Capital Structure and Corporate Governance

There has been reasonable consensus among practitioners and academicians about the importance of good corporate governance in the economy. Corporate governance has received much attention in the emerging market economies, like India in later half of the nineties. In a recent study for India, find that corporate governance significantly influences the firm performance. Corporate ownership structure can act as an incentive device for reducing the agency costs associated with the separation of ownership and management can be used to protect property rights of the firm.


Introduction
There has been reasonable consensus among practitioners and academicians about the importance of good corporate governance in the economy. Corporate governance has received much attention in the emerging market economies, like India in later half of the nineties. In a recent study for India, find that corporate governance significantly influences the firm performance. Corporate ownership structure can act as an incentive device for reducing the agency costs associated with the separation of ownership and management can be used to protect property rights of the firm.
A large body of literature does confirm the evidence that corporate governance, particularly the role of ownership structure, is crucial in determining the incentive of insiders to expropriate minority shareholder. The impact of corporate governance on the firm value has been extensively studied in recent years. The literature has highlighted the role of ownership structure that has the impact on the firm value. Most of the literature on corporate governance is concentrated in explaining the firm performance and its determinants. Yet, little is known as to how the corporate governance influences firm's financing policies (capital structure).
The paper aims to bridge research gap by providing a direct empirical test of the hypothesis. We hypothesize that the firms with poor corporate governance mechanisms tend to have higher level of debt than equity in their portfolio and vise-versa. We are particularly interested in the role of firm's ownership structure with connection to its capital structure. Our main research objective is to test whether there are links between the capital structure and corporate governance. If so, does debt constrain or facilitate entrenchment? The study of the relation between capital structure and corporate governance is advantageous, not only to enrich our understanding about whether or not firms that are vulnerable to expropriation issue more debt to have more resources to use for private interests but also which ownership groups viz. foreign, corporate, director, institutional have positive or negative impact on the debt equity ratio of a firm. This paper also sheds lights on the other possible agency issues in determining the firm's financing decisions. These agency problems may arise between the firm's controlling shareholders and the debt providers and between the debt suppliers and their minority shareholders. For example, the controlling shareholder of a firm and the firm's debt providers on the relationship of corporate governance mechanism and capital structure of corporate firms in India. To our best knowledge there is no study in this context for India.
Since the pioneer work of Modigliani and Miller [3] proposed the concept, that the general characteristics of a firm's ownership structure can affect performance has received considerable attention but few studies have looked at the relationship between ownership structure and capital structure. Corporate debt policy has also been viewed as an internal control mechanism, which can reduce agency conflicts between management and shareholders, particularly the agency costs of free cash flow as suggested by [4]. Jensen [4] argues that managers with substantial amounts of free cash flow are likely to engage in non-optimal activities. Grossman and Hart [5] suggest that debt is a disciplinary device that may be used to reduce the agency costs of free cash flow. However, as Myres [6] demonstrates, debt can also have undesirable effects such as inducing managers to forego positive net present value projects. Jensen and Meckling [7] argue that managerial shareholding can reduce managerial incentives to consume perquisites, expropriate shareholder's wealth and to engage in other non-maximizing behavior and thereby helps in aligning the interests between management and shareholders. This paper examines the link between capital structure and shareholding pattern for a panel of more than 2000 publicly traded Indian corporate firms over the years 1994 to 2000. We develop our regression framework based on the capital structure theory, suggested by corporate finance models. We include the factors that may affect the firm's capital structure into our empirical specifications. These factors are age, size, tangibility, marketing, advertising, distribution, Rand D expenses, and profitability. We also include industry dummies to control the industry effect on firm's leveraging. The industry classification is defined based on National Sample Survey Organization's National Industrial Classification, 1998.
We have contributed in four ways to the existing literature. First, we employ an econometric framework that specifically controls for firm specific unobserved heterogeneity and aggregate macroeconomic shocks. Second, our econometric methodology allows us to control for the unobserved firm heterogeneity caused by the ownership structure and other observed variables. This approach also provides evidence in favor of the fixed effect approach. Thirdly, it uses exact shareholding by different groups of owners, controlling for change in firm value due to small change in shareholding pattern (not exactly changing the dominance of a group), as in most of the cases shareholders dominance does not change dramatically. Finally, this paper is the very first study in case of India, which investigates the relationship between the ownership and capital structure.
Unlike the specifications of Faccio, we do not include the marketbased variables, which are calculated based on the stock prices such as volatility and Tobin's Q because we are strongly convinced by the argument suggested by Joh, to exclude the market-based measures when the stock market appears to be less efficient. We consider that stock markets in India are in line with this proposition. For this analysis of Indian data, accounting measures of performance are likely to be better measures of performance than share market based measures for at least these three reasons. First, researchers have shown existence of some market inefficiencies even in the developed countries. This suggests that the stock prices in India are not likely to reflect all available information, peculiarly during the period of study. Second, a firm's accounting profitability s more directly reflected to its financial survivability than its market value, and many studies have used accounting measures to predict bankruptcy, or financial distress. Thirdly, most of the stocks do not trade regularly, which may result in inappropriate pricing of shares. Some stock market scams, also happened during the period of study, which makes us believe that stock prices during the period of study were prone to price manipulations.
Though the accounting measures may not take into account the future prospects of firm endurance but they do take into account the current scenario of financial strength. The share market measures of firm performance such as Tobin's q may run into severe problems in the context of emerging market economy specially India, as most of the firms, go for debt-financing in these economies rather than using finance from the share market. As a result, share market measures may not reflect the actual profits made by the investors on their investments. Moreover, stock price information is not readily available for the period of study for all the firms. Most of the stocks trade irregularly on the exchange and have very low levels of liquidity. We believe that such traded price may not provide actual information about the firm value for the thinly traded stocks. Moreover, share prices may not reflect true value of firms because it is driven by many factors, which may not be efficient, for example: noise trading, portfolio insurance, high transaction costs, and other factors unrelated to firm performance may induce randomness in stock prices. Stock market in India also faces high volatility during end of financial year due to annual central and state government budget announcements. Since market prices determine their values based on accounting information provided by the firms in their un-audited quarterly financial results, and audited annual financial results. A market measure of performance will also suffer from the drawbacks of the accounting performance measures, as well as problems of inefficient capital market. The declaration of the annual audited financial results, for the same period (April-March), does not happen at the same time. Stock market may reflect prices adjusted to the information only after the declaration, which is different for different firms. Thus, taking the last closing price for calculation of the market value of firm may not be desirable, whereas, annual audited financial results are for the same period.
The firm level panel data for our study is primarily obtained from the CMIE, the Center for Monitoring the Indian Economy. The data used in the analysis consists of all manufacturing firms listed on The Stock Exchange, Bombay (BSE), for which we could get their historical shareholding pattern for the period of study. Public Sector firms and firms within financial services are not included in the analysis. We confine our analysis to BSE listed firms only because all the listed firms are required to follow the norms set by SEBI for announcing the financial accounts. The BSE also has the second largest number of domestic quoted companies on any stock exchange in the world after NYSE, and more quoted companies than either the London or the Tokyo stock exchange.
We analyze data from 1994 to 2000 1 . We also restrict our analysis to firms that have no missing data (on sales, age, shareholding pattern, PBDIT and assets) for at least two consecutive years. 2 There are 2575 firms (5224 firm years) in our sample, for which there is data required for at least two consecutive years. 3 Our final sample consists framework for more than four decades, which forms the foundation of the corporate governance system in India. Numerous initiatives have been taken by Securities Exchange Board of India (SEBI) to enhance corporate governance practice, in fulfillment of the twin objectives: investor protection and market development, for example: streamlining of the disclosure, investor protection guidelines, book building, entry norms, listing agreement, preferential allotment disclosures and lot more.
Although the Indian Corporate Sector is a mix of government and private firms (which are again a mix of firms owned by business group families, and multi nationals and stand-alone firms), it has not suffered from the cronyism that has dominated some of the developing economies (read East-Asian economies). Accounting system in India is well established and is similar to those followed in most of the development economies.

Empirical Analysis
Himmelberg, Hubbard, and Palia [9] have argued that regression of firm performance on ownership variables is potentially missspecified because of the presence of the firm heterogeneity. Specifically, if some of the unobserved determinants of firm performance are also determinants of ownership, then ownership might spuriously appear to be a determinant of firm performance. Zhou [10] has argued that the firm-fixed effect is not necessary in terms of ownership, as the ownership structure in general does not vary over time for a specific firm. Similar arguments may be valid while analyzing the impact of corporate governance (ownership structure) on firm's capital structure. However, in Indian case, the argument made by Zhou [10] against the use of firm-level panel data analysis is not valid. Kumar provide detailed discussion on this issue and provided an explicit test to justify the inclusion of firm-fixed effects in both forms, namely, in terms of control variables as well as in terms of ownership structures. The study provides an explicit F-test for presence of fixed effect for control variables, ownership structure, separately as well as jointly. Percentage shareholding of different investors may be correlated, because, share ownership by Foreign, Institutional, Corporate and Director, along with the shares of 'other top 50 shareholders' and 'others not included above' adds up to '100' percent. In order to avoid the problem of multi collinearity, this study uses only four main shareholders, i.e. foreign, institutional, corporate, and director.
In this paper, we use firm-level fixed-effects panel data methodology. Primarily because, this model allows us to control for both year-variant but firm invariant omitted variable as well as firm variant but timeinvariant variables.
This leads us to the estimation of the following equation: Where (Ownership) it variables measures the fraction of the equity of firm i, lying between 0 and 100, that is owned by different group of owners in period t. The (Control) it variables are firm-specific factors, which may also have influence on the capital structure.
By using panel data method one is better able to control for the effects of missing or unobserved variables. Specifically, under the fixed effects model, the intercepts are allowed to be different for different cross-sections and hence the effects of the omitted variables can be captured. The effects of the omitted variables are driven by either individual time-invariant variables or period individual-invariant variables. The individual time-invariant variables are variables that of 2517 firms with 5,117 observations. We perform our analysis after restricting the proxy for capital structure (debt intensity) to lie between 1st and 99 th percentile to tackle the problem of outliers, which may be influential. This leaves us with 5017 observations for 2478 firms. The traditional aspect of the agency cost theory suggests that insider ownership aligns the interest of management and other stakeholders of the firm Jensen and Meckling [7], as managers become self-constrained and avoid rent extraction, since they too have to bear the costs of such activities in a proportion of their ownership stake in the firm. Recent studies document that the controlling shareholders have significant discretion and power to expropriate minority shareholders, as high ownership precludes takeover threats and thus decreases firm value [8]. Because, majority owners can redistribute wealth, in both efficient and inefficient ways from other minority shareholders, whose interests need not coincide. This suggests a non-linear relation between block-holders share ownership and firm leverage. In other words, the costs of the concentrated ownership may exceed its benefits. We therefore, include four ownership variables: the managerial shareholding (director), institutional investors shareholding (institutional), foreign investors shareholding (foreign), and corporate shareholding (corporate) with their squares to examine the presence of ownership effect. The squares of the ownership variables are included to distinguish the change in their effect after a certain threshold, i.e. non-linear impact of ownership structure on capital structure.
Our sample includes more than 2000 firms from India. The significant increase of our sample coverage mainly comes from the extensive manual works in overcoming the data restrictions on the ownership structure information, which often required supplementary data collected from the annual reports of the firms, such as the information on historical shareholding pattern, business groups or families and their relatives. We provide the evidence by using firmlevel panel data that allow us not only to econometrically control for individual firm heterogeneity but also to give more data that are informative, more degrees of freedom, and more efficiency.

Data and Summary Statistics
For our study of effects of ownership structure (shareholding pattern) on capital structure, in emerging economy, we focus our attention on Indian corporate sector. We choose this as an experimental setting as Indian corporate sector offers several distinct advantages over other emerging market economies.
The Indian Corporate Sector has large number of corporate firms, lending itself to large sample statistical analysis. It is large by emerging market standards and the contribution of the industrial and manufacturing sectors (value added) is close to that in several developed economies. Unlike several other emerging markets, firms in India, typically maintain their shareholding pattern over the period of study, making it possible to identify the ownership affiliation of each sample firm with clarity. It is largely a hybrid of the outsider systems 4 and the insider systems 5 of corporate governance. The legal framework for all corporate activities including governance and administration of companies, disclosures, shareholders rights, has been in place since the enactment of the Companies Act in 1956 and has been fairly stable during the period of study. The listing agreements of stock exchanges have also been prescribing on-going conditions and continuous obligations for companies. 6 India has had a well-established regulatory are the same for given cross-sectional units over time but vary across cross-sectional units (intangible assets, managerial skill). The period individual-invariant variables are variables that are same for all cross-sectional units at a given time but vary over the time (macroeconomic scenario). All these omitted variables may correlate with the independent variable.
Econometric technique employed in the analysis overcomes the possible heterogeneity and omitted variable problems, which often arise with cross-section analysis. In addition, the various measures of the dependent variables and the independent control variables that we use for our robustness checks can significantly mitigate the possible measurement errors. The hypothesized relationship between firm size and leverage is mixed. On the one hand, the larger firms usually have a higher debt ratio because it is usually easier for large firms to borrow from the banks or to raise debt in the capital markets. Further, larger firms can diversify their operations; therefore, the default risk might decrease which results in high debt ratio. On the other hand, information asymmetry is likely less severe for larger firms than for smaller firms. The outside investors might find it easier to get more information about the firms. This allows larger firm to raise equity directly from the capital markets, allowing large firms to have lower leverage.
The debt financing is still the prevalent method in the emerging market economies, where the financial system operates mainly under the bank-based economies. In the world of asymmetric information, the firm's tangible fixed assets can be often served as the collateral to lower the risk of the lenders who suffer from the agency cost of debt. Firms who have greater proportion of fixed assets tend to have higher debt ratio. We incorporate year dummy to control for unobserved macroeconomic effects. Detailed discussion of the variable construction is provided in the Appendix 1 and 2. Unless otherwise stated, we use debt intensity as our proxy for the capital structure in the regression analysis.

Descriptive Statistics
We present a detailed structure at the 2-digit level industrial classification of our data in Table 1, which clearly reflects the unbalanced nature of the panel. Table 1 also depicts that most of the firms included in our sample belongs to NIC-1, NIC-2 or in the NIC-3 according to 1-digit industrial classification. Based on the industrial classification of National Sample Survey Organization (NSSO), India's National Industrial Classification 1998. The mean level of debt intensity is 0.2409 with a maximum of 4.0632, whereas minimum level of debt equity is -1361.67 with median at 0.82. Total borrowing varies from 0 to Rs. 11520.24 Crore with a standard deviation of 395.95 and kurtosis 284.109. The mean level of PBDIT is 28.9 Crore whereas maximum is 4788.44 Crore and a minimum of -127.94 Crore, standard deviation of 123.62 and kurtosis of 525.67795. This once again reinforces wide variation that exists in our sample. Table 3 reports the results of the cross-sectional regression analysis with 1-digit industry dummies, for each year of the sample. To the best of our knowledge, no other study has used a panel data framework to analyze the impact of corporate governance on capital structure. We find that results vary across years in case of ownership variable's impact on debt intensity. Foreign ownership has non-linear impact on firm performance in 1994, and in 1996. The institutional investors' share has positive linear effect and negative effect in squares in 1996, and in 1997, 1998 square term becomes insignificant. Group firms are found to have significantly higher debt level in 1998, 1999, and in 2000. However, we note that Tangibility and LnSale have significantly positive impact for all the years. We also find that industry dummies are significant at 1% level for all the years. In sum, our cross-sectional results indicate that none of the ownership variables has consistently significant effect over the years.

Regression Results
We report results of pooled OLS with one digit industry dummy in Table 4 (column 1). In pooled regression without any time dummy, we find that 'foreign', 'institutional' play significant role in the firm's capital structure choices and their impact is non-linear, positive in levels, and negative in squares. Square of corporate ownership have positive and significant impact of debt intensity ratio. Column 2 of Table 4 reports the results with two-digit industry dummy (NIC-2 digit). The results in terms of impact of ownership variables are almost same as in Column 1. We repeat the same exercise with time dummies. We report the results in Columns 3 and 4 of Table 4, for NIC-1 digit and NIC-2 digit, respectively. Once, again results are qualitatively same. We also document the evidence that industry and time dummies are significant, separately and jointly. From the results of pooled OLS, we find that there is significant impact of ownership structure on capital structure of the firm. We now proceed with the fixed-effects panel-data model. We report the results of our regression analysis in Table 5, this analysis we use ROA as a measure of firm performance. Column 1 of Table 5 reports the result of the fixed-effect analysis for the full sample. Institutional ownership and square of foreign ownership have significant negative impact on the debt intensity of the firm. Square of institutional ownership have positive (significant at 12%) impact of debt intensity. In Column 2 of Table 5, we report the findings of the regression after restricting the sample to lie between 1% and 99% of debt intensity to take care of the outlier's effect. Column 3 reports the result after restricting the sample for 10% and 90%. Column 4 of the Table 5 reports the results after restricting the sample for only those firms for which debt intensity is positive. Our results remain same qualitatively; however, institutional ownership loses its significant in some cases. Square of foreign is found to have significantly positive impact on debt intensity, consistently. ROA has negative and significant impact on firm debt. This finding is in lines with the existing literature suggesting that the firms with high performance tend to have lower level of debt in their portfolio. Age has non-linear impact on firm debt; positive (insignificant) and negative in square (significant), suggesting that the younger firms rely on debt more than the equity, this trend reverses once they become older. This result is plausible as the older firms have the history of performance and they are known in the market, therefore they may have lower cost of capital if raised in form of equity than debt (intangible assets). We also find "Tangibility" to have positive and highly significant impact on debt level of a firm. In    Table 5, we present the findings of the regression analysis when we have introduced the interaction between the group dummy with ownership structure of the firms. We find that group firms with higher foreign ownership, institutional ownership tend to have lower debt level. However, this negative impact of group dummy on debt is found for all the ownership categories though insignificant.
For a robustness test of our findings, we re-run the above models (Table 5) with PBDIT as a measure of performance rather than ROA and present the findings in Table 6. Column 1 of Table 6 reports the result of the fixed-effect analysis for the full sample. Column 2 of Table  6 we report the findings of the regression after restricting the sample to be between 1% and 99% of debt intensity to take care of the outlier effect. Column 3 reports the result after restricting the sample for 10% and 90%. Column 4 of the Table 6 reports the results after restricting the sample for only those firms for which debt intensity is positive. In Column 5 of Table 6, we present the findings of the regression analysis with the interaction term (interaction of the group dummy with ownership structure of the firms). Our findings remain almost similar (qualitatively) to the findings from Table 5. This once again reinforces our findings with regard to the impact of corporate governance practices on the debt intensity of a firm.
To check whether ownership's collinearity has anything to do with the obtained results, we use each ownership group separately. In Table 7, we present our findings when we use only ownership variables as explanatory variable in the model of capital structure. One may argue that since the ownership variables may be correlated with each other, the previous results may be problematic because of collinearity. However, when we use one variable in the regression analysis, we may not be able to capture the impact of bilateral relationship between two or more group of owners, and hence may not get the clear picture. We report the results for each group of owners separately as well as jointly. Column 1 of Table 7 presents the results when we use only foreign ownership as explanatory variables Column 2 for directors, Column 3 for Institutional, Column 4 for corporate investors. It is clear from the table that in such case only, square of foreign has negative and significant impact on debt intensity. Results remain similar even when we use all the ownership variables for the full sample (Column 5), for the sample restricted between 1% and 99% based on the debt intensity (Column 6), and for sample restricted between 10% and 90% (Column 7). However, one may note that in the Column 7, we also find the   p-values are reported in parentheses.  institutional has negative and significant impact on debt intensity. The results suggest that even if we do not take account of other variables, which may have influence on the debt holding (capital structure) of a firm, ownership structure do play a significant role [11][12][13][14][15][16][17].

Conclusions
The previous research in agency theory does confirm that the corporate governance, in particular the role of ownership structure, can affect firm performance by mitigating agency conflicts between managers and shareholders. This study extends the agency framework and tests the hypotheses, which concern the relationship between ownership structure and capital structure. Using the firm-level panel data of 2251 listed firms from India; we find that the firm-level corporate governance has nonlinear relationship with the firm's capital structure. The results provide evidence that the distribution of equity ownership among directors and external shareholders has a significant relationship with debt equity ratio. This provides support for the active monitoring hypothesis, which proposes that external block holders have greater incentives and an ability to monitor management. The results also indicate a curvilinear relationship between level of insider's ownership and debt equity relationship.
The higher debt ratio of the weaker corporate governance suggests that debt can facilitate expropriation in the economies where the institutions appear to be ineffective. Our empirical results shed new lights on the importance of these ownership structure and group specific factors, but how these factors affect the firm's debt structure remains for future studies. Still, little knowledge is available as to the mechanism of entrenchment that leads to the firm's financing choice. The pyramid or cross-holding structures can be partially used to explain the phenomenon, because the direct ownership structure is still common for Indian firms. Further, clinical analyses in the form of case studies might need to be carried out to further explore this issue.
The results have considerable implication regarding the capital structure debate. By arguing for a link between the ownership structure and capital structure and through empirical support, this paper adds to an understanding of variation in capital structure. Moreover, the analysis of corporate governance in the financial institutions and its impacts on the firms will be very helpful, in particular for regulators to propose concrete measures for improving the financial system.