 2000 to 2004. The standard errors are clustered at company level, whereas the robust t-statistics are put in parentheses. *, **, and *** define the significance level at respec-tively, 10%, 5%, and 1%. The market adjusted return has one significant cumulative abnormal return, where the market model has six significant cumulative abnormal returns in the even window. Both models have an significant cumulative abnormal return for the period zero till five day post, which is the period that the board members are added.

Observations CAR

Market adjusted return

CAR Market Model

Event Window:

Five days prior 777 -0.0004 _(-0.45) -0.0003 _(-0.32)

Four days prior 777 0.0002

(-0.17)

0.0006 (0.51)

Three days prior 777 0.0007

(-0.50)

0.0004 (0.28)

Two days prior 777 -0.0006 _(-0.37) 0.0008 _(0.44)

One day prior 777 0.0011

(0.56)

0.0024 (1.22)

Event day 777 0.0023 _(1.02) 0.0043* _(1.84)

One day post 777 0.0013 _(0.55) 0.0045* _(1.78)

Two days post 777 0.0016

(0.61)

0.0052 (1.93)

Three days post 777 0.0025

(0.94)

0.0059** (2.12)

Four days post 777 0.0017

(0.60)

0.0057* (1.94)

Five days post 777 0.0022

(0.72)

0.0068** (2.15)

Five days prior – one day prior 3,885 -0.0002

(-0.27)

0.0008 (1.17)

Event day - five days post 4,662 0.0020*

(1.79) 0.0054*** (4.81) All 8,547 0.0009 (1.46) 0.0033*** (4.83)

Table 5.2 provides the information whether the CARs are significantly larger than zero. The event window is set five days before and after the date of the announcement of appointing a director. The second column, estimated based on the market adjusted return,

shows a significant level of 10% for the abnormal return of the combined days from the day of the event up till five days after. Where the abnormal return estimated by the mar-ket model has a minimum significance level of 10% for the event day and the five days after separately and combined. These findings support our hypothesis that the announce-ment of adding a board member or politician results in a higher stock return. In addition to these findings, figure B.2 in Appendix B shows the cumulative average abnormal re-turn for the market model and market adjusted rere-turn model. The increase in the average stock return on the day of the event (for both estimation models) and continue this posi-tive line up till day five is in line with Table 5.2 regarding the posiposi-tive abnormal return.

5.2. Main regressions

For testing the first hypothesis, the announcement of appointing a board member to a company increases the stock return, columns one and four in table 5.3 should be con-sidered. The first regression (column one) estimates the announcement effect of adding a board member on stock return by the market adjusted model. The effect is measured by the dummy member, that splits the event window into two parts. Five days before the announcement takes the value zero, where the announcement day and the five days after take value 1. The coefficient of the variable is 0.0021 and has a significance level of 0.01. Column three also shows a small positive effect of the announcement of appointing a director on stock returns, 0.0046 with a significance level of one percent. This Indicates that the announcement of appointing a board member has a significant small positive effect on stock returns. In other words, investors react positive to an appointment of a director. One of the reasons could be that investors experience that appointing a board member leads to changes or innovation in the company. This finding is also supported by the studies of Cebula and Rossie (2015) and Huang et al., (2008), where both of the stud-ies found a larger effect than this study.

Regressions two and five (column two and five of table 5.3) show the result for the second hypothesis. The hypothesis states that the announcement of appointing a board member with a political background increases the stock return. In the second column of

member with political background is -0.0096 with a significance level of 0.01. This out-come is in contrast with the hypothesis because it is slightly negative, instead of a posi-tive coefficient. This negaposi-tive effect means that investors do not expect that the appoint-ment of a former politician leads to extra value for the company. Although the finding does not support the hypotheses, it can be explained what the reason is of the negative effect on stock return. First of all, it could be the small sample size of this study that causes the negative effect. Second, on the basis of the study of Fan et al., (2007) a lack of knowledge of the politician or bad leadership in their previous job could cause the inves-tors to be more suspicious of the role of the politician. Furthermore, a well-known person could also increase the risk of a company. Former politicians can involve higher risk for the company with respect to corruption and bribery, therefore investors could be suspi-cious about the appointment (World-Check, 2008). Column five of the market model also shows the result of the analysis for the second hypothesis, with a coefficient of 0.0004. As the coefficient is insignificant it has no power of explanation and therefore there is no support for the hypothesis.

Column three and six show the results of the analysis for the third hypothesis. The third hypothesis states that the appointment of a board member with political background for companies in a high regulated industry increases the stock return of the company. The reason for this is that companies in high regulated industries have to comply and apply with more regelation and rules, therefore a person with more lobbying power is more effective in these industries. In column three of the adjusted market return model the in-teraction variable regulated industry multiplied by the variable politician shows a signifi-cance level of 0.01 and a coefficient of 0.0208. Column six of the market model also shows a significance level of 0.01 and a slightly positive coefficient of 0.0107. So both models give a positive abnormal return for the announcement of appointing a director in high regulated industries.

In addition, the significant control variables in the regressions are the board size, the independent dummy, Gender, and the market capitalization. Gender is for all the re-gressions significant and has a small negative coefficient, which means that when the appointment of a new board member is a female the stock return will drop. This finding contradicts the finding of Farrell and Hersch (2003). They found a positive abnormal

Tabel 5.2: The effect of adding a former politician to the board on stock returns

This table looks at the announcement effect of appointing a board member (with political background) on stock returns. The regressions use daily data from 2000 to 2004. The cumulative abnormal return is the dependent variable in all six regressions. Member is a dummy variable that is one at the announcement date and the five days after and zero for the five days before, the same applies for the “Politician (time 0-5)”. The third row is an interaction variable of regulated industry and politician. Gender is a dummy variable that is one when it is a woman that is appointed to the board. Furthermore, Independency is a dummy vari-able that takes the value one if the director has no connections to the company beforehand. Regulated in-dustry is a dummy variable that takes value one, if the companies business is in a regulated inin-dustry. The difference between columns 1-3 and 4-6 is that the first columns are based on the market adjust return model, whereas columns 4-6 is based on the market model. The standard errors are clustered at company level, whereas the robust t-statistics are put in parentheses. *, **, and *** define the significance level at respectively, 10%, 5%, and 1%.

Dependent variable: Cumulative abnormal return

Market adjusted return model Market model

(1) (2) (3) (4) (5) (6)

Board Member Politician Politician

(Interaction) Board Member Politician

Politician (interaction) Member 0.0021*** 0.0046*** (4.122) (3.37) Politician -0.0096*** -0.0163*** 0.0004 -0.0030*** (-13.35) (-18.59) (0.52) (-3.11) Politician*Regulated 0.0208*** 0.0107*** industry (13.49) (6.25) Gender (female) -0.0039*** -0.0030*** -0.0028*** -0.0054** -0.0032*** -0.0032*** (-2.58) (-4.43) (-4.22) (-2.84) (-4.36) (-4.26) Age -0.0002** -0.0002 -0.0001 -0.0003*** -0.0001 -0.0001 (-2.25) (-0.62) (-0.023) (-2.78) (-0.94) (-0.94) Independency -0.0051*** -0.0011* -0.0009 -0.0032* -0.0020*** -0.0021*** (-2.94) (-1.68) (-1.22) (-1.86) (-2.28) (-3.03) Market capitalization 0.0004 0.0009*** 0.0010*** -.0010* 0.0001 0.0002 (0.79) (4.15) (4.85) (-1.77) (0.72) (0.76) Regulated Industry 0.0015 0.0027*** 0.0014 -0.0021 0.0021*** 0.0005 (0.98) (4.70) (0.84) (-1.34) (3.32) (0.82) Board size 0.0001 -0.0003** -0.00027*** -0.0005* -0.0004*** -0.0004*** (0.34) (-2.59) (-2.60) (-1.68) (-3.46) (-3.44) Constant 0.0087 0.0087 -0.004 0.0005*** .00548 0.0054 (0.84) (0.84) (-0.39) (1.31) (1.32) (1.32) Observations 8,547 8,547 8,547 8,547 8,547 8,547 Adj. R-squared 0.033 0.037 0.047 0.045 0.052 0.086

mod-return for the market adjusted mod-return model in column two and three. However, in column four of the market model the coefficient is negative and significant. Column two till five show a negative effect of board size on stock return. The outcome that a larger board has a negative effect on the stock returns is supported by the studies of Bennedsen et al. (2008) and Hermalin and Weisbach (2003). Larger boards, as mention before, could lead to a free-riding problem of board members or indecisiveness of the board. As for the de-pendency dummy variable, it indicates that appointing an independent board member has a negative effect on the stock returns. Implying that investors have more confidence in board members that have a history at the company in comparison to unrelated directors.

By summarizing the six regressions it can be concluded that there is small posi-tive announcement effect of adding a member to the board on stock return. Also, the an-nouncement of adding a former politician to the board has a negative effect according to the market adjusted return model. Moreover, the interaction variable of regulated industry and politician have a positive effect on the stock return. Furthermore, It is also worth no-ticing that the adjusted r-squared is low, even though it goes marginally up by adding the control variables it still remains remarkably low.

6. Conclusion & Discussion

This study endeavoured to give an answer to the question what the effect of appointing a former politician to the board of directors on stock returns is. In earlier studies several effects of corporate governance on stock returns were analysed, but non of the studies actually have tested the announcement effect of politician to the board on stock returns. However, closely related studies of Cebula and Rossi (2015) and Huang et al., (2008) examined the announcement effect of appointing a board member on stock returns. The findings of both studies show the same outcome, namely a positive effect of appointing a board member on stock returns. Furthermore, Hillman (2005) found a positive correlation between (former) politicians on the board of directors and company value. The three studies clarify that appointing a board member has a positive effect on the stock market,

therefore investors consider board members as an added value to the company. In these previous studies the background of the appointed board members was not taken into ac-count. As mentioned in the literature review though, missing lobbying power can be an incentive for a company to appoint a former politician to the board of directors. For this matter, it can be assumed that appointing a former politician to the board of directors would lead to an even greater value. In line with this reasoning the main question was formulated, whether an appointment of a former politician leads to a higher stock return.

First, the three hypothesis will be discussed before the main question will be an-swered. The first hypothesis states that the announcement of appointing a board member has a positive effect on the stock return. The result of the analysis supports this hypothe-sis and the findings of Cebula and Rossi (2015) and Huang et al., (2008), that there is a positive announcement effect of appointing a board member to the board of directors on stock returns. Cebula and Rossi (2015) state that investors interpreted the appointment of a member to a change in strategy of the company, whereas Huang et al., (2008) clarify that the new member brings in new expertise which will innovate the company and there-fore have a positive effect on the stock return.

In line with the reasoning of appointing a board member for their expertise is the second hypothesis formulated, asserting that the appointment of a former politician to the board of directors has a positive effect on the stock return. In addition, the resource de-pendency theory is relevant for this study because it explains that a company requires a former politician to obtain more resources. In terms of appointing a politician it is the resource of lobbying power. A company that has the possibility to influence the govern-ment by a former politician could reduce the costs for the company through lower tax costs or specific subsidies. In contrast, appointing a well known person entails the risk of fraud, bribery and corruption, as is defined in the PEP theory. Hence, the resource de-pendency theory expects a positive effect on the stock return, whereas the PEP argument expects a negative effect. These two contradictory theories and the three mentioned stud-ies were the basis for this study. The results of the analyses show a negative effect for the appointment of a former politician to the board of directors on the stock return, which supports the PEP theory. This outcome does not support the second hypothesis, since a

vestors are discouraged about the appointment of a former politician to the board of di-rectors, because the stock return drops slightly when a politician is appointed. On the other hand, it can be considered that investors judge the appointment of a politician as a risky move because of a lack of knowledge of the politician or possible threats with re-spect to the PEP theory.

Although a negative effect is found for the appointment of a former politician to the board of directors on stock return, an additional hypothesis is stated with respect to the politician and stock return. Assuming that former politicians are appointed because of their lobbying power, the effect on stock return of the former politician will be higher in industries that are highly regulated. Therefore, the third hypothesis is the effect of the announcement of adding a former politician to the board of directors on stock return for companies in highly regulated industries. The findings of the analysis show that appoint-ing a former politician to the board has a positive effect on the stock return for companies that are highly regulated. This implies that investors consider former politician valuable when a company is highly regulated. This outcome is in line with the reasoning of Huang et al., (2008), that a new member brings in new expertise and therefore adds value to the company. Conclusively, this finding suggest that adding a politician to the board is valu-able for the investors whenever it is in a highly regulated industry.

Possible limitations regarding this study will be discussed in this paragraph. First of all, in this study a relatively small sample size is used and therefore the outcome of the regression could be a biased outcome. Secondly, the dates for the announcement of for-mer politicians and other appointed board members are hand-collected. Hence, the accu-racy of hand-collected data is lower than the accuaccu-racy of data that is retrieved from data-bases, due to possible human errors. Furthermore, in this study it is assumed that former politicians have lobbying power, whereas non-former politicians have no lobbying pow-er. This is an extreme assumption because a general board member without being a for-mer politician could easily have built a great connection to the government or have lob-bying power as well. Additionally, this study shows results for the United States. Howev-er, countries differ in regulation or institutional background and therefore it is possible that results for other countries will lead to a different outcome.

As this study is the first that analyses the announcement effect of appointment of board members regarding political background, several directions for further research are still open. For example, examining the effect of an appointment of a former politician to the board of directors on stock return in the long run can be of interest. Another study could test whether there is a difference between a developed and a developing country. In general, in a developing country the corruption level is higher than in developed coun-tries, and therefore a study could analyse if the level of regulation has an influence on the appointment of a former politician to the board of directors of a company. Overall, these follow-up studies could contribute to clarify whether a board member with a political background is of value to the company and its investors. The current study can be seen as a first step towards answering this question.

7. References

Alhaji, I. A., & Yusoff, W. F. (2012). Insight of Corporate Governance Theories. Journal

of Business & Management, 1(1), 52-63.

Bennedsen, M., Kongstedb, H. C., & Nielsen, K. M. (2008). The causal effect of board size in the performance of small and medium-sized firms. Journal of Banking &

Finance, 32(6), 1098–1109.

Black, B. S., de Carvalho, A. G., & Gorga, É. (2012). What matters and for which firms for corporate governance in emerging markets? Evidence from Brazil (and other BRIK countries). Journal of Corporate Finance, 18(4), 934-952.

Brown, S. J., & Warner, J. B. (1985). Using daily stock returns: The case of event studies. Journal of Financial Economics, 14(1), 3-31.

Carrington, I., & Johnston, R. B. (2006). Protecting the financial system from abuse: Challenges to banks in implementing AML/CFT standards. Journal of Money

Laundering Control, 9(1), 48-61.

Carter, D. A., Simpkins, B. J., & Simpson, W. G. (2003). Corporate Governance, Board Diversity, and Firm Value. The Financial Review, 28(1), 33-53.

Cebula, R. J., & Rossi, F. (2015). Stock market reactions to announcements of the board of directors: evidence from Italy. Applied Economics, 49(20), 2102-2118.

Chan, K. C., Du, J., Lin, Y., Peng, Q., & Tang, X. (2016). Politically connected directors and firm value: Evidence from forced resignations in China. The North American

Journal of Economics and Finance, 37, 148–167.

de Jong, A., Dejong, D. V., Mertens, G., & Wasley, C. E. (2005). The role of self-regulation in corporate governance: evidence and implications from The Netherlands. Journal of Corporate Finance, 11(3), 473–503.

Edlin, A. S., & Stiglitz, J. E. (1995). Discouraging Rivals: Managerial Rent-Seeking and Economic Inefficiencies. The American Economic Review, 85(5), 1301-1312. Eisenberg, T., Sundgren, S., & Wells, M. T. (1998). Larger board size and decreasing

firm value in small firms. Journal of Financial Economics, 48(1), 53-54.

Fama, E. F., & French, K. R. (1988). Dividend yields and expected stock returns. Journal

Fan, J. P., Wong, T. J., & Zhang, T. (2007). Politically connected CEOs, corporate governance, and Post-IPO performance of China's newly partially privatized firms. Journal of Financial Economics, 84(2), 330-357.

Farrell, K. A., & Hersch, F. L. (2003). Additions to corporate boards: the effect of gender. Journal of Corporate Finance, 11(1-2), 85-106.

FATF. (2012a). International Standards on Combating Money Laundering and the Financing of Terrorism & Proliferation. Paris, France.

FATF. (2012b). Specific Risk Factors in Laundering the Proceeds of Corruption. Paris, France.

Giroud, X., & Mueller, H. M. (2010). Does corporate governance matter in competitive industries? Journal of Financial Economics, 95(3), 312-331.

Goldman, E., Rocholl, J., & So, J. (2008). Do Politically Connected Boards Affect Firm Value? Review of Financial Studies, 22(6), 2331-2360.

Gomes, J., Kogan, L., & Zhang, L. (2003). Equilibrium Cross Section of Returns.

Journal of Political Economy, 111(4), 693-732.

Hermalin, B., & Weisbach, M. (2001). Boards of Directors as an Endogenously Determined Institution: A Survey of the Economic Literature. National Bureau of

Economic Research.

Hillman, A. J. (2005). Politicians on the Board of Directors: Do Connections Affect the Bottom Line? Journal of Management, 31(3), 461-481.

Huang, H.-H., Hsu, P., Khan, H. A., & Yu, Y.-L. (2008). Does the Appointment of an Outside Director Increase Firm Value? Evidence from Taiwan. Emerging Markets

Finance and Trade, 44(3), 66-80.

Jiang, X., & Lee, B. S. (2012). Do Decomposed Financial Ratios Predict Stock Returns and Fundamentals Better? Financial Review, 47(3), 531-564.

Kiel, G. C., & Nicholson, G. J. (2003). Board composition and corporate performance: How the Australian experience informs contrasting theories of corporate governance. Corporate Governance: An International Review, 11(3), 189–205. KPN. (2014). KPNCorporate. Retrieved February 9, 2015, from KPN:

http://corporate.kpn.com/pers/persberichten/kpn-benoemt-jan-kees-de-jager-tot-Kreuger, A. O. (1974). The Political Economy of the Rent-Seeking Society. The

American Economic Review, 64(3), 291-303.

Larcker, D. F., Ormazabal, G., & Taylor, D. J. (2011). The market reaction to corporate governance regulation. Journal of Financial Economics, 101(2), 431-448.

Lewellen, J. (2004). Predicting returns with financial ratios. Journal of Financial

Economics, 74(2), 209-235.

LexisNexis. (n.d.). LexisNexis Academic. Retrieved 2016, from LexisNexis: http://academic.lexisnexis.nl.proxy.uba.uva.nl:2048/

Maaiveld. (2015). ECONOMIE & POLITIEK. (Financieel Dagblad) Retrieved February 9, 2015, from FD.: http://fd.nl/economie-politiek/1091381/jolande-sap-terug-in-den-haag

Nguyen, B. D., & Nielsen, K. M. (2010). The value of independent directors: Evidence from sudden deaths. Journal of Financial Economics, 98(3), 550-567.

OCC. (2016). About the OCC. Retrieved from Office of the Comptroller of the Currency; U.S. Departement of the Treasury: http://www.occ.gov/about/what-we-do/mission/index-about.html

Pontiff, J., & Schall, L. D. (1998). Book-to-market ratios as predictors of market returns.

Journal of Financial Economics, 49, 141-160.

Warner, J. B., & Watts, R. L. (1988). Stock prices and top management changes. Journal

of Financial Economics, 20, 461-492.

World-Check. (2006). REPUTATION DAMAGE: The Price Riggs Paid.

World-Check. (2008). “Politically Exposed Person” Refining the PEP Definition.

WRDS. (n.d.). Wharton Research Data Services. Retrieved 2016, from The Wharton School, University of Pennsylvania: https://wrds-web.wharton.upenn.edu/wrds/ Yermack, D. (1996). Higher market valuation of companies with a small board of

8. Appendix

8.1. Appendix A

Table A.1: List of companies and directors and their announcement of appointment

This table shows 463 S&P 500 companies that are used for this study. Furthermore, the announcement date and name of the added director (777 announcements) to the company, as well as whether the director has a political background, is provided in the table. Political background is defined as whether a director has served as bureaucrat, diplomat or politician (101 directors have a political background).

Company Name Director Date Politician

3M ROBERT S MORRISON 11Nov2002 No

3M VANCE D COFFMAN 14May2002 No

3M W JAMES MCNERNEY JR 05Dec2000 Yes

3M CO MICHAEL L ESKEW 07Apr2003 No

ABBOTT LABORATORIES JACK M GREENBERG 08Dec2000 No

ABBOTT LABORATORIES RICHARD A GONZALEZ 28Jul2000 No

ABBOTT LABORATORIES ROXANNE S AUSTIN 08Dec2000 No

ADC TELECOMMUNICATIONS RICHARD R ROSCITT 29Jan2001 No

ADC TELECOMMUNICATIONS INC ROBERT E SWITZ 02Jun2003 No

ADOBE SYSTEMS ANTONIO M PEREZ 20Sep2000 No

AES PHILIP A ODEEN 20Mar2003 Yes

AETNA JEFFERY E GARTEN 28Jan2000 No

AETNA R DAVID YOST 25Oct2002 Yes

AETNA INC EDWARD J LUDWIG 27Jun2003 No

AFLAC JOSEPH M (MAX) CLELAND 01Jul2003 Yes

AFLAC MARVIN R SCHUSTER 22May2000 No

AFLAC ROBERT B JOHNSON 06May2002 No

AFLAC TAKATSUGU MURAI 20May2000 No

AGILENT TECHNOLOGIES HEIDI KUNZ 24Feb2000 No

AGILENT TECHNOLOGIES JAMES G CULLEN 14Apr2000 No

AIR PRODUCTS & CHEMICALS INC W DOUGLAS FORD 20Nov2003 Yes

AIR PRODUCTS AND CHEMICALS CHARLES H NOSKI 19Oct2000 No

ALBERTO-CULVER -CL B SAM J SUSSER 25Jan2001 No

ALBERTO-CULVER CO GOVERNOR JIM EDGAR 24Oct2002 Yes

ALBERTSON'S BONNIE GUITON HILL 07Dec2001 No

ALBERTSON'S JON C MADONNA 13Dec2002 No

ALBERTSON'S LAWRENCE R JOHNSTON 24Apr2001 Yes

ALLEGHENY ENERGY LEWIS B CAMPBELL 10Oct2000 No

ALLEGHENY TECHNOLOGIES BRIAN P SIMMONS 11Jul2002 Yes

ALLEGHENY TECHNOLOGIES GEORGE J KOURPIAS 14Jul2000 Yes

ALLERGAN ANTHONY H WILD 03Aug2000 Yes

ALLIED WASTE INDUSTRIES J TOMILSON HILL 25Jan2002 Yes

ALLIED WASTE INDUSTRIES LEON D BLACK 05Jul2000 Yes

ALLSTATE JACK M GREENBERG 05Feb2002 No

ALTERA JOHN P DAANE 27Nov2000 Yes

ALTERA ROBERT J FINOCCHIO JR 14Jan2002 Yes

ALTERA CORP KEVIN J MCGARITY 15Mar2004 No

ALTRIA GROUP MATHIS CABIALLAVETTA 28Aug2002 No

AMBAC FINANCIAL GROUP JILL M CONSIDINE 31Mar2000 Yes

AMBAC FINANCIAL GROUP ROBERT J GENADER 23Jan2001 No

AMERADA HESS JOHN J O'CONNOR 15Oct2001 No

AMEREN CORP SUSAN S ELLIOTT 10Oct2003 Yes

AMERICAN ELECTRIC POWER RICHARD L SANDOR 26Apr2000 No

AMERICAN ELECTRIC POWER THOMAS V SHOCKLEY III 26Apr2000 Yes

AMERICAN EXPRESS URSULA M BURNS 26Jan2004 No

AMERICAN GREETINGS -CL A CHARLES A RATNER 23Jun2000 No

AMERICAN GREETINGS -CL A JACK KAHL 22Dec2000 No

AMERICAN INTERNATIONAL GROUP DONALD P KANAK 04Dec2003 No

AMERICAN INTERNATIONAL GROUP WILLIAM S COHEN 02Feb2004 Yes

AMERICAN POWER CONVERSION ELLEN B RICHSTONE 19Feb2004 No

AMERICAN POWER CONVERSION JOHN F KEANE 19Jun2001 Yes

AMERISOURCEBERGEN CHARLES H COTROS 15Jan2002 Yes

AMERISOURCEBERGEN JANE E HENNEY 07Jan2002 Yes

AMERISOURCEBERGEN RODNEY H BRADY 29Aug2001 No

AMERISOURCEBERGEN CORP KURT J HILZINGER 05Mar2004 No

AMGEN PATRICIA C SUELTZ 08Jan2002 Yes

AMSOUTH BANCORPORATION CHARLES D MCCRARY 18Oct2001 No

AMSOUTH BANCORPORATION CLEOPHUS THOMAS 22Jul2002 Yes

ANADARKO PETROLEUM PRESTON M GEREN III 31Jul2000 Yes

ANALOG DEVICES JAMES A CHAMPY 19May2003 No

ANDREW DENNIS L WHIPPLE 07Aug2001 No

ANDREW GUY M CAMPBELL 09Feb2000 No

ANDREW CORP PHILIP W COLBURN 18Feb2003 No

APPLE COMPUTER ARTHUR D LEVINSON 16Aug2000 No

APPLIED MATERIALS MINORU MORIO 22Mar2001 No

APPLIED MATERIALS STAN SHIH 22Feb2000 No

APPLIED MICRO CIRCUITS KEVIN N KALKHOVEN 29Oct2001 No

APPLIED MICRO CIRCUITS L WAYNE PRICE 19Mar2001 No

ARCHER-DANIELS-MIDLAND HERMAN DE BOON 27Oct2000 No

ARCHER-DANIELS-MIDLAND SANDRA ANDREAS MCMURTRIE 01Nov2001 No

ARCHER-DANIELS-MIDLAND CO ALAN L BOECKMANN 04Nov2004 No

ASHLAND RALPH E GOMORY 21Sep2000 No

ASHLAND ROGER W HALE 18Jul2001 No

AT&T CHARLES H NOSKI 21Feb2002 No

AT&T FRANK C HERRINGER 10Apr2002 No

AT&T LOUIS A SIMPSON 19Jul2000 No

AT&T MASAKI YOSHIKAWA 19Jun2001 No

AT&T SHIRLEY A JACKSON 20Jun2001 No

AT&T CORP HERBERT L HENKEL 23Feb2004 No

AT&T CORP WILLIAM F ALDINGER III 16Jul2003 No

AUTODESK PER-KRISTIAN HALVORSEN 05Apr2000 Yes

AUTOMATIC DATA PROCESSING LESLIE A BRUN 30Jan2003 No

AUTOMATIC DATA PROCESSING R GLENN HUBBARD 19Mar2004 No

AUTONATION INC ALAN S DAWES 30Jun2003 No

AUTOZONE MARSHA J EVANS 26Feb2002 No

AUTOZONE W ANDREW MCKENNA 14Mar2000 No

AVAYA HENRY B SCHACHT 23Oct2000 No

AVAYA PAULA STERN 02Dec2002 No

AVAYA PHILLIP A ODEEN 30Sep2002 No

AVAYA INC ANTHONY P TERRACCIANO 25Feb2003 No

AVERY DENNISON BRUCE KARATZ 30Oct2001 Yes

AVERY DENNISON DEAN A SCARBOROUGH 27Apr2000 No

BAKER HUGHES CLARENCE P CAZALOT JR 14Mar2002 No

BAKER HUGHES J LARRY NICHOLS 07Mar2001 No

BALL CORP ERIK H VAN DER KAAY 28Jan2004 No

BANK OF AMERICA C STEVEN MCMILLAN 25Apr2001 No

BANK OF AMERICA PETER V UEBERROTH 28Jun2001 No

BANK OF AMERICA CORP EDWARD L ROMERO 26Jun2003 Yes

BANK ONE DAVID C NOVAK 20Feb2001 No

BANK ONE JAMES DIMON 28Mar2000 Yes

BANK ONE ROBERT I LIPP 28Feb2003 No

BANK ONE ROBERT I LIPP 28Feb2003 No

BARD (C.R.) HERBERT L HENKEL 19Apr2002 No

BARD (CR) INC THEODORE E MARTIN 09Oct2003 No

BECTON DICKINSON BERTRAM L SCOTT 24Sep2002 No

BECTON DICKINSON & CO EDWARD F DEGRAAN 22Apr2003 No

BELLSOUTH JOSEPH M MAGLIOCHETTI 02Mar2000 No

BELLSOUTH JOSEPH M MAGLIOCHETTI 02Mar2000 No

BEMIS CO DAVID S HAFFNER 06May2004 No

BEST BUY MARK C THOMPSON 29Mar2000 No

BEST BUY CO INC MARY A TOLAN 03Mar2004 No

BEST BUY CO INC MATHEW H PAULL 03Sep2003 No

BEST BUY CO INC RONALD JAMES 03Mar2004 Yes

BIOGEN ECKHARD PFEIFFER 22May2002 No

BIOGEN LAWRENCE C BEST 11Feb2003 No

BLACK & DECKER M ANTHONY BURNS 02Jan2001 No

BMC SOFTWARE JON E BARFIELD 16Jul2001 No

BMC SOFTWARE KATHLEEN A O'NEIL 11Nov2002 No

BMC SOFTWARE ROBERT E BEAUCHAMP 10Jan2001 No

BOEING JOHN M SHALIKASHVILI 01May2000 Yes

BOSTON SCIENTIFIC UWE E REINHARDT 07May2002 No

BROADVISION KLAUS S LUFT 15Feb2000 No

BROWN-FORMAN INA BROWN BOND 01Nov2002 No

BRUNSWICK GRAHAM H PHILLIPS 06Feb2002 No

BRUNSWICK RALPH C STAYER 01May2002 No

BURLINGTON RESOURCES WILLIAM E WADE JR 16Jul2001 No

BURLINGTON RESOURCES INC JAMES A RUNDE 21Jan2004 No

CALPINE GERALD GREENWALD 17Jul2001 Yes

CALPINE KENNETH T DERR 19Apr2001 No

CAMPBELL SOUP PAUL R CHARRON 08Oct2003 Yes

CAMPBELL SOUP CO JOHN F BROCK 23Sep2004 No

CAPITAL ONE FINANCIAL CORP LEWIS HAY III 27Oct2003 Yes

CATERPILLAR INC EDWARD B RUST JR 18Feb2003 No

CENDANT CHERYL D MILLS 20Jun2000 Yes

CENDANT SHELI Z ROSENBERG 05Apr2000 No

CENDANT WILLIAM S COHEN 23Jan2001 Yes

CENDANT CORP GEORGE HERRERA 16Dec2003 Yes

CENDANT CORP RONALD L NELSON 14Apr2003 No

CENTERPOINT ENERGY DERRILL CODY 07May2003 No

CENTERPOINT ENERGY INC ROBERT T O'CONNELL 03Jun2004 No

CENTEX THOMAS J FALK 27May2003 No

CENTURYTEL JOSEPH R ZIMMEL 31Dec2002 No

CERIDIAN WILLIAM J CADOGAN 15Feb2000 No