NOTHING HERE

mailto:[email protected]

mailto:[email protected]

Determinants of Social Disclosure Quality in Taiwan: An Application of Stakeholder Theory

Tzu-Kuan Chiu • Yi-Hsin Wang

Received: 28 February 2013 / Accepted: 19 March 2014

� Springer Science+Business Media Dordrecht 2014

Abstract This study adopts a stakeholder theory frame-

work to examine determinants of social reporting quality

and empirically test the ability of the theory to explain

disclosure quality in an emerging economy. Using a sample

of 246 listed companies and a hand-collected dataset that

included 2 years of data based on survey questions

reflecting international disclosure trends, we apply an

aggregate measure of quality with five facets to a variety of

corporate social responsibility areas. The results support

the application and demonstrate that measures of stake-

holder power, strategic posture, economic resources, firm

size, and media visibility are related to social disclosure

quality. This paper adds to the scarce evidence on social

reporting in Taiwan and provides a useful method for

evaluating disclosure quality. It also illustrates the impact

of two dominant foreign stakeholder groups on social

disclosures in Taiwan: (1) organizational buyers in the

global supply chain and (2) listing and social rating agen-

cies in international capital markets.

Keywords Social reporting � Disclosure quality � Stakeholder theory � Supply chain � Taiwan

Introduction

Over the past two decades there has been a steady rise in

the demand for corporate information. With the growth of

public awareness and media coverage of environmental,

social, and ethical issues, firms have sought to enhance the

scope and depth of their social disclosure accordingly. In

broad terms, corporate social disclosure (CSD) refers to

corporate reporting that focuses on environmental, social,

and ethical issues (Gray et al. 1995; Adams et al. 1998;

Deegan 2002; Reverte 2009). It involves such areas as

climate change abatement, human rights, employee rela-

tions, product liability, philanthropy, and corporate gov-

ernance. Firms usually disclose information regarding their

activities in these areas in their annual reports, press

releases, or company website, or in separate social reports

[corporate social responsibility (CSR) report or sustain-

ability report].

In contrast to mandatory financial reporting, the social

reporting context is less prescriptive, thereby making the

motivation underlying disclosure by managers a topic of

inquiry. In general, CSD is viewed as a means to safeguard

corporate reputation and to influence perceptions of the

firm in the minds of stakeholders, such as regulators,

consumers, shareholders, civil society groups, and social

investors (Deegan 2002; Milne 2002; Chen and Roberts

2010; Islam and Deegan 2010). By sharing information

with outside parties on the noneconomic impacts of the

firm’s business operations, disclosure can diminish infor-

mational asymmetries between the firm and its stakehold-

ers, mitigate potential adverse regulatory pressures, and

leave the firm better placed to take advantage of future

investment opportunities. Moreover, several studies have

emphasized the role of social disclosure in maintaining the

legitimacy of business organizations within an institutional

T.-K. Chiu (&) Shanghai Advanced Institute of Finance, Shanghai Jiao Tong

University, Shanghai, China

e-mail: [email protected]

Y.-H. Wang

Department of Accountancy, National Taipei University, San

Shia, New Taipei City, Taiwan

123

J Bus Ethics

DOI 10.1007/s10551-014-2160-5

framework characterized by an increasing interest in CSR

activities (Adams et al. 1998; Neu et al. 1998; Haniffa and

Cooke 2005; De Villier and Van Staden 2006).

Prior studies have revealed that there are a number of

firm and industry characteristics which are potential

determinants of CSD practices and which affect the relative

costs and benefits of disclosure. More specifically, the

existing literature indicates that these costs and benefits are

related to pressures from external agents (Roberts 1992;

Milne 2002; Cormier et al. 2005; Islam and Deegan 2010;

Huang and Kung 2010), the firm’s vulnerability to those

pressures as captured by its size and media exposure

(Roberts 1992; Cormier and Magnan 2003), its resource

availability (Roberts 1992; Brammer and Pavelin 2008;

Branco and Rodrigues 2008), as well as ownership and

governance structures (Haniffa and Cooke 2005; Van der

Laan Smith et al. 2005; Khan et al. 2013).

Stakeholder theory is considered to be an influential

theory within the domain of social and environmental

accounting research (SEAR), and many studies have

employed this theory to explain the organization–society

relationship (Ullmann 1985; Roberts 1992; Van der Laan

Smith et al. 2005; Chen and Roberts 2010; Huang and

Kung 2010). The theory extends the manager’s roles and

responsibilities beyond profit maximization to include

interests and claims of nonshareholding stakeholder

groups. Within this framework, the long-run survival and

success of the firm requires stakeholders’ support, and CSD

is often used as a tool of communication between the

management and its stakeholders to win the necessary

support. However, not all stakeholder groups are of equal

importance to the company, differing in the power they

possess, the legitimacy of their claims, and the urgency of

their demands (Mitchell et al. 1997). Within the limited

time and resources available, managers may only respond

to the most pressing demands of one or two powerful

stakeholder groups, ignoring requests from other groups.

Past studies in the SEAR field have primarily focused on

the Anglo–American context, using countries such as the

UK, Australia, and the USA as the subject of analysis

(Roberts 1992; Gray et al. 1995; Deegan and Gordon 1996;

Adams et al. 1998; Neu et al. 1998). It has been argued,

however, that different societies define the role of firms and

stakeholders differently and that this difference affects the

extent, issues, and quality of CSD (Van der Laan Smith

et al. 2005; Matten and Moon 2008). In recent years, data

from studies on other institutional contexts in emerging

economies have supplemented the literature on CSD (Pu-

rushothaman et al. 2000; Xiao et al. 2004; Chapple and

Moon 2005; Haniffa and Cooke 2005; De Villier and Van

Staden 2006; Khan et al. 2013). These data have stressed

the impact of the macro context on social reporting,

including cultural background, national business system,

regulatory framework, and governance structure. More-

over, the existing literature shows that CSD in these

countries is not only limited in the range of themes

addressed, but is also largely restricted to human resources

and community involvement, with minimal reporting on

climate change and environmental issues. Further, disclo-

sures are normally expressed in general and vague terms

with little concrete information provided; consequently, the

overall quality of disclosure is poor.

This study examines the potential determinants of CSD

practices in Taiwan, an emerging economy where only

scant evidence exists on CSD practices to date. Within a

stakeholder theory framework, we modify the contingency

model proposed by Ullmann (1985) to explain the observed

differences in CSD quality across Taiwanese firms.

According to this model, three dimensions crucial to CSD

are stakeholder power, the firm’s strategic posture toward

CSD, and economic resources. As the theory needs to

embrace a contextual approach, we use Mitchell’s dynamic

model of stakeholder relations (Mitchell et al. 1997) to

identify the so-called ‘dominant’ stakeholders who possess

the relationship attributes of power and legitimacy in the

Taiwanese setting. Two dominant foreign stakeholder

groups, namely, organizational buyers in the global supply

chain and listing agencies in international capital markets,

are of special interest as these two groups affect the mode

of response that firms take to address stakeholder influ-

ences and their specific demands on reporting quality.

Firms displaying an active posture toward CSD through

deliberate measuring, verifying, collating, and publishing

of CSR information clearly attempt to influence their

organization’s status with these dominant stakeholders.

Specifically, Taiwan has a well-developed electronics

industry that serves as outsourced vendors for the global IT

industry and international brands such as Apple, Intel, and

Hewlett-Packard (Lau 1994; Sturgeon and Lee 2005). In

terms of reporting, these firms face high pressure from the

global supply chain and are required to provide specific

information on whether they set formal targets, quantify

social and environmental impact, and report specific actions;

they are also subject to external audit. In addition, quite a few

Taiwanese firms, including Delta Electronics, China Steel,

and Taiwan Mobile, are listed on leading international stock

exchanges or selected as components of global ethical indi-

ces. The opening up of Taiwan’s capital market has increased

the demand on firms to strengthen disclosure quality

regarding their CSR activities. It follows that pressure from

the two dominant foreign stakeholder groups serves as the

main drivers for quality disclosure in Taiwan.

Prior studies often use ‘content analysis’ to quantify the

data and analyze the determinants of the extent of social

reporting (Deegan and Gordon 1996; Patten 2002; Branco

and Rodrigues 2008; Huang and Kung 2010; Khan et al.

T.-K. Chiu, Y.-H. Wang

123

2013), but relatively little attention has been paid to dis-

closure quality. In this paper, we make a distinction

between the extent and the quality of disclosure and select

the latter as the focus of analysis. Although there is no

well-accepted disclosure quality index for CSD, the exist-

ing literature distinguishes between ‘soft talk’ disclosures,

characterized by general, qualitative, nonverifiable state-

ments concerning corporate aims, and ‘higher quality’

disclosures that have attributes of being specific (in time

frame, action, and monetary value), verifiable, and for-

ward-looking (Freedman and Stagliano 1992; Patten 2002;

Cormier et al. 2005; Van der Laan Smith et al. 2005;

Brammer and Pavelin 2008). Consistent with this view, we

use five facets of quality to analyze the reporting behavior

of Taiwanese firms: (1) the firm’s disclosure of CSR pol-

icies, (2) specification of goals, (3) formulation of initia-

tives, (4) quantification of impact and reporting of

progress, and (5) presentation of external certificates and

audits.

Using a sample of 246 firms listed on the Taiwan Stock

Exchange (TWSE) drawn from a diverse range of indus-

trial sectors, we analyze firms’ disclosure quality for two

consecutive years, 2010 and 2011, in the social responsi-

bility areas of human rights, employee relations, climate

change abatement, and environment impact. Data are col-

lected from a set of survey questions in line with interna-

tional disclosure trends, so that the influence of the foreign

institutional context on Taiwanese firms through regula-

tions imposed by dominant stakeholders can be better

observed. The study uses a variety of information sources,

including firms’ annual reports, websites, and CSR reports.

The empirical results show that disclosure quality in

Taiwan is unsatisfactory, with the worst areas being cli-

mate change abatement and environment impact, a situa-

tion which is related to the state’s low concern for social

aspects, poor environmental scrutiny, and insufficient reg-

ulation and lax enforcement on the part of the respective

regulatory authorities. This ensures that the responsiveness

of Taiwanese firms to external forces in the local envi-

ronment is limited, resulting in a situation where the

quality of social reporting lags behind that of industrialized

countries. However, while strong domestic scrutiny bodies

and regulations are lacking, the more globalized Taiwanese

firms, which are subject to higher disclosure standards due

to their international listing status or global supply chain

position, respond actively to the institutional context by

enhancing disclosure quality in relation to the thus-gener-

ated stakeholder power and the resulting legitimacy

requirements. Those who do not face such disclosure

pressures do not respond.

The results of the empirical tests are of interest for

several reasons. First, the significance of the model indi-

cates that stakeholder theory is an appropriate foundation

for empirical analyses of social disclosure quality in Tai-

wan. Second, our overall findings support the contingency

framework developed by Ullmann (1985) that is important

in CSD research. Third, the results compliment, in the

Taiwanese setting, Mitchell’s specification (Mitchell et al.

1997) that power and legitimacy are criteria that can be

used to identify the influential stakeholder groups. Fourth,

the results from the set of control variables indicate that

high-quality disclosure is associated with larger firms, as is

more frequent media exposure. Fifth, the results support

the argument that current period levels of CSD relate to

prior period measures of the independent and control

variables employed in this study.

The remainder of the paper is organized as follows. In

the following section we review the literature in this field

and provide relevant evidence. In the section Social Dis-

closure in the Taiwanese Context we elaborate on con-

textual factors in Taiwan and derive testable hypotheses.

This is followed by a section describing the data and

estimation methods (Data and Method of Estimation) and

the section reporting the findings, with analysis. A number

of conclusions are drawn in the final section. The Appendix

provides details on how the data elements were generated.

Literature Review: Stakeholder Theory and Relevant

CSD Evidence

This study adopts a stakeholder framework to explain the

observed differences in CSD quality across Taiwanese

companies. CSD provides information on the impact of a

company’s corporate social responsibility activities on a

broad range of constituencies or stakeholders. Stakeholder

theory is considered to be an important theory in CSD

research, and many prior studies have employed this theory

to explain the organization–society relationship (Ullmann

1985; Roberts 1992; Milne 2002; Van der Laan Smith et al.

2005; Chen and Roberts 2010; Huang and Kung 2010).

According to Freeman’s now classic definition, a

stakeholder is ‘any group or individual who can affect or is

affected by the achievement of an organization’s objec-

tives’ (Freeman 1984). Stakeholders are those who have ‘a

stake’ in an organization and have something ‘at risk’, and

they usually include stockholders, creditors, employees,

customers, suppliers, public interest groups, governmental

bodies, and the community. While the neo-classical view

of the corporation focuses on shareholder supremacy and

firm profit maximization (Friedman 1970; Jensen 2002), in

contrast, the stakeholder concept is intended to ‘broaden

management’s vision of its roles and responsibilities

beyond the profit maximization functions to include inter-

ests and claims of non-stockholder groups’ (Mitchell et al.

1997). Under this approach, the long-run survival and

Social Disclosure Quality in Taiwan

123

success of the company requires the support of its stake-

holders, and a principal function of the manager is to

handle stakeholders’ needs, expectations, and demands, as

well as to balance conflicts among them.

Based on the stakeholder theory of strategic manage-

ment, Ullmann (1985) presents a contingency model to

spell out correlations between CSD and three-dimensional

factors. The first dimension of the model, stakeholder

power, explains that the firm will be responsive to the

intensity of stakeholder demands. If social reporting is to

be an effective management strategy for dealing with

stakeholders, a positive relationship between stakeholder

power and CSD is expected. The second dimension is the

firm’s strategic posture toward social demands, which

describes the mode of response of a company’s key deci-

sion-makers to the demands of social reporting. Ullmann

(1985) dichotomizes strategic posture as active or passive.

A company whose management tries to influence their

organization’s status with key stakeholders through

developing deliberate disclosure programs, engaging in

continuous monitoring activities, and institutionalizing

reporting activities (in terms of a dedicated CSD unit and

report) possesses an active posture. Alternatively, if the

company’s management does not address to stakeholder

influences, then it is perceived to possess a passive strategic

posture. Thus, the more active the strategic posture, the

greater the expected CSD.

The model’s third dimension concerns the company’s

past and current economic performance. The financial

resources of the company are an important determinant of

CSD in two ways. First, financial resources determine the

relative weight of social demands and the attention they

receive from top decision-makers. In periods of low prof-

itability and in situations of high debt, the importance

placed on meeting CSD goals may be secondary to meeting

the economic demands that impact directly on a company’s

continued viability. Second, the availability of financial

resources affects the financial capability to undertake

costly disclosure programs related to collecting, measuring,

and auditing data. Therefore, given certain levels of

stakeholder power and strategic posture, the better the

economic performance of a company, the better its CSD.

However, given the constraints imposed by the limited

resources, in meeting the demands of the various stake-

holder groups a company’s manager may not accord all

stakeholders the same level of importance. Different cri-

teria have been suggested as determinants of how and why

managers should allocate priorities to competing stake-

holder demands, and the best way to determine the

appropriate level of management. In defining the stake-

holder–management relationship, Mitchell et al. (1997)

develop a dynamic theory of stakeholder relations and

argue that stakeholder identification and salience are a

function of the stakeholder possessing one or more of the

three relationship attributes: (1) the stakeholder’s power to

influence the firm, (2) the legitimacy of the stakeholder’s

relationship with the firm, and (3) the urgency of the

stakeholder’s claim on the firm.

Regarding the first factor in Mitchell’s specification,

resource dependency theory suggests that power accrues to

those parties who control vital resources for the firm,

thereby creating power differentials among stakeholders

(Pfeffer 1981). Furthermore, a party in a relationship has

power when it has—or can gain access to—a certain

means, such as utilitarian means based on material or

financial resources, to impose its will in the relationship.

For the second factor, a stakeholder group achieves legit-

imacy if it has legitimate claims on the firm, where the

basis of the legitimacy of the relationship may derive from

contract, exchange, legal right, legal title, moral right, or

at-risk status (Hill and Jones 1992). An entity may have a

legitimate claim on the firm, but unless it has either power

to enforce its will in the relationship or a perception that its

claim is urgent, it will not achieve salience for the firm’s

managers. Hence, the third factor, urgency, is related to the

degree to which stakeholder claims call for immediate

action. This factor can be characterized by time sensitivity

(the claims needing to be given immediate attention) and

criticality (the claims being vital and highly important).

Nonetheless, gaining stakeholders’ support and approval

requires a dialogue between the management of a company

and its stakeholders. CSD, which is publicly disclosed

information about a company’s social responsibility, pro-

vides a tool of communication between the firm and its

stakeholders (Dierkes and Antai 1985; Gray et al. 1995;

Neu et al. 1998; Deegan 2002). In this regard, corporate

managers play a unique role in stakeholder relations. They

constitute the group of individuals who make the strategic

decision(s) to allocate the limited resources of the firm in

the manner which they perceive to be most consistent with

stakeholder claims. The relationship factors, power, legit-

imacy, and urgency influence the level of importance a

company attaches to the claims of different stakeholder

groups. The more critical the manager perceives the

stakeholders’ claims to be, the higher the chances that the

stakeholder demands will be addressed. Viewing CSD as a

strategic plan by companies to manage stakeholder rela-

tions, we can expect a positive relationship between the

level and quality of social disclosures and the importance a

company attaches to its stakeholders.

In addition to stakeholder identification, Mitchell et al.

(1997) further generate a typology of stakeholders based on

the possession of one, two, or all three of these attributes

and then prioritize their salience to managers of the firm.

One class of stakeholders, that composed of the dominant

stakeholders, is of particular interest to us in the present

T.-K. Chiu, Y.-H. Wang

123

context. This is the situation where stakeholders are both

powerful and legitimate, and the combination of two

attributes leads the stakeholders to adopt an active stance,

with a corresponding increase in firms’ responsiveness to

the stakeholder’s interests. Thus, we expect that the firm

will have formal mechanisms in place that not only

acknowledge the importance of its relationship with dom-

inant stakeholders but which also constantly monitor the

relationship. For instance, companies that acknowledge the

importance of social reporting demanded by legitimate,

powerful stakeholders may organize a special team to

centralize the data collection system to measure, verify,

collate, and publish such information. Moreover, when

dominant stakeholders acquire the last attribute, urgency,

they will become highly salient stakeholders, and managers

will have an immediate mandate to both attend to and give

priority to that stakeholder’s specific claim.

In applying extant theories to specific regions and

countries, prior studies also stress the importance of taking

a more contextual and nationally contingent approach to

social responsibility (Chapple and Moon 2005; Matten and

Moon 2008). In the present setting, contextual factors, such

as socioeconomic background, political standing, legal

framework, and the tradition of interrelations among dif-

ferent social actors, may determine the power of a stake-

holder group and the legitimacy of its claims, which in turn

affect managers’ response to their relevant stakeholders

through the extent and quality of social disclosures. It is

important to explain the conditions under which managers

do consider certain classes of entities as stakeholder.

Therefore, in the following section we examine the con-

textual factors in Taiwan that could impact the constella-

tion of dominant stakeholders, as well as the nature of the

relationships between firms and these stakeholders. The

dominant stakeholder groups thus identified will be used to

represent stakeholder power in Ullmann’s model (Ullmann

1985).

Social Disclosure in the Taiwanese Context:

Development of Hypotheses

Contextual Factors

Taiwan is one of the leading trading nations in East Asia,

and its economic development is based on a well-devel-

oped education system, a strong research and development

sector, a comparatively sound banking and financial sys-

tem, and a technology-intensive model predominantly

based on small- and medium-sized enterprises (Lau 1994).

The Taiwan has a highly export-led economy and has a

well-developed electronics industry serving as outsourced

vendors for the global IT industry (Sturgeon and Lee

2005). From 2008 to 2012, electronics exports consistently

accounted for 40 % of total exports, and roughly one-half

of the companies listed on the TWSE are from the elec-

tronics industry. 1

However, the country has a long

unsolved issue of national identity with a unique political

status in the world. Taiwan is not admitted into interna-

tional and regional organizations and bodies (such as the

United Nations and the World Bank) in a state capacity

because of China’s resistance to national recognition. This

means that Taiwan is not entitled to sign global agreements

(such as the Kyoto Protocol and trade agreements) and is

thus not constrained by related international regulations.

In terms of domestic regulations, it seems reasonable to

characterize Taiwan as a country of insufficient regulations

and lax enforcement, which in turn implies that the

domestic legislative and regulatory bodies are not an

influential stakeholder group for Taiwanese firms. Legis-

lative paralysis due to continuous bickering between the

camps of the two major political parties (the KMT Party

and the Democratic Progressive Party) in the parliament

has resulted in the shelving of many important laws, of

which the Greenhouse Gas Reduction Act is one. A lack of

effective carbon emission regulations and a failure on the

part of the authorities to fulfill their duties has created a

situation where firms face minimal disclosure requirements

and minimal regulatory pressures from local governments.

The Taiwanese government has also failed to enact

significant legislation relating to corporate governance.

Instead, it has formulated a number of voluntary disclosure

rules, such as the Regulations Governing the Exercise of

Powers by Audit Committee of Public Companies. For

CSR activities in general, the TWSE issued a Code for

CSR practice in 2010, but its application remains on a

‘comply or explain basis.’ Although the Financial Super-

visory Commission encourages public firms to disclose

CSR activities in annual reports, the majority of companies

report their CSR cursorily in a few sentences.

Socially responsible investment (SRI) has recently

emerged as an alternative to the traditional investment

dogma, emphasizing accountability, transparency, and

ethical business practices (Starr 2008; Louche and Ly-

denberg 2010). Institutional investors are viewed as

important drivers of SRI in Europe and the USA (Cox and

Wicks 2011; EUROSIF 2012). However, the stock market

in Taiwan is still dominated by small individual investors

with a low social awareness and limited concern for

1 Electronics exports as a percentage of total exports came to

38.66 % in 2008, 40.50 % in 2009, 41.65 % in 2010, 40.03 % in

2011, and 41.37 % in 2012. These data are website-referenced:

statistical data provided by Director-General of the Office of Budget,

Accounting, and Statistics of Executive Yuan at http://www.dgbas.

gov.tw/ct.asp?xItem=28536&ctNode=3292&mp=1 (Accessed 10 Jan

2013).

Social Disclosure Quality in Taiwan

123

environmental issues, making SRI a rare investment prac-

tice. This can be seen from the continuous shrinking of the

asset volume of the only SRI fund in Taiwan, the Fubon

CSR fund, since its initiation in 2009. Hence, it is rea-

sonable to say that, at least in Taiwan’s local market, SRI

has not been a driver of CSD.

Watchdog groups, including civil society groups and

nongovernmental organizations (NGOs), are often viewed as

important players in the business–society relationship

(Teegen et al. 2004; Chen and Bouvain 2009; Arenas et al.

2009). However, these groups are weak in Taiwan. Con-

sumers and environmental NGOs are largely inactive, using

mild methods with moderate appeals. Of the internationally

leading environmental organizations Greenpeace and

Friends of the Earth, the former has carried out projects in

Taiwan but was unable to bring about any significant

changes to the operating methods of local firms, while the

latter has not established a local branch. As regards the local

Environmental Protection Union, it has only ever appealed

to the government; there is as yet no record of any petition

directed against corporations. Taiwan’s citizens have

established various local environmental organizations, such

as the Green Alliance, to raise awareness of climate change

and promote energy-saving. However, it can be inferred

from their websites and the frequency of updates that very

few activities have been carried out.

Overall, the extent and quality of social disclosure varies

substantially across Taiwanese firms. This wide range in

social disclosure seems to be related to the various degrees

of pressure that firms face from, and the various degrees of

responsiveness that firms display to, influential foreign

stakeholders. For example, the disclosure quality of elec-

tronics firms operating as outsourced vendors for interna-

tional brands is quite satisfactory, particularly in

environmental areas, as they are subject to more stringent

standards and face higher reporting pressure. On the other

hand, disclosure practices in those Taiwanese industries

which have difficulty extending their business overseas and

thus are not yet in line with international standards, such as

the financial sector, are lagging behind.

Specifically, the very first piece of information that pro-

fessional buyers in the global supply chain ask Taiwanese

firms to provide is their answers to, and scores on, the survey

questions of the Carbon Disclosure Project (CDP). CDP is an

international NGO providing a global system for firms to

measure, disclose, manage, and share vital environmental

information, with the goal of ‘[facilitating] a dialogue, sup-

ported by quality information, from which a rational

response to climate change will emerge’ (CDP 2008). The

CDP questionnaires reflect a best practice framework for the

information that companies should measure and report

regarding the impact of climate change on their businesses.

For example, in CDP 6, firms are asked to identify strategic

risks and opportunities, to report actual absolute greenhouse

gas emissions, to compare performance against plans to

reduce emissions, and to elaborate responsibility and man-

agement approach to climate change (CDP 2008).

There are other noneconomic criteria to be met by Tai-

wanese vendors in the global supply chain, such as the Elec-

tronic Industry Code of Conduct and a number of ethical

souring codes of conduct. Moreover, buyers normally request

that potential vendors provide stand-alone CSR reports which

need to follow the disclosure guidelines of the Global

Reporting Initiative (GRI) Institute, the standard-setting

organization for corporate social reporting, with complete and

concrete information on environmental impact, human rights,

employee relations, and product liability. As a result, Tai-

wanese electronics companies like ASUS, LiteOn Group, AU

Optronics, and Wistron all face such pressure and are effec-

tively forced to comply to avoid losing orders.

Prior studies have shown that international capital

markets can be a driver of CSD in emerging economies

(Xiao et al. 2004; Haniffa and Cooke 2005). A large

number of Taiwanese companies are either listed on

international stock exchanges (such as the New York Stock

Exchange) or selected as components of global ethical

indices [such as the Dow Jones Sustainability Index

(DJSI)] to access capital worldwide. Their reporting quality

is superior to that of their counterparts for both ex-ante and

ex-post reasons. Ex-ante, firms with poor disclosure quality

cannot meet the requirements of renowned international

exchanges or pass the screening threshold of global ethical

indices. Taking the DJSI as an example, disclosure quality

is regarded as a precondition for being socially responsible

by SAM, a social rating agency which carries out research

and data gathering in accordance with defined methodol-

ogy to select constituent companies for the DJSI. Firms are

evaluated on the basis of the content, context, and coverage

of the environmental and social reporting included on their

websites and in dedicated reports, such as the CSR reports,

annual reports, and other special publications (SAM 2011).

After being listed or selected, firms will attract more ana-

lyst coverage as well as being subject to higher disclosure

standards; this is the ex-post reason.

Development of Hypotheses

Below, we spell out the explanatory variables and derive

testable hypotheses. There are three stakeholder power

variables, one strategic posture variable, two economic

performance variables, and two control variables.

Stakeholder Power Variables

Ullmann presents stakeholder power as the first dimension

of the model, but provides little guidance on how to

T.-K. Chiu, Y.-H. Wang

123

identify this in reality. Here, we employ Mitchell’s speci-

fication (Mitchell et al. 1997) and argue that there are three

stakeholder groups which should be included in the Tai-

wanese setting. Among the three, two groups possess the

attributes of power and legitimacy, and are thus dominant

stakeholders.

Dominant Stakeholder Group: Buyers in the Global

Supply Chain

When environmental and social concerns become noneco-

nomic criteria in socially responsible organizational buying,

ethical sourcing forms the basis for buyers to select vendors

in the global supply chain (Drumwright 1994). Buyers in the

global supply chain control contracts and orders that may be

necessary for the continued operation of a company. In

particular, large customers and buyers can often influence

corporate behavior through the exercise of bargaining/buy-

ing power in contractual relationships. Among other factors,

such buyers/customers are concerned with working condi-

tions, labor rights, code of conduct, and environmental

issues, all of which have a direct effect on financial perfor-

mance and/or their reputation and, consequently, their own

image (Roberts 2003; Islam and Deegan 2010). They thus

tend to inquire about the ethical corporate practices of their

clients and demand transparency as well as quality reporting.

In the case of Taiwan, buyers in the global supply chain

form a dominant stakeholder group that possesses both

power and legitimacy: the utilitarian power of these buyers

is based on their bargaining strength; their contracts and

orders make their disclosure requests legitimate claims.

The combination of two attributes achieves salience of this

group for the firm’s managers, with a corresponding

increase in their responsiveness to its reporting quality

demands. This results in the following hypothesis:

H1 The quality of CSD tends to be higher in firms

associated with reporting pressure from the global supply

chain.

Dominant Stakeholder Group: Listing and Social

Rating Agencies in International Capital Markets

Freeman (1984) recognized the ability of government to

influence corporate strategy and performance and discussed

the role of legislative bodies as stakeholders whose inter-

ests must be addressed by management. Firms may use

disclosure as a strategy to satisfy government demands and

to reduce the risk of governmental intrusions that may

adversely affect the firm’s value. In the case of firms in

emerging economies with insufficient domestic regula-

tions, as they increase their exposure to international stock

markets, they are likely to change their disclosure practices

in response to the higher standards being imposed (Scott

1994; Chapple and Moon 2005; Haniffa and Cooke 2005).

For Taiwanese companies, regulators in international

capital markets, including listing agencies and social rating

agencies, can be characterized as a dominant stakeholder

group in deference to the legitimate reporting claims they

have upon the listed/selected or to-be listed/selected firms

and their control of a vital resource, capital, as well as their

ability to exercise power. In Mitchell’s typology (Mitchell

et al. 1997), the two-attribute stakeholders expect their

claims to be fulfilled and receive much of managers’

attention, with a corresponding increase in firms’ respon-

siveness to the stakeholder’s interests. Thus, the level of

engagement between management and these dominant

stakeholders is likely to be higher. Accordingly,

H2 The quality of CSD tends to be higher in firms

associated with reporting pressure from international cap-

ital markets.

Influential Stakeholder Group: Shareholders

The degree to which the ownership of company stock is

concentrated in the hands of a few large investors or dis-

persed among many has been proposed as an influence on

disclosure policy (Ullmann 1985; Roberts 1992; Cormier

et al. 2005; Brammer and Pavelin 2008; Gamerschlag et al.

2011). Disperse corporate ownership, especially by inves-

tors with social concerns, heightens the pressure on man-

agers to enhance the extent and quality of CSD. On the

other hand, the responsiveness of firms with a concentrated

ownership structure to the public’s request for information

might be lower, since the controlling shareholders can still

obtain the needed information through special channels,

and this tends to reduce the firm’s incentive to disclose.

Following Ullmann (1985), we include the shareholding

stakeholder group in the disclosure model and hypothesize

that the wider the dispersion of corporate ownership, the

better the quality of social disclosure. Hence, we test the

following hypothesis:

H3 The quality of CSD tends to be higher with increasing

dispersal of share ownership.

Strategic Posture Variable

The firm’s strategic posture toward social reporting

describes its mode of response, which can be characterized

as active or passive. In the case of dominant stakeholders,

the combination of power and legitimacy leads the domi-

nant stakeholders to adopt an active stance, with a corre-

sponding increase in the firm’s responsiveness to the

stakeholder’s interests. A manager trying to influence his

organization’s relationship with these stakeholders may

Social Disclosure Quality in Taiwan

123

take an active posture by developing deliberate disclosure

programs and engaging in continuous monitoring activities.

We may even expect that the firm will institutionalize the

practice by having formal mechanisms in place to measure,

verify, collate, and publish the information. Alternatively,

if the manager does not address the stakeholder influence,

then the firm is perceived to possess a passive strategic

posture. This results in the following hypothesis:

H4 The quality of CSD tends to be higher in firms

associated with a positive strategic posture toward CSD.

Financial Resource Variables: Profit and Leverage

We referred above to the importance of the third dimen-

sion, i.e., the financial resources of the company, in Ull-

mann’s three-dimensional contingency framework

(Ullmann 1985). Several authors have explored the role of

surplus financial resources, or organizational slack, in

determining the firm’s strategies for social disclosure, and

have found that the presence of slack promotes CSD

(Cormier et al. 2005; Brammer and Pavelin 2008; Huang

and Kung 2010). We propose two measures of corporate

financial health for inclusion in this study to test the impact

of prior economic performance on a company’s social

disclosure quality: (1) the company’s profitability and (2)

its leverage. Stakeholder theory predicts a positive asso-

ciation between the company’s profitability and social

disclosure quality, and a negative association between the

company’s leverage and social disclosure quality. We thus

have the following hypotheses:

H5 The quality of CSD tends to be higher with increasing

profitablity of the firm.

H6 The quality of CSD tends to be lower with increasing

leverage of the firm.

Control Variables

Organizational size and visibility are often proposed as firm-

level drivers of social disclosure, and many researchers have

suggested that firm size and media exposure are likely to act

as intervening variables and should be controlled for in

empirical tests (Roberts 1992; Neu et al. 1998; Purushoth-

aman et al. 2000; Brammer and Pavelin 2008; Branco and

Rodrigues 2008; Reverte 2009). Also, arguments can be

made that firm size and visibility represent some aspect(s) of

stakeholder power, strategic posture, and economic

resource. Hence, we include both firm size and media

exposure as control variables and hypothesize that they are

positively related to disclosure quality:

H7 The quality of CSD tends to be higher with increasing

size of the firm.

H8 The quality of CSD tends to be higher with increasing

media exposure of the firm.

Data and Method of Estimation

We use a hand-collected set of specific CSD data. The

sample is based on firms listed on the TWSE, with data

being extracted from these firms’ publicly disclosed

information, including corporate websites, annual reports,

and CSR reports, by means of content analysis. Data on the

underlying firm-specific characteristics are taken from the

Taiwan Economic Journal (TEJ) database.

As of the end of 2010, there were 758 firms listed on the

TWSE. We divide these into three groups based on reve-

nues and randomly select one out of every three firms from

each of the high, medium, and low revenue groups. This

results in 84 firms for each revenue group, with a total of

252 firms. After deleting firms with missing data, we are

left with 82 firms in each group, making the final sample in

this study 246 firms, with a total of 492 firm–year

observations.

Table 1 presents the industries and the number of firms

analyzed in our sample. As noted, our sample is drawn

from a diverse range of industries, including the food and

beverage, textile, automotive manufacturing and steel

industries, among many others, but electronics (122 firms)

accounts for half of the sample analyzed. As of the end of

2010, half of the companies listed on the TWSE were

electronics firms (379/758 firms). By this yardstick, the

sample analyzed is broadly representative.

In 2010 and 2011, only 40 and 39 sample firms,

respectively, published separate CSR reports. Of those

firms not issuing reports, a number published CSR sections

on company websites, but others lacked any coverage at

all. As listed companies are required to issue annual reports

and financial reports, all sample firms provided such data.

We use 31 December of each year as the cutoff date for

data, as firms issue CSR reports for the previous year in the

latter half of the year.

Dependent Variables: CSD Quality

Empirical study of the determinants of disclosure quality

requires the use of a suitable method for measuring quality,

or the quantification of the qualitative. CSD studies often

employ a form of conceptual content analysis of disclosed

information whereby the existence or incidence of a partic-

ular predetermined concept is recorded and analyzed

(Krippendorf 2004). This technique has been widely used to

assess the extent of disclosure (e.g., Gray et al. 1995) and

includes the counting of words, sentences, and proportion of

pages, depending on the unit of analysis. However, it should

T.-K. Chiu, Y.-H. Wang

123

be noted that the quality of disclosure and the extent of dis-

closure may not be equal measures of quality. Extent often

refers to the amount, but lengthy reporting may contain

irrelevant information or use misleading words, resulting in

incomplete or poor quality disclosure.

This paper focuses on CSD quality and extends the

analyses of published studies (Cormier et al. 2005; Van der

Laan Smith et al. 2005; Brammer and Pavelin 2008). We

keep the quality concept in line with content analysis (e.g.,

Deegan and Gordon 1996; Patten 2002) as well as the

reporting requirements of the GRI Institute as specified in

guidelines G3-1 (Global Reporting Initiative 2011). Here,

standard disclosure covers strategy and profile, manage-

ment approach, and performance indicators, as organiza-

tions are required to disclose policies and goals, describe

initiatives, quantify impacts of operation, report progress,

and present external assessment.

We use a coding scale to quantify the quality of infor-

mation and provide justifications, as follows. First, while

many other disclosure studies focus on the amount of

disclosure, we employ a ‘qualitative scale’ to ensure that

irrelevant or redundant information is not considered.

Second, information from a firm’s CSR reports, annual

reports, and websites is read and analyzed to arrive at a

comprehensive disclosure quality score. This process

allows for the integration of various types of information

into a single score to facilitate cross-firm comparison.

From the aforementioned contextual factors it is clear

that, as Taiwanese firms move from the local to the global

environment, the type of social reporting required by pro-

fessional buyers in the supply chain and listing agencies in

international capital markets is pro-active, timely, concrete,

accurate, and verifiable. This justifies our use of the qual-

ity—rather than the extent-of disclosure as the focus of

analysis. Hence, disclosure quality, the dependent variable

in the model, is measured by an index score. Five facets of

quality are considered, including the firm’s disclosure

of CSR policies, goals, and initiatives, quantification of

impact and reporting of progress, and presentation of

external certificates and audits.

The index score is obtained from data elements gener-

ated from a set of survey questions in line with the inter-

national trends on social reporting, so that the differences

in reporting the quality of firms with various degree of

pressure from, and various degree of responsiveness to, the

international environment can be better contrasted. For this

purpose, we use the well-known 100 Best Corporate Citi-

zens (or 100 Best) rating and re-interpret this as a ‘dis-

closure rating.’ 2

100 Best is a CSR rating of U.S. firms by

CR magazine, in which the best 100 CSR performers from

the Russell 1000 index are cited. Being guided by the

American Corporate Responsibility Officers Association

(CROA), its Methodology Committee includes a wide

range of experts and is thus highly respected (Brammer

et al. 2009; SustainAbility 2010; Scalet and Kelly 2010). 3

The method is rigorous and objective, as it is announced

openly on the magazine’s website, and the survey questions

reflect the CSR practices of industrialized countries and the

related disclosure trends.

The data are generated from survey questions on several

CSR categories. Data in each category fall into one of the

following three subcategories: policy, disclosure, and per-

formance. After our amendments and re-interpretation, the

2 For details on how to construct a corporate social performance

database based on the methodology and survey questions of the 100

Best, please see Chiu (2014). This paper also explains why part of the

data thus obtained can be reinterpreted as CSD data. 3

Information on the methodology and protocols for 2012 can be

found at the following URL: http://www.thecro.com/files/CR%

20Magazine%20Corporate%20Citizen%20Methodology%202012.

pdf and http://www.thecro.com/files/CR%20Corporate%20Citizen

ship%20Protocol,%202012.pdf. Accessed 10 Jan 2013).

Table 1 Sample description

Sample size

Number of firms initially included in the analysis (randomly

selected from 758 firms listed on the TWSE in 2010)

252

Less firms lacking data 6

Total number of firms included in the final analysis 246

Industry-wide distribution a

Food & beverage 8

Plastics 7

Textiles 14

Electrical equipment & machinery 12

Electronics b

122

Chemical 7

Glass & ceramics 1

Steel 10

Rubber 4

Automotive 1

Construction & materials 13

Biotech & healthcare 5

Transportation & logistics 8

Oil & gas 5

Finance & insurance 12

Trading 4

Tourism 1

Others 12

a The industry affiliation recorded in the Taiwan Economic Journal

(TEJ) data based follows the industrial classification of the Taiwan

Stock Exchange (TWSE) with minor revisions b

The electronics industry includes semiconductors, electronic com-

ponents, computer and peripherals such as IT (telecommunications,

and optoelectronics)

Social Disclosure Quality in Taiwan

123

data used in this study come from a set of 291 questions

(items) covering five facets of disclosure quality: policy,

target, initiative, progress, and external certificates and

audits. Table 2 presents the six CSD categories, climate

change abatement, employee relations, environmental

impact, corporate governance, human rights, and philan-

thropy, the five quality facets for each category, and the

number of survey questions for each facet in a given cat-

egory. The Appendix provides the details of the amend-

ments and a set of sample questions.

There are several points to note from Table 2. First, no

questions on policies, goals, or certificates are asked in the

category of corporate governance and employee relations.

This absence is due to the obvious reasons that most

countries have established mandatory rules in these areas

and required firms to set up relevant systems, implying that

questions on ‘voluntary’ certification are irrelevant. Sec-

ond, in terms of employee relations, most states drew up

detailed rules and regulations long ago, so the questions

focus on voluntary initiatives rather than on the required

corporate policies and goals. Third, the focus with respect

to philanthropy is on policy, as the voluntary nature of

philanthropic deeds makes ‘certificates and audits’ inap-

propriate. Fourth, the category of climate change abate-

ment has no questions on ‘certificates and audits’ for

similar reasons.

To enhance the reliability of the measurements, we

followed previous studies and trained two groups of

assistants and asked them to calculate CSD scores sepa-

rately (Cormier et al. 2005). The training includes clarifi-

cation of the questions, explanation of related regulations,

and deciphering of the CSR reports. Reliability tests indi-

cate that the between-group average variance is lower than

1 %, which shows a strong scoring consistency.

Before analyzing the potential determinants of CSD, we

need to look at the general disclosure conditions in Taiwan.

Table 3 reports the survey results of the six disclosure

categories for 2010 and 2011. As the results illustrate,

social disclosure of Taiwanese firms is unsatisfactory; on

average, firms can only answer about 15 % of the 291

questions, with the best responder disclosing 164 data

elements in 2011 and the worst 22 in 2010. Among the

various categories of CSD, the categories of ‘climate

change abatement’ and ‘environmental impact’ have a

quite poor disclosure rate, with an average of 6 or 7 %, as

well as a wide cross-firm variation. In contrast, social

disclosure on ‘employee relations’ is much better with an

average reporting rate of approximately 40 %. The results

are consistent with previous findings: environmental reg-

ulations and stakeholder scrutiny in emerging countries are

insufficient, resulting in sub-par disclosure. In contrast,

firms are quite skilled in managing employee relations, and

disclosure certainly reflects this. Moreover, countries in the

early stage of CSD development tend to attach more

importance to employee relations (Chapple and Moon

2005; Khan et al. 2013), with environmental disclosure

being not only brief, but also full of vague language and

broad terms (Haniffa and Cooke 2005; De Villier and Van

Staden 2006; Mahadeo et al. 2011).

The result for ‘philanthropy’ is counter-intuitive as only

an average of two items are disclosed out of eight in that

category. This seems to run contrary to the large volume of

information on philanthropic activities, such as holding

cultural, volunteer, and disaster relief activities, as released

on company websites and CSR reports. The international

mainstream practice is to ‘integrate charitable activities

into a firm’s core businesses, and provide relevant products

and services,’ and a disparity between the awareness of

such activities by Taiwanese firms and the mainstream

does exist. For example, none of the key performance

indicators in the GRI G3-1 is related to ‘philanthropy’

(Global Reporting Initiative 2011), and the survey ques-

tions of the 100 Best mainly focus on philanthropic policy.

Taiwanese firms, however, have not updated their prac-

tices, and their out-dated concepts as well as lack of policy

formulation at the organizational level all result in poor

disclosure quality in this category.

Regarding corporate governance, firms on average could

only answer 0.15 questions in 2010, but the number

increased to 1.01 in 2011. This is due to the fact that a

number of firms set up independent audit committee or

compensation committee in 2011, driving up the disclosure

rate in this category. As regards environmental impact and

climate change, we note that these are the poorest disclo-

sure areas. The high carbon emissions as well as poor

corporate reporting of related policies and goals reflect the

insufficiency of the governmental regulations in Taiwan as

Table 2 Corporate social disclosure categories, quality facets, and number of questions

Facet Category a

CC ER EN CG HR PH

Policies 18 0 19 0 27 8

Goals 1 0 1 0 1 0

Initiatives 2 45 18 6 3 0

Progress reports 36 14 76 0 5 0

Certificates and audits 0 0 10 0 1 0

Total number of

questions

57 59 124 6 37 8

% (of total of 291

questions)

19.59 20.28 42.61 2.06 12.71 2.75

a Categories: CC, Climate change abatement; ER, employee rela-

tions, EN, environmental impact; CG, corporate governance; HR,

human rights; PH, philanthropy

T.-K. Chiu, Y.-H. Wang

123

well as their poor enforcement. Good scores seem to be

driven by disclosure demands from foreign stakeholder

groups. A few firms have made significant improvements

in environmental disclosure from 2010 to 2011, which is

reflected in the substantial increase in the maximum scores

obtained.

Measurement of Independent Variables

With some ideas about social disclosure in Taiwan, we are

now in a position to examine the potential determinants of

CSD quality. Data are either taken from the TEJ or

extracted from firms’ annual reports, with variables being

measured as follows: firm size is measured by total assets,

financial leverage by the long-term debt-to-equity ratio,

corporate profits by return on assets, and deviations from

ownership by the percentage of controlling shareholding

divided by earning distribution ratio. As regards firm

exposure, we use the incidence of news media coverage of

the firm as a measure, calculated as the average number of

times the respective firm was reported in Taiwan’s two

best-selling newspapers (United Daily News and China

Times). For reporting pressure from international capital

markets on CSD quality, we use a binary dummy variable

(1 for yes, 0 for no) to indicate whether the company

possesses an international listing status or is a constituent

on a global ethical index, as a measure of the firm’s dis-

closure pressure in foreign capital markets.

The measure of buyers in the global supply chain as a

dominant stakeholder group is difficult to operationalize.

Previous studies have used advertising fees as a variable to

represent the strength of customer relations (Huang and

Kung 2010). This variable is not suitable in our setting as,

firstly, we are more concerned with demands from orga-

nizational buyers than from individual consumers, and,

secondly, firms’ fee data are internal and therefore hard to

obtain. However, from the perspective of Taiwanese firms,

we can use the electronics industry as a measure to reflect

the perceived strength of regulatory pressure imposed by

the global supply chain on CSD quality.

This measure also serves an additional function: prior

studies have shown that the nature of business activities or

industry affiliation (high or low environmental impact) can

affect a firm’s disclosure practice. Therefore this infor-

mation needs to be either controlled as an intervening

variable or included as an explanatory variable, depending

on the theoretical framework (Deegan and Gordon 1996;

Patten 2002; Brammer and Pavelin 2008; Gamerschlag

et al. 2011). As defined by Taiwan’s Bureau of Energy

based on the country’s level of industrialization with its

corresponding industrial structure, the four high environ-

mental-impact industries are electronics, paper and pulp,

chemical materials, and non-metallic minerals. 4

However,

the electronics industry and high environmental-impact

industries are highly correlated and, therefore, to include

the former as an explanatory variable and the latter as a

control variable in the same model may result in significant

Table 3 Survey results of the six disclosure categories (2010,

2011)

Disclosure categories Maximum

(N)

Minimum

(N)

Mean

(N)

Median

(N)

Mean/total

questions (%)

Year: 2010

Climate change abatement: 57 questions 38 0 4.12 0 7.23

Environmental impact: 124 questions 51 0 7.67 3 6.19

Employee relations: 59 questions 41 16 23.43 23 39.71

Human rights: 37 questions 15 3 3.88 3 10.49

Corporate governance: 6 questions 2 0 0.15 0 2.50

Philanthropy: 8 questions 2 0 1.89 2 23.63

Total: 291 questions 132 22 41.13 31.00 14.13

Year: 2011

Climate change abatement: 57 questions 42 0 4.43 0 7.77

Environmental impact: 124 questions 77 0 8.84 3 7.13

Employee relations: 59 questions 43 17 24.55 24 41.61

Human rights: 37 questions 16 3 5.27 4 14.24

Corporate governance: 6 questions 2 0 1.01 1 16.83

Philanthropy: 8 questions 4 0 1.81 2 22.63

Total: 291 questions 164 24 45.91 34 15.78

4 The definition is based on the 2009 Energy Statistics Handbook

released by the Bureau of Energy of the Taiwanese Ministry of

Economic Affairs (page 18, note 1). The energy consumption of these

four industries accounts for 65 % of the total for all manufacturing

industries in Taiwan.

Social Disclosure Quality in Taiwan

123

statistical difficulties. As the electronics industry is the

most internationally exposed among the four, and the one

most expected to enhance disclosure quality, we have

selected this industry as a proxy for the strength of

reporting pressure from the global supply chain, which

suffices to fulfil the aforementioned function as well.

As to a proxy for the strategic posture variable, previous

studies have used philanthropic activities (Roberts 1992;

Muller and Kräussl 2011), public affair officers (Roberts

1992), or press releases (Magness 2006). In our model,

when dominant stakeholders are seen as expecting dedi-

cated reporting, a company with active strategic posture

would set up a formal department or organize a special-

purpose team to engage in the continuous measuring,

verifying, collating, and publishing of such information.

For companies which have independent CSR departments

in place, the handling deliberate disclosure programs is

often part of the departmental function. Moreover, we

distinguish those that use the relatively traditional philan-

thropic mode from those that better institutionalize and

embed the ways in which their CSR is employed. In this

sense, a positive posture is only captured by the existence

of independent CSR departments or special-purpose teams,

but not by company philanthropic foundations which

mostly focus on cultural and educational activities in the

Taiwanese setting. Data on the firms’ strategic posture are

taken from the annual reports.

Regression Model

The statistical analysis in this study includes the use of

linear regression models to analyze the relationship

between CSD index scores and the influencing factors

referred to in the previous section, including three stake-

holder power variables, one strategic posture variable, two

economic resource variables, and two control variables.

The dependent variables are the four quality index scores:

total disclosure (TOTD), climate change abatement and

environmental disclosure (CCEND), employee relations

disclosure (ERD), and human rights disclosure (HRD).

They should be read together with Table 4, where CCEND

is the sum of ‘climate change abatement’ and ‘environ-

mental impact’, and the other three are self-explanatory.

Table 4 Description of variables, expected signs, and data sources

Variable (expected sign) Description Data source

Dependent variable

TOTD Total disclosure Hand collected

CCEND Climate change abatement and environmental impact disclosure Hand collected

ERD Employee relations disclosure Hand collected

HRD Human rights disclosure Hand collected

Independent variable

Stakeholder power

Sup (?) Using the electronics industry as a measure to capture the impact

of the global supply chain on CSD quality: 1 = yes,

0 = otherwise

TEJ

Cap (?) Using foreign listing or global ethical index component as a

measure to capture the impact of international capital markets on

CSD quality: 1 = yes, 0 = otherwise

Annual reports

Own (?) Deviations from ownership (percentage of controlling shares

divided by earning distribution ratio)

TEJ

Strategic posture

Dept (?) Existence of an independent CSR department or a special-purpose

team: 1 = yes, 0 = otherwise

Annual reports

Economic resource

Prof (?) ROA, measured as ratio of income (after taxes before interests) to

total assets

TEJ

Lev (-) Ratio of debt to net worth TEJ

Control variable

Size (?) Total assets TEJ

Media (?) The average number of instances of being reported in Taiwan’s

two leading newspapers

United Daily News

and China Times

CSD corporate social disclosure, CSR corporate social responsibility, ROA return on assets, TEJ Taiwan Economic Journal

T.-K. Chiu, Y.-H. Wang

123

Throughout our analysis, we introduce a 1-year lag

between the dependent and independent variables, i.e.,

disclosure quality in 2010 and 2011 is determined by

stakeholder power, strategic posture, economic resources,

as well as control variables in 2009 and 2010, respectively.

This lag exists due to the dynamic nature of strategic

planning, the focus of stakeholder theory on meeting the

long-term interests of stakeholders, and the empirical

findings of several previous studies (Roberts 1992; Bram-

mer and Pavelin 2008).

To test the hypotheses, our full model specifications

have the following structure:

Disi;t ¼b0 þ b1Supt�1 þ b2Capt�1 þ b3Ownt�1 þ b4Deptt�1 þ b5Proft�1 þ b6Levt�1 þ b7Sizet�1 þ b8Mediat�1 þ et

where Disi is the ith disclosure quality index score, Sup is a

binary dummy variable for pressure from the global supply

chain, Cap is a binary dummy variable for pressure from

international capital markets, Own is stockholder power,

Dept is a binary variable for the firm’s strategic posture,

Prof is profits, Lev is financial leverage, Size is firm size,

and Media is media exposure. Table 4 presents the various

explanatory variables, expected signs, and data sources.

Analysis of Empirical Results

Before turning to the regression analysis, upon which we

base our hypothesis tests, we present, in Table 5, descrip-

tive statistics of the variables in our study based on 492

firm–year observations. As can be seen from the average,

maximum, and minimum number of items disclosed by

firms, disclosure quality is poor in general with large dif-

ferences across firms. Using TOTD to illustrate, reporting

is poor when the data show large variations: on average

they disclose 43.53 items (14.96 %) out of the total 291

questions; the best performer discloses 174 items

(59.79 %), and the worst 22 items (7.56 %). The result

should be representative, since the sample analyzed con-

sists of firms of various sizes.

Among the descriptive statistics, the mean of Sup indi-

cates that 50 % of sample companies are under reporting

pressures from buyers in the global supply chain. The mean

of Cap illustrates that 12 % of the companies sampled that

have to cope with reporting requirements from interna-

tional capital markets face more stringent standards than

local ones. As regards the strategic posture variable, Dept,

its mean shows that 13 % of sample companies exhibit a

positive strategic posture toward CSD by having formal

mechanisms in place. The skewness and kurtosis indices

and Kolmogorow–Smirnov normality tests (or K–S Lil-

liefors) indicate that our data for continuous independent

variables are not normally distributed. As such, following

previous studies, we use Van der Waerden transformation

to convert them to normal scores before conducting the

regression analysis (see, for example, Haniffa and Cooke

2005; Mahadeo et al. 2011). Also, relative to other inde-

pendent variables, the numerical values for firm size are

huge, which might cause excessive impacts on the overall

estimation results. To avoid this possibility, we use the

natural logarithm of the variable in the regression.

Table 6 presents the bivariate correlations between the

dependent variables, the four disclosure index scores

(TOTD, CCEND, ERD, and HRD), and the independent

variables. With the exception of the economic resource

variables Prof and Lev, all bivariate correlations between

disclosure quality indices and the independent variables

possess the expected sign, indicating that the related

hypotheses will be supported. The stakeholder power

variables Sup, Cap, and Own, the strategic posture variable

Table 5 Descriptive statistics of dependent and independent

variables

*** Significance is \ 0.01 according to the Kolmogorow–

Smirnov (K–S Lilliefors) test

for normality, therefore the data

are not normally distributed a

See Table 4 for a complete

description of the variables

Independent and

dependent variables a

Mean Standard

deviation

Minimum Maximum Skewness Kurtosis K–S

(Lilliefors)

TOTD 43.53 27.90 22 174 2.19 4.00 0.33***

CCEND 12.53 21.84 0 117 2.28 4.43 0.35***

ERD 23.99 4.15 16 43 1.21 2.11 0.17***

HRD 4.58 2.92 3 16 2.37 4.83 0.38***

Sup 0.50 0.46 0 1 -0.85 -1.28 0.44***

Cap 0.12 0.32 0 1 2.38 3.67 0.53***

Own 2.49 4.79 0 50.44 6.21 48.78 0.38***

Dept 0.13 0.34 0 1 2.21 2.88 0.52***

Prof 5.66 9.06 -34.82 85.76 1.18 15.11 0.12***

Lev 147.76 377.63 2.33 4376.57 5.77 42.61 0.36***

Size (1b NT) 100.24 424.48 0.13 5005.40 7.65 70.98 0.41***

Media 26.72 77.45 0 568 4.33 20.30 0.38***

Social Disclosure Quality in Taiwan

123

Dept, and the control variables Size and Media are all

significantly correlated with the dependent variables. For

the economic resource variables Prof and Lev, some of the

correlations with the dependent variables possess the

expected sign, such as that between TOTD and Lev

(q = -0.026), but some do not, such as the one between HRD and Prof (q = -0.023). A number of other correla- tions are also significant at the 1 % level, such as those

between Size and Media (q = 0.401) and Prof and Lev (q = -0.181). However, none of the variance inflation factors exceed the critical value of 10, indicating that

multicollinearity is not a problem in our study.

In discussing our regression results relating to the

determinants of CSD, we begin with the full model spec-

ification before presenting the model that omits the stra-

tegic posture variable Dept. We do so to aid comparison

with previous studies deploying a stakeholder framework

that made such omission (Van der Laan Smith et al. 2005;

Huang and Kung 2010).

Table 7 reports the results across the four index scores

for CSD quality. For the full model specification, when

considered individually from any disclosure quality index,

TOTD, CCEND, ERD, or HRD, the signs of the estimated

coefficients are all in line with our expectations: disclosure

quality is positively related to the three stakeholder power

variables Sup (b1), Cap (b2), and Own (b3), positively related to the strategic posture variable Dept (b4), posi- tively related to the economic resource variable Prof (b5), negatively related to the other economic resource variable

Lev (b6), and positively related to the two control variables Size (b7) and Media (b8). Although they have the expected signs, some estimates remain statistically insignificant. For

instance, using the full model with the dependent variable

TOTD, the three stakeholder power variables, the one

strategic posture variable, and the two control variables are

all significant at either the 1 or 5 % level, but the two

economic resource variables Prof and Lev are insignificant.

Specifically, for the first dominant stakeholder group,

i.e., buyers in the global supply chain, we find a significant

link to all four disclosure indices. This complements the

previous empirical finding that pressure from international

buyers is a crucial factor for CSD (Roberts 2003; Islam and

Deegan 2010; Khan et al. 2013). For the other dominant

stakeholder group, i.e., listing or social rating agencies in

international capital markets, the same significant link is

also found. Given that prior research views access to

international capital markets as a driver of corporate

reporting (Scott 1994; Cormier and Magnan 2003; Cormier

et al. 2005; Haniffa and Cooke 2005; Reverte 2009;

Gamerschlag et al. 2011), this finding suggests that Tai-

wanese firms confronted with more intense disclosure

requirements are more likely to enhance the quality of their

CSD. For both dominant stakeholder groups, the fit of theT a

b le

6 P

e a rs

o n

c o

rr e la

ti o

n c o

e ffi

c ie

n ts

In d

e p

e n

d e n

t v

a ri

a b

le s

In d

e p

e n

d e n

t a n

d d

e p

e n

d e n

t v

a ri

a b

le s

T O

T D

C C

E N

D E

R D

H R

D S

u p

C a p

O w

n D

e p

t P

ro f

L e v

S iz

e

S u

p 0

.1 5

2 *

* *

0 .1

4 8

* *

* 0

.1 7

6 *

* *

0 .0

9 5

* *

C a p

0 .5

8 7

* *

* 0

.5 8

1 *

* *

0 .5

2 2

* *

* 0

.4 3

8 *

* *

0 .1

3 3

* *

*

O w

n 0

.2 7

4 *

* *

0 .2

7 7

* *

* 0

.2 0

1 *

* *

0 .2

2 1

* *

* 0

.1 0

9 *

* *

0 .1

6 3

* *

*

D e p

t 0

.6 6

3 *

* *

0 .6

4 9

* *

* 0

.5 2

6 *

* *

0 .6

3 0

* *

* 0

.0 2

0 0

.3 8

3 *

* *

0 .1

8 7

* *

*

P ro

f 0

.0 4

0 0

.0 4

1 0

.0 4

8 -

0 .0

2 3

0 -

0 .0

0 3

- 0

.1 1

3 *

* *

0 .0

7 8

*

L e v

- 0

.0 2

6 -

0 .0

4 8

0 .0

3 5

0 .0

5 0

- 0

.2 4

3 *

* *

- 0

.0 0

1 0

.1 2

4 *

* *

0 .0

1 4

- 0

.1 8

1 *

* *

S iz

e 0

.1 4

2 *

* *

0 .1

1 3

* *

* 0

.1 9

3 *

* *

0 .1

8 4

* *

* -

0 .1

9 2

* *

* 0

.1 7

9 *

* *

0 .0

5 3

0 .4

6 1

* *

* -

0 .0

7 4

0 .5

9 3

* *

*

M e d

ia 0

.5 1

1 *

* *

0 .4

9 5

* *

* 0

.4 6

4 *

* *

0 .4

3 4

* *

* 0

.0 2

5 0

.5 8

1 *

* *

0 .1

4 3

* *

* 0

.5 4

5 *

* *

0 .0

8 6

* 0

.1 1

8 *

* *

0 .4

0 1

* *

*

* ,

* *

, *

* *

in d

ic a te

st a ti

st ic

a l

si g

n ifi

c a n

c e

a t

th e

9 0

, 9

5 ,

a n

d 9

9 %

c o

n fi

d e n

c e

le v

e l,

re sp

e c ti

v e ly

S e e

T a b

le 4

fo r

a c o

m p

le te

d e sc

ri p

ti o

n o

f th

e v

a ri

a b

le s

T.-K. Chiu, Y.-H. Wang

123

hypothesized relationship seems to mean that the posses-

sion of two attributes, power and legitimacy, assures that

expectations on reporting quality will be met, resulting in a

significant association between their salience and disclo-

sure quality.

However, for shareholding stakeholders, the positive

hypothesized effect of ownership dispersion on disclosure

quality is restricted to three quality indices only: TOTD,

CCEND, and ERD. It would not appear that having greater

ownership dispersion makes a Taiwanese firm more likely

to enhance disclosure quality in human rights issues

(q = 0.35). This may indicate that the shareholding stakeholders only have a moderate concern for human

rights issues, and a fit with the hypothesized relationship

would require the concern to be stronger.

For the strategic posture variable Dept, we find a sig-

nificant link to all four disclosure quality indices. The

significance of the variable seems to indicate that corporate

institutionalization of social disclosure by having some

formal mechanisms in place, such as a special-purpose

team responsible for collecting data, measuring perfor-

mance, and publishing dedicated reports, leads a firm to

have a better disclosure quality. The positive association

holds not only for total disclosure but also for other dis-

closure categories as well. The result is consistent with

previous findings related to CSR practice in general

(Roberts 1992; Chapple and Moon 2005).

For the two economic resource variables, our findings

provide limited support for the hypotheses that firms

exhibiting relatively strong economic performance in prior

periods, as measured by higher returns on equity (Prof) and

lower debt-to-equity ratios (Lev), are more likely to have

better disclosure quality. The insignificant link that we found

between TOTD and Prof (q = 0.57) or between TOTD and Lev (q = 0.17) seems to mean that organizational slack is not a compelling factor in determining firms’ strategies for

disclosure quality. The firm’s strategic decision on reporting

quality may be influenced by a variety of factors; a fit with the

hypothesized relationship between TOTD and economic

performance would require the association to be stronger so

that some other factors would be depressed. Our results on

the link between CSD and financial performance are con-

sistent with earlier evidence that either an insignificant link

or a reverse association is present (e.g., Reverte 2009;

Branco and Rodrigues 2008).

As regards the control variables, our results support the

suggestions that firm size and media exposure may act as

intervening variables in empirical tests focusing on CSD

(e.g., Purushothaman et al. 2000; Branco and Rodrigues

2008; Reverte 2009). These findings may be explained in

part by the arguments that size and firm visibility are

macro-level proxies for aspects of stakeholder power or

strategic posture toward CSD.T a

b le

7 R

e su

lt s

o f

re g

re ss

io n

a n

a ly

si s

o n

th e

e x

p la

n a to

ry v

a ri

a b

le s

o f

d is

c lo

su re

q u

a li

ty

D e p e n d e n t

v a ri

a b le

sa In

d e p

e n

d e n

t v

a ri

a b

le sa

R 2

M o

d e l

F

C o n st

a n t

S u p

(b 1 )

C a p

(b 2 )

O w

n (b

3 )

D e p

t (b

4 )

P ro

f (b

5 )

L e v

(b 6 )

S iz

e (b

7 )

M e d

ia (b

8 )

T O

T D

- 1

2 .4

7 (0

.4 8

) 8

.3 2

(0 .0

0 )*

* *

2 2

.4 3

(0 .0

0 )*

* *

2 .1

2 (0

.0 1

)* *

3 6

.8 5

(0 .0

0 )*

* *

0 .7

1 (0

.5 7

) -

1 .6

8 (0

.1 7

) 1

.8 6

(0 .0

1 )*

* 3

.6 7

(0 .0

0 )*

* *

0 .6

2 9

8 .6

6 *

* *

- 5

0 .2

3 (0

.0 1

) 1

0 .3

9 (0

.0 0

)* *

* 2

7 .6

0 (0

.0 0

)* *

* 2

.5 2

(0 .0

1 )*

* 1

.7 0

(0 .2

4 )

- 0

.7 2

(0 .6

2 )

3 .6

2 (0

.0 0

)* *

* 5

.8 8

(0 .0

0 )*

* *

0 .4

7 6

2 .0

6 *

* *

C C

E N

D -

2 1

.4 7

(0 .1

3 )

5 .8

0 (0

.0 0

)* *

* 1

8 .7

2 (0

.0 0

)* *

* 1

.7 5

(0 .0

1 )*

* 2

8 .7

3 (0

.0 0

)* *

* 0

.2 4

(0 .3

8 )

- 0

.9 9

(0 .3

2 )

1 .0

4 (0

.0 9

)* 2

.7 6

(0 .0

0 )*

* *

0 .5

9 8

5 .9

5 *

* *

- 5

0 .9

1 (0

.0 0

)* *

* 7

.4 1

(0 .0

0 )*

* *

2 2

.7 5

(0 .0

0 )*

* *

2 .0

6 (0

.0 1

)* *

0 .6

4 (0

.5 9

) -

0 .2

5 (0

.8 3

) 2

.4 2

(0 .0

0 )*

* *

4 .4

8 (0

.0 0

)* *

* 0

.4 4

5 4

.5 8

* *

*

E R

D 9

.0 3

(0 .0

0 )*

* *

1 .8

7 (0

.0 0

)* *

* 2

.1 7

(0 .0

0 )*

* *

0 .2

6 (0

.0 7

)* 3

.3 2

(0 .0

0 )*

* *

0 .3

4 (0

.1 1

) -

0 .2

3 (0

.2 1

) 0

.5 6

(0 .0

0 )*

* *

0 .6

7 (0

.0 0

)* *

* 0

.5 1

6 2

.8 9

* *

*

5 .6

3 (0

.0 7

)* 2

.0 6

(0 .0

0 )*

* *

2 .6

3 (0

.0 0

)* *

* 0

.3 0

(0 .0

5 )*

0 .4

3 (0

.0 5

)* -

0 .1

4 (0

.5 2

) 0

.7 2

(0 .0

0 )*

* *

0 .8

7 (0

.0 0

)* *

* 0

.4 6

5 7

.9 9

* *

*

H R

D -

0 .6

0 (0

.7 7

) 0

.5 5

(0 .0

2 )*

* 1

.3 3

(0 .0

0 )

* *

* 0

.1 0

(0 .3

5 )

4 .3

1 (0

.0 0

)* *

* 0

.2 1

(0 .1

7 )

- 0

.4 5

(0 .0

0 )*

* *

0 .1

8 (0

.0 5

)* 0

.3 8

(0 .0

7 )*

0 .5

0 6

1 .2

7 *

* *

- 5

.0 1

(0 .0

4 )*

* 0

.7 9

(0 .0

0 )*

* *

1 .9

3 (0

.0 0

)* *

* 0

.1 4

(0 .2

3 )

0 .3

2 (0

.0 6

)* -

0 .3

4 (0

.0 5

)* *

0 .3

8 (0

.0 0

)* *

* 0

.4 8

(0 .0

1 )*

* 0

.3 2

3 2

.4 6

* *

*

* ,

* *

, *

* *

S ta

ti st

ic a l

si g

n ifi

c a n

c e

a t

9 0

, 9

5 ,

a n

d 9

9 %

c o

n fi

d e n

c e

le v

e l,

re sp

e c ti

v e ly

. W

h e n

th e

sy m

b o

l d

ir e c ti

o n

is p

re d

ic te

d ,

a o

n e -t

a il

e d

te st

is u

se d ,

o th

e rw

is e

a tw

o -t

a il

e d

te st

is u se

d

p v

a lu

e s

a re

g iv

e n

in p

a re

n th

e si

s a

S e e

T a b

le 4

fo r

a c o

m p

le te

d e sc

ri p

ti o

n o

f th

e v

a ri

a b

le s

Social Disclosure Quality in Taiwan

123

The F values of the full models are all significant at 1 %

level, as illustrated by the model F, with 98.66 for the

dependent variable TOTD and 85.95 for CCEND. These

results indicate that our models have very good fits. As for

R 2

values, the explanatory variables combined can explain

62, 59, 51, and 50 %, of the total variation of TOTD,

CCEND, ERD, and HRD, respectively.

Moving from the full to the partial model in which the

strategic posture variable Dept is omitted, the results show

some similarities, but also some differences. First, similar

to the full model, for any disclosure index in the partial

model, stakeholder power variables Sup and Cap, and

control variables Size and Media are all significant at either

the 1 or 5 % level. Second, in contrast to the full model,

when we omit Dept, the economic resource variable Prof

has a significant positive effect on the disclosure quality of

ERD (q = 0.05) as well as on HRD (q = 0.06). Third, the most noticeable difference between the full and partial

model is the F values, with the values dropping signifi-

cantly in the partial model. For example, for the model

with dependent variable TOTD, F drops from 98.66 in the

full model to 62.06 in the partial model. This large drop in

the F value seems to indicate that the full model is a better

specification than the partial one. Further, moving from the

full to the partial model, the R 2

values also drop quite a bit,

which reinforces the findings for the F values, indicating

that the strategic posture variable Dept is an important

determinant of social disclosure quality in Taiwan.

The results of the empirical tests are of interest for

several reasons. First, the significance of the model pro-

vides evidence that stakeholder theory is an appropriate

foundation for empirical analyses of CSD quality in Tai-

wan. Second, the overall findings support the contingency

framework developed by Ullmann (1985), which occupies

an important place in CSD research. Third, the results

complement Mitchell’s specification (Mitchell et al. 1997)

that a combination of power and legitimacy helps to

identify the dominant stakeholder groups in the Taiwanese

setting. Fourth, the results support the argument that cur-

rent period levels of CSD relate to prior period measures of

stakeholder power, strategic posture toward CSD, and

economic resources. Also, the significance of the individ-

ual variables supports arguments regarding links between

CSD and the specific empirical measures employed in this

study.

Conclusion

This paper examined the voluntary social reporting of an

Asian emerging economy, Taiwan, analyzed the potential

determinants of disclosure quality, and interpreted the

empirical results based on a stakeholder theory framework.

In applying extant theories to the Taiwanese setting, we

took a contextually contingent approach to determine two

dominant stakeholder groups who possess power and

legitimacy and thereby influenced managers’ responsive-

ness to their transparency demands through the quality of

social disclosure. Our analysis distinguished between a

number of quality facets in a variety of disclosure areas,

with an emphasis on the extent to which CSD is compre-

hensive, reliable, and forward-looking. Using a sample of

246 listed companies and a hand-collected dataset con-

taining 2 years of data based on the survey questions of the

renowned 100 Best Corporate Citizens to reflect ongoing

international disclosure trends, we analyzed how reporting

practice is related to stakeholder power, the firm’s strategic

posture toward CSD, economic resources, and control

variables.

We find that the disclosure quality of Taiwanese firms is

generally unsatisfactory, being particularly inferior in areas

related to environmental impact and climate change

abatement. Such inadequacy is due to insufficient legisla-

tion and lax enforcement of existing regulations. The

reporting quality of firms seems to be influenced by three

stakeholder power variables, one strategic posture variable,

and two economic resource variables, as well as the two

intervening variables. However, our findings also demon-

strate a high variability across firms. While active domestic

scrutiny bodies and regulations are lacking, firms con-

fronted with more intense disclosure requirements either

from organizational buyers in the global supply chain or

listing agencies in international capital market are more

likely to enhance the quality of CSD. Those who do not

face disclosure pressures from the foreign institutional

environment do not respond.

This study contributes to the literature on CSD in the

following ways. First, it adds to the scarce evidence on

social reporting in Taiwan. Second, we provide a useful

method for evaluating reported quality, which reflects

whether a firm’s disclosure discusses CSR policies, goals,

initiatives, progress, and audits. Third, the analysis draws

upon a quite large sample from a diverse range of industrial

sectors and uses a variety of information sources, thus

permitting a comprehensive exploration of the impact of

firm and industry characteristics upon disclosure quality.

Fourth, we illustrate the impact of two dominant stake-

holder groups, buyers in the global supply chain and listing

agencies in international capital markets, on voluntary

social disclosure, in an emerging country with insufficient

local laws and weak civil society scrutiny.

This paper suffers from limitations that could be

addressed in future work. First, we apply an aggregate

measure of quality with five facets to the voluntary dis-

closures made by the sample companies. An alternative

approach would be an independent focus on each

T.-K. Chiu, Y.-H. Wang

123

individual indicator of quality. This latter approach would

permit insight into whether the facets are complements or

substitutes, as well as reveal the extent to which each is

associated with a particular influential stakeholder group,

mode of strategic posture, or form of economic resource.

Secondly, although we introduce time through the use of a

simple 1-year lag structure and 2 years of data, our analysis

remains cross-sectional in nature. The quality of the study

would benefit were more years of data available, since

longitudinal analysis would help to shed light on the

evolving pattern of disclosure quality of Taiwanese firms

(Haniffa and Cooke 2005; Gamerschlag et al. 2011; Ma-

hadeo et al. 2011). Third, compared to industrial nations in

Europe and the USA, the environmental laws of Taiwan are

clearly lagging behind. This means that international dis-

closure requirements take precedence over domestic ones

for globalized Taiwanese firms. This paper uses ‘interna-

tional listing’ as the test. Future studies could add other

variables, such as firms’ environmental penalties, to see if

these potentially affect corporate environmental reporting.

Finally, many studies have stressed the importance of

culture on corporate disclosure (Chapple and Moon 2005;

Haniffa and Cooke 2005; Van der Laan Smith et al. 2005)

and searched for appropriate variables to measure the

abstract dimensions of culture. We stress the importance of

macro factors, but an in-depth analysis is beyond the scope

of this paper. Further analysis is needed to explore the

influence that cultural characteristics have on voluntary

disclosure in Taiwan.

Appendix

The 100 Best Corporate Citizens (100 Best ) has been

compiled once a year since 2000 to advance corporate

accountability and responsibility. Being advised and guided

by the CROA Methodology Committee (composed of a

cross-section of CSR experts, including practitioners, aca-

demics, NGOs, and investment firms), the rating reflects the

view of the professional community in relation to global

CSR trends. The methodology is objective and quantitatively

oriented, and the change is evolutionary in that the Com-

mittee has set itself the target of evolving the metrics for one

category each year to facilitate meaningful year-over-year

comparisons. Details of the methodology, reasons for

amendments and survey questions are announced on CR

magazine’s website for public review.

The 100 Best 2012 encompasses 318 data elements

(from 318 questions) spread across seven categories. Data

in each category fall into one of three dimensions: policy,

disclosure, and performance. The categories, data elements

per category, and their respective weights are climate

change abatement (CC), 59 (16.5 %); environmental

impact (EN), 130 (19.5 %); employee relations (ER), 64

(19.5 %); human rights (HR), 39 (16 %); philanthropy

(PH), 9 (9 %); corporate governance (CG), 9 (7 %);

financial performance (F), 8 (12.5 %). The methodology

weighs the seven categories differently to account for dif-

ferent relative values. For example, ‘environment’ and

‘climate change’ together have a total of 189 data elements

and represent 36 % of the rating; these two categories

reflect the increasing emphasis on managing the costs

associated with environmental impact and climate change,

including the imminent pricing of carbon. In contrast,

categories with fewer questions and lower weight are

formed for two reasons. First, mandatory activities and

compliance cannot be used as a major differentiator among

companies, so only nine data elements on corporate gov-

ernance are included in the rating. Second, if the Com-

mittee considered that certain activities are no longer part

of the mainstream practice of CSR, then it would not

include too many related questions. For example, ‘check-

book philanthropy’ has become an outdated practice, and

this explains why ‘philanthropy’ only has nine data ele-

ments with a focus on policy to judge the comprehen-

siveness of the firm’s CSR management system.

The original 100 Best is a CSR rating, rather than a CSD

rating. It covers the three dimensions on policy, disclosure

and performance for obvious reasons. Firms should define

their CSR activities, establish policy, set up development

goals, and determine their scope of application. Then they

need to disclose related information, including reporting on

progress towards goals, and the environmental and social

impacts of their operations. Finally, based on the actual

CSR performance, cross-firm comparisons are made, with a

ranking.

As our study focuses on CSD, the questions of the 100

Best have to be re-interpreted, mainly in the ‘policy’ and

‘performance’ dimensions. To this end, we have done two

modifications. First, the original questions on the policy

dimension of the 100 Best are concerned with policy for-

mulated by the firm in the areas of climate change abate-

ment, environmental impact, and philanthropy, and this

approach involves policy content, implementation and

scope of application. We interpret these policy questions

into ‘disclosure’, because corporate policies should be

openly disclosed. This interpretation can help achieve the

aims of this study. Second, the original questions on per-

formance of the 100 Best use other publicly available

sources of information (such as the paid database Corpo-

rate Library and Morningstar) to make cross-firm com-

parisons. As performance ranking is not linked to the aims

of this paper, we removed all of the performance questions,

including questions on financial performance. Removal of

the finance questions does not affect the results of our

study, as they (such as the current ratio) fall under

Social Disclosure Quality in Taiwan

123

mandatory disclosure. The re-interpretations and amend-

ments will not alter the essence of the 100 Best, since

disclosure has been the most significant part of the original

evaluation. After these changes, the remaining data ele-

ments (obtained from questions) in each category are: cli-

mate change abatement, 57; environment, 124; employee

relations, 59; human rights, 37; philanthropy, 8; and cor-

porate governance, 6. A total number of 291 questions are

used in our study to obtain data to assess CSD quality.

Data of the 100 Best are of two types: true/false for the

policy and disclosure questions, and numerical for the

performance questions. ‘True’ counts as one point; ‘False’

and blank fields count as zero point. Numerical values

(from performance questions) are compared with the

numerical answers of all other firms in order to generate a

ranking. Our re-interpretations and amendments do not

affect the data type for the policy and disclosure questions,

data for which are still of the binary type: if firms formu-

lated policy or disclosed information, they score 1; other-

wise 0. We attach the same weight to each of the remaining

six CSR categories and allow the number of questions in

each category reflect its relative importance. With the

removal of all (27) performance questions from the 100

Best 2011, it follows that the 291 CSD questions used in

this study result in a binary data system, with the highest

possible index score being 291, and the lowest 0.

The rating of the 100 Best is based on objective, publicly

disclosed information. In 2012, the CROA contracted out

the data collection work to IW Financial, a professional

consulting firm serving the ESG (environmental, social,

and governance) investment community. IW Financial

collects companies’ data from several sources, including

corporate websites, CSR reports, the Carbon Disclosure

Project, as well as paid databases such as the Foundation

Center and Corporate Library, answering survey questions,

computing total scores, and calculating the rankings. The

final decision, however, rests with CROA.

This paper follows the 100 Best in using public infor-

mation to assess the quality of CSD. Data are collected

from corporate websites, annual reports, financial reports,

and CSR reports. As some of the survey questions are

about specific practices in the USA, we need to re-interpret

the questions to fit the local conditions. For example, the

original questions of the 100 Best involved practices of the

U.S. Environmental Protection Agency, so this is changed

to Taiwan’s Environmental Protection Agency. In addition,

parts that involved the 401(K) retirement plan in the USA

are changed to the local ‘portable pensions’ in Taiwan;

parts about U.S. employee benefits in the form of education

grants need to be changed to ‘vocational education and

continuing education’; finally, the relevant requirements of

the U.S.’ Sarbanes–Oxley Act are changed to ‘the Exercise

of Powers by the Public Companies’ Auditing Committee’.

In accordance with the main body of this paper, the

overall CSD quality is reflected in disclosures in the fol-

lowing five facets: policies, goals, initiatives, progress

reports, and audits. Table 2 in the main text reports the

number of questions of these five facets of disclosure

quality in each of the six CSR categories. For example, the

category of climate change abatement has 18 questions on

policies, one on goals, two on initiatives and thirty-six on

progress reports. Table 8 provides a number of sample

Table 8 Sample questions

Climate change abatement

Policies: Does the company disclose that it has an enterprise

level climate change policy?

Goals: Does the company disclose the quantitative targets set

by its climate change policy?

Initiatives: Does the company disclose that it is involved in the

Carbon Mitigation initiative?

Reports: Does the company disclose its total direct greenhouse

gas emissions? Indirect emissions?

Environmental impact

Policies: Does the company disclose that its environmental

policy has a commitment to stakeholder involvement?

Goals: Does the company disclose the quantitative targets set

by its environmental policy?

Initiatives: Does the company disclose that it has a consumer

product recycling program?

Reports: Does the company disclose its total energy use? Does

the company disclose its energy conservation?

Certificates: Does the company disclose its Green Building

certificates? ISO 14000 certificates?

Human rights

Policies: Does the company disclose that its human rights

policy includes a commitment to public reporting?

Goals: Does the company disclose quantitative targets set by its

human rights policy?

Initiatives: Does the company disclose whether it is a signatory

to the United Nation’s Global Compact?

Reports: Does the company disclose the number of personnel

involved in policy implementation?

Certificates: Does the company disclose its SA 8000 certified

facilities?

Employee relations

Initiatives: Does the company disclose its employee benefits

covered by insurance?

Reports: Does the company disclose the composition of its

workforce by location?

Corporate governance

Initiatives: Does the company disclose whether it has an

independent Compensation Committee?

Philanthropy

Policies: Does the company disclose whether technical

assistance is part of its philanthropic giving?

T.-K. Chiu, Y.-H. Wang

123

questions of the 100 Best after being reinterpreted for

‘disclosure’ and is broken down into five facets of disclo-

sure quality for the six CSR categories.

References

Adams, C. A., Hill, W., & Roberts, C. B. (1998). Corporate social

reporting practices in Western Europe: Legitimating corporate

behavior? British Accounting Review, 30(1), 1–21.

Arenas, D., Lozano, J., & Albareda, L. (2009). The role of NGOs in

CSR: Mutual perceptions among stakeholders. Journal of

Business Ethics, 88(1), 175–197.

Brammer, S., Brooks, C., & Pavelin, S. (2009). The stock perfor-

mance of America’s 100 Best Corporate Citizens. Quarterly

Review of Economics and Finance, 49(3), 1065–1080.

Brammer, S., & Pavelin, S. (2008). Factors influencing the quality of

corporate environmental disclosure. Business Strategy and the

Environment, 17(2), 120–136.

Branco, M. C., & Rodrigues, L. L. (2008). Factors influencing social

responsibility disclosure by Portuguese companies. Journal of

Business Ethics, 83(4), 685–701.

Carbon Disclosure Project (CDP) (2008). Carbon Disclosure Project

2008, Asia ex-Japan report. Hong Kong: Association for

Sustainable & Responsible Investment in Asia (ASrIa).

Chapple, W., & Moon, J. (2005). Corporate social responsibility in

Asia: A seven-country study of CSR Website reporting. Business

and Society, 44(4), 415–441.

Chen, S., & Bouvain, P. (2009). Is corporate responsibility converg-

ing? A comparison of corporate responsibility reporting in the

USA, UK, Australia, and Germany. Journal of Business Ethics,

87(1), 299–317.

Chen, J., & Roberts, R. (2010). Towards a more coherent under-

standing of the organization–society relationship: A theoretical

consideration for social and environmental accounting research.

Journal of Business Ethics, 97(4), 651–665.

Chiu, T. (2014). A solution to data issue for emerging economies: A

baseline corporate social responsibility database for Taiwan.

China: Mimeo, Shanghai Advanced Institute of Finance, Shang-

hai Jiao Tong University.

Cormier, D., & Magnan, M. (2003). Environmental reporting

management: A continental European perspective. Journal of

Accounting and Public Policy, 22(1), 43–62.

Cormier, D., Magnan, M., & Van Velthoven, B. (2005). Environ-

mental disclosure quality in large German companies: Economic

incentives, public pressure or institutional conditions? European

Accounting Review, 14(1), 3–39.

Cox, P., & Wicks, P. (2011). Institutional interest in corporate

responsibility: Portfolio evidence and ethical explanation. Jour-

nal of Business Ethics, 103(1), 143–165.

De Villier, C., & Van Staden, C. J. (2006). Can less environmental

disclosure have a legitimizing effect? Evidence from Africa.

Accounting, Organizations and Society, 31(8), 763–781.

Deegan, C. (2002). The legitimizing effect of social and environ-

mental disclosures: A theoretical foundation. Accounting, Audit-

ing and Accountability Journal, 15(3), 282–311.

Deegan, C., & Gordon, B. (1996). A study of the environmental

disclosure practices of Australian corporations. Accounting and

Business Research, 26(3), 187–199.

Dierkes, M., & Antai, A. (1985). The usefulness and use of social

reporting information. Accounting, Organization and Society,

19(1), 29–34.

Drumwright, M. (1994). Socially responsible organizational buying:

Environmental concern as a noneconomic buying criterion.

Journal of Marketing, 58(3), 1–19.

European Sustainable Investment Forum (EUROSIF) (2012). Euro-

pean SRI Study 2012. Brussels, Belgium: EUROSIF.

Freeman, R. E. (1984). Strategic management: A stakeholder

approach. Boston: Pitman.

Freedman, M., & Stagliano, A. J. (1992). European unification,

accounting harmonization, and social disclosures. The Interna-

tional Journal of Accounting, 27(2), 112–122.

Friedman, M. (1970). The social responsibility of business is to

increase its profits. New York: The New York Times Magazine.

13 September.

Gamerschlag, R., Moller, K., & Verbeeten, F. (2011). Determinants

of voluntary CSR disclosure: Empirical evidence from Germany.

Review of Management Science, 5(2–3), 233–262.

Global Reporting Initiative (GRI) (2011). Sustainability Reporting

Guidelines, version 3-1. Amsterdam: Global Reporting

Initiative.

Gray, R., Kouhy, R., & Lavers, S. (1995). Corporate social and

environmental reporting: A review of the literature and a

longitudinal study of UK disclosure. Accounting, Auditing and

Accountability Journal, 8(2), 47–77.

Haniffa, R., & Cooke, T. (2005). The impact of culture and

governance on corporate social reporting. Journal of Accounting

and Public Policy, 24(5), 391–430.

Hill, C., & Jones, T. (1992). Stakeholder-agency theory. Journal of

Management Studies, 29(2), 131–154.

Huang, C., & Kung, F. (2010). Drivers of environmental disclosure

and stakeholder expectation: Evidence from Taiwan. Journal of

Business Ethics, 96(3), 435–451.

Islam, M., & Deegan, C. (2010). Media pressure and corporate

disclosure of social responsibility performance information: A

study of two global clothing and sports retail companies.

Accounting and Business Research, 40(2), 131–148.

Jensen, M. (2002). Value maximization, stakeholder theory, and the

corporate objective function. Business Ethics Quarterly, 12(2),

235–256.

Khan, A., Muttakin, M., & Siddiqui, J. (2013). Corporate governance

and corporate social responsibility disclosure: Evidence from an

emerging economy. Journal of Business Ethics, 114(2),

207–223.

Krippendorf, K. (2004). Content analysis: An introduction to its

methodology. London: Sage Publications.

Lau, L. (1994). The competitive advantage of Taiwan. Asia Pacific

Business Review, 1(1), 90–112.

Louche, C., & Lydenberg, S. (2010). Responsible investing. In J.

Boatright (Ed.), Finance ethics: Critical issues in theory and

practice (pp. 393–418). Hoboken, NJ: Wiley.

Magness, V. (2006). Strategic posture, financial performance and

environmental disclosure: An empirical test of legitimacy theory.

Accounting, Auditing & Accountability Journal, 19(4), 540–563.

Mahadeo, J. D., Oogarah-Hanuman, V., & Soobaroyen, T. (2011). A

longitudinal study of corporate social disclosures in a developing

economy. Journal of Business Ethics, 104(4), 545–558.

Matten, D., & Moon, J. (2008). ‘Implicit’ and ‘explicit’ CSR: A

conceptual framework for a comparative understanding of

corporate social responsibility. Academy of Management Review,

33(2), 404–424.

Milne, M. J. (2002). Positive accounting theory, political costs and

social disclosure analyses: A critical look. Critical Perspectives

on Accounting, 13(3), 369–395.

Mitchell, R., Agle, B., & Wood, D. (1997). Toward a theory of

stakeholder identification and salience: Defining the principle of

who and what really counts. Academy of Management Review,

22(4), 853–886.

Social Disclosure Quality in Taiwan

123

Muller, A., & Kräussl, R. (2011). Doing good deeds in times of need:

A strategic perspective on corporate disaster donations. Strategic

Management Journal, 32(9), 911–929.

Neu, D., Warsame, H., & Pedwell, K. (1998). Managing public

impressions: Environmental disclosures in annual reports.

Accounting, Organizations and Society, 23(3), 265–282.

Patten, D. M. (2002). The relation between environmental perfor-

mance and environmental disclosure: A research note. Account-

ing, Organization and Society, 27(8), 763–764.

Pfeffer, J. (1981). Power in organizations: A critical essay. New

York: Random House.

Purushothaman, M., Tower, G., Hancock, P., & Taplin, R. (2000).

Determinants of corporate social reporting practices of listed

Singapore companies. Pacific Accounting Review, 12(2),

101–124.

Reverte, C. (2009). Determinants of corporate social responsibility

disclosure ratings by Spanish listed firms. Journal of Business

Ethics, 88(2), 351–366.

Roberts, R. (1992). Determinants of corporate social responsibility

disclosure: An application of stakeholder theory. Accounting,

Organizations and Society, 17(6), 595–612.

Roberts, S. (2003). Supply chain specific? Understanding the patchy

success of ethical sourcing initiatives. Journal of Business

Ethics, 44(2/3), 159–170.

Scalet, S., & Kelly, T. (2010). CSR rating agencies: What is their

global impact? Journal of Business Ethics, 94(1), 69–88.

Scott, T. (1994). Incentives and disincentives for financial disclosure:

Voluntary disclosure of defined benefit pension plan information

by Canadian firms. The Accounting Review, 69(1), 26–43.

Starr, M. (2008). Socially responsible investment and pro-social

change. Journal of Economic Issues, 42(1), 51–73.

Sturgeon, T., & Lee, J. (2005). Industry co-evolution: A comparison

of Taiwan and North American electronics contract manufac-

turers. In S. Berger & R. Lester (Eds.), Global Taiwan: Building

Competitive Strengths in a New International Economy. Ar-

monk, NY: M.E Sharpe.

SustainAbility (2010). Rate the raters phase two: taking inventory of

the ratings universe. Available at: http://www.sustainability.

com/projects/rate-the-raters. Accessed 1 Oct 2013.

Sustainable Asset Management (SAM). (2011). Corporate sustain-

ability assessment questionnaire: DJSI 2011—test companies.

Zurich: SAM Research Inc.

Teegen, H., Doh, J. P., & Vachani, S. (2004). The importance of

nongovernmental organizations (NGOs) in global governance

and value creation: An international business research agenda.

Journal of International Business Studies, 35(6), 463–483.

Ullmann, A. A. (1985). Data in search of a theory: A critical

examination of the relationship among social performance,

social disclosure, and economic performance of U.S. Firms.

Academy of Management Review, 10(3), 540–557.

Van der Laan Smith, J., Adhikari, A., & Tondkar, R. H. (2005).

Exploring differences in social disclosures internationally: A

stakeholder perspective. Journal of Accounting and Public

Policy, 24(2), 123–151.

Xiao, J., Yang, H., & Chow, C. (2004). The determinants and

characteristics of voluntary Internet-based disclosures by listed

Chinese companies. Journal of Accounting and Public Policy,

23(3), 191–225.

T.-K. Chiu, Y.-H. Wang

123

2015.pdf

• Determinants influencing CSR practices in small and medium sized MNE subsidiaries: A stakeholder perspective
• Introduction
• Literature review and theory development
• Theoretical background
• Importance of CSR for SMMSs in emerging economies
• Primary stakeholders
• Consumers
• Internal managers and employees
• Governments
• Suppliers
• Investors
• Secondary stakeholders
• Competitors
• Media
• Local community
• NGOs
• Methodology
• Sample and data collection
• Variable measurement
• Results and discussion
• Managerial relevance
• Conclusions
• Variable measurements
• References