A-PLUS WRITER AS THE QUESTION

Factors Influencing Social Responsibility

Disclosure by Portuguese Companies Manuel Castelo Branco Lúcia Lima Rodrigues

ABSTRACT. This study compares the Internet (corpo-

rate web pages) and annual reports as media of social

responsibility disclosure (SRD) and analyses what influ-

ences disclosure. It examines SRD on the Internet by

Portuguese listed companies in 2004 and compares the

Internet and 2003 annual reports as disclosure media. The

results are interpreted through the lens of a multi-theo-

retical framework. According to the framework adopted,

companies disclose social responsibility information to

present a socially responsible image so that they can

legitimise their behaviours to their stakeholder groups and

influence the external perception of reputation. Results

suggest that a theoretical framework combining legiti-

macy theory and a resource-based perspective provides an

explanatory basis for SRD by Portuguese listed

companies.

KEY WORDS: annual reports, internet, legitimacy

theory, resource-based perspectives, social responsibility

disclosure, Portugal

Introduction

Most of the empirical studies on social responsibility

disclosure (SRD) have focused on the annual report,

which is considered to be the most important tool

used by companies to communicate with their

stakeholders (see, for example, Gray et al., 1995b;

Neu et al., 1998). However, the Internet has become

an important medium through which companies can

disclose information of different natures, and thus

some recent studies have been made analysing com-

panies’ web pages as a SRD medium (see, for

example, Frost et al., 2005; Patten, 2002a; Patten and

Crampton, 2004; Williams and Pei, 1999). Explora-

tion of companies’ web pages as a SRD medium is

now as essential as the exploration of annual reports to

understand SRD disclosure practices.

The purpose of this study is to understand SRD,

both on the Internet and in annual reports, by

developing and testing a series of hypotheses. The

nature of SRD in annual reports and on the

Internet by a sample of companies with shares listed

on the Portuguese Stock Exchange (Euronext –

Lisbon) is analysed. Using content analysis, SRD is

classified in terms of theme (environment, human

resources, products and customers and community

involvement).

Companies are considered to engage in corporate

social responsibility (CSR) activities and disclosure

because of two different kinds of motivations. Some

companies expect that having good relations with

their stakeholders will lead to increased financial

returns by assisting in developing valuable intangible

assets (resources and capabilities). These assets can

be sources of competitive advantage because they

can differentiate a company from its competitors.

Other companies engage in CSR activities and

disclosure to conform to stakeholder norms and

expectations about how operations should be con-

ducted, thus constituting mainly a legitimacy

instrument used by a company to demonstrate its

adherence to such norms and expectations. Al-

though some companies engage in CSR activities

and disclosure because their managers’ personal

values are aligned with CSR values, this aspect will

not be explored in this study.

Whereas the first kind of motivations may be

explored through a resource-based perspective ana-

lytical lens (see, for example, Branco and Rodrigues,

2006; Hasseldine et al., 2005; Toms, 2002), the

second kind is consistent with social and political

theory explanations, in particular legitimacy theory

(see, for example, Deegan, 2002; Deephouse and

Carter, 2005; Neu et al., 1998; Patten and

Crampton, 2004; Zimmerman and Zeitz, 2002).

Journal of Business Ethics (2008) 83:685–701 � Springer 2008 DOI 10.1007/s10551-007-9658-z

The results are interpreted through the lens of a

multi-theoretical framework which combines these

two perspectives, according to which companies

disclose social responsibility information mainly to

present a socially responsible image so that they can

legitimise their behaviours to stakeholder groups and

influence the external perception of reputation.

Companies with a higher visibility seem to exhibit

greater concern to improve corporate image through

SRD. Results suggest that the framework proposed

may be an explanation of SRD by Portuguese listed

companies.

This paper examines empirical evidence from

Portugal for two reasons. First, we want to add to

the scarce research on SRD by Portuguese compa-

nies by providing new empirical data. Most of the

present literature is based in Anglo-Saxon countries

and evidence should be added about other geo-

graphic, cultural and institutional contexts. In con-

trast to the understanding of SRD from common

law English-speaking countries (Australia, Canada,

UK, USA), the determinants of SRD in Continental

Europe, particularly in Portugal, are still relatively

unknown. Second, we want to analyse if there are

reasons to expect that listed companies in less

developed countries, such as Portugal, will behave in

a different manner than companies in more devel-

oped countries.

According to Lopes and Rodrigues (2007, p. 30),

Portugal is one of the least developed countries in

the euro-area and a small OECD country. It presents

specific features regarding its capital market, com-

panies’ financing structure and corporate governance

systems, providing for a different institutional setting

from most developed and capital market-oriented

countries, where most of the SRD studies have

been made. In particular, the degree of family

ownership is significant and financing policies are

bank oriented.

Notwithstanding the particular characteristics of

Portugal, the results of this study suggest that factors

which influence SRD practices of Portuguese listed

companies are not significantly different than those

which influence SRD of companies in more

developed countries. This is consistent with the

results of Cormier and Magnan (2003), which lead

them to suggest that the similitude in the way in

which disclosure strategies are determined, irre-

spective of a given country’s socio-cultural envi-

ronment, is ‘‘an illustration of the strong impact of

globalised stock markets on fostering convergence in

corporate practices’’ (op. cit., p. 58).

In the following section, the theoretical frame-

work used is presented. Thereafter follow sections

on hypotheses development, methodology, results

and discussion. Finally, some conclusions are drawn.

Theoretical framework

Two major influences on companies’ SRD are

acknowledged in this study: those related to the

socio-political context within which companies

operate, and those related to economic incentives.

The theoretical framework adopted incorporates

both influences, by adopting institutional theory

perspectives, specifically legitimacy theory (see, for

example, Deegan, 2002; Deephouse and Carter,

2005; Neu et al., 1998; Patten and Crampton, 2004;

Zimmerman and Zeitz, 2002), and resource-based

perspectives (see, for example, Branco and

Rodrigues, 2006; Hasseldine et al., 2005; Toms,

2002). Some authors provide important studies in

which similar combinations are attempted (see, for

example, Bansal, 2005).

In this study, companies are considered to engage

in some form of stakeholder management, driven by

two different kinds of motivations. Some companies

believe that being seen as socially responsible will

bring them a competitive advantage, allowing them

to achieve better economic results. They expect that

having good relations with their stakeholders will

lead to increased financial returns by assisting in

developing valuable intangible assets (resources and

capabilities) which can be sources of competitive

advantage because such assets can differentiate a

company from its competitors. These motivations

are consistent with a resource-based perspective

analytical lens.

Other companies engage in CSR activities and

disclosure because of external pressures. They either

conform to what other companies do, because they

believe that not doing so would harm them in terms

of their profitability and survival, or respond to

discrediting events, which they believe to be detri-

mental to their profitability and survival and must be

addressed to mitigate their effects. CSR activities and

disclosure appear as mechanisms these companies use

686 Manuel Castelo Branco and Lúcia Lima Rodrigues

to act and be seen acting within the bounds of what

is considered acceptable according to the expecta-

tions of stakeholders on how their operations should

be conducted. Social responsibility activities and

disclosure constitute mainly a legitimacy instrument

used by a company to demonstrate its adherence to

such expectations. These motivations are consistent

with social and political theory explanations, in

particular legitimacy theory.

From a resource-based perspective the benefits of

CSR are, to a great extent, related to their effect on

corporate reputation (Branco and Rodrigues, 2006).

Companies with a good social responsibility repu-

tation are able to improve relations with external

actors such as customers, investors, bankers, suppliers

and competitors. They also attract better employees

or increase current employees’ motivation and

morale as well as their commitment and loyalty to

the company, which in turn may improve financial

outcomes. Disclosure of information on a company’s

behaviours and outcomes regarding social responsi-

bility helps to build a positive image with stake-

holders.

SRD is particularly important in enhancing the

effects of CSR on corporate reputation. It might be

considered a signal of improved social and envi-

ronmental conduct and hence reputation in those

fields because disclosure influences the external

perception of reputation. It will be difficult for

companies investing in social responsibility activities,

likely to create positive reputation, to realise the

value of such reputation without making associated

disclosures (Hasseldine et al., 2005; Toms, 2002).

Probably the most important weakness of

resource-based perspectives is related to the lack of

understanding they provide on the influence that the

relationships between a company and its environ-

ment have on the company’s success (Branco and

Rodrigues, 2006). This is why in this study a

resource-based perspective is combined with social

and political theories, in particular legitimacy and

stakeholder theories. However, these theories are

considered complementary rather than alternative or

opposite (Gray et al., 1995a, p. 52).

The institutional perspective of legitimacy theory

is one of the dominant theories in SRD research

(see, for example, Deegan, 2002; Patten and

Crampton, 2004; Neu et al., 1998). The analytical

focus of legitimacy theory’s institutional perspective

is on social legitimacy, which refers to the accep-

tance of a company by its social environment, and

external constituents. Companies consider the

expectations of various social constituents in their

behaviour to achieve social legitimacy. Legitimacy

‘‘is a social judgment of appropriateness, acceptance,

and desirability’’ (Zimmerman and Zeitz, 2002,

p. 418). The importance of social legitimacy comes

from the theoretical assumption that companies are

embedded in the social environment in which they

operate, and that their performance and expectations

are affected by the environment. The company’s

success, even survival, is determined by this

interface.

SRD is seen, from such a perspective, as one of

the strategies used by companies to seek acceptance

and approval of their activities from society. It is seen

as an important tool in corporate legitimation strat-

egies. It is used to establish or maintain the legiti-

macy of the company because it may influence

public opinion and public policy. Legitimacy theory

suggests that SRD provides an important way of

communicating with stakeholders, to convince them

that the company is fulfilling their expectations

(even when actual corporate behaviour remains at

variance with some of these expectations).

One problem regarding attempts to combine

different bodies of theory to explain organisational

behaviour is that they are often incommensurable or

incompatible in some important aspects. The theo-

ries often focus on different core concepts. A multi-

theoretical framework should focus on common

core concepts.

Legitimacy theory and resource-based perspec-

tives are believed to be useful as they can be con-

ceived as using what Campbell et al. (2003, p. 559)

call ‘‘the stakeholder metanarrative’’. Thus, these

perspectives can be explored by using stakeholder

theory insights. On the other hand, organisational

legitimacy and organisational reputation are consid-

ered to have similar antecedents, social construction

processes and consequences (Deephouse and Carter,

2005). This study refers to these two interrelated

concepts: that of legitimacy, which is explored from

an institutional perspective; and that of reputation,

which is explored from a resource-based perspective.

For the purposes of this study, the fundamental

aspect is that legitimacy requires a reputation that

must be retained. It requires a company to convince

Factors Influencing SRD by Portuguese Companies 687

its relevant publics that its activities are congruent

with their values. Thus, reputation and legitimacy

are inextricably linked, and in this study the

distinction between the two will not be explored

further.

Development of hypotheses

In what follows, explanations for SRD based on the

theoretical framework presented in the previous

section are developed by selecting the most relevant

factors influencing SRD. To analyse the usefulness

of the theoretical framework proposed above, this

study adopts the strategy of examining a sample of

companies and using a variety of proxies for a

company’s social visibility related to its characteris-

tics and media exposure. Variables are chosen to

represent particular aspects of social visibility, and in

each case, an expectation regarding its relationship to

SRD is stated based on prior literature.

International experience

International experience is developed by operating

in, and depending upon, foreign markets (Bansal,

2005). The importance of international experience

as a determinant of SRD can be explained from the

perspective of social and political theories (Choi,

1999) and a perspective which is resource based

(Bansal, 2005).

The manner in which the role of a company, and

its stakeholders, is defined in a country, will

undoubtedly affect SRD practices. Operating in

foreign markets requires companies to consider na-

tional differences in customer needs, which are

influenced by the culture and customs of that

country. Companies are exposed also to a greater

extent to the laws, rules and regulations governing

trade within different countries.

In less developed countries one would expect that

a company which does a larger amount of business

abroad is exposed to a broader spectrum of stake-

holder influences and to the international commu-

nity scrutiny. Given the trend of pro-social

responsibility international initiatives, such exposure

is likely to lead to more proactive corporate initia-

tives with respect to the social responsibility issues.

H1: There will be a positive relationship between

the degree of international activity and SRD.

Company size

SRD is related to corporate size, with larger com-

panies disclosing more than smaller ones (see, for

example, Adams et al., 1998; Archel, 2003; Neu

et al., 1998; Patten, 1991; Purushothaman et al.,

2000). Size is also used commonly as a proxy for

public visibility. Larger companies are more sus-

ceptible to scrutiny from stakeholder groups since

they are highly visible to external groups and more

vulnerable to adverse reactions among them; and

larger companies, on average, are more diversified

across geographical and product markets, thus hav-

ing larger and more diverse stakeholder groups

(Brammer and Pavelin, 2004a, p. 704). It is also

more likely that larger, more visible companies will

consider social responsibility activities and disclosure

as a way of enhancing corporate reputation.

H2: There will be a positive relationship between

size and SRD.

Industry affiliation

Another commonly used proxy for social visibility is

industry affiliation. This was found to be related to

SRD by legitimacy theory studies. Industries with

high public visibility, or a potentially more impor-

tant environmental impact, or having less favourable

public images were found to disclose more social

responsibility information than their counterparts

(see, for example, Adams et al., 1998; Archel, 2003;

Clarke and Gibson-Sweet, 1999; Patten, 1991).

There are reasons to suspect that industry affilia-

tion is related to certain categories of SRD. Com-

panies in industries with larger potential

environmental impact are more likely to provide

environmental information, and companies in

industries with high visibility among final consumers

are more likely to consider important issues of

community involvement and disclose information

related to such involvement (Clarke and Gibson-

Sweet, 1999).

688 Manuel Castelo Branco and Lúcia Lima Rodrigues

Thus, this study suggests that the classifications of

industries should be refined to provide more reliable

tests. Two types of proxies for social exposure related to

industry affiliation which were proposed in previous

studies are used: ‘‘consumer proximity’’ (see, for

example, Campbell et al., 2006; Clarke and Gibson-

Sweet, 1999) and ‘‘environmental sensitivity’’ (see, for

example, Archel, 2003; Patten, 2002b). The different

proxies for social exposure are believed to be related to

different SRD categories: community disclosure is

expected to be related positively with a measure of

proximity to the final consumer, whereas environ-

mental disclosure is expected to be related positively

with a measure of environmental sensitivity.

Consumer proximity

The nearer a company is to the individual consumer,

the more probable is its name to be known to most

members of the general public, and hence, the

greater will be its social visibility. Thus, it is hy-

pothesised that community involvement disclosure is

associated with the measure of a company’s prox-

imity to the final consumer.

H3a: There will be a positive relationship between

community involvement disclosure and the

consumer proximity measure.

Environmental sensitivity

Companies in industries that have a larger potential

impact on the environment are considered to be

subject to greater pressures with respect to envi-

ronmental concerns than companies in industries

with less risk in terms of environmental impact.

Therefore, companies in environmentally sensitive

industries are more likely to disclose environmental

information than companies in less environmentally

sensitive industries.

H3b: There will be a positive relationship between

environmental disclosure and the environ-

mental sensitivity measure.

Media exposure

Several studies suggest that individual companies’

media exposure, which is used as a proxy for social

visibility, is likely to be associated to higher levels of

SRD (Bansal, 2005; Brammer and Pavelin, 2004b,

2007, in press; Bewley and Li, 2000; Cormier et al.,

2004, 2005). The total amount of media coverage

raises companies’ visibility, making them the object

of further public attention and scrutiny (Bansal,

2005, p. 203).

H4: There will be a positive relationship between

SRD and the media exposure measure.

Control variables

Control variables, which are designed to account for

other potential influences on SRD practices and

have been analysed in the SRD literature, are

introduced. Prior researchers argue that social

responsibility activities and disclosure are dependent

on the availability of financial resources within a

company (for example, Brammer and Pavelin, 2007,

in press; Roberts, 1992). Following Brammer and

Pavelin (2007, in press), profitability and leverage are

used in this study to capture the availability of

financial resources within a company. These vari-

ables are included as control variables.

Profitability

Several empirical studies have concluded that profit-

ability does not appear to be a significant determinant

of SRD (for example, Archel, 2003; Brammer and

Pavelin, 2007, in press; Patten, 1991; Purushothaman

et al., 2000). From a legitimacy theory perspective,

profitability can be considered to be related positively

or negatively to SRD (Neu et al., 1998). On the other

hand, from a stakeholder perspective (Roberts, 1992),

economic performance is expected to be associated

positively with social responsibility activities and dis-

closure. In view of the existence of these results and

different interpretations, the association between this

variable and SRD is tested without making any a pri-

ori assumption about the sign of such association (see,

for example, Archel, 2003; Bewley and Li, 2000;

Purushothaman et al., 2000).

Leverage

The power of creditors as a stakeholder group

depends upon the degree to which a company relies

Factors Influencing SRD by Portuguese Companies 689

on debt financing (Roberts, 1992). Noting a lack of

conclusiveness in the studies which explore this

relationship, Purushothaman et al. (2000, p. 112)

point out that companies with high leverage may

have closer relationships with their creditors and use

other means to disclose social responsibility infor-

mation. Thus, in this study the association between

this variable and SRD is tested without making any

a priori assumption about the sign of such association.

Methods

Empirical models

The statistical analysis conducted in this study

includes the use of multiple linear regression models

to analyse the relationship between total SRD and

each one of its categories, both in annual reports and

on the Internet, and the influencing factors referred

to in the previous section. Ten models are exam-

ined. The models all use the same influencing factors

discussed above. However, there are ten different

dependent variables: total SRD in annual reports and

on the Internet (SRDAR and SRDI), environ-

mental disclosure in annual reports and on the

Internet (EDAR and EDI), human resources dis-

closure in annual reports and on the Internet

(HRDAR and HRDI), products and customers

disclosure in annual reports and on the Internet

(PCDAR and PCDI), community involvement

disclosure in annual reports and on the Internet

(CIDAR and CIDI).

The approach adopted in the empirical analysis is

summarised by the following general form of the

models:

SRD disclosure index = f (international experi-

ence, size, media exposure, consumer proximity,

environmental sensitivity, control variables)

The general form of the models examined is thus

stated as:

DISCi ¼ �0i þ �1iIEi þ �2iSi þ �3iMEi þ �4iCPi þ �5iESi þ �6iProfiti þ �7iLevi þ ui

where, for company i: DISCi: SRD index

(SRDARi; SRDIi; EDARi; EDIi; HRDARi;

HRDIi; PCDARi; PCDIi; CIDARi; CIDIi); IEi:

international experience; Si: size; MEi: media

exposure; CPi: consumer proximity; ESi: environ-

mental sensitivity; Profiti: profitability; Levi: lever-

age; ui: error term.

Operationalisation of variables

International experience

Following Bansal (2005), international experience is

measured by the percentage of sales outside Portugal

to total sales as reported in the segment data of the

financial statements (see also Choi, 1999).

Company size

Because there are no theoretical reasons which

might clearly justify choosing a particular measure of

size (Hackston and Milne, 1996, p. 87), the measure

used in this study is total assets, as reported on the

balance sheet (Brammer and Pavelin, 2004b; Haniffa

and Cooke, 2005).

Industry affiliation

Because it is suggested in this study that the classi-

fications of industries should be refined to provide

more reliable tests of the theoretical framework used,

two types of proxies for social exposure related to

industry affiliation are used: ‘‘consumer proximity’’

and ‘‘environmental sensitivity’’. The different

proxies for social exposure are believed to be related

to different SRD categories: community disclosure is

expected to be related positively to a measure of

consumer proximity, whereas environmental dis-

closure is expected to be related positively to a

measure of environmental sensitivity.

Consumer proximity. In this study, a binary measure

(high profile and low profile) is used. High-profile

companies are those that are better known to the

final consumer and whose names are expected to be

known to most members of the general public.

Based on prior literature, high-profile companies are

identified as those in the following sectors: house-

hold goods and textiles, beverages, food and drug

retailers, telecommunication services, electricity, gas

distribution, water and banks. All others are con-

sidered ‘‘low profile’’. A one/zero variable is used to

designate companies from these industries: one if the

company is from a high-profile sector, and zero if it

is from a low-profile sector.

690 Manuel Castelo Branco and Lúcia Lima Rodrigues

Environmental sensitivity. In this study, ‘‘more sensi-

tive’’ industries are considered to be those with more

risk of being criticised in environmental matters

because of their activities involving higher risk of

environmental impact (such has natural resource

depletion or pollution). Thus, based on prior liter-

ature, the following ‘‘more sensitive’’ sectors are

identified: mining, oil and gas, chemicals, con-

struction and building materials, forestry and paper,

steel and other metals, electricity, gas distribution

and water. All others are considered as ‘‘less sensi-

tive’’. A one/zero variable is used to designate

companies from these industries: one if the company

is from a more sensitive industry and zero if it is from

a less sensitive industry.

Media exposure

To develop a measure of the companies’ media

exposure, the number of articles in two Portuguese

newspapers was counted. Company exposure was

measured by perusing ‘‘Jornal de Notı́cias’’ and

‘‘Expresso’’, for the period between 1 January 2001

and 31 December 2003. The search facilities present

on the web pages of these newspapers were used. A

search was carried out for each company, using as a

keyword, the name of the company. The search

results were examined to exclude articles that did not

relate specifically to social responsibility issues.

Control variables

Profitability and leverage are used in this study to

capture the availability of financial resources within

the company. These two variables are used as control

variables.

Profitability. When measuring corporate performance

one can use accounting or market-based measures.

In contrast with accounting-based measures, market-

based measures are less subject to bias by managerial

manipulation and they do not rely on past perfor-

mance (McGuire et al., 1988, p. 859). However,

they are based on investors’ viewpoints on company

performance, thus ignoring other stakeholder groups

(ibid.). This is the main reason for adopting an

accounting-based variable in this study.

Thus, return on assets (ROA) is used as a measure

for economic performance (Belkaoui and Karpik,

1989; Bewley and Li, 2000; Brammer and Pavelin,

2007, in press; Cormier et al., 2004; Patten, 1991).

ROA is measured by the ratio of Net income/total

assets (Belkaoui and Karpik, 1989).

Leverage. Leverage is measured by the ratio of Total

debt/total assets (see, for example, Belkaoui and

Karpik, 1989; Brammer and Pavelin, 2007, in

press).

Sample

The sample used in this study comprises listed

companies, as they are more likely to disclose social

responsibility information and have a web page that

provides SRD. To be included in the sample for this

study, a company had to:

• have its shares listed on the Portuguese Stock Exchange (Euronext – Lisbon) by the end of

2003,

• have its 2003 annual report available for re- view, and

• have an accessible corporate web page on the Internet by August 2004.

The initial sample included all companies listed on

Euronext – Lisbon at 31 December 2003. From the

initial 57 listed companies (50 of them listed on the

main market and seven on the second market), a

final sample of 49 companies was identified, as de-

scribed in Table I.

The companies included in the sample are clas-

sified according to sectors using the FTSE Global

Classification System. This classification system

comprises the several sectors which are considered in

Table II. Construction and building materials is the

sector which presents the largest number of com-

panies (8 companies and around 16% of the total).

Banks and Forestry and paper are the sectors which

follow in terms degree of importance (each of them

with 6 and about 24% taken together).

Data collection

To measure the level of social responsibility infor-

mation disclosed by sample companies, this study

uses ‘‘content analysis’’. This technique consists of

classifying the information disclosed into several

Factors Influencing SRD by Portuguese Companies 691

categories of items which capture the aspects of so-

cial responsibility one wants to analyse.

The simplest form of content analysis consists of

detecting the presence or absence of information

(see, for example, Haniffa and Cooke, 2005; Patten,

2002b; Purushothaman et al., 2000). One of the

main shortcomings of this form of content analysis is

that it does not allow the measurement of the extent

TABLE I

Identification of the sample

Description Main market Second market Overall

Companies listed on Euronext Lisbon at 31 December 2004 50 7 57

Less

Companies not subject to Portuguese law 2 0 2

Companies without annual report for 2003 1 0 1

Companies without web page 1 1 2

Companies with web page under construction or maintenance 3 0 3

Final sample 43 6 49

TABLE II

Nature of SRD by sectors

Companies Environmental

disclosure

Human resources

disclosure

Products and

consumers

disclosure

Community

involvement

disclosure

Annual

reports

Web

pages

Annual

reports

Web

pages

Annual

reports

Web

pages

Annual

reports

Web

pages

Sector n % n % n % n % n % n % n % n %

Automobiles and parts 0 0 0 0 1 100 0 0 1 100 0 0 0 0 0 0

Banks 3 50 2 33 6 100 2 33 6 100 5 83 5 83 5 83

Beverages 1 50 2 100 2 100 2 100 2 100 1 50 0 0 1 50

Chemicals 2 100 2 100 2 100 2 100 1 50 2 100 1 50 2 100

Construction and building materials 6 75 3 38 7 88 4 50 6 75 6 75 3 38 2 25

Electricity 1 100 1 100 1 100 1 100 1 100 1 100 1 100 1 100

Electronic and electrical equipment 1 100 0 0 1 100 1 100 1 100 1 100 0 0 0 0

Engineering and machinery 1 100 1 100 1 100 1 100 0 0 1 100 1 100 1 100

Food and drug retailers 2 67 2 67 3 100 2 67 2 67 1 33 1 33 2 67

Food producers and processors 0 0 0 0 2 100 1 50 1 50 0 0 0 0 0 0

Forestry and paper 2 33 3 50 3 50 3 50 2 33 1 17 1 17 1 17

Household goods and textiles 0 0 0 0 1 100 1 100 0 0 1 100 0 0 1 100

Leisure, entertainment and hotels 1 33 1 33 3 100 0 0 1 33 0 0 0 0 0 0

Media and photography 0 0 0 0 2 67 0 0 1 33 0 0 0 0 1 33

Software and computer services 0 0 0 0 4 100 2 50 1 25 3 75 0 0 0 0

Telecommunication services 2 100 2 100 2 100 2 100 1 50 2 100 1 50 2 100

Transport 1 33 1 33 3 100 1 33 2 67 1 33 0 0 1 33

Total 23 47 20 41 44 90 25 51 29 59 26 53 14 29 20 41

%: Percentage of disclosing companies in the sector (except for the final line in the table where disclosing companies as a

percentage of total sample is reported).

692 Manuel Castelo Branco and Lúcia Lima Rodrigues

of information disclosure and, therefore, the coded

data do not reflect the emphasis that companies

attach to each information item (Zéghal and Ahmed,

1990, p. 42). However, the number of different

topics discussed is considered as a reasonable measure

of management’s willingness to provide social

responsibility information in general (Bewley and Li,

2000, p. 206). On the other hand, we consider it to

be a more appropriate method than counting of

sentences, words or proportion of pages when one is

comparing such different media of disclosure as

annual reports and web pages.

Thus, the analysis of the SRD is made using an

equal-weighted index, that is, a scoring system

which assigns a point for each SRD theme pertain-

ing to any of the categories considered. Disclosure

scores for each company are added and not weigh-

ted, because it is assumed that each item of disclosure

is equally important.

Listed companies’ 2003 annual reports were ana-

lysed. Only the sections of the annual report where the

disclosure of social responsibility information is vol-

untary were analysed, namely the chairman’s report or

letter to the shareholders and the management report.

Each of the companies’ web pages was accessed

and analysed during the month of August 2004. The

entire web pages were examined. All links were

followed, but for the following exclusions:

• neither on-line copies of the annual report (Patten and Crampton, 2004) nor on-line

copies of social and/or environmental re-

ports, where available, were included in the

web page analysis;

• links to external press release disclosures were also not followed (but press releases of

the companies were examined for SRD)

(Patten and Crampton, 2004);

• links to company publications such as news- letters or product catalogues were not fol-

lowed.

The reason for the exclusions referred to in the

preceding paragraph is the purpose of collecting

segregated data on the two media analysed (Frost

et al., 2005, p. 91). This is done because the focus is

on the comparison of the social responsibility

information that companies choose to highlight on

their web pages with similar information disclosed

on their annual reports.

Several empirical studies in the area were of great

utility in developing the SRD index used (see, for

example, Archel, 2003; Adams et al., 1998; Gray

et al., 1995b, Hackston and Milne, 1996; Patten,

1991; Purushothaman et al., 2000; Williams and Pei,

1999). SRD refers in this study to disclosures in the

following four categories:

• environmental; • human resources; • products and consumers; • community involvement.

Environmental disclosure comprises disclosures

relating to environmental policies, environmental

management system and environmental awards

(including ISO 14001 and Eco Management and

Audit Scheme – EMAS), the environmental impacts

of products and processes, environment-related

expenditures, the environmental benefits of products,

conservation of natural resources and recycling

activities, and disclosures concerning energy

efficiency. Human resources disclosure covers such

issues as employee numbers and remuneration,

employee share ownership, employee consultation,

training and education, employment of minorities or

women, and trade union information. Products and

consumers disclosure encompasses disclosures related

to product quality (for example, third-party recog-

nition for the quality of the company’s products) and

consumer relations (for example, customer com-

plaints). Community involvement disclosure includes

disclosures relating to sponsorship (e.g. of art exhib-

its), as well as charitable donations and activities.

The total maximum score is of 30. The maximum

score for each of the categories considered is of 11 for

environmental disclosure, 9 for human resources dis-

closure, 5 for products and consumers disclosure and 5

for community involvement disclosure (see Table III).

To avoid penalising companies for not disclosing

items considered irrelevant to them, these items were

excluded. This is the case with Banks and Software and

computer services sectors, particularly regarding some of

the environmental disclosure items (pollution arising

from use of product, discussion of specific environ-

mental laws and regulations, prevention or repair of

Factors Influencing SRD by Portuguese Companies 693

damage to the environment, environmental aesthetics

and energy efficiency of products) and some of the

products and consumers disclosure items (safety and

customer safety practices). The same is thought to be the

case with companies from the Leisure, entertainment

and hotels sectors, but only regarding environmental

disclosure items.

Thus, the disclosure score indexes are constructed

to take into account these considerations:

Xmj

i¼1

di

N

This index expresses the level of disclosure for a

company j, where N is the maximum number of

relevant items a company may disclose and di is equal

to 1 if the indicator i is disclosed, and 0 otherwise.

TABLE III

SRD areas

Categories and items of disclosure Annual reports Web pages

n % n %

Environmental disclosure

Environmental policies or company concern for the environment 16 32.65 17 34.69

Environmental management, systems and audit 14 28.57 10 20.41

Pollution from business operations 11 22.45 9 18.37

Pollution arising from use of product 4 8.16 9 18.37

Discussion of specific environmental laws and regulations 4 8.16 1 2.04

Prevention or repair of damage to the environment 3 6.12 5 10.20

Conservation of natural resources and recycling activities 11 22.45 8 16.33

Sustainability 11 22.45 12 24.49

Environmental aesthetics 4 8.16 3 6.12

Conservation of energy in the conduct of business operations 10 20.41 7 14.29

Energy efficiency of products 1 2.04 2 4.08

Human resources disclosure

Employee Health and Safety 16 32.65 9 18.37

Employment of minorities or women 2 4.08 0 0.00

Employee training 29 59.18 5 10.20

Employee assistance/benefits 6 12.24 4 8.16

Employee remuneration 26 53.06 5 10.20

Employee profiles 32 65.31 19 38.78

Employee share purchase schemes 14 28.57 3 6.12

Employee morale 10 20.41 6 12.24

Industrial relations 7 14.29 0 0.00

Products and consumers disclosure

Product safety 6 12.24 7 14.29

Product quality 25 51.02 23 46.94

Disclosing of consumer safety practices 1 2.04 4 8.16

Consumer complaints/satisfaction 12 24.49 10 20.41

Provision for disabled, aged, and difficult-to-reach consumers 3 6.12 5 10.20

Community involvement disclosure

Charitable donations and activities 10 20.41 14 28.57

Support for education 14 28.57 13 26.53

Support for the arts and culture 12 24.49 14 28.57

Support for public health 3 6.12 5 10.20

Sponsoring sporting or recreational projects 10 20.41 13 26.53

%: Disclosing companies as a percentage of total sample.

694 Manuel Castelo Branco and Lúcia Lima Rodrigues

When the disclosure score index is equal to 0, it

indicates that company i does not disclose any item.

Index values equal to i = 1, … , mj mean that a level of disclosure is provided, and mj is the maximum

number of indicators di disclosed by a company j.

In the case of Banks and Software and computer

services sectors, environmental information total

score is 6; and for products and consumers, the total

score is 3 (SRD total score of 23). In the case of

companies from the Leisure, entertainment and

hotels sector, environmental information total score

is 6 (SRD total score of 25).

By comparison with other sectors, the financial

services sector has significantly lower direct envi-

ronmental impact. This is used by some authors as an

argument to exclude banks and finance companies

even in studies which analyse all the various com-

ponents of SRD (Archel, 2003). Simpson and

Kohers (2002, p. 101) characterise the banking

industry as having a limited direct pollution of the

environment and a relatively homogeneous pro-

duction process where product safety and employee

safety are minimal concerns.

In the case of banks, which represent an important

part of the sample used in this study, we have con-

sidered that environmental disclosure comprises

disclosures relating to environmental policies, envi-

ronmental management systems and environmental

awards (including ISO 14001 and Eco Management

and Audit Scheme – EMAS), lending and investment

policies (considered as pollution from business

operations), conservation of natural resources and

recycling activities, sustainability and disclosures

concerning energy efficiency. Pollution arising from

use of product and energy efficiency of products are

not concerns that banks have and, as far as we know,

there are no relevant specific environmental laws and

regulations pertaining to the banking sector.

Results and discussion

Descriptive analysis

Results suggest that companies prefer the annual

report as an SRD medium. 11 companies do not

present social responsibility information on the

Internet, whereas in the annual report the same

happens with only 5 companies. For the annual

report 44% of companies disclose 3 or 4 categories of

social responsibility information, and 45% only dis-

close information related to one or two of the cat-

egories considered. On the Internet, 38% of the

companies disclose 3 or 4 categories of social

responsibility information, and 39% only disclose

information related to one or two of the categories

considered.

Results in Table II indicate that the kind of social

responsibility information that more companies dis-

close in their annual reports is human resources

information (90%), followed by products and con-

sumers information (59%) and environmental

information (47%). On the Internet, the kind of

information that more companies provide is prod-

ucts and consumers information (53%), followed by

human resources information (51%).

Comparison of the information disclosed on the

Internet with similar information disclosed in the

annual reports in Table II indicates that community

involvement information is not disclosed as fre-

quently both on the Internet and the annual reports.

Community involvement disclosure is the only

category regarding which the Internet is the pre-

ferred media of disclosure by companies (40% of the

companies disclose this information on the Internet,

whereas only 29% of them use the annual report to

do so).

These results are similar to those reported by

Clarke and Gibson-Sweet (1999) and can be inter-

preted in a similar way. For example, Banks and

Telecommunication services are sectors with a high

visibility among consumers, and community rela-

tions disclosure is an important part of the SRD

made by companies in these sectors. As expected,

while few banks disclose environmental information,

the percentage of retailers and telecommunication

services disclosing such information is higher com-

pared to banks.

On the other hand, as Clarke and Gibson-Sweet

(1999) suggest, some industries have a larger po-

tential impact on the environment but are not as

close to the final consumer, and the public is less

aware of them. A company less well known to the

public, and involved in activities with larger po-

tential impact on the environment, would have less

reason to justify its existence to society by means of

community disclosures than a better known one.

This seems to be the case of companies in the

Factors Influencing SRD by Portuguese Companies 695

Chemicals, Construction and building materials and

Forestry and paper sectors: a larger percentage of

them disclose more environmental information than

community involvement information.

What seems more difficult to explain (and was not

expected) are the results for environmental disclo-

sures by companies in some environmentally sensi-

tive sectors, such as Construction and building

materials and Food producers and processors. These

companies do not disclose more SRD than compa-

nies from other sectors, as might be expected.

However, the fact that the companies included in

the sample are listed on the second market, and are

thus less visible, may explain the lack of disclosure, at

least in part.

Comparison of the information disclosed on the

Internet with similar information disclosed in the

annual reports in Tables II and III indicates that

environmental information and human resources

information are more evident in annual reports than

on the Internet, whereas the reverse is the case with

community involvement information. The differ-

ence only seems to be relevant in the case of human

resources information (the annual report is the pre-

ferred medium of disclosure) and community

involvement information (the Internet is the pre-

ferred medium of disclosure). With respect to

products and consumers information it is difficult to

say whether it has a stronger presence in annual

reports or on the Internet.

As Zéghal and Ahmed (1990) argue, the choice of

a medium for information disclosure is dependent on

the target public for whom the message is intended.

Because annual reports are directed at investors and

human resources are an important resource, it is

natural for investors to be interested in it. On the

other hand, because company web pages are aimed

at a broader public, including consumers, it is natural

for companies to give prominence to community

involvement and products/consumers information.

Table IV presents the results of the test on the

difference between SRD in the annual reports and

on the Internet. Both the Wilcoxon-signed rank

test and the paired sample t-test indicate significant

differences in total SRD and human resources dis-

closure (more disclosure in annual reports). Al-

though the difference in community involvement

disclosure is not statistically significant, there is an

important difference between the Internet and an-

nual reports as disclosure media (more disclosure on

the Internet).

Analysis of the main results

An analysis of the Kolmogorov-Smirnov (K-S

Lilliefors) and the Shapiro-Wilk normality test sta-

tistics suggests that dependent variables and contin-

uous independent variables are not distributed

normally. To bring the variables closer to normality

for the purpose of the regression analysis, the

dependent and independent continuous variables are

transformed by computing normal scores using Van

der Waerden’s transformation (Haniffa and Cooke,

2005). The regression analysis is performed with the

transformed variables.

Multiple regression is used to test the hypotheses

developed above. However, before conducting the

TABLE IV

Paired sample t-tests and Wilcoxon-signed rank test for SRD in annual reports and on the Internet

Paired sample t-test Wilcoxon test

Mean difference t Sig. (2-tailed) Z Asymp. Sig. (2-tailed)

Total SRD 0.063 3.452 0.001 )2.997a 0.003 Environmental disclosure 0.020 0.901 0.372 )0.316a 0.752 Human resources disclosure 0.206 6.949 0.000 )5.132a 0.000 Products and consumers disclosure )0.008 )0.186 0.853 )1.333b 0.844 Community involvement disclosure )0.041 )1.183 0.243 )1.183b 0.183

a Based on positive ranks (disclosure in annual reports > disclosure on the Internet).

b Based on negative ranks (disclosure in annual reports < disclosure on the Internet).

696 Manuel Castelo Branco and Lúcia Lima Rodrigues

analysis several diagnostic tests, such as normality

tests, autocorrelation tests, heteroscedasticity tests

and multicollinearity tests, are performed to ensure

valid conclusions are drawn based on the multiple

regression results. If the tests are not satisfied then

corrective procedures are performed. The Jarque-

Bera normality test is performed on the residuals of

all the models.

The possible existence of multicollinearity is tes-

ted based on the correlation matrix incorporating all

the independent variables (transformed data) as well

as computing the variance inflation factor (VIF).

Results indicate that multicollinearity is unlikely to

be a problem. In addition, results suggest that in

none of the regressions the highest VIF is above 3,

confirming that there is no need to be concerned

about the correlation between the independent

variables.

To test for unequal variances, White’s general

heteroscedasticity test is performed on each set of

results. 1

All chi-squared test statistics are not signif-

icant at a five percent level. Thus, the tests suggest

that a widespread heteroscedasticity problem does

not exist in the data and no corrective procedure is

undertaken to combat its presence in the data.

Multiple regression analysis is used for multivari-

ate testing of the hypotheses. Each of the dependent

variables, SRD in annual reports and on the Internet,

is regressed against the transformed independent

variables. The categorical variables are also included.

Table V presents the results of the regression models

pertaining to total SRD.

Table V reports the results of regressing the

independent variables on the dependent variable

total SRD. The F values for the two models are

significant at 0.01 level. This suggests that the

independent variables considered, when taken

together, explain total SRD and its categories taken

individually. However, this does not mean that each

of the independent variables contributes to the

explanation of the dependent variables.

The adjusted R 2 ’s suggest that approximately 34%

(in the case of annual reports) and 37% (in the case of

the Internet) of the variation in the SRD scores

between the companies can be explained by the

independent variables included in the regression

models. Only two of the independent variables are

significant in each model: size in both, media

exposure in the case of annual reports and leverage

in the case of the Internet. The coefficients of total

TABLE V

Results of the regression models for total SRD

Independent variables Disclosure media: Annual reports Disclosure media: Web pages

Coefficient estimate Coefficient estimate

(Constant) 0.109 )0.201 International experience 0.027 0.146

Leverage )0.256 )0.332** Profitability )0.091 )0.227 Size 0.490** 0.404**

Media exposure 0.376* 0.224

Environmental sensitivity )0.21 0.181 Consumer proximity )0.151 0.435

R 2

= 0.432; Adj. R 2

= 0.335;

Durbin-Watson = 1.762

R 2

= 0.465; Adj. R 2

= 0.374;

Durbin-Watson = 1.937

F = 4.460; p = 0.001 F = 5.100; p = 0.000

White heterosced. test:

Obs * R 2

= 36.933; p = 0.292

White heterosced. test:

Obs * R 2

= 28.025; p = 0.713

Jarque-Bera test: JB = 2.631;

p = 0.268

Jarque-Bera test: JB = 0.601;

p = 0.740

*Significant at the 0.10 level (2-tailed).

**Significant at the 0.05 level (2-tailed).

Factors Influencing SRD by Portuguese Companies 697

assets and media exposure are positive. This indi-

cates, as hypothesised, that as the value of these

variables increases so does a company’s SRD score.

Leverage is significant in the case of total SRD on

the Internet, presenting a negative coefficient which

suggests that the higher the leverage in a company,

the lower its SRD.

Thus, at an aggregated level, the supported

research hypotheses in the case of SRD in annual

reports are those related to size (H2) and media

exposure (H4), whereas in the case of SRD on the

Internet the supported hypothesis is the one related

to size (H2).

Regarding the results of regressing the indepen-

dent variables on each category of SRD, 2

from the

eight regression models only those which have hu-

man resources disclosure as a dependent variable are

non-significant at the 1% level. In the case of human

resources disclosure in annual reports the regression

is significant at the 2% level, whereas in the case of

similar disclosure on the Internet the regression is

non-significant.

The explanatory power of the regression ranges

from 7.5% for the human resources disclosure on

the Internet to 49.5% for community involvement

disclosure on the Internet. As for the importance of

the independent variables in explaining variation

between companies’ disclosure, the size variable is

significant with positive coefficients in almost all

the regression models. The exceptions are the

models which have as dependent variables products

and consumers disclosure in annual reports and

human resources disclosure on the Internet. The

media exposure variable is significant with positive

coefficients only when products and consumers

disclosure is the independent variable (both in

annual reports and on the Internet). Consumer

proximity is significant with a positive coefficient

in the case of community involvement disclosure

on the Internet, which leads to a conclusion that

H3a is accepted in the case of disclosure on the

Internet. In addition, leverage is significant with a

negative coefficient in the case of human resources

disclosure on the Internet, and profitability is sig-

nificant with a positive coefficient in the case of

products and consumers disclosure in annual

reports.

Table VI summarises the results on hypothesis

testing.

Consistent with previous studies size and media

exposure, which are considered as proxies for social

visibility, have in general a positive relationship with

total SRD. These results are consistent with the

expectations resulting from the theoretical frame-

work proposed and with previous SRD studies.

The non-significant relation between SRD and

international experience in both media of disclosure

considered is an unexpected result. These results are

consistent with those of Choi (1999), who analysed

environmental disclosure practices of Korean listed

TABLE VI

Summary of the results from the hypotheses testing

Variables Hypotheses Annual reports Web pages

International experience Positive relation Non-significant Non-significant

Company size Positive relation Positive relation Positive relation

Non-significant: products

and consumers

Non-significant: human

resources

Media exposure Positive relation Positive relation: products

and consumers and total

SRD

Positive relation: products

and consumers

Non-significant Non-significant

Environmental visibility Positive relation Non-significant Non-significant

Consumer proximity Positive relation Non-significant Positive relation:

community involvement

Non-significant

698 Manuel Castelo Branco and Lúcia Lima Rodrigues

companies and did not find a significant association

between international experience (which he calls

‘‘foreign customers’ influence’’) and disclosure.

Regarding consumer proximity, a significant

positive relation is only discernible in the case of

Internet community involvement disclosure.

Although a positive relation in the case of annual

reports’ community involvement disclosure is also

hypothesised, these results are considered to be

consistent with the theoretical framework proposed.

Because company web pages are aimed at a broader

public than annual reports, it is natural for companies

to give prominence to the Internet as a media of

disclosing their community involvement activities.

Environmental visibility is not a factor which

explains the differences in environmental disclosure

among companies. This is an unexpected finding.

The theoretical framework proposed leads to the

expectation that higher environmental visibility is

associated to higher levels of disclosure, and the

findings of previous SRD studies are consistent

with such expectation. These findings are a sign

that companies with a more limited environmental

impact are also disclosing environmental informa-

tion. For example, banks are increasingly disclosing

this kind of information. This is probably explained

by the fact that in the last two decades the focus has

been on environmental responsibility and disclo-

sure.

Portuguese firms seem to be quite sensitive to

public perceptions, as proxied by their media visi-

bility and their size, when determining their SRD

strategies. These findings are well documented in the

literature, both in Anglo-Saxon countries (see, for

example, Bewley and Li, 2000; Neu et al., 1998;

Patten, 1991) and in continental Europe (Cormier

and Magnan, 2003; Cormier et al., 2005). These

results are also consistent with literature from other

countries (Choi, 1999; Purushothaman et al., 2000).

Cormier and Magnan (2003, p. 58), who have

analysed French firms’ environmental reporting

practices, suggest that ‘‘corporate disclosure strategies

seem to be determined in a similar way, irrespective

of a given country’s socio-cultural environment’’.

These authors consider that this ‘‘is an illustration

of the strong impact of globalised stock markets

on fostering convergence in corporate practices’’

(ibid.).

Concluding remarks

This study analyses some factors which influence

SRD by a sample of companies listed on the

Portuguese Stock Exchange (Euronext – Lisbon),

using a theoretical framework which combines

legitimacy theory and a resource-based perspective.

According to this framework, managers increasingly

need to consider SRD as a signal of improved social

and environmental conduct in those fields because

disclosure influences the external perception of

reputation. By demonstrating that they operate in

accordance with social and ethical criteria, compa-

nies can build reputation, whereas failing to do so

can be a source of reputational risk.

Portuguese companies attribute greater impor-

tance to annual reports as disclosure media than to

the Internet. Noticeable differences are related to the

much higher presence of human resources infor-

mation in annual reports than on the Internet and

the higher presence of community involvement

information on the Internet than in annual reports.

These results are probably related to the fact that

annual reports are directed at investors and it is

natural for investors to be interested in human

resources. On the other hand, because company web

pages are aimed at a broader public, it is natural for

companies to give prominence to community

involvement information.

Evidence seems to suggest that companies with

higher visibility exhibit greater concern to improve

corporate image through SRD both on the Internet

and in annual reports. In addition, in sectors with a

high visibility among consumers there is greater con-

cern for community involvement activities and dis-

closure. There is thus some support for the use of a

combination of legitimacy theory with resource-based

perspectives to explain SRD by Portuguese listed

companies.

An interesting result is related to the seeming lack

of significant difference between the factors influ-

encing SRD practices of Portuguese listed compa-

nies when compared to companies from more

developed countries. We consider that there is no

reason to expect that companies in the less devel-

oped capital markets will behave in a significantly

different manner than companies in more developed

capital markets.

Factors Influencing SRD by Portuguese Companies 699

We interpret the findings as a result of the con-

vergence in corporate practices which is promoted

by the impact of globalised stock markets (Cormier

and Magnan, 2003, p. 58) and has as consequence a

seeming lack of importance of general contextual

factors in determining disclosure practices of listed

companies. Although further analyses are required to

validate this claim, it is a promising avenue for future

research (ibid).

Furthermore, we believe that such similitude in

the way in which disclosure practices are determined

is not as likely to happen in the case of unlisted

companies. SRD practices of listed companies are

less subject to general contextual factors than those

of unlisted companies. An interesting possible

extension of this study would be to use a sample of

companies which are not listed, including small and

medium-sized companies.

Other possible extensions of this study, which are

not mutually exclusive, may be envisaged to add

new insights to the analysis of SRD by companies.

One such possible extension is related to the use of

more refined content analysis procedures. Another

possible extension is an in-depth analysis of catego-

ries of SRD, which very likely would involve

variations to the theoretical framework proposed.

Finally, the use of a larger sample would be an

important way of adding new insights to the analysis

of SRD by Portuguese companies.

Notes

1 Only the results of the White heteroscedasticity test

using cross terms are reported. 2

Detailed results of regressing the independent vari-

ables on each category of SRD are available from the

authors on request.

References

Adams, C. A., W.-Y. Hill and C. B. Roberts: 1998,

‘Corporate Social Reporting Practices in Western

Europe: Legitimating Corporate Behaviour’, The

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

Archel, P.: 2003, ‘La divulgación de la información social

y medioambiental de la gran empresa española en el

perı́odo 1994–1998: situación actual y perspectives’,

Revista Española de Financiación y Contabilidad 117,

571–601.

Bansal, P.: 2005, ‘Evolving Sustainability: A Longitudinal

Study of Corporate Sustainable Development’, Strategic

Management Journal 26(3), 197–218.

Belkaoui, A. and P. G. Karpik: 1989, ‘Determinants of

the Corporate Decision to Disclose Social Informa-

tion’, Accounting, Auditing and Accountability Journal

2(1), 36–51.

Bewley, K. and Y. Li: 2000, ‘Disclosure of Environ-

mental Information by Canadian Manufacturing

Companies: A Voluntary Disclosure Perspective’,

Advances in Environmental Accounting and Management 1,

201–226.

Branco, M. C. and L. L. Rodrigues: 2006, ‘Corporate

Social Responsibility and Resource Based Perspec-

tives’, Journal of Business Ethics 69(2), 111–132.

Brammer, S. and S. Pavelin: 2004a, ‘Building a Good

Reputation’, European Management Journal 22(6), 704–

713.

Brammer, S. and S. Pavelin: 2004b, ‘Voluntary Social

Disclosures by Large UK Companies’, Business Ethics:

A European Review 13(2/3), 86–99.

Brammer, S. and S. Pavelin: 2007, ‘Factors Influencing

the Quality of Corporate Environmental Disclosure’,

Business Strategy and the Environment (in press).

Campbell, D., B. Craven and P. Shrives: 2003, ‘Volun-

tary Social Reporting in Three FTSE Sectors: A

Comment on Perception and Legitimacy’, Accounting,

Auditing and Accountability Journal 16(4), 558–581.

Campbell, D., G. Moore and P. Shrives: 2006, ‘Cross-

Sectional Effects in Community Disclosure’, Account-

ing, Auditing and Accountability Journal 19(1), 96–114.

Choi, J.-S.: 1999, ‘An Investigation of the Initial Vol-

untary Environmental Disclosures Made in Korean

Semi-Annual Financial Reports’, Pacific Accounting

Review 11(1), 73–102.

Clarke, J. and M. Gibson-Sweet: 1999, ‘The Use of

Corporate Social Disclosures in the Management of

Reputation and Legitimacy: A Cross Sectoral Analysis

of UK Top 100 Companies’, Business Ethics: A European

Review 8(1), 5–13.

Cormier, D., I. M. Gordon and M. Magnan: 2004,

‘Corporate Environmental Disclosure: Contrasting

Management’s Perceptions with Reality’, Journal of

Business Ethics 49(2), 143–165.

Cormier, D. and M. Magnan: 2003, ‘Environmental

Reporting Management: A Continental European

Perspective’, Journal of Accounting and Public Policy

22(1), 43–62.

Cormier, D., M. Magnan and B. Van Velthoven: 2005,

‘Environmental Disclosure Quality in Large German

Companies: Economic Incentives, Public Pressures or

700 Manuel Castelo Branco and Lúcia Lima Rodrigues

Institutional Conditions?’, European Accounting Review

14(1), 3–39.

Deegan, C.: 2002, ‘The Legitimising Effect of Social and

Environmental Disclosures – A Theoretical Founda-

tion’, Accounting, Auditing and Accountability Journal

15(3), 282–311.

Deephouse, D. L. and S. M. Carter: 2005, ‘An Exami-

nation of Differences Between Organizational Legiti-

macy and Organizational Reputation’, Journal of

Management Studies 42(2), 329–360.

Frost, G., S. Jones, J. Loftus and S. Van Der Laan: 2005,

‘A Survey of Sustainability Reporting Practices of

Australian Reporting Entities’, Australian Accounting

Review 15(1), 89–96.

Gray, R., R. Kouhy and S. Lavers: 1995a, ‘Corporate Social

and Environmental Reporting: A Review of the Litera-

ture and A Longitudinal Study of UK Disclosure’,

Accounting, Auditing and Accountability Journal 8(2), 47–77.

Gray, R., R. Kouhy and S. Lavers: 1995b, ‘Methodo-

logical Themes: Constructing a Research Database of

Social and Environmental Reporting by UK Compa-

nies’, Accounting, Auditing and Accountability Journal

8(2), 78–101.

Hackston, D. and M. J. Milne: 1996, ‘Some Determinants

of Social and Environmental Disclosures in New

Zealand Companies’, Accounting, Auditing and Account-

ability Journal 9(1), 77–108.

Haniffa, R. M. and T. E. Cooke: 2005, ‘The Impact of

Culture and Governance on Corporate Social

Reporting’, Journal of Accounting and Public Policy 24(5),

391–430.

Hasseldine, J., A. I. Salama and J. S. Toms: 2005,

‘Quantity Versus Quality: The Impact of Environ-

mental Disclosures on the Reputations of UK PLCS’,

The British Accounting Review 37(2), 231–248.

Lopes, P. T. and L. L. Rodrigues: 2007, ‘Accounting for

Financial Instruments: An Analysis of the Determi-

nants of Disclosure in the Portuguese Stock

Exchange’, The International Journal of Accounting

42(1), 25–56.

McGuire, J. B., A. Sundgren and T. Schneeweis: 1988,

‘Corporate Social Responsibility and Firm Financial

Performance’, Academy of Management Journal 31(4),

854–872.

Neu, D., H. Warsame and K. Pedwell: 1998, ‘Managing

Public Impressions: Environmental Disclosures in

Annual Reports’, Accounting, Organizations and Society

23(3), 265–282.

Patten, D. M.: 1991, ‘Exposure, Legitimacy, and Social

Disclosure’, Journal of Accounting and Public Policy 10(4),

297–308.

Patten, D. M.: 2002a, ‘Give or Take on the Internet: An

Examination of the Disclosure Practices of Insurance

Firm Web Innovators’, Journal of Business Ethics 36(3),

247–259.

Patten, D. M.: 2002b, ‘Media Exposure, Public Policy

Pressure, and Environmental Disclosure: An Exami-

nation of the Impact of Tri Data Availability’,

Accounting Forum 26(2), 152–171.

Patten, D. M. and W. Crampton: 2004, ‘Legitimacy and

the Internet: An Examination of Corporate Web Page

Environmental Disclosures’, Advances in Environmental

Accounting and Management 2, 31–57.

Purushothaman, M., G. Tower, R. Hancock and

R. Taplin: 2000, ‘Determinants of Corporate Social

Reporting Practices of Listed Singapore Companies’,

Pacific Accounting Review 12(2), 101–133.

Roberts, R. W.: 1992, ‘Determinants of Corporate Social

Responsibility Disclosure: An Application of Stake-

holder Theory’, Accounting, Organizations and Society

17(6), 595–612.

Simpson, W. G. and T. Kohers: 2002, ‘The Link Between

Corporate Social and Financial Performance: Evidence

from the Banking Industry’, Journal of Business Ethics

35(2), 97–109.

Toms, J. S.: 2002, ‘Company Resources, Quality Signals

and the Determinants of Corporate Environmental

Reputation: Some UK Evidence’, The British

Accounting Review 34(3), 257–282.

Williams, S. M. and C.-A. H. W. Pei: 1999, ‘Corporate

Social Disclosures by Listed Companies on Their Web

Pages: An International Comparison’, The International

Journal of Accounting 34(3), 389–419.

Zéghal, D. and S. A. Ahmed: 1990, ‘Comparison of

Social Responsibility Information Disclosure Media

Used by Canadian Firms’, Accounting, Auditing and

Accountability Journal 3(1), 38–53.

Zimmerman, M. A. and G. J. Zeitz: 2002, ‘Beyond

Survival: Achieving New Venture Growth by Build-

ing Legitimacy’, Academy of Management Review 27(3),

414–431.

Manuel Castelo Branco

Faculty of Economics,

University of Porto,

Rua Dr. Roberto Frias,

4200-464 Porto, Portugal

E-mail: [email protected]

Lúcia Lima Rodrigues

School of Economics and Management,

University of Minho,

Campus de Gualtar,

4710-057 Braga, Portugal

E-mail: [email protected]

Factors Influencing SRD by Portuguese Companies 701