Agency Theory, Corporate Governance and Dividend Payout … Policy and Governance in... · Agency Theory, Corporate Governance and Dividend ... linear relationship between dividend

Agency Theory, Corporate Governance and Dividend

Payout in New Zealand

Travis Brown Department of Accountancy and Finance

University of Otago

Helen Roberts† Department of Accountancy and Finance

University of Otago

15 December, 2016

JEL Classification: G34, G353

Keywords: Corporate governance, Agency theory, Dividend payout,

____________________ †Footnote the corresponding author and also any acknowledgements from presentations and input that other seminar participants have made.

Abstract

This paper examines the effect of internal corporate governance on the likelihood and the level

of dividend payout in New Zealand. We find a positive relationship between the level of

dividend payout and internal corporate governance. Interestingly, our results also support a non-

linear relationship between dividend policy and beneficial director ownership levels. We find

evidence to suggest that lower dividend payout coincides with decreasing agency costs arising

from increased director-shareholder alignment for beneficial director share ownership less than

26%. The relationship becomes negative when ownership exceeds 56%. The high levels of

ownership invoke managers to invest surplus cash to maximize firm value, with dividend policy

acting as a substitute for director oversight. Entrenchment exists for ownership levels between

26% and 56%. We also find that external ownership combined with strong board governance

may mitigate agency problems.

1. Introduction

Given New Zealand’s (NZ) climate of strong investor protection, we examine if there is still a

reliance on dividends to substitute for poor corporate governance.1 This study investigates

internal corporate governance and the consequential effects upon the dividend payout levels of

NZ firms. Consistent with Harford and Maxwell (2012) and Agrawal (2009), we report that

investors do not need to rely on dividends to compensate for a lack of good corporate

governance. Specifically, higher levels of dividends occur in conjunction with stronger board

governance and a less influential CEO. This results contrasts with earlier work by John and

Knyazeva (2006) who report that dividends act as a substitute for weak governance.

While corporate governance as a determinant of dividend policy is documented in the

literature (John and Knyazeva, 2006; Rozeff, 1982) this relationship has not been examined in

the NZ market. In particular, given the openness, relative efficiency, and free flow of

information and capital flows of the NZ stock market, it provides a unique diversification

opportunity for international investors seeking dividend income (Tajaddini, Crack and Roberts,

2015). It is worth noting that the NZ corporate finance environment is regulated by the NZ

Stock Exchange and the Financial Markets Authority. CEOs of listed firms may also be

directors, however following the introduction of the Best Practice Code in 2004, CEO-Chair

duality is no longer permitted. This contrasts with publicly listed firms in the US where the CEO

is also a director. CEO board involvement in NZ is quite heterogeneous with some CEOs

choosing to have no board involvement at all. Mean CEO board participation for our sample is

66%. The variation in CEO board involvement provides a richer dataset for testing the

association between governance and dividend policy.

Internal governance across NZ listed firms may also vary. The small population means there

is a limited pool of director/managerial talent and while directors have roughly the same number

of directorships as other comparable countries, they have a higher proportion of unlisted

directorships (Boyle and Ji, 2013). Given the higher average number of firms that directors need

to monitor, the time and effort given to each directorship may be less. Conflicts of interest can

also arise due to cronyism between directors and CEOs (Ryan and Wiggins, 2004). Collegial

boardroom pressure and an expectation of maintaining the status can also lead to complacency

and a willingness to simply approve majority recommendations (Main, O’Reilly and Wade, 1995).

This behavior can lead to boards becoming more even more influenced by CEO

1 Porta et al (1999) found New Zealand was one of many countries that exhibited strong investor protection - proxied by an anti-director index

recommendation, including the approval of dividend decisions that are sub-optimal for investors.

Such decisions may be associated with weak internal governance and compromise shareholder

value.

2. Background material and extant literature

Dividends have been explained as a means to communicate information to shareholders or to

satisfy payout demands from heterogeneous dividend clienteles (Allen and Michaely, 2003).

However, further work by DeAngelo, DeAngelo and Skinner (2004) questions both the signaling

and the clientele considerations, after reporting dividends being paid by a small number of large

US firms. DeAngelo and DeAngelo (2006) then propose an alternative view whereby the need to

distribute the firm’s free cash flow drives the optimal payout policy. Incorporating agency theory

(Jensen, 1986) with evolution in the firm’s investment opportunity set (Fama and French, 2001;

Grullon, Michaely and Swaminathan, 2002) they propose a life-cycle theory where firms alter

dividends through time. Evidence consistent with the life-cycle approach confirms a positive

relation between the propensity to pay dividends and the ratio of retained earnings to total equity

(DeAngleo, DeAngelo and Stulz, 2006). However, none of these theories can explain the

reported decline in the residual propensity to pay dividends in the US (Fama and French, 2001).

Denis and Osobov (2008) provide additional evidence that the mix of earned to contributed

equity determines dividend policy for developed markets around the world but does not support

signaling as a determinant of dividend policies.2 In particular, the concentration of dividends

among the largest, most profitable firms is consistent with the life-cycle theory prediction but

contrary to the signaling and clientele explanations. We extend this literature by examining the

effect of internal and external corporate governance on the likelihood and level of dividend

payout in NZ. Agency theory suggests that firms pay dividends to mitigate the costs associated

with suboptimal managerial behavior. Dividends reduce firm free cash flow, restricting cash

available for discretionary spending by managers (Jensen, 1986). Firms with high managerial

ownership experienced increased alignment, lowering agency costs (Jensen and Meckling, 1976).

Dividends also force managers to raise investment capital in the external markets, subjecting

managers to additional monitoring by investors and investment banks.

2 The study uses data from Worldscope over the period 1989 – 2002 but due to data limitations Denis and Osobov (2008) only performs tests using dividend data for the United States, Canada, the United Kingdom, Germany, France and Japan.

Corporate governance may also affect the alignment of shareholder-manager interests. Internal

governance refers to the composition of the board and firm ownership. External corporate

governance refers to the managerial labour market and the market for corporate control

(takeover market). Corporate governance can be viewed a substitute device for the monitoring

aspect of dividend policy (Rozeff, 1982).

Dividends may act as a substitute for good corporate governance because cutting dividends is

costly for the firm (Jensen, 1986), dividends require effort to be maintained, and they create an

effective pre-commitment policy. Two competing hypotheses concerning the relation between

agency theory and dividends exist (La Porta et al., 2000). Under the outcome hypothesis small

shareholders under strong investor protection may pressure managers to pay out earnings. This

implies that better corporate governance is associated with higher levels of dividends. Evidence

of a positive association between investor protection and dividends supports this view (Agrawal

2009). Conversely, the substitution hypothesis proposes that dividends reduce agency costs and

require more frequent external financing. The benefits of decreased agency costs are traded off

against the monitoring costs due to the increased use of capital markets (Rozeff, 1982). Chae et

al. (2009) show that firms with lower external financing constraints tend to increase the payout

ratio and experience an improvement in their corporate governance. They find that governance

is reversed depending on the relative sizes of agency and external financing costs. John and

Knyazeva (2006) report that if dividend policy is set to maximize shareholder value, dividend

payout should decrease with the strength of corporate governance.

Good governance limits the potential for managers to act sub-optimally, lowering agency

costs and the dividends required to mitigate them. The empirical literature also supports two

alternative results. First, managers may become insulated from internal disciplining mechanisms

when there is higher insider (executive) ownership or weak corporate governance (Farinha,

2003). Entrenched managers can cause deviations in optimal payout policy by preferring to pay

lower dividends to avoid raising funds in the external capital markets and outside monitoring. If

dividend policy is used to monitor entrenchment-related agency costs, a positive relation

between insider ownership and dividend policy is observed (Weston, 1979; McConnel and

Servaes, 1990). A negative relation implies that ownership and payout are substitute mechanisms

to reduce suboptimal behavior (Hu and Kumar, 2004). Farinha (2003) identifies a U-shaped

association characterized by a critical level of entrenchment for managerial holdings at 30%.

Further evidence endorsing dividend policy as a disciplining mechanism in countries with

different legal systems and distinct agency problems shows the relation between dividends and

insider ownership following the pattern negative-positive-negative in firms with an Anglo-Saxon

tradition (Farinha et al., 2009). Second, institutional ownership may force firms to pay higher

dividends (Short et al, 2002). This result is consistent with both the agency perspective and

institutional investors preferring to free ride on other external monitoring activities. Conversely,

Zheckhauser and Pound (1990) report that institutional ownership is an effective monitoring

mechanism. Overall the evidence shows that insider ownership can align shareholder and

manager interests up to a critical ownership level. Additional evidence shows that board

independence affects dividend policy but is not a substitute control mechanism (Young, 2000).

3. Data and Sample Selection

We obtain board and ownership data and financial data from company annual reports. Market

capitalization, data for dividend policies, earnings, sales, takeover bids and substantial security

holder ownership is sourced from disclosures on the NZX Research database which provides

information on publicly listed NZ firms. The sample includes all firms listed on the NZX from

2004 – 2012 for which data on corporate governance and dividends are available. Financial and

regulated utility firms have been excluded. Firms with a negative dividend payout ratio are also

excluded. The definitions of all variables are given in Table 1. The summary statistics for the 563

firm-year observations in the sample are reported in Table 2.

[Insert Table 1 here]


The data in Table 2 shows that 80% of the firms in the sample pay a dividend and the mean

dividend payout ratio is 52.7% for the three-year average payout period ad 49.0% for the five-

year average payout period. The maximum dividend payout for both these periods is greater than

one. This is a reasonably common observation in NZ and is predominantly the result of unusual

and excessively low earnings while keeping dividends constant. The CEO is a member of his

own compensation committee 11% of the time and CEO board members account of 66.4% of

the sample. CEO Chairman is quite uncommon, making up only 1.8% of the sample. Average

board size is six directors. Busy directors account for 61% of the sample. Inside and block

owners make up 17.1% and 27.7%, respectively. Mean institutional ownership is quite small at

only 5.7%. Mean director share ownership is 15.7% while combined block and institutional

owners account for a mean of 33.4% of shares held. CEOIND has a mean value of 0.736,

ranging from 0 to 1. On average CEOs appear to be less influential and therefore represent a

higher level of internal governance for firms. Board influence, BD_IND takes a mean value of

0.431. The lower measure indicates that there may be traits associated with weak governance in

relation to a firm’s board of directors. The mean market-to-book ratio is 2.32; the mean growth

in sales is 12.6%. These values suggest that firms will need to access the capital markets regularly

to sustain the current growth trajectory. The low value for the CORP variable, mean = 0.2149,

indicates there is very little external disciplinary effect by the way of takeover threats for the

sample.

Table 3 reports the correlation coefficients between the independent variables. The positive

correlation between the propensity to pay (DIV) and board size is consistent with the positive

correlation between firm size and board size (0.56). There is also evidence that the likelihood of

paying out dividends decreases when a firm has higher growth opportunities. The opportunity

cost of paying dividends will be higher in this case and the cash may have a higher value if is

reinvested into the business. The corporate governance proxies indicate that if the directors are

busy and there is a lack of institutional ownership, the firm is more likely to pay a dividend as

summarized by BD_IND. The negative correlation between CEO_IND and the likelihood of

paying a dividend implies that as CEO influence on the board decreases, the level of dividend

increases.


Table 4 reports the summary statistics and the result of the t-tests for differences in the means

across varying degrees of institutional (INST) and block ownership (BLOCK). The varying

degrees of external ownership are compared to the base case of both blockholders and

institutional holders present. The propensity to pay dividends ranges from 77.09% (both block

and institutional holders) to 83.3% (only institutional holders). This difference is statistically

significant. Mean director beneficial ownership (BODOWN) is lower for firms with only

institutional shareholdings (4.76%) compared to those with only blockholdings (18.56%). Board

governance (BD_IND) is higher for firms with no block or institutional holdings. Institutional

ownership is significantly higher when there are no blockholdings and vice versa.


4. Methodology

4.1 Hypothesis development

Firms can alter the governing board and the ownership structure to suit their specific needs.

Variation in firm ownership structure and board composition may be related to different levels

of agency costs. These decisions make up the internal governance of the firm. We examine the

relation between internal governance and dividend policy.

Research Question I: Does internal corporate governance affect the dividend decisions of NZ

publicly listed firms?

Research Question II: Is there an empirical relationship between the strength of internal

corporate governance and the level of dividend payout for NZ publically listed firms?

Firms with weaker governance are more likely to pay higher levels of dividends as they are

subject to higher agency costs (John and Knyazeva, 2006). Higher dividend payments act as a

substitute for good governance. Firms with stronger governance will be less likely to pay

dividends (John and Knyazeva, 2006). We expect to find a negative relation between the level of

dividends and the strength of internal governance.

Hypothesis Ia: Firms with weaker levels of internal corporate governance will exhibit a greater

tendency to pay dividends than firms with stronger internal corporate governance.

Rozeff (1982) reports a positive relation between dividend payout and agency costs. Higher

agency costs associated with weaker internal governance are mitigated by higher dividend payout.

Dividends can compensate for a lack of monitoring, discipline and misalignment of interests that

are evident in firms with weak governance.

Hypothesis IIa: As a firm’s internal governance increases, the level of dividend payout

decreases.

The managerial entrenchment hypothesis refers to managers in positions that are relatively safe

and secure. There is little threat of takeover and these firms typically have weaker governance

embodied by a lack of monitoring by blockholders and institutions. Entrenched managers may

be less willing to pay dividends for fear of scrutiny by capital markets. However, a firm with

strong corporate governance will, in theory, pay more dividends. This positive relation between

managerial entrenchment and dividend payout has been documented in several studies including

Farinha (2003), Brown and Caylor (2004), and Kowaleski et al. (2007). Agency costs may also

increase with greater entrenchment exposure. A dividend policy can help to alleviate these costs.

Dividend payout theory states that the relation between internal governance and propensity to

pay dividends should be positive. This leads to the following hypotheses:

Hypothesis Ib: Firms with weaker levels of internal governance will exhibit less of a tendency

to pay dividends than firms with stronger internal governance.

Hypothesis IIb: As a firm’s internal corporate governance increases, the level of dividend

payout increases.

The external corporate governance environment may also impact these hypotheses. In support

of Hypothesis Ia and Hypothesis IIa, the transparent nature of the managerial supply market

means that there could be a stronger disciplinary effect present. Managers may feel more

compelled to act in the best interests of shareholders and payout dividends when the internal

corporate governance is weak. Alternatively, because the managerial supply market in NZ is

smaller compared to other developed countries, there may be an entrenchment effect. In this

case managers may feel more secure in their respective roles and seek to maximize personal

interests. This view is consistent with the non-optimal dividend payout predicted in Hypothesis

Ib and Hypothesis IIb.

4.2 General Model Specification

There are two parts to the analysis. The first relates to research question one, the second to

research question two.

Research Question 1.

We propose the following logit model to examine whether internal corporate governance affects

the dividend decisions of NZ listed firms. The model explores the determinants of the

probability that a firm pays a positive dividend against the probability that a firm pays no

dividend.

ln � ��

=

��+��_��+��_��+�� !�+�"��#�� +$%��&�'()��+$*'�(�+$+��',�+$- �.��+$/0��+$��,)'�#!1�+$�� ,�' � +2� (1)

Research Question 2.

We employ the following OLS multivariate regression model to investigate the relationship

between the strength of corporate governance and the level of dividend payout for NZ publicly

listed firms.

��&,(3� =$�+$��_��+$��_��+$�� !�+$"��#�� +$%��&�'()��+$*'�(�+$+��',�+$- �.��+$/0��+$��,)'�#!1�+$�� ,�' � +2� (2)

4.3 Dependent Variables

DIVPAY is the ratio of total annual ordinary cash dividends to normalized after-tax earnings.

DIVPAY is zero when no dividends are paid and one in the case of 100% of after-tax earnings

are paid out in the form of dividends. DIVPAY is truncated below zero to account for negative

dividend payout. Consistent with Chae, Kim and Lee (2009), the results include those cases

when the dividend payout ratio exceeds one. Bae, Chang and Kang (2012) find evidence that NZ

has the highest dividend payout on average when compared to 33 other countries. In order to

correctly proxy for a firm’s target payout ratio special dividends and imputation credits are

excluded. One-off income statement items which may misattribute earnings are also avoided by

using normalized earnings. Including an average of DIVPAY in the form of DIVPAY3 and

DIVPAY5 also reduces possible noise within the data.

4.4 Explanatory Variables

Ownership Structure

Three variables BODOWN, INST and BLOCK are used to determine the effect of different

ownership structures on dividend policy and internal governance relation. BODOWN is

expected to be positively associated with stronger monitoring of management. Dividend policy is

expected to be less important in reducing agency costs as BODOWN increases. INST and

BLOCK are external ownership variables. A large block holder will be more incentivized to

monitor the performance and activities of the company because of the size of the investment.

Reddy et al. (2010) report strong evidence of increased monitoring at the level of blockholdings

increases and an associated reduction in agency costs (similar to Shleifer and Vishny, 1988).

INST should increase the level and quality of firm monitoring. Crane et al. (2016) report that

high institutional ownership significantly increases the dividends paid. The effect is stronger for

firms with high agency costs.

4.5 Internal Governance Indices

Original Indices

We construct a corporate governance index to test the effect of corporate governance on

dividend policy for publically listed NZ firms. The index helps to prevent variable interactions

from distorting results.3 CEO_IND uses CEOREMUM, CEOBOD and CEOCHAIR to

measure the influence a firm’s CEO has upon the board and remuneration committee. BD_IND

represents board size, the number of executive directors and the number of busy directors.

These two indices proxy for the strength of internal governance used in equation (1) and

equation (2). As a check on robustness of our results we include two alternative indices for

CEO_IND and BD_IND. The indices are explained in detail in Table 1.

4.6 Control Variables

The seven control variables are selected to control for any influence that they may have on

dividend policy, or alternatively, they are viewed as proxies for agency costs. PGROWTH is the

five-year geometric mean for past growth in total sales. Firms that have experienced high growth

historically are less willing to pay out dividends compared to firms with lower growth. MB is the

market-to-book ratio of equity. As a firm’s growth opportunities increase, the opportunity cost

of paying out cash to shareholders increases. Firms with higher opportunity costs will have lower

dividend payout on average (Smith and Watts, 1992). LEVERAGE is the ratio of book value of

total debt to book value of total assets. As a firm takes on more leverage the dividend payout will

decrease.4 SIZE is the natural logarithm of firm market capitalization. Larger firms have a higher

dividend yield on average however they also experience higher agency costs. Dividend policy

may be used to align the interests of managers and shareholders. ROA is the three-year mean

ratio of after-tax earnings to total assets. We expect a positive relation between ROA and

dividend payout.

DISPERS is defined as the 100 percentile of holdings minus the top five non-custodial holdings.

Firms with more dispersed shareholders are likely to have greater agency costs and higher

dividend payout to alleviate these costs. CORP measures the number of takeover bids upon a

company in the last five years. The variable proxies for cross-sectional variation between external

disciplining factors and their potential effects on dividend policy.

5. Empirical Results

5.1 Logistic Model Analysis

Table 5 reports the results for the logistic regression on the pooled sample for the period 2004 to

2012 as stated in equation (1). Each model includes year dummies. All the models are statistically

3 John and Knyazeva (2006), Kowalewski et al. (2008), Bae, Chang and Kang (2012) and numerous others report a governance index is optimal to us as a proxy for corporate governance. We model our two indices upon the INT_GOV index used in John and Knyazeva (2006). 4 Smith and Watts (1992) and Crane et al., (2016) hypothesise that as leverage is predominant in larger firms, and larger firms usually have a higher payout, there may be a positive relation between leverage and dividend payout.

significant at the 1% level. The highly significant control variables, LEVERAGE, ROA, SIZE,

MB and PGROWTH are all associated with a firm’s propensity to pay dividends. Models (1) and

(3) report a marginally significant, positive coefficient for the BD_IND variable lending partial

support for Hypothesis Ib. As board governance increases the firm is more likely to pay a

dividend. This is consistent with the managerial entrenchment hypothesis (Farinha, 2003; Brown

and Caylor, 2004).


5.2 Ordinary Least Squares Analysis

Panels A and B of Table 6 report the coefficients for the ordinary least squares regression model

given by equation (2). Each model includes year dummies. All the models are highly significant.

Panels A and B report significant positive coefficients for both CEO_IND and BD_IND.5

These are consistent with firms demonstrating stronger governance pay higher dividends. The

finding is consistent with Hypothesis IIb.


BODOWN, BODOWN2 and BODOWN3 are all significant in the majority of the DIVPAY5

models and marginally significant in the DIVPAY3 models. The signs on the coefficients are

negative, positive and negative, respectively. This result is consistent with work by Morck et al.

(1988), Beiner et al., (2006) and Farinha et al., (2009).6 The estimated turning points in the model

are 25.4% - 26.3% and 53.5% - 55.6%. These findings suggest that for beneficial director share

ownership below 26%, the benefits associated with the alignment to shareholder interests may

exceed the private consumption benefits and shirking behavior of management. Between 26%

and 56% beneficial director ownership the relationship is positive, consistent with managerial

entrenchment. Dividends act as a compensating factor to offset the higher agency costs (Farinha,

2003). Without dividends the agency benefits associated with perquisite consumption and

shirking (that ultimately destroy shareholder wealth) can exceed the agency costs of such actions.

When the level of beneficial director share ownership exceeds 56% the relationship is again

negative. Director and shareholder interests are aligned again and dividend policy is not required

as a monitoring device. The negative relation is consistent with the substitution hypothesis (John

and Knyazeva, 2006). Another explanation is that as managers become more insulated, the

dividend payout and propensity decreases, consistent with agency theory. Paying dividends

5 Similar results are also found using the alternate indices CEO_IND1 and BD_IND1. 6 Farinha et al., (2009) reports that for firms with an Anglo-Saxon tradition the relation between dividends and insider ownership follows the pattern negative-positive-negative.

increases the rate at which the firm must seek capital from the external markets, which also

decreases cash that can be used for private consumption by managers. Evidence from

antitakeover legislation shows that this behavior is more prevalent in poorly governed firms

(Francis et al., 2011).

The significant negative BLOCKINST coefficient indicates that external ownership is an

effective monitor that can reduce the need to pay high dividends (Shleifer and Vishny, 1986).

The negative relationship also provides support for the managerial entrenchment hypothesis

(Morck et al., 1988). The significant positive BLOCKINST*BD_IND interaction term shows

that firms with strong board governance and higher levels of external ownership pay higher

dividends. Consistent with the results from the logit analysis, ROA and PGROWTH are highly

significant. Profitability is associated with higher dividend payout and higher growth prospects

are associated with lower dividends (Smith and Watts, 1992; Rozeff, 1982).

5.3 Robustness Checks

Our results do not adjust for potential heteroscedasticity in the error terms. While graphs of the

residuals do not indicate any deviations from the underlying assumptions for the models we

verify our results using estimations for the regression models that control for heteroscedasticity.

The results are remarkably similar to those already reported. Following Thompson (2011), when

sample sizes are small (such as for our sample), it is possible to find statistically significance even

when it does not exist. Hence we do not include firm level clustering or clustering over time. To

do so would be adding unnecessary bias into the analysis.

We also run an alternative regression model to estimate equation (2) using five principal

components based on the control variables. The results are shown in Table 7. Consistent with

the results reported in Table 6, the CEO_IND and BD_IND coefficients are positive and

significant. However the beneficial director ownership coefficients are only significant for some

of the DIVPAY5 models reported in Panel B of Table 7.The turning points for the models are

between 17.5% - 19% and 55.7% - 58.2%, with the same signs as the estimation in Table 6. The

results support the strong, positive relation between the level of dividend payout and strong

board governance and external ownership.


Finally we run a Tobit regression censored from below to account for possible bias in the

regression model due to the number of firms that do not pay dividends (23% of our sample).

The estimation is reported in Table 8. Consistent with the regression results in Tables 6 and 7

there is evidence of a significant positive relation with CEO_IND and BD_IND. Better

corporate governance in terms of less influential CEOs and better governed boards are

associated with higher levels of dividends. The coefficients for beneficial director share

ownership (BOD_OWN) are in the same direction and highly significant for the DIVPAY5

model, in particular. The turning points are 24.0% - 25.4% and 43.1% - 45.4%.


The BLOCKINST, BLOCKINST*BD_IND, ROA and PGROWTH coefficients are all

consistent with the earlier estimations reported in Table 6. Consistent with Smith and Watts

(1992) and Rozeff (1982), MB is significant and negative. Firms with higher growth

opportunities are likely to pay lower dividends.

6. Conclusion

This study examines the relationship between the propensity to pay and the level of dividend

payout for publicly listed NZ firms. The NZ market provides an interesting, previously untested

setting to explore the governance-dividend payout relationship. The market for corporate control

in NZ is much less active than in overseas markets and, as a result may have a weaker

disciplining effect on managers. In addition, the market for managerial labour is small and highly

transparent. On the one hand, the disciplining mechanism provided by the managerial labour

market may be stronger in such an environment. Conversely, managers may feel secure in their

jobs knowing that the pool for managerial talent is small and restricted.

The NZ market may also have weaker internal governance. There is a small pool of director

talent, and firms often contain complex cross holdings and interlocked directorships. This can

mean that directors serve on multiple boards, which may weaken their capacity to act

independently in the best interests of respective shareholders. Moreover, investments which are

not value maximizing may be endorsed by busy directors who are privy to information from

multiple board meetings and affiliations.

Several legislative changes may have strengthened NZ’s internal governance environment. In

1993, the new Companies Act came into effect, raising the standard of responsibility for

directors. Mandatory public disclosure of CEO compensation was introduced in 1997.

Furthermore, the amendments brought about by the New Zealand Stock Exchange in 2003 were

also designed to improve the quality of corporate governance in NZ publicly listed firms.7 It is

also worth noting that the NZ dividend imputation tax credit system means that, at the intra-

market level, the firm’s dividend policy is not affect by tax. Given New Zealand’s (NZ) climate

of strong investor protection, we examine if there is still a reliance on dividends to substitute for

poor corporate governance.

We find strong evidence of a relation between internal governance and the likelihood of a NZ

publically listed firm paying a dividend. Moreover, there is evidence to suggest that the level of

dividend payout is strongly related to the firm’s internal corporate governance, beneficial director

ownership and being privy to both strong board governance and external ownership. An

important finding in this study is the level of dividends being significantly and non-linearly

related to beneficial director ownership. Consistent with work by Farinha et al., (2009) we report

a negative-positive-negative pattern at director ownership levels of 26% and 56%.8 For beneficial

director ownership less than 26%, there is a negative relation suggesting that decreasing agency

costs from increased director-shareholder alignment offset the need for higher dividend

payments. Beneficial director ownership levels between 26% and 56% exhibit a significant

positive relationship consistent with managerial entrenchment. Higher agency costs associated

with entrenchment can be mitigated through higher levels of dividends (Farinha, 2003).

Beneficial director ownership exceeding 56% results in the relationship becoming significantly

negative again. When directors own higher proportions of a firm’s equity, their alignment with

shareholder wealth increases and there is less need to use dividend policy to align director and

shareholder interests.

Our findings also show that NZ firms with strong board governance and low levels of CEO

influence tend to have higher levels of dividend payout. The propensity to pay dividends is also

marginally related to stronger board governance. External ownership is found to be negatively

related to the level of dividend payout, consistent with the idea that there is a significant

monitoring ability present. According to Morck et al. (1988) the negative relation provides

empirical support for the managerial entrenchment hypothesis. Considering external ownership

and good board governance jointly shows a very strong positive relation with the level of

7 These became mandatory in 2004 under the Best Practice Code. 8 The turning points reported in Farinha and Lopez-de-Foronda (2009) for the Anglo-Saxon sample consisting of the USA, UK and Irish firms were 36% and 95% during the period 1996 – 2000.

dividend payout. These findings indicate that in the presence of strong board governance,

external blockholders have the ability to force higher levels of dividends.

Future research may wish to examine in detail the impact of the Companies Act 1993 and its

various amendments and determine its effect on the dividend propensity and dividend payout

levels in NZ. We report preliminary evidence from this study of a shift towards a stronger

corporate governance environment, but there is no steadfast empirical evidence of a causal link

to the Companies Act 1993. While Hossain et al., (2001) do look at the Act their study examines

the relation between firm performance and the board. It does not investigate dividend policy.

The results of such research will be useful for regulators to use as a measure for both the

effectiveness and the response time of related legislation.

P a g e | 5

References

Agrawal, A. K., 2009. The impact of investor protection law on corporate policy: Evidence from

the Blue Sky laws.

Bae, S. C., Chang, K., & Kang, E. (2012). Culture, corporate governance, and dividend policy:

international evidence. Journal of Financial Research, 35(2), 289-316.

Boyle, G., and Ji, X., 2013. New Zealand corporate boards in transition: Composition, activity

and incentives between 1995 and 2010. Pacific Accounting Review, 25(3), 235-258.

Chae, J., Kim, S., Lee, E. J., 2009. How corporate governance affects payout policy under agency

problems and external financing constraints. Journal of Banking and Finance, 33, 2093-2101.

Crane, A. D., Michenaud, S., & Weston, J. P. (2016). The effect of institutional ownership on

payout policy: Evidence from index thresholds. Review of Financial Studies, forthcoming.

DeAngelo, H., DeAngelo, L., Stulz, R.M., 2006. Dividend policy and the earned/contributed

capital mix: a test of the life-cycle theory. Journal of Financial Economics 81, 227–254.

Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. The Journal of Law and

Economics, 26(2), 301-325.

Fama, E. F., & French, K. R. (2001). Disappearing dividends: changing firm characteristics or

lower propensity to pay? Journal of Financial economics, 60(1), 3-43.

Farinha, J., 2003. Dividend policy, corporate governance and the managerial entrenchment

hypothesis: An empirical analysis. Journal of Business Finance and Accounting, 30(9), 1173-1209.

Farinha, J., Lopez-de-Foronda, O., 2009. The relation between dividends and insider ownership

in different legal systems: international evidence. The European Journal of Finance, 15(2), 169-189.

Hossain, M., Prevost, A. K., & Rao, R. P. (2001). Corporate governance in New Zealand: The

effect of the 1993 Companies Act on the relation between board composition and firm

performance. Pacific-Basin Finance Journal, 9(2), 119-145.

Hu, A., & Kumar, P. (2004). Managerial entrenchment and payout policy. Journal of Financial and

Quantitative Analysis, 39(04), 759-790.

Jo, H., Pan, C., 2009. Why are firms with entrenched managers more likely to pay dividends?

Review of Accounting and Finance, 8(1), 87-116.

Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs

and ownership structure. Journal of Financial Economics, 3(4), 305-360.

Jensen, M. C. (1986). Agency cost of free cash flow, corporate finance, and takeovers. Corporate

Finance, and Takeovers. American Economic Review, 76(2).

John, K., and Knyazeva, A., 2006. Payout policy, agency conflicts, and corporate

governance. Agency Conflicts, and Corporate Governance (May 2006).

Kowalewski, O., Stetsyuk, I., & Talavera, O. (2008). Does corporate governance determine

dividend payouts in Poland?. Post-Communist Economies, 20(2), 203-218.

Main, B. G. M., O’Reilly, C. A., and Wade, J., 1995. The CEO, board of directors and executive

compensation: Economic and psychological perspectives. Industrial and Corporate Change 4(2), 293-

332.

Morck, R., Shleifer, A., & Vishny, R. W. (1988). Management ownership and market valuation:

An empirical analysis. Journal of Financial Economics, 20, 293-315.

Reddy, K., Locke, S., & Scrimgeour, F. (2010). The efficacy of principle-based corporate

governance practices and firm financial performance: An empirical investigation. International

Journal of Managerial Finance, 6(3), 190-219.

Rozeff, M. S. (1982). Growth, beta and agency costs as determinants of dividend payout

ratios. Journal of Financial Research, 5(3), 249-259.

Shleifer, A., & Vishny, R. W. (1986). Large shareholders and corporate control. The Journal of

Political Economy, 461-488.

Short, H., Zhang, H., & Keasey, K. (2002). The link between dividend policy and institutional

ownership. Journal of Corporate Finance, 8(2), 105-122.

Smith, C. W., & Watts, R. L. (1992). The investment opportunity set and corporate financing,

dividend, and compensation policies. Journal of Financial Economics, 32(3), 263-292.

Tajaddini, R., Crack, T. F., & Roberts, H. (2015). Price and Earnings Momentum, Transaction

Costs, and an Innovative Practitioner Technique. International Review of Finance, 15(4), 555-597.

Ryan, H. E., and Wiggins, R. A., 2004. Who is in whose pocket? Director compensation, board

independence, and barriers to effective monitoring. Journal of Financial Economics, 73, 497-524.

Zeckhauser, R. J., & Pound, J. (1990). Are large shareholders effective monitors? An

investigation of share ownership and corporate performance. In Asymmetric Information, Corporate

Finance, and Investment (pp. 149-180). University of Chicago Press.

Table 1. Variable Definitions

Dependent Variables

DIV Dummy variable that takes the value of one if the firm pays a dividend and zero otherwise.

DIVPAY3 The mean ratio of total ordinary cash dividends paid to normalized earnings during the previous three years.

DIVPAY5 The mean ratio of total ordinary cash dividends paid to normalized earnings during the previous five years.

Explanatory Variables

CEOCHAIR Dummy variable that takes the value of one if the CEO is chairman of the board and zero otherwise.

CEOBOD Dummy variable that takes the value of one if the CEO is a member of the board and zero otherwise.

CEOREMUN Dummy variable that takes the value of one if the CEO is a member of the remuneration committee and zero otherwise.

BDSIZE The total number of directors on a firm’s board.

BUSY The percentage of directors with three or more directorships.

INSIDE The percentage of inside director on the board. Inside directors are defined as executive directors.

BLOCK

The common stock ownership of the largest non-institutional blockholder. A non-institutional blockholder is defined as any person(s), company, or other entity (such as a city council or government department) with ownership of at least 5% of total ordinary shares that does not fit the definition of an institutional holder.

INST

The common stock ownership of the largest institutional blockholder. An institution is defined as any investment bank, investment fund, insurance group, or mutual fund likely to actively trade their position with ownership of at least 5% of total ordinary shares.

BLOCKINST The summation of block shareholdings and institutional shareholdings.

BODOWN The total beneficial stock ownership held by directors and associated persons as a proportion of the total value of common equity.

CORP The number of takeover bids on a company in the previous five years.

Variables for BD_IND

INVBDSIZE The maximum number of board directors in the sample minus the actual number of directors for the firm, divided by the maximum number of board directors in the sample

NONINSIDE 1-proportion of inside (INSIDE) directors on the board. NONBUSY 1-proportion of busy (BUSY) directors on the board.

Original Indices

CEO_IND (3-C)/3 where C represents the sum of the dummy variables CEOBOD, CEOREMUN, and CEOCHAIR and is a value between 0 and 1. A higher value of CEO_IND should indicate a potentially less influential CEO.

BD_IND

(12-BD)/12, where BD is the board governance index constructed as follows. Observations are sorted into quartiles for each of the following new variables: INVBDSIZE, NONINSIDE, and NONBUSY. Each firm year observation is assigned a value of 1, 2, 3, or 4 depending on which quartile it ranks in for each variable (1 for the top quartile and 4 for the

lowest quartile). For each firm, the assigned values of each new board structure variable are summed to give the score BD. Higher values of BD_IND indicate better governance.

Alternate Indices

CEO_IND1

This variable takes the value of 1 if the CEO is not a director (CEOBOD=0), and zero otherwise. A value of one for this index indicates potentially stronger governance and a value of zero indicates weaker governance. This is based upon the rationale that a CEO who is a member of a board may have considerable influence over the board and its decisions. CEOREMUN is not included in this definition of governance.

BD_IND1

(300-BD)/300, where BD is the board governance index constructed as follows. Each of the following variables: INBDSIZE, NONINSIDE, and NONBUSY are multiplied by 100. For each firm, BD is assigned the sum of these values. Lower values of BD_IND1 indicate potentially better governance.

Control Variables

LEVERAGE Book value of total debt to book value of total assets.

ROA The three-year mean ratio of after-tax earnings to total assets.

CORP The number of takeover bids for a company in the last five years.

SIZE The natural logarithm of the firm’s market capitalization.

MB The ratio of market-to-book value of equity and is a proxy for growth opportunities.

PGROWTH The five-year geometric mean of past growth in total sales.

DISPERS 100% - the top five non-custodial shareholdings.

Table 2. Summary Statistics

This table reports the mean, median, maximum and minimum values for each of the variables used in the study. All of the variables are defined in Table 1. The pooled sample contains 563 firm-year observations.

Observations Mean Median Std. Dev. Min Max

Dependent Variables DIV 563 0.799 1.000 0.401 0.000 1.000

DIVPAY3 536 0.526 0.525 0.420 0.000 1.414

DIVPAY5 540 0.490 0.477 0.380 0.000 1.210

Main Explanatory Variables

CEOREMUN 563 0.110 0.000 0.313 0.000 1.000

CEOBOD 563 0.664 1.000 0.473 0.000 1.000

CEOCHAIR 563 0.018 0.000 0.132 0.000 1.000

BDSIZE 563 6.073 6.000 1.569 3.000 12.000

BUSY 563 0.614 0.667 0.300 0.000 1.000

INSIDE 563 0.171 0.167 0.149 0.000 0.667

BLOCK 563 0.277 0.191 0.252 0.000 0.970

INST 563 0.057 0.000 0.087 0.000 0.805

BLOCKINST 563 0.334 0.264 0.238 0.000 0.970

BODOWN 563 0.157 0.037 0.213 0.000 0.919

Internal Governance Indices

CEO_IND 563 0.736 0.667 0.217 0.000 1.000

CEO_IND1 563 0.336 0.000 0.473 0.000 1.000

BD_IND 563 0.431 0.417 0.151 0.083 0.750

BD_IND1 563 0.454 0.475 0.122 0.091 0.756

Control Variables

LEVERAGE 563 0.442 0.411 0.343 0.000 1.410

ROA 563 0.008 0.051 0.155 -0.507 0.161

CORP 563 0.215 0.000 0.560 0.000 3.000

SIZE 563 11.730 11.778 1.715 7.219 16.281

MB 563 2.319 1.653 2.003 0.460 8.216

PGROWTH 563 0.126 0.062 0.242 -0.169 0.939

DISPERS 563 0.599 0.655 0.266 0.046 0.990

Table 3. Pearson Correlation Coefficients The sample includes all NZ publicly listed firms from 2004-2012 based on available data. The variables are defined as in Panel A and in the Appendices. Pairwise correlations that are significant at the 5% level have been identified by **. The correlations are based on 563 observations.

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

[1] CEOREMUN 1.00

[2] CEOBOD 0.25** 1.00

[3] CEOCHAIR 0.00 0.10** 1.00

[4] BDSIZE -0.07 0.14** -0.03 1.00

[5] BUSY -0.05 -0.119** 0.03 0.11** 1.00

[6] INSIDE 0.13** 0.61** 0.06 -0.01 -0.13** 1.00

[7] BLOCK 0.07 -0.26** 0.03 -0.14** 0.06 -0.01 1.00

[8] INST -0.07 0.00 -0.07 0.05 0.09** -0.02 -0.32** 1.00

[9] BLOCKINST 0.04 -0.27** 0.01 -0.13** 0.10** -0.02 0.94** 0.02 1.00

[10] BODOWN -0.11** 0.08 0.12** -0.15** -0.16** 0.25** 0.09** -0.03 0.08** 1.00

[11] CEO_IND -0.66** -0.86** -0.27** -0.06 0.11** -0.52** 0.15** 0.05 0.18** -0.03 1.00

[12] CEO_IND1 -0.25** -1.00 -0.10** -0.14** 0.12** -0.61** 0.26** 0.00 0.27** -0.08 0.86** 1.00

[13] BD_IND 0.00 -0.40** -0.06 -0.59** -0.50** -0.48** 0.06 -0.08 0.04 -0.02 0.30** 0.40** 1.00

[14] BD_IND1 -0.02 0.21** 0.04 0.49** 0.82** 0.29** -0.01 0.09** 0.02 -0.09** -0.15** -0.21** -0.85** 1.00

[15] LEVERAGE 0.00 0.03 0.00 -0.06 0.14** 0.01 -0.18** 0.08 -0.17** 0.03 -0.02 -0.03 -0.03 0.09** 1.00

[16] ROA 0.04 0.15** -0.07 0.27** 0.28** 0.11** 0.18** -0.01 0.18** 0.00 -0.11** -0.15** -0.34** 0.38** -0.11** 1.00

[17] CORP -0.13** -0.18** 0.01 0.06 0.27** -0.11** 0.13** 0.11** 0.18** 0.01 0.19** 0.18** -0.14** 0.21** -0.06 0.07 1.00

[18] SIZE 0.02 0.17** -0.04 0.56** 0.29** -0.01 -0.05 0.11** -0.02 -0.26** -0.12** -0.17** -0.41** 0.46** -0.09** 0.41** 0.11** 1.00

[19] MB -0.02 0.16** 0.06 -0.06 -0.19** 0.11** -0.25** 0.03 -0.25** 0.01 -0.12** -0.16** 0.03 -0.14** 0.01 -0.29** -0.12** 0.13** 1.00

[20] PGROWTH 0.07 0.16** 0.23** -0.10** -0.10** 0.17** -0.12** 0.01 -0.12** 0.24** ,-0.19** -0.16** -0.01 -0.05

0.00 -0.26** -0.08** -0.11** 0.29** 1.00

[21] DISPERS 0.09** 0.35** -0.01 0.08 -0.03 0.16** -0.75** 0.15** -0.74** -0.18** -0.29** -0.35** -0.12** 0.07 0.19** -0.07 -0.13** 0.15** 0.22** 0.06 1.00

Table 4. Beneficial Director, Institutional and Blockholder Shareholder Statistics In this table we report the statistics and results of a t test for difference of means for the breakdown of firms where there are only blockholders (BLOCK>0, INST=0), only institutional holders (BLOCK=0, INST>0), blockholders and institutional holders (BLOCK>0, INST>0), and where there are neither (BLOCK=0, INST=0). For Panel A B and C; the significance is in regards to the base case of where there are both blockholders and institutions present. For Panel D; the significance is in regards to the difference on Panel B and Panel C. We report figures for SIZE (natural log of market capitalisation), DIV (proportion of firms paying a dividend), the mean payout of DIVPAY3 and DIVPAY5, BODOWN (proportion of shares held beneficially by directors), CEO_IND (ownership level of CEO), BD_IND (board governance), and the holdings of both Institutional (INST) and blockholders (BLOCK). *** Indicates Significance at 1%, ** Indicates Significance at 5%, * Indicates Significance at 10%

Panel A: All Firms

Panel B: Dividend Paying Firms

Panel C: Non-Dividend Paying Firms

Panel D: Difference of Means

Block=0 INST=0 (N=17)

Block=0 INST>0 (N=78)

Block>0 INST=0 (N=289)

Block>0 INST>0 (N=179)













SIZE 11.77 12.94 11.26 11.96 11.92 13.27*** 11.45*** 12.45 11.07 11.28** 10.47 10.31 0.85 1.99*** 0.97*** 2.15***

DIV = 1 (%) 82.35 83.33** 80.62 77.09

Mean DIVPAY3 0.34 0.56* 0.56** 0.47 0.44** 0.66 0.73*** 0.63

Mean DIVPAY5 0.33 0.49 0.53*** 0.44 0.41** 0.58 0.68*** 0.59

BODOWN (%) 6.78* 4.76*** 18.56 16.57 4.29** 4.68*** 16.35 15.55 18.43 5.19** 27.74* 19.98 -14.14*** -0.51 -11.39*** -4.43

CEO_IND 0.75 0.63*** 0.74 0.77 0.71 0.61*** 0.75 0.77 0.89 0.72 0.73* 0.79 -0.17 -0.11** 0.05 -0.02

BD_IND 0.51*** 0.37** 0.46*** 0.41 0.51*** 0.35** 0.45*** 0.39 0.53 0.45 0.50 0.49 -0.02 -0.10*** -0.05** -0.10***

INST (%)

15.23***

11.31 12.54***

10.10 28.68***

15.39

-16.14***

-5.29***

BLOCK (%)

38.08*** 25.49

42.49*** 25.75

19.73* 24.62

22.76*** 1.13

BLOCKINST (%)

15.23*** 38.08 36.81 12.54*** 42.49*** 35.86 28.68* 19.73*** 40.01 -16.14*** 22.76*** -4.15

Table 5. Logistic Model This table reports the regression output for the logit models fitted over the data for the pooled data sample of 563 observations from 2004 to 2012. The coefficients are reported first with the corresponding standard error in grey beneath each estimate. The control variables are PGROWTH = the five-year geometric mean of past growth in total sales, MB = the ratio of the market-to-book value of equity and is a proxy for growth, LEVERAGE = the ratio of the book value of total debt to the book value of total assets, SIZE = the natural logarithm of market capitalization, ROA = the three-year mean ratio of after-tax earnings to total assets, DISPERS = 100% minus the sum of the top five noncustodial holdings, and CORP = the number of takeover bids on a company during the previous five years. CEO_IND measures the influence of the CEO. Higher values of CEO_IND indicate a potentially less influential CEO, and better governance. BD_IND is the board governance index. Higher values indicate better governance. BODOWN measure the percentage of shares held beneficially by the board of directors. BLOCKINST measures the percentage of shares held by BLOCK and INST owners, these investors represent the external shareholders of the firm. CEO_IND1 measures CEO influence. CEO_IND1 takes the value one if the CEO is not a director and zero otherwise. A value of one for CEO_IND indicates a potentially less influential CEO. BD_IND1 is an alternative board governance index. Higher values of BD_IND1 indicate better governance. *** Significant at 1%, ** Significant at 5%; * Significant at 10%.

MODEL 1 2 3 4 5

INTERCEPT -3.189 -1.661 -2.326 -1.014 -1.779

2.222 1.820 2.316 1.928 2.399

CEO_IND -1.247

-1.317

-1.203

0.916 0.947 0.956

BD_IND 2.537*

2.395*

0.958

1.331 1.366 2.521

CEO_IND1

-0.118

-0.141 0.429 0.444

BD_IND1

-2.405

-2.226 1.676 1.698

BODOWN 1.012 -0.461 1.857 0.492 -0.508

5.995 5.938 6.044 6.001 6.373

BODOWN2 5.918 10.205 3.370 6.945 3.256

19.766 19.749 20.405 20.449 20.815

BODOWN3 -15.017 -18.394 -13.576 -16.200 -13.348

16.746 16.809 17.732 17.829 18.305

BLOCKINST -0.697 -0.821 -9.759 -9.744 -10.367

1.258 1.219 6.654 6.611 7.407

BLOCKINST2

26.093 26.722 28.738

17.614 17.486 18.128

BLOCKINST3

-19.990 -21.023 -21.955

13.503 13.392 13.960

BODOWNBD_IND

5.528

4.697

BLOCKINSTBD_IND

-0.275

5.117

LEVERAGE 2.535*** 2.613*** 2.667*** 2.790*** 2.600***

0.852 0.858 0.892 0.899 0.901

ROA 15.239*** 15.094*** 15.758*** 15.639*** 15.470***

2.135 2.107 2.249 2.218 2.349

CORP 0.469 0.384 0.590 0.517 0.689

0.419 0.424 0.440 0.449 0.468

SIZE 0.366*** 0.350** 0.351** 0.331** 0.335**

0.138 0.140 0.138 0.140 0.145

MB -0.296*** -0.295*** -0.306*** -0.303*** -0.295***

0.101 0.100 0.101 0.101 0.104

PGROWTH -2.643*** -2.510*** -2.574*** -2.442*** -2.534***

0.676 0.671 0.684 0.680 0.682

DISPERS -1.138 -1.113 -1.141 -1.114 -0.926

1.045 1.017 1.051 1.028 1.078

Year Dummies YES YES YES YES YES

Number of obs. 563 563 563 563 563

Pseudo R² 0.564 0.560 0.568 0.564 0.571

Likelihood Ratio -123.005*** -124.221*** -121.937*** -123.074*** -121.173***

Table 6. OLS Regression Results This table reports the regression output for the four OLS models fitted over the pooled data sample of 563 observations from 2004 to 2012. Panel A reports the results for the models fitted using the dependent variable DIVPAY3. Panel B reports the results for the models fitted using the dependent variable DIVPAY5. The coefficients are reported first with the corresponding standard error in grey beneath each estimate. Year dummy variables are included in the estimated models. The control variables are PGROWTH = the five-year geometric mean of past growth in total sales, MB = the ratio of the market-to-book value of equity and is a proxy for growth, LEVERAGE = the ratio of the book value of total debt to the book value of total assets, SIZE = the natural logarithm of market capitalization, ROA = the three-year mean ratio of after-tax earnings to total assets, DISPERS = 100% minus the sum of the top five noncustodial holdings, CORP = the number of takeover bids on a company during the previous five years. CEO_IND measures the influence of the CEO. Higher values of CEO_IND indicate a potentially less influential CEO, and better governance. BD_IND is the board governance index. Higher values indicate better governance. BODOWN measure the percentage of shares held beneficially by the board of directors. BLOCKINST measures the percentage of shares held by BLOCK and INST owners, these investors represent the external shareholders of the firm. *** Significant at 1%, ** Significant at 5%; * Significant at 10%.

Panel A: DIVPAY3 Panel B: DIVPAY5

MODEL 1 2 3 4 5 6 7 8

INTERCEPT 0.230 0.353* 0.146 0.271 0.295* 0.415** 0.232 0.354**

0.193 0.197 0.196 0.201 0.169 0.171 0.171 0.175

CEO_IND 0.186** 0.198*** 0.181** 0.190** 0.190*** 0.209*** 0.204*** 0.217***

0.077 0.076 2.330 0.077 0.067 0.066 0.068 0.067

BD_IND 0.289** -0.257 0.293** -0.196 0.236** -0.343* 0.204** -0.303*

0.117 0.202 2.470 0.204 0.102 0.175 0.104 0.177

BODOWN -0.858* -0.929* -0.949* -0.998* -0.963** -1.007** -0.963** -1.023**

0.477 0.510 -1.960 0.514 0.414 0.441 0.418 0.443

BODOWN2 2.439 2.991* 2.255 2.864* 2.811** 3.358** 2.260 2.898**

1.625 1.615 1.360 1.649 1.412 1.393 1.427 1.418

BODOWN3 -2.173 -2.637* -1.803 -2.359* -2.505** -2.969** -1.829 -2.419**

1.395 1.386 -1.270 1.421 1.215 1.199 1.231 1.225

BLOCKINST -0.298*** -0.950*** 0.530 -0.151 -0.330*** -1.022*** 0.064 -0.603

0.101 0.209 0.930 0.601 0.087 0.177 0.489 0.513

BLOCKINST2 -1.296 -1.328 0.235 0.069

-0.820 1.580 1.349 1.346

BLOCKINST3

0.409 0.577

-0.891 -0.591

0.330 1.254 1.060 1.061

BODOWNBD_IND -0.145 -0.166 -0.200 -0.143

0.441 0.444 0.383 0.386

BLOCKINSTBD_IND 1.538***

1.410***

1.644***

1.447***

0.429 0.435 0.366 0.373

LEVERAGE -0.064 -0.063 -0.030 -0.036 -0.073 -0.067 -0.029 -0.036

0.070 0.071 -0.420 0.072 0.062 0.061 0.062 0.062

ROA 1.292*** 1.108*** 1.298*** 1.130*** 1.188*** 0.991*** 1.189*** 1.015***

0.117 0.126 11.170 0.126 0.103 0.110 0.102 0.110

CORP 0.014 0.015 0.016 0.015 0.002 0.003 0.008 0.007

0.027 0.027 0.570 0.027 0.024 0.024 0.024 0.024

SIZE 0.013 0.022* 0.011 0.019 0.011 0.020* 0.008 0.016

0.011 0.012 0.980 0.012 0.010 0.010 0.010 0.010

MB 0.006 0.002 0.007 0.003 0.006 0.002 0.008 0.004

0.009 0.009 0.860 0.009 0.008 0.008 0.007 0.008

PGROWTH -0.295*** -0.304*** -0.303*** -0.310*** -0.282*** -0.292*** -0.287*** -0.294***

0.068 0.067 -4.460 0.067 0.059 0.058 0.059 0.058

DISPERS -0.139 -0.117 -0.133 -0.116 -0.202*** -0.177** -0.187** -0.168**

0.087 0.087 -1.520 0.088 0.076 0.076 0.076 0.076

Year Dummies YES YES YES YES YES YES YES YES

Number of obs. 536 536 536 536 540 540 540 540

Adjusted R² 0.357 0.370 0.362 0.373 0.395 0.416 0.406 0.421

F 15.12*** 14.68*** 14.20*** 13.72*** 17.76*** 17.68*** 16.99*** 16.65***

Table 7. Robustness Check: OLS Regression Estimated Using Principal Components This table reports the regression output for the four OLS models fitted over the pooled data sample of 563 observations from 2004 to 2012 using principal components for the control variables. Panel A reports the results for the models fitted using the dependent variable DIVPAY3. Panel B reports the results for the models fitted using the dependent variable DIVPAY5. The coefficients are reported first with the corresponding standard error in grey beneath each estimate. Year dummy variables are included in the estimated models. The control variables are PGROWTH = the five-year geometric mean of past growth in total sales, MB = the ratio of the market-to-book value of equity and is a proxy for growth, LEVERAGE = the ratio of the book value of total debt to the book value of total assets, SIZE = the natural logarithm of market capitalization, ROA = the three-year mean ratio of after-tax earnings to total assets, DISPERS = 100% minus the sum of the top five noncustodial holdings, CORP = the number of takeover bids on a company during the previous five years. CEO_IND measures the influence of the CEO. Higher values of CEO_IND indicate a potentially less influential CEO, and better governance. BD_IND is the board governance index. Higher values indicate better governance. BODOWN measure the percentage of shares held beneficially by the board of directors. BLOCKINST measures the percentage of shares held by BLOCK and INST owners, these investors represent the external shareholders of the firm. *** Significant at 1%, ** Significant at 5%; * Significant at 10%.


MODEL 1 2 3 4 5 6 7 8

INTERCEPT 0.211** 0.504*** 0.131 0.428*** 0.191** 0.480*** 0.135 0.426***

0.087 0.107 0.098 0.119 0.076 0.092 0.086 0.103

CEO_IND 0.187** 0.208*** 0.183** 0.200*** 0.199*** 0.230*** 0.214*** 0.237***

0.077 0.076 0.078 0.077 0.068 0.066 0.068 0.067

BD_IND 0.352*** -0.369* 0.361*** -0.324 0.299*** -0.436** 0.273** -0.415**

0.119 0.202 0.121 0.205 0.104 0.176 0.106 0.179

BODOWN -0.441 -0.615 -0.519 -0.673 -0.587 -0.748* -0.560 -0.737*

0.482 0.507 0.490 0.513 0.420 0.439 0.425 0.443

BODOWN2 1.452 2.518 1.335 2.506 1.911 2.939** 1.388 2.569*

1.650 1.628 1.685 1.669 1.439 1.407 1.462 1.440

BODOWN3 -1.479 -2.336* -1.211 -2.207 -1.871 -2.703** -1.267 -2.283*

1.419 1.399 1.453 1.438 1.241 1.212 1.264 1.245

BLOCKINST -0.255*** -1.129*** 0.496 -0.440 -0.262*** -1.136*** -0.001 -0.867*

0.096 0.198 0.582 0.605 0.084 0.169 0.504 0.518

BLOCKINST2

-1.292 -1.362

0.391 0.105

1.623 1.600 1.392 1.368

BLOCKINST3

0.532 0.756

-0.884 -0.463

1.288 1.270 1.094 1.078

BODOWNBD_IND

-0.326

-0.359

-0.321

-0.290

0.440 0.443 0.383 0.387

BLOCKINSTBD_IND

2.060***

1.996***

2.073***

1.964***

0.410 0.414 0.352 0.357

PC1 -0.164*** -0.150*** -0.161*** -0.150*** -0.152*** -0.138*** -0.147*** -0.135***

0.014 0.014 0.014 0.014 0.012 0.012 0.012 0.012

PC2 0.076*** 0.074*** 0.079*** 0.076*** 0.062*** 0.061*** 0.066*** 0.063***

0.017 0.016 0.017 0.016 0.014 0.014 0.014 0.014

PC3 -0.001 0.000 -0.006 -0.003 -0.004 -0.003 -0.009 -0.006

0.015 0.015 0.015 0.015 0.013 0.013 0.013 0.013

PC4 -0.048*** -0.040** -0.046 -0.040** -0.0525*** -0.044*** -0.046*** -0.041***

0.016 0.016 0.016 0.016 0.014 0.014 0.014 0.014

PC5 0.033* 0.018 0.033* 0.018 0.034** 0.019 0.034** 0.019

0.019 0.020 0.019 0.020 0.017 0.017 0.017 0.017


Number of obs. 536 536 536 536 540 540 540 540

Adjusted R² 0.3249 0.354 0.328 0.354 0.360 0.398 0.367 0.399

F 14.55*** 14.96*** 13.41*** 13.76*** 16.96*** 17.98*** 15.85*** 16.59***

Table 8. Robustness Check: Tobit Regression Results This table reports the regression output for the four OLS models fitted over the pooled data sample of 563 observations from 2004 to 2012 using principal components for the control variables. Panel A reports the results for the models fitted using the dependent variable DIVPAY3. Panel B reports the results for the models fitted using the dependent variable DIVPAY5. The coefficients are reported first with the corresponding standard error in grey beneath each estimate. Year dummy variables are included in the estimated models. The control variables are PGROWTH = the five-year geometric mean of past growth in total sales, MB = the ratio of the market-to-book value of equity and is a proxy for growth, LEVERAGE = the ratio of the book value of total debt to the book value of total assets, SIZE = the natural logarithm of market capitalization, ROA = the three-year mean ratio of after-tax earnings to total assets, DISPERS = 100% minus the sum of the top five noncustodial holdings, CORP = the number of takeover bids on a company during the previous five years. Higher values indicate better governance. BODOWN measure the percentage of shares held beneficially by the board of directors. BLOCKINST measures the percentage of shares held by BLOCK and INST owners, these investors represent the external shareholders of the firm. *** Significant at 1%, ** Significant at 5%; * Significant at 10%.


1 2 3 4 5 6 7 8

INTERCEPT 0.013 0.119 -0.086 0.015 0.141 0.241 0.076 0.173

0.237 0.240 0.239 0.244 0.199 0.200 0.200 0.203

CEO_IND 0.214** 0.224** 0.225** 0.230** 0.228*** 0.242*** 0.262*** 0.269***

0.093 0.092 0.093 0.093 0.079 0.078 0.079 0.078

BD_IND 0.421*** -0.077 0.393*** -0.040 0.332*** -0.216 0.260** -0.201

0.141 0.255 0.142 0.255 0.118 0.213 0.120 0.212

BODOWN -1.265** -1.309** -1.283** -1.342** -1.420*** -1.405*** -1.323*** -1.359***

0.602 0.625 0.604 0.625 0.511 0.528 0.509 0.526

BODOWN2 4.487** 4.762** 3.840* 4.199* 4.926*** 5.226*** 3.916** 4.320**

2.164 2.149 2.164 2.163 1.856 1.827 1.845 1.833

BODOWN3 -4.638** -4.830** -3.790* -4.085** -4.921*** -5.128*** -3.810*** -4.147**

1.957 1.939 1.952 1.947 1.698 1.667 1.684 1.670

BLOCKINST -0.407*** -0.931*** 0.273 -0.166 -0.423*** -1.026*** -0.363 -0.839

0.122 0.250 0.687 0.714 0.102 0.205 0.566 0.585

BLOCKINST2 -0.370 -0.606 1.511 1.189

1.873 1.875 1.532 1.533

BLOCKINST3

-0.513 -0.241

-1.978* -1.611

1.472 1.476 1.196 1.201

BODOWNBD_IND -0.041 -0.020 -0.194 -0.087

0.529 0.529 0.444 0.444

BLOCKINSTBD_IND

1.283**

1.123**

1.480***

1.236***

0.528 0.532 0.434 0.438

LEVERAGE 0.093 0.096 0.159 0.152 0.084 0.098 0.163* 0.161*

0.103 0.103 0.106 0.106 0.086 0.085 0.088 0.087

ROA 3.465*** 3.224*** 3.495*** 3.282*** 2.718*** 2.463*** 2.740*** 2.524***

0.313 0.320 0.312 0.321 0.223 0.228 0.222 0.228

CORP 0.033 0.033 0.038 0.036 0.013 0.012 0.021 0.019

0.031 0.031 0.031 0.031 0.027 0.026 0.026 0.026

SIZE 0.017 0.025* 0.013 0.020 0.013 0.022* 0.007 0.015

0.014 0.014 0.014 0.015 0.012 0.012 0.012 0.012

MB -0.023* -0.027** -0.021* -0.025** -0.019*** -0.023** -0.016 -0.020**

0.012 0.013 0.012 0.013 0.010 0.010 0.010 0.010

PGROWTH -0.603*** -0.610*** -0.618*** -0.623*** -0.581*** -0.590*** -0.590*** -0.595***

0.103 0.103 0.103 0.103 0.087 0.086 0.086 0.086

DISPERS -0.184* -0.156 -0.148 -0.128 -0.232*** -0.204** -0.189** -0.170*

0.106 0.107 0.106 0.107 0.090 0.090 0.089 0.089


Number of obs. 536 536 536 536 540 540 540 540

Pseudo R² 0.464 0.471 0.472 0.477 0.558 0.573 0.574 0.584

log likelihood -239.275*** -236.242*** -235.770*** -233.47 -175.392*** -169.617*** -169.380*** -165.352***

Agency Theory, Corporate Governance and Dividend Payout … Policy and Governance in... · Agency Theory, Corporate Governance and Dividend ... linear relationship between dividend

Documents