To pay or not to pay: what matters the most for dividend payments?

Ranajee Ranajee (ICFAI Foundation for Higher Education, Hyderabad, Telangana, India)
Rajesh Pathak (IBS-Hyderabad, ICFAI Foundation for Higher Education, Hyderabad, India) (Goa Institute of Management, Goa, India)
Akanksha Saxena (ICFAI Foundation for Higher Education, Hyderabad, Telangana, India)

International Journal of Managerial Finance

ISSN: 1743-9132

Publication date: 3 April 2018

Abstract

Purpose

The purpose of this paper is to test the stickiness of payout policy across times for Indian firms, by identifying the determinants of dividend payout (for amount of dividends as well as probability of dividends) and examine their predictive consistency through good and bad times, affiliation categories, amid controls for idiosyncratic characteristics. The authors also examine the scantly explored effects of financial constraints on firms’ dividend decisions.

Design/methodology/approach

The authors use various regression models, i.e. panel, Tobit and logit models; and amid control for firm-specific characteristics throughout the analysis.

Findings

The authors observe payout levels on average increasing with time for Indian firms. Further, group firms pay higher dividends compared to standalone firms. Firms’ leverage, profitability, non-promoters holdings, growth prospects and dividend event are apparently the important determinants of payout ratio and are mostly, but not always, consistent through times and firms’ categories, for both the amount as well as the likelihood of dividend payments. Financial constraints have an overall negative impact on dividends with significantly varying magnitude across periods of stability, crisis and recovery. Firms’ age and size are positive and significant factors for dividends level decisions in Indian firms, which is consistent with the life-cycle theory. However, inconsistent size and age effect is observed in determining the likelihood of dividend payment.

Research limitations/implications

This study adds to the growing literature on the changing trends and contributing factors of firms’ dividend payout policy.

Originality/value

This study provides evidence on predictive consistency of payout policy of firms and its determination with the change in the external economic condition.

Keywords

Citation

Ranajee, R., Pathak, R. and Saxena, A. (2018), "To pay or not to pay: what matters the most for dividend payments?", International Journal of Managerial Finance, Vol. 14 No. 2, pp. 230-244. https://doi.org/10.1108/IJMF-07-2017-0144

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


1. Introduction

The distribution of residual profit among residual claimants is dividend payout decision. Dividend payout has important implications as it gratifies the existing shareholders and entices the newer ones. Dividend payout decisions are a trade-off between ploughing back the profits, for better future, by investment into growth assets or returning it, immediately to investors, it actually belongs. The extant literature reports different confounding annotations concerning dividend distribution and its impact on firm value. Veteran value investors like Graham and Dodd (1940) reported a direct relation between dividends and market price of the share. Walter (1963) recommended payout on the basis of life stage of the firms whereas, Gordon (1962) believed risk averse investors prefer certain returns, hence, use higher discount rate for future returns. As per signaling hypothesis, managers use dividends to communicate better prospect of the company as compared to the competitors (Bhattacharya, 1979). Clientele effects entail mangers using dividends as a measure to satisfy the demand of payouts from congregated investors (Allen and Michaely, 2003). Modigliani and Miller, in their well-known dividend irrelevance theory, reported dividends to be irrelevant for firm valuation. Although payout policy is an important corporate finance decision, its ir/relevance from the companies or investors viewpoint is an unresolved puzzle, as reported by Black(1976).

The corporate payout policy can be understood as a function of willingness to pay and ability to pay by the company to the residual claimants. In reporting of willingness to pay, Fama and French (2001) found a sizeable decline in the fraction of US firms paying dividends during 1978-1999. Denis and Osobov (2008) and Ferris et al. (2009) echo similar findings, i.e. recording changes in the idiosyncratic characteristics of firms as the reason for the decline in dividend pay-out. De Angelo et al. (2004) raised suspicion on signaling clientele viewpoint as key determinants of dividend. They seem to be closer to Lintner (1956) suggestion that managers view earnings stability as one of the critical factors in dividend decisions. According to De Angelo et al. (2006), optimal payout policy is driven by the need to distribute the firm’s free cashflow. They propose a life-cycle theory that combines elements of Jensen’s (1986) agency theory with evolution in the firm’s investment opportunity set. The alternative theory reports that, in the early stage of life cycle, firms pay few or no dividends because their earnings are volatile and the investment needs exceed their cashinflows. In later years, internally generated funds exceed investment opportunities and earnings become relatively stable. To reduce the agency cost and possibility of wastage of free cash flow, firms chose to optimally pay out the excess funds to shareholders.

This paper builds on the above arguments and attempts to explore the factors affecting corporate payout policy and test their consistency in the world where uncertainty is the new normal. Firm’s uniqueness does play a role in effective and efficient use of factors of production to attain its objectives, but cannot guarantee level, growth, and sustainability of profit. To determine earning levels itself is a challenge, it is difficult to answer questions such as the following: Whether payout decisions are firm specific? Do firm categories matter for these decisions? What is the relevance of economic conditions for dividend payments? Is it related to the forces of economic expansion and contraction? Is payout policy time dependent? What really explains the payout ratios of firms’ consistently through good and bad times or if there exist only temporal predictors of dividends?

The objective of the paper is to understand the dividend payout patterns through good and bad times and examine predictive consistency of its determinants for both the dividend levels and the dividend event. We use data of Indian firms over a prolonged period of 2001-2015, including the recent financial crisis. First, we test the stickiness of firms’ payout across economic conditions (periods of stability, crisis and recovery). Institutional voids in emerging markets (Khanna and Palepu, 1997) make it vital to analyze whether payout policy of group-affiliated firms differs from that of standalone firms, which we verify next. Further, we revisit determinants of dividend payout (for event of dividends payment as well as extent of dividends) and examine their predictive consistency across different economic conditions. Finally, we investigate how the dividend payouts differ for firms that are financially constrained as compared to unconstrained firms.

The study has several distinct features. First, the study is based in emerging markets context, which is infamously known for higher level of information asymmetry. Hence, dividend payout as a device to reduce information asymmetry should have greater impact as cash dividends can take care of the institutional voids. Second, reputed business groups use similar mechanism to exercise control and to expropriate minority shareholders. The paper studies the impact of ownership on payout policy. Third, as all business groups firms are not the same, we investigate the implications of firm-specific variables not only on the level of dividend but also on the dividend event. Fourth, the study tests dividend payout in different economic situations to see if dividend payout is different across them as well as the role of financial constraints in payout decisions.

We use various regression models, i.e. panel, Tobit and logit models; and control for firm-specific characteristics throughout the analysis. Our study adds to the growing literature on the changing trends and contributing factors of firms’ payout policy. The study provides evidence on predictive consistency of payout policy of firms and its determination with the change in the external economic condition.

The rest of the paper is organized as follows. The next section discusses the hypothesis development followed by data and variables used in the study in Section 3. The empirical methods and results are presented in Section 4. Section 5 concludes the study.

2. Literature review and hypothesis development

The extant literature on dividends and its determinants reveals several facets of firms’ payout policy. We present a summary of relevant existing studies in this area suitable for our investigation.

2.1 Stickiness of dividends

Lintner (1956), using responses from managers of 28 companies, reported unwillingness of managers in cutting dividends. The study concludes that based on sustainable earnings, managers of mature firm decide on target payout ratio and stick to it even if they have to go the extra mile. Managers signal prospects of future earning to shareholders by dividend announcement; hence, a decrease is not preferable (Bhattacharya, 1979; John and Williams, 1985; Bernheim, 1991). Following the dividend discount models of stock valuation, a stock price is the sum total of all the future dividends, discounted at their opportunity cost. As per some of existing studies on dividend patterns, four out of five firms keep their quarterly dividend constant (Aharony and Swary, 1980; Loderer and Mauer, 1992; Nissim and Ziv, 2001). A decrease in productivity should not result in change in dividends (Baker and Powell, 1999).

Using survey evidence of 384 executives, Brav et al. (2005) reported dividends conveying positive information about the company and is not only preferred by existing retail investors but also attracts the newer ones. Therefore, mangers, instead of cutting on dividends, would rather compromise on investment decisions or would take help of financing alternatives available to them. De Angelo et al. (2006) reported the lack of enthusiasm in the managers in cutting dividends for the firms with long history of dividend payment. It is evident that stock market’s reaction to the dividend cuts is negative, which makes it sticky. Larkin et al. (2016) reported payment of premiums to the shares of consistent dividend paying firms. Cost of capital of the firms paying stable dividend is reported to be lower than comparable firm in the same study.

Sticky dividends also build the threat to the continued existence of the firms. If the external environment is not supportive for the firm’s operating activities, then keeping the dividend unceasing would lead to bankruptcy (Kumar and Lee, 2001). Based on good times and bad time analogy, Karpavičius (2014) favored a volatile dividend stream and advised the firms to cut dividends rather than default. Combining the arguments above, we investigate whether dividends are sticky or not for Indian firms. In fact, our test of stickiness for Indian firms is tougher as it includes crisis period. We posit that strong clientele effects primarily led by tax- free dividend income to stockholders in India are likely to make Indian dividend paying firms continue maintaining the same even during the crisis. Moreover, the reluctance of managers to reduce dividend (Lintner, 1956) and resilience of payout reported by Floyd et al. (2015) strengthens our belief. Thus, we propose the following hypothesis:

H1.

Dividends are sticky through periods of stability, crisis and recovery for Indian firms.

2.2 Group affiliation and dividends

Business group is a collection of firms, with common business family as the promoter of the group. They share resources and are interlinked with each other. Common and well-known group name provides the credibility and reputation to the member firms which makes the problem of information asymmetry and market imperfections non-existent for them. Manos et al. (2007) and La Porta et al. (2000) reported dividend payout to be a function of information asymmetry and external finance. The role of group becomes important when we use it in the context of dividend payment, as group affiliation has implication for both the above factors.

Furthermore, inside monitoring and information sharing among the group members reduce the information asymmetry, whereas the presence of internal capital market for group firms reduces the dependence of the firms on external financial market for dividend decisions. Hence, group firms should consistently pay a higher dividend as compared to standalone firms. Similar results are reported in previous studies (Gopalan et al., 2007; Khanna and Palepu, 2000; Manos et al., 2012). Gopalan et al. (2014) reported the similar results for 22 countries of Asia and Europe. Manos et al. (2012) argued the contrarian view by describing group firms as self-determining firms independent of external capital market’s disciplining mechanism. As group firms donot need external financial market for their investment needs, they are not required to signal anything to the capital market or the residual claimants. Hence, they can afford to pay a lower or no dividends as compared to standalone firms. Combining the above arguments, we investigate whether the group affiliation affects the dividend payout of a firm. Thus, we propose the following hypothesis:

H2.

Business group firms consistently pay higher dividends than the standalone firms.

2.3 Determinants of dividends

While discussing determining factors of dividend payout, it is vital to consider both the amount of dividends as well as event of dividend payment. Aivazian et al. (2003) reported the dependence of payout policy on firm-level variables. Fama and French (2001) reported likelihood of paying dividends by firms and its relation to size, growth opportunities, and profitability. Denis and Osobov (2008) investigated payout in six countries and revealed that the likelihood of paying dividends is strongly associated with the ratio of retained earnings to total equity. Moreover, most dividend paying firms are those whose equity consists primarily of retained earnings and very few firms pay dividends when retained earnings are negative. Reportedly, following firm-specific variables, across time and space, they have more often been identified as predictors of dividend payout levels and event.

2.3.1 Firm size (size)

Smaller firms face higher borrowing constraints and more likely to suffer from financial distress due to higher information asymmetries (Berger et al., 2001). Lloyd et al. (1985) and Vogt (1994) reported higher dividend payout ratio of large firms as they are mature firms, which can access the external capital market and are not dependent on internal financing for dividend decisions. Thus, large size firms are more likely to pay dividends.

2.3.2 Leverage

Analyzing the effect of leverage on dividend decision, Barclay et al. (1995) found higher level of financial leverage to be deterrent for financial flexibility which increases the level of financial risk and hence reduces the level of dividends. Ferreira and Vilela (2004) found the greater ability of highly levered firms in tapping the external market for funding and report a higher dividend payout by them.

2.3.3 Profitability

Profitable firms tend to pay high dividends and they pay more often than not (Amidu and Abor, 2006; Baker et al., 1985). However, Gopalan et al. (2014) reported a negative relation between profitability and dividend payout because of availability of large number of investment project to the profitable firms.

2.3.4 Firms’age

Lifecycle theory argues that firms adjust their dividend policy through time, in response to changes in investment opportunities (DeAngeloet al., 2006). Therefore, younger firms tend to pay lower dividends due to abundance of investment opportunities coupled with low internally generated capital. The fund requirements for a growing firm are more; therefore, they are likely to retain larger portion of earnings and payout less (Amidu and Abor, 2006). Later in the firm’s lifecycle, internally generated cash exceeds investment opportunity so higher dividends are paid to mitigate free cash flow problems. Evidence in support of the theory is found by De Angelo et al. (2006) and Denis and Osobov (2008). However, Von Eije and Megginson (2008) found no correlation between retained earnings (to total equity) and the likelihood of paying dividends.

2.3.5 Growth (M/B)

Growth firms require funds for investment in assets. When internally generated funds are not sufficient to finance the investment needs, external financing is used. External financing is costly and incorporates transaction cost. Rozeff’s (1982) cost minimization model establishes that managers want to minimize the transaction cost of external financing, and hence report a negative correlation between growth and dividend payout.

2.3.6 Cash holdings

Chen et al. (2014) showed that firms paying dividends in the previous year hold more cash in the current period, whereas Ferreira and Vilela (2004) reported dividend paying firms hold less cash.

2.3.7 Non-promoters non-institutional (NPNI) holdings

The signaling theory explains dividend announcement by managers of a firm as signaling mechanism to reduce information asymmetry regarding better prospects for the firm in future, and hence reports a positive relation between them (Bhattacharya, 1979; Dasilas and Leventis, 2011; Miller and Rock, 1985). According to Denis and Osobov (2008), higher earning firms donot need to signal anything to the market ex ante, whereas Li and Zhao (2008) using US stock data reported a negative relationship between dividend payout and information asymmetry. Manos et al. (2012) reported similar results. NPNI, in literature, is proxy for information asymmetry, since the higher holding by individuals results in ineffective monitoring thereby causing high levels of information asymmetry. Consistent with pecking order theory and Manos et al. (2012), we expect a negative coefficient of this variable in explaining dividend decisions. Combining the above arguments, we investigate whether firm-specific characteristics affect the dividend payout of a firm with the help of following hypotheses:

H3a.

Firm-specific characteristics affect dividend payout levels.

H3b.

The effect of firm-specific characteristics on payout levels varies through sub-periods i.e. stable, crisis and recovery.

H3c.

Firm-specific variables impact the likelihood of dividend event.

2.4 Impact of financial constraints on dividends

Koo et al. (2017) reported relationship between financial constraint and dividend payout policy. Firms with financial constraint plough back the earnings than paying it as dividends. A precautionary or speculative motive of the firm prevents the firms with volatile cash flow from paying the dividends. Chay and Suh (2009) found negative relationship between the uncertainty about the firm’s cash flow and dividend payout. Floyd et al. reported the opposite view in their study of US firms, where the authors reported an increased dividend payout by the firms and found crisis of 2008 to be an insignificant event for payout policy of the firms. Combining the above arguments, we investigate whether the financial constraint affects the dividend payout of a firm. Thus, we propose the following hypothesis:

H4a.

Financial constraint affects the dividend payout negatively.

H4b.

The effect of financial constraint on the payout varies through sub-periods, i.e. stable, crisis and recovery.

3. Data and variables

The data used in the study are collected from the Centre for Monitoring the Indian Economy Prowess database, a publicly available firm-level database (Ghosh, 2016). Our sample covers all firms listed on National Stock Exchange and Bombay Stock Exchange for the period of April 2001-March2016. We exclude financial firms, utility firms, and government firms as their decisions are largely influenced by regulations and government policies. We collect annual data for firm characteristics measured by different variables. Subsequently, we delete several firms from the sample such as extremely misreported data on some of the relevant variables, data of firms with no reported information on relevant variables, etc. We winsorize all variables at the bottom and the top at 1percent level to mitigate the effect of outliers. After these adjustments, the final panel consists of 29,189 firm-year observations.

Consistent with the literature, our dependent variable is payout ratio measured by ratio of dividends per share to earnings per share; however, for investigation of likelihood of dividend payment, we use “Dividend_dummy,” i.e. a dummy variable that takes value of 1 for dividend paying firm and 0 otherwise. Our prime explanatory variables are “Time dummies” for recovery and crisis periods with stable period as period of reference. The dummies for crisis and recovery periods take value of 1 for the corresponding year’s observations and 0 otherwise. We, based on behavior of major Indian indices, consider period 2001-2008 as stable period, period 2009-2010 as crisis period and period 2012-2016 as recovery period (see Figure 1) for classification of firm year observations across times. Our identification of crisis period, i.e. years 2009-2010, serves the purpose of alienating crisis effects adequately as the data frequency is annual, collected at the end of March every year thereby impounding the effects, since the inception of crisis in September, 2008 till the first quarter of 2010 when it largely weakened. Nonetheless, we also conduct the analysis considering period 2009-2011 as crisis period and results (not reported for brevity purpose but available on request) remain qualitatively similar.

Besides, we proxy group affiliation through “Group_dummy” that takes value of 1 for group-affiliated firms and 0 for standalone firms. Financial constraint of the firm is alternatively measured by HP index (Headlock and Pierce index, 2010) and by cash-flow volatility (CFVOL). We also use industry dummies to control for invariant industry effects. We follow the two-digit National Industry Classification to define industry. We interact the potential determinant variables with time and group dummies to test for consistency of the relationship.

Other important explanatory variables include firm size, leverage, profitability, age, growth prospects, cash holdings, NPNI holdings, etc. The key variables of the study are listed and defined in Table I.

4. Empirical methodology and results

4.1 Model specification

We apply panel regression and logistic regression in our study for examining payout level decisions (how much to pay) and likelihood of payout decision (to pay or not to pay), respectively. However, we alternatively apply a more suitable censored Tobit regression for dividend level (as the values this variable takes is greater or equal to 0) investigation to verify robustness of our results. Our basic model of estimation is presented as follows:

Y i t = α i + Σ α i T i m e _ d u m m i e s i + Σ β j det e r min a n t s j + Σ γ k D e t e r min a n t s × T i m e d u m m i e s k + ε i t
where Yit is the value of dividend payout for firm i in year t, or dividends_dummy taking value 0 or 1, representing dividend levels and dividend events, respectively, in separate regressions. The αi would provide change in dividends through time; βj would help identifying the important determinants of dividend decisions, whereas γk,would explain the consistency of determinants through sub-periods of study. Besides, we also alter the model specifications slightly to attend all the objectives identified in the study and that can be traced from tables of results for both level and likelihood decisions.

4.2 Results and discussion

4.2.1 Univariate analysis

Descriptive statistics of the key variables of study are reported in Table II. For the overall time periods, the average payout for Indian firms is observed to be 20.7 percent with median 11.8 percent. NPNI, our main proxy for information asymmetry, is found to have an average of 41.9 percent for all the Indian firms with standard deviation 20.3 percent. Average cash volatility is 56.6 percent and is negatively skewed like size of the firm, whereas average ROA is 6.7 percent. Leverage and M/B ratio of firms are on average 37.6 and 2.41 percent, respectively.

In Table III, we present the average dividend payout of all the firms in our sample for the quintiles formed on the variables – financial constraint and cash volatility. We rank firms based on these variables and form quintiles. We then compute average dividend payout for every quintile to observe and analyze how the average dividend payment changes with changes in these variables.

We compute in Table III the mean dividend payout for overall and sub-periods considered in our analysis; stability period (2001-2008), crisis period (2009-2010) and recovery period (2011-2016). As expected, for the overall sample period, average payout of the firm decreases with financial constraint of the firm; that is firms with more financial constraints are paying less dividend to shareholders. Surprisingly, such relationship is not observed in case of cash volatility, where it is expected that firms with high cash volatility to decrease the average payout. The effect of crisis can thus be observed from the fall of overall dividend payout, which is consistent with financial constraint averages too. However, firms classified on cash flow volatility exhibit peculiar patterns where the average payout of firms with low cash flow volatility has risen during crisis.

Table IV presents the correlation table of key variables employed in the study. The size and age of the firm correlate positively with each other. The CFVOL surprisingly correlate negatively with financial constraint variable, positively with firm size and negatively with ROA. NPNI has a negative association with firm size and CFVOL and positive correlation with financial constraints variable. The magnitudes of other associations are not found meaningful.

We also present the behavior of annual averages of key variables in Figure 2. Here, all variables except DP (payout ratio that appears against right axis) are plotted against left axis. It shows that MB is one of the most volatile variables in the group. ROA turns negative during crisis period as expected. The average DP ratio increases with time as opposed to NPNI, whereas the leverage has inverted U shape. The yearly averages of leverage are indicative of leverage build up in years immediately preceding the financial crisis and thereafter a drop down in succeeding years.

4.2.2 Multivariate analysis

Table V presents the panel results examining varying levels of dividend payment through good and bad time and in the presence of group and dividend event control. Model (1) includes the time dummies only with control to firm fixed effects. We observe coefficients for recovery period being statistically significant and positive. The intercept signifies the average dividend payout for Indian firms during stable time i.e. on average 19.2 percent, and the negative but insignificant coefficient of crisis dummy indicates that though dividends drop slightly during crisis time but do not differ significantly compared to that of stability period. This is consistent with the stickiness behavior of dividends. Besides, the payout of firms during recovery period is on an average 4.8 percent higher than that of stable period resulting in an average of 24 percent (0.192+0.048).

These results are consistent with the results of Floyd et al., who examined dividend payments of US firms and suggested no impact of 2008 crisis. They also observe increase in dividend payout in post-crisis periods.

Model (2) is extended through inclusion of dummy for group-affiliated firms as it is observed in finance literature that business group firms take different decisions. The results are consistent with the previous model and further suggest that group firms on average pay 5.7 percent higher dividends compared to standalone firms in all times. Model (3) includes a dummy for dividend event as explanatory variable where it takes a value of 1 if the company paid dividend last year and 0 otherwise. The empirics suggest that firms that paid dividend last year pay around 15.5 percent higher dividend than others do in stable period. Moreover, these firms also cut on dividend during bad times and their group affiliation moderately matters for payout level decisions. Model (4) examines the model with control to firm age and size besides firm fixed effects and provides results similar to model 3. The R2 improves significantly for this model possibly due to cross sectional control for age and size and age appears driving payout levels positively.

Table VI presents the Tobit regression results examining the determinants of dividends and their consistency through periods of stability, crisis and recovery. Column (1) reports empirics of the predictors of payout levels to the control of firm specific size and age factors. The profitable firms are found paying higher dividends, whereas high information asymmetry measured by NPNI associates with decrease in dividends. The dividend event last year positively affects the payout levels in subsequent years. whereas growth potential measured by MB ratio affects payout levels negatively.

Column (2) reports consistency of predictors examined through interactions with time dummies, whereas column (3) shows the effects for group firms vs standalone firms. Leverage is a significant determinant of payout level only during stable time with a negative sign. NPNI effects on dividends differ significantly during crisis period as compared to other times. The profitability of firms also has positive but varying effects on dividends level across times, whereas the effect of MB ratio turns positive during recovery period. For group firms, the effects are consistent mostly except for NPNI where the relation turns positive. It suggests that the high information asymmetry in group-affiliated firms leads to higher payout levels.

Table VII reports the results of payout decisions under financial constraints and cash-flow uncertainty. From column (1), we infer that financially constraint firms pay lower dividends during stable periods however; the payout levels have improved during crisis and recovery period implying that the behavior of dividend is unchanged through time even for constrained firms. Column (2) results indicate that firms with more uncertain cash flows experience significant fall in payout levels during crisis and recovery periods. For the group firms, interaction coefficients in column (3) suggest that a group-affiliated constrained firms pay moderately lower dividends than unconstrained firms. Moreover, a group firm with uncertain cash flows pays significantly lower dividends (around 15 percent) compared to standalone firms, this can be attributed to the huge internal funding needs of member firms in times of high cash-flow uncertainty.

Table VIII provides the logistic regression results for determinants of dividend event across sub-periods of study. We observe that likelihood of dividends increases for firms with high cash holding and the same holds true for a group firms consistently across the sub-periods of study. The likelihood of dividend payout reduces significantly for firms exhibiting volatile cash flows in all times. The information asymmetry levels in a firm increase the chance of dividend payments significantly only during crisis period and reduce it in other times. High leverage reduces the likelihood of dividend in good times but surprisingly during crisis and recovery it induces the likelihood of dividend event. Profitable firms seemingly are more likely to pay dividend in good times and abstain from paying in troublesome periods.

The idiosyncratic factors, such as firm size and age, unlike in other periods, turn ineffective during crisis in determining payout event. The likelihood of dividend event is observed high for constrained firms during stable periods but less during crisis and recovery periods. Thus, from all empirical results, we observe the inconsistency of most predictors of dividends level and event both in terms of their magnitude and direction through periods of stability, crisis and recovery.

5. Conclusions

The paper identifies the determinants of dividend payout and examines their consistency across time periods (stable, crisis and recovery) in a multivariate regression framework. The analysis is conducted using dummy variables for sub-periods and firm categories in a panel regression setting along several controls. Using all firms listed on stock exchanges in India for the period of 2001-2015, we find that firms exhibit stickiness behavior for dividends payout. Further, we observe that group firms on average pay 5.7 percent higher dividends as compared to standalone firms in all times. Firms that paid dividend last year pay around 15.5 percent higher dividend than others do in stable period. Moreover, these firms also cut on dividend during bad times and their group affiliation moderately matters for payout level decisions. Size and age affect payout levels positively. Leverage is a significant determinant of payout level only during stable time with a negative sign. The profitability of firms also has positive but varying effects on dividends level across times, whereas the effect of MB ratio turns positive during recovery period. NPNI effects on dividends differ significantly during crisis period as compared to other times. Average payout of the firm decreases with financial constraint of the firm. Group firm with uncertain cash flows pays significantly lower dividends (around 15 percent) compared to standalone firms, this can be attributed to the huge internal funding needs of member firms in times of high cash-flow uncertainty. Overall, we find support for all outlined hypothesis in the study.

The study spells out the prominent impact of idiosyncratic factors on dividend payout, an important corporate finance function. Besides, it also reveals the varying effects of the factors during good and bad times. The study, overall, highlights that the payout policy of a firm is dependent on several firm-specific characteristics. Moreover, they affect the payout decisions differently in different times. The analysis based on disaggregation of firms on the basis of group affiliation and financial constraint also reveal that dynamic effect of various firm classifications plays a vital role in determining both the payout levels and payout decision. The future research on payout policy can be examining the role of risk aversion of mangers and controlling mechanism used by board of directors in determining dividend payout policy.

Figures

Plot of Nifty50 and sensex values

Figure 1

Plot of Nifty50 and sensex values

Plot of annual averages of key variables

Figure 2

Plot of annual averages of key variables

Description of variables

Variables Definition
Dividend payout Dividend paid per share/earnings per share
Firm size(Size) A natural log of total assets
Leverage Ratio of book debts to total assets
Profitability (ROA) Ratio of operating profits to total assets
Age Age is measured by natural log (data year-incorporation year)
Growth (M/B) Average price to book ratio of firm
Cash holding The ratio of cash to total asset net of cash and cash equivalents
NPNI Total holding percentage of non-promoters and non-institutions
Dividend dummy A dummy variable takes a value 1 for firms that pay dividend, and 0 otherwise
Dum_Stable Dummy variable takes the value 1 for observations pertaining to year 2001-2008 and 0 otherwise
Dum_Recovery Dummy variable takes the value 1 for observations pertaining to year 2011-2015 and 0 otherwise
Group_dummy Dummy variable that takes a value 1 for firms that belong to a business groups and 0 for standalone firms
Financial constraint (HP) Headlock and Pierce (HP, 2010); index uses quadratic relation between size and constraints and linear relation between age and constraints to develop the following equation: −0.737×size−0.043×size2−0.040×age
Cash-flow_Vol (CFVOL) Cash flow volatility is measured by standard deviation of cash flow over five years, scaled by total assets

Note: The table presents the definition of variables used in this study

Descriptive statistics

Stats DP Cash_vol MB Lev. ROA NPNI. Size Age Fin_const.
Mean 0.207 0.566 2.414 0.376 0.067 0.419 7.467 26.044 −8.232
SD 0.821 0.622 19.014 1.464 0.858 0.203 2.179 20.027 2.989
SE(mean) 0.006 0.004 0.125 0.009 0.005 0.001 0.013 0.117 0.018
P50 0.118 0.510 0.930 0.302 0.061 0.407 7.582 21.000 −8.187
P25 0.000 0.000 0.390 0.147 0.018 0.265 6.011 13.000 −10.082
Skewness 63.838 −1.895 82.577 40.547 16.399 0.374 −0.193 1.723 −0.318
P75 0.273 1.080 2.150 0.455 0.119 0.556 8.887 31.000 −6.093

Notes: TableII presents the summary statistic for key variables used in the study. It reports mean, standard error of mean (SE), median, 25th and 75th percentiles (P25 and P75), Skewness and standard deviation (SD). Variables reported are dividend payout(DP), Non promoter non institutional holding (NPNI), volatility of cash (cash_vol), price to book ratio (MB), Book leverage (Lev), Return on assets (ROA), Size of the firm (Log of total assets ), age of the firm (log of Age), and financial constraint variable measured by HP index (Fin. Const)

Dividend payout patterns under financial constraint and cash-flow uncertainty

Rank_finconstr Rank_cash_volatility
Timeline 1 2 3 4 5 1 2 3 4 5 Total
2001-2008 0.266 0.227 0.189 0.137 0.084 0.159 0.183 0.195 0.200 0.222 0.191
2009-2010 0.214 0.196 0.211 0.176 0.098 0.252 0.207 0.197 0.137 0.114 0.187
2011-2016 0.285 0.246 0.204 0.266 0.185 0.316 0.254 0.224 0.261 0.117 0.241

Notes: Table III presents the average DPS based on firms ranked on financial constraint and cash volatility quintiles. The averages are reported across sub-periods defined by stability (2001-2008), crisis (2009-2010) and recovery (2012-2015) for low (1) to high (5) quintiles for both variables. The sub-periods wise averages are reported in last column

Correlation table

NPNI Cash_vol MB Lev. ROA Size Age Fin_const. DP
NPNI 1.000
Cash_vol −0.358 1.000
MB −0.153 0.029 1.000
Lev. 0.061 0.016 −0.046 1.000
ROA 0.098 −0.381 0.108 −0.119 1.000
Size −0.485 0.559 0.120 0.076 −0.182 1.000
Age −0.176 0.154 0.072 −0.067 −0.042 0.344 1.000
Fin_const. 0.486 −0.547 −0.120 −0.067 0.177 −0.997 −0.347 1.000
DP −0.068 0.002 0.030 −0.053 0.011 0.080 0.073 −0.076 1.000

Notes: TableIV reports correlation coefficients of key variables, i.e. dividend payout(DP), non-promoter non-institutional holding (NPNI), volatility of cash (cash_vol), price to book ratio (MB), Book leverage (Lev), Return on assets (ROA), Size of the firm (Log of total assets ), age of the firm (log of Age), and financial constraint variable measured by HP index (Fin. Const). Numbers in italic indicate significant values at 5 percent level

Panel regressions results using dummy variables

1 2 3 4
Dum_crisis −0.005 (−0.29) −0.009 (−0.49) −0.044 (−2.37)*** −0.050 (−2.61)***
Dum_recovery 0.048 (3.58)*** 0.044 (3.31)*** −0.010 (−0.74) −0.022 (−1.48)
Dum_grp 0.057 (4.33)*** 0.022 (1.65)*
Div_dum 0.155 (11.27)*** 0.145 (9.70)***
Size 0.004 (0.99)
Age 0.001 (3.36)***
_cons 0.192 (24.42)*** 0.168 (17.33)*** 0.112 (10.28)*** 0.068 (2.47)***
R2 0.030 0.058 0.070 0.326
Firm fixed effects Yes Yes Yes Yes

Notes: *,***Significant at 10 and 1 percent levels, respectively

Tobit regressions results

1 2 3
Dum_crisis −0.171(−8.54)*** −0.205 (−3.22)***
Dum_recovery −0.213 (−13.98)*** −0.222 (−5.2)***
NPNI −0.002 (−4.54)*** −0.001 (−1.92)* −0.004 (−7.99)***
Lev −0.002 (−1.39) −0.169 (−3.52)*** −0.001 (−0.33)
ROA 0.555 (9.21)*** 0.395 (5.81)*** −0.090 (−1.79)*
Crisis_npni −0.001 (−0.48)
Crisis_roa 0.542 (1.69)*
Crisis_lev −0.006 (−0.05)
Rec_npni −0.002 (−2.64)***
Rec_roa 0.963 (5.24)***
Rec_lev −0.022 (−0.29)
MB −0.225 (−6.47)*** −0.008 (−3.25)***
Crisis_mb 0.007 (1.53)
Rec_mb 0.009 (2.72)***
Size 0.041 (9.96)*** 0.039 (9.31)*** 0.018 (4.31)***
Age 0.003 (9.03)*** 0.003 (9.12)*** 0.002 (6.85)***
div_dum 0.671 (34.82)*** 0.685 (34.84)*** 0.604 (32.08)***
dum_grp −0.049 (−1.38)
Grp_npni 0.006 (8.28)***
Grp_lev −0.189 (−2.74)***
Grp_roa −0.165 (−1.44)
Grp_mb 0.000 (−0.06)
_cons −0.743 (−18.02)*** −0.756 (−17.43)*** −0.597 (−13.13)***
pseudo R2 0.055 0.097 0.099

Notes: The coefficients of key variables and of their interactions with time dummies and group dummy are shown in column 1, 2 and 3, respectively. *,***Significant at 10 and 1 percent levels, respectively

Tobit regressions results for financial constraints and uncertainty

1 2 3
dum_crisis 0.359 (3.64)*** 0.304 (3.3)***
dum_recovery −0.041 (−0.54) 0.189 (3.93)***
fin_constr −0.312 (−8.28)*** −0.629 (−11.66)***
constr_crisis 0.0471 (4.62)***
constr_rec. 0.0140 (1.8)*
cash_vol 0.022 (0.82) −0.100 (−3.48)***
cashvol_crisis −0.414 (−4.62)***
cashvol_rec. −0.376 (−7.73)***
dum_grp −0.068 (−0.79)
grp_cashvol −0.149 (−3.48)***
grp_constr −0.018 (−1.77)*
Size 0.549 (10.51)*** 0.087 (14.19)*** 0.990 (13.31)***
Age 0.003 (9.6)*** 0.003 (6.42)*** 0.003 (7.87)***
div_dum 0.555 (27.35)*** 0.712 (28.92) *** 0.657 (26.56)***
_cons −2.121 (−22.85)*** −1.338 (−31.56) *** −2.928 (−22.77)***
psedoR2 0.0496 0.0489 0.0525

Notes: The coefficients of key variables and of their interactions with time dummies and group dummy are shown in column 1, 2 and 3, respectively. *,***Significant at 10 and 1 percent levels, respectively

Logitregression results on predictors of dividend event and their consistency

All Stable Crisis Recovery
cash_vol −1.678 (−38.46)*** −1.199 (−20.39)*** −0.922 (−4.47)*** −1.591 (−9.49)***
NPNI −0.013 (−12.04)*** −0.013 (−9.77)*** 0.010 (2.49)*** −0.013 (−3.40)***
Lev −0.137 (−2.46)*** −0.767 (−6.74)*** 0.053 (0.40) 0.269 (1.60)
ROA 0.213 (1.74)* 2.698 (11.35)*** −0.732 (−1.81)* −2.299 (−4.57)***
Cash_hold 0.120 (2.81)*** 0.202 (2.94)*** 0.072 (0.94) 0.157 (1.90)*
dum_grp 0.081 (1.85)* 0.012 (0.22) 1.014 (7.73)*** 0.769 (5.35)***
Size 0.887 (4.19)*** 2.044 (8.56)*** −1.433 (−1.37) −3.030 (−2.66)***
Age 0.005 (4.64)*** 0.006 (4.54)**** −0.005 (−1.24) −0.001 (−0.24)
fin_constr 0.218 (1.42) 1.099 (6.36)*** −1.674 (−2.23)*** −2.723 (−3.31)***
_cons −4.670 (−13.11)*** −6.350 (−15.72)*** −3.087 (−1.68) 1.472 (0.75)
Pseudo R2 0.380 0.249 0.245 0.266

Notes: The coefficients of key variables during overall period and across sub-periods (stable, crisis and recovery) are reported. *,***Significant at 10 and 1 percent levels, respectively

References

Aharony, J. and Swary, I. (1980), “Quarterly dividend and earnings announcements and stockholders’ returns: an empirical analysis”, The Journal of Finance, Vol. 35 No. 1, pp. 1-12.

Aivazian, V., Booth, L. and Cleary, S. (2003), “Do emerging market firms follow different dividend policies from US firms?”, Journal of Financial Research, Vol. 26 No. 3, pp. 371-387.

Allen, F. and Michaely, R. (2003), “Payout policy”, Handbook of the Economics of Finance, Vol. 1, pp. 337-429.

Amidu, M. and Abor, J. (2006), “Determinants of dividend payout ratios in Ghana”, The Journal of Risk Finance, Vol. 7 No. 2, pp. 136-145.

Baker, H.K. and Powell, G.E. (1999), “How corporate managers view dividend policy”, Quarterly Journal of Business and Economics, pp. 17-35.

Baker, H.K., Farrelly, G.E. and Edelman, R.B. (1985), “A survey of management views on dividend policy”, Financial Management, pp. 78-84.

Barclay, M.J., Smith, C.W. and Watts, R.L. (1995), “The determinants of corporate leverage and dividend policies”, Journal of Applied Corporate Finance, Vol. 7 No. 4, pp. 4-19.

Berger, A.N., Klapper, L.F. and Udell, G.F. (2001), “The ability of banks to lend to informationally opaque small businesses”, Journal of Banking & Finance, Vol. 25 No. 12, pp. 2127-2167.

Bernheim, B.D. (1991), “Tax policy and the dividend puzzle”, Rand Journal of Economics, Vol. 22, pp. 455-476.

Bhattacharya, S. (1979), “Imperfect information, dividend policy, and the bird in the hand fallacy”, The Bell Journal of Economics, pp. 259-270.

Black, F. (1976), “The dividend puzzle”, The Journal of Portfolio Management, Vol. 2 No. 2, pp. 5-8.

Brav, A., Graham, J.R., Harvey, C.R. and Michaely, R. (2005), “Payout policy in the 21st century”, Journal of Financial Economics, Vol. 77 No. 3, pp. 483-527.

Chay, J.B. and Suh, J. (2009), “Payout policy and cash-flow uncertainty”, Journal of Financial Economics, Vol. 93 No. 1, pp. 88-107.

Chen, D., Li, S., Xiao, J.Z. and Zou, H. (2014), “The effect of government quality on corporate cash holdings”, Journal of Corporate Finance, Vol. 27, pp. 384-400.

Dasilas, A. and Leventis, S. (2011), “Stock market reaction to dividend announcements: evidence from the Greek stock market”, International Review of Economics & Finance, Vol. 20 No. 2, pp. 302-311.

DeAngelo, H., DeAngelo, L. and Skinner, D.J. (2004), “Are dividends disappearing? Dividend concentration and the consolidation of earnings”, Journal of Financial Economics, Vol. 72 No. 3, pp. 425-456.

DeAngelo, H., DeAngelo, L. and Stulz, R.M. (2006), “Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory”, Journal of Financial Economics, Vol. 81 No. 2, pp. 227-254.

Denis, D.J. and Osobov, I. (2008), “Why do firms pay dividends? International evidence on the determinants of dividend policy”, Journal of Financial Economics, Vol. 89 No. 1, pp. 62-82.

Fama, E.F. and French, K.R. (2001), “Disappearing dividends: changing firm characteristics or lower propensity to pay?”, Journal of Financial Economics, Vol. 60 No. 1, pp. 3-43.

Ferreira, M.A. and Vilela, A.S. (2004), “Why do firms hold cash? Evidence from EMU countries”, European Financial Management, Vol. 10 No. 2, pp. 295-319.

Ferris, S.P., Sen, N. and Unlu, E. (2009), “An international analysis of dividend payment behavior”, Journal of Business Finance & Accounting, Vol. 36 Nos 3‐4, pp. 496-522.

Floyd, E., Li, N. and Skinner, D.J. (2015), “Payout policy through the financial crisis: The growth of repurchases and the resilience of dividends”, Journal of Financial Economics, Vol. 118 No. 2, pp. 299-316.

Ghosh, S. (2016), “Small business, lending relationships and crisis: evidence from Indian micro data”, Decision, Vol. 43 No. 1, pp. 1-15.

Gopalan, R., Nanda, V. and Seru, A. (2014), “Internal capital market and dividend policies: evidence from business groups”, The Review of Financial Studies, Vol. 27 No. 4, pp. 1102-1142.

Gopalan, R., Nanda, V.K. and Seru, A. (2007), “Do business groups use dividends to fund investments?”, available at: https://ssrn.com/abstract=967328 http://dx.doi.org/10.2139/ssrn.967328 (accessed July 5, 2017).

Gordon, M.J. (1962), The Investment, Financing, and Valuation of the Corporation, RD Irwin.

Graham, B. and Dodd, D. (1940), Security Analysis (The Classic 1940 Second Edition), McGraw-Hill.

Jensen, M.C. (1986), “Agency costs of free cash flow, corporate finance, and takeovers”, The American Economic Review, Vol. 76 No. 2, pp. 323-329.

John, K. and Williams, J. (1985), “Dividends, dilution, and taxes: a signalling equilibrium”, The Journal of Finance, Vol. 40 No. 4, pp. 1053-1070.

Karpavičius, S. (2014), “Dividends: relevance, rigidity, and signaling”, Journal of Corporate Finance, Vol. 25, pp. 289-312.

Khanna, T. and Palepu, K. (1997), “Why focused strategies may be wrong for emerging markets”, Harvard Business Review, Vol. 75 No. 4, pp. 41-48.

Khanna, T. and Palepu, K. (2000), “Is group affiliation profitable in emerging markets? An analysis of diversified Indian business groups”, The Journal of Finance, Vol. 55 No. 2, pp. 867-891.

Koo, D.S., Ramalingegowda, S. and Yu, Y. (2017), “The effect of financial reporting quality on corporate dividend policy”, Review of Accounting Studies, pp. 1-38.

Kumar, P. and Lee, B.S. (2001), “Discrete dividend policy with permanent earnings”, Financial Management, pp. 55-76.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. (2000), “Investor protection and corporate governance”, Journal of Financial Economics, Vol. 58 No. 1, pp. 3-27.

Larkin, Y., Leary, M.T. and Michaely, R. (2016), “Do investors value dividend-smoothing stocks differently?”, Management Science.

Li, K. and Zhao, X. (2008), “Asymmetric information and dividend policy”, Financial Management, Vol. 37 No. 4, pp. 673-694.

Lintner, J. (1956), “Distribution of incomes of corporations among dividends, retained earnings, and taxes”, The American Economic Review, Vol. 46 No. 2, pp. 97-113.

Lloyd, W.P., Jahera, J.S. and Page, D.E. (1985), “Agency costs and dividend payout ratios”, Quarterly Journal of Business and Economics, pp. 19-29.

Loderer, C.F. and Mauer, D.C. (1992), “Corporate dividends and seasoned equity issues: an empirical investigation”, The Journal of Finance, Vol. 47 No. 1, pp. 201-225.

Manos, R., Murinde, V. and Green, C.J. (2007), “Leverage and business groups: evidence from Indian firms”, Journal of Economics and Business, Vol. 59 No. 5, pp. 443-465.

Manos, R., Murinde, V. and Green, C.J. (2012), “Dividend policy and business groups: evidence from Indian firms”, International Review of Economics & Finance, Vol. 21 No. 1, pp. 42-56.

Miller, M.H. and Rock, K. (1985), “Dividend policy under asymmetric information”, The Journal of Finance, Vol. 40 No. 4, pp. 1031-1051.

Nissim, D. and Ziv, A. (2001), “Dividend changes and future profitability”, The Journal of Finance, Vol. 56 No. 6, pp. 2111-2133.

Rozeff, M.S. (1982), “Growth, beta and agency costs as determinants of dividend payout ratios”, Journal of financial Research, Vol. 5 No. 3, pp. 249-259.

Vogt, S.C. (1994), “The cash flow/investment relationship: evidence from US manufacturing firms”, Financial Management, pp. 3-20.

Von Eije, H. and Megginson, W.L. (2008), “Dividends and share repurchases in the European Union”, Journal of Financial Economics, Vol. 89 No. 2, pp. 347-374.

Walter, J.E. (1963), “Dividend policy: its influence on the value of the enterprise”, The Journal of Finance, Vol. 18 No. 2, pp. 280-291.

Further reading

Chen, Q., Chen, X., Schipper, K., Xu, Y. and Xue, J. (2012), “The sensitivity of corporate cash holdings to corporate governance”, The Review of Financial Studies, Vol. 25 No. 12, pp. 3610-3644.

Miller, M.H. and Modigliani, F. (1961), “Dividend policy, growth, and the valuation of shares”, The Journal of Business, Vol. 34 No. 4, pp. 411-433.

Corresponding author

Rajesh Pathak can be contacted at: rajesh.pathak84@gmail.com