THE INTERPLAY BETWEEN DIRECTOR COMPENSATION AND CEO COMPENSATION

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 THE INTERPLAY BETWEEN DIRECTOR COMPENSATION AND CEO COMPENSATION Dan Lin, Takming Universiy of Science and Technology Lu Lin, Takming Universiy of Science and Technology ABSTRACT This paper empirically examines he deerminans of direcor compensaion and CEO compensaion and invesigaes wheher direcor compensaion has an effec on CEO compensaion. Based on 713 firms (or 2,852 firm-years) beween 2007 and 2010, we find ha CEO enure is relaed o he abiliy of he CEO in influencing he board s pay deerminaion process. However, siing on he board does no srenghen he CEO s power over he board during he pay negoiaion process. More imporanly, we find evidence of a muual back scraching relaionship beween CEO and he board of direcors. Excess direcor compensaion and CEO compensaion are posiively relaed. The resuls hus suppor Jensen s (1993) argumen ha as he CEO is involved in he selecion of direcors, he monioring role of he board of direcors becomes less effecive. JEL: J33, M52 KEYWORDS: Direcor Compensaion, CEO Compensaion, Board of Direcors INTRODUCTION D ue o he conflics of ineress beween ouside shareholders and managers in he modern corporae srucure, he board of direcors has he fundamenal role of monioring managers o ensure ha managers ac in he ineres of shareholders. However, as he CEO is ofen involved in he selecion of direcors, Jensen (1993) argues ha he board direcors may no be an effecive monior. The board of direcors may become more aligned wih he CEO, hereby compromising he independence of he board. Brick e al. (2006) furher sugges ha when he board of direcors is highly compensaed, hey are less likely o conduc criical monioring of he CEO, referred o as muual back scraching. According o Hermalin and Weisbach (1998), he CEO may also use barriers o monioring, including large boards, inside direcors, CEO dualiy, CEO enure, and CEO membership in nominaing commiee, in an aemp o maximize his compensaion. Therefore, one objecive of his sudy is o examine wheher direcor compensaion has an effec on CEO compensaion by uilizing he excess direcor compensaion variable, which is he residual from he direcor compensaion model. Afer he financial crisis of 2008, he fa ca problem highlighed he execuive compensaion issue. Recenly here have been increasing concerns abou he escalaion in execuive compensaion (Dong and Ozkan, 2008). In paricular, he subsanial rises in execuive pay have far exceeded he increases in underlying firm performance (Gregg e al., 2005). The review of CEO compensaion by Frydman and Jener (2010) shows ha here was a dramaic increase in compensaion levels from he mid 1970s o he early 2000s in he US. Especially in he 1990s, he annual growh raes were more han 10% by he end of he decade. The increase in execuive compensaion is also eviden in firms of all sizes while larger firms have experienced greaer growh. The high level of CEO pay in he U.S. has herefore brough abou considerable debae and a lo of aenion from academia and policy makers regarding execuive compensaion, in paricular, he pay-seing process and he effeciveness of he compensaion conracs. The compensaion packages of he op execuives are se by he board of direcors. Afer he financial crisis, he boards of collapsed firms are asked o hold full responsibiliy because hey have no conduced 11

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 appropriae supervision over op execuives. In his regard, his sudy incorporaes he characerisics of he board of direcors and he effec of direcor compensaion, in addiion o CEO characerisics, when examining he deerminans of CEO compensaion. In shor summary, he objecive of his sudy is wofold. Firs, we analyze he deerminans of direcor compensaion. Based on he direcor compensaion model, we derive he residuals (i.e., excess direcor compensaion ). Secondly, we examine wheher excess direcor compensaion and a se of CEO and direcor characerisics (such as CEO enure, CEO shareholdings and board size) are relaed o CEO compensaion. While he deerminans of CEO compensaion and he pay-for-performance relaionship (Jensen and Murphy, 1990; Main e al., 1996; Brick e al., 2006; Ozkan, 2007) have been exensively researched, he compensaion srucure of he board of direcors as a governance mechanism has received less aenion (Cordeiro e al., 2000; Gregg e al., 2005), in paricular, he inerplay beween direcor compensaion and CEO compensaion (Brick e al., 2006). Accordingly, his sudy makes an imporan conribuion by linking direcor compensaion wih CEO compensaion and examines wheher here is a muual back scraching relaionship beween CEO and he board of direcors. Tha is, wheher he CEO receives higher compensaion when he direcors are paid more. Specifically, we include an excess direcor compensaion variable in he CEO compensaion model. If here is a posiive relaionship beween excess direcor compensaion and CEO compensaion, hen a muual back scraching relaionship beween he board of direcors and he CEO exiss. If a negaive relaionship is observed, i means ha he direcors are effecive moniors of he op managemen. In addiion, his sudy conribues o he lieraure by adoping muliple measures when analyzing direcor compensaion. This allows us o examine he direcor compensaion from differen perspecives. Unlike CEO compensaion, as here is more han one person siing on he board of direcors, he board of direcor compensaion may be measured by he oal direcor compensaion for he enire board, he average direcor compensaion, and he compensaion of he highes paid direcor. Mos of previous sudies rely on one single measure (for example, Becher e al., 2005; Fernandes, 2008) or differeniae compensaion by cash and sock compensaion only (for example, Cordeiro e al., 2000; Brick e al., 2006). These sudies may suffer from he weaknesses inheren in a paricular measure. For example, oal direcor compensaion for he enire board may be influenced by he size of he board. The average direcor compensaion ignores he dispersion wihin each firm and may be disored by exreme values. Using he compensaion of he highes paid direcor may someimes be measuring he compensaion of he CEO. Therefore, i is imporan o consider differen measures. Based on 713 firms (or 2,852 firm-years) beween 2007 and 2010, we find suppor for he muual back scraching relaionship beween he CEO and he direcors. Specifically, excess direcor compensaion and CEO compensaion are posiively relaed. The evidence hus suggess ha he direcors are no good moniors of he CEO. The resuls also suppor Jensen s argumen. As direcors are seleced by he CEO, he effeciveness of direcors monioring of he op managemen is weakened. The remainder of his paper is organized ino five secions. In Secion 2, we review he prior empirical lieraure on direcor and CEO compensaion and develop he hypoheses esed in his sudy. In Secion 3, we describe he daa, mehodology and sample characerisics. In Secion 4, we presen he resuls on direcor compensaion and CEO compensaion. A conclusion is provided in Secion 5. LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT In modern economies, mos companies are characerized by he separaion of ownership and conrol where he ownership is held by diverse shareholders and he conrol is in he hands of op execuives. As a 12

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 resul, shareholders are no able o monior managers acions direcly. According o he agency heory, hese companies are likely o suffer from agency problems. Tha is, managers as he agens may no always ac in he ineres of shareholders (i.e., he principals), hereby giving rise o he conflics of ineress. The governance srucure of he firms, as argued by he agency heoriss, can miigae he poenial agency problem beween managers and shareholders arising from he separaion of ownership and conrol, and herefore, influence he way firms se execuive compensaion packages (Murphy, 2009). In fac, he board of direcors who is responsible for providing advice o he managemen and assising wih sraegy developmen plays a key governance role in monioring op managemen (Fama and Jensen, 1983). The board of direcors also has an essenial role in seing CEO compensaion (Finkelsein and Hambrick, 1988; Boyd, 1994; Barkema and Gomez-Mejia, 1998; Carpener and Sanders, 2002; Chhaochharia and Grinsein, 2009). Therefore, one objecive of his sudy is o examine wheher he board of direcors has influences over CEO compensaion. An early paper by Finkelsein and Hambrick (1988) provides a synhesis on CEO compensaion and suggess ha here are wo main se of facors ha affec CEO compensaion: firs, he marke facors, including managerial labor marke, marginal producs of CEOs, CEO discreion, firm size, firm performance, and human capial; secondly, he power and preferences of he board and CEO. Consisen wih his view, Ozkan (2007) finds ha corporae governance mechanisms have a significan effec on he level of CEO compensaion. Specifically, measures of board and ownership srucures explain a significan amoun of cross-secional variaion in CEO oal compensaion. Barkema and Gomez-Mejia (1998) propose a general research framework on he relaionship beween pay and performance. They argue ha crieria, such as he marke, peer compensaion, individual characerisics, a firm s governance srucure (including ownership srucure, board of direcors, remuneraion commiee, and marke for corporae conrol), and coningencies (such as a firm s sraegy, R&D level, marke growh, indusry concenraion and regulaion, and naional culure), can enhance our undersanding of he deerminans of execuive pay. Moreover, he managerial power heory argues ha excessive CEO pay is due o he greaer power of execuives over direcors ha allows he former o se heir own pay and exrac rens (Bebchuk e al, 2002; Bebchuk and Fried, 2004). An implicaion of he heory is ha enhancing he independence of he board can improve corporae governance and preven managers from exracing rens in he form of higher pay (Guhrie e al., 2012). Therefore, he firs objecive of his sudy is o examine he deerminans of direcor compensaion. Then, we invesigae if CEO characerisics and direcor characerisics, including excess direcor compensaion, have influences over CEO compensaion. Specifically, his sudy adds o he lieraure on execuive compensaion by invesigaing he effec of direcor compensaion on CEO compensaion and esing if here is a muual back scraching relaionship beween he CEO and he board of direcors. The hypoheses of his sudy are developed below. Direcor Compensaion Following Hill and Phan (1991), his sudy uses CEO enure o proxy for CEO s abiliy o exercise influence over he board of direcors. Previous sudies (Hermalin and Weisbach, 1991; Shivdasani and Yermack, 1999) have suggesed ha CEOs can exer influence over he direcor selecion process. Ryan and Wiggins (2001) argue ha he level of CEO enrenchmen and CEO power over he board of direcors increase wih CEO enure. Specifically, hey find ha firms wih long-enured CEOs (i.e., more enrenched managers) discourage board scruiny of managemen and provide weaker incenives o direcors o monior managemen. Therefore, CEO enure is expeced o be negaively associaed wih direcor compensaion. Tha is, he following hypohesis is proposed. 13

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 H1a: CEO enure will be negaively relaed o direcor compensaion. CEO direcor is an imporan corporae governance variable ha accouns for he CEO influence over he board. Previous sudies (Boyd, 1995; Daily and Schwenk, 1996; Conyon and Peck, 1998; Cordeiro and Veliyah, 2003) have mosly used CEO chairman as he proxy; ha is, wheher he CEO is also he chairman of he board of direcors. However, his sudy argues ha even in he case where he CEO is no he chairman and is simply a board of direcor, he sill has he abiliy o exer influence on he board. Hence, his sudy argues ha using a broader definiion, CEO direcor, is a beer proxy. To es for he influence of CEO over he board of direcors, we include a dummy variable, if he CEO also holds a board sea. When a CEO is also a board of direcor, he board is likely o be enrenched. Brick e al. (2006) find ha direcors of firms wih a uniary leadership srucure (ha is, he CEO and he Chairman are he same person) receive higher oal compensaion han direcors of firms wih a dual leadership srucure where he roles of CEO and he chairman are performed by differen persons. They argue ha his is because he uniary leadership srucure reflecs weak governance. Accordingly, we offer he following hypohesis. H1b: CEO direcor will be posiively relaed o direcor compensaion. Firms wih larger boards are expeced o be associaed wih higher direcor compensaion for wo reasons. Firsly, as he number of direcors increases, he oal board compensaion will increase. Secondly, firms wih larger boards are ypically more complex firms and herefore should give higher pay o heir direcors. Therefore, a posiive relaionship beween board size and direcor compensaion is proposed. H1c: Board size will be posiively relaed o direcor compensaion. CEO Compensaion As CEOs build a power base and gain voing conrol over ime, hey may exer greaer influence over board composiion. Consequenly, CEOs may be able o demand compensaion packages ha serve heir own ineress raher han he shareholders (Hill and Phan, 1991; Cordeiro and Veliyah, 2003; Ozkan, 2011). Moreover, Finkelsein and Hambrick (1996) sugges ha he enure of an execuive can affec and proxy for his aiudes owards risk. This is because long-enured execuives have esablished high firm-specific human capial and become less mobile (Hill and Phan, 1991). They will be unwilling o ake on any unnecessary risks ha are likely o bring more harms han benefis. Hill and Phan (1991) furher argue ha he posiive relaionship beween pay and firm risk will be sronger he longer he enure of he CEO. Hence, CEO enure is expeced o be posiively associaed wih CEO compensaion. H2a: CEO enure will be posiively relaed o CEO compensaion. A CEO who is also a board of direcor is likely o obain higher pay since he can no only paricipae in bu also exer influence over he board s pay deerminaion process. Therefore, a posiive relaionship is expeced beween CEO compensaion and CEO direcor. H2b: CEO direcor will be posiively relaed o CEO compensaion. The level of CEO shareholdings shows he exen o which he wealh of he CEO is linked wih firm value and is relaed o he exen of agency problems faced by companies (Dong and Ozkan, 2008). CEOs wih greaer shareholdings in he firm will have sronger incenives o boos he firm s sock value. Therefore, less incenive compensaion is needed for aligning he ineress of CEO and shareholders. Accordingly, CEO shareholdings can ac as a subsiue for CEO compensaion (Cordeiro and Veliyah, 2003) and a negaive relaionship is expeced beween CEO compensaion and CEO shareholdings. 14

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 H2c: CEO shareholdings will be negaively relaed o CEO compensaion. Male CEOs are expeced o receive higher compensaion han female CEOs given ha he CEO marke is predominaed by males. Therefore, we offer he following hypohesis. H2d: Male CEO will be associaed wih higher CEO compensaion. The size of he board affecs he effeciveness of he board in monioring managemen. For example, when he board size grows large, more resource neworks and professional views can be brough o board. However, hese advanages may be overwhelmed by he efficiency losses in communicaion, decision-making and coordinaion beween board members as he number of board members increases. In oher words, a large board may in effec reduce he effeciveness of board monioring and herefore be associaed wih higher CEO compensaion. Consisen wih he laer view, Core e al. (1999) repor ha larger boards pay more o heir CEOs in erms of boh cash compensaion and oal compensaion. Based on a sample of 414 UK companies beween 2003 and 2004, Ozkan (2007) also repors ha firms wih larger board size are associaed wih higher CEO compensaion, measured by oal compensaion and cash compensaion. Moreover, Gues (2010) examines a comprehensive and long period daase of 1,880 UK firms over he period 1983-2002 and repors a posiive relaionship beween board size and he rae of increase in execuive compensaion, providing suppor for he argumen ha large boards suffer from he problems of less efficien decision-making and poor communicaion. Therefore, his sudy expecs a posiive relaionship beween board size and CEO compensaion. H2e: Board size will be posiively relaed o CEO compensaion. To examine he impac of direcor compensaion on CEO compensaion, we include he residuals from he direcor compensaion model in he CEO compensaion model, i.e., he excess direcor compensaion. While he pay of he CEO is deermined by he board of direcors, he CEO is involved in he selecion of he board of direcors. Therefore, his sudy expecs a muual back scraching relaionship beween he CEO and he board of direcors; ha is, a posiive relaionship beween excess direcor compensaion and CEO compensaion. Specifically, his sudy ess if CEOs receive a higher pay when direcors are being paid higher. H2f: Excess direcor compensaion will be posiively relaed o CEO compensaion. Conrol Variables To conrol for oher variables documened in previous lieraure as imporan in deermining compensaion levels, he following variables are also included in he models. Firm size conrols for he fac ha larger firms which are ypically more complex will require direcors o spend more ime and pu more effor in monioring managers. In oher words, larger firms are associaed wih greaer complexiy and informaion processing demands and herefore, direcors of larger firms are expeced o receive higher compensaion. Hence, a posiive relaionship is expeced beween direcor compensaion and firm size. Similarly, CEOs of larger firms have greaer responsibiliy, require more effor, and herefore are expeced o be more highly compensaed (Smih and Was, 1992; Core e al. 2003). The sudy by Conyon (1997) has repored a significanly posiive relaionship beween firm size and CEO compensaion levels. Accordingly, a posiive relaionship is also expeced beween CEO compensaion and firm size. Agency heory suggess ha one way o align he ineress of managers wih ha of shareholders is o ie he compensaion conracs o firm performance (Firh e al., 2006; Chhaochharia and Grinsein, 2009); ha is, o creae a pay-for-performance linkage. In oher words, o moivae direcors o acively monior managers on behalf of shareholders, direcors should be rewarded when firm performance is high. 15

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 Therefore, we expec a posiive relaionship beween direcor compensaion levels and firm performance. Similarly, making he CEOs hold accounable for firm performance is essenial for moivaing he CEOs o iniiae sraegies ha boos firm value. Hence, a posiive relaionship beween CEO compensaion and firm performance is also expeced. The pay of direcors and CEOs is likely o be se wih reference o he pay of oher direcors and CEOs in he same indusry. Hilburn (2010) repors ha direcors of echnology companies have higher pay han heir counerpars a general indusry companies. Therefore, he differences in indusry srucures, complexiy and indusry cusoms are likely o affec he level of compensaion (Hempel and Fay, 1994). Hence, his sudy includes a dummy variable for indusry secors o conrol for iner-indusry differences in compensaion levels. Year dummies are also included in our models o conrol for unobserved differences beween years. The inclusion of hese dummies can capure common facors ha are driven by indusry- and economy-wide effecs. DATA AND METHODOLOGY The daa used in his sudy are obained from he Sandard and Poor s ExecuComp daabase. To be included in he sample, he sample firms mus have all he required financial informaion, such as oal asses, sales, ROA and ROE, CEO compensaion, and direcor compensaion daa. As he informaion on direcor compensaion in ExecuComp daabase is more complee from he year 2006 and onwards, he sample period for his sudy is se beween 2007 and 2010. Previous lieraure has suggesed ha banks are likely o face greaer poenial conflics of ineress han indusrial firms due o is disinc characerisics such as he exisence of deposi insurance, high deb-o-equiy raios and asse-liabiliy issues (Becher e al., 2005). Since he naure of financial services indusry is differen from ha of indusrial firms, firms belonging o he financial services indusry are excluded from he sample. Therefore, our sample begins wih a oal of 940 firms (or 3760 firm-years). Afer eliminaing 28 firms wih missing daa and 199 firms in he finance, insurance and real esae indusries (ha is, Division H of he SIC division srucure), he final sample consiss of 713 firms (or 2,852 firm-years). The hypoheses are esed using pooled ime-series cross-secional regression analysis. The wo models esed in his sudy are oulined below. Model 1 is on direcor compensaion and Model 2 is on CEO compensaion. ln + ( DIRCOMPi, ) = αi, + β1ceotenurei, + β2ceodiri, + β3ln( BSIZEi, ) β ln( FSIZEi ) + β5performancei + INDUSDUM + YEARDUM 4,, 1 The dependen variable (DIRCOMP) of Model 1 is measured in hree ways, he oal direcor compensaion, he average compensaion of direcors, and he compensaion of he highes paid direcor. Firsly, he oal direcor compensaion is he direcors oal compensaion for he enire board, including cash fees, sock awards, opion awards, non-equiy incenive plan compensaion, change in pension value and non-qualified deferred compensaion earnings, and oher compensaion provided by ExecuComp daabase. The reason for measuring direcor compensaion for he enire board is ha i is he board collecively ha moniors for and acs on behalf of he shareholders. Secondly, he average direcor compensaion is he per capia compensaion of direcors (Fernandes, 2008), where he compensaion is measured in oal and includes cash fees, sock awards, opion awards, non-equiy incenive plan compensaion, change in pension value and non-qualified deferred compensaion earnings, and oher compensaion provided by ExecuComp daabase. One weakness wih his measure is ha measuring direcor compensaion as an average ignores he dispersion wihin each firm. (1) 16

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 As mos sudies focus on he CEO who holds he op paying job, his sudy also analyzes he highes paid person on he board; i.e., he hird measure of direcor compensaion in his sudy. Gregg e al. (1993) who examine he relaionship beween direcors pay and corporae performance also adop his measure. Formally, he compensaion of he highes paid direcor is he oal compensaion of he highes paid direcor, where oal compensaion includes cash fees, sock awards, opion awards, non-equiy incenive plan compensaion, change in pension value and non-qualified deferred compensaion earnings, and oher compensaion provided by ExecuComp daabase. The definiions of independen and conrol variables are as follows. CEO enure (CEOTENURE) is measured by he number of years he CEO had held he posiion in a given company. An alernaive measure for CEO enure is he age of he CEO (CEOAGE), which is expeced o have srong posiive correlaion wih CEO enure and also proxies for CEO experience. CEO direcor (CEODIR) is a dummy variable ha equals one if he CEO is also a board of direcor. Board size (BSIZE) is measured by he number of direcors on he board. Firm size (FSIZE) is measured by oal asses and sales. Firm performance (PERFORMANCE) is measured by he reurn on asses (ROA) and reurn on average equiy (ROE), which are lagged one year in order o avoid measuring he effec of compensaion on performance. The lagged performance measure can also accoun for he fac ha direcor compensaion paid in one year is usually deermined by he firm performance in he previous year. ROA has been widely used in previous sudies on execuive compensaion and corporae governance as a proxy for firm performance. ROA shows how efficien he firm is in uilizing is asses (Finkelsein and Hambrick, 1996; Finkelsein and Boyd, 1998; Carpener and Sanders, 2002). On he oher hand, ROE can beer reflec firm performance from he shareholders poin of view. Therefore, in his sudy, models are esimaed separaely using boh measures. Indusry (INDUSDUM) is deermined by SIC division srucure, ranging from Division A o J (Descripions for he SIC division srucure are oulined below. Division A: agriculure, foresry, and fishing; Division B: mining; Division C: consrucion; Division D: manufacuring; Division E: ransporaion, communicaions, elecric, gas, and saniary services; Division F: wholesale rade; Division G: reail rade; Division I: services; Division J: public adminisraion.) Noe ha Division H, he finance, insurance, and real esae indusries, is excluded from he sample. In his sudy we also include year dummies (YEARDUM). Based on Model 1, we derive he excess direcor compensaion (EXDIRCOMP), which is he residual from he direcor compensaion model when oal direcor compensaion is used as he dependen variable. The excess direcor compensaion measures he exen of direcor under- or overpaymen. This variable is hen included in he second model on CEO compensaion, as oulined below, o es he impac of direcor compensaion on CEO compensaion. ( CEOCOMP ) ln i, = αi, + β1ceotenurei, + β2ceodiri, + β3ceoholdingi, + β 4CEOGENDERi, + β5ln( BSIZEi, ) + β6exdircompi, + β7ln( FSIZEi, ) (2) + β PERFORMANCE + INDUSDUM + YEARDUM 8 i, 1 The dependen variable (CEOCOMP) of Model 2 is measured in wo ways, CEO oal compensaion and CEO cash compensaion. Ozkan (2011) suggess ha firm performance may affec cash and equiy-based componens of compensaion differenly. I is imporan o incorporae muliple measures for compensaion. In his sudy, he CEO oal compensaion comprises salary, bonus, oher annual paymen, resriced sock grans, long-erm incenive payous, value of opions graned and all oher paymens provided by ExecuComp daabase. The second measure, CEO cash compensaion, consiss of salary and bonus. 17

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 The addiional variables inroduced in he second model are defined as follows. CEO shareholdings (CEOHOLDING) is calculaed as he shares owned by he CEO, excluding opions ha are exercisable or will become exercisable wihin 60 days, divided by he number of common shares ousanding. CEO gender (CEOGENDER) is a dummy variable ha equals one if he CEO is male. Excess direcor compensaion (EXDIRCOMP) is he residual from he direcor compensaion model where he dependen variable is he oal direcor compensaion. Table 1 presens he descripive saisics of CEO characerisics and CEO compensaion for 713 sample firms. The average and median age of CEOs is boh 55, ranging from 34 o 80. The mean CEO ownership is 1.53% and ranges from 0 o 75.8% of ousanding shares. CEO enure, which measures he number of years he CEO had held he posiion in a given company, has an average of 7.2 years and ranges from 0 o 47 years. The mean (or median) value of cash compensaion, which consiss of salary and bonus, received by he CEOs of our sample firms is $1,116,474 (or $875,158). The oal compensaion has an average of $5,838,773 and ranges from $30,002 o $128,706,100. In our sample, abou 96.6% of CEOs are male and 96.8% of CEOs also hold a board sea. Table 1: Descripive Saisics of CEO Characerisics and CEO Compensaion Mean Median Max Min SD CEO characerisics CEO age 55 55 80 34 6.67 CEO shareholdings (%) 1.53 0.29 75.80 0.00 4.91 CEO enure (years) 7.22 5.00 47.00 0.00 6.68 CEO cash compensaion ($'000) 1,116.5 875.2 77,926 7.1 2,466.3 CEO oal compensaion ($'000) 5,838.8 4,076.8 128,706 30.0 6,722.0 CEO gender Male 2754 96.56% Female 98 3.44% Toal 2852 100.00% CEO is also a board of direcor Yes 2760 96.77% No 92 3.23% Toal 2852 100.00% This able repors he descripive saisics of CEO characerisics and CEO compensaion for 713 firms (or 2,852 firm-years) beween 2007 and 2010. CEO shareholdings is calculaed as shares owned by he CEO, excluding opions ha are exercisable or will become exercisable wihin 60 days, divided by he oal number of common shares ousanding. CEO cash compensaion includes salary and bonus. CEO oal compensaion includes salary, bonus, oher annual, oal value of resriced sock graned, oal value of sock opions graned (using Black-Scholes), long-erm incenive payous, and all oher compensaion. The descripive saisics for firm characerisics and direcor compensaion are shown in Table 2. The average board size is 9 and ranges from 3 o 26 direcors. The average firm size, measured by oal asses, is $9,898 million and $7,849 million if measured by sales. Firm performance is measured by ROA and ROE. The average ROA and ROE are 3.99% and 9.99%, respecively. The mean and median average direcor compensaion per board is $181,794 and $166,643, respecively. The mean oal direcor compensaion per board is $1,597,003 and ranges from $33,374 o $14,685,740. 18

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 Table 2: Descripive Saisics of Firm Characerisics and Direcor Compensaion Mean Median Max Min SD Firm characerisics Board size 9 8 26 3 2.46 Toal asses ($m) 9,898.2 2,345.1 797,769 10.0 36,640 Sales ($m) 7,849.1 2,034.8 425,071 0.1 25,797 ROA (%) 3.99 5.17 52.85-163.38 11.77 ROE (%) 9.99 12.24 524.38-906.03 36.19 Direcor compensaion per board DIRCOMP_Average ($'000) 181.8 166.6 1,796 3.6 114.6 DIRCOMP_Maximum ($'000) 305.6 221.4 7,779 13.6 430.3 DIRCOMP_Toal ($'000) 1,597.0 1,402.6 14,686 33.4 1,081.1 This able repors he descripive saisics of firm characerisics and direcor compensaion for 713 firms (2,852 firm-years) beween 2007 and 2010. DIRCOMP_Average is he average direcor compensaion for each firm (or each board), ha is, he per capia compensaion of direcors. DIRCOMP_Maximum is he compensaion of he highes paid direcor in each firm. DIRCOMP_Toal is he oal direcor compensaion for he enire board. Direcor compensaion is defined o include cash fees, sock awards, opion awards, non-equiy incenive plan compensaion, change in pension value and non-qualified deferred compensaion earnings, and oher compensaion Table 3 repors he descripive saisics for he componens of direcor compensaion. Beween 2007 and 2010, here is a oal of 24,604 direcor-years. The average cash fees paid o direcors is $71,708,000. The direcors in our sample receive an average of $73,103,000 in sock awards, $28,111,000 in opion awards, and $515,000 in non-equiy incenive plan. The oal direcor compensaion has an average of $185,118,000 and ranges from -$1,299,073,000 o $7,778,702. The negaive oal compensaion can be aribued o he negaive amouns in sock and opion awards and he negaive change in pension value and non-qualified deferred compensaion earnings. Table 3: Descripive Saisics of Direcor Compensaion Mean Median Max Min SD Componens of direcor compensaion Cash fees ($'000) 71.71 68.39 777.2 0.0 44.35 Sock awards ($'000) 73.10 60.69 7,612.0-362.1 94.70 Opion awards ($'000) 28.11 0.00 4,939.6-1,886.1 95.45 Non-equiy incenive ($'000) 0.52 0.00 2,619.0 0.0 29.02 Pension change ($'000) 0.96 0.00 406.0-805.3 12.23 Oher compensaion ($'000) 10.68 0.00 6,004.4 0.0 94.02 Toal compensaion ($'000) 185.12 165.50 7,778.7-1,299.1 182.22 This able repors he descripive saisics of direcor compensaion for 24,604 direcor-years beween 2007 and 2010. Direcor compensaion is classified as cash fees, sock awards, opion awards, non-equiy incenive plan, change in pension value and non-qualified deferred compensaion earnings, and all oher compensaion. Cash fees are direcor fees ha are earned or paid in cash. Sock awards are measured by he value of sock-relaed awards (e.g. resriced sock, resriced sock unis, phanom sock, phanom sock unis, common sock equivalen unis ec.) ha do no have opion-like feaures. Opion awards are measured by he value of opion-relaed awards (e.g. opions, sock appreciaion righs, and oher insrumens wih opion-like feaures). Non-equiy incenive is measured by he value of amouns earned during he year pursuan o non-equiy incenive plans. Pension change is composed of above-marke or preferenial earnings from deferred compensaion plans and aggregae increase in acual value of defined benefi and acual pension plans during he year. Oher compensaion includes perquisies and oher personal benefis, conribuions o defined conribuion plans, life insurance premiums, gross-ups and oher ax reimbursemens, discouned share purchases, consuling fees, awards under chariable award programs ec. Table 4 repors he correlaions beween variables. Overall, he CEO and direcor compensaion are posiively relaed o board size, firm performance, measured by ROA and ROE, and firm size, measured by oal asses and sales. The CEO shareholdings are negaively associaed wih CEO compensaion, suggesing a subsiuion effec beween CEO shareholdings and CEO compensaion (Cordeiro and Veliyah, 2003). Consisen wih he expecaion, he CEO enure, a proxy for CEO power, is posiively relaed o CEO compensaion and negaively relaed o direcor compensaion. 19

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 Table 4: Correlaion Marix 1 2 3 4 5 6 7 8 9 10 11 12 13 1.CEO age 1 2.CEO shareholdings 0.11 *** 1 3.CEO enure 0.42 *** 0.38 *** 1 4.Board size 0.07 *** -0.24 *** -0.22 *** 1 5.Asses 0.03-0.06 *** -0.05 ** 0.32 *** 1 6.Sales 0.05 *** -0.07 *** -0.07 *** 0.29 *** 0.67 *** 1 7.ROE -1 0.00-0.01 0.01 0.07 *** 0.05 *** 0.08 *** 1 8.ROA -1 0.00 0.02 0.02 0.04 ** 0.03 * 0.07 *** 0.67 *** 1 9.CEOCOMP_CASH 0.10 *** -0.02 0.06 *** 0.10 *** 0.14 *** 0.11 *** 0.02 0.01 1 10.CEOCOMP_TOT 0.12 *** -0.12 *** 0.00 0.34 *** 0.28 *** 0.32 *** 0.11 *** 0.09 *** 0.65 *** 1 11.DIRCOMP_AVE 0.01-0.15 *** -0.03 * 0.10 *** 0.16 *** 0.15 *** 0.09 *** 0.07 *** 0.15 *** 0.37 *** 1 12.DIRCOMP_MAX -0.03 * -0.07 *** -0.07 *** 0.10 *** 0.06 *** 0.06 *** 0.02 0.02 0.05 *** 0.16 *** 0.73 *** 1 13.DIRCOMP_TOT 0.05 *** -0.19 *** -0.10 *** 0.51 *** 0.33 *** 0.30 *** 0.11 *** 0.08 *** 0.18 *** 0.51 *** 0.85 *** 0.66 *** 1 This able repors he correlaions of variables used in he regression analysis for a sample of 713 firms during he period 2007-2010. CEOCOMP_CASH denoes CEO cash compensaion. CEOCOMP_TOT denoes CEO oal compensaion. DIRCOMP_AVE denoes he average direcor compensaion. DIRCOMP_MAX denoes he compensaion of he highes paid direcor. DIRCOMP_TOT denoes he oal direcor compensaion. RESULTS Table 5 repors OLS esimaion resuls for direcor compensaion. In Panel A, he dependen variable is oal direcor compensaion, measured by he direcors oal compensaion for he enire board. In Panel B, he dependen variable is he average compensaion of direcors, which is measured as he per capia compensaion of direcors, where he compensaion is measured in oal. In Panel C, he dependen variable is he oal compensaion of he highes paid direcor. For each measure of direcor compensaion (i.e., in each panel), Model 1 is esimaed four imes as we have adoped alernaive measures for CEO enure (i.e., CEO enure and CEO age), firm size (i.e., oal asses and sales), and firm performance (i.e., ROE and ROA). The regression esimaes in Table 5 show ha CEOs wih shorer enure or younger age are significanly associaed wih higher direcor compensaion a he 1% level. This finding is consisen wih our predicion ha shor-enured CEOs have less abiliy o exercise influence over he board of direcors. The resul is consisen across hree measures of direcor compensaion. Inconsisen wih our expecaion, CEO direcor dummy variable is negaively associaed wih he direcor compensaion, significan a he 1% level. In oher words, he direcor compensaion is higher when he CEO is no a member of he board. The resul suggess ha wihou he influence of CEO over he board, direcors are able o se higher compensaion o favor hemselves. Board size is significanly posiively relaed o oal direcor compensaion and he compensaion of he highes paid direcor a he 1% level. However, i is significanly negaively relaed o he average compensaion of direcors. This is because as he number of board members increases, he oal direcor compensaion per board evens ou, leading o a negaive relaionship. Firm size, measured by oal asses and sales, are also are significanly posiively relaed o direcor compensaion. Ineresingly, he sudy by Song and Xu (2007) based on a sample of Chinese lised companies finds ha he oal compensaion received by board of direcors is negaively associaed wih board size, CEO enure and he proporion of inside direcors. They sugges ha when he board lacks independence, he execuives will dominae over direcors, resuling in less compensaion o direcors. Consisen wih Song and Xu (2007), his sudy finds ha direcors of larger firms receive more compensaion. 20

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 Table 5: Analysis of Direcor Compensaion Panel A: Dependen variable: ln(dircomp_oal) Panel B: Dependen variable: ln(dircomp_average) 1 2 3 4 1 2 3 4 Inercep 3.877 *** 5.074 *** 4.003 *** 3.864 *** 3.877 *** 5.074 *** 4.003 *** 3.864 *** (55.437) (21.725) (53.237) (51.494) (55.437) (21.725) (53.237) (51.494) CEOTENURE -0.008 *** -0.008 *** -0.008 *** -0.008 *** -0.008 *** -0.008 *** (-10.478) (-10.750) (-10.497) (-10.478) (-10.750) (-10.497) ln(ceoage) -0.339 *** -0.339 *** (-7.337) (-7.337) CEODIR -0.101 *** -0.108 *** -0.098 *** -0.101 *** -0.101 *** -0.108 *** -0.098 *** -0.101 *** (-5.935) (-6.132) (-5.891) (-6.275) (-5.935) (-6.132) (-5.891) (-6.275) ln(bsize) 0.843 *** 0.889 *** 0.941 *** 0.843 *** -0.157 *** -0.111 *** -0.059 ** -0.157 *** (34.922) (37.732) (33.208) (31.848) (-6.497) (-4.709) (-2.067) (-5.940) ln(assets) 0.196 *** 0.198 *** 0.198 *** 0.196 *** 0.198 *** 0.198 *** (44.784) (44.852) (40.955) (44.784 ) (44.852) (40.955) ln(sales) 0.172 *** 0.172 *** (36.447) (36.447) ROE -1 0.001 *** 0.001 *** 0.001 *** 0.001 *** 0.001 *** 0.001 *** (4.032) (3.813) (2.609) (4.032) (3.813) (2.609) ROA -1 0.001 * 0.001 * (1.754) (1.754) Indusry and year dummies Yes Yes Yes Yes Yes Yes Yes Yes Adjused R 2 0.546 0.544 0.521 0.545 0.303 0.299 0.264 0.301 Panel C: Dependen variable: ln(dircomp_max) 1 2 3 4 Inercep 4.208 *** 6.150 *** 4.297 *** 4.202 *** (24.094) (14.841) (24.679) (23.425) CEOTENURE -0.015 *** -0.015 *** -0.015 *** (-8.503) (-8.603) (-8.465) ln(ceoage) -0.555 *** (-6.132) CEODIR -0.114 *** -0.126 *** -0.112 *** -0.115 *** (-3.472) (-3.854) (-3.364) (-3.481) ln(bsize) 0.178 *** 0.258 *** 0.245 *** 0.177 *** (10.925) (12.204) (11.519) (10.251) ln(assets) 0.152 *** 0.155 *** 0.153 *** (68.617) (84.134) (59.841) ln(sales) 0.136 *** (71.922) ROE -1 0.000 0.000 0.000 (1.369) (0.904) (0.674) ROA -1 0.000 (0.116) Indusry and year dummies Yes Yes Yes Yes Adjused R 2 0.208 0.198 0.195 0.207 This able presens he regression analysis of direcor compensaion for 713 firms beween 2007 and 2010, where he direcor compensaion includes cash fees, sock awards, opion awards, non-equiy incenive plan, change in pension value and non-qualified deferred compensaion earnings, and all oher compensaion. In Panel A, B and C, he dependen variables are oal direcor compensaion, average direcor compensaion and he compensaion of he highes paid direcor, respecively. CEOTENURE is measured by he number of years he CEO had held he posiion in a given company. CEOAGE is he age of he CEO. CEODIR is a dummy variable ha equals one if he CEO is also a board of direcor. BSIZE is measured by he number of direcors on he board. ASSETS and SALES are measures for firm size. ROE and ROA measure firm performance and are lagged one year. -saisics (in parenheses) are calculaed using Whie's (1980) heeroskedasiciy-consisen sandard errors. ***, ** and * indicae coefficien is significan a he 1, 5 and 10% level, respecively. 21

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 The resuls show ha direcor compensaion, measured by oal direcor compensaion and average compensaion of direcor, is higher when firms have beer pas performance, supporing he argumen ha compensaion conracs should be linked o firm performance. However, when he direcor compensaion is measured by he compensaion of he highes paid direcor, he significan relaionship wih pas firm performance disappears. In oher words, highly paid direcors are ofen no paid based on heir performance. This finding suppors he recen call for reviewing he compensaion packages of fa ca direcors (Dong and Ozkan, 2008). The evidence also suggess ha for highly paid direcors, he pay-for-performance linkage ofen does no exis. In paricular, Gregg e al. (2005) argue ha he subsanial rises in execuive pay have far exceeded he increases in underlying firm performance. Moreover, in erms of he firm performance measures, we find ha ROE is a beer predicor of direcor compensaion han ROA. This can be explained by he fac ha ROE can beer reflec how well a firm performs from he shareholders poin of view. The OLS esimaion resuls for CEO compensaion by incorporaing he effec of direcor compensaion are presened in Table 6. The dependen variable is CEO oal compensaion for Panel A and CEO cash compensaion for Panel B. The former measure comprises he CEO s salary, bonus, oher annual paymen, resriced sock grans, long-erm incenive payous, value of opions graned and all oher paymens. The laer measure consiss of salary and bonus only. For each measure of CEO compensaion (i.e., in each panel), Model 2 is esimaed four imes as we have adoped alernaive measures for CEO enure (i.e., CEO enure and CEO age), firm size (i.e., oal asses and sales), and firm performance (i.e., ROE and ROA).The resuls show ha excess direcor compensaion is significanly posiively relaed o CEO compensaion a he 1% level. This finding suppors our hypohesis ha CEOs receive higher pay when he direcors are paid higher. Accordingly, he evidence is consisen wih he argumen of a muual back scraching relaionship beween he CEO and he board of direcors (Brick e al., 2006). The resuls also sugges ha direcors are no good moniors of he op managemen and suppor Jensen s (1993) argumen ha he effeciveness of direcors monioring role can be weakened by he fac ha direcors are seleced by he CEO. Consisen wih he expecaion, CEO enure and CEO age are posiively relaed o CEO compensaion. Alhough he level of significance is weaker when he CEO compensaion is measured by CEO oal compensaion, CEO enure and CEO age are significanly relaed o CEO cash compensaion a he 1% level. Inconsisen wih our expecaion, CEO direcor dummy variable is negaively relaed o CEO compensaion a he 5% significance level. Tha is, CEO compensaion is higher when he CEO does no hold he board sea. Therefore, he observed high compensaion received by CEOs ha we observe oday canno be explained by heir presence on he board of direcors. Moreover, he resul does no suppor he argumen ha dual leadership where he roles of CEO and he chairman are performed by differen people is associaed wih beer governance and herefore lower CEO compensaion. Addiionally, he resuls demonsrae ha CEO shareholdings are significanly negaively associaed wih CEO oal compensaion a he 1% level, providing suppor for he hypohesis ha CEO shareholdings and CEO compensaion conracs are subsiue mechanisms for aligning he ineress of CEO and shareholders (Cordeiro and Veliyah, 2003). However, CEO shareholdings are insignificanly associaed wih CEO cash compensaion. This is because he cash componen of CEO compensaion conracs does no link CEO wealh wih firm value, and herefore, does no have he subsiuion effec like CEO oal compensaion. Ineresingly, we find ha he gender of CEOs is significanly relaed o CEO cash compensaion bu no CEO oal compensaion. Specifically, he resuls show ha male CEOs receive higher cash compensaion. Board size and firm size are significanly posiively relaed o CEO compensaion a he 1% level, consisen wih he hypohesis. Since larger firms are ypically more complex and have larger boards, CEOs of larger firms are herefore more highly compensaed. Ineresingly, we find ha boh measures of firm performance, ROE and ROA, canno explain CEO oal compensaion, herefore, providing evidence agains he pay-for-performance linkage ha have been 22

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 raised by he popular press. Consisen wih he finding of his sudy, Ozkan (2007) does no find a significan relaionship beween CEO compensaion and firm performance based a sample of large UK companies for he fiscal year 2003/2004. Table 6: Analysis of CEO Compensaion Panel A: Dependen variable: ln(ceocomp_oal) Panel B: Dependen variable: ln(ceocomp_cash) 1 2 3 4 1 2 3 4 Inercep 4.087 *** 3.095 *** 4.352 *** 4.088 *** 5.155 *** 2.946 *** 5.280 *** 5.146 *** (23.987) (12.027) (24.535) (24.673) (29.570) (15.665) (29.230) (28.565) CEOTENURE 0.002 0.003 * 0.002 0.007 *** 0.007 *** 0.007 *** (1.636) (1.917) (1.635) (13.410) (14.502) (15.361) ln(ceoage) 0.261 *** 0.587 *** (4.819) (43.189) CEODIR -0.131 ** -0.125 ** -0.124 ** -0.130 ** -0.047 ** -0.033-0.043 ** -0.049 ** (-2.172) (-2.084) (-2.094) (-2.142) (-2.286) (-1.641) (-2.241) (-2.575) CEOHOLDING -0.009 *** -0.009 *** -0.011 *** -0.009 *** -0.002-0.001-0.003-0.002 (-6.668) (-5.787) (-7.335) (-6.661) (-1.269) (-0.405) (-1.471) (-1.228) CEOGENDER 0.050 0.041 0.038 * 0.047 0.028 *** 0.009 0.019 ** 0.034 *** (1.449) (1.145) (1.908) (1.304) (2.816) (0.854) (2.520) (3.410) ln(bsize) 0.143 *** 0.131 *** 0.310 *** 0.145 *** 0.120 *** 0.087 *** 0.191 *** 0.115 *** (3.671) (3.021) (7.351) (3.551) (17.770) (17.702) (15.542) (20.068) EXDIRCOMP 0.404 *** 0.407 *** 0.412 *** 0.405 *** 0.084 *** 0.092 *** 0.088 *** 0.083 *** (13.220) (12.899) (13.181) (13.130) (6.692) (8.069) (7.032) (6.865) ln(assets) 0.433 *** 0.431 *** 0.433 *** 0.223 *** 0.219 *** 0.225 *** (109.245) (113.227) (118.890) (45.888) (46.982) (56.629) ln(sales) 0.395 *** 0.208 *** (53.933) (37.799) ROE -1 0.000 0.000 0.000 0.000 ** 0.000 ** -0.001 ** (0.885) (0.944) (0.147) (-2.332) (-2.327) (-2.383) ROA -1 0.001-0.002 *** (1.594) (-10.616) Indusry and year dummies Yes Yes Yes Yes Yes Yes Yes Yes Adjused R 2 0.615 0.616 0.565 0.615 0.412 0.422 0.386 0.414 This able presens he regression analysis of CEO compensaion for 713 firms beween 2007 and 2010. In Panel A and B, he dependen variables are CEO oal compensaion and CEO cash compensaion, respecively. CEOTENURE is measured by he number of years he CEO had held he posiion in a given company. CEOAGE is he age of he CEO. CEODIR is a dummy variable ha equals one if he CEO is also a board of direcor. CEOHOLDING is calculaed as shares owned by he CEO divided by he number of common shares ousanding. CEOGENDER is a dummy variable ha equals one if he CEO is male. BSIZE is measured by he number of direcors on he board. EXDIRCOMP is he residual from he oal board compensaion model where he dependen variable is he oal board compensaion. ASSETS and SALES are measures for firm size. ROE and ROA measure firm performance and are lagged one year. -saisics (in parenheses) are calculaed using Whie's (1980) heeroskedasiciy-consisen sandard errors. ***, ** and * indicae coefficien is significan a he 1, 5 and 10% level, respecively. CONCLUDING COMMENTS The global financial crisis in 2008 sheds ligh on he significance of reviewing he compensaion packages of op execuives. Based on a sample of 713 US firms beween 2007 and 2010, his sudy examines he deerminans of direcor and CEO compensaion based on a number of board of direcor and CEO characerisics. We also invesigaes wheher here is a muual back scraching relaionship beween he CEO and he board of direcors by analyzing he relaionship beween direcor compensaion and CEO compensaion. Specifically, his sudy proposes wo empirical models. The firs is on direcor compensaion and he second is on CEO compensaion. 23

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 The resuls show ha CEOs wih shorer enure or younger age are associaed wih higher direcor compensaion bu lower CEO compensaion. This finding provides suppor for he argumen ha CEO enure or CEO age is relaed o CEO s abiliy o influence he board s pay deerminaion process. Ineresingly, we find ha CEO who also holds a board sea is no associaed wih higher CEO compensaion. The resul hus indicaes ha siing on he board of direcors does no srenghen he CEO s power over he board during he pay negoiaion process. More imporanly, he resuls sugges ha CEOs receive higher pay when he direcor compensaion is higher, supporing he muual back scraching relaionship beween he CEO and he board of direcors. There is also a subsiuion effec beween CEO oal compensaion and he level of CEO ownership. Finally, firms wih larger board size and firm size give higher pay o heir direcors and CEOs. One limiaion of his sudy is ha due o he consrain on he availabiliy of board of direcors daa, he sample period of his sudy is limied o four years only. Fuure research could exend he sample period by dropping he board size variable o see if similar resuls can be reached. REFERENCES Barkema, H.G. and L.R. Gomez-Mejia1(1998) Managerial compensaion and firm performance: A general research framework, Academy of Managemen Journal, vol. 41, p. 135-145. Bebchuk, L.A. and J. Fried (2004) Pay wihou performance: The unfulfilled promise of execuive compensaion, Cambridge, MA: Harvard Univ. Press. Bebchuk, L.A., J. Fried and D. Walker (2002) Managerial power and ren exracion in he design of execuive compensaion, Universiy of Chicago Law Review, vol. 69, p. 751-846. Becher, D.A., T.L. Campbell II and M.B. Frye (2005) Incenive compensaion for bank direcors: The impac of deregulaion, Journal of Business, vol. 78, p. 1753-1777. Boyd, B.K. (1994) Board conrol and CEO compensaion, Sraegic Managemen Journal, vol. 15, p. 335-344. Boyd, B.K. (1995) CEO dualiy and firm performance: A coningency model, Sraegic Managemen Journal, vol. 16, p. 301-312. Brick, I.E., O. Palmon and J.K. Wald (2006) CEO compensaion, direcor compensaion, and firm performance: Evidence of cronyism? Journal of Corporae Finance, vol. 12, p. 403-423. Carpener, M.A. and W.M.G. Sanders (2002) Top managemen eam compensaion: The missing link beween CEO pay and firm performance? Sraegic Managemen Journal, vol. 23, p. 367-375. Chhaochharia, V. and Y. Grinsein (2009) CEO compensaion and board srucure, Journal of Finance, vol. 64, p. 231-261. Conyon, M.J. (1997) Corporae governance and execuive compensaion, Inernaional Journal of Indusrial Organizaion, vol. 15, p. 493-509. Conyon, M.J. and S.I. Peck (1998) Board conrol, remunearion commiees, and op managemen compensaion, Academy of Managemen Journal, vol. 41, p. 146-157. Cordeiro, J. and R. Veliyah (2003) Beyond pay for performance: A panel sudy of he deerminans of CEO compensaion, American Business Review, vol. 21, p. 56-66. 24

The Inernaional Journal of Business and Finance Research VOLUME 8 NUMBER 2 2014 Cordeiro, J., R. Veliyah, E.J. Eramus and M. Tax (2000) An empirical invesigaion of he deerminans of ouside direcor compensaion, Corporae Governance: An Inernaional Review, vol. 8, p. 268-279. Core, J.E., W.R. Guay and D.F. Larcker (2003) Execuive equiy compensaion and incenives: A survey, Federal Reserve Bank of New York Economic Policy Review, vol. 9, p. 27-50. Core, J.E., R.W. Holhausen and D.F. Larcker (1999) Corporae governance, chief execuive officer compensaion, and firm performance, Journal of Financial Economics, vol. 51, p. 371-406. Daily, C.M. and C. Schwenk (1996) Chief execuive officers, op managemen eams, and boards of direcors: Congruen or counervailing forces? Journal of Managemen, vol. 22, p. 185-208. Dong, M. and A. Ozkan (2008) Insiuional invesors and direcor pay: An empirical sudy of UK companies, Journal of Mulinaional Financial Managemen, vol. 18, p. 16-29. Fama, E.F. and M.C. Jensen (1983) Separaion of ownership and conrol, Journal of Law and Economics, vol. 26, p. 301-325. Fernandes, N. (2008) EC: Board compensaion and firm performance: The role of independen board members, Journal of Mulinaional Financial Managemen, vol. 18, p. 30-44. Finkelsein, S. and B. Boyd (1998) How much does he CEO maer? The role of managerial discreion in he seing of CEO compensaion, Academy of Managemen Journal, vol. 41, p. 179-199. Finkelsein, S. and D.C. Hambrick (1988) Chief execuive compensaion: A synhesis and reconciliaion, Sraegic Managemen Journal, vol. 9, p. 543-558. Finkelsein, S. and D.C. Hambrick (1996) Sraegic leadership: Top execuives and heir effecs on organizaion, New York: Wes Publishing. Firh, M., P.M.Y. Fung and O.M. Rui (2006) Corporae performance and CEO compensaion in China, Journal of Corporae Finance, vol. 12, p. 693-714. Frydman, C. and D. Jener (2010) CEO compensaion, SSRN Working Paper. Available a: hp://ssrn.com/absrac=1582232. Gregg, P., S. Jewell and I. Tonks (2005) Execuive pay and performance in he UK 1994-2002, Unpublished Working Paper, Universiy of Brisol and Universiy of Exeer. Gregg, P., S. Machin and S. Szymanski (1993) The disappearing relaionship beween direcors' pay and corporae performance, Briish Journal of Indusrial Relaions, vol. 31, p. 1-9. Gues, P. (2010) Board srucure and execuive pay: Evidence from he UK, Cambridge Journal of Economics, vol. 34, p. 1075-1097. Guhrie, K., J. Sokolowsky and K.M. Wan (2012) CEO compensaion and board srucure revisied, Journal of Finance, vol. 67, p. 1149-1168. Hempel, P. and C. Fay (1994) Ouside direcor compensaion and firm performance, Human Resource Managemen, vol. 33, p. 111-133. 25

D. Lin and L. Lin IJBFR Vol. 8 No. 2 2014 Hermalin, B.E. and M.S. Weisbach (1991) The effecs of board composiion and direc incenives on firm performance, Financial Managemen, vol. 20, p. 101-121. Hermalin, B.E. and M.S. Weisbach (1998) Endogenously chosen boards of direcors and heir monioring of he CEO, American Economic Review, vol. 88, p. 96-118. Hilburn, W. (2010) Trends in direcor compensaion, Bloomberg Businessweek, Ocober 19, 2010. Hill, C.W.L. and P. Phan (1991) CEO enure as a deerminan of CEO pay, Academy of Managemen Journal, vol. 34, p. 707-717. Jensen, M.C. (1993) The modern indusrial revoluion, exi, and he failure of inernal conrol sysems, Journal of Finance, vol. 48, p. 831-880. Jensen, M.C. and K.J. Murphy (1990) Performance pay and op-managemen incenives, Journal of Poliical Economy, vol. 98, p. 225-264. Main, B.G.M., A. Bruce and T. Buck (1996) Toal board remuneraion and company performance, Economic Journal, vol. 106, p. 1627-1644. Murphy, K.J. (2009) Compensaion srucure and sysemic risk, Working Paper FEB 34-09, USC Marshall School of Business. Ozkan, N. (2007) Do corporae governance mechanisms influence CEO compensaion? An empirical invesigaion of UK companies, Journal of Mulinaional Financial Managemen, vol. 17, p. 349-364. Ozkan, N. (2011) CEO compensaion and firm performance: An empirical invesigaion of UK panel daa, European Financial Managemen, vol. 17, p. 260-285. Ryan, H.J. and R. Wiggins (2001) The influence of firm- and manager-specific characerisics on he srucure of execuive compensaion, Journal of Corporae Finance, vol. 7, p. 101-123. Shivdasani, A. and D. Yermack (1999) CEO involvemen in he selecion of new board members: An empirical analysis, Journal of Finance, vol. 54, p. 1829-1853. Smih, C. and R. Was (1992) The invesmen opporuniy se and corporae financing, dividend, and compensaion policies, Journal of Financial Economics, vol. 32, p. 263-292. Song, Z. and Y. Xu (2007) The relaionship beween direcor compensaion and board independence: Evidence from China, Proceedings of 2007 Inernaional Conference on Managemen Science and Engineering. BIOGRAPHY Dan Lin is an assisan professor a he Takming Universiy of Science and Technology, and can be reached a mcylin@akming.edu.w. Lu Lin, he corresponding auhor of his paper, is an assisan professor a he Takming Universiy of Science and Technology, and can be reached a suzannelin@akming.edu.w. 26