Do Co‐Opted Directors Influence Dividend Policy?

• Date: 01 June 2018
• DOI: http://doi.org/10.1111/fima.12196
• Author: Sang Mook Lee,Pornsit Jiraporn
• Published date: 01 June 2018

Do Co-Opted Directors Influence

Dividend Policy?

Pornsit Jiraporn and Sang Mook Lee∗

Weexplore how co-opted directors affect dividend policy. Co-opted directorsare those appointed

after the incumbent chief executive officer (CEO) assumes office. Our results show that co-

opted directors lead to a weaker propensity to pay dividends and, for dividend-paying firms,

significantly lower dividend payouts. We also show that board co-option has more explanatory

power for dividend policy than does the traditional measureof board effectiveness, that is, board

independence. Exploiting the passage of the Sarbanes-Oxley Act as a naturalexperiment, we show

that the effect of board co-option on dividend policy is more likely causal, rather than merely an

association.

Dividend policy is irrelevant under the assumption of perfect capital markets (Miller and

Modigliani, 1961). Over the past 50 years, researchers have attempted to relax this assumption

by introducing several market imperfections. One important market imperfection is the existence

of agency conflicts, wherein managers may adopt a suboptimal dividend policy that does not

maximize shareholders’ wealth. Agency problems, however, can be alleviatedby effective gover-

nance mechanisms. The board of directors is usually regarded as one of the primary governance

mechanisms in the firm and thus plays a critical role in mitigating potential agency conflicts.

In this article, we explore how board effectiveness influences dividend policy.A majority of the

literature on board effectiveness focuses on board size and composition. Unfortunately, despite

much research, the evidence on the effect of board independence on corporate outcomes is mixed

and weak. We investigate the role of board effectivenessusing a novel measure, that is, board co-

option. Coles, Daniel, and Naveen (2014) demonstrate that the degree of board co-option better

explains corporate outcomes such as executive compensation and chief executive officer (CEO)

turnover than do traditional measures of board independence. The degree of board co-option

is defined as the percentage of directors appointed after the CEO assumes office. Directors

appointed by the current CEO may not impose stringent oversight over the CEO as the CEO

was involved in their appointment. In other words, these directors are “co-opted” by the CEO.

Therefore, the larger the number of co-opted directors on the board, the less effective the board

is supposed to be in mitigating agency conflicts.

Agency theory suggests that opportunistic managers tend to dislike paying dividends as divi-

dends reduce free cash flow they could otherwise exploit. Thus, in firms with co-opted boards,

shareholders may be less likely to receivedividends or may receive smaller dividends than would

be the case in firms with strong governance. Our empirical evidence strongly supports this

We sincerely thank an anonymous referee and Raghavendra Rau (Editor) for their helpful comments and suggestions.

We also thank Lalitha Naveen for sharing the data on co-opted directors. Part of this research was carried out when

PornsitJiraporn served as a visiting scholar at The National Institute of Development Administration (NIDA) in Bangkok,

Thailand.

∗PornsitJiraporn is an Associate Professor of Finance in the Schoolof Graduate Professional Studies, Pennsylvania State

University in Malvern, PA.Sang Mook Lee is an Assistant Professor of Finance in the School of Graduate Professional

Studies, Pennsylvania State University in Malvern, PA.

Financial Management •Summer 2018 •pages 349 – 381

350 Financial Management rSummer 2018

hypothesis. Board co-option leads to a significantly weaker propensity to pay dividends. For

dividend-paying firms, board co-option is associated with significantly smaller dividend payouts.

The results hold even after controlling for firm characteristics as well as board independence.

In fact, board co-option explains the dividend-paying behavior of a firm much better than does

board independence.

In terms of economic magnitude, a rise in the percentage of co-opted directors by one standard

deviation lowersthe propensity to pay dividends by 4.62%. For dividend-payingf irms, an increase

in board co-option by one standard deviation reduces dividend payouts by 6.16%. Thus, the effect

of board co-option on dividend policy is not only statistically significant but is also economically

meaningful. It is crucial to note that when board co-option is included in the regression analysis,

board independence becomes less significant in explaining dividend payouts. This is particularly

interesting because the debate on board effectiveness in past decades has been dominated by

board independence as the dominant measure of board efficacy. Our study shows that, at least as

far as dividend policy is concerned, board co-option has stronger explanatory power than does

board independence.

To reduce endogeneity issues, we follow several steps. First, to minimize a potential omitted-

variablebias, we control for the one-year lagged value of dividend payouts.Because both dividends

at time tand dividends at time t−1 are influenced by the same unobservable time-invariant

characteristics, inclusion of the lagged value mitigates the omitted-variable bias. We obtain

consistent results when the lagged value is included. Second, we exploit the passage of the

Sarbanes-Oxley Act (SOX) and the associated listing requirements by NASDAQ and the New

York Stock Exchange (NYSE) as a natural experiment. Prior research shows that firms appointed

additional independent directors to comply with the mandate requiring a majority of independent

directors (Linck, Netter, and Yang, 2009). This represents an exogenous shock that raises board

co-option. Our natural experiment reveals that board co-option leads to lower dividend payouts.

Because a natural experiment based on an exogenous regulatory shock is substantially less

susceptible to endogeneity, the effect of board co-option on dividend policy is likely causal.

It can be argued that dividends do not represent total cash distributions to shareholders because

many firms distribute cash through stock repurchases as well. Consequently, weinvestigate total

payouts, calculated as the sum of both dividends and stock repurchases. The evidence shows that

the presence of co-opted directors leads to lower total payouts, consistent with the evidence for

dividends. Wenext investigatethe choice between dividends and stock repurchases. Agency theory

suggests that managers prefer repurchases over dividends because unlike dividends, repurchases

are not firm commitments. In contrast, dividends, once paid, usually must be maintained. Our

results show that more board co-option leads to a much weaker propensity to pay dividends

relative to using repurchases. Interestingly, our analysis also shows that board independence does

not explain very well the choice between dividends and repurchases, whereas board co-option

does, in a way consistent with the predictions of agency theory.

Finally, we investigatethe role of board co-option in regulated f irms. Regulated firms are likely

to be less vulnerable to agency problems because regulation represents an additional layer of

monitoring that prevents managers from exploiting shareholders. Consistent with this notion, we

find that the effect of board co-option on dividends is not significant in f inancial and utility firms.

The results of our study contribute to the literature in several areas. First, we contribute to

the literature in dividend policy. Our results support the agency explanation of dividend policy.

Co-opted boards represent a weakened governance mechanism that exacerbates agency costs

and allows managers to retain more free cash flow within the firm. This is consistent with the

notion that managers dislike dividends, as dividends deprive them of potential free cash flows

they could otherwise exploit. Second, we contribute to the literature in corporate governance.

Jiraporn & Lee rDo Co-Opted Directors Influence Dividend Policy? 351

The debate on the role of the board, however, has been dominated by board independence. We

show that board independence does not explain dividend policy very well. Board co-option, on

the contrary, exhibitsmuch stronger explanatory power.Our results strongly confir m the findings

in Coles et al. (2014), who propose that more attention should be given to co-opted directors as

a new measure of board effectiveness. Third, our results contribute to the debate over the costs

and benefits of SOX. Proponents of SOX argue that it helps strengthen corporate governance.

Opponents argue that it imposes heavy and unnecessary burdens on firms, particularly small

firms, and that its effectiveness is still ambiguous. We contribute to this debate by showing that

the focus of SOX, that is, board independence, may not completelyrepresent board effectiveness.

Rather, board co-option may be more relevant for certain corporate outcomes, such as dividend

policy.

I. Literature Review and Hypothesis Development

Dividend payouts are hypothesized to reduce agency conflicts by reducing the amount of free

cash flow, which could be used by managers for their private benefits rather than for maximizing

shareholders’ wealth (Grossman and Hart, 1980; Easterbrook, 1984; Jensen, 1986; DeAngelo,

DeAngelo, and Stulz, 2006). Dividends may also help mitigate agency conflicts by exposing

firms to more frequent monitoring by the primary capital markets, as paying dividends increases

the probability that new equity must be issued more often (Easterbrook, 1984). Based on the

literature, we develop two competing hypotheses that explain the effect of board co-option on

dividend payouts.

A. Outcome Hypothesis

The outcome hypothesis argues that board co-option leads to a weaker tendency to pay div-

idends. For dividend-paying firms, board co-option is associated with lower dividend payouts.

Co-opted directors are less likely to monitor managers in a stringent manner. Thus, managers are

more likely to adopt a dividend policy that maximizes their private benefits. Because managers

inherently prefer not to pay dividends as dividendsdeprive them of free cash flow they otherwise

could exploit, this hypothesis predicts that more board co-option leads to lowerdividend payouts.

Per this view, dividend policy is an outcome of a weak governance mechanism, that is, a co-opted

board.

Consistent with this hypothesis, several prior studies show that weak (strong) governance is

associated with smaller (larger) dividend payouts. For instance, Renneboog and Szilagyi (2006)

report that firms with strong shareholders force higher dividend payouts in Dutch firms. Michaely

and Roberts (2006) conclude that strong governance encourages higher and more consistent

payouts. La Porta et al. (2000) find strong support for the outcome model. Using the governance

metrics provided by the Institutional Shareholder Services, Jiraporn, Kim, and Kim (2011) report

that shareholders of firms with strong corporate governance can force managers to disgorge cash

in the form of larger dividend payouts.

B. Substitution Hypothesis

The substitution hypothesis is predicated on an argument made by La Porta et al. (2000) and

relies critically on the need for firms to raise money in the external capital markets, at least

occasionally. To be able to raise capital in attractive terms, it may be necessary for firms to