ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Insider Ownership and Firm Value

This paper examines the effect of insider ownership on corporate value in India for the period of 2000-01 to 2003-04, using 1,833 Bombay Stock Exchange listed firms by investigating the relationship between insider's equity holding and firm value. While the "convergence of interest" or "monitoring" hypothesis predicts a positive relationship, the "entrenchment" hypothesis predicts a negative one. This paper also provides evidence that the relationship between insider shareholding and firm value is not linear in nature and documents a significant non-monotonic relationship between the two. Tobin's Q first increases, then declines and finally rises as ownership by insiders rises. It also confirms that foreign promoter/ collaborator shareholding has a significant positive impact on firm value.

Insider Ownership and Firm Value

Evidence from Indian Corporate Sector

This paper examines the effect of insider ownership on corporate value in India for the period of 2000-01 to 2003-04, using 1,833 Bombay Stock Exchange listed firms by investigating the relationship between insider’s equity holding and firm value. While the “convergence of interest” or “monitoring” hypothesis predicts a positive relationship, the “entrenchment” hypothesis predicts a negative one. This paper also provides evidence that the relationship between insider shareholding and firm value is not linear in nature and documents a significant non-monotonic relationship between the two. Tobin’s Q first increases, then declines and finally rises as ownership by insiders rises. It also confirms that foreign promoter/ collaborator shareholding has a significant positive impact on firm value.

MANOJ PANT, MANORANJAN PATTANAYAK

A
fter the successful takeover of Arcelor by L N Mittal, the Tata group chairman Ratan Tata made a statement that, “The steel industry is fragmented and considerably vulnerable. The only safeguard is to increase the founding family’s stake over time…the increase in stake is to ensure that management control remains with promoters” (The Economic Times, July 7, 2006). The entire Reliance fiasco started with a statement by Mukesh Ambani where he admitted that ownership is an issue in Reliance Industries. Finally, the company was split into two groups. While discussing issues on corporate governance, Rahul Bajaj said, “More than 75 per cent of large listed Indian companies are family owned, in which a family has a significant shareholding in the company…companies where management has little or no stake in the company constitute less than 5 per cent of large, listed companies. In a company managed by ‘owners’, there is a very strong motivation for management to work for a longterm share price increase” [Business Standard, July 29, 2005]. These three unrelated facts have one commonality, that is, the issue of ownership, control rights and shareholder’s wealth maximisation.

Which governance structure is appropriate for an emerging economy like India? Do family-run business entities perform better than professional-run firms? Does the widely held firm command better stock prices than one with ownership concentration? As the Indian stock market is expanding in terms of volume of trading and market capitalisation, these questions need to be addressed by regulatory bodies like the Securities and Exchange Board of India (SEBI). Yet, very little work exists on these issues in the Indian context. In this paper, we will look at some of these issues in an econometric framework.

The issue of ownership, control rights and firm value so far has drawn wide attention from academia in advanced countries. All theoretical and empirical research on the relationship between equity ownership and performance is influenced by the separation thesis of Berle and Means (1932). However, extant literature reveals that concentrated holding is widely prevalent around the world [Holderness 2005; Claessens et al 2000; Laporta et al 1998]. The convergence-of-interest or monitoring hypothesis predicts a positive relationship between ownership concentration and firm performance; at the same time the entrenchment hypothesis proposes a negative one. Some authors argue that both the effects operate at different levels of shareholding, thus resulting in a non-linear relationship between insider ownership level and performance. However, the veracity of several predictions can only be empirically examined.

In India, ownership structure is highly concentrated in the hands of family members and their acquaintances. The post-liberalisation capital market in India is showing steady progress with a large number of small shareholders participating in the equity market. When such a fortune is concentrated in the hands of a few people, it draws our immediate attention. The east Asian crisis reminds people how a poor governance structure can ruin an economy within no time. The family based governance model failed the economies of the east Asian countries and created havoc around the world [Claessens 2000; Lemmon and Lins 2003]. Therefore, it is imperative to study the performance of firms in India where family and kinship ties still dominate business decisions. Against this backdrop, the current paper measures the performance of Indian firms where insider ownership stake plays a key explanatory variable. Due to a large shareholding concentration in the hand of families, professional managers have no option but to execute the decree of insiders/promoters of the firm. In a majority of firms, there is no separation of chief executive officers and chair; therefore, family supremacy is more overt than covert. With a moderately developed capital market, a non-existent market for corporate control and weak product market competition, India presents a unique opportunity and challenge to study the behaviour of insiders.

To outline the organisation of the paper, Section I reviews the related empirical literature. The major hypotheses of the study are stated in Section II. In Section III, we have explained the variables used in the econometric model. The descriptive results are analysed in Section IV. The empirical model and results are discussed in Section V and a summary of major findings is reported in Section VI.

I Insider Ownership and Firm Value: Evidence

There are a large number of studies on ownership, governance structure and firm performance [for a survey see, Shleifer and Vishny 1997; Classens et al 2002; Holderness 2003]. However,

Figure 1

Y Firm Value

0 Z1

Z2 Per cent of

X Convergence Convergence Insider Share of Interest of Interest

Entrenchment

most of the attention has been given to corporate governance problems in developed countries. Here we have attempted to bridge this gap by focusing on an emerging economy like India.

Demsetz (1983) maintains that ownership structure is an endogenous outcome of the maximisation process. Therefore, every change in ownership level is made in order to maximise shareholder profit. Consequently, the ownership concentration and the profit rate are unrelated. Demsetz and Lehn (1985) and Demsetz and Villalonga (2001) in their respective studies find ownership concentration and firm performance to be unrelated. Holderness et al (1999) find that ownership has a non-linear impact on firm performance. Morck, Shleifer and Vishny (1988, henceforth MSV) attempt to capture the convergence-of-interest and entrenchment effect in a sample of 371 Fortune 500 firms. Using a piecewise linear regression function they find a significant non-monotonic relationship between board ownership and firm performance. McConnell and Servaes (1990) examine the relation between Tobin’s Q and ownership structure for a sample 1,173 US firms for 1976 and 1,093 firms for 1986. They find strong evidence of a curvilinear relation between insider ownership and Q.

Short and Keasey (1999) study the relation between managerial ownership and firm performance from the UK perspective. They find that for both the return on shareholders’ equity (RSE) and market value to book value ratio (VAL), the coefficients of director shareholding (DIR), DIR2 and DIR3 are positive, negative and positive respectively and all are statistically significant. Faccio and Lasfer (2000) also document similar findings for high growth firms in UK. Cho (1998) is very critical about the implicit assumption of exogenous ownership structure in several studies. He has used a system of equations for insider ownership, corporate value and investment. His result shows that investment affects corporate value, which in turn affects the ownership structure but not vice-versa. Himmelberg, Hubbard and Palia [HHP 1999] using a fixed effect panel data method to control for unobserved firm level heterogeneity conclude that managerial ownership has no statistically significant effect on firm performance.

There are few studies available on India which are explicitly designed to address the issues of the institutional structure of production. Most of the studies which are undertaken in the context of India are at the industry level and structured in the traditional neo-classical framework where institutional mechanisms are suppressed. In the later phase of the 1990s, some Indian scholars have tried to fill this gap by exclusively focusing on India. Studies by Chhibber and Majumdar (1998, 1999), Majumdar and Chhibber (1998), Patibandla (2006), Sarkar and Sarkar (1999, 2000), Khanna and Palepu (1999, 2000, 2004) are the few studies on the Indian corporate governance system.

Chhibber and Majumdar (1999) study the impact of foreign shareholding on a firm’s profitability and contrast the pre-reform and post-reform character of such holdings. After controlling for a variety of firm and environment-specific factors, the study finds that post-reform foreign ownership holdings have significant effects on firm profitability. Khanna and Palepu (1999) find that insider ownership has a positive and significant impact on firm value while director’s holding has no perceptible impact. The most descriptive study on Indian corporate governance has been done by Sarkar and Sarkar (1999) using micro level data of 1,613 firms for the year 1995-96. Director’s shareholding shows a non-linear relation with firm value – i e, it first decreases up to 25 per cent and increases thereafter. In the case of corporate bodies’ shareholding, beyond 15 per cent Q-ratio shows a positive association with firm value. For foreign shareholdings, there is a monotonic increase in shareholder value. Sarkar and Sarkar (2005) provide evidence that promoter shareholding has no impact on firm value in case of low growth firms while it has a positive impact on firm value for high growth firms.

II Major Hypotheses

The empirical and theoretical literature suggests that the relationship between managerial/insider ownership and firm value is non-linear in nature. This non-linearity is ascribed to the convergence of interest and entrenchment hypothesis. When the shareholding of insiders is very low, the entrenchment effect is non-operational due to less control over the decision-making process of the firm. However, once they gain controlling authority in the firm, they can entrench themselves or pursue non-value maximising activities. As per the entrenchment hypothesis, more equity ownership by a manager/insider may lead to lower financial performance. But with majority ownership (say, more than 50 per cent), their interest is better aligned with the interests of the firm. They have to bear 50 per cent of the loss of each penny forgone. It may be noted that incentive effects operate positively at all levels of ownership. Each increment in shareholding induces the manager/insider to perform. However, it is the dominance of the entrenchment effect over the incentives effect at a medium level of ownership that drives the value of the firm downwards. Considering the above arguments, we test for any non-linearity in the relationship between insider ownership and market value. More formally, the hypothesis is stated below: Hypothesis I: Firm performance is a non-monotonic function of share ownership by insiders. In other words, firm value first increases, then decreases and thereafter increases with insider ownership.

In a graph (Figure 1), the hypothesis can be described as follows:

Foreign promoters/collaborators act as strategic partners for a domestic corporation when they come up with technological expertise. The technological and organisational advantages of foreign firms and their ability to operate internationally bring reputational advantages vis-à-vis domestically owned firms. As foreign firms operate globally, family dominance is less in their firm. They recruit professionals to look after their overseas affairs. This brings professionalism into the firm and family dominance may weaken. In some countries, foreign firms also receive several fiscal incentives from the host governments. This may translate into higher market value for the firm. Therefore, we hypothesise that firms with foreign promoters/collaborators tend to have higher market value vis-à-vis domestically owned firms. Hypothesis II: Firms with a foreign promoter/collaborator will have a higher market value than completely domestically owned firms.

III Data Base and Explanation of Variables

The sample for the study is drawn from Prowess, a database developed and maintained by the Centre for Monitoring Indian Economy (CMIE). Our initial sample consists of 1,833 listed firms in India. Most of the firms are listed on the Bombay Stock Exchange (BSE) and National Stock Exchange (NSE) of India. For each firm we have four years of observations, i e, 2000-01 to 2003-04. Out of a large panel we have eliminated those for whom

(a) sales data, (b) shareholding pattern information, and (c) selected stock indicators like share prices and market capitalisation are missing. All together we have 1,833 firms with 7,330 observations. The sample consists of both large and small firms.

Explanation of Variables

Performance measures: In the corporate governance framework, most authors rely on market based performance indicators of the firm (Tobin’s Q) to measure performance. Tobin’s Q is defined as the ratio of the market value of equity, i e, number of shares times the secondary market price plus preference capital plus the book value of its debt – to the book value of the fixed assets of the firm. This market value exhibits the discounted present value of its expected future income stream whereas the book value of equity shows the investment in assets by shareholders in the assets utilised to generate that income stream. Therefore, the Q ratio, taking into account the future prospects of the firm provides a measure of the management’s ability to generate a certain income stream from an asset base [Short and Keasey 1999]. It is argued that accounting measures such as return of assets (ROA), return of equity (ROE) and return on sales (ROS) are noisy as all firms do not follow the same accounting procedure in emerging economies.

Variables of Interest

The major variables of interest for the present study are the amount of share ownership by insiders. INS: The fraction of equity shares held by insiders/promoters. INSSQ and INSCU are the square and cube of insider share respectively. Promoter share reflects direct insider holding in the firm. In the Indian context it is the family holding in the firm. FORPRES: A categorical variable FORPRES is generated, which is equal to 1 if there is foreign promoter share within promoter share. In our sample, there are 345 firms where a foreign promoter holds a share. FORS<26 per cent: A categorical variable that is equal to 1 if foreign promoter share within promoter/insider share is less than 0.26, 0 otherwise. FORS:>26-51 per cent: A categorical variable that is equal to 1 if foreign promoter share within promoter/insider share is more than or equal to 0.26 and less than 0.51, 0 otherwise. FORS:>51 per cent: A categorical variable that is equal to 1 if foreign promoter share within promoter/insider share is more than or equal to 0.51, 0 otherwise. INS.SP: Here, Ins.sp1 to Ins.sp3 are the three spline variables of insider ownership which are discussed later in the paper.

As we have already stated, the performance of a firm is determined by a host of factors. In the ownership-performance model if we do not control for these factors, then the emerging relationship will be a spurious one. The selection of the control variable is a tedious one as there are numerous factors which affect firm performance. To make the task simple, we have relied upon two studies – Cho (1998) and HHP (1999). Besides these, we have considered several other control variables which are either India specific or significant on theoretical grounds. The constructed variables are given in Table 1.

IV Descriptive Results

Ownership Classification and Market Value of Firms

We view the ownership structure of firms as a way to organise and run economic activity. We hypothesise that the success and continuance of any form of ownership organisation depends upon its performance, i e, real or perceived. In Table 2, we present the performance of firms when the insider is the majority shareholder and otherwise. When a shareholder holds more than 50 per cent of the share in a firm, he/she has strategic control over the firm. As per the Indian Companies Act 1956; with more than

Table 1: Explanation of Variables

Variables Abbreviation/Model Name Measurement
Sales LeverageAge Operating income Capital expenditure Research and development intensity Selling intensity Capital intensity Liquidity BSE 500 LN(S) LEV LN(Age) Y/S I/K R&D/K ADV/K K/S LIQUIDITY BSE 500 Firm size is measured using natural logarithm of sales revenue for each year. Leverage is defined as the ratio of long term debt to total assets.Age is defined as the number of years between the observation year and the firm’s incorporation year.Operating income is a measure of profitability and measured as the ratio of cash profits to sales.This is the capital expenditure or new fixed assets creation in the firm. It is measured as the ratio of expenditure in purchase of new fixed assets to gross fixed assetsThis is measured as the ratio of R&D expenditure to capital. This is defined as the ratio of advertising and promotional expenditures to capital.This is defined as the ratio of gross fixed assets to sales. Turnover rate is used as a proxy for liquidity. Turnover is measured as the annualaverage number of shares traded over total number of shares A dummy variable is generated to distinguish between BSE-500 firms and others. Ifthe firm belongs to the BSE 500 list then it is coded as 1, else it is zero.
Economic and Political Weekly April 21, 2007 1461

a 50 per cent ownership stake, an investor can pass an ordinary resolution which governs among others – adoption of annual accounts, matters related to the alteration of the capital structure of the company, issues related to the appointment of auditors and their remuneration, appointment of directors and their remuneration and issues related to the voluntary winding up of a company. Thus, it is useful to examine the performance of firms when property rights are fully devolved to insiders vis-à-vis firms where ownership is widely diffused.

In our sample there are 3,722 (930 firms) observations where the insider stakes less than 51 per cent and 3,608 observations (902 firms) where the insider stake is more than or equal to 51 per cent. In Table 2 (Panel A), we show the performance of firms when insider share is less than 51 per cent and Panel B shows the performance of firms when insiders’ share is more or equal to 51 per cent. There are three kinds of performance parameters

– (a) market related or Tobin’s Q and its variants like Mktbook, Mbvr, Mtbe, Metba, MVA, (b) accounting ratios like ROA, ROE and earnings per share (EPS), and (c) agency cost measures like operating expense ratio and asset turnover. The average (median) Q-ratio is 0.69 (0.55) when insider share is less than 51 per cent whereas the respective figures are 0.73 (0.57) when shareholding increases to 51 per cent. All other market related indicators exhibit similar results. ROA is 0.03 decimal points more when the insider has 51 per cent or more stakes in the firm. Though ROE is higher when insider has less than 51 per cent stake in the firm, the difference is not statistically significant as given in the last column of Table 2. EPS and market value added (MVA) are more for higher insider ownership firms. In the literature, agency cost is directly measured by operating expense ratio and asset turnover [Ang, Cole and Lin 2000; Singh and Davidson 2003]. Operating expense will be higher and asset turnover will be low in the presence of high agency costs. When the insider is not the majority shareowner, operating expense ratio is 1.50 and asset turnover is 0.82. When the insider stake is equal to or more than 51 per cent, operating expense ratio is less, i e, 1.08 and asset turnover is more, i e, 0.97. As we can see from the last column, except ROE and MVA, the difference in the mean value of all the performance indicators between the two ownership categories is statistically significant. Therefore, as the insider ownership stake increases or property rights devolve fully to insiders, the performance of the firm also improves.

V Empirical Specifications and Results

In this section our objective is to specify an appropriate functional form for the relationship between the insider ownership structure of firms and their value, which will enable us to test our hypotheses. According to previous empirical evidence, the hypotheses examined here describe the potential non-linear effects of insider ownership structure on firm value. Our first hypothesis predicts a cubic relationship between value and ownership concentration as a result of joint consideration of incentive alignment and entrenchment effects.

To examine the ownership effect on firm value, we regress the Q-ratio, our measure of profitability, on measures of insider ownership share, as well as other variables which should affect Tobin’s Q. We have used two methods – (a) a cubic (polynomial) equation method and (b) the spline method. Each method has its own advantages and disadvantages. Adopting a cubic form of the relationship between firm value and insider ownership allows us to investigate both the convergence of interest and entrenchment effects according to the level of insider ownership. This method enables us to determine the optimal break points within the system, which is not the case in the spline specification. The primary specification is:

Performanceit = α + β (ownership)it + γXit + θt + δt+ εit

Where (ownership)it measures the fraction of the equity of firm i, lying between 0 and 1, which is owned by insiders in period t. The control variables Xit are industry specific factors. Here, industry specific fixed effects are δt’ time effects θt and a random unobserved component εit. To introduce non-linearites as is confirmed by MSV (1988), McConnell and Servaes (1990, 1995), Kole (1995) and Short and Keasey (1999), we have used the polynomial function of insider ownership such as (ownership)2it’ (ownership)3it. Therefore, we can expand our basic model to:

Performanceit = α + β1 (ownership)it + β2 (ownership)2 +

itβ3 (ownership)3it + γXit + θt + δi+ εit

Completing the regression with the notations and control variables explained earlier, the above model can be developed as follows:

Performanceit= α + β1 (INS)it + β2(INS)2it + β3 (INS)3it + γ1Ln(s)it

  • + γ2Ln(s)2 + γ3(Y/S)it + γ4(LEV)it + γ5(LEV)2it + γ6(R&D/K)it
  • + γ7(ADV/K)it + γ8(I/K)it + γ9(K/S)it + γ10Ln(age)it
  • + γ11(FORPRES)it + γ12(FORS26 – 51%)it + γ13(FORS > 51%)it
  • + γ14(BSE500)it + θt + δt+ εit
  • MSV (1988) use a piecewise linear regression model to study the non-linear relationship between board ownership and firm performance. Henceforth, it has been widely used in such kind of studies, including some work on India [Sarkar and Sarkar 2000; Chhibber and Majumdar 1998]. Here, we can describe the model

    Table 2: Ownership-Performance

    Panel-A Panel-B t-Stat of Insider Ownership Insider Ownership Difference Share<51 Per Cent Share>51 Per Cent in Mean Mean Median Mean Median

    Tobin’s Q 0.69 0.55 0.73 0.57 -2.49*
    Mktbook 0.99 0.97 1.18 1.03 -9.50*
    Mbvr 0.69 0.31 0.99 0.41 -3.77*
    Mtbe 4.62 0.71 5.87 1.40 -3.41*
    Metba 0.32 0.13 0.39 0.17 -4.32*
    ROA 0.08 0.08 0.11 0.11 -7.83*
    ROE 1.91 0.47 1.82 0.94 0.57
    Operating expense 1.50 0.89 1.08 0.87 3.35*
    Asset turnover 0.82 0.68 0.97 0.87 -8.12*
    EPS 3.88 0.53 6.24 2.02 -1.98*
    MVA 39.37 -2.87 46.63 4.00 -1.33

    NB: The last column shows statistical significance of difference in means of less than 51 per cent insider ownership and greater than 51 per cent insider ownership, * Significant at 1 per cent level. Mktbook = (book value of assets – book value of equity + market value of equity)/book value of assets, Mbvr = (market value of equity)/(book value of equity + reserves), Metba = market value of equity/book value of assets, ROE = gross profit/ total paid-up equity capital, ROA = gross profit/total assets, operating expense = (cost of sales + administration and other costs)/sales, asset turnover = net sales/average total assets, MVA = (market value of equity/ no of outstanding shares) – (book value of equity/no of outstanding shares) EPS = profit after tax/no of outstanding equity.

    Figure 2: Predicted Tobin’s Q and Insider Share moves from alignment to entrenchment and to alignment as their

    (Pooled Data Cubic Estimation)

    ownership stake in the firm increases. Even if each ownership variable taken individually is significant, it is pertinent to show the joint significance of these variables. Below each model, we

    1.2 1 .8 .6 .4Predicted Values and 95 per centCI for Tobin’s Q

    provide the P-value of F-statistics for the joint hypothesis that all the three insider ownership coefficients are jointly zero (i e, joint exclusion test). For model 1, the P-value is 0.00 which rules out the null hypothesis.

    In Figure 2; we plot the predicted Tobin’s Q and insider share. The graph shows that corporate value increases with insider ownership but begins to decline at higher levels of ownership. Finally, corporate value increases at very high levels of insider

    0 .2 .4 .6 .8 1 ownership. Our finding is consistent with convergence of interest

    Insider Ownership and entrenchment effects at increasing levels of insider ownership. When we have calculated the turning points for this baseline

    ••• pred_Q –––– lower ........ Upper

    specification, the turning points on the cubic function of insider share are at 0.29 (maxima) and 0.56 (minima).

    which has been used as follows: Performanceit = α + β1 (Ins.sp1)it + β2 (Ins.sp2)it + β3 (Ins.sp3)it

    + γ1Ln(s)it + γ2Ln(s)2 + γ3(Y/S)it + γ4(LEV)it + γ5(LEV)2

    it

  • + γ6(R&D/K)it+ γ7(ADV/K)it+ γ8(I/K)it+ γ9(K/S)it+ γ10Ln(age)it
  • + γ11(FORPRES)it + γ12(FORS26 – 51%)it + γ13(FORS > 51%)it
  • + γ14(BSE500)it + θt + δt+ εit We have generated variables containing a linear spline of insider ownership with two knots at 0.20 and 0.49. More formally, let INSi, i=1,...,n, be the variables to be created ki, i=1,...,n-1 be the corresponding knots, and INS (insider share) be the original variable. Then, INS1 = min (INS,k1) and INSi = max {min(INS,ki), ki-1}–ki-1 where i=2,...,n. In the marginal spline specification, the definitions are: INS1 = INS and INSi = max(0,INS – ki-1) where i=2,...,n. Marginal specification is preferred as when the generated variables are used in estimation, its coefficient represents the change in the slope from the preceding interval. For example, Ins.sp2 in the marginal specification shows the change in slope from after insider ownership 0.20 to before insider ownership 0.49. So, it tests for the differences in slopes. In the non-marginal spline specification, Ins.sp2 would have measured the slopes for the interval, i e, 0.20 to 0.49. Another advantage of this is that it makes it possible to test whether the change in slope is significant,
  • i e, if the effect of ins.sp2 is not significant then the effect of insider ownership does not change after the break point.

    Empirical Results and Analysis

    In Table 3, we report the pooled ordinary least squares (OLS) result of insider ownership and firm performance in four different model specifications, i e, models 1-4. The first model reports the results from a baseline specification using only ownership variables in which Tobin’s Q is regressed on insider ownership (INS), insider ownership squared (INSSQ) and insider ownership cubed (INSCU). In the parenthesis, below each coefficient, we report the heteroskedasticity consistent t-statistics.

    The results from model 1 confirm our hypothesis that the relationship between the performance of firms and insider ownership is cubic in form. The point estimate of each ownership variable is statistically significant. The coefficients on INS and INS3 are positive while that on INS2 is negative. This result provides support for the general functional form of the relationship between firm value and insider ownership that is management

    In the case of US firms, McConnell and Servaes (1990) find a significant concave relationship between managerial ownership and firm value. In their sample for 1976, the maximum is reached when insider ownership is 49.4 per cent and it is 37.6 per cent for 1986. To note here, McConnell and Servaes have considered a quadratic equation model which can only test for bell shapes that cover the entire ownership interval from 0 to 100 per cent. However, the evidence provided by MSV (1988) using a piecewise linear regression shows that the relationship is cubic. Consequently, regressing squared insider ownership will result in a model specification error, if the relationship is of a cubic form. Short and Keasey (1999) using a third degree polynomial find the maxima at 13 per cent and minima at 42 per cent in the firm value model for UK firms. Similarly, Faccio and Lasfer

    Table 3: Pooling Data OLS Regression Estimates of Tobin’s Qon Insider Ownership and Other Firm Characteristics

    Pooled Data (Dependent Variable: Tobin’s Q) Variable Model 1 Model 2 Model 3 Model 4

    INS 1.1574(3.60)* 0.7751(2.53)* 0.7227(2.36)* 0.9326(3.07)* INSSQ -3.0371(-3.81)* -1.9550(-2.56)* -1.9466(-2.55)* -2.4861(-3.30)* INSCU 2.3769(4.03)* 1.7023(2.99)* 1.6932(2.98)* 1.9950(3.55)* LN(S) --0.0604(-4.81)* -0.0635(-5.03)* -0.0631(-5.04)* LN(S)2 -0.0071(3.42)* 0.0074(3.59)* 0.0073(3.60)* Y/S -0.0005(1.55)§ 0.0005(1.70)** 0.0007(2.05)* LEV -0.6270(10.77)* 0.6402(11.22)* 0.6828(12.22)* LEVSQ -0.0988(2.84)* 0.0913(2.73)* 0.0863(2.67)* R&D/K -4.6277(5.15)* 4.5609(5.08)* 4.5621(5.15)* ADV/K -0.0623(2.30)* 0.0618(2.34)* 0.0566(2.33)* I/K -0.0381(1.76)** 0.0417(1.84)** 0.0443(1.88)** K/S --0.0003(-1.99)* -0.0004(-2.38)* -0.0004(-2.38)* LN(AGE) --0.0275(-2.10)* -0.0290(-2.21)* -0.0348(-2.66)* LIQUIDITY -15.5405(3.79)* 15.9149(3.87)* 16.3626(3.97)* FORPRES --0.1403(6.64)* -FORS:

    26-51per cent ---0.2013(5.30)* FORS:

    >51 per cent ---0..4084(7.15)* BSE 500 -0.4130(13.24)* 0.3903(12.32)* 0.3666(11.47)* R-square 0.0046 0.2109 0.2170 0.2297 F Stat: Prob > F 0.00 0.00 0.00 0.00 Prob>F:

    (insider)# 0.00 0.00 0.00 0.00 Year dummy: N0 Yes Yes Yes Obs. 7329 7305 7305 7305

    Notes: Heteroskedasticity consistent t-statistics are in parentheses. Standard errors are calculated using White’s heteroskedasticity consistent variance-covariance matrix.

    * Indicates significance at 5 per cent level, ** indicates significance at 10 per cent level, § indicates significance at 15 per cent level.

    # For models P-value of joint exclusion test of insider ownership is given, for which the null hypothesis is that all the insider-ownership variables are jointly zero.

    Figure 3: Operational Efficiency Score

    (Operating Expense Ratio and Asset Turnover)

    pgp0 2 4 6 8 10 5 4 3 2 1 0 Efficiency Score

    Firm Size (Deciles) —– Av_opexp • —— Av_assturn

    -

    (2000) document the turning points at 19.68 per cent (maxima) and 54.12 per cent (minima) in UK firms. In the case of India, Sarkar and Sarkar (2000), using a piece-wise linear model, find the firm value to decrease till 25 per cent of insider share and to increase thereafter. Deb and Chaturvedula (2004), studying 433 firms of Standard and Poor’s CNX 500 for the year 2003, document that in India firm value increases up to 30 per cent of insider ownership and between 30 and 60 per cent firm value decreases, which increases thereafter. In India, as the above mentioned study shows and our findings also confirm, managers/ insiders entrench at a very high level of ownership. This difference in findings could be attributed to institutional differences between advanced countries and India [Short and Keasey 1999; Faccio and Lasfer 2000]. Several studies documented that ownership in the US is less concentrated [Denis and McConnell 2003; Roe 2005; La Porta et al 1998]. A highly diffused ownership structure enables the manager or insider to entrench at low levels of shareholding as non-insider shareholdings are diffused [MSV 1988]. Second, as MSV (1988), Agrawal and Knoeber (1996) and Cho (1998) have collected samples from Fortune 500 list, i e, large firms, it is not surprising to see managers to entrench at low levels of insider ownership because major block holdings will be absent in such firms.

    In India, as we have already seen, ownership is highly concentrated. This governance structure allows insiders to entrench only at higher levels of ownership. When insiders have fewer stakes in the firm, there may be some other block-holders who prevent them from dissipating shareholders’ wealth. Stulz’s (1988) reasoning is also quite appealing in the context of India, although from a different perspective. When insiders have lower stakes in a firm, the likelihood of a successful takeover is quite high, thus forcing insiders to perform. However, once they have a substantial stake in the firm so that the probability of a takeover is low, they may try to consume or divert the firm’s resources at the cost of ordinary shareholders. In reality, the market for corporate control is at its nascent stage in India [Chojer 2005; Raju et al 2004].

    In model 2, we have used several control variables and year dummy. The sign and significance of ownership variables remain unaltered. Our result also indicates a convex non-linear impact of firm size (i e, sales) on firm value. Firm value (Q-ratio) first decreases and then increases in log of sales (lsales) which is consistent with the curvilinear findings of HHP (1999). Whether economies of scale or organisational efficiency (X-efficiency) is the reason for positive association of value and size for large firms require a more scientific investigation. In an earlier study [Pant and Pattanayak 2005], we find the absence of scale economies in Indian industry. Therefore, internal capital market advantages [as suggested by Short and Keasey 1999] and organisational efficiency of large firms [Williamson 1985] may be the reason for a higher Q-ratio.

    In Figure 3, we show the average operational efficiency of a firm on the basis of the operating expense ratio and asset turnover. The x-axis shows the size decile (measured by sales) of the firm and the y-axis measures the efficiency score. It is apparent from the figure that the operating expense ratio declines with an increase in the size of firms. On the other hand, as the size of the firm increases, the asset turnover ratio increases. Therefore, this suggests that large firms are operationally more efficient than small firms.

    Leverage or debt intensity is positive as the point estimate is

    0.63 with high statistical significance. To test for non-linearity, the models include a squared leverage term which is also positive and statistically significant in model 2. A high commitment of fixed debt payment helps in alleviating the excess cash flow problems. This is also consistent with the signalling argument [Ross 1977] and free cash flow theory [Jensen 1986]. Next, the coefficient of research and development (R&D) intensity is positive and statistically significant. Here, we see that higher spending in R&D leads to a higher Q-ratio which suggests that high R&D firms are innovative and profitable. The point estimate of advertising or promotional expenditure is 0.06 and statistically significant at less than 1 per cent level. Advertising expenditure captures the effect of intangible assets along with R&D expense. These variables control for an upward bias in the Q-ratio that result from the use of book value of total assets in the denominator of Q-ratio. Higher expenditure in advertising helps firms in building their reputation and also acts as an entry deterrent for new entrants in the industry.

    Current expenditure in fixed capital (I/K) is found to be positive and significant. We used the variable “age” to control for life cycle effects. As in other studies, we find younger firms to be more valuable than older firms. The final variable in model 2 is BSE 500, a categorical variable equal to 1 if the firm is one of the Bombay Stock Exchange 500 (BSE 500) firms. As expected, we find the dummy variable to be positive and highly significant. On an average, the market value of BSE 500 firms is 41 per cent more in comparison to non-BSE 500 firms.

    In model 3, we include another categorical variable, i e, foreign presence (FORPRES). If the promoter shareholding includes shares of a foreign promoter/collaborator, we code it as 1 otherwise zero. A foreign promoter/collaborator is a strategic partner for a domestic corporation when he provides superior technology know-how. The technological and organisational advantages of foreign firms and their ability to operate internationally brings reputational advantages vis-à-vis domestically owned firms. In our sample, there are 345 firms with foreign promoter/collaborator shareholding. The categorical variable shows that firms with a foreign promoter share tend to have higher market value. Operating income is positive and statistically significant. It is a measure of profitability and measured as the ratio of cash profit to sales. Consistent with our expectation, the firm value is higher; the higher is the profit rate of the firm. For all other variables, the signs and significance remain intact. When we calculate the turning points of insider share in model 3 (after including several control variables), the turning points are found at 0.32 (maxima) and 0.45 (minima).

    Chhibber and Majumdar (1999) confirm that before 1991, foreign ownership has no discernible impact on firm profitability.

    Figure 4: Predicted Tobin’s Q and Insider Share are found at 0.30 (maxima) and 0.52 (minima). Models 6, 7 and 8

    (Fixed effect estimation with control variables)

    are the replication of models 2, 3 and 4 with industry fixed effects. All the ownership variables signs and significance confirm to earlier model results. However, the effect of current expenditure in fixed assets (I/K) on firm value disappears at the conventional

    Predicted Values and 95 per cent

    0 .2 .4 .6 .8 1

    Insider Ownership pred_q lower ...... upper

    CI for Tobin’s Q

    level of significance once we control for industry effects. Age has no statistically significant impact on firm value in these

    .8

    models. Firm value is found to be monotonically increasing with leverage. It is revealing as most of the previous studies on India

    find a negative effect of debt holding on profitability. Confirming our previous models, we find the coefficient of lsales to be negative and lsales2 to be positive.

    .6

    .4

    In Figure 4, we plot the predicted Tobin’s Q and insider ownership for the model 8. Unlike in the figure, here we include all control variables and industry effects. We find the value of maxima at 0.32 of insider ownership and minima at 0.52.

    This is due to several restrictions on the amount of foreign holdings in the firm. Post-1991, they find foreign ownership influences profit rate positively only when foreign stake exceeds 51 per cent in the firm. Any foreign stake less than 51 per cent has no significant impact on profitability. To test Chhibber and Majumdar’s findings here, we have generated three categorical variable, i e, FORS<26 per cent, FORS26-51 per cent and FORS>51 per cent. The first dummy variable. i e, FORS<26 per cent is our base category. In model 4, the other two dummy variables have positive sign with high statistical significance. Unlike Chhibber and Majumdar (1999) we find even when foreign promoter stake lies between 26 per cent and 51 per cent, the firms exhibit higher Q-ratios.

    Models with Industry Fixed Effects

    In such kind of studies it is inevitable to control for industry effects as the incentive alignment and entrenchment problems differ across investment opportunity sets. Also, panel estimation allows one to control for industry specific shocks and a firm’s sensitivity to such shocks which is not possible in cross-section data. To avoid possible spurious correlation between ownership and the Q-ratio through industry effects we have used a twodigit level industry dummy based on national industrial classification (1998) which is akin to standard industrial classification (SIC). For the sake of brevity, we have not reported the intercept as each industry has its own intercept. Using firm level fixed effects is problematic as it removes all cross-section variation which is important in the present study. Firm specific dummy variables eliminate all between firm variations from the data. It only exploits the within group variation in the data as the mean demeaning procedure eliminates between variation. Since the cross-sectional variation is important in governance studies, we prefer the industry dummy rather than firm level dummy variables [Hermalin and Weisbach 1991, Zhou 2001]. The industry dummy exploits the within variation in each sector across the firms.

    The results with industry specific fixed effects are presented in four a model specification, models 5 to 8 in Table 4. Model 5 is the replication of model 1 without any control variables. As per our hypothesis, the signs of INS and INS3 are positive and sign of INS2 is negative. It confirms our previous findings that insiders move from alignment, to entrenchment to alignment as their ownership stake in the firm increases. All the ownership variables found to be jointly significant. Here, the turning points

    Models with Spline Specification

    Following MSV (1988) we adopt a piecewise linear model to exactly identify the non-monotonicity in the relation between insider stock ownership and firm performance. This model allows for the effect of holdings to change at different threshold points known as spline nodes. A piecewise linear specification and/or cubic model is better than a quadratic functional form to test for non-monotonicity in the ownership and performance relationship. However, it is difficult to say whether the piecewise specification or cubic regression model is better. Even though Morck et al used the piecewise functional form as early as 1988, authors like McConnell and Servaes (2005) and Short and Keasey (1999) adopt a quadratic or cubic functional form in their study. Therefore it is a matter of choice or convenience to choose the model in one’s study. In a spline specification, one can choose the knots

    Table 4: Fixed Effects Regression Estimates of Tobin’s Q onInsider Ownership and Other Firm Characteristics

    Dependent Variable: Tobin’s Q Industry Fixed Effects Variable Model 5 Model 6 Model 7 Model 8

    INS 1.0216(3.11)* 0.7533(2.42)* 0.7049(2.26)* 0.9452(3.05)* INSSQ -2.6574(-3.35)* -1.7712(-2.34)* -1.7868(-2.36)* -2.3905(-3.19)* INSCU 2.1500(3.70)* 1.5422(2.77)* 1.5536(2.79)* 1.8943(3.44)* LN(S) --0.0679(-5.23)* -0.0694(-5.35)* -0.0683(-5.32)* LN(S)2 -0.0086(4.05)* 0.0087(4.13)* 0.0086(4.13)* Y/S -0.0005(1.62)** 0.0005(1.69)** 0.0007(2.07)* LEV -0.7117(12.57)* 0.7206(12.94)* 0.7668(14.06)* LEVSQ -0.0817(2.56)* 0.0754(2.45)* 0.0694(2.33)* R&D/K -4.1301(4.63)* 4.1119(4.60)* 4.1453(4.68)* ADV/K -0.0608(2.30)* 0.0601(2.33)* 0.0545(2.32)* I/K -0.0213(1.23) 0.0250(1.38) 0.0266(1.43)§ K/S --0.0004(-2.52)* -0.0004(-2.78)* -0.0005(-2.77)* LN(AGE) -0.0037(0.27) 0.0026(0.19) -0.0006(-0.04) LIQUIDITY -11.3492(2.66)* 11.8262(2.77)* 12.1547(2.85)* FORPRES --0.1397(6.41)* FORS>26percent ---0.1943(5.09)* FORS>51percent ---0..4170(7.26)* BSE 500 -0.3991(12.89)* 0.3780(12.07)* 0.3550(11.26)* Adj R-square 0.0336 0.2288 0.2344 0.2470 F Stat: Prob > F 0.00 0.00 0.00 0.00 Prob>F:(insider)# 0.00 0.00 0.00 0.00 Year dummy: No Yes Yes Yes Obs 7329 7305 7305 7305

    Notes: Heteroskedasticity consistent t-statistics are in parentheses. Standarderrors are calculated using White’s heteroskedasticity consistentvariance-covariance matrix.

    * Indicates significance at 5 per cent level, ** indicates significance at10 per cent level, § indicates significance at 15 per cent level.

    # For models, P-value of joint exclusion test of insider ownership isgiven, for which the null hypothesis is that all the insider-ownershipvariables are jointly zero.

    as per certain practical reasons and it is amenable to comparison with previous studies. In this study we choose the knots at [00.20], [0.20-0.49] and above 0.49. As MSV (1988) state, the chosen nodes are arbitrary. However, here we have tried to link with the Indian property rights regime as per the companies act, 1956. Also, the chosen knots, to some extent, are being influenced by our cubic model results. The three model specifications are given in models 9-11 in Table 5.

    The first model in Table 5 is considered for a comparison purpose with the next two models. When shareholding level is less than 26 per cent, one cannot block special resolutions. When shareholdings go beyond 26 per cent, one can stop the passing of a special resolution. A special resolution requires the support of three-fourths majority of shareholders present and entitled to voting as per section 189 of Indian Companies Act, 1956. When the shareholding level exceeds 50 per cent, one can pass ordinary resolutions which govern most of the activities of the firm. Thus varying degrees of control power are associated with each level of ownership stake, thus we expect its effect to be reflected in firm performance. Accordingly we have generated three spline nodes at [0-0.26], [0.26-0.51] and above in model 9. As we can see in the model, the sign of each ownership variable is as per our hypothesis, i e, Ins sp1 and Ins sp3 are positive and Ins sp2 is negative. According to our hypothesis, insiders perform at very low and high levels of shareholding but shirk at the medium level of holding. However, none of the ownership variables is significant at the conventional level of significance.

    In models 10 and 11, the spline nodes are fixed at [0-0.20], [0.20-0.49] and above. We have selected the cut-off point after extensive search at several points, including the turning points that emerged in our cubic estimation. Like MSV (1988) we rely on the model where we find the root mean square error to be minimum. In our cubic model, we find the values of maxima to vary from 0.28 to 0.32 and the values of minima to vary from

    0.45 to 0.54. But the spline specification best fits the aforementioned range. Here 0.20 is much below 0.26 but 0.49 is only .02 point less than 0.51 where one gets simple majority. However, once insiders have a 20 per cent share or more, they have enough property rights and they can collude with some small stakeholders to block special resolutions when they want so.

    Our baseline specification, which does not include any control variables is model 10. We report heteroskedasticity consistent t-statistics below each estimated coefficients. As per the hypothesis of the study, we see the firm value to increase when insider stake is less than 20 per cent and beyond 49 per cent but it declines in the range of 20 to 49 percent. In model 11, all of the explanatory variables are included. The sign and significance of each ownership coefficient is consistent with our previous findings. For each 1 per cent increase in ownership between 0 and 20 per cent, firm value increases by an average 0.005 and for each 1percent increase in ownership from 20 per cent to 49 per cent, firm value declines by 0.007 points. The coefficient of Ins sp2 shows the change in slope from the preceding interval, i e, 0-0.20. It is

    0.677 and statistically significant. As insider stake increases beyond 49 per cent, we find an increase in the Q-ratio but at a slower rate of 0.003 for each 1 per cent increase in ownership. Therefore, when ownership goes above 49 per cent, the differential impact or the marginal change in slope is positive and significant. The joint exclusion test of insider ownership variables confirms that all three insider ownership variables taken together are significant.

    VI Summary and Conclusion

    This study confirms that with an increase in insider ownership stake, firm value initially increases. At the initial level of ownership, either insiders do not have a substantial stake to entrench themselves or they have incentives to perform more to acquire more ownership stakes. The market discipline may force the insiders to pursue value maximisation, despite their lack of personal incentives to do so at this low level of stake in the firm. Also, it is possible that they want to show their performance; otherwise they may be targeted by some other management group for takeover. But, when their stake exceeds 20 per cent they have a reasonably high stake in the firm. They play a crucial role in the decision-making process of the firm. They have the incentive to consume at office or divert firm resources to the entity where they have exclusive ownership right. With a substantial fraction of firm shares, which confers enough voting power, they may satisfy their value non-maximising objectives without endangering their position in the firm.

    It is very hard to note the exact source of entrenchment as earlier studies ascribed it to a number of factors: status as a founder, increasing voting power, increased tenure with attachment to the firm, lower employment of professional managers, and dominance of insider over outside directors in the board. As and when their ownership stake goes beyond 49 per cent, there is a convergence of interest with the firm. They have very high interest in the firm as they have to bear maximum loss for each dollar loss. Also, as per the monitoring hypothesis, with greater ownership stake insiders keep an eye on other constituents of the firm and the firm gets rid of the free rider problem associated

    Table 5: Piecewise Linear Fixed Effects Regression ofTobin’s Q on Insider Ownership and Other Firm Characteristics

    Industry Fixed Effects (Dependent Variable: Tobin’s Q) Variable Model 9$ Model 10 Model 11

    Ins.sp1 0.1618(0.91) 0.5892(2.31)* 0.5460(2.37)* Ins.sp2 -0.1404(-0.61) -0.6775(-2.04)* -0.6140(-2.11)* Ins.sp3 0.2738(1.63)§ 0.4386(2.38)* 0.3689(2.21)* LN(S) -0.0705(-5.45)* --0.0697(-5.44)* LN(S)2 0.0087(4.16)* -0.0086(4.15)* Y/S 0.0007(2.13)* -0.0007(2.11)* LEV 0.7780(14.40)* -0.7764(14.38)* LEVSQ 0.0674(2.28)* -0.0678(2.29)* R&D/K 4.1316(4.67)* -4.1211(4.66)* ADV/K 0.0554(2.35)* -0.0553(2.34)* I/K 0.0261(1.40) -0.0258(1.39) K/S -0.0005(-2.79)* --0.0005(-2.79)* LN(AGE) 0.0010(0.07) -0.0018(0.13) LIQUIDITY 12.2702(2.88)* -12.2083(2.86)* FORPRES ---FORS >26 per cent 0.1964(5.13)* -0.1973(5.16)* FORS >51 per cent 0.4109(7.11)* -0.4111(7.10)* BSE 500 0.3565(11.26)* -0.3567(11.26)* Adj R-square 0.2455 0.0389 0.2458 F Stat: Prob > F 0.00 0.00 0.00 Prob>F: (insider)# 0.00 0.00 0.00 Year dummy: Yes Yes Yes Obs 7305 7329 7305

    Notes: Heteroskedasticity consistent t-statistics are in parentheses. Standard errors are calculated using White’s heteroskedasticity consistent variance-covariance matrix.

    * Indicates significance at 5 per cent level, ** indicates significance at 10 per cent level, § indicates significance at 15 per cent level.

    # For models, P-value of joint exclusion test of insider ownership is given, for which the null hypothesis is that all the insider-ownership variables are jointly zero.

    $ Model 9 is adopted following India’s corporate law or legal regime where certain control rights devolve at (0-0.26), (0.26-0.51) and above.

    with dispersed ownership. Unlike other Asian countries, concentrated ownership has not destroyed firm value. The major governance problem in India is not the typical agency problem between managers and shareholders. It is the conflict between dominant insiders and atomistic outside investors. Our study proposes that higher insider ownership is beneficial to the firm as well as to the economy as it maximizes the shareholder wealth. Therefore, the policymakers must maintain utmost care while formulating policy which may dilute the ownership structure in the firm. From the results it appears that collaboration of domestic firms with foreign firms is value augmenting. The foreign ownership stake in domestic firms bring sophisticated technology and advanced management techniques. It helps in improving the firm performance and positively affects firm value.

    [IX

    Email: manojp@mail.jnu.ac.in mpattanayak@gmail.com

    References

    Agrawal, Anup and Charles R Knoeber (1996): ‘Firm Performance andMechanisms to Control Agency Problems between Managers andShareholders’, The Journal of Finance and Quantitative Analysis, 31:3, pp 377-97.

    Ang, James S, Rebel A Cole and James Wuh Lin (2000): ‘Agency Costsand Ownership Structure’, The Journal of Finance, LV:1, pp 81-105.Bajaj, Rahul (2005): ‘How Independent Can a Director Be?’ Business Standard, July, 29.Berle, Adolf A and Gardiner C Means (1932): The Modern Corporation andPrivate Property, MacMillan, New York.

    Chhibber, Pradeep and Sumit K Majumdar (1998): ‘State as Investor andState as Owner: Consequences for Firm Performance in India’, Economic Development and Cultural Change, 46, pp 561-80.

    – (1999): ‘Foreign Ownership and Profitability: Property Rights, Control,and the Performance of Firms in Indian Industry’, Journal of Law and Economics, XLII, pp 209-305.

    Cho, Myeong Hyeon (1998): ‘Ownership Structure, Investment, and theCorporate Value: An Empirical Analysis’, Journal of Financial Economics, 47, pp 103-21.

    Chojer, Ruchi (2005): ‘The Market for Corporate Control: A Survey’, SEBI Bulletin, 3, December:12, pp 5-16.

    Claessens, Stijn, Simeon Djankov and Larry H P Lang (2000): ‘The Separtaionof Ownership and Control in East Asian Corporations’, Journal of Financial Economics, 58, pp 81-112.

    Claessens, Stijn, Simeon Djankov, Joseph P H Fan and Larry H P Lang (2002):‘Disentangling the Incentive and Entrenchment Effects of LargeShareholdings’, The Journal of Finance, LVII:6, pp 2741-71.

    Deb, Saikat Sovan and Chakrapani Chaturvedula (2004): ‘Ownership Structureand Firm Value: Empirical Study on Corporate Governance System ofIndian Firms’, SSRN abstract_id=594221.

    Demsetz, Harold (1983): ‘The Structure of Ownership and the Theory ofthe Firm’, The Journal of Law and Economics, XXVI, June, pp 375-93.Demsetz, Harold and Belen Villalonga (2001): ‘Ownership Structure andCorporate Performance’, Journal of Corporate Finance, 7, pp 209-33.

    Demsetz, Harold and Kenneth Lehn (1985: ‘The Structure of CorporateOwnership: Causes and Consequences’, Journal of Political Economy, 93:6, pp 1155-77.

    Denis, Diane K and John J McConnell (2003): ‘International CorporateGovernance’, European Corporate Governance Institute working paperNo 05/2003.

    Faccio, Mara and M Ameziane Lasfer (2000): ‘Managerial Ownership, BoardStructure and Firm Value: The UK Evidence’, http://ssrn.com/abstract=179008.

    Hermalin, Benjamin E and Michael S Weisbach (1991): ‘The Effects of BoardComposition and Direct Incentives on Firm Performance’, Financial Management, 20, pp 101-12.

    Himmelberg, Charles P, R Glenn Hubbard and Darius Palia (1999):‘Understanding the Determinants of Managerial Ownership and the LinkBetween Ownership and Performance’, Journal of Financial Economics, 53, pp 353-84.

    Holderness, Clifford G (2003): ‘A Survey of Block Holders and CorporateControl’, FRBNY Economic Policy Review, pp 51-64.

    – (2005): ‘A Contrarian View of Ownership Concentration in the United States

    and Around the World’, SSRN abstract id 686175. Holderness, Clifford G, Randall S Kroszner and Dennis P Sheehan (1999): ‘Were the Good Old Days That Good? Changes in Managerial StockOwnership Since the Great Depression’, Journal of Finance, LIV:1, pp 435-69.

    Jensen, M C and W Meckling (1976): ‘Theory of the Firm: ManagerialBehaviour, Agency Costs and Capital Structure’, Journal of Financial Economics, 3, pp 305-60.

    Jensen, Michael C (1986): ‘Agency Costs of Free Cash Flow, Corporate Finance,and Takeovers’, American Economic Review, 76, May, pp 323-29.

    Khanna, Tarun and Krishna Palepu (1999): ‘Emerging Market BusinessGroups, Foreign Investors and Corporate Governance’, NBER WorkingPaper Series No 6955, pp 1-33.

  • (2000): ‘Is Group Affiliation Profitable in Emerging Markets? An analysisof Diversified Indian Business Groups’, The Journal of Finance, 55:2, pp 867-91.
  • (2004): ‘The Evoluton of Concentrated Ownership in India: Broad Patternsand a History of the Indian Software Industry’, NBER Working PaperSeries No 10613.
  • Kole, Stacey R 1995): ‘Measuring Managerial Equity Ownership: A Comparisonof Sources of Data’, Journal of Corporate Finance, 1, pp 413-35.

    LaPorta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert WVishny (1998): ‘Law and Finance’, Journal of Political Economy, 106, pp 1112-55.

    Lemmon, Michaell L and Karl V Lins (2003): ‘Ownership Structure, CorporateGovernance, and Firm Value: Evidence from the East Asian Financial Crisis’, The Journal of Finance, 58:4, pp 1445-68.

    Loderer, Claudio and Kenneth Martin (1997): ‘Executive Stock Ownershipand Performance: Tracking faint Traces’,Journal of Financial Economics, 45, pp 223-55.

    Majumdar, Sumit K (1997): ‘The Impact of Size and Age on Firm-LevelPerformance: Some Evidence from India’, Review of Industrial Organisation, 12, pp 231-41.

    Majumdar, Sumit K and Pradeep Chhibber (1999): ‘Capital Structure andPerformance: Evidence from a Transition Economy on an Aspect ofCorporate Governance’, Public Choice, 98, pp 287-305.

    McConnell, John J and Henri Servaes (1990): ‘Additional Evidence on EquityOwnership and Corporate Value’, Journal of Financial Economics, 27, pp 595-612.

    – (1995): ‘Equity Ownership and Two Faces of Debt’, Journal of Financial Economics, 39, pp 131-57.

    McConnell, John J, Henri Servaes and Karl V Lins (2005): ‘Changes in EquityOwnership and Changes in the Market Value of the Firm’, London Business School Working Paper, pp 1-43.

    Morck, Randall, Andrei Shleifer and Robert W Vishny (1988): ‘ManagementOwnership and Market Valuation: An Empirical Analysis’, Journal of Financial Economics, 20, pp 293-316.

    Pant, Manoj and Manoranjan Pattanayak (2005): ‘Does Openness PromoteCompetition: A Case Study of Indian Manufacturing’, Economic and Political Weekly, XL: 39, Septmber 24-30, pp 4226-31.

    Patibandla, Murali (2006): ‘Equity Pattern, Corporate Governance andPerformance: A Study of India’s Corporate Sector’, Journal of Economic Behaviour and Organisation, 59:1, pp 29-44.

    Raju, M T and L S Deepthi (2004): ‘Market for Corporate Control and Takeover Regulations: Trends and Analysis’, SEBI working paper, 10, pp 1-42.

    Roe, M J (2005): ’The Inevitable Instability of American CorporateGovernance’, The Corporate Governance Law Review, 1, pp 1-19.

    Ross, S A (1977): ‘The Determination of Financial Structure: The IncentiveSignalling Approach’, Bell Journal of Economics, 8, pp 23-40.

    Sarkar, Jayati and Subrata Sarkar (1999): ‘The Governance of IndianCorporates’ in Kirit S Parikh (ed), India Development Report, Oxford University Press, Delhi, pp 201-18.

  • (2000): ‘Large Shareholder Activism in Corporate Governance in DevelopingCountries: Evidence from India’, International Review of Finance, 1:3, pp 161-94.
  • (2005): ‘Debt and Corporate Governance in Emerging Economies: EvidneceFrom India’, IGIDR working paper No. 2005-007.Shleifer, Andrei and Robert W Vishny (1997): ‘A Survey of CorporateCovernance’, Journal of Finance, LII:2, pp 737-83.
  • Short, Helen and Kevin Keasey (1999): ‘Managerial Ownership and thePerformance of Firms: Evidence from the UK’, Journal of CorporateFinance, 5, pp 79-101.

    Singh, Manohar and Wallace N Davidson (2003): ‘Agency Costs, OwnershipStructure and Corporate Governance’, Journal of Banking and Finance, 27, pp 793-816.

    Stulz, Rene (1988): ‘Managerial Control of Voting Rights: Financing Policiesand the Market for Corporate Control’, Journal of Financial Economics, 20, pp 25-54.

    Williamson, Oliver E (1985): The Economic Institutions of Capitalism: Firms,Markets, Relational Contracting, The Free Press, London.

    Zhou, Xianming (2001): ‘Understanding the Determinants of ManagerialOwnership and the Link between Ownership and Performance: Comment’,Journal of Financial Economics, 62, pp 559-71.

    To read the full text Login

    Get instant access

    New 3 Month Subscription
    to Digital Archives at

    ₹826for India

    $50for overseas users

    Comments

    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top