The dynamic relationship between managerial ownership and firms performance in vietnams

Findings from this study also present evidence to confirm the view that the reduction of lagged managerial ownership level would send a signal in relation to the quality of firm, and wou

UNIVERSITY OF ECONOMICS INSTITUDE OF SOCIAL STUDIES

VIETNAM- NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

THE DYNAMIC RELATIONSHIP BETWEEN

MANAGERIAL OWNERSHIP AND FIRM’S PERFORMANCE IN VIETNAM

BY NGUYEN THI THANH AN

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, NOVEMBER 2016

UNIVERSITY OF ECONOMICS INSTITUDE OD SOCIAL STUDIES

VIETNAM- NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

THE DYNAMIC RELATIONSHIP BETWEEN

MANAGERIAL OWNERSHIP AND FIRM’S PERFORMANCE IN VIETNAM

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

Nguyen Thi Thanh An

ii

ACKNOWLEDGEMENTS

First of all, the thesis had implemented with the assistance and encouragement of my principal supervisor Dr Vo Hong Duc I would like to emphasize the thankful gratitude and appreciation to him His guidance helps me develop my dissertation throughout all stages of processing thesis He also devotes his valuable time for me and my team to encourage and share his experience in study as well as in real life

In addition, I would like thank Dr Truong Dang Thuy who is patient and sympathetic in provision of econometric technique and supporting in any question My thanks also go to all

of staffs at VNP, especially Ms Xuan Hong has already assisted the students with her dedication

Furthermore, I also acknowledged all of classmates in Class 21 I had the opportunity to work and learn in an intimate and supportive class

Above all, I most deeply thank to my parents who always support in any project and intention and love me without condition over 26 years

HCMC, November, 2016

Nguyen Thi Thanh An

iii

ABSTRACT

Corporate governance is generally considered as a key factor for the operational success for enterprises One of the most frequently discussed concepts among corporate governance factors is the ownership structure, in particular, managerial ownership Various empirical studies have been conducted to consider and examine the impact of diverse ownership structure aspects on firm’s performance which is proxied by various dimensions including accounting return, market evaluation and probability of bankruptcy

Findings from various empirical studies on the above issue are mixed due to the

following reasons First, it is difficult to capture appropriately the power of managers in

making decisions In the developing countries, managers also control the firm via the ownership of related parties As such, this study is conducted to measure managerial ownership in terms of (i) direct ownership and (ii) indirect ownership to achieve the better

measurement Second, many authors argued that managerial ownership should be treated as

endogenous parameter and the relationship between managerial ownership and firm’s performance is non-monotonic To deal with endogeneity of managerial ownership, this study

focuses on the effect of the change in managerial ownership (managers’ decision to

purchasing and selling stocks) on the change in firm’s performance

To achieve these objectives, a panel data of 285 listed firms on Ho Chi Minh City Stock Exchange (HOSE) for the period from 2010 to 2015 is utilized Findings from this study indicate that the percentage of stocks owned by managers and their related parties significantly fluctuated In addition, the actual managerial ownership level tends to move away from the optimal level due to the existence of adjustment costs Furthermore, managers are likely to sell stock when the entire financial market performed well Notwithstanding, the managers do not purchase stoFcks in the case of illiquid market and deteriorating performance Findings from this study also present evidence to confirm the view that the reduction of lagged managerial ownership level would send a signal in relation to the quality

of firm, and would also provide the negative impact on firm’s performance regarding market evaluation However, the study fails to provide empirical evidence in relation to the change in managerial ownership against the change in firm’s return on total assets the accounting- based measurement

Key words: Managerial ownership, Firm’s performance, Endogeneity, Market-based

measurement, Accounting-based measurement

iv

TABLE OF CONTENTS

DECLARATION i

ACKNOWLEDGEMENTS ii

ABSTRACT iii

TABLE OF CONTENTS iv

LIST OF TABLES vii

LIST OF FIGURES viii

LIST OF ABBREVIATION ix

CHAPTER 1 INTRODUCTION 1

1.1 Problem statement 1

1.2 Research objectives 3

1.3 Research questions 3

1.4 Contributions of the thesis 4

1.5 Research Scope 4

1.6 Structure of the thesis 4

CHAPTER 2 LITERATURE REVIEW 6

2.1 The theoretical background of managerial ownership and firm’s performance 6

2.1.1 The agency approach 6

2.1.1.1The incentive effect 7

2.1.1.2The entrenchment effect 7

2.1.2 The managerial discretion approach 8

2.1.3 The timing approach 10

2.2 Endogeneity of managerial ownership 10

v

2.3 The empirical evidences of relationship between managerial ownership and firm’s

performance and limitations 12

2.3.1The research in worldwide and the limitations 12

2.3.1.1The exogenous managerial ownership 12

2.3.1.2The endogenous managerial ownership 14

2.3.2The empirical evidence in Vietnam 17

2.4 The corporate governance of Vietnamese listed firms 18

2.5 The conceptual framework 21

CHAPTER 3 RESEARCH METHODOLOGY AND DATA 22

3.1 Data sources 22

3.2 Measurement variables 23

3.2.1Definition and measurements of firm’s performance 23

3.2.1.1Accounting – based measurements 23

3.2.1.2Market–based measurements 24

3.2.2 Definition and measurement of managerial ownership 24

3.3 Research methodology 25

3.4 The empirical model 26

3.4.1 The determinants of firm’s performance and optimal managerial ownership 26

3.4.1.1The determinants of firm’s performance 26

3.4.1.2The determinants of optimal managerial ownership level 28

3.4.1.3 The movement of actual managerial ownership 31

3.4.2 The explanation of the large change in managerial ownership 32

3.4.3The dynamic relationship between managerial ownership and firm’s performance 34

CHAPTER 4 RESULTS AND DISCUSSIONS 36

4.1 Data description 36

vi

4.1.1 Descriptive statistics 36

4.1.2 Correlation analysis 41

4.2 The determinants and movement of managerial ownership 41

4.2.1 The determinants of managerial ownership 44

4.2.2 The movement of actual managerial ownership. 46

4.3 The explanation of the large change (decrease or increase) 47

4.3.1 The statistics by group 47

4.3.2 The likelihood regression of large change (increase or decrease) against the change in firms’ characteristics and market condition. 50

4.4 Dynamics of managerial ownership and firm’s performance 53

4.4.1Firm’s performance: accounting-based measurement 53

4.4.2Firm’s performance: market-based measurement. 55

CHAPTER 5 CONCLUSIONS AND POLICY IMPLICATIONS 60

5.1 Concluding remarks 60

5.2 Policy implications 61

5.2.1 The implications for enterprises 61

5.2.2 The implications for Vietnam’s authority and the Government 62

5.3 The limitations and further research 62

5.3.1 The limitations 62

5.3.2 The further research 62

REFERENCES 64

APPENDICES 70

vii

LIST OF TABLES

Table 3.1 Tests are ultilized to find the appropriate model 26 Table 3.2 The determinants of optimal managerial ownership level 28 Table 3.3 The summary of variables employed in Probit model 32 Table 4.1 Summary statistics of firm’s characteristics of 285 firms listed on HOSE from

2010 to 2015) period 37 Table 4.2 Statistical summary of variables separated by year 38 Table 4.3 Statistical summary of variables separated by industry 39 Table 4.4 Correlation coefficients between managerial ownership and firm’s attributes 40 Table 4.5 The relationship between level of managerial ownership and firm’s

performance 43 Table 4.6 The determinants of managerial ownership level 45 Table 4.7 The movement actual mangerial ownership level toward to estimated optimal

level 46 Table 4.8 Statistical summary of data by data source 49 Table 4.9 Large change in managerial ownership against change in firm’s attributes and

market condition 52 Table 4.10 The effect of lagged managerial ownership change on firm’s performance

change in terms of accounting-based measurement 53 Table 4.11 The effect of lagged change in managerial ownership on firm’s performance in

terms of market-based measurement 57

viii

LIST OF FIGURES

Figure 1.1 The relationship between insider ownership and firm’s performance 8

Figure 1.2 The management structure of shareholding company 19

Figure 1.3 The internal governance structure of a listed company 20

Figure 1.4 The conceptual framework 21

Figure 4.1 The nonlinear relation between the lagged change in managerial ownership and the change in return on total assets (ROA) 53

Figure 4.2 The nonlinear relation between the lagged change in managerial ownership and the change in market evaluation of firm’s performance (Tobin’s Q) 55

ix

LIST OF ABBREVIATION

AMEX American Stock Exchange

CEO Chief Executive Officer

GMM Generalized method of moments

GMS General Meeting of Shareholders

FGLS Flexible generalized least squares

HOSE Ho Chi Minh Stock Exchange

IFC International Finance Corporation

NASDAQ National Association of Securities Dealers Automated Quotations

NYSE New York Stock Exchange

OLS Ordinary least square

R&D Research & Development

ROE Return on equity

P/E Price/ Earnings ratio

2SLS Two Stage Least Square

SSC State Securities Commission

UPCOM Unlisted Public Company Market

1

Chapter 1 INTRODUCTION

1.1 Problem statement

In corporate world, the separation between principals (owners) and the agents (managers), or the ownership and control of the firm, induces the agency problem Managerial ownership could generally be used as an effective mechanism to reduce this problem As such, the optimal level of managerial ownership has attracted great attention from both academics and practitioners in a modern corporate governance framework The ultimate objective of any firm is to maximize the firm’s value or to maximize owner’s propensity However, in some cases, under managerial discretion, they will maximize their welfare (Kuhnen & Zwiebel, 2006; Lambrecht & Myer, 2007) Three considerable approaches which have been used to attempt to explain the mechanism of impact of managerial ownership on firm’s performance: (i) agency approach, (ii) managerial discretion approach, and (iii) timing approach Most of the empirical studies presented a non-linear relationship between managerial ownership and firm’s performance McConnell and Servaes (1990) found the reversed U-shaped relationship between Tobin’s Q and the level of insider ownership MO also provides a positive effect on firm’s performance up to the threshold of 40 percent to 50 percent Morck, Shleifer, and Vishny (1988) examined this relationship with sectional data of

500 Fortune firms and concluded that the W-shaped between the level of MO and Tobin’s Q was found However, Kole (1995) and Himmelberg, Hubbard, and Palia (1999) argued that cross-sectional data cannot capture adequately the change in firm’s environment and endogeneity of MO Many recent studies attempted to deal with endogeneity issue of MO with different approaches, however, some limitations still exists and many arguments in interpretation the result provoked

Kole and Lehn (1997) addressed an important issue of governance structure that has change over time They also investigated the question as to what forces drive the change In reality, Holderness, Kroszner and Sheehan (1998) observed and detected the managerial ownership in the US listed firms changed over time Under similar circumstance, a new

approach in relation to the relationship between the change in managerial ownership and firm’s performance via the change in Tobin’s Q in American firms investigated by

Fahlenbrach and Stulz (2008) The study attempted to examine the effect of the dynamic

In Vietnam, several studies investigated the impact of the structure of ownership including the existence of state ownership, and foreign ownership on firm’s performance (Le

& Phung, 2013) Their findings were an inversed U-shaped relationship between state ownership level and Tobin’s Q, and foreign ownership can boost firm’s performance In addition to that, the empirical results confirmed that there was no statistically significant evidence of managerial ownership on firm’s performance in accounting- based measurement (ROA, ROE) (Do & Wu, 2014; Nguyen & Giang, 2015) Nevertheless, the negative effect of managerial ownership on firm’s performance in market aspects (P/E ratio) could come from expropriation of block-holder It will be impossible to capture absolutely the power of managers via the proportion of their shares In some cases, managers also have the power on making decision in business via their relatives’ ownership and their represented stock of certain organization To the extent that, the existence of family –corporations in Vietnam are not rare, and according to the survey of IFC (2012), family ownership provided negative effect on accounting benefit rate Family ownership is one of special block-holder in Vietnam (Tsao, Chen, Lin, & Hye, 2009; Nguyen & Giang, 2015) While Holderness, Kroszner and Sheehan (1998) defined the managerial ownership included direct and indirect ownership To

be more specific, direct ownership indicated the percentage of stock which managers hold the title, receives any pecuniary benefit from their ownership (including dividend, capital gain), and they also exercise the voting right Otherwise, indirect managerial ownership was considered as the percentage of stocks held by their related people and their represented organizations The previous studies on the relationship between managerial ownership and

1.2 Research objectives

The main aims of this study are two folds: (i) estimating the optimal level for managerial ownership; and (ii) examining and quantifying the relationship between managerial ownership and firm’s performance In addition, this study attempts to explain the behavior of managers and their relatives in relation to selling and purchasing firm’s share As such, the research objectives for this study can be summarized as below

(i) Estimating the optimal level for managerial ownership based on firm’s characteristics and market’s environment;

(ii) Observing the movement of actual managerial ownership level toward the optimal level;

(iii) Figuring out the factors which impact the decisions on the selling and purchasing stocks which have been used by the board of directors themselves and/or their related parties;

(iv) Examining the relationship between the change in managerial ownership and the

change in firm’s performance (both aspects are considered as: market-based

(forward-looking) measurement and an accounting- based (backward-looking) measurement)

1.4 Contributions of the thesis

Several considerable and highlight contributions of this study could be found as follow: (i) First, a new measurement of managerial ownership (both direct and indirect

ownership) is conducted This result created a detailed database of managerial ownership of Vietnamese listed firms

(ii) Second, provided more empirical evidence about the relationship between

managerial ownership and firm’s performance with a different approach compared

to previous studies The contemporary approach envisaged on the changes (increase or decrease) in managerial stock ownership and the changes in firm’s performance instead of the level of two mentioned variables like traditional approach

1.5 Research scope

In general, the study investigates this relation by exploiting database which consisted of

285 non-financial firms listed on HOSE in the period from 2010 to 2015 Nevertheless, in several econometric regressions, some observations are excluded due to the inefficiently matched data

1.6 Structure of the thesis

Chapter1: Introduction

In this chapter, the gap of previous researches, the objectives of study, scope of study, the motivation, and contributions of this study are presented

Chapter 2: Literature review

The chapter reviews the existing theoretical framework, mechanism and empirical evidences of the relationship between managerial ownership and firm’s performance The legal framework of corporate governance in Vietnam, as well as the measurements

of firm’s performance and managerial ownership is also clarified

Chapter 3: Research methodology and data

This chapter exhibits the methodology, quantitative econometric models as well as explains the measurement and definition of variables

5

Chapter 4: Results and discussions

The statistical description, results of estimation and interpretations of the result are discussed and compared to previous studies

Chapter 5: Conclusions and policy implications

Some remarkable findings and proposed policy implications are presented Limitations and further research also could be found in this chapter

6

Chapter 2 LITERATURE REVIEW

In this chapter, the relatively theoretical framework, mechanism of managerial ownership and firm’s performance, together with empirical evidences around the world and Vietnam are presented The structure of these contents is organized as following:

(i) The related theoretical framework helps us explain the mechanism of this relationship;

(ii) The empirical evidences of the relationship between managerial ownership and firm performance around the world and Vietnam with cross-sectional or panel data (might be found);

(iii) The corporate governance of Vietnamese listed firms

2.1 The theoretical background of managerial ownership and firm’s performance

2.1.1 The agency approach

Agency theory, according to Jensen and Meckling (1976) claimed that ownership structure can effect to agency cost They stated that there is conflict between managers (agents) and shareholders (principals) in managing firm’s operation Managers would be responsible for all their decisions even though they do not serve full profit of firms Therefore, with the power, they could pursue the investment plans to generate benefit themselves even these spending can be harm for all firms (shareholders) In the one hand, managerial ownership could be one of the solutions to alleviate the conflict between agents and principals Lower fraction of stock held by managers could result in the lower monitor cost Zwiebel (1995) proposed that the monitor carried by owners could not eliminate absolutely the harm from manager’s decision Beyer, Czarnitzki, and Kraft (2012) proposed that increasing level of managerial ownership will result in managers (agents) behaving like owners (principals) It implies that principal-agent divergence would reduce when the level managerial ownership increased up to the threshold In the other hand, they also argued that the managerial discretion appeared when board of directors accumulated enough power After that, they pursue their own benefit rather than maximizing firm value Cosh, Fu, and Hughes (2007) also advocated the combination of effects between innovation and entrenchment during the increasing managerial ownership level

7

To sum up, an increase in managerial ownership creates both countervailing interest alignment and entrenchment effect So, there are two opposing impacts: the incentive and entrenchment effects of managerial ownership and firm’s performance The effects areprovoked by the separation between principal(s) (the owners) and agent(s) (managers) when the agents make the decision to maximize principal(s)’ welfare

2.1.1.1 The incentive effect

Jensen and Meckling (1976) argued that managers hold stocks; they have an incentive

to adopt investment strategies that are benefit for all company by augmenting the cash flow of firm and reducing outside payment The higher percentage of managerial ownership was

consistent with the higher firm’s performance

Leland and Pyle (1977) claimed that managerial ownership can also serve as a signal for company quality They argued that insiders ownshares to maximize their welfare, so they are also risk-averse which indicates that they consider carefully their investment opportunities Managers increase their percentage of shareholding to signalize more valuable firm and outsidersare convinced that it is good investment In addition to that, Stulz (1988) argued that higher managerial ownership caused the lower probability of hostile takeovers

2.1.1.2 The entrenchment effect

There is a negative relationship between managerial ownership and profitability or firm value, especially at considerably high level of managerial ownership The higher managerial ownership level is, the more difficulty for outside owners control the management, so

managers could make decisions for their own benefit instead of the whole firms

Morck, Shleifer and Vishny (1988), and Stulz (1988) argued that with higher voting right, the appearance of entrenchment effect which indicates the higher managerial ownership would induce the negative impact on firm’s performance Since, it is difficult for external and minority shareholders to monitor and control the firm

Moreover, Hirshleifer and Thakor (1994) indicated that ineffective management has the roots of the inefficiently valuable information getting from takeover market Other argument presented by Fahlenbrach and Stulz (2009), who claimed that the cost of holding more share tends to increase since managers’ portfolio becomes less diversified They are willing to hold more stock when the compensation is proportional or more

Generally, agency theory indicates that there is trade-off between advantage and disadvantage of higher managerial ownership So, there is a nonlinear relationship between managerial ownership and firm’s performance and exists an optimal level of managerial

8

ownership McConnell and Servaes (1990) predicted the inversed U-shaped relationship between managerial ownership and firm performance in terms of market-based measurement They explained that at the lower level of managerial ownership, incentive effect was outstanding the entrenchment effect Then negative effect becomes dominated along with increasing managerial ownership Larcker, Randall and Itner (2003) predicted that firm’s characteristics such as the stage in firm lifecycle, R&D expenditure, assets structure, and growth opportunities could affect to the optimal level of managerial ownership

The effect of managerial ownership on firm’s performance could be demonstrated as figure1.1

Figure 1.1 The relationship between insider ownership and firm’s performance

Firm’s performance

INOWNS Convergences of Entrenchment Convergences of interests

interests effect

Source: Iturralde , Maseda , and Arosa (2011)

2.1.2 The managerial discretion approach

Hambrick and Finkelstein (1987) initially introduced managerial discretion theory which defined as the latitude of managers in making strategic decision process Three reasons

of managerial discretion were identified Firstly, the variance of environment influences on enterprises and managers For example, firms with higher ratio R&D expenditure and advertising spending would signalize for more space of managerial discretion Another reason

is the effect of the characteristics of organization on managers’ actions Two crucial factors are available resources and inertial forces respectively The initial element implies that the inadequately financial resource would limit the manager’s strategic decision The later factorrefers the limitation of manager’s latitude which was especially important in large

9

organization and strong corporate culture The last factor is managers themselves Finkelstein and Boyd (1998) modeled the managers’ compensation and discretion and they discovered that the higher firm’s performance would follow the higher payment of managerial discretion Stulz (1990) and Zwiebel (1996) developed further managerial discretion theory They claimed that managers will select the magnitude of ownership to maximize their utilities instead of enterprises value With this approach, managerial ownership is endogenous instead

of exogenous under managerial discretion theory The cost always is consistent with managers extracting firm’s cash flow when managers carry out for their own benefit To extent to the negative relationship between managerial ownership and firm’s performance are detected since they utilized their discretion to maximize their objectives (Fama, 1980; Jensen, 1986; Brush, Bromiley, & Hendrickx, 2000)

According to Fahlenbrach and Stulz (2008), three key motivations are explored when managers held stock under managerial discretion approach

The financing motivation

When enterprises exercised constrained financial resources, especially start-up firms, managerial ownership could have cheaper cost of capital compared to other external sources Firms with higher information asymmetries will be limited to access other external resources like bank loan or equity from outsider investors When the firm becomes mature, the cost of issuing share or bank loan reduces, so financing motivation of managers is secondary role Therefore, they predicted that the percentage of stock held by managers will decrease over time

The bonding motivation

The interest of managers and minority shareholders can align via the managerial ownership as long as the fraction of their stock do not excess of specific threshold The bonding motivation is emphasized in the case managers being lower reputation or wider managerial discretion space When the firm operated with higher ratio intangible assets, lower growth opportunities or managers being more well-known, the role of this motivation was diminished This phenomenon occurred when organizations operated stably and relatively full grown

The control motivation

Board of directors tends to acquire more shares when their control is threatened Increasing managerial ownership likely occurs when business is not performing well and the

2.1.3 The timing approach

The concentrated content of timing approach is that the insiders will have more information of operation of enterprises compared to outside investors, so they could gain abnormal return It indicates that the managers probably purchase firm stock when the firms performed well likely signifying the overvalued firm share and vice versa (Jenter, 2005)) The market timing theory suggested that managers efficiently beat the market to get extraordinary return This approach provides the same results with the managerial discretion approach, however, two approaches have different root McConnell, Servaes and Lin (2008) investigated the relation change in insider ownership and abnormal return by observing 6-day interval return and their conclusion was change in insider ownership impacting on firm’s performance

2.2 Endogeneity of managerial ownership

Roberts and Whited (2013) proposed that endogeneity issue comes from the correlation between the error term and one of explanatory variables The first reason is omitted variables, which correlated with the error term and independent variable Secondly, the endogenous variables have roots of imperfect measurement to capture one of the proxies (measurement error) Additionally, simultaneity which indicated two variables are determined mutually also induced endogeneity

Demsetz and Lehn (1985) initially established assumptions that under contracting environment, managerial ownership considered as endogenous variable because the ownership level was influenced by the maximizing of firm value of manager’s decision Jensen and Meckling (1976) also claimed that ownership structure, specifically managerial

to monitor the activities of administrators If the proxy for quality of monitoring technology does not include, the negative relationship between managerial ownership and market-based performance of enterprises is probably spurious relation.

Additional example of firm heterogeneity is considered as the fraction of intangible assets When all things being equal, the companies that intensified intangible fixed assets would require higher managerial ownership to limit the space of managerial discretion To extent the market performance measured by Tobin’s Q, for instance, while the denominator is the book value of total assets, the numerator measures as the market value of assets There will be a huge gap between market value and book value of intangible assets, but the former is likely higher than the later Therefore, the unobserved ratio of intangible assets causes the positively spurious relation between managerial ownership and Tobin’s Q

Himmelberg, Hubbard, and Palia (1999) constructed the generally econometric model to explain endogeneity of managerial ownership due to the heterogeneity of firm environment

xit and uit denote observable and unobservable characteristics for firm i at year t We assume unobservable characteristics being time –invariant

𝑃𝐸𝑅𝑖,𝑡 = 𝛼0+ 𝛼1 𝑚𝑖,𝑡+ 𝛼2𝑥𝑖,𝑡+ 𝜏𝑖,𝑡 (3b)

If we use equation (3b) to test the relationship between managerial ownership and firm’s performance, the parameters (or only one of parameter) are (is) consistent as long as the error term 𝜏it in the equation (3b) is uncorrelated with managerial ownership (independent variable) and firm’s performance (dependent variable) Nevertheless, the level of managerial ownership depends on unobservable characteristics, so 𝜏it is correlated with managerial ownership (m)

In term of econometrics, the result could be:

𝐸 (𝑚𝑖,𝑡, 𝜏𝑖,𝑡) = 𝐸 ((𝛽1𝑥𝑖,𝑡+ 𝛾1𝑢𝑖) (𝜕𝛾2+ 𝛾3)𝑢𝑖) = 𝛾1(𝜕𝛾2+ 𝛾3)𝛿𝑢2 (4)

As a result, we cannot estimate equation (3b) by OLS because the coefficients are inconsistent and biased So, the research focused on the change instead of level of managerial ownership

2.3 The empirical evidences of relationship between managerial ownership and firm’s performance and limitations

2.3.1 The research in worldwide and the limitations

2.3.1.1 The exogenous managerial ownership

There are vast number of empirical studies investigated the relationship between managerial ownership and firm’s performance using cross-sectional data, and these researches showed different shapes of the relationship However, general conclusion could be that relationship between firm’s performance and managerial ownership is nonlinear

Morck et al (1988) examined the relationship between managerial ownership and Tobin’s Q with cross-sectional data of 500 Fortune firms which were likely large-scale enterprises and operating stably They explored the threshold of managerial ownership fluctuating from zero percent up to 5 percent which provided positive effect on Tobin’s Q is one of proxies for firm’s performance in terms market-based measurement Then, the Tobin’s

Q would reduce if managerial ownership increases up to 25 percent and the repeatedly

it confirmed such non-monotonic relation

Short and Keasey (1999) strengthened the mentioned argument by providing more empirical evidence in United Kingdom They employed two proxies of firm’s performance that are the ratio market of total assets over the book value of equity (VAL) and return on equity (ROE) regressed against a cubic function of level of managerial ownership To the extent market-based measurement, VAL fluctuated with the shape of relation likely the result

of Morck et al (1988) Nevertheless, managerial ownership (percentage) thresholds varied distinctively More specifically, the first positive threshold is 12.99 percent Then, the negative relation took place until the managerial ownership reaching 41.99 percent-threshold while the positive impact would recover In terms of accounting-based measurement such as ROE, the same correlated graph with VAL was explored with diverse turning points which were 15.58 percent and 41.84 percent, respectively

Kole (1995) conducted experiments of 352 firms instead of 500 Fortune firms like study

of Morck et al (1988) and explored that N-shaped relationship between level of managerial ownership and Tobin’s Q, but the turning points considerably altered The positive effect of managerial ownership on Tobin’s Q is more significantly sustained in small firm than large firm

Hermalin and Weisbach (1991) analysed the triennial data of 142 firms listed on NYSE

in 1971, 1974, 1977, 1980 and 1983 in the order and the inversed W-shaped relation between managerial ownership and Tobin’s Q had been found To be specific, the beginning is positive relation of managerial ownership up to 1 percent, thereafter this relationship reverses when managerial ownership fluctuates in the scope from 1 percent to 5 percent When managerial ownership continues increase excess of 5 percent, positive effect accounts a dominant

2.3.1.2 The endogenous managerial ownership

As mentioned in the previous section, Demsetz and Lehn (1985) provided some arguments that under contracting environment, the importance of unobserved heterogeneity need to be emphasized Himmelberg, Hubbard, and Palia (1999) investigated the relationship between managerial ownership and firm’s performance with panel data including 600 US firms at random from 1982 to 1992 period which took endogeneity issue into consideration They employed fixed effect model to estimate the relationship between the fraction insider ownership and firm value The result of study showed that no econometrically significant impact and regression results in previous studies probably reflect the spurious (or false) relation

The endogenous managerial ownership has been approved by vast number of researchers and some econometric solutions to overcome this issue have been conducted Notwithstanding, the explanation of estimated results that the change in managerial ownership has been taken advantage to change firm value provoked the controversy

The first mentioned remedy is to use panel data and fixed effects Fixed effect can accommodate unobserved heterogeneity when the authors assume unobserved heterogeneity being time-invariant The disadvantage of this method is that it could be inefficient to eliminate the spurious relationship between managerial ownership and firm’s performance absolutely since the FE model insinuatingly assumed that the omitted variables are time-invariant The limitation of fixed effect argued by Zhou (2001), since he observed and found that very small and slow changes of managerial ownership in fiscal year accompanied with dramatically change in firm value, especially Tobin's Q He also stated that it is difficult to explain the relationship between two subjects, because power of these tests is less effective In essence, Holderness, Kroszner and Sheehan (1998) summarized and proposed that the level of managerial ownership of firms listed on NYSE and AMEX had change from time to time

15

The overtime change included systematic and unsystematic change The Zhou’s explanation indicated that the origin of systematic change could be the change in contracting firm’s environment Additionally, the root of unsystematic change should come from random incentives such as market conditions Therefore, FE model cannot completely eliminate the spurious relation

An alternative remedy should be employed that is instrument variable (IV) in several studies The authors based on Hermalin and Weisbach’s study (1991) which suggested the lagged control variable as instrument variable of managerial ownership The natural logarithm

of sales, and its square considered as proxy of size of firm as well as idiosyncratic risk (SIGMA, SIGDUM) considered as proxy of managerial risk could be potential candidates of instrument variables They also argued that the Hausman test just answered the question as whether the existence of endogeneity problem or not, without confirmed the validation of instrument variable It is difficult to find completely efficient instrument variable of managerial ownership since lagged unobserved explanatory variables could continue to be endogenous Therefore, instrument variable (IV) did not eliminate completely the problem and provided inefficiently reliable estimation

The study of Cui and Mak (2002) investigated the relationship between managerial ownership and firm’s performance in intensive R&D firms listed on AMEX, NASDAQ and NYSE in two year 1996 and 1998 They employed the many alternative piecewise regressions and used Two Stage Least Square (2SLS) to examine the result The W-shaped relationship between managerial ownership and Tobin’s Q was found which is opposite the result of Hermalin and Weisbach’s study was exposed above However, unlike previous research, the result of regression was inconsistent in two considered aspects of firm’s performance (market-based and accounting-based measurement)

Next, Drakos and Karathanassis (2004) claimed that after controlling the fixed effect and random effect, there was insignificant relation in 59 Greek listed firms in from 1996 to

1998 period The study based on exogenous managerial ownership assumption, however, the deficiency of data leads to the impossible to implement robust regression Therefore, Drakos and Bekiris (2010) continued to investigate the impact of ownership structure on firm’s performance of 146 firms listed on Athens stock exchange from 2004 to 2010 They improved their research by employed by simultaneous equations and 2SLS and 3SLS (Three Stage Least Square) with lager number of firms than the past sample to overcome the endogenous issue The result of study showed that the positive relation between Tobin’s Q and managerial

16

ownership possess in all three regressions, and dividend payout ratio is used as the instrument variable Wu (1973) and Hausman (1978) proposed the test to check the valid of instrument variable, and all instruments pass this test which indicates instrument variables being exogenous

Last but not least, Mueller and Spitz (2002) examined the effect of managerial ownership in 1300 medium and small German firms over the period from 1997 to 2000 The Generalized Method of Moments (GMM) and dynamic panel data utilized in order to take impact of past firm’s performance on contemporaneous one into account The result of study consolidated the inversed U-shaped relation between managerial ownership and firm’s performance and the positive relationship between past firm’s performance and concurrent ones They found that managers worried about the concentrated on holding one specific share because of their demand of diversified portfolio They also concluded that managerial ownership would provide a signal of firm quality

Coles, Lemmon, and Meschke (2012) explored the relationship by using parsimonious, structural model to obtain productivity parameters and managerial ownership estimation (Q* denotes for the optimal productivity parameter) The similar relation between managerial ownership and firm’s performance was found However, the simulated impact of unobservable variable year by year (such as productivity) on Tobin’s Q was nonlinear Additionally, structural productivity parameters also used some firm’s attributes such as firm size, the ratio of R&D or advertising expenditure and book leverage as proxies Together with, they raise questions about the misspecification of model and did not completely eliminate the endogeneity issue Roberts and Whited (2013) acknowledged this limitation and endogenous issues in the financial corporate and suggested some other measures to overcome the problem of endogeneity

Due to the limit of econometric techniques in dealing with endogeneity issue and explanation the results, some recommendations are concentration on changes in managerial ownership instead of level like previous studies Himmelberg, Hubbard, and Palia (1999) suggested that taking into considerations of managerial ownership changes help us understand

more deeply about the relationship between managerial ownership and firm’s performance

Additionally, Fahlenbrach and Stulz (2008) investigated the relationship between the change

in managerial ownership and the change in Tobin’s Q of American firms Unlike previous

researches, the study attempted to include the dynamic change of managerial ownership They found that after firms operating efficiently, managers tend to reduce the percentage of their

17

ownership because they need more diversified portfolio However, the rate of managerial ownership does not increase in the case of poor firm’s performance They also argued that the information about the change in managerial ownership fully absorbed by the market after a period so change in managerial ownership in year t-1 should affect to Tobin’s Q in year t The same result confirmed by Helwege, Pirinsky and Stulz (2007) that managers would sell their stock after firms being good performance and stock market being liquidity They also utilized Probit regression to estimate the probability of insider ownership reduction regard of the change of firm’s characteristics

McConnell, Servaes and Lins (2008) found that managers did not purchase firm share to move their ownership rate toward the optimal managerial ownership level as the agency theory approach Moreover, a regression of abnormal return against the change in insider ownership, square of insider ownership level and the interaction of the change and level of insider ownership was conducted The result was positive coefficient of change in insider ownership and negative estimators of two later variables, which was similar to Fahlenbrach and Stulz’s study (2008) which carried out by employed event study approach Generally, the change in managerial ownership can be explained partially by the managerial discretion theory

Firdaus and Kusumastuti (2012) investigated the same issues in the relation to firm’s life cycle They found that ownership structure (insiders ownership specifically) impacted on firm’s performance differently in each stage (growing firms, mature firms and stagnant firms)

of firm’s life cycle

2.3.2 The empirical evidence in Vietnam

Do and Wu (2014) investigated the impact of ownership structure on firm’s performance, which included non-financial firms listed on Ho Chi Minh Exchange from 2009

to 2012 Their evidence convinced that no statistically significant evidence of managerial ownership on firm’s performance in terms of accounting-based measurement such as ROA and ROE In addition to that, the entrenchment effect is dominant in state-owned enterprises (SOEs) compared to other type of enterprises There was a limitation, that is, firm’s performance measured by ROA and ROE (accounting–based measurement) which were historical performance can be affected by accounting method and estimates such as depreciation expenditures The authors employed regression with panel data which estimated directly the impact managerial ownership on firm’s performance As argument in previous section, regression has trouble in endogeneity problem or spurious relation

18

Nguyen and Giang (2015) investigated the related ownership concentration and firm’s performance in 34 construction and material construction firms on HOSE, HNX and UPCOM In terms of econometrics, POLS and FE used to estimate this relationship Two proxies for firm’s performance: price over earning per share (P/E) and ROA (return on assets) are employed The result showed that concentration ownership has no impact on ROA which represents historical performance, but negative impact on P/E They argued that negative effect could come from expropriation of block-holder The improvement of this study is firm’s performance could be measured in both aspects: backward (historical perspective) and forward looking (future perspective)

The U-shaped relation between the rate of firm shares owned by board of director and ROA was shown in an empirical study with the sample of 77 firms listed on HOSE from 2006

to 2011 (Vo & Phan, 2013) The Flexible Generalized Least Squares (FLGS) estimation employed and found the threshold of managerial ownership being 22 percent More specifically, the negative relation of MO and ROA takes place until the level of share owned

by board of director excess of 22 percent, and then opposite effect is predominant

Hoang, Nguyen and Hu (2016) explored the impact of ownership structure which includes managerial ownership and firm’s performance of non-financial firms listed on HOSE

in (2007 – 2015) period The proxy of firm’s performance was Tobin’s Q and system- GMM were also employed to eliminate the spurious relation The cubic model included the lagged firm’s performance (t-1), managerial ownership, its squares, its cubic, and some common firm’s attributes The first element indicated past performance could impact on contemporaneous ones The study provided the evidence that confirmed the non-monotonic relationship between the levels of managerial ownership to Tobin’s Q Unsurprisingly; the result is consistent with previous studies N-shaped relation implied the positive effect on firm’s performance at the lower and higher managerial ownership, and at the reasonable level

of managerial ownership, negative effect on firm’s performance was found

2.4 The corporate governance of Vietnamese listed firms

For emerging markets, the enhancement of corporate governance was emphasized to protect the investors, and ensured market transparency Listed firms have operated under two tiers of corporate governance which were General Meeting of Shareholders (GMS) and Board

of Management (BOM), respectively The Law on Enterprise 2014 which was effective from the beginning of 2015 has some change compared to Law on Enterprise 2005 to improve transparency of information and quality of management

19

The minimum percentage requirement of voting right and quorum reduced the gap between Vietnamese Law Enterprise with jurisdictions (IFC) The Law of Enterprise 2014 also stipulated that minimum of 20 percent of members in board of directors is independent and supervised operating activities of enterprise In addition, the issues of transparency and disclosure information of internal and large shareholders received greater attention in the new legislation Law of Enterprise 2014 made amendment that CEO, chairman or other management personnel have to disclose their ownership themselves and related party of ownership in firms even though in other firms as long as total percentage of stock holding excess 10 percent

Figure 1.2 The management structure of Shareholding Company

Source: Le and Walker (2008)

Le and Walker (2008) proposed that “the legal framework and institutional foundation for the capital markets in Vietnam are in an early stage of development” They also argued that the listed firms have to abide the Law of Securities 2006 and that firms experienced the inefficient flexibility and accountability The Law of Enterprise 2014 reformed in some points

to improve the flexibility in administrative procedure and merger and acquisition (M&A) activities such as allowing multiple legal representatives instead of sole person; merging the business without limitation of type firms

General Director

(CEO)

20

Figure 1.3 The internal governance structure of a listed company

This figure represents the relation in listed firm including powers and responsibilities of parties The solid line represents for appointment and dismissal right, and the dashed lines represents for monitoring function

Source: Le and Walker (2008)

According to the mechanism, to separate the control and supervisory of firm operation, requirement of one third of members of Board of Management must be non-executive independent member

Management

21

2.5 The conceptual framework

Figure 1.4 The conceptual framework

Source: Author’s analysis

Conclusion of this chapter

This chapter provided the relevant theories and empirical evidences to explore the relationship between ownership structure and firm’s performance The endogenous issue has also been recognized and some remedies are also suggested to solve problem However, the explanation of regression results has provoked disputation Therefore, the study continues investigating an alternative approach by focusing on change instead of the level variables like previous studies

Change in Managerial Ownership (Increase/ Decrease)

Market-based measurement (∆ Tobin’s Q)

Change in firm performance

Accounting based measurement (∆𝑅𝑂𝐴)

Change in Market environment:

22

Chapter 3 RESEARCH METHODOLOGY AND DATA

To achieve research objectives, this chapter details the some contents as follows: the sample size and data source employed in study Next, the measurement of variables and constructed empirical model are also found in this chapter

3.1 Data sources

In Vietnam, firms have been listed on two stock exchanges, HNX (Hanoi stock exchange) and HOSE (Ho Chi Minh Stock Exchange) There are some different criteria about the firm size, the accumulated accounting profit for the listed firms, so in order to get homogenous sample, the research only focuses on firms listed on HOSE The study also excluded securities firms and financial companies out the scope of research because they operated with different rules about capitalization, capital structure, and specific regulations such as the level of ownership ceiling of group of related shareholders Due to the deficiently financial information from availably official data sources, the data collected from some sources such as: annual report, firm’s prospectus, security firms’ website Data was collected with a best effort of author, especially the rate of shares owned by board of directors and related parties The lack of information of listed and delisted firms in the observable span leads to the unbalanced panel data which was exploited in this study At the end of 2015, there were 341 companies listed on HOSE, and author removed 28 financial companies; 28 firms violated security regulation or inefficient requirement information, so it were only 285 companies from 2010 to 2015 included in sample

In this study, managerial ownership (MO) includes direct and indirect ownership which measured by the percentage of share owned by board of directors themselves and the related parties which defined at Article 28 of Circular 52/2012- BTC and amended by Circular 155/2015-BTC Indirect ownership is also defined as the percentage of share that the members of board directors deputized for organization (Neely, Gregory & Platts, 1995) Other argument had done by Koufopoulous, Zoumbos and Argyropoulous (2008) claimed that in management scheme, performance could be viewed in two terms quantification and accounting So, firms should been managed their operational processes to achieve their objectives Measuring performance of businesses in order to assess the level of success in

23

management of firm’s resources in a certain period got consensus from great number researchers (Demirbag & Zaim, 2006; Gedennes & Sharma, 2002) The measurements of performance concept have been developed which also help compare firm’s achievement in span of time

The corporate governance scheme played a key role in the development of firm’s performance, and established appropriately corporate governance would enhance the expected

firm’s performance (Ehikioya, 2009) In practice, measurement of firm’s performance reveals

valuable information and communicates with other entities, especially outsiders about the effectiveness of firm’s operation With firm’s performance measurements, they enable quantification, and understand much easily a complex concept as well as more convenient for the evaluation (Lebas, 1995)

3.2 Measurement variables

3.2.1 Definition and measurements of firm’s performance

3.2.1.1 Accounting – based measurements

Accounting–based measurements generally view towards profitability aspect which used for comparison with competitors or in the relation with risk In addition, these indicators are often influenced by the estimates of future expenditure such as depreciation or provision Such indicators also limited because of accounting conventions and the method of record of

value of assets (Kapopoulos & Lazaretou, 2009)

ROA (Return on assets - one of accounting-based measurements) defined as the ratio of earning after tax over book value of total assets which represented the ability generated profit

of total assets ROA was used to demonstrate for firm’s performance (Hu & Zhou, 2008; Mehran, 1995; Demsetz & Lehn, 1985; and Vo & Nguyen, 2014) Moreover, Ibrahim and AbdulSamad (2011) claimed that ROA reflects the effectiveness in using the firm’s assets to serve the economic benefit of its shareholders regardless of the type of resources financing to total assets (capital structure) According the statistical result of Macrothink Institute, ROA is the most common measurement of firm’s performance employed by scholars

24

3.2.1.2 Market–based measurements

The market-based measurements reflect the investors’ expectation about the ability of generating profit of firms in the future (forward-looking) While, accounting-based indicators measure the short term profit, in contrast, market-based focused on the expected the long-run profitability (Bozec, Dia, & Bozec, 2010)

Tobin’ Q is extensively used to capture the firm’s performance in market aspect This ratio calculated by the ratio market value of firm over the replacement value of assets which devised by Tobin and Brainard (1969) Econometrically, the numerator is total market value

of common, preferred stock and total liabilities, and the denominator is the book value of total asset (the replacement cost of production capacity) However, inefficiently Vietnamese debt market, we cannot obtain the market value of liabilities so Tobin’s Q can be calculated by:

𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 = Market value of common and preferred stock+book value of liabilities

Tobin’s Q has also been a common measurement of firm’s performance and was widely used by large number of authors, for example (McConnella, Servaes, & Lins, 2008; Kole, 1997; Fahlenbrach & Stulz, 2009; Coles, Lemmon, & Meschke, 2012; Firdaus & Kusumastuti, 2013; Hoang, Nguyen, & Hu, 2016)

3.2.2 Definition and measurement of managerial ownership

There are different measurements of managerial ownership concept Initially, managerial ownership is defined as the proportion of stock held by all block holders and insiders which included all managers and offices of firms (Holderness, 2008) Another definition of managerial ownership is as the proportion of stock held by board of directors, excluding stock option (Cho, 1998) In empirical researches, Agrawa and Knoeber (1996) utilized the less common proxy for managerial ownership which calculated by the fraction of stock held by CEO Other measurement of managerial ownership would be the total percentage of stock held by board directors and their families (Short & Keasey, 1996) However, most of mentioned measurements just captured the direct ownership In order to clarify the pattern of managerial ownership, Securities and Exchange Act (SEC) in the year of

1934 required the public firms register the percentage of stock held by board of directors Moreover, SEC’s definition of managerial ownership included indirect and direct ownership

To be clearer, direct ownership indicates managers hold the title, voting rights as well as

25

pecuniary benefit of stock like dividend or capital gain (or loss) While indirect ownership means that managers can control firm via the voting right even though they do not hold stocks’ title and pecuniary benefit Shares held by family members and their representative organization which consider as indirect ownership

This study followed the definition developed by Holderness, Kroszner and Sheehan (1998) which measured total indirect and direct ownership of members of board of directors excluding chief accountant as managerial ownership level

3.3 Research methodology

To answer the objective questions, three parts of analysis are implemented In the first part, we investigated the determinants of optimal level of managerial ownership, and the actual level moving toward the optimal level or not The second part is explanatory elements

of selling and purchasing stock related parties and managers’ decision are investigated by using Probit regression The last part exhibits the relationship between the change in managerial ownership and change in firm’s performance which will be formulated by using POLS, FE, and RE The Stata 12 software employed to investigate and analyze data Initially,

the descriptive statistics, correlation analysis and calculating VIF utilized to summarize the

data by years and by industries to explore these relations However, to assess more specific direction and magnitude of the impact of these factors, the multivariate regressions are used The generally econometric model could be represented as following:

be carried out respectively in study

The result

H 0 : POLS is not rejected H 0 : POLS is not rejected POLS

H 0 : POLS is not rejected H 0 : POLS is rejected RE model

H 0 : POLS is rejected H 0 : POLS is not rejected FE model

H 0 : POLS is rejected H 0 : POLS is rejected H 0 : RE is reject FE model

H 0 : POLS is rejected H 0 : POLS is rejected H 0 : RE is not reject RE model

Source: Author’s summary

However, some diagnostics are checked to get the best linear unbiased estimator (BLUE) The Wald test is conducted to check group wise heteroskedasticity and the Wooldridge test checking autocorrelation So, the robust standard error conducts to get more efficient estimators The magnitude of estimators is unchanged, but the value of standard error reduces

3.4 The empirical model

3.4.1 The determinants of firm’s performance and optimal managerial ownership

3.4.1.1 The determinants of firm’s performance

The multivariable regression advocated by the vast number of scholars which is employed to estimate the firm’s performance as the following:

𝑃𝐸𝑅𝑖,𝑡= 𝛽0+ 𝛽1𝑀𝑂𝑖,𝑡+ 𝛽2𝑀𝑂𝑖,𝑡2 + 𝛽3𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽4𝐿𝐸𝑉𝑖,𝑡+ 𝛽5 𝐺𝑅𝑂𝑊𝑖,𝑡 + 𝛽6𝑅𝐼𝑆𝐾𝑖,𝑡+ 𝜀 (6)

where:

 PER is firm’s performance measured in two aspects accounting-based and

market-based measurement

 MO is the level of managerial ownership measured by the proportion of share held

by managers and their related parties

 SIZE is the natural logarithm of annual sales

 LEV is the ratio of book value of long-term debt over total assets

 GROW is selective proxies for growth as RDTA and investment ratio (CAPEXTA)

 RISK is the idiosyncratic risk (SIGMA)

Leverage

The relationship financial leverage and firm’s performance was presented on great number of studies Ilyuklin (2015) found an empirical evidence in Russia to emphasize the positive impact of leverage on firm’s performance thank to the monitor of creditors Safieddine and Titman (1999) also advocated positive impact of leverage because leverage is considered as takeover defense Nevertheless, controversy findings about the relationship between leverage and firm’s performance was consistent with pecking order theory (Javed, Rao, Akram, & Nazir, 2015; Gleasona, Mathur, & Mathur, 2000)

Growth

Two proxies of growth in model are exploited in this study are ratio R&D expenditure over total assets (RDTA) ratio and budget expenditure over total assets (CAPEXTA) A large number of studies proved the positive impact of growth opportunities on firm’s performance because investors will be willing to put much money to invest in enterprises with higher growth opportunities The higher Tobin’s Q usually exists in firms in accordance with larger R&D expenditure and higher investment ratio (King & Santor, 2008; Dimitrios & Psillaki, 2010; Jiraporn & Liu, 2008)

Risk

Some empirical evidences revealed the negative effect of risk on firm’s performance in terms of market evaluation Some arguments would explain the root of this relationship To be specific, the stock of firms operating riskier would be underestimated by investors (Zeitun & Tian, 2007) Bloom and Milkovich (1998) indicated that the excessive volatility of cash flow, profits or higher financial distress cost could lead to higher probability of bankruptcy

28

3.4.1.2 The determinants of optimal managerial ownership level

The model exploited by Himmelberg, Hubbard, and Palia (1999) which was widely recognized and approved by the vast number of scholars

The general model in order to estimate the optimal managerial ownership level should

be

𝑀𝑂𝑖,𝑡∗ = 𝛽 1 + 𝛽 2 ln(𝑆) 𝑖,𝑡 + 𝛽 3 (ln(𝑆))𝑖,𝑡2 + 𝛽 4 𝐾𝑇𝐴 𝑖,𝑡 + 𝛽 5 𝐾𝑇𝐴𝑖,𝑡2 + 𝛽 6 𝑅𝐷𝑇𝐴 𝑖,𝑡 +

𝛽7𝑌𝑆𝑖,𝑡+ 𝛽8𝑆𝐼𝐺𝑀𝐴𝑖,𝑡+ 𝛽9𝐼𝑁𝐷𝑈𝑀𝑖+ 𝑣𝑖+ 𝜀𝑖,𝑡 (7) Managerial ownership level (m) is measured by the transformed managerial ownership (natural logarithm (1-m)) which m is the percentage of shareholding by the board of director

in two terms direct and indirect ownership All of firms listed on HOSE are classified into nineteen industries, so eighteen dummies variables are utilized to show the differences among industries The definition and measurement of variables in equation (7) are presented in table below

Table 3.2 The determinants of optimal managerial ownership level

Dependent variable

Ln (1-m) Transformed managerial

ownership

Ln (1-m), where m is the percentage of indirect

ownership and direct ownership of managers

Control variables

Ln 2 (S) The square of firm size Ln 2 (Sales)

KTA The ratio of tangible assets

over the total assets

Tangible assets Total assets KTA 2 The square of the ratio

tangible assets over the total assets

YS Ratio of operating income

over sales (Operating Income

RDTA Ratio of R&D expenditure

over total assets (R&D expenditure

RDUM R&D dummy RDUM is equal 1 in the case of availability of

R&D expenditure and otherwise being equal

to zero CAPEXTA Investment expenditure 𝐶ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑓𝑖𝑥𝑒𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 + 𝑑𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛

𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 SIGMA Idiosyncratic stock price risk Standard deviation error term taken from

CAPM regression

INDUM The code uses to divide our

sample by the industries

18 dummy variables are utilized for 19 industries

Source: Author’s summary

29

In static model for panel data, 3 regressions are used: POLS, FE, and RE In addition, industry dummy variables are also included in regression to show the industry effect The model helps us predict the optimal level of managerial ownership for each firm

The independent variables in equation (7)

Size

Himmelberg, Hubbard, and Palia (1999) stated that the size of firm posed ambiguous effect on the optimal level of managerial ownership He argued that larger enterprises with more complex operation and organizational structure could induce higher monitor and agency cost So, the result could be higher level of managerial ownership In addition, large firms usually hire proficient administrators which also imply that they are wealthier as well as more likely hold higher proportion of firm share Otherwise, the authors documented that larger firms could take advantage of economies of scale in tracking firm’s operation or serve the monitor of rating agency, securities commission and banks This argument was advocated in some previous studies (Fama & French, 1995; Do & Wu, 2015) It suggested that larger enterprises are consistent with lower optimal managerial ownership level The measurement

of firm size could be the natural logarithm of total assets (Abor, 1999) or annual sales (Himmelberg, Hubbard, & Palia, 1999) In this study, Ln (S) is exploited to measure the size

of firm and smoothies data And, Ln2(S)utilized to capture the nonlinear relationship

The scope for discretion spending

The growth opportunities

It is easier to monitor against firms with higher “hard” investment because tangible assets are more observable than intangible assets The ratio tangible assets over total assets (KTA) is used to measure the hard capital, and (KTA)2 also added to explore the nonlinear relationship which also employed by (Wiwattanakantang, 2013) In addition, the ratio R&D expenditure to tangible assets (RDTA) captured “soft” investment In addition, their dummy variable (RDUM) utilized to show non-reporting R&D expenditure firms Since some enterprises do not separate R&D expenditure, so some observations miss this information The exclusion of non-reported R&D firms could lead to the selection bias, because the sample just focused on the intensive R&D firms Gertler and Hubbard (1988) provided the empirical evidence demonstrating a positive relationship between level of R&D intensity and the optimal managerial ownership level Some researchers also employed the ratio R&D expenditure over total assets (Coles, Daniel, & Lalitha, 2006; Coles, Lemmon, & Meschke, 2012) or over annual sales (Bebchuk, Cremers, & Peyer, 2011) to demonstrate the growth