Impact of Dividend Payment on Firms’ Leverage: Moderating Role of Firm Size

Abstract

This article investigates the impact of firms’ dividend payment decision on its capital structure. It also examines whether firm size moderates the relationship between dividend decision and capital structure. It relies on a large dataset of 4,116 listed firms over the period 2002–2019 in the Indian scenario comprising 37.21% firm-year observations as dividend payers and 62.79% firm-year observations as dividend non-payers. Fixed effect regression is used for analysis based on the results of the Hausman test for model specification. The study observes negative impact of dividend payment on firms’ leverage ratios. Ability of dividend payment to mitigate the free cash flow agency conflict together with the availability of substantial volume of retained earnings to meet the financing requirements internally are found to be possible explanations for lesser demand of debts in case of dividend paying firms. The study also finds heterogeneous impact of dividend payment on leverage ratios for small and large firms, thus confirming the moderating role of firm size. The higher level of free cash flow for dividend payers compared to non-payers in case of small firms is found to be the only possible explanation for such results. Financial managers may refer to the findings of the study as a guide while deciding their financing choices.

Keywords

Capital Structure Dividend Policy Firm Size Leverage Ratio Pecking Order Theory Trade-off Theory

Introduction

The Chief Financial Officers in the corporate world often engage in taking principal financial decisions and capital structure is one of them. It denotes the proportion of different sources of finance in the total capital of firms. Deciding the optimum debt-equity mix in total capital is critical, as it influences the cost of capital and the overall value of firms. The trade-off theory argues that firms’ financing decisions are guided by their choice of optimum capital structure. The use of debt as a source of finance provides some important benefits to firms such as—(a) interest on debt is tax-deductible, and therefore the inclusion of more debt reduces the tax liability of firms (Modigliani & Miller, 1958), (b) debt reduces the owner-manager agency conflicts by making managers disciplined (Jensen & Meckling, 1976) and (c) the payment of interest on debt is mandatory as against the payment of dividend on equity capital, the former lessens the free cash flow available to them, and hence the agency costs associated with the same (Jensen, 1986). On the contrary, increasing debt in the capital structure has certain drawbacks which include—(a) increasing debt beyond a certain limit increases the financial distress and insolvency risk for firms (Karus & Litzenberger, 1973), (b) debt causes shareholder-debtholder agency conflicts where the former acts for their best at the costs of the latter (Jensen & Meckling, 1976) and (c) increasing proportion of debt leads to debt overhang and consequently causes loss of future investment opportunities with positive net present values (Myers, 1977). Therefore, a firm’s ideal capital structure is a point where the benefits and costs of using an additional unit of debt trade-off.

An ample number of studies have been carried out exploring the factors that determine firms’ optimal debt-equity choice post the seminal work of Modigliani and Miller (1958). An increasing number of research studies conducted during the last six decades have identified several determinants of firms’ capital structure which include firms’ size, the tangibility of assets, industry median leverage ratio (IndMed), profitability, growth, uniqueness, non-debt tax shield (NDTS), etc. (Huang & Ritter, 2009; Leary & Roberts, 2005; Qian et al., 2009; Rajan & Zingales, 1995). Besides, the dividend payout ratio is another important factor that influences firms’ capital structure choice. The arguments on the impact of dividend payout on firms’ leverage ratios are ambiguous in prior studies (Frank & Goyal, 2009). The linkage between firms’ dividend policy and their capital structure choice is guided by two important theoretical arguments namely the agency theory argument and the signalling theory argument. As per the arguments of agency theory, debt is one of the important weapons to minimize the agency costs related to free cash flow (Jensen, 1986). Since the timely payment of interest on debt is mandatory, it reduces the free cash available to the managers thus minimizing the chances of misutilization of funds. Payment of dividends can substitute debt in mitigating the free cash flow agency conflicts as firms can reduce their cash hoardings by paying dividends to their shareholders (Fama & French, 2002). Further, by minimizing the reliance on debt, firms can also avoid problems related to excessive debts such as financial distress and bankruptcy risks (Karus & Litzenberger, 1973), and debt overhang and under-investment problems (Myers, 1977). Accordingly, agency theory predicts a negative relationship between dividend payment and leverage ratio.

Like agency theory, the signalling theory also suggests a negative impact of dividend payment on firms’ leverage ratio. However, unlike agency theory, the arguments of signalling theory are based on the information asymmetry between insiders and outsiders. The equity market is argued to be worst affected by information asymmetry between insiders and outsiders. Signalling theory claims that dividends can disseminate credible information about the firm to outsiders which ultimately minimizes the information costs of accessing the equity market. Accordingly, keeping other things constant, cheaper access to the equity market suggests an adverse impact of dividend payments on firms’ leverage ratio (Antoniou et al., 2008a, 2008b; Guha-Khasnobis & Bhaduri, 2002). On the contrary, another dimension of signalling theory suggests a favourable impact of dividend payment on financial leverage. Payment of dividends signals firms’ confidence about stability in their future earnings. Firms paying dividends also signal their better financial health and hence greater debt-taking capacity (Guha-Khasnobis & Bhaduri, 2002). Favourable earnings prospects and better financial health makes dividend-paying firms less risky which in turn allows them to have better financial accessibility in the financial market compared to dividend-non-paying firms (Byoun, 2008; Dang, 2013; Frank & Goyal, 2009; Ghose & Kabra, 2016; Guha-Khasnobis & Bhaduri, 2002). Hence, these arguments suggest that dividend-paying firms can borrow from the market at lesser costs and therefore, afford to maintain a higher leverage ratio compared to their opposite counterparts.

The empirical evidence on the impact of dividend payout on financial leverage is mixed in the existing literature. Few studies have investigated this issue in the context of US firms. While investigating the effect of tax benefits on firms’ corporate financing choices, Graham (2000) has observed a conservative leverage policy for US firms paying dividends which means dividend-paying firms rely lesser on debt compared to dividend non-paying firms. Fama and French (2002) investigated the relationship between target payout ratio and leverage ratio on the premise of the argument that the former can substitute the latter in minimizing agency costs of free cash flow. Negating the theoretical expectation, the findings of their study could not establish strong evidence of an adverse relationship between the two. However, Byoun (2008) has found a significant negative influence of dividend payment on the leverage ratio thus stressing the substitution role of dividend payment in reducing agency costs of free cash flow. Frank and Goyal (2009) and Antoniou et al. (2008b) have also observed a similar negative impact of dividend payments on US firms’ leverage ratio. In yet another study, Antoniou et al. (2008a) carried out an investigation in the context of market-based (UK) and bank-based (Germany) economies and observed that the impact of dividend payout ratio on leverage is country dependent. Their results reveal a significant negative and positive impact of dividend payout ratio on firms’ bank debt ratio in the context of the UK and German firms respectively. Finally, in the context of Indian firms, Guha-Khasnobis and Bhaduri (2002) observed insignificant impacts of dividend payout on both long-term and short-term leverage ratios. Therefore, it is evident that the existing empirical evidence is inconclusive with respect to the influence of firms’ dividend payout on their financial leverage.

Another important dimension in the relationship between dividend policy and capital structure overlooked in the existing literature is the moderating role of firms’ size. The underlying arguments in extant theories of capital structure and dividend policy, like agency costs, information asymmetry, adverse selection costs, signalling through dividend payment, etc., suggest that firms’ size may moderate the relationship between the dividend payout and leverage. Larger firms enjoy better financial accessibility compared to smaller firms as they are usually financially stable and more diversified than smaller firms. Besides, smaller firms face severe asymmetric information costs and thus have higher costs of issuing debts or equity compared to larger firms (Byoun, 2008; Ozkan, 2001; Rajan & Zingales, 1995; Titman & Wessels, 1988). Since dividend payout disseminates a favourable signal in the financial market (Myers & Majluf, 1984), small firms, with higher information costs, can use dividend payments for increasing their accessibility in the debt market. These arguments suggest a positive impact of dividend payment on leverage for smaller firms. Large firms, on the contrary, with better financial accessibility and lesser information costs may use dividend payments to substitute debt in mitigating agency costs of free cash flow. Further, with lower asymmetric information costs, they may give more preference to equity and less preference to debt to avoid agency costs associated with the latter thus having a negative impact on firms’ leverage.

Though there is no dearth of studies on identifying the capital structure determinants (Chaklader & Chawla, 2016; Chakraborty, 2010, 2013; Ganguli, 2013; Ghose, 2017; Ghose & Kabra, 2016, 2017, 2018; Handoo & Sharma, 2014; Sharma & Paul, 2015), studies investigating the impact of dividend payment on leverage ratio is scant in the context of listed firms in India. Further, there are no studies examining the influence of firms’ size in the relationship between dividend payment and leverage. The present study fills these gaps in the existing literature.

The remainder of the article is organized as below. The second section provides a description of the methodological framework which includes sub-sections on sample firms, measurement of variables and econometric modelling. The third section contains the empirical results. Finally, the fourth section concludes the article.

Methodological Framework

Sample Firms

The initial sample for the study includes all firms listed in any of the Indian stock exchanges and data of which are available in the ‘Capitaline Plus’ database during 2002–2019. Some standard restrictions are applied to decide the final sample (Byoun, 2008; Dang et al., 2014; Ozkan, 2001; Rajan & Zingles, 1995; Wojewodzki et al., 2017). Financial and utility firms are excluded from the study for having different capital structures. Firms without data for three consecutive years are excluded from the study. Firms with the value of net sales and total assets equal to zero or less than zero are also dropped from the study. Finally, firms with missing data on any of the needed variables are not considered for the study. The final dataset contains 4116 firms with minimum observations of 3 years and maximum observations of 18 years. The dataset contains 46,754 observations of which 17,397 firm-year observations (i.e., 37.21% observations) are dividend payers and 29,357 firm-year observations (i.e., 62.79% observations) are dividend non-payers.

Variables and Their Measurements

The dependent variable of the study is the firms’ leverage ratio and the independent variable of the study is the firms’ dividend payment status. Besides, the study uses seven principal elements of capital structure that are firms’ size, growth, industry median leverage ratio (IndMed), non-debt tax shield (NDTS), profitability, tangibility and uniqueness. The study employs the most used measures of variables from the existing literature. The definitions of dependent, independent and control variables of the study are shown in Table 1.

Table 1.

Description of Variables.

Variable Type	Variable	Definitions of Variable
Dependent	LEV1 (Total Debt) LEV2 (Long- term Debt) LEV3 (Short- term Debt)	Total debt divided by total assets Long-term debt divided by total assets Short-term debt divided by total assets
Independent	PAYERS	‘1,’ if the firm pays dividend in a particular year and ‘0’ otherwise

Control	Size	Natural logarithm of total assets
	Tangibility	PPE divided by total assets
	Profitability	EBIT divided by total assets
	NDTS	Depreciation divided by net sales
	Growth	Percentage change in total assets from previous year to current year


	Uniqueness	Selling expenses divided by net sales
	IndMed	Median values of leverage ratio for each industry on yearly basis based on NIC 2008 classification

Source: The authors.

Statistical and Econometric Modelling

The study uses a panel data regression model for examining the impact of dividend payment on firms’ leverage controlling the impact of some conventional capital structure determinants (Drobetz & Wanzenried, 2006; Frank & Goyal, 2003; Huang & Ritter, 2009; Qian et al., 2009; Rajan & Zingales, 1995). The fixed effect and random effect models are two important panel data regression models which control for the firm and time-specific heterogeneities. The study uses the Hausman test to select between the random effect and fixed effect model. Precisely, the study employs the following model for empirical analysis:

D P V_{i t} = α + β P A Y E R S_{i t} + λ C N T V_{i t} + ω_{i t}

(1)

Where the ‘i’ implies the number of firms in the sample and the ‘t’ implies the number of years. DPV stands for the dependent variable, which is the firms’ leverage ratio measured either in terms of long-term debt, short-term debt or total debt. The constant in the model is denoted by ‘α’. PAYERS is the dummy variable that takes the value ‘1’ for dividend payers and ‘0’ otherwise, and the coefficient of PAYERS is represented by ‘β’. CNTV implies a group of control variables including some principal elements of capital structure traced in earlier studies and time-specific effects. The coefficients of control variables are denoted by ‘λ’. The study controls the effect of seven principal elements of capital structure, which are firms’ size, growth, industry median leverage ratio (IndMed), non-debt tax shield (NDTS), profitability, tangibility and uniqueness. To control the influence of extreme observations, following previous studies (Flannery & Rangan, 2006), winsorization of all the control variables is done at the 1st and 99th percentiles. Finally, ‘ω_it’ denotes the composite error term of the model. The study used the procedure of White (1980) for adjustment of the standard errors for heteroscedasticity.

Empirical Results

Dividend Payment and Leverage

This section provides the empirical findings of the influence of dividend payment status on leverage in general. The study reports the summary statistics of variables, leverage comparison between payers and non-payers, and correlation matrix in Table 2, Table 3 and Table 4, respectively. A quick look at the results of Table 2 reveals that the average total debt to total assets ratio (LEV1) for Indian firms is 34% whereas, the average ratio of long-term debt (LEV2) and short-term debt (LEV3) to total assets are 16% and 17% respectively. It is evident from Table 3 that dividend non-payers have higher mean leverage compared to dividend payers with respect to all measures of leverage and all the differences are statistically significant at the 1% level¹. These results possibly provide support to the argument that dividends may serve as a substitute for debt in mitigating agency costs of free cash flow (Frank & Goyal, 2009). Moreover, a higher payout ratio may also be associated with lesser costs of equity due to positive signalling of future growth prospects. However, the results of univariate analysis are only indicative in nature as it overlooks the possible impacts of other explanatory variables on the dependent variable. The correlation matrix provided in Table 4 reveals interdependence among the dependent and independent variables. Nevertheless, the variance inflation factor (VIF) analysis indicates that the multicollinearity is not severe as VIF values are within the threshold value of 10 (Gujarati et al., 2012).

Table 2.

Descriptive Statistics.

Variables	Mean	Median	SD
LEV1	0.34	0.25	0.58
LEV2	0.16	0.06	0.31
LEV3	0.17	0.10	0.24
Size	4.43	4.31	2.23
Tangibility	0.31	0.27	0.25
Profitability	0.06	0.07	0.13
NDTS	0.03	0.02	0.02
Growth	0.18	0.06	0.56
Uniqueness	0.04	0.02	0.05
IndMed_TD	0.24	0.25	0.15
IndMed_LTD	0.08	0.06	0.09
IndMed_STD	0.11	0.12	0.08

Source: The authors.

Table 3.

Leverage (Payers vs Non-Payers).

Variables	Payers		Non-Payers		Payers vs Non-Payers (t-test)
Variables	Mean	Median	Mean	Median	Payers vs Non-Payers (t-test)
LEV1	0.26	0.24	0.41	0.26	−0.15 (−27.94)*
LEV2	0.12	0.07	0.19	0.05	−0.07 (−22.86)*
LEV3	0.14	0.11	0.19	0.10	−0.05 (−22.33)*

Source: The authors.

Note: (1) In case of t-test, figures outside parentheses are difference in mean value of variables between Payers and Non-Payers and inside parentheses are t-statistics.

(1) * denotes test statistics are statistically significant at 1% level of significance.

Table 4.

Correlation Matrix.

Variables	LEV1	LEV2	LEV3	Size	Tangibility	Profitability	NDTS	Growth	Uniqueness	IndMed_ LEV1	IndMed_ LEV2	IndMed_ LEV3
LEV1	1
LEV2	0.78*	1
LEV3	0.68*	0.21*	1
Size	−0.07*	−0.03*	0.00	1
Tangibility	0.21*	0.27*	0.13*	0.13*	1
Profitability	−0.20*	−0.14*	−0.15*	0.27*	−0.01	1
NDTS	0.19*	0.20*	0.15*	0.06*	0.59*	−0.01*	1
Growth	−0.07*	−0.04*	−0.08*	0.04*	−0.09*	0.16*	−0.14*	1
Uniqueness	−0.04*	−0.01*	−0.04*	0.16*	0.10*	0.13*	0.09*	−0.01**	1
IndMed_ LEV1	0.18*	0.20*	0.19*	0.12*	0.42*	0.05*	0.31*	−0.05*	0.06*	1
IndMed_ LEV2	0.16*	0.24*	0.09*	0.04*	0.42*	0.04*	0.30*	−0.02*	0.07*	0.81*	1
IndMed_ LEV3	0.14*	0.09*	0.24*	0.16*	0.30*	0.04*	0.24*	−0.06*	0.05*	0.84*	0.42*	1

Source: The authors.

Note: (1) * and ** denote test statistics are statistically significant at 1% and 5% level, respectively.

The results of the fixed effect regression model are reported in Table 5 for all three measures of leverage, that is, LEV1, LEV2 and LEV3. The chi-square statistics of Hausman tests are statistically significant in all the models which indicate the appropriateness of the fixed effect models to the random effect models. Accordingly, the study reports only the results of fixed effect models. The study conducts two Wald tests to examine the validity of explanatory variables used in the model. Wald Test 1 (i.e., Model Sig.) and Wald Test 2 (i.e., Time Effect Sig.) respectively investigate whether explanatory variables and time dummies used in the models are jointly significant. The respective null hypothesis for Wald test 1 and Wald test 2 is that the explanatory variables and time dummies have zero joint significance. As shown in the tables, the F-statistics obtained from Wald tests are statistically significant at 1% level in all the cases which means the explanatory variables and time effects used in the models jointly explain variations in the dependent variables. Finally, the adjusted R² values vary between 0.55 and 0.65 which are reasonably good.

The results in Table 5 show that constants have positive coefficients for all measures of leverage and the probability values suggest that the coefficients are statistically significant at 1% level of significance. The positive coefficients of constants suggest, keeping other things constant, Indian firms maintain a certain level of debt. The coefficient of the primary independent variable of the study, that is, PAYERS is negative and statistically significant for all the measures of leverage. The significant negative coefficients of PAYERS support the findings of univariate analysis and suggest lesser leverage for dividend-paying firms compared to their opposite counterparts. These results are inconsistent with the findings of some prior studies (Antoniou et al., 2008b; Fama & French, 2002; Guha-Khasnobis & Bhaduri, 2002) who found an insignificant influence of dividend payout on firms’ leverage. The following theoretical arguments possibly justify these results: first, dividend-paying firms, consistent with the arguments of signalling theory, may have better access to the equity market and thus rely more on equity as a substitution of debt compared to dividend non-paying firms. However, observed results, that is, lesser leverage for dividend payers, are inconsistent with another signalling argument that dividend payments positively impact leverage because of lesser costs of borrowing (Byoun, 2008; Dang, 2013; Ghose & Kabra, 2016; Guha-Khasnobis & Bhaduri, 2002). Second, as per the arguments of agency theory, debt is useful in minimizing free cash flow agency conflicts (Jensen, 1986). However, since debt has its own agency costs such as debt overhang and underinvestment (Myers, 1977), firms may use dividend payments as a substitute for debt in mitigating free cash flow agency conflicts and thus, having a negative impact of the former on the latter. To investigate the relevance of different theoretical arguments, the study compares the equity capital, borrowing cost, free cash flow and retained earnings of dividend-paying and non-paying firms². Table 6 reports the results of the comparison. First, it is evident from the results that the equity capital of payers is significantly lesser than the non-payers which is inconsistent with the argument that the former use equity as a substitute for debt. Second, the borrowing costs of dividend payers are lesser than the non-payers which supports the argument of dividend signalling theory. Third, there is no difference in free cash flow between the two groups of firms which is consistent with the argument that dividend payment substitutes debt in mitigating agency conflict of free cash flow. Finally, dividend payers have a substantially higher level of retained earnings than non-payers which is consistent with the argument of the pecking order theory. A higher level of retained earnings also explains the lesser level of leverage for dividend payers even with lesser borrowing costs. Overall, evidence suggests that the ability of dividend payment to mitigate free cash flow agency conflict together with the availability of a substantial volume of retained earnings to meet the financing requirements internally possibly minimizes the demand for debts in the case of dividend-paying firms and thus, negatively affects the leverage ratios of the firm.

Table 5.

Results (Overall).

Variables	LEV1	LEV2	LEV3
Constant	0.30 (6.81)*	0.08 (3.49)*	0.11 (6.09)*
Payers	−0.02 (2.35)**	−0.01 (−2.74)**	−0.02 (4.98)*
Size	−0.07 (6.26)*	−0.01 (−2.62)**	−0.01 (2.43)**
Tangibility	0.42 (8.05)*	0.30 (11.61)*	0.09 (3.86)*
Profitability	−0.50 (8.63)*	−0.18 (−6.54)*	−0.22 (9.68)*
NDTS	0.39 (1.99)**	0.44 (3.88)*	0.12 (1.45)
Growth	−0.01 (2.98)*	−0.00 (−0.78)	−0.01 (3.26)*
Uniqueness	0.14 (1.45)	0.04 (0.67)	0.05 (1.14)
IndMed	0.68 (7.90)*	0.63 (6.79)*	0.64 (8.92)*
Model Sig. (Wald Test 1)	18.70*	29.95*	22.56*
Time-Effect Sig. (Wald Test 2)	6.60*	5.00*	3.24*
Hausman Test	131.22*	209.11*	133.45*
Adjusted R²	0.65	0.61	0.55

Source: The authors.

Note: (1) Figures in parentheses are t-statistics. (2) Figures against Wald tests are F-statistics obtained from the tests. (3) Figures against Hausman Tests are chi-square statistics.

Table 6.

Characteristics of Payers and Non-Payers.

Variables	Overall	Payers	Non-Payers	Payers vs Non-Payers (t-test)	Remarks
Equity Capital	0.35	0.07	0.51	−0.43 (−60.45)*	Inconsistent with the substitution hypothesis
Borrowing Cost	0.03	0.02	0.03	−0.01 (−17.75)*	Consistent with the signalling hypothesis
Free Cash Flow	−0.05	−0.06	−0.04	−0.02 (0.08)	Consistent with the free cash flow hypothesis
Retained Earnings	−0.01	0.45	−0.28	0.72 (52.23)*	Consistent with the pecking order hypothesis

Source: The authors.

Note: (1) In case of t-test, figures outside parentheses are difference in mean value of variables between Payers and Non-Payers and inside parentheses are t-statistics.

(2) * denotes test statistics are statistically significant at 1% level of significance.

Results for control variables are mostly consistent with the findings of prior empirical studies. Size has negatively significant coefficients for all measures of leverage which suggest that larger firms include a lesser proportion of debt in the capital structure than smaller firms. The plausible explanation for the inverse impact of size on leverage is that large firms, in comparison to small firms, face lesser adverse selection costs in accessing the equity market (Mukherjee & Mahakud, 2010; Rajan & Zingales, 1995). Besides, large firms are usually mature firms with lower growth rates and higher profitability and therefore, may rely more on retained earnings than external financing to meet their financing requirements, thus having a negative impact on leverage. Tangibility has positively significant coefficients for all measures of leverage. The trade-off theory argues that the availability of a higher level of tangible assets provides firms the collateral for securing debts. Collateral grants security to lenders in the event of defaults and as a result, they charge lesser return on their investments. Therefore, these results are consistent with the arguments of trade-off theory and the findings of prior empirical studies (Dang et al., 2012; Drobetz & Wanzenried, 2006; Frank & Goyal, 2003; Handoo & Sharma, 2014; Korajczyk & Levy, 2003; Mukherjee & Mahakud, 2012).

The coefficients of profitability have negative signs which are in line with the predictions of pecking order theory (Myers, 1984; Myers & Majluf, 1984). The sources of external finance (both debt and equity) suffer from adverse selection costs due to information asymmetry between insiders and outsiders and, because of that, firms generally prefer internally generated funds over external sources of finance. As a result, firms with higher profitability maintain a lower level of debt. This result confirms the findings of studies like Korajczyk and Levy (2003), Flannery and Rangan (2006), Manos et al. (2007), Qian et al. (2009), Chakraborty (2010), Mukherjee and Mahakud (2012) and Ganguli (2013). NDTS has significant positive coefficients for total debt and long-term debt which are against the expectations of trade-off theory but similar to the findings of Bradley et al. (1984), Chakraborty (2010) and Chakraborty (2013). The possible explanation for this result may be the high positive correlation between tangibility and NDTS.

The growth of a firm has a significant adverse impact on total debt (LEV1) and short-term debt (LEV3) which implies that firms with more growth opportunities maintain lesser total debt via lesser short-term debts. The negative impact of growth confirms the argument of agency theory. Growth firms suffer more due to excessive debt in the form of forgoing future projects with positive net present values, that is, underinvestment problem. In order to avoid this situation, they preserve their financing (debt) capacity by using lesser debt at present (DeAngelo & DeAngelo, 2007; Myers, 1977). Moreover, firms with higher growth opportunities also suffer less from free cash flow agency conflicts. Growing firms using lesser short-term debt possibly indicates their strategy to avoid the future underinvestment scenario due to refinancing risk. Uniqueness is found to have an insignificant impact on all measures of leverage which is inconsistent with the argument that unique firms maintain a lower level of debt to avoid higher costs of liquidation (Titman & Wessels, 1988). Finally, the coefficient of IndMed is positive and statistically significant in all the models which implies that firms’ capital structure is substantially influenced by industry practices (Frank & Goyal, 2009).

Dividend Payment and Leverage: Moderating Role of Firm Size

This section reports the results of the second objective of the study, that is, to investigate the moderating role of firm size on the relationship between dividend payment status and leverage. To fulfil this objective, the study divides the sample firms into two groups namely large firms and small firms³. The study then creates two dummy variables namely SMALL and LARGE to separate the small and large firms respectively. SMALL is assigned with the value 1 in case the firm falls in the category of small firms and 0 otherwise. Similarly, LARGE is assigned with the value 1 in case the firm falls in the category of large firms and 0 otherwise. For testing the moderating effect of firm size, the study then extends the basic regression model (Model 1) by incorporating the interaction of SMALL and LARGE with explanatory variables as below:

D P V_{i t} = α + β S M A L L \times P A Y E R S_{i t} + λ S M A L L \times C N T V_{i t} + φ L A R G E \times P A Y E R S_{i t} + δ L A R G E \times C N T V_{i t} + ω_{i t}

(2)

In Model 2, β and λ will respectively denote the impacts of PAYERS and control variables on firms leverage in the case of small firms, and φ and δ will show the same results, respectively, in the case of large firms⁴. A statistically significant difference in coefficients of SMALL × PAYERS and LARGE × PAYERS will recommend that the influence of dividend payment status on firms’ leverage is not identical for small and large firms, and firm size plays moderating role in the relationship between dividend payment and leverage. Similarly, if coefficients for SMALL × CNTV and LARGE × CNTV are significantly different that will suggest that the determinants of capital structure function differently based on firm size.

Table 7 reports the comparison of leverage ratios between dividend-paying and non-paying firms separately for small and large firms. The results in Table 7 reveal that dividend payers have a significantly lesser level of leverage in the case of both small and large firms. These results give a preliminary indication of dividend payers’ incentive to maintain a lower level of debts than the non-payers irrespective of the size of firms. However, given the limitations of t-tests, the study relies on multivariate regression analysis for making inferences. The results of the extended regression model (Model 2), showing the moderating effects of firm size, is provided in Table 8. Table 8 also reports the results of various Wald tests. Wald Test 1 through Wald Test 8 show the statistical significance of the difference in coefficients of all explanatory variables between small and large firms. Wald Test 9 through Wald Test 12 respectively show whether all explanatory variables, explanatory variables with small firm dummy interaction, explanatory variables with large firm dummy interaction and time dummies are jointly significant under the null hypothesis of zero joint significance. It is evident from the results in Table 8 that coefficients of SMALL × PAYERS are statistically insignificant, whereas coefficients of LARGE × PAYERS are negatively significant for all measures of leverage. These results suggest that dividend payment does not influence the level of leverage in the case of small firms but adversely affects the level of leverage in the case of large firms. Further, the results reveal that the coefficients of PAYERS in the case of small and large firms are significantly different from each other for all the measures of leverage (Wald Test 1) which confirms the moderating role of firm size in the relationship between dividend payment and leverage. Further investigation (Table 9) reveals that dividend payers have lesser equity and borrowing costs, and higher retained earnings irrespective of firm size which is consistent with the findings of the overall analysis. The only difference between the small and large firms is the position of free cash flow. In the case of small firms, dividend payers have significantly higher free cash flow compared to non-payers which indicates that dividend payment cannot completely substitute debt to mitigate free cash flow agency conflict, and, thus, there is no negative influence of dividend payment on leverage. On the contrary, in the case of large firms, there is no significant difference in free cash flow between dividend payers and non-payers which is consistent with the argument that dividend payment possibly substitutes debt to minimize free cash flow agency conflict and thus, negatively affects leverage. Nevertheless, the insignificant impact of dividend payment on leverage is inconsistent with the argument that small firms make use of dividend signalling to increase their leverage.

Table 7.

Leverage Based on Firms’ Size.

Variables	Overall		Payers		Non-Payers		Payers vs Non-Payers (t-test)
Variables	Mean	Median	Mean	Median	Mean	Median	Payers vs Non-Payers (t-test)
Small Firms
LEV1	0.37	0.14	0.18	0.13	0.40	0.14	−0.22 (−12.65)*
LEV2	0.16	0.01	0.06	0.01	0.18	0.01	−0.11 (−12.34)*
LEV3	0.16	0.03	0.11	0.07	0.17	0.03	−0.05 (−7.96)*
Large Firms
LEV1	0.36	0.31	0.28	0.26	0.47	0.39	−0.18 (−29.55)*
LEV2	0.18	0.11	0.14	0.09	0.23	0.15	−0.09 (−24.63)*
LEV3	0.17	0.13	0.14	0.11	0.22	0.17	−0.08 (−28.99)*

Source: The authors.

Note: (1) In case of t-test, figures outside parentheses are difference in mean value of variables between Payers and Non-Payers and inside parentheses are t-statistics.

(1) * denotes test statistics are statistically significant at 1% level of significance.

Table 8.

Results (Moderating Effect of Firm Size).

Variables	LEV1	LEV2	LEV3
Constant	0.18 (3.10)*	0.05 (2.16)**	0.14 (6.26)*
Small_PAYERS	0.01 (0.67)	−0.001 (−0.23)	−0.001 (−0.61)
Small_Size	−0.13 (6.21)*	−0.03 (−3.12)*	−0.03 (−3.00)**
Small_Tangibility	0.40 (4.29)*	0.26 (5.47)*	0.09 (2.22)**
Small_Profitability	−0.48 (5.56)*	−0.22 (−5.08)*	−0.14 (−4.35)*
Small_NDTS	−0.28 (0.75)	0.04 (0.18)	0.03 (0.22)
Small_Growth	−0.01 (0.64)	−0.01 (−1.35)	0.00 (0.77)
Small_Uniqueness	0.04 (0.28)	−0.02 (−0.18)	0.04 (0.59)
Small_IndMed	0.73 (3.25)*	0.67 (3.97)*	0.35 (1.55)
Large_PAYERS	−0.03 (3.31)*	−0.02 (−3.88)*	−0.02 (−3.68)*
Large_Size	−0.001 (0.40)	−0.001 (−0.10)	−0.001 (−0.62)
Large_Tangibility	0.44 (7.87)*	0.37 (14.00)*	0.05 (2.03)**
Large_Profitability	−0.56 (6.52)*	−0.10 (−2.32)**	−0.38 (−10.22)*
Large_NDTS	0.77 (3.46)*	0.81 (6.16)*	0.01 (0.13)
Large_Growth	−0.02 (2.43)**	0.00 (0.59)	−0.02 (−6.22)*
Large_Uniqueness	0.26 (1.61)	0.10 (1.25)	0.08 (0.98)
Large_IndMed	1.37 (6.65)*	0.88 (6.85)**	0.34 (2.54)**
PAYERS (Wald Test 1)	12.33*	3.86**	2.54***
Size (Wald Test 2)	85.82*	9.23*	8.83*
Tangibility (Wald Test 3)	0.40	4.11**	0.82
Profitability (Wald Test 4)	0.69	4.15**	23.95*
NDTS (Wald Test 5)	17.80*	9.14*	0.01
Growth (Wald Test 6)	11.68**	2.07	23.19*
Uniqueness (Wald Test 7)	1.83	0.97	0.16
IndMed (Wald Test 8)	19.54*	1.42	0.001
Model Sig. (Wald Test 9)	14.85*	27.02*	14.93*
Small Firm Interaction Sig. (Wald Test 10)	16.85*	16.03*	6.70*
Large Firm Interaction Sig. (Wald Test 11)	39.95*	36.52*	21.51*
Time-Effect Sig. (Wald Test 12)	6.02*	2.80*	2.80*
Hausman Test	145.49*	215.32*	165.72*
Adjusted R²	0.66	0.62	0.55

Source: The authors.

Note: (1) Figures in parentheses are t-statistics. (2) Figures against Wald tests are F-statistics obtained from the tests. (3) Figures against Hausman Tests are chi-square statistics.

Table 9.

Characteristics of Payers and Non-Payers (Small and Large Firms).

Variables	Overall	Payers	Non-Payers	Payers vs Non-Payers (t-test)	Remarks
Small Firms
Equity Capital	0.72	0.19	0.78	−0.59 (−23.28)*	Inconsistent with the substitution hypothesis
Borrowing Cost	0.02	0.01	0.02	−0.01 (−3.15)*	Consistent with the signalling hypothesis
Free Cash Flow	0.003	0.037	−0.002	0.039 (8.03)*	Inconsistent with the free cash flow hypothesis
Retained Earnings	−0.47	0.40	−0.58	0.98 (20.10)*	Consistent with the pecking order hypothesis
Large Firms
Equity Capital	0.07	0.04	0.11	−0.07 (−34.83)*	Inconsistent with the substitution hypothesis
Borrowing Cost	0.03	0.03	0.05	−0.02 (−42.22)*	Consistent with the signalling hypothesis
Free Cash Flow	−0.29	−0.07	−0.64	0.57 (1.18)	Consistent with the free cash flow hypothesis
Retained Earnings	0.33	0.46	0.13	0.32 (43.71)*	Consistent with the pecking order hypothesis

Source: The authors.

The study also finds evidence confirming the moderating role of firm size in the relationship between leverage and its conventional determinants. Size has a significant negative influence on leverage only in the case of small firms but has insignificant influence in the case of large firms. These results possibly indicate that with the increase in size, small firms tend to rely more on equity capital thus having a negative impact on leverage. Tangibility favourably affects leverage and the impact is higher in the case of large firms. Though, the differential impact is significant only in the case of long-term debt (Wald Test 3). Availability of collateral coupled with larger size possibly gives better access to debt for larger firms. Consistent with the pecking order arguments, profitability has a significant negative impact on all measures of leverage irrespective of firm size. However, for small firms, the negative impact is more in the case of long-term debt and less in the case of short-term debt compared to large firms. These results possibly indicate that small firms have higher incentives to reduce long-term commitments, whereas large firms have higher incentives to reduce short-term commitments. The differential costs and agency conflicts associated with different debt maturity structures possibly explain these results. NDTS has a significant positive impact on long-term debt and total debt only in the case of large firms. A positive correlation between tangibility and NDTS possibly explains these results. Growth negatively affects short-term debt and total debt only in the case of large firms. These results suggest that large firms with higher growth opportunities reduce their reliance on short-term debt possibly to avoid the refinancing risk and the consequent underinvestment scenario. Uniqueness has an insignificant impact on leverage irrespective of firm size. Finally, industry median leverage positively affects leverage irrespective of firm size and the impact is more in the case of large firms compared to small firms (for total debt only). Better financial accessibility for larger firms possibly explains this result. Overall, the results suggest that the impact of dividend payment and other conventional determinants on leverage is not homogeneous for firms of different sizes. The size of the firm not only has a direct influence but also has an indirect influence on leverage through its moderating effect on the conventional determinants of leverage.

Conclusion

Theoretical arguments underpinning firms’ capital structure and dividend policy suggest a relationship between firms’ dividend policy and capital structure decisions. This article investigates the impact of a firm’s dividend payment decisions on its capital structure. It also examines whether firm size moderates the relationship between dividend decisions and capital structure. The study observes that dividend payment has a negative impact on all measures of leverage, that is, total debt, long-term debt and short-term debt which suggest a lesser proportion of debt in the capital structure for dividend payers than non-payers. Further investigation reveals that dividend payers have lesser equity capital and borrowings cost, and higher retained earnings compared to non-payers. However, there is no difference in free cash flow between the two groups of firms. Lesser equity capital is inconsistent with the hypothesis that payers rely more on equity capital, whereas lesser borrowing cost for payers is consistent with the dividend signalling hypothesis. The ability to minimize free cash flow agency conflict without debt and meet the financing requirement through internally generated funds possibly makes dividend payers less reliant on external debt. The study also observes that firm size influences the relationship between dividend payment decisions and capital structure choice. The negative influence of dividend payment is significant only in the case of large firms but not for small firms. The comparison of underlying characteristics of dividend payers and non-payers reveals similar results like the whole sample analysis irrespective of firms’ size barring one exception. In the case of small firms, dividend payers have a higher level of free cash flow compared to non-payers which possibly indicates that dividend payments do not completely substitute debt in solving free cash flow agency conflicts.

The findings regarding the control variables are mostly consistent with the arguments of conventional capital structure theories. Size, profitability and growth have negative impacts on leverage, whereas tangibility, NDTS and industry median leverage have positive influences on leverage. The negative impact of size may be attributed to lower adverse selection costs in the equity market, whereas the adverse impact of profitability supports the pecking order arguments. The inverse impact of growth indicates lesser reliance of growing firms on debt and is consistent with the agency costs (underinvestment problem) argument. On the other hand, the positive impact of tangibility can be associated with the availability of more collateral and lesser default risk for firms with higher tangible assets, whereas the positive coefficient of industry median leverage ratio reflects the significant influence of industry peers on firms’ capital structure. Nevertheless, the positive impact of NDTS rejects the arguments of the trade-off theory. The study also confirms the significant role of firm size on the impact of conventional capital structure determinants.

The present study is a useful contribution to the existing literature for at least three important reasons. First, there is no study in the Indian context specifically investigating the relationship between firms’ dividend policy and capital structure choice. The present study provides useful insights on this topic by examining the impact of the former on the latter and by revealing the underlying firm characteristics driving such impact. Second, prior studies surprisingly ignore the moderating effect of firm size in the relationship between various firm characteristics including dividend payment and firms’ capital structure. The present study confirms that the impacts of dividend payment status and other firm characteristics on capital structure are not homogeneous for small and large firms. Finally, there is hardly any study in the Indian context based on a large dataset of 4,116 firms covering a period of 18 years. Therefore, the findings of the present study are expected to better reflect the financing choices of Indian firms. Moreover, recently there are a number of reforms in the Indian economy like the introduction of the Insolvency and Bankruptcy Code, rolling out of Goods and Services Tax, further opening up of the economy for foreign investment etc. Given these developments, the present study revalidates the findings of prior empirical studies with the recent dataset. The findings of the study are expected to be useful to financial managers in deciding their financing choices.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

References

Antoniou

, Guney

, & Paudyal

(2008a). The determinants of corporate debt ownership structure: Evidence from market-based and bank-based economies. Managerial Finance, 34(12), 821–847.

Antoniou

, Guney

, & Paudyal

(2008b). The determinants of capital structure: Capital market-oriented versus bank-oriented institutions. Journal of Financial and Quantitative Analysis, 43(1), 59–92.

Bessler

, Drobetz

, Haller

, & Meier

(2013). The international zero-leverage phenomenon. Journal of Corporate Finance, 23, 196–221.

Bradley

, Jarrell

G. A.

, & Kim

E. H.

(1984). On the existence of an optimum capital structure: Theory and evidence. The Journal of Finance, XXXIX(3), 857–878.

Byoun

(2008). How and when do firms adjust their capital structures towards targets? The Journal of Finance, LXIII(6), 3069–3096.

Chaklader

, & Chawla

(2016). A study of determinants of capital structure through panel data analysis of firms listed in NSE CNX 500. Vision, 20(4), 267–277.

Chakraborty

(2010). Capital structure in an emerging stock market: The case of India. Research in International Business and Finance, 24, 295–314.

Chakraborty

(2013). Does capital structure depend on group affiliation? An analysis of Indian firms. Journal of Policy Modeling, 35, 110–120.

Dang

V. A.

(2013). An empirical analysis of zero-leverage firms: New evidence from the UK. International Review of Financial analysis, 30(1), 189–202.

10.

Dang

V. A.

, Kim

, & Shin

(2012). Asymmetric capital structure adjustments: New evidence from dynamic panel threshold models. Journal of Empirical Finance, 19(4), 465–482.

11.

Dang

V. A.

, Kim

, & Shin

(2014). Asymmetric adjustment towards optimal capital structure: Evidence from a crisis. International Review of Financial Analysis, 33, 226–242.

12.

DeAngelo

, & DeAngelo

(2007). Capital structure, payout policy and financial flexibility (Working Paper No. FBE 02 -06). University of Southern California.

13.

Drobetz

, & Wanzenried

(2006). What determines the speed of adjustment to the target capital structure? Applied Financial Economics, 16, 941–958.

14.

Fama

E. F.

, & French

K. R.

(2002). Testing trade-off and pecking order predictions about dividends and debt. The Review of Financial Studies, 15(2), 1–33.

15.

Flannery

M. J.

, & Rangan

K. P.

(2006). Partial adjustment towards target capital structures. Journal of Financial Economics, 79, 469–506.

16.

Frank

M. Z.

, & Goyal

V. K.

(2003). Capital structure decisions: Which factors are reliably important? Financial Management, 38(1), 1–37.

17.

Frank

M. Z.

, & Goyal

V. K.

(2009). Testing the pecking order theory of capital structure. Journal of Financial Economics, 67, 217–248.

18.

Ganguli

S. K.

(2013). Capital structure—Does ownership structure matter? Theory and Indian evidence. Studies in Economics and Finance, 30(1), 56–72.

19.

Ghose

(2017). Impact of business group affiliation on capital structure adjustment speed: Evidence from Indian manufacturing sector. Emerging Economy Studies, 3(1), 54–67.

20.

Ghose

, & Kabra

K. C.

(2016). Capital structure determinants of the Indian manufacturing sector: An empirical study. International Journal of Applied Management Research, 3(1&2), 68–83.

21.

Ghose

, & Kabra

K. C.

(2017). Capital structure, group affiliation and financial constraints: Indian evidence. Prajnan: Journal of Social and Management Sciences, 46(1), 9–36.

22.

Ghose

, & Kabra

K. C.

(2018). Dynamic capital structure adjustment and business group affiliations: Indian evidence. Business Perspectives and Research, 6(1), 27–41.

23.

Graham

J. R.

(2000). How big are the tax benefits of debt? The Journal of Finance, 55(5), 1901–1941.

24.

Guha-Khasnobis

, & Bhaduri

(2002). Determinants of capital structure in India (1990–1998): A dynamic panel data approach. Journal of Economic Integration, 17(4), 761–776.

25.

Gujarati

D. N.

, Porter

D. C.

, & Gunasekar

(2012). Basic econometrics (5th ed.). Tata McGraw-Hill.

26.

Handoo

, & Sharma

(2014). A study of determinants of capital structure in India. IIMB Management Review, 26, 170–182.

27.

Huang

, & Ritter

J. R.

(2009). Testing theories of capital structure and estimating the speed of adjustment. Journal of Financial and Quantitative Analysis, 44(2), 237–271.

28.

Jensen

M. C.

(1986). Agency costs of free cash flow, corporate finance and takeovers. The American Economic Review, 76(2), 323–329.

29.

Jensen

M. C.

, & Meckling

W. H.

(1976). Theory of firm: Managerial behavior, agency cost and ownership structure. Journal of Financial Economics, 3(4), 305–360.

30.

Karus

, & Litzenberger

R. H.

(1973). A state-preference model of optimal financial leverage. The Journal of Finance, 28(4), 911–922.

31.

Korajczyk

R. A.

, & Levy

(2003). Capital structure choice: Macroeconomic conditions and financial constraints. Journal of Financial Economics, 68, 75–109.

32.

Leary

M. T.

, & Roberts

M. R.

(2005). Do firms rebalance their capital structures? The Journal of Finance, LX(6), 2575–2619.

33.

Manos

, Murinde

, & Green

C. J.

(2007). Leverage and business groups: Evidence from Indian firms. Journal of Economics and Business, 59(5), 443–465.

34.

Modigliani

, & Miller

M. H.

(1958). The cost of capital, corporation finance and the theory of investment. The American Economic Review, XLVIII(3), 261–297.

35.

Mukherjee

, & Mahakud

(2010). Dynamic adjustment towards target capital structure: Evidence from Indian companies. Journal of Advances in Management Research, 7(2), 250–266.

36.

Mukherjee

, & Mahakud

(2012). Historical market-to-book ratio and corporate capital structure. Global Business Review, 13(2), 339–350.

37.

Myers

S. C.

(1977). Determinants of corporate borrowings. Journal of Financial Economics, 5, 147–175.

38.

Myers

S. C.

(1984). The capital structure puzzle. The Journal of Finance, XXXIX(3), 575–592.

39.

Myers

S. C.

, & Majluf

(1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187–221.

40.

Ozkan

(2001). Determinants of capital structure and adjustment to long run target: Evidence from UK company panel data. Journal of Business Finance and Accounting, 28(1&2), 175–198.

41.

Qian

, Tian

, & Wirjanto

T. S.

(2009). Do Chinese publicly listed companies adjust their capital structure toward a target level? China Economic Review, 20(4), 662–676.

42.

Rajan

R. G.

, & Zingales

(1995). What do we know about capital structure? Some evidence from international data. The Journal of Finance, L(5), 1421–1460.

43.

Sharma

, & Paul

(2015). Does liquidity determine capital structure? Evidence from India. Global Business Review, 16(1), 84–95.

44.

Titman

, & Wessels

(1988). The determinants of capital structure choice. The Journal of Finance, 43, 1–19.

45.

White

(1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 48(4), 817–838.

46.

Wojewodzki

, Poon

W. P. H.

, & Shen

(2017). The role of credit rating on capital structure and its speed of adjustment: An international study. The European Journal of Finance, 24(9), 735–760.