Ownership Concentration, Managerial Ownership and Firm Performance: Evidence from Turkey

Transcription

1 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey Pınar Evrm Mandacı, Guluzar Kurt Gumus* Abstract: Ths study examnes the effects of ownershp concentraton and manageral ownershp on the proftablty and the value of non-fnancal frms lsted on the Istanbul Stock Exchange (ISE) n the context of an emergng market We measure the frm s performance by Return on Assets (ROA) and Tobn s Q ratos, where the former measures proftablty and the latter the value of the frm In addton, we gve detaled nformaton on the man characterstcs of the ownershp structures of the frms n our sample and fnd that ownershp of Turksh frms s hghly concentrated In addton, the unlsted holdng companes have the hghest average percentage of shares, whch supports the belef that ndvduals or famles establsh the holdng companes n order to control ther lsted frms After controllng for nvestment ntensty, leverage, growth and sze, we fnd that ownershp concentraton has a sgnfcantly postve effect on both frm value and proftablty, whle manageral ownershp has a sgnfcantly negatve effect on frm value Keywords: Corporate governance, ownershp concentraton, manageral ownershp, frm performance, Turkey JEL: G34; G15 DOI: /v Introducton Good corporate governance depends on the combnaton of the protecton of the rghts of nvestors and proper ownershp concentraton It has been observed that ownershp concentraton s hgh n less developed countres, where the rghts of nvestors are not protected due to the outrght lack of or nsuffcent regulaton provded by the relevant laws (see Shlefer and Vshny, 1997 and La Porta et al, 1999) The relatonshp between ownershp structure and frm performance provdes an dea about the effectveness of alternatve corporate governance mechansms Grossman and Hart (1986) argued that when the ownershp structure of a frm s overly dffused, shareholders are not motvated to montor management decsons closely, because the benefts that they can attan are mostly lower than the cost they would have to afford to control the managers Yet, ths settng may nfluence performance negatvely On the other sde, Shlefer and Vshny (1986) argued that when the ownershp structure s concentrated, shareholders wll control the actvtes of the managers easly, thereby avodng neffcency n management, and mprovng frm performance However, accordng to the agency theory, Jenson and Mecklng (1976) argued that hgh concentraton may smultaneously lead major shareholders to gve prorty to ther own nterests, and * Pınar Evrm Mandacı Dokuz Eylul Unversty, Faculty of Busness, Department of Accountng and Fnance e-mal: pnarevrm@deuedutr Guluzar Kurt Gumus Dokuz Eylul Unversty, Faculty of Busness, Department of Accountng and Fnance e-mal: guluzarkurt@deuedutr Aprl 2010 Unauthentcated 57

2 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey subsequently agency problems 1 may occur between the shareholders and managers In order to mnmze agency problems, shareholders have to endure agency costs 2 In addton, accordng to them, manageral ownershp prevents conflcts of nterest between the managers and owners and ncreases the value of the frm Sgnfcant manageral ownershp can algn managers nterests wth those of the outsde shareholders so that managers can have strong ncentve to pursue value-maxmzng behavor (algnment effects) In contrast, Demsetz (1983) argued that too large an ownershp stake by managers could potentally lead them to worry more about ther own nterests, not those of outsde shareholders, hence decreasng the frm s value (entrenchment effects) In ths paper we am to measure the effects of manageral ownershp and ownershp concentraton on the proftablty and value of Turksh non-fnancal frms lsted on the Istanbul Stock Exchange (ISE) In Turkey, the lsted corporatons are mostly owned by famles Standards of corporate governance and nvestor protecton are lower n Turkey than n the US and other major countres Hence, we hope that ths study wll add an nterestng dmenson to the relaton between these varables and performance n a developng country under a poor governance system Our paper s organzed as follows The second part conssts of a lterature revew In the thrd part, data and summary statstcs are presented Methodology and emprcal results are presented n the fourth part, and the fnal part offers concludng comments 2 Lterature Revew Although ntally Berle and Means (1932) suggested a postve correlaton between ownershp concentraton and performance, some of the followng studes dd not observe a relaton between these two varables (see Demsetz, 1983; Demsetz and Lehn, 1985 and etc) The study by Demsetz and Lehn (1985) that examned the relatonshp between accountng proft rate and percentage of shares owned by the fve and ten largest shareholders where ownershp structure s treated as an endogenous varable found no evdence of a relaton between these varables for US companes They argued that although greater ownershp concentraton results n 1 It s lkely that managers may place personal goals ahead of corporate goals (Gtman, 2006: 20) 2 The costs borne by stockholders to mantan a governance structure that mnmzes agency problems and contrbutes to the maxmzaton of owner wealth (Gtman, 2006: 20) stronger ncentves to montor, the expected gan from actve montorng and the cost of alternatve ownershp structures vary across frms Morck et al (1988) gnored the endogenous ssue and smlar to Demsetz and Lehn (1985) found no sgnfcant relaton n the lnear regressons they calculated by usng accountng profts and Tobn s Q as an alternatve measure of performance Shlefer and Vshny (1997) argued that makng the ownershp structure relatvely centralzed could promote the shareholder s controllng ablty and therefore the exstence of the bg shareholder was favorable to the growth of a company s value Hmmelberg et al (1999) argued that the emprcal fndngs mght be the result of unobservable frm heterogenety, whch mght affect both ownershp concentraton and performance These unobserved exogenous frm characterstcs mght nduce a spurous relatonshp between Tobn s Q and ownershp concentraton They found no relaton between these two varables after estmatng frm fxed effects Loderer and Martn (1997) and Demsetz and Vllalonga (2001) found no nfluence of ownershp concentraton on performance for US frms Ther fndng s consstent wth the vew that whle exacerbatng some agency problems, dffuse ownershp also yelds compensatng advantages that generally offset such problems On the other hand, Morck et al (1988) and Hrak et al (2003) for Japanese frms, and Gorton and Schmd (2000) for German companes, found a postve relaton between ownershp concentraton and frm value In addton to studes on developed countres, most of the studes for developng countres found a postve relaton between ownershp concentraton and performance Claesses and Djankov (1999) examned Czech companes and argued that the more concentrated the ownershp, the hgher the frm s proftablty; ths fndng sgnfed the same postve relaton as ndcated n studes on Czech frms by Claessens (1997) and Wess and Nktn (1998) However, ther fndngs were ambguous when the type of ownershp was taken as the control varable Smlar to ther prevous study n 1997, Xu and Wang (1999) examned Chnese lsted frms and found a postve correlaton between the shareholdng rato of the frst fve and ten bg shareholders and performance Sun et al (2002) proved that relatvely bgger holdng companes and other bg shareholders whch had a certan concentraton degree could help mprove the performance of Chnese frms Barbers et al (1996) also found a postve relaton for Russan frms Joh (2003), who examned Korean frms, found that after controllng 58 Unauthentcated SEE Journal

3 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey frm and ndustry characterstcs, frms wth low ownershp concentraton recorded low proftablty Yammeesr et al (2006) examned Ta non-fnancal frms, and as the lterature proved prevously, found a postve assocaton between concentrated ownershp and performance Smlarly, Omran et al (2008), n ther studes on a group of Arab countres (Egypt, Jordan, Oman, Tunsa), found that ownershp concentraton was postvely correlated wth varous performance measures, and that large-sze frms were more lkely to acheve better performance However, parallel to the fndngs of Demsetz and Lehn (1985), and Hmmelberg et al (1999), Chen et al (2005) found for Hong Kong frms that concentrated ownershp was not assocated wth better operatng performance or hgher frm valuaton Comparably, Gunasekarage et al (2007) proved that ownershp concentraton s negatvely related to frm performance n Chna when market-to-book rato s used as a performance ndcator Gursoy and Aydogan (2002) examned the mpact of ownershp concentraton and ownershp mx on frm performance of Turksh nonfnancal frms between 1992 and 1998 and found that hgher concentraton led to better market performance but lower accountng performance They used prce-toearnngs rato and stock returns to measure market performance There are many studes wth contendng results on the relatonshp between manageral ownershp and performance Among them, Jensen and Mecklng (1976) stated that when manageral ownershp ncreased, conflct would decrease and performance would ncrease In contrast, Fama and Jensen (1983) and Stulz (1988) argued that greater stock ownershp by managers ncreased the power of the nternal owners and decreased the power of the external owners n terms of nfluencng performance Morck et al (1988) found a sgnfcant non-monotonc relatonshp between Tobn s Q and board member ownershp Accordngly, the relaton ncreases between 0% and 5%, decreases between 5% and 25%, and ncreases beyond 25% Ther outcomes were also sgnfcant when some control varables such as R&D and advertsng ratos, leverage, sze, growth and ndustry dummes were ncluded n models However, they were not sgnfcant when accountng proft rates were used as an alternatve performance measure Smlar to the study of Morck et al (1988), McConnel and Servaes (1990), Hermaln and Wesbach (1991), Cho (1998) and most recently Floracks et al (2009) found a postve relaton for low levels of ownershp and a negatve relaton for hgh levels of ownershp Yet, unlke Morck et al (1988), McConnel and Servaes (1990) found that ths relatonshp was also sgnfcant when they used accountng profts nstead of Tobn s Q In contrast to other studes, Floracks et al (2009) found a negatve relatonshp by usng a semparametrc estmaton approach As n Morck et al (1988), Holderness et al (1999) found a sgnfcant postve relaton between frm performance and manageral ownershp wthn the 0% to 5% range of manageral shareholdngs; but n contrast to Morck et al (1988), they dd not fnd a statstcally sgnfcant relaton beyond 5% Demsetz and Lehn (1985), Loderer and Martn (1997), and Hmmelberg et al (1999) as well as Demsetz and Vllalonga (2001) dd not fnd a sgnfcant relaton between manageral ownershp and performance Most of those studes examned the assocaton between nsder ownershp and performance account for the endogenety of ownershp structure except Morck et al (1988) and McConnell and Servaes (1990) Demsetz and Vllalonga (2001) examned US frms and ther evdence supported the belef that ownershp structure was endogenous but beled the belef that ownershp structure affected frm performance They argued that f there were compensatng advantages n a frm, there would be no systematc relaton between manageral shareholdngs and frm performance In addton, they argued that t mght ndcate that ths relatonshp depends on locaton, specal local laws and governance practces There s no study whch examnes the relatonshp between manageral ownershp and the performance of Turksh frms As can be seen from the above, the emprcal results on the effects of manageral ownershp and ownershp concentraton on frm performance are conflctng In addton, the prevous studes focused mostly on large ndustralzed countres, whch completed ther nsttutonalzaton process; therefore, ther outcomes mght not be relevant for developng countres In ths study, we try to fll ths gap by examnng ths ssue for a developng country, namely Turkey 3 Data and Summary Statstcs Our sample ncludes all non-fnancal frms lsted on the ISE n the year 2005 The number of frms n our sample s 203 We excluded banks and leasng, nvestment, nsurance and holdng companes snce ther fnancal tables are dfferent from non-fnancal frms We Aprl 2010 Unauthentcated 59

4 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey collected data on market values of the sample frms from the Monthly (December) Bulletns of the ISE We used annual company reports ssued by the ISE to obtan data on ownershp structure (ownershp concentraton, manageral ownershp, etc) We obtaned the remanng data by usng the balance sheets and ncome statements from the ISE s webste In lne wth prevous studes (such as Morck et al,1988; McConnel and Servaes, 1990 and etc), we also focused on one year of data by takng nto consderaton the fact that the ownershp structure of the frm does not vary frequently In addton, the Internatonal Fnancal Reportng Standards have been appled n Turkey snce the begnnng of 2005 The frms are requred to ncorporate the new standards and to prepare more detaled annual reports whch provde more data to fnance researchers and consequently facltate the analyss of frms ownershp structures Hence, ths study takes 2005 to be the startng lne Table 1 depcts the number of the dfferent types of shareholders of the present study s sample frms operatng n dfferent ndustres Accordng to ths table, ndvduals and famles, unlsted holdng companes and unlsted non-fnancal frms pre-domnate the ownershp postons n the sample frms Furthermore, ths predomnance s mostly seen n the textle ndustry Addtonally, most of the shareholders are foregn companes n the food ndustry, lsted holdng companes n the metal products ndustry and lsted non-fnancal frms n the stone and sol ndustry The ownershp postons of the state, foundatons and labor unons are very low compared to other partes Table 2 denotes the average percentage of shares held by the owners of the sample frms n dfferent ndustres The hghest average percentage of shares s held by unlsted holdng companes, whch supports the belef that ndvduals or famles establsh the holdng companes n order to control ther lsted frms It s followed by the unlsted non-fnancal frms and ndvduals and famles respectvely Table 3 depcts the proporton of shares held by the Board members and managers as well as ther relatves In Turkey t s often observed that the famly members are the CEOs, Boards of Drectors or top managers of the frms As a result, management control s n the hands of these famly members (see Yurtoğlu, 1998 and Demrağ and Serter, 2003) Table 3 denotes that CEOs, Boards of drectors and top managers have almost 838% of shares outstandng And ther relatves have almost 3% of the outstandng shares Board members and general Industres Indvdual Holdng Comp (lsted) Holdng Comp (unlsted) Non-fnancal (lsted) Non-fnancal (unlsted) Fnancal (lsted) Fnancal (unlsted) Foregn State Foundatons/ Retrement Funds Labor Unons Food Textle Wood and Paper Chemcal Stone and Sol Metal Man Metal Products Other Manufacturng Frms Technology Educaton, Sport and Health Telecommuncaton Wholesale and Retal Electrcty and Constructon Total Table 1: The Number and Types of Shareholders of Non-fnancal Frms n Dfferent Industres (Year 2005) 60 Unauthentcated SEE Journal

5 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey Industry Number of Frms Ind/Famly Lsted Holdng Non-fnancal frm Fnancal Frm Unlsted Total Lsted Unlsted Total Lsted Unlsted Total Foregn State Foundatons retrement funds Labor unons/ Cooperatves Food Textle Wood and Paper Chemcal Stone and Sol Metal Man Metal Products Other Manufacturng Frms Technology Educaton, Sport and Health Telecommuncaton Wholesale and Retal Electrcty and Constructon Average Table 2: Ownershp Structure of Non- fnancal Frms n Dfferent Industres (n %, Year 2005) managers have a consderable share n the technology ndustry (2405 percent) The average of openness-to-publc s 33 percent wthn sample frms Table 4 gves the percentage of shares owned by the largest three, fve and ten shareholders for the sample frms n dfferent ndustres It shows that the largest shareholder has around 48 percent of shares, whch ndcates that one person or an nsttuton has almost half of a lsted company The averages for the largest three, fve and ten are 6127; 6420 and 6593 respectvely Moreover, the hghest ownershp concentraton s n educaton, sport and health ndustry for the largest shareholder and n the telecommuncaton ndustry for the largest three, fve and ten shareholders Industry Share of Board Members and General Manager Share of Famly Members Open-to- Publc (%) Food Textle Wood and Paper Chemcal Stone and Sol Metal Man Metal Products Other Manufacturng Frms Technology Educaton, Sport and Health Telecommuncaton Wholesale and Retal Electrcty and Constructon Average Table 3: Manageral Ownershp n Non-fnancal frms (n %, Year 2005) Aprl 2010 Unauthentcated 61

6 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey Ownershp Concentraton Industry Top 10 Top 1 Shareholder Top 3 Shareholders Top 5 Shareholders Shareholders Food Textle Wood and Paper Chemcal Stone and Sol Metal Man Metal Products Other Manufacturng Frms Technology Educaton, Sport and Health Telecommuncaton Wholesale and Retal Electrcty and Constructon Average Table 4: The Ownershp Concentraton Rates for Non-fnancal Frms (n %, Year 2005) We used two measures of performance as dependent varables, Return on Assets (ROA) and Tobn s Q ratos, where the former measures proftablty and the latter the value of the frm We used two ndependent varables, the percentage of shares held by the largest three shareholders and the percentage of shares held by the managers We observed that the studes on Turksh frms (such as Gursoy and Aydogan, 2002 and Demrağ and Serter, 2003) prefer usng the share of the largest three shareholders to measure ownershp concentraton Dependng on these prevous studes as well as our outcomes about the concentraton rates presented n Table 4, we used the percentage of shares held by the largest three shareholders to measure ownershp concentraton Our control varables are the nvestment ntensty, leverage, growth and sze whch are assumed to have an effect on frm performance In addton to these we employ ndustry dummy varables n order to pont out whether the performance measures dffer across ndustres The study s varables and ther defntons are presented n Table 5 Dependent Varables Defnton TOBIN S Q Market value of assets (total debt plus market value of equty)/ total assets ROA Net Income/Total Assets Independent Varables CON3 Total share of the largest three shareholders n the frm OWNER Manageral Ownershp: Percentage of Shares Owned by the managers Control Varables CAPEXP Captal Expendtures/Sales (nvestment ntensty) DEBTTA Total Debt/Total Equty (Leverage) GROWTH Average growth n net sales over three-year perod ( ) SIZE Logarthm of total assets INDUSTRY DUMMIES Educaton-health; electrcty; food-beverage; chemcal-petroleum-plastc; metal productsmachnery; wood-paper-prntng; non-metal mneral products; technology; textleleather; wholesale and retal trade; transportaton Table 5: Varables and Defntons 62 Unauthentcated SEE Journal

7 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey ROA TOBIN S Q CAPEXP DEBTTA GROWTH Mean Maxmum Mnmum Std Dev Table 6: Descrptve Statstcs ROA CON3 OWNER CAPEXP DEBTTA GROWTH SIZE ROA 1 CON OWNER CAPEXP DEBTTA GROWTH SIZE ***, **, * ndcate 1 %, 5 % and 10 % sgnfcance Table 7A Correlaton Matrx (between ROA and ndependent varables) TOBINSQ CON3 OWNER CAPEXP DEBTTA GROWTH SIZE TOBINSQ 1 CON OWNER CAPEXP DEBTTA GROWTH SIZE ***, **, * ndcate 1 %, 5 % and 10 % sgnfcance Table 7B Correlaton Matrx (between Tobns Q and ndependent varables) The descrptve statstcs for varables are presented n Table 6 For the year 2005, the average ROA was 005 whle the hghest ROA was 059 Mean of Debt-to-Total Assets ndcates that Turksh companes prefer fnancng ther companes wth captal nstead of debt The hghest standard devaton belongs to captal expendtures and Tobn s Q, showng the hghest varance among companes Tables 7 A and B show the correlaton matrx for each dependent varable The correlaton coeffcents among ndependent varables are low Addtonally, correlaton coeffcents between ndependent and dependent varables are consstent wth the drecton of the relaton and coeffcents found n the regressons analyss 4 Methodology and Emprcal Results We appled multple regresson analyss to measure the effects of ownershp concentraton and manageral ownershp on frm performance We developed two dfferent groups of hypotheses on the relatonshp of ownershp concentraton and manageral ownershp wth frm performance In parallel to most of the prevous studes on developng countres, we expected a postve relatonshp between ownershp concentraton and frm performance Snce the ownershp concentraton s hgh n Turkey, the shareholder can easly control the managers and force them to focus on the maxmzaton of the shareholders wealth Agency theory suggests that when the manageral ownershp ncreases, the conflct of nterest between the managers and owners wll decrease and frm performance wll ncrease Although most of the prevous studes dd not observe a relaton between manageral ownershp and frm performance, n lne wth the theory, we expected a postve relatonshp between them The hypotheses of the study are stated below; H1 0 The ownershp concentraton s not sgnfcantly related to frm performance H1 1 The ownershp concentraton s postvely correlated to frm performance Aprl 2010 Unauthentcated 63

8 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey H2 0 The level of manageral ownershp s not sgnfcantly related to frm performance H2 1 The level of manageral ownershp s postvely correlated to frm performance Dfferent models were constructed to explore the effects of the ndependent varables In models where the ROA and TOBINSQ are the dependent varables, frst all varables other than ndustry dummes were ncluded n the analyss, and then the ndustry dummes were ncluded to nvestgate the ndustry effects on frm value and proftablty The four models of the study are presented below of educaton, food and beverage and wholesale ndustres on frm performance After we ncluded the ndustry dummes, the adjusted R 2 ncreased to 034 from 018 The postve relatonshp between ownershp concentraton and performance supports the study of Shlefer and Vshny (1997) who state that snce nvestor protecton s low n developng countres, ownershp concentraton s accepted as an alternatve corporate governance tool n these countres Our fndngs are also parallel to prevous studes on developng countres such as Barbers et al (1996), Claesses and Djankov (1999), Joh (2002), Yammeesr et al (2006) However, our result on Model 1: ROA Model 2: ROA = 0 1CON3 2OWNER 4CAPEXP 5DEBTTA 6GROWTH 7 β SIZE + ε = 0 1CON3 2OWNER 4CAPEXP 5DEBTTA 6GROWTH 7 β SIZE + D + ε Model 3: TOBINSQ Model 4: TOBINSQ İ = 0 1CON3 2OWNER 4CAPEXP 5DEBTTA 6GROWTH 7 β SIZE + ε = 0 1CON3 2OWNER 4CAPEXP 5DEBTTA 6GROWTH 7 β SIZE + D + ε Table 7 presents the regresson results for ROA and Tobn s Q In Models 1 and 2, we used ROA as a measure of proftablty and found a sgnfcant postve effect among the largest three shareholders on frm proftablty n the models n whch the dummes were excluded Debt-to-total assets and growth varables also had sgnfcant effects n all models Addtonally, the dummy for the educaton ndustry has a sgnfcant effect, but manageral ownershp and captal expendtures dd not have an effect on proftablty When we added the dummy varables n the model, the adjusted R 2 ncreased to 0276 from 0182 In Models 3 and 4, we used Tobn s Q as an ndcator of frm value and found that the largest three shareholders had a sgnfcant postve effect, and that manageral ownershp had a sgnfcant negatve effect on frm value When we ncluded the dummy varables, we dd not fnd a sgnfcant relatonshp between manageral ownershp and frm value In addton, Debt-to-total assets and sze varables were the other factors that affected frm value negatvely Captal expendture and growth dd not have a sgnfcant effect on frm value The model wth dummy varables of ndustres ponted out the sgnfcant effects the relatonshp between the ownershp concentraton and proftablty s the opposte of that of Gursoy and Aydogan (2002) whch found a negatve relatonshp between ownershp concentraton and proftablty The fndngs on the effects of manageral ownershp are much more conflcted n the lterature Our fndng does not support the agency theory The negatve relaton between manageral ownershp and frm value mght be explaned by Demsetz (1983), who argued that too much manageral ownershp could potentally lead managers to worry more about ther own nterests, and not those of outsde shareholders, hence decreasng frm value 5 Concludng Comments The emprcal results on the effects of manageral ownershp and ownershp concentraton on frm performance are conflctng Prevous studes focused mostly on large ndustralzed countres, whch completed ther nsttutonalzaton process and therefore, ther outcomes mght not be relevant for developng countres In ths study, we try to fll ths gap by examnng the effects of ownershp concentraton and manageral 64 Unauthentcated SEE Journal

9 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey CON3 OWNER CAPEXP DEBTTA GROWTH SIZE Regressor Educaton-Health Electrcty Food-Beverage Chemcal-Petroleum-Plastcs Metal Products-Machnery Wood-Paper-Prntng Non-Metal Products Technology Textle-Leather Wholesale And Retal Trade Transportaton C ROA TOBINSQ Model 1 Model 2 Model 3 Model ** (1546) 1656* (2056)**) (-0464) (-2091)** (1053) 1126 (1019) *** (-1986)** -4074*** (-2535)** *** (2471)** 1991** (1613) ** (1930)* -2288** (-2833)*** (2304)** (4977)*** (-1528) (0544) (-0089) (2099)** (0527) (1005) (0106) (0205) (-0824) (0351) (0973) (1609) (1286) (1158) (-0937) (0463) (0257) (1731)* (1411) (0799) ** (-1770)* 3522*** (3742)*** Adjusted R ***, **, * ndcate 1 %, 5 % and 10 % sgnfcance Table 7: Regresson Results ownershp on the performance of Turksh frms lsted on the ISE n the context of a developng country We measured frm performance by ROA (Return on Assets) and Tobn s Q, where the former measures proftablty and the latter frm value We tred to test the relatonshps between these performance measures and the percentage of shares owned by the largest three shareholders, and the percentage of shares held by board members and general managers In addton to these ndependent varables, we used control varables ncludng nvestment ntensty, leverage, growth and sze and ndustry dummes, whch are assumed to have an effect on frm performance In addton, we analyzed the ownershp structure of the sample frms and found that the hghest average percentage of shares was held by the unlsted holdng companes, unlsted non-fnancal frms and ndvduals and famles, respectvely, whch confrms the wdespread belef that n Turkey ndvduals or famles set up ther unlsted frms n order to control ther lsted companes Supportng the studes of Gursoy and Aydogan (2002) and Gönenç (2004), we found that the ownershp of Turksh frms s hghly concentrated Our regresson results show that ownershp concentraton has a sgnfcant postve effect on both frm value and proftablty Ths result may support the dea of Schlefer and Vshny (1997) who state that snce nvestor protecton s low n developng countres, ownershp concentraton s accepted as an alternatve corporate governance mechansm n these countres On the other hand, n contrast to the agency theory, we found a negatve relaton between manageral ownershp and frm value, whch mght support the argument of Demsetz, who states that too much manageral ownershp could potentally lead managers to worry more about ther own nterests and decrease frm value The fndng does not support the dea of Jensen and Mecklng (1976), who argue that as manageral ownershp ncreases, the conflct between the managers and owners wll decrease and performance wll ncrease The results of the study suggest that Turksh frms can ncrease ther performance by ncreasng ther ownershp concentraton and by decreasng manageral ownershp Aprl 2010 Unauthentcated 65

10 Ownershp Concentraton, Manageral Ownershp and Frm Performance: Evdence from Turkey We beleve, despte the unqueness of the results, that ths study proves that the subject deserves greater attenton and more detaled analyss We dd not add other agency varables such as the number of ndependent drectors, board composton, etc, snce the full data s not avalable n annual reports Future studes mght collect ths data by drectly contactng the frms In addton, future studes mght nclude fnancal frms and holdng companes References Barbers, N, M Boycko, A Shlefer, N Tsukanova (1996) How does prvatzaton work? Evdence from the Russan shops Journal of Poltcal Economy, 104, Berle, A, G Means (1932) The Modern Corporaton and Prvate Property Harcourt, Brace, &World, New York Chen, Z, Y-L Cheung, A Strourats, AWS Wang (2005) Ownershp concentraton, frm performance, and dvdend polcy n Hong Kong Pacfc-Basn Fnance Journal, 13, Cho, M-H (1998) Ownershp structure, nvestment, and the corporate value: An emprcal Analyss Journal of Fnancal Economcs, 47, Claessens, S (1997) Corporate governance and equty prces: Evdence from the Czech and Slovak Republcs Journal of Fnance, 52, Claesses, S, S Djankov (1999) Ownershp concentraton and corporate performance n the Czech Republc Journal of Comparatve Economcs, 27, Demrağ, I, M Serter (2003) Ownershp Patterns and Control n Turksh Lsted Companes Corporate Governance: An Internatonal Revew, 11, Demsetz, H (1983) The structure of ownershp and the theory of the frm Journal of Law and Economcs, 26, Demsetz, H, K Lehn (1985) The structure of corporate ownershp: Causes and consequences Journal of Poltcal Economy, 93, Demsetz, H, B Vllalonga (2001) Ownershp structure and corporate performance Workng paper UCLA Fama E, M Jensen (1983) Separaton of ownershp and control Journal of Law and Economcs, 26, Floracks, C, A Kostaks, A Ozkan (2009) Manageral ownershp and performance Journal of Busness Research, 62, Gtman, L (2006) Prncples of Manageral Fnance Addson Wesley Longman Gorton, G, FA Schmd (2000) Unversal bankng and the performance of German frms Journal of Fnancal Economcs, 58, Gönenç, H (2004) VIII Natonal Fnance Symposum, October, Istanbul Techncal Unversty Istanbul, Grossman, S, O Hart (1986) The costs and benefts of ownershp: A theory of vertcal and lateral ntegraton Journal of Poltcal Economy, 94, Gunasekarage, A, K Hess, A Hu (2007) The nfluence of the degree of state ownershp and the ownershp concentraton on the performance of lsted Chnese companes Research n Internatonal Busness and Fnance, 21, Gursoy, G, K Aydogan (2002) Equty ownershp structure, rsk takng and performance Emergng Markets Fnance and Trade, 38, 6-25 Hermaln, B E, M S Webach (1991) The effect of board composton and drect ncentves on frm performance Fnancal Management, 21, Hmmelberg, C, RG Hubbard, D Pala (1999) Understandng the determnants of manageral ownershp and the lnk between ownershp and performance Journal of Fnancal Economcs, 53, Hrak, T, H Inoue, A Ito, F Kurok, H Masuda (2003) Corporate governance and frm value n Japan: Evdence from 1985 to 1998 Pacfc-Basn Fnance Journal, 11, Holderness, C, R Kroszner, D Sheehan (1999) Were the good old days that good? Evoluton of manageral stock ownershp and corporate governance snce the Great Depresson Journal of Fnance, 54, Jensen, M, W Mecklng (1976) Theory of the frm: manageral behavor, agency costs, and ownershp structure Journal of Fnancal Economcs, 3, Joh, SW (2003) Corporate governance and frm proftablty: Evdence from Korea before the, economc crss Journal of Fnancal Economcs, 68, La Porta, R, FLopez-de-slanes, A Shlefer, R Vshny (1999) Corporate ownershp around the world Journal of Fnance, 54, Loderer, C, and K Martn (1997) Executve stock ownershp and performance: trackng fant traces Journal of Fnancal Economcs, 45, McConnell, J, H Servaes (1990) Addtonal evdence on equty ownershp and corporate value Journal of Fnancal Economcs, 27, Morck, R, A Shlefer, R Vshny (1988) Management ownershp and market valuaton: an emprcal analyss Journal of Fnancal Economcs, 20, Omran, M M, A Bolbol, A Fatheldn (2008) Corporate governance and frm performance n Arab equty markets: does ownershp concentraton matter? Internatonal Revew of Law and Economcs, 28, Shlefer, A, R Vshny (1986) Large shareholders and corporate control Journal of Poltcal Economy, 94, Shlefer, A, R Vshny (1997) A survey of corporate governance Journal of Fnance, 52, Stulz, RM (1988) Manageral control of votng rghts: Fnancng polces and the market for corporate control Journal of Fnancal Economcs, 20, Sun, Q, W Tong, J Tong (2002) How does government ownershp affect frm performance? Evdence from Chna s prvatzaton experence Journal of Busness Fnance and Accountng, 29, 1 27 Wess, A, and G Nktn (1998) Performancee of Czech companes by ownershp structure Unversty of Mchgan: Wllam Davdson Insttute, The Davdson Insttute Workng Paper 186 Xu, X,Y Wang (1997) Ownershp structure, corporate governance, and corporate performance: The case of Chnese stock companes World Bank Workng Paper 1794, World Bank Xu, X, Y Wang (1999) Ownershp structure and corporate governance n Chnese stock companes Chna Economc Revew, 10, Yammeesr, J, S J Lodh, S K Herath (2006) Influence of ownershp structure and corporate performance pre-crss: evdence from Thaland Internatonal Journal of Electronc Fnance, 1, Yurtoglu, BB (1998) Ownershp Structure of Turksh Lsted Frms ISE Fnance Award Seres 1: Unauthentcated SEE Journal