Reputation as a governance mechanism? Evidence from payout policy of insider‐controlled firms in Taiwan

• Author: Liang‐wei Kuo
• Date: 01 October 2017
• Published date: 01 October 2017
• DOI: http://doi.org/10.1111/jbfa.12261

DOI: 10.1111/jbfa.12261

Reputation as a governance mechanism? Evidence

from payout policy of insider-controlled firms

in Taiwan

Liang-wei Kuo

Departmentof Finance, National Chung Cheng

University,Taiwan

Correspondence

Liang-weiKuo, Department of Finance, National

ChungCheng University, No.168, Sec. 1, Uni-

versityRd., Min-Hsiung, Chia-Yi 62120, Taiwan.

Email:glendalw@ccu.edu.tw

Abstract

This paper examines the effect of entrenched insiders’ reputational

concerns on corporate payout policy in Taiwan, a market in which

typical public firms are controlled by a single dominant shareholder

who is subject to weak takeover threats and has incentives and

abilities to extract private benefits by oppressing minority equity

holders. The reputation-building hypothesis predicts that firms with

higher expropriation risk by a controlling shareholder make more

payouts to credibly commit not to expropriate minority sharehold-

ers, thereby establishing reputation in the capital market for risk

diversification and low-cost external financing. I show that corpo-

rate payout intensity is significantly and positively correlated with

measures related to the moral hazard of dominant owners. The rep-

utation effect manifests in firms that most value it; the interaction

analyses indicate that younger, smaller,or growth firms with higher

controlling shareholder expropriation risk pay more cash dividends.

Moreover, firms are less likely to omit dividends and more likely

to resume dividends when their controlling shareholders are more

entrenched. Finally,I show that the value of cash dividends is higher

for firms with higher controlling shareholder expropriation risk and

that expected dividend increases in these firms are value enhancing.

KEYWORDS

payout policy, agency problem, controlling shareholder, expropria-

tion, control rights, corporate governance

1INTRODUCTION

Prior literature widely shows that manager–shareholder agency conflicts cause financial policies to be relevant to firm

value. FollowingJensen and Meckling (1976), the free-cash-flow-centric theory suggests that firms choose debt and/or

dividend policies ex ante to minimize agency costs that arise from managers pursuing their own agendas at the expense

of shareholders’ interests (e.g.,Grossman & Hart, 1982; Easterbrook, 1984; Jensen, 1986; Stulz, 1990). However, more

recent literature argues that static ex-ante optimal financial policies may not incentivize entrenched managers, who

J Bus Fin Acc. 2017;44:1443–1476. wileyonlinelibrary.com/journal/jbfa c

2017 John Wiley & Sons Ltd 1443

1444 KUO

in reality possess most of the residual control rights over corporate decision-making and can only be disciplined at

costs borne by shareholders. That is, entrenched managers may reverse their previous decisions ex post if explicit or

implicit disciplinary mechanisms are not present. Through dynamic model setups, this literature obtains efficient equi-

librium choices that also pertain to dividend policy (Zwiebel, 1996; Fluck, 1999; Allen et al., 2000; Myers,2000;). To wit,

entrenched managers voluntarily pay dividends to credibly constrain their empire-building ambition in orderto deter

potential control challenges by outsiders.

Because dividends reduce cash resources at managers’ disposal that may otherwise be wasted or diverted, they

are presumably more preferred by investors in markets where agency problems are more severe. In these markets,

public corporations are typically controlled by a single dominant shareholder who is often involved in management

and amplifies his or her effective control through control mechanisms such as dual-class shares, pyramidal structures

and cross-holdings. The lack of active takeover marketsand weak legal protection of (minority) shareholders further

aggravatethe conflicts between outside investors and inside controlling shareholders as manifested by a variety of tun-

neling activities documented in the literature from inefficient investment to outright theft (e.g.,Johnson et al., 2000a;

Johnson et al., 2000b).1

From the perspective of minority shareholders in these firms, a dollar distributed to equity holders is not equiva-

lent to a dollar kept in the firm; hence, dividends used to mitigate expropriationby dominant shareholders are of great

importance. Nevertheless, controlling shareholders can significantly influence dividend decisions, and thus whether

their incentives are consistent with positive payouts determines the role dividends play in addressing agency prob-

lems. Gomes’ (2000) multiperiod signaling model provides a theoretical rationale. He suggests that controlling share-

holders in public firms have incentives to build reputations for not divertingcash flows in order to later diversify their

idiosyncratic risk by trading at prices that are otherwise discounted for the anticipated expropriation concerns. This

reputation effect is more valuable for firms with a higher degree of moral hazard. Dividend payouts are a credible sig-

nal insofar as equity markets facilitate multiperiod cash flow realizations and trading activities. Gomes also stresses

that risk diversification, rather than access to equity capital at lower costs, motivates dominant owners to build a rep-

utation for treating minority shareholders well because equity markets are not a major source of external funds in

countries lacking effective governance mechanisms.

I empirically test the payout policy implications of the reputation effect for entrenched controlling shareholders in

Taiwan.I choose this market for its market and institutional features. First, the majority of public firms have a single

large controlling shareholder, who holds at least 10% of the company’s voting stakes and uses control mechanisms,

such as pyramidal structures, to enhance effective control with a relativelylower level of cash flow rights. The owner-

ship of the second largest shareholder in these companies is normally incommensurable, so that takeoverthreats and

monitoring are weak or non-existent.However, despite the weak legal protection of shareholder rights compared with

that of developed countries such as the United States, minority individual investors in Taiwan supply substantial cap-

ital to the primary market and actively trade shares in the secondary market. Forinstance, in 2012 they contributed

more than 40% of capital to listed companies, and their aggregate transactions accounted for more than 60% of total

annual trading value in the TaiwanStock Exchange (TWSE).2Notwithstanding the rather active equity marketin terms

of turnover ratio,3equity financing is the main source of funds for firms; in 2012 equity comprised less than 8% of the

sources of funds as opposed to 40% for debt.4La Porta et al. (1999), among others,5show that these characteristics

are not atypical. In addition, the risk of expropriation of minority shareholders by the dominant shareholder reflects

1Examples of tunneling activities by controlling shareholders include asset sales at prices favorableto other corporations they control, transfer of funds or

assets out of the companies for their personal use, inefficient investments with large private benefits to themselves, and transferpricing with other entities

theycontrol.

2Thefigures are from the statistics data published by the TWSE at http://www.twse.com.tw/en/statistics/statistics.php?tm=07.

3According to statistics provided in the regular report by the TWSE, the turnoverratio of Taiwan stock market in terms of trading value was 97.33 in 2012,

comparedwith 86.66 on the New York Stock Exchange, 92.41 on the TokyoStock Exchange, and 95.73 on the Shanghai stock market.

4Thestatistics are compiled from data published by the Central Bank of Taiwan at http://www.pxweb.cbc.gov.tw/dialog/statfile1L.asp?lang=1&strList=L.

5La Porta et al. (1999) examine ownership structures of large corporations in 27 wealthy economies and find that firms in most countries are controlled

by dominant owners, whose voting power is significantly in excessof their cash flow ownership. Similar findings also appear in a contemporaneous study by

Claessens et al. (2000) for the separation of ownership and control in 2,980 corporations from 9 East Asian countries. Recent evidencein Carney and Child

KUO 1445

negatively on the valuation of firms in this market (Yeh,Shu, & Guo, 2008). As such, dividends can serve as a signal to

outside investors that controlling shareholders are voluntarilyconstraining their entrenchment activities. Second, Tai-

wan’s imputation tax system reduces the tax disadvantages of individual investorsreceiving dividend income because

the imputation tax credit for corporate taxes paid by the companies distributing dividends can be used to offset indi-

viduals’ income tax liability.6Consequently,tax considerations may not be a significant factor when firms make payout

decisions. Takentogether, these market characteristics, along with tax institutions, make the Taiwan market a suit-

able environment in which to investigatewhether dominant owner moral hazard explainscorporate payout policy and

whether payouts can lower the agency costs of controlling shareholders.

La Porta et al.’s (2000) outcome hypothesis, based on agency theory, serves as an alternative to the reputation-

building hypothesis. This outcome hypothesis suggests that dividends are an outcome of the operationof the country-

level legal protection of minority shareholders and the firm-level quality of governance mechanisms. Minority share-

holders presumably are better protected by an effective legal system and by the active monitoring of the board of

directors or other institutional investors, providing more power to force firms to pay out dividends. Hence, the out-

come hypothesis predicts that dividend payments are higher for firms in countries with strong legal protection and

effective enforcement of shareholder rights. Within a country,firms with a higher (lower) quality of internal or exter-

nalgovernance mechanisms are associated with a higher (lower) payout level, ceteris paribus. Several studies empirically

test the outcome hypothesis using either a multicountry sample (e.g., La Portaet al., 2000; Faccio et al., 2001; Kalcheva

& Lins, 2007) or a non-US single-country analysis (e.g., Adjaoud & Ben-Amar,2010) and find evidence that is consistent

with the hypothesis’ predictions.

The outcome hypothesis is conditioned on the strength of corporate governance. The reputation-building hypoth-

esis, although implicitly related to the quality of governance, focuses on entrenched insiders’ use of dividends as a

commitment to build reputation in the capital market, either for their personal diversification objectives or to finance

future firm needs. Thus, this study tests the reputation-building hypothesis against the outcome hypothesis in a set-

ting characterized by a less effective investor rights protection system compared with the United States, the United

Kingdom, or other developed economies, and byweaker firm governance against the single dominant shareholder.7

I use three proxiesto measure the entrenchment level of a firm’s largest ultimate owner. The first is the deviation, or

wedge, between the dominant shareholder’s voting rights and his or her cash flow rights. As previously noted, control-

enhancing mechanisms can separate ownership from control. The entrenchment of the dominant owner increases his

or her power over corporate resources and exacerbates agency conflicts, as evidenced by the negative relationship

between the wedge and firm value (Claessens et al., 2002; Lemmon & Lins, 2003). The second and third proxies are

motivated by the observation that dominant shareholders can also control a firm through direct participation in man-

agement, which, when combined with controlling the monitoring function, insulates them from internal challenges. I

define the dominant shareholder as directly participating in management if he or she or a family member serves as

CEO of the firm, and I define the dominant shareholder as controlling both management and monitoring functions if he

or she or a family member serves as both CEO and chairperson of the board. Tomy knowledge, the last two measures

are novel and have not been jointly considered with the control–cash flow wedge in research on the agency problems

in firms with controlling shareholders.

Using a sample of corporations publicly listed on the TWSE and TaipeiExchange (TPEx), the two primary securi-

ties markets in Taiwan, between 2000 and 2012, I test the hypothesis that more entrenched dominant shareholders

pay out more dividends to build reputation in the equity markets. I find that the measures of the extent of control-

ling shareholder expropriationrisks are significantly and positively correlated with payout intensity. In particular,firms

with a high deviation between votingand cash flow rights and firms with a CEO related to the ultimate shareholder pay

(2013)explores the changes in ownership and control from 1996 to 2008 of the same set of companies as Claessens et al. and find that political change in some

countries maydrive changes in ownership structure yet the fundamental characteristics do not change drastically. La Porta et al. (1997) report that the ratio

ofthe stock market capitalization held by minorities to gross GNP is more than 40% in their sample of 49 countries.

6InTaiwan, corporate income tax prior to 1 January 2010 was 25% and was subsequently reduced to 17%. Personal income tax takes a progressiverate from

5%to 40%.

7Ithank an anonymous referee for suggesting that I consider the alternative outcome hypothesis and test it against the reputation-building hypothesis.