Accepted Manuscript

Drivers of innovation capital disclosure in intellectual capital statements: Evidence

from Europe
Lucia Bellora, Thomas W. Guenther
PII:

S0890-8389(13)00042-5

DOI:

10.1016/j.bar.2013.06.002

Reference:

YBARE 625

To appear in:

The British Accounting Review

Please cite this article as: Bellora, L., Guenther, T.W, Drivers of innovation capital disclosure in
intellectual capital statements: Evidence from Europe, The British Accounting Review (2013), doi:
10.1016/j.bar.2013.06.002.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

SC

RI
PT

Drivers of innovation capital disclosure

in intellectual capital statements:
Evidence from Europe

M
AN
U

Bellora, Lucia
Guenther, Thomas W. 1

AC
C

EP

TE
D

(Technische Universitt Dresden, Germany)

Lucia Bellora (PhD student)

Thomas W. Guenther (Full professor)
Technische Universitt Dresden (TU Dresden)
Faculty of Business Management and Economics
Chair of Business Management, esp. Management Accounting and Control
D-01062 Dresden
Germany
Corresponding author:
Lucia Bellora
Phone: +49 351 463 35274
Email: lucia.bellora@tu-dresden.de

ACCEPTED MANUSCRIPT
2

Drivers of innovation capital disclosure in intellectual capital statements:

Evidence from Europe

RI
PT

Abstract
Innovation is one of the major determinants of competitive success. As a result, there is demand for information on the innovation activities of firms among investors, other stake-

SC

holders and the public. Using content analysis, this paper examines the innovation capital
disclosure (INCD) characteristics (i.e. disclosure quantity and quality) in the intellectual capi-

M
AN
U

tal statements (ICS) of 51 European for-profit firms. Additionally, the relationships between
INCD characteristics and industry, firm size, region of registered office and the disclosure
guidelines adopted are analysed. Our content analysis detects an average of 29.16 items on
innovation capital (INC) per ICS. These are mainly qualitative, non-financial and historically
orientated. Furthermore, as expected, industry, firm size, region and disclosure guidelines

TE
D

drive the quantity of disclosure. Prior empirical studies of voluntary disclosure in documents
other than ICS have also suggested a relationship between firm size and disclosure quality.
Interestingly, our results for INCD in ICS do not support this relationship. This provides ten-

EP

tative evidence for a similar qualitative level of innovation capital disclosure across firm size.
Furthermore, our findings show mostly homogeneous disclosure patterns across the regions

AC
C

in Europe, suggesting that multinational efforts towards fostering INCD has made the ICS
phenomenon more a European than a local phenomenon.
Keywords: content analysis, disclosure, innovation capital, intellectual capital statements
We are indebted to two anonymous reviewers, to the associate editor and to Mike Jones for their valuable comments on previous versions of the paper. The authors appreciate the useful comments of Jan Mouritsen and of
the participants in the 2008 EIASM Workshop on Visualising, Measuring and Managing Intangibles and Intellectual Capital in Hasselt (Belgium) on a first draft of this paper.

ACCEPTED MANUSCRIPT
2

1.

Introduction
Innovation capital (INC) is that part of the intellectual capital of a firm that describes the

ability of a firm to generate and use innovative solutions, and related results in terms of intel-

RI
PT

lectual property rights and other tangible, intangible and financial assets (Edvinsson and
Malone, 1997). The transformation of western economies from an industrial to a knowledgebased structure increases the relevance of INC. On a macro-economic level, innovation has

SC

become a major driver of economic growth, while traditional growth enhancers such as investment in physical assets are declining in importance (OECD, 2010). On a micro-economic

turns (Bowen, Rostami and Steel, 2010).

M
AN
U

level, innovations are recognised as means of gaining competitive advantage and higher re-

In light of the macro- and micro-economic relevance of innovation efforts, related disclosure is of interest for both individuals and organisations. However, mandatory disclosure according to IFRS or local GAAP is restrictive, as usually only data on research and develop-

TE
D

ment (R&D) expenses have to be reported (e.g. IAS 38). Thus, voluntary disclosure can provide additional information (e.g. European Commission, 2006). While there are a number of
studies dealing with the voluntary disclosure of intellectual capital in general or human capi-

EP

tal disclosure in particular, we did not find any studies dealing with voluntary innovation
capital disclosure (INCD). In our paper, we address this gap by analysing the INCD behav-

AC
C

iour of a sample of European firms.

Disclosure behaviour can be explained using a cost/benefit approach. Potential benefits are
described through various theories. According to legitimacy theory (developed by Parsons,
1960 and Weber, 1978; suggested for intellectual capital research by Guthrie, Petty, Yongvanich, and Ricceri, 2004), firms voluntarily report information to the public in order to improve their reputations (and thus the necessary inflow of capital, labour and customers) by
demonstrating that socially desirable actions, such as those leading to innovations, are being

ACCEPTED MANUSCRIPT
3

taken (Adams, Hill, and Roberts, 1998; Cho and Patten 2007; Clarkson, Li, Richardson, and
Vasvari, 2008; Guthrie et al., 2004). Closely linked to this, the stakeholder approach
(Freeman, 1984) suggests the necessity of providing information to individuals and organisa-

RI
PT

tions affecting, or being affected by, the firms actions (Clarkson, 1995). According to
agency theory, voluntary disclosure narrows the information gap between managers (agents)
and investors (principals) (Healy and Palepu, 2001). As a consequence, investors uncertainty

SC

about expected returns and thus the costs of capital decrease (e.g. Botosan, 1997). Besides
these benefits, the direct costs as well as indirect costs of disclosure play a major role in

M
AN
U

firms disclosure decisions. Direct costs include costs for the preparation of the disclosure
(e.g. the collection of information by employees) and for dissemination (e.g. printing reports),
whereas indirect costs include opportunity costs (e.g. costs derived from the postponement of
other activities of the firm due to the need to allocate human and financial resources to the
disclosure process) and proprietary costs (e.g. costs derived from revealing competition-

TE
D

sensitive information). Hence, firms will decide voluntarily to disclose information on INC as
long as the benefits of disclosure equal or outweigh its costs (Depoers, 2000; Leuz and Wysocki, 2008). This cost/benefit approach explains differences in disclosure levels between

EP

firms. Empirical research cannot directly measure these costs and benefits. However, exploring differences in disclosure levels due to different firm characteristics (i.e. drivers) allows

AC
C

for some tentative inferences on the cost/benefit relationship for specific firm samples
(Depoers, 2000). In our paper, we address industry, size, region of registered office and the
disclosure guidelines adopted as potential drivers.
The costs and benefits of disclosure can vary not only on the basis of how much information about a topic is disclosed by a firm (disclosure quantity), but also on the basis of how the
information is disclosed (disclosure quality) (Beretta and Bozzolan, 2008; Marston and
Shrives, 1991). In our paper, we address both disclosure quantity, on the basis of frequency

ACCEPTED MANUSCRIPT
4

counts, and disclosure quality, on the basis of different types of disclosures (financial vs. nonfinancial, historical vs. forward-looking, etc.).
Among the different disclosure channels where INCD might be found, we focus on intel-

RI
PT

lectual capital statements (ICS). These are documents issued on a voluntary basis for external
reporting purposes, enabling companies to present their intangible assets and therewith their
innovation efforts in a narrative, visual and numeric form. ICS usually incorporate narra-

SC

tives that describe how intangible assets are created, mantained and used within the firm as
well as some indicators intended to make the firms intangible assets more comparable to

M
AN
U

other firms (Mouritsen, Larsen and Bukh, 2001c).

Our paper contributes to the research in three ways. First, we carry out a quantitative content analysis of ICS. Content analysis is a methodology that codifies text into categories and
quantifies the frequency of occurrences within each category (Krippendorff, 2004). The current literature provides some studies on what topics are addressed in ICS. We add to the ex-

TE
D

isting research by investigating the frequency and quality level of the reported content. Thus,
practice will be able to use our results in order to expand the quality of ICS disclosure.
Second, the previous literature on voluntary disclosure has mainly focused on intellectual

EP

capital as a whole, providing a broad understanding of intellectual capital disclosure (Guthrie,

Ricceri and Dumay, 2012). Few studies have paid attention to only one category of intellec-

AC
C

tual capital, i.e. human capital (e.g. Abeysekera and Guthrie, 2004b). However, research explicitly addressing other intellectual capital categories (e.g. INC) to our best knowledge is
lacking. This stands in contrast to the fact that innovation plays a prominent role in the enhancement of a firms value creation (e.g. Lev, 2001).
Furthermore, innovations play a specific role within the network of interrelated intellectual
capital (Lev, 2001). For example, patents (internal capital) become relevant for firm value if
they are used by employees (human capital) for the development of innovative products. Em-

ACCEPTED MANUSCRIPT
5

ployee training (human capital) is required for the development of successful innovations.
The establishment of a network of cooperation and experience exchange with other firms
(external capital) is a prerequisite for the achievement of knowledge necessary for the devel-

RI
PT

opment of innovations (e.g. Mouritsen, Bukh and Bang, 2008; Mouritsen, Johansen, Larsen
and Bukh, 2001b). In other words, INC is the result of the emphasis placed on different types
of intellectual capital, such as human or external capital. Our study especially considers INC

SC

as an intellectual capital category and offers a basis for future research on the role of other
categories of intellectual capital in the intellectual capital network.

M
AN
U

Third, our study does not only rely on the description of the disclosure of INC in ICS, but
rather searches for possible explanations of differences between firms concerning disclosure
quantity and quality. From our sample of 51 ICS from Nordic, German-speaking and southern European countries, encompassing ICS from firms from various industries, of varying
sizes and employing diverse disclosure guidelines, we make inferences about different

TE
D

cost/benefit relationships governing different types of firms. While drivers of intellectual

capital disclosure have already been considered in prior studies (e.g. Vergauwen, Bollen and
Oirbans, 2007), our analysis is the first to offer evidence on the drivers of INCD in ICS. Dis-

EP

closure guideline-setters can use our results to evaluate their efforts in directing INCD towards high-quality data and harmonising the use of ICS across countries.

AC
C

The remainder of this paper is organised as follows. Comparable studies on voluntary intellectual capital disclosure are reviewed in section 2; section 3 develops the hypotheses; section 4 describes the content analysis carried out; the empirical evidence is presented in section 5; and section 6 provides a discussion of the results and concludes.
2.

Literature review
Traditionally, firms have concentrated their management activities on tangible and finan-

cial assets. More recently, the focus has widened to intangible assets such as human capital or

ACCEPTED MANUSCRIPT
6

INC (Lev, 2001; OECD, 2010). As a reaction to these developments, different institutions
and committees have recommended a stronger future orientation and a focus on non-financial
items in disclosure (e.g. AICPA, 1994; ICAEW, 2005; IASB, 2007). In parallel with these

RI
PT

efforts, various governmental institutions and researchers have provided different ICS disclosure guidelines since the mid-1990s. Appendix 1 presents the most common guidelines consistent with those surveyed by the European Commission (2006) or Ricceri (2008). These

SC

guidelines have been identified on the basis of extensive literature and internet research for
European guidelines proposing a consolidated numeric and narrative report for external dis-

M
AN
U

closure purposes by for-profit firms. All these guidelines refer to the tripartite distinction between internal, external and human capital proposed by Edvinsson et al. (1997), but differ in
the degree and in the way in which INCD is addressed (European Commission, 2006). For
example, while the Austrian Research Centers (2000) guideline is designed for researchintensive firms and thus it devotes considerable attention to research-, development- and

TE
D

technology-related disclosure, Danish Guidelines (Danish Ministry of Science Technology

and Innovation, 2003) and the Putting IC into Practice project (Thorleifsdottir and Claessen,
2006) address a wider range of firms and thus dedicate less attention to INCD. Approaches to

EP

disclosure also differ. While the Modelo Intelect (Euroforum, 1998), the German Federal
Ministry of Economics and Technology (2008) and the Austrian Research Centers (2000)

AC
C

guidelines favour a quantification approach by indicators, Danish guidelines (Danish Ministry of Science Technology and Innovation, 2003) and the Putting IC into Practice project
(Thorleifsdottir et al., 2006) dedicate much emphasis to storytelling and thus to a more qualitative approach to disclosure (European Commission, 2006).
This reorientation from tangible and financial assets to intangible assets is also reflected in
the research on disclosure. Guthrie et al. (2012) give an overview on a decade of intellectual
capital accounting research. Campbell and Abdul Rahman (2010) and Striukova, Unerman

ACCEPTED MANUSCRIPT
7

and Guthrie (2008) offer useful overviews of previous content analyses of intellectual capital
disclosure by for-profit firms. Their overviews show three research streams. The first stream
deals with the disclosure of intellectual capital in annual reports without focusing on a par-

RI
PT

ticular intellectual capital category (this includes various studies since 1999). The category
system employed is informed by the work by Guthrie, Petty, Ferrier and Wells (1999) and is
based on the tripartite distinction between internal, external and human capital. Guthrie et al.

SC

(1999) analyse the intellectual capital disclosure of 20 Australian firms in annual reports.
This study has since been replicated by other researchers for other countries (e.g. Brennan

M
AN
U

(2001) for 11 Irish firms; April, Bosma and Deglon (2003) for 20 South African firms; Bozzolan, Favotto and Ricceri (2003) for 30 Italian firms; Abdolmohammadi (2005) for 58 US
firms over five years and Guthrie, Petty and Ricceri (2006) for 100 Hong Kong and 50 Australian firms; Campbell et al. (2010) for one U.K. firm for the 1978-2008 period; Li, Mangena and Pike (2012) for 100 U.K. firms). Bontis (2003) applies a different framework to

TE
D

investigate 10,000 Canadian firms. Overall, these studies either focus only on a quantification
of disclosure or additionally identify potential drivers of disclosure. In general, external capital disclosure is prevalent over other intellectual capital disclosure categories. In most cases,

EP

the sample encompasses large firms from a single country.

A second stream addressed the disclosure of one particular type of intellectual capital in

AC
C

annual reports. This research is focused on human capital in annual reports (Abeysekera and
Guthrie, 2004a; Olsson, 2001; Subbarao and Zeghal, 1997). We did not find evidence of studies explicitly addressing other types of intellectual capital such as, for example, INC. Furthermore, all the studies in this stream of research were descriptive and they quantified the
disclosure of human capital without either considering drivers for disclosure or measuring
disclosure quality.

ACCEPTED MANUSCRIPT
8

Recent findings demonstrate that these two streams of research are limited since they both
focus only on annual reports. Striukova et al. (2008) show that the disclosure of intellectual
capital in annual reports is not a good proxy for the overall disclosure of intellectual capital in

RI
PT

different types of corporate reporting channels (e.g. webpages, interim reports, etc.). This
limitation is partially addressed by a third stream of research, which considers intellectual
capital disclosure in ICS instead of that in annual reports (Table 1).

SC

[insert Table 1 about here]

This stream includes a limited number of studies qualitatively analysing the content of ICS

M
AN
U

(compared with the numerous studies dealing with intellectual capital disclosure in annual
reports), despite the clear focus of ICS on intellectual capital disclosure. Overall, these studies outline how different firms present their intellectual capital in ICS. However, the studies
differ in the ways in which they investigate the ICS under review and in their perspectives on
INCD. While Ordez de Pablos (2002, 2003, 2005) investigates ICS in terms of indicators

TE
D

provided, the approach by Mouritsen et al. (2001c), Mouritsen, Larsen and Bukh (2001a) and
Mouritsen et al. (2008) is directed more broadly to understanding the relationships among
knowledge narratives, knowledge management challenges and indicators. Therefore, the per-

EP

spectives on INCD that can be drawn from these studies are different. While the Ordez de
Pablos (2002, 2003, 2005) studies consider only the indicators of INC (e.g. number of new

AC
C

products), Mouritsen et al. (2001a, 2001c, 2008) analyse both quantitative and qualitative
INCD and also refer to the narrative part of ICS. These latter studies, particularly the analysis
of the ICS of Maxon Telecom (Mouritsen et al., 2008), show that INCD is not confined to
indicators from one intellectual capital category, but is rather distributed across internal (e.g.
number of patents), external (e.g. number of customers involved in the innovation process)
and human capital (e.g. number of innovation-related training sessions). Furthermore, this

ACCEPTED MANUSCRIPT
9

study finds that INCD is also a part of the narratives that explain how relationships between
intellectual capital resources generate innovations.
In summary, the studies in Table 1 provide our first evidence for the use of ICS for INCD.

RI
PT

In our study, we also focus on ICS in our INCD investigation for three reasons. First, ICS are
exclusively aimed at presenting firms intangible assets, including INC (Mouritsen et al.,
2001c). A survey reveals that one of the major motives for working with ICS is to highlight

SC

the firms innovation efforts (Mouritsen et al., 2001c). Second, ICS do not have space restrictions and they seem thus to be more appropriate than other disclosure documents with limited

M
AN
U

space when researching the voluntary disclosure of INC (Striukova et al., 2008). Third, ICS
disclosure is accessible and affordable for firms of any size (e.g. there are no printing costs,
since ICS are mostly only available online) (Mouritsen et al., 2001c), which fosters the analysis of the disclosure of large and small firms.

Our study adds to the qualitative content analyses of the third stream of research by using

TE
D

a quantitative approach for both collecting data and drawing inferences. Furthermore, we
support the second stream of research by drawing attention to a different type of intellectual
capital, i.e. INC, in addition to the analyses of human capital and its disclosure drivers.
Development of hypotheses

EP

3.

Our paper aims to answer the questions of (1) the quantity and quality of information pro-

AC
C

vided on INC in ICS and (2) the relationship between firm characteristics and these disclosure characteristics. For the second research question, we expect a relationship between disclosure on one hand and the industry, size and region of registered office of the firm and the
disclosure guidelines adopted by the firm on the other hand. The rationale for these hypotheses is outlined in the following.
Following legitimacy theory, firms in high R&D intensity industries need to be licensed
by the environment in which they operate (European Commission, 2006). This may drive

ACCEPTED MANUSCRIPT
10

them to provide more, higher-quality INCD than firms that do not need this license due to
their memberships of a low R&D intensity industry (Patten, 1991). This relationship is also
supported by the stakeholder approach, which suggests that firms in high R&D intensity in-

RI
PT

dustries will profit from communicating their potential by disclosing their innovation efforts
in order to attract resources from stakeholders. Furthermore, as the potential of a firms R&D
activities is better known to managers than it is to investors, the information asymmetry be-

SC

tween agents and principals is more pronounced in high R&D intensity firms than it is in low
R&D intensity firms. Therefore, high R&D intensity firms can reduce their industry-inherent

M
AN
U

information asymmetry by providing detailed disclosures of INC. In summary, we expect that

firms in high R&D intensity industries achieve a higher level of INC disclosure than firms in
low R&D intensity industries. Previous intellectual capital disclosure research has also found
an association between disclosure and industry (Boesso and Kumar, 2007; Bozzolan et al.,
2003). This leads to:

TE
D

H1(a): The quantity of INCD in ICS is higher for firms in high R&D intensity industries than
for firms in low R&D intensity industries.
H1(b): The quality of INCD in ICS is higher for firms in high R&D intensity industries than

EP

for firms in low R&D intensity industries.

Moreover, we suggest that firm size is associated with INCD. From legitimacy theory, we

AC
C

infer that the actions taken by larger firms have a higher overall impact on the society in
which they operate, which leads to higher legitimisation and results in a higher disclosure
pressure on larger firms than on smaller firms. Similarly, larger firms usually address a larger
number of stakeholders and stakeholder groups, which forces them to comply with higher
disclosure standards than is the case for smaller firms. At the same time, agency theory is
ambivalent about the association between disclosure and size. On one hand, statistics
(European Commission, 2008a) identify a higher percentage of R&D expenses and innova-

ACCEPTED MANUSCRIPT
11

tion output with larger firms, implying higher innovation-related information asymmetry and
thus higher information expectations from investors. On the other hand, smaller firms might
be interested in a higher disclosure level as a means to overcome information asymmetry be-

RI
PT

cause they are relatively unknown to investors (Leuz et al., 2008). The costs of disclosure are
likely to be proportionally higher for smaller firms, as they have to bear fixed disclosure costs
without the possibility of benefiting from the economies of scale available to larger firms

SC

(Leuz et al., 2008). Prior intellectual capital disclosure researchers have noted an association
between disclosure and firm size (Boesso et al., 2007; Bozzolan et al., 2003). In light of this

M
AN
U

ambiguous theoretical background, two non-directional hypotheses are formulated:

H2(a): The quantity of INCD in ICS is related to the size of the firm.
H2(b): The quality of INCD in ICS is related to the size of the firm.
Despite the efforts of several governmental and non-governmental institutions to increase
the use of ICS for INCD, implementation remains variant between European regions

TE
D

(European Commission, 2006). These diverse disclosure practices are influenced by regional
cultures (Chaminade and Johanson, 2003). In line with legitimacy theory, agency theory and
the stakeholder approach, the public, investors and stakeholders in countries that have a

EP

stronger intellectual capital tradition, such as Nordic countries, will be more sensitive to the
importance of the disclosure of INC and will therefore expect higher levels of disclosure.

AC
C

Furthermore, the (perceived) costs of INCD may also vary depending on region-specific disclosure attitudes. For example, Chaminade et al. (2003) suggest that southern European countries have a stronger fear of competitive disadvantages when disclosing information on intellectual capital than Nordic countries. Previous intellectual capital disclosure studies support
the relationship between disclosure and the region of legal domicile of a firm (i.e. the location
of the registered office of a firm) (Bozzolan, OReagan and Ricceri, 2006; Guthrie et al.,

ACCEPTED MANUSCRIPT
12

2006; Vandermaele, Vergauwen and Smits, 2005; Vergauwen and van Alem, 2005). Therefore, we formulate:
H3(a): The quantity of INCD in ICS is related to the region of registered office of the firm.

RI
PT

H3(b): The quality of INCD in ICS is related to the region of registered office of the firm.
For ICS, disclosure guidelines have been issued by various governmental and nongovernmental institutions as well as by researchers (Appendix 1 lists common European ICS

SC

disclosure guidelines). Firms generally incorporate the suggestions of these guidelines in their
ICS, as the guidelines are expected to summarise the expectations of investors, other stake-

M
AN
U

holders and the public. The guidelines differ according to the quantity and quality of INCD
they propose. For example, the Putting IC into Practice project (Thorleifsdottir et al., 2006)
only names a few INCD indicators as examples..The German Federal Ministry of Economics
and Technology (2008) guideline also suggests discussing the drivers of intellectual capital
(e.g. product and process innovation) concerning quantity, quality and systematic manage-

TE
D

ment (Bornemann and Alwert, 2007). This enlarges the quantity of recommended INCD.
Furthermore, the Austrian Research Centers (2000) guideline particularly addresses research
firms, thus suggesting a high proportion of indicators related to the INCD area. The disclo-

EP

sure guidelines also differ on the quality of the recommended INCD. For example, on one
hand, there are the quantification-orientated German Federal Ministry of Economics and

AC
C

Technology (2008) and Modelo Intelect (Euroforum, 1998) guidelines, while on the other
hand there is the narrative-oriented Putting IC into Practice (Thorleifsdottir et al., 2006)
guideline (e.g. European Commission, 2006; Ordez de Pablos, 2005). Additionally, the
Putting IC into Practice (Thorleifsdottir et al., 2006) and Danish Ministry of Science, Technology and Innovation (2003) guidelines suggest discussing the short- and long-term goals of
the firm, thus providing a future orientation for ICS, while other guidelines such as the ones
of the Austrian Research Centers (2000), the guidelines of the German Federal Ministry of

ACCEPTED MANUSCRIPT
13

Economics and Technology (2008) or the Modelo Intelect (Euroforum, 1998) mostly refer to
past achievements. From the above, we anticipate that:
H4(a): The quantity of INCD in ICS is related to the disclosure guidelines adopted.

4.

RI
PT

H4(b): The quality of INCD in ICS is related to the disclosure guidelines adopted.
Research design

4.1. Identification and measurement of INCD

SC

To enhance the comparability of this study with previous intellectual capital research, the
category system used to allocate the disclosures of ICS is informed by Guthrie et al. (1999),

M
AN
U

which has been adopted by numerous studies, including the recent ones by Campbell et al.
(2010) and Striukova et al. (2008). This category system distinguishes between internal, external and human capital. This allows us to reflect on the role of different types of intellectual
capital in the process of innovation generation (Lev, 2001). We capture those disclosures that
relate to the entire innovation process (including the disclosure of the involvement of cus-

TE
D

tomers in the development of innovations, training for employees to impart the skills needed
to generate and implement new product ideas, the shape of new product development processes and the results in terms of successful new products developed with cooperation partners

EP

or customers or within the firm). The categorisation was based on previous studies and adjusted to fit the context of INCD (for similar adjustments, see Bozzolan et al., 2003 or

AC
C

Striukova et al., 2008).1 We summarise the categories and subcategories, as well as related
definitions, in Appendix 2.
ICS were analysed manually. This is consistent with Beattie and Thomson (2007), Krippendorff (2004) and Weber (1990), as we rely on the superior ability of humans to interpret
and translate text (we deal with ICS in German and English) to identify firm-specific terms
and to negotiate synonyms and multiple meanings.

ACCEPTED MANUSCRIPT
14

Our hypotheses address the quantity and quality of INCD. The quantity of disclosure was
measured in previous content analyses by using as measurement units either the volume of
space devoted to an item (in terms of words, sentences or paragraphs) (e.g. Bozzolan et al.,

RI
PT

2003; Guthrie and Petty, 2000) or the frequency of disclosure (e.g. Beattie et al., 2007;
Striukova et al., 2008). In the present study, the method of counting the frequencies of disclosure is applied, with each INCD item counted independent of its length. This means that

SC

sometimes long, complex sentences may address one item if only one topic is mentioned or
several items if different topics are mentioned (Weber, 1990, p. 22). There are various ration-

M
AN
U

ales behind this decision. First, we analyse ICS both in German and in English and follow
Campbell, Beck and Shrives (2005), who find that quantifying disclosure based on the
amount of space leads to invalid inferences due to the different etymological structures in the
two languages. Second, counting the number of words, sentences or paragraphs to measure
disclosure is hardly applicable when analysing and comparing ICS, whose lengths vary con-

TE
D

siderably from report to report and where the costs of additional space (e.g. more pages per
report) is negligible (Striukova et al., 2008). Third, counting instances of disclosures acknowledges the possibility that multiple INC items are disclosed within one sentence, para-

EP

graph or page (Beattie et al., 2007).

The quantity of disclosure is determined by the number of times an INC item is reported

AC
C

per ICS. Therefore, multiple disclosures and duplicates are explicitly recorded (Beattie et al.,
2007). Similar to Botosan (1997), we formulate a quantity index for INCD (INCDQUANT)
for firm j over the i=1;2;3 categories (internal, external and human capital) as follows:

where
INCDQUANTij=instances of the disclosure of INC in category i for firm j.

ACCEPTED MANUSCRIPT
15

Additionally, disclosure researchers have noted the importance of differentiating the types
of information provided by an item (e.g. Beretta et al., 2008). Following Boesso et al. (2007),
we consider the type, nature and time-reference of disclosure and formulate the quality index

RI
PT

for INCD (INCDQUAL) as follows:

Kj=number of total text units of the ICS of firm j;

SC

where

M
AN
U

typek j= 0 if the text unit k does not provide any item on INC; 1 if the text unit provides a
qualitative item; 2 if the text unit provides a quantitative item;

naturek j= 0 if the text unit k does not provide any item on INC; 1 if the text unit provides a
financial item; 2 if the text unit provides a non-financial item;

outlookk j= 0 if the text unit k does not provide any item on INC; 1 if the text unit provides an

TE
D

historical item; 2 if the text unit provides a forward-looking item.

Consistent with Boesso et al. (2007), INCDQUAL thus assigns a score of 2 to INCD if
the text unit addresses a quantitative item, as firms make efforts to translate narratives into

EP

accountable enumerators of INC; a score of 1 is assigned if the text unit addresses a qualitative item. Furthermore, non-financial disclosures are awarded a score of 2, as financial infor-

AC
C

mation is often already available in mandatory reporting; a score of 1 is assigned if the text
unit addresses a financial item. Finally, we assign a score of 2 to forward-looking disclosure,
as this kind of information can give insights into the firms future ability to create value
through its INC, while a score of 1 is achieved for a text unit addressing a historical item
(AICPA, 1994; IASB, 2007; ICAEW, 2005; Boesso et al., 2007; Hooks, Coy and Davey,
2002; Mouritsen et al., 2001a).
4.2. Reliability and validity

ACCEPTED MANUSCRIPT
16

As content analysis involves a certain degree of subjectivity, measurement reliability and

validity should be considered in depth. Therefore, we follow the assessment criteria outlined
by Krippendorff (2004) (see Table 2) in order to safeguard the reliability and validity of the

RI
PT

data and enhance credibility. The table shows that the content analysis is highly reliable in
terms of stability, reproducibility and accuracy. Furthermore, even if validity per se can be
evaluated only tentatively, the table provides positive signals for the plausibility of our results

[insert Table 2 about here]

M
AN
U

4.3. Sample

SC

and inferences.

Following Striukova et al. (2008), we consider voluntary disclosure in documents besides

annual reports and focus in particular on ICS. Because the main interest of the present paper
is information disclosed to external users, the collection of ICS was based on public availability. To fulfil this criterion, two requirements needed to be satisfied: first, the firm issuing the

TE
D

ICS had to be mentioned by one of the guideline-setters in a publicly available list as a user
or participant in the project developing the guideline (e.g. on the web site or in a project report of the guideline-setter). Therefore, we surveyed common intellectual capital disclosure

EP

guidelines (Appendix 1) and found lists of participants for the following guidelines: the Austrian Research Centres Model; Danish ICS Guidelines; the guidelines by the German Federal

AC
C

Ministry of Economics and Technology; the Spanish Modelo Intelect; and the guidelines in
Putting IC into Practice. Second, the firm mentioned in the participant list must provide the
ICS on its webpage.2

Data collection was carried out between December 2008 and January 2009. According

to participant lists, nearly 260 identifiable ICS should be available. Because of the linguistic
capacities of the researchers, ICS not in English or German were excluded from the sample.
After reviewing public availability and language, data collection led to the detection of 126

ACCEPTED MANUSCRIPT
17

ICS. In the next step, research institutes, universities and other academic entities, as well as
non-profit organisations, were excluded. The reason for this exclusion was twofold. First, the
economic process of innovation of a for-profit firm is hardly comparable with the innovation

RI
PT

process of a non-profit firm or a research institute. Second, universities and other academic
entities are often obliged by law to disclose ICS according to a predefined structure and content (e.g. Austrian Ministry of Science and Research, 2002), thus not facilitating the analysis
of genuinely voluntary disclosure, which is the focus of our analysis. This selection proce-

SC

dure led to a final sample of 51 ICS from 51 for-profit firms.3 This sample size is comparable

M
AN
U

with other content analyses of ICS (see Table 1). As ICS disclosure is voluntary, the available
population of firms providing this type of disclosure is smaller than the population of firms
available for analyses of mandatory disclosure.

For each firm, we considered the latest ICS published. Owing to learning effects, the last
published ICS should give the most realistic picture of the information that firms are able and

TE
D

willing to disclose. Furthermore, random sample analyses of the ICS of a specific firm over a
number of years suggest that the content and structure of ICS remain relatively constant over
time, consistent with similar findings for annual reports (Lang and Lundholm, 1993). Eight

EP

European countries were included in the sample: Germany (n=24), Denmark (n=10), Iceland
(n=5), Sweden (n=4), Finland (n=3), Austria (n=2), Spain (n=2) and Norway (n=1).4 Only

AC
C

five of the firms in the sample were listed on the stock market, supporting the expectations
suggested by legitimacy theory and the stakeholder approach that all firms not only listed
firms provide the public with information about their activities and performances. The sample is summarised in Table 3.
[insert Table 3 about here]
5.

Results

5.1. INCD: Descriptive results on the quantity and quality of disclosure

ACCEPTED MANUSCRIPT
18

Table 4 displays the descriptive results of our study. We found 1,487 items dealing with
INC in the 51 ICS analysed. On average, each ICS discloses 29.16 items on INC. The highest
proportion of these items relates to the human capital category, closely followed by internal

RI
PT

and external capital. The most frequent subcategory in the human capital category is the innovativeness of employees (following the categories and definitions of INC in appendix 2),
where the qualitative or quantitative results of the innovation activities of employees are ad-

SC

dressed (e.g. the description of new products or services developed by employees or related
scientific publications of employees). The most frequent subcategory in the internal capital

M
AN
U

category is that on management processes, where firms describe their structures, procedures
and organisational practices for innovation activities (e.g. the predefined stages or average
duration of a development process). Customers and market is the most addressed subcategory in the external capital category. Here, firms disclose the involvement of customers in the
innovation process as well as the financial impact of innovations (e.g. the processes to in-

from new products).

TE
D

clude customers innovative ideas in the products to be developed or the financial returns

[insert Table 4 about here]

EP

We find a predominance of qualitative, non-financial, historically orientated disclosures.

Data on the type of disclosure flow into the calculation of the INCDQUAL. INCDQUANT

AC
C

and INCDQUAL indexes for our sample are summarised in Table 5. These data provide the
first insights into heterogeneous INCD behaviours in ICS. INCDQUANT ranges from a
minimum of one to a maximum of 109 INC items per ICS. INCDQUAL ranges from a minimum of five to a maximum of 527. The ICS with the lowest INCDQUANT and INCDQUAL
scores are by the Trio Hair & Company (a Germany-based hairdresser corporation), which
only disclosed the existence of its procedures to enhance firm innovativeness (internal capital). The firm with the highest INCDQUANT and INCDQUAL scores is Indra (a Spain-based

ACCEPTED MANUSCRIPT
19

technology corporation), whose disclosure includes, for example, R&D expenses (internal
capital), cooperation with other firms on product and process development (external capital)
and the number of employees working in R&D (human capital). The potential drivers of dif-

RI
PT

ferences in INCDQUANT and INCDQUAL are tested in the following section.

[insert Table 5 about here]
5.2. INCD: Bivariate results on potential drivers

SC

In section 3, we developed hypotheses about the potential drivers of INCD in ICS. Owing
to the categorical nature of the independent variables and exploratory character of our analy-

M
AN
U

sis, the conjectured relationships are examined using t-tests (for independent variables composed of two groups) or ANOVA (for independent variables composed of more than two
groups).5

For independent variables, we build groups as follows. For the industry variable, we distinguish between firms in high R&D intensity industries and firms in low R&D intensity indus-

TE
D

tries on the basis of the results of the Community Innovation Survey by the European Commission (2008a).6

For the firm size variable, we distinguish smaller firms from larger ones on the basis of a

EP

median split by size in terms of employee headcount in the disclosure year. Among the several measures available as size proxies (e.g. total assets, market capitalisation and turnover;

AC
C

Bozzolan et al., 2003), the employee headcount measure is the most suitable for our study for
two reasons. First, as we are exploring intellectual capital, the personnel available represent
human capital and is more adequate as a measure for size than tangible or financial assets. An
increasing number of employees implies an increasing number of stakeholders (the employees themselves, labour unions, local communities, etc.) interested in firm disclosure (Boesso
et al., 2007). Additionally, ICS requires support from the workforce involved in the disclosure process, which tends to be easier for firms with a larger headcount. Second, highly inno-

ACCEPTED MANUSCRIPT
20

vative, entrepreneurial firms may not be listed on the capital market and do not have to provide publicly financial statements, thus hampering data availability. Employee headcount, by
contrast, was provided in each ICS in the sample.

RI
PT

For geographical location, we follow Chaminade et al. (2003) and distinguish between
firms from German-speaking countries, Nordic countries and southern European countries.

Finally, in analysing differences in INCD arising from the adoption of different disclosure

SC

guidelines, we distinguish between those firms adopting German guidelines (German Federal
Ministry of Economics and Technology, 2008), those adopting Putting IC into Practice

M
AN
U

guidelines (Thorleifsdottir et al., 2006) and firms adopting other guidelines.

[insert Table 6 about here]

The bivariate results are displayed in Table 6. Consistent with H1(a) and H1(b), we find
that firms in high R&D intensity industries disclose significantly more INC items (INCDQUANT) and to a higher qualitative level (INCDQUAL) than firms in low R&D intensity

TE
D

industries (both p<0.1).7

We find a significant difference (p<0.1) in the quantity of disclosure between smaller and
larger firms, which is consistent with our conjecture in H2(a). Our sample provides evidence

EP

of greater disclosure by firms with a higher number of employees than by those with a lower
number of employees. There were no significant differences in INCDQUAL between firms

AC
C

with above the median and below the median number of employees, contradicting our hypothesis H2(b). Thus, a high level of quality in the disclosure of ICS does not seem to relate
to firms with more employees.
Interestingly, we find significant (p<0.1) regional differences in the quantity (H3(a)) but
not in the quality (H3(b)) of innovation. German-speaking and southern European countries
seem to disclose significantly more innovation-related items. However, there is no significant
difference in INCDQUAL across regions.

ACCEPTED MANUSCRIPT
21

Finally, we identify a significant relationship (p<0.05) between the disclosure guidelines

adopted by a firm and the amount as well as the quality of items disclosed on INC in ICS.
This is consistent with our hypotheses H4(a) and H4(b) and indicates that disclosure follow-

RI
PT

ing the guidelines by the German Federal Ministry of Economics and Technology (2008) as
well as other disclosure guidelines leads to a higher quantity and quality of INCD
(Thorleifsdottir et al., 2006).
Discussion of the results and limitations

SC

6.

There has been a lack of prior research into INCD, and this contrasts with the micro- and

M
AN
U

macro-economic importance of innovation. Our paper investigates 51 European ICS as instruments to disclose information about INC with regard to (1) the quantity and quality of
INCD and (2) the characteristics likely to drive firm differences in INCD quantity and quality.

Each ICS encompasses at least one and on average 29.16 items on INC. Overall, the bene-

TE
D

fits of the disclosure of INC in ICS seem to exceed its costs. This inference can be further
explored by considering how much the three INCD categories are addressed in the ICS. Human capital is the category with most INCD in ICS. Nevertheless, we find a balanced distri-

EP

bution across the three categories. This empirically supports the proposition that INC is created by investments in all intellectual capital categories (Lev, 2001). These intellectual capital

AC
C

categories are highly interwoven and they create a network that facilitates innovation and
value creation (Mouritsen et al., 2008).
Non-financial, qualitative, historical INCD seems to show the most advantageous
cost/benefit relationship. Financial information may be considered a matter for mandatory
reporting but not for voluntary disclosure, as, for example, some firms disclose R&D expenses in financial reports. The fear of proprietary costs may deter firms from quantitative,
verifiable disclosure. At the same time, qualitative texts allow us to explain how the innova-

ACCEPTED MANUSCRIPT
22

tiveness goals of a firm are achieved. Owing to space constraints in other disclosure documents such as in annual reports, this qualitative information is usually not addressable elsewhere.

RI
PT

One further point deserves attention. In contrast to the purpose of ICS as a disclosure instrument that provides insights into the future value creation of firms (e.g. Mouritsen et al.,
2001a), we find a low degree of forward-looking INCD in our ICS sample. The benefits de-

SC

rived from fulfilling the information expectations of stakeholders about the future value of
the firm and the related intentions concerning innovation efforts seem insufficient to outper-

M
AN
U

form the proprietary costs of disclosing information about future innovation behaviour. Similar results have also been found by other disclosure studies (Boesso et al., 2007; Mouritsen et
al., 2001c; Ordez de Pablos, 2005; Striukova et al., 2008), indicating that from this point of
view INCD does not differ from other forms of intellectual capital.

We find industry, size, region of registered office of the firm and disclosure guidelines

TE
D

adopted to be significantly related to the quantity of INCD in ICS, while only industry and
disclosure guidelines adopted make a difference to the quality.
According to the industry membership variable and following our theoretical background,

EP

one explanation for the relation is that firms in high R&D intensity industries are under
higher pressure from the public, investors and other stakeholders to legitimate themselves as

AC
C

part of this industry and to provide value-relevant information on innovation than firms in
low R&D intensity industries. In other words, if the costs of disclosure are presumed equal,
high R&D intensity firms profit more from a high quantity and high quality disclosure of INC
than other firms. Furthermore, high R&D intensity firms may have more information to be
disclosed than firm with low R&D intensity or, in other words, lower costs of disclosure, as
their information availability in the field of INC may be higher.. These findings are consistent
with previous results on intellectual capital disclosure in annual reports where, for example,

ACCEPTED MANUSCRIPT
23

Bozzolan et al. (2003) find intellectual capital disclosure to be higher in high technology industry firms than in other firms. The qualitative analysis of Maxon Telecoms ICS by
Mouritsen et al. (2008) also suggests that firms whose primary focus lies on research, devel-

RI
PT

opment and technology stress INCD in their disclosures, both using narratives and using indicators.

Similarly, the benefits of INCD disclosure may be higher and the costs of INCD disclosure

SC

may be lower for firms with more employees. Larger firms gain more attention from society,
investors and other stakeholders. Owing to the variety of attention and thus of interest in INC,

M
AN
U

larger firms will have to disclose more. Additionally, firms with higher headcount can usually
exploit economies of scale in disclosure, as they can create specialised departments for management control and reporting with better capabilities in data procurement and processing.
This enables firms with more employees to disclose more to lower costs per unit of information. This result confirms previous findings on voluntary disclosure in annual reports (e.g.

TE
D

Boesso et al., 2007; Bozzolan et al., 2003).

A result that deserves particular attention is that the quality of disclosure does not differ
significantly between firms with a low or a high number of employees. From this result, we

EP

infer that the benefits and costs of disclosing high quality items are similar for all firms. This
result is worth mentioning in relation to the efforts undertaken by the European Commission

AC
C

(2006) and described in the RICARDIS report to encourage SMEs to use ICS to disclose their
innovation activities. Compared with the findings by Striukova et al. (2008), who analyse a
wide range of firm reports but not ICS and find an overall higher disclosure quality with larger firms (measured by comparing FTSE 100, FTSE 250 and FTSE small cap firms), our
results stress the homogeneous applicability of ICS for both large and small firms.
While the quantity of disclosure differs significantly across regions, we find no significant
differences in the quality of disclosure provided. Interestingly, regions with most INCD are

ACCEPTED MANUSCRIPT
24

southern Europe and German-speaking countries. This contrasts with the findings of Chaminade et al. (2003), who argues that there are higher (perceived) proprietary costs in southern
countries and lower ones in Nordic countries. Differences in quantity do not translate into

RI
PT

higher INCD quality for southern and German-speaking countries, indicating homogeneous
disclosure quality across European regions. These findings are not consistent with the previous narrative findings on ICS (i.e. without conducting a quantitative content analysis) by

SC

Ordez de Pablos (2002, 2003), who claims, for example, that disclosure in Spain is, compared with Nordic countries, more qualitative, which would proportionally decrease INCD-

M
AN
U

QUAL.

Finally, both the quantity and quality of INCD disclosure differs across firms adopting different disclosure guidelines. In particular, the German Federal Ministry of Economics and
Technology (2008) guidelines lead to higher disclosure quantity and quality on INC than
those of the Putting IC into Practice project (Thorleifsdottir et al., 2006). This result seems

TE
D

to occur because of the high degree of detail with which the German guidelines address the
drivers of intellectual capital (including innovation) as well as the quantification and graphical depiction of the quantity, quality and systematic management of these drivers

EP

(Bornemann et al., 2007). In light of the different attempts by multinational projects to homogenise the use of ICS (e.g. European Commission, 2006), our results provide evidence for

AC
C

the still low success of these efforts.

We acknowledge a number of limitations inherent in our study. The generalisability of this
research is limited to INCD decisions in ICS in European for-profit firms. However, our findings suggest avenues for further exploration by researchers on disclosure in ICS and for other
forms of intellectual capital. Further limitations relate to the methods employed in this study.
First, the indexes used, as well as the proxies for our independent variables, might be criticised (e.g. concerning the weightings of the INCDQUAL index or the use of staff headcount

ACCEPTED MANUSCRIPT
25

as a proxy for firm size). Nevertheless, we preferred to rely on indexes found to be helpful in
prior disclosure studies and which had thus already been validated instead of developing our
own, new disclosure index. At the same time, we tested whether our results were susceptible

RI
PT

to change if we measured INCDQUAL in a different way. We standardised INCDQUAL

with the number of pages, words and characters per ICS. None of these different measures of
INCDQUAL changed our results except for the disclosure guidelines adopted. These do not

SC

show significant differences for disclosure quality when using all alternative measures. This
is attributable to the significant differences in length across ICS based on different guidelines.

M
AN
U

For the proxies for the independent variables, the use of employee headcount as a proxy
for firm size, for example, may be criticised. In section 5.2, however, we explained our methodological choices from a content and from a data availability perspective. Future studies
with better data availability are called for to test for the robustness of our results by using
other proxies for independent variables.

TE
D

Second, as is true for practically all content analyses due to their subjectivity, caution
needs to be exercised in the interpretation and generalisation of our results. Nonetheless, we
made intensive efforts to maximise the reliability and validity of the content analysis carried

EP

out closely following the guidelines of Krippendorff (2004) and Weber (1990).
Notwithstanding these limitations, our paper contributes to the research in several ways.

AC
C

First, we show that ICS are used to disclose information about INC with an average frequency of 29.16 items per ICS. This illustrates that firms publishing ICS are able to achieve
one of the major goals identified by Mouritsen et al. (2001c), namely to highlight innovation
efforts. Similar to other disclosures, the type of disclosure remains mostly qualitative and
historical, thus not tapping into the full potential of ICS. Second, we find that INCD in ICS is
carried out by using items relatively equally distributed across internal, external and human
capital categories. Therefore, we show how INC is interrelated with different types of intel-

ACCEPTED MANUSCRIPT
26

lectual capital and provide quantitative evidence for the argument that innovations can be
generated by enhancing the intellectual capital categories available to the firm. Third, we
contribute to the research on disclosure by identifying the drivers of INCD quantity as indus-

RI
PT

try, size, region of registered office and disclosure guidelines. In particular, we show that the
quality of the information disclosed is homogeneous across firm sizes and regions. This demonstrates that ICS are a disclosure instrument that is applicable across different types of firms

SC

(industry, size, region) with a similar disclosure quality.

Our findings also have implications for policy and practice. First, small firms especially

M
AN
U

may identify ICS as instruments adequate for meeting the demand for INCD, as can be seen
from the overall use of INCD items in the sample ICS. Second, policymakers may achieve
insights into current disclosure practices on INC and find ways to further improve policymaking. For example, policymakers in Nordic countries could recognise in their data the necessity to increase the disclosure of INC in order to increase the perception of their economies

TE
D

innovativeness. Finally, guideline-setters may recognise that further efforts are needed in
order to satisfy the call for homogenisation in disclosure guidelines and thus in disclosure

Notes:
1

EP

behaviour.

This adjustment resulted from a discursive process between one of the authors and two other experienced

AC
C

content analysts following Beattie et al. (2007). The adjustment was necessary as the category systems previously used in the literature addressed overall intellectual capital and not specifically INC.
2

Additionally, we searched the internet for additional ICS not included in publicly available lists of firms
involved as users or participants in projects developing disclosure guidelines, but this did not lead to an increase in the available ICS.

Data collection led to 52 ICS. The analysis of INCDQUANT led to the exclusion of one ICS of a firm disclosing more than 8,000 INC items. The firm was a hospital corporation that disclosed in a 152-page report
research project data on each of its 117 hospitals. Its inclusion would have distorted our results.

Our sample identification is validated by the RICARDIS report, which found no evidence for the current
availability of ICS in Italy, France, the Netherlands, and Eastern European countries (European Commission,
2006). Furthermore, despite the attention drawn by several institutions to intangible assets and their disclo-

ACCEPTED MANUSCRIPT
27

sure (ACCA, 2006; ICAEW, 2005), we found no evidence of the use of ICS disclosure among British firms
(see also the findings by Striukova et al., 2008). Potential explanations for missing ICS disclosure in UK
may arise from the common law tradition (compared to the code law tradition of other countries), leading to
a more comprehensive disclosure in compulsory financial reports than in other regions and to a stronger fo-

RI
PT

cus on shareholders than on stakeholders in reporting in Britain when compared to other countries (see La
Porta, Lopez-de-Silanes, Shleifer, 2008).
5

T-test and ANOVA require a normal distribution of the dependent variables. The one-sample Kolmogorov
Smirnov test showed that both INCDQUANT and INCDQUAL follow a normal distribution.

Since the firms in our sample have very different underlying reporting standards, e.g. owing to different
requirements from diverse local GAAPs or for listed and non-listed firms, this hampers accounting data

SC

availability; thus, firm-specific R&D expenses and R&D intensity could not be measured. Therefore, we
used industry-specific R&D intensity as a proxy for firm-specific R&D intensity. Industry-specific R&D intensity was taken from the Community Innovation Survey 2006 (European Commission, 2008a) and meas-

M
AN
U

ured as total innovation expenditure as a percentage of total turnover. Firms in industries with a R&D intensity higher than the industry average were considered to have a high R&D intensity, while firms in industries
with a lower R&D intensity were considered to have a low R&D intensity.

An increase in the sample size increases the statistical power and significance of the presented results. Owing to our limited sample size of n=51, we consider the results at a level of p<0.1 to be significant (Schwab,
2005). In the disclosure research field, reporting p-levels of 0.01, 0.05 and 0.1 can be often found in high-

EP

TE
D

ranked journals (e.g. the studies of Boesso et al., 2007; Depoers, 2000; Toms, 2002).

AC
C

ACCEPTED MANUSCRIPT
28

References
Abdolmohammadi, M. (2005). Intellectual capital disclosure and market capitalization. Journal of Intellectual Capital, 6(3), 397416.

RI
PT

Abeysekera, I., Guthrie, J. (2004a). Human capital reporting in a developing nation. The British Accounting Review, 36(3), 251268.

Abeysekera, I., Guthrie, J. (2004b). How is intellectual capital being reported in a developing

SC

nation?, In Hopper, T., Hoque, Z. (Ed.), Research in Accounting in Emerging Economies,

Supplement 2: Accounting and Accountability in Emerging and Transition Economies (pp.

M
AN
U

149169). Amsterdam et al.: Elsevier.

ACCA (2006). SME intangible assets, available on the internet at

http://www.accaglobal.com/pubs/general/activities/research/research_archive/rr-093-001.pdf.
Accessed 05.07.2010.

Adams, C. A., Hill, W.-Y., Roberts, C. B. (1998). Corporate social Reporting practices in

TE
D

Western Europe: legitimating corporate behaviour?. British Accounting Review, 30(1), 1-21.
AICPA (1994). Improving business reporting A customer focus: Meeting information needs
of investors and creditors. Jersey City: AICPA.

EP

April, K. A., Bosma, P., Deglon, D. (2003). Intellectual capital measurement and reporting:

180.

AC
C

Establishing a practice in South African mining. Journal of Intellectual Capital, 4(2), 165

Austrian Ministry of Science and Research (2002). University Act, available on the internet
at http://www.bmwf.gv.at/fileadmin/user_upload/wissenschaft/recht/englisch/E_UG.pdf. Accessed 04.07.2010.

Austrian Research Centers (2000). Intellectual Capital Report 1999, research report of the
Austrian Research Centers. Seibersdorf: Austrian Research Centers.

ACCEPTED MANUSCRIPT
29

Backhuijs, J. B., Holterman, W. G. M., Oudman, R. S., Overgoor, R. P. M., Zijlstra, S. M.

(1999). Reporting on intangible assets, final report for the benefit of the Dutch Ministry of
Economic Affairs and the intangible assets pilot project sounding board group. Amsterdam:

RI
PT

Dutch Ministry of Economic Affairs.

Beattie, V., Thomson, S. J. (2007). Lifting the lid on the use of content analysis to investigate
intellectual capital disclosure. Accounting Forum, 31(2), 129163.

SC

Beretta, S., Bozzolan, S. (2008). Quality versus quantity: The case of forward-looking disclosure. Journal of Accounting, Auditing & Finance, 23(3), 333375.

M
AN
U

Boesso, G., Kumar, K. (2007). Drivers of corporate voluntary disclosure A framework form
Italy and the United States. Accounting, Auditing & Accountability Journal, 20(2), 269296.
Bontis, N. (2003). Intellectual capital disclosure in Canadian corporations. Journal of Human
Resource Costing and Accounting, 7(1/2), 920.

Bornemann, M., Alwert, K. (2007). The German guideline for intellectual capital reporting:

TE
D

method and experiences. Journal of Intellectual Capital, 8(4), 563576.

Botosan, C. A. (1997). Disclosure level and the cost of equity capital. The Accounting Review, 72(3), 323349.

EP

Bowen, F. E., Rostami, M., Steel, P. (2010). Timing is everything: A meta-analysis of the
relationships between organizational performance and innovation. Journal of Business Re-

AC
C

search, 63(11), 11791185.

Bozzolan, S., Favotto, F., Ricceri, F. (2003). Italian annual intellectual capital disclosure
An empirical analysis. Journal of Intellectual Capital, 4(4), 543558.
Bozzolan, S., OReagan, P., Ricceri, F. (2006). Intellectual capital disclosure (ICD) in listed
companies A comparison of practice in Italy and the UK. Journal of Human Resource Costing and Accounting, 10(2), 92113.

ACCEPTED MANUSCRIPT
30

Brennan, N. (2001). Reporting intellectual capital in annual reports: Evidence from Ireland.
Accounting, Auditing & Accountability Journal, 14(4), 423436.
Campbell, D., Abdul Rahman, M. R. (2010). A longitudinal examination of intellectual capi-

RI
PT

tal reporting in Marks & Spencer annual reports, 19782008. The British Accounting Review,
42(1), 5670.

Campbell, D., Beck, A. C., Shrives, P. (2005). A note on comparative language interrogation

SC

for content analysis: The example of English vs. German. The British Accounting Review,
37(3), 339350.

M
AN
U

Chaminade, C., Johanson, U. (2003). Can guidelines for intellectual capital management and
reporting be considered without addressing cultural differences? Journal of Intellectual Capital, 4(4), 528542.

Cho, C. H., Patten, D. M. (2007). The role of environmental disclosures as tools of legitimacy: A research note. Accounting, Organizations and Society, 32(7-8), 639-647.

TE
D

Clarkson, M. (1995). A stakeholder framework for analysing and evaluating corporate social
performance. Academy of Management Review, 20(1), 92117.
Clarkson, P. M., Li, Y., Richardson, G. D., Vasvari, F. P. (2008). Revisiting the relation be-

EP

tween environmental performance and environmental disclosure: An empirical analysis. Accounting, Organizations and Society, 33(4-5), 303-327.

AC
C

Danish Ministry of Science Technology and Innovation (2003). Intellectual Capital Statements The new guideline. Copenhagen: Danish Ministry of Science, Technology, and Innovation.

Depoers, F. (2000). A cost-benefit study of voluntary disclosure Some empirical evidence

from French listed companies. The European Accounting Review, 9(2), 245263.
Edvinsson, L., Malone, M. S. (1997). Intellectual Capital Realizing your companys true
value by finding its hidden brainpower. New York: HarperCollins.

ACCEPTED MANUSCRIPT
31

Euroforum (1998). El Modelo Intelect. Madrid: Instituto Universitario Euroforum Escorial.

European Commission (2005). The new SME definition - User guide and model declaration,
available on the internet at

RI
PT

http://ec.europa.eu/enterprise/enterprise_policy/sme_definition/sme_user_guide.pdf. Accessed 04.07.2010.

European Commission (2006). RICARDIS: Reporting Intellectual Capital to Augment Re-

SC

search, Development and Innovation in SMEs.

European Commission (2008a). Community Innovation Survey, available on the internet at

ase. Accessed 18.03.2010.

M
AN
U

http://epp.eurostat.ec.europa.eu/portal/page/portal/science_technology_innovation/data/datab

European Commission (2008b). NACE Rev. 2 Statistical classification of economic activities in the European Community, available on the internet at

http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-RA-07-015/EN/KS-RA-07-015-

TE
D

EN.PDF. Accessed 18.03.2010.

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

Times Prentice Hall.

EP

German Federal Ministry of Economics and Technology (2008). Wissensbilanz Made in

Germany Leitfaden. Leitfaden 2.0 zur Erstellung einer Wissensbilanz. Berlin: German

AC
C

Federal Ministry of Economics and Technology.

Guthrie, J., Petty, R. (2000). Intellectual capital Australian annual reporting practices.
Journal of Intellectual Capital, 1(3), 241251.
Guthrie, J., Petty, R., Ferrier, F., Wells, R. (1999). There is no accounting for intellectual
capital in Australia: A review of annual reporting practices and the internal measurement of
intangibles within Australian Organizations, OECD Conference Report. Amsterdam.

ACCEPTED MANUSCRIPT
32

Guthrie, J., Petty, R., Yongvanich, K., Ricceri, F. (2004). Using content analysis as a research
method to inquire into intellectual capital reporting. Journal of Intellectual Capital, 5(2),
282-293.

RI
PT

Guthrie, J., Petty, R., Ricceri, F. (2006). The voluntary reporting of intellectual capital
Comparing evidence from Hong Kong and Australia. Journal of Intellectual Capital, 7(2),
254271.

SC

Guthrie, J., Ricceri, F., Dumay, J. (2012). Reflections and projections: A decade of intellectual capital accounting research. British Accounting Review, 44(1), 68-82.

M
AN
U

Healy, P. M., Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the
capital markets: A review of the empirical disclosure literature. Journal of Accounting &
Economics, 31(13), 405440.

Hooks, J., Coy, D., Davey, H. (2002). The information gap in annual reports. Accounting,
Auditing & Accountability Journal, 15(4), 501522.

TE
D

IASB (2007). Intangible Assets, available on the internet at

http://www.iasb.org/NR/rdonlyres/6A1793C3-81F3-4A62-B0D4EC9BF7A5772F/0/IntangibleassetsprojectupdateDec07.pdf. Accessed 08.04.2008.

EP

ICAEW (2005). UK corporate reporting of intellectual capital, available on the internet at

http://www.icaew.com/index.cfm/route/122595/icaew_ga/en/Technical_and_Business_Topic

AC
C

s/Thought_leadership/UK_corporate_reporting_of_intellectual_capital. Accessed 05.07.2010.

Japanese Ministry of Economy Trade and Industry (2005). Guidelines for Disclosure of Intellectual Assets Based Management, available on the internet at
http://www.meti.go.jp/english/information/downloadfiles/cIPP0403e.pdf. Accessed
04.07.2010.
Kaplan, R. S., Norton, D. P. (1996). Linking the balanced scorecard to strategy. California
Management Review, 39(1), 5379.

ACCEPTED MANUSCRIPT
33

Krippendorff, K. (2004). Content analysis An introduction to its methodology. Thousand

Oaks: Sage.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. (2008). The economic consequences of legal

RI
PT

origins. Journal of Economic Literature, 46(2): 285-332.

Lang, M., Lundholm, R. (1993). Cross-sectional determinants of analysts ratings of corporate
disclosure. Journal of Accounting Research, 31(2), 246271.

SC

Leuz, C., Wysocki, P. D. (2008). Economic consequences of financial reporting and disclosure regulation: A review and suggestions for future research (Working Paper), available on

M
AN
U

the internet at http://www.mit.edu/wysockip/www/papers/LW2008.pdf, Accessed

30.12.2012.

Lev, B. (2001). Intangibles Management, measurement, and reporting. Washington:

Brookings.

Li, J., Mangena, M., Pike, R. (2012). The effect of audit committee characteristics on intel-

TE
D

lectual capital disclosure. British Accounting Review, 44(2), 98-110.

Marston, C. L., Shrives, P. J. (1991). The use of disclosure indices in accounting research A
review article. The British Accounting Review, 23(3), 195210.

Report).

EP

Meritum (2000). Guidelines for managing and reporting on intangibles (Intellectual Capital

AC
C

Mouritsen, J., Bukh, P., Bang, H. (2008). Understanding intellectual capital in an innovative
medium-sized firm: The case of Maxon Telecom. Australian Accounting Review, 15(36), 30
39.

Mouritsen, J., Larsen, H. T., Bukh, P. N. (2001a). Valuing the future Intellectual capital
supplements at Skandia. Accounting, Auditing & Accountability Journal, 14(4), 399422.

ACCEPTED MANUSCRIPT
34

Mouritsen, J., Johansen, M., Larsen, H., Bukh, P. (2001b). Reading an intellectual capital
statement: Describing and prescribing knowledge management strategies. Journal of Intellectual Capital, 2(4), 359383.

RI
PT

Mouritsen, J., Larsen, H. T., Bukh, P. N. D. (2001c). Intellectual capital and the "capable
firm": Narrating, visualising and numbering for managing knowledge. Accounting, Organizations and Society, 26(78), 735762.

the Nordika Projekt, available on the internet at

SC

Nordic Industrial Fund (2001). Intellectual capital - Managing and reporting A report from

M
AN
U

http://www.nordicinnovation.net/_img/Nordika_report_-_final.pdf. Accessed 12.12.2008.

Nordic Industrial Fund (2003). How to develop and monitor your company's intellectual capital: Tools and actions for the competency-based organisation the Frame Project, available
on the internet at http://www.nordicinnovation.net/_img/IC_rapport_engelsk.pdf. Accessed
15.12.2008.

TE
D

OECD (2010). The OECD Innovation Strategy Getting a head start on tomorrow, available
on the internet at http://www.oecd-ilibrary.org/science-and-technology/the-oecd-innovationstrategy_9789264083479-en. Accessed 28.05.2010.

EP

Olsson, B. (2001). Annual reporting practices: Information about human resources in corporate annual reports in major Swedish companies. Journal of Human Resource Costing and

AC
C

Accounting, 6(1), 3952.

Ordez de Pablos, P. (2002). Evidence of intellectual capital measurement form Asia,

Europe and the Middle East. Journal of Intellectual Capital, 3(3), 287302.
Ordez de Pablos, P. (2003). Intellectual capital reporting in Spain A comparative view.
Journal of Intellectual Capital, 4(1), 6181.
Ordez de Pablos, P. (2005). Intellectual capital reports in India Lessons from a case
study. Journal of Intellectual Capital, 6(1), 141149.

ACCEPTED MANUSCRIPT
35

Parsons, T. (1960). Structure and process in modern societies. Glecoe: Free Press.
Patten, D. M. (1991). Exposure, legitimacy, and social disclosure. Journal of Accounting and
Public Policy, 10(4), 297308.

RI
PT

Ricceri, F. (2008). Intellectual capital and knowledge management - Strategic management

of knowledge resources. Abingdon: Routledge.

Schmalenbach working group Accounting and Reporting of Intangible Assets (2005). Cor-

Business Review, Special Issue, 57(1), 65100.

SC

porate reporting on intangibles A proposal from a German background. Schmalenbach

M
AN
U

Schwab, D. P. (2005). Research methods for organizational studies. Mahwah, London: Lawrence Erlbaum.

Society for Knowledge Economics (2005). Australian guiding principles on extended performance management, available on the internet at

http://www.ske.org.au/download/Australian-Guiding-Principles.pdf. Accessed 04.07.2010.

TE
D

Striukova, L., Unerman, J., Guthrie, J. (2008). Corporate reporting of intellectual capital
Evidence from UK companies. The British Accounting Review, 40(4), 297313.
Subbarao, A. V., Zeghal, D. (1997). Human resources information disclosure in annual re-

2(2), 5373.

EP

ports An international comparison. Journal of Human Resource Costing and Accounting,

AC
C

Sveiby, K.-E. (1997). The new organizational wealth Managing and measuring knowledgebased assets. San Francisco: Berrett-Koehler.
Thorleifsdottir, A., Claessen, E. (2006). Putting intellectual capital into practice Nordic
harmonized knowledge indicators, available on the internet at
http://www.nordicinnovation.net/_img/putting_ip_into_practice.pdf. Accessed 15.12.2008.
Toms, J. S. (2002). Firm resources, quality signals and the determinants of corporate environmental reputation: Some UK evidence. The British Accounting Review, 34(3), 257282.

ACCEPTED MANUSCRIPT
36

Vandermaele, S. N., Vergauwen, P. G. M. C., Smits, A. J. (2005). Intellectual capital disclosure in The Netherlands, Sweden and the UK A longitudinal and comparative study. Journal of Intellectual Capital, 6(3), 417426.

RI
PT

Vergauwen, P. G. M. C., van Alem, F. J. C. (2005). Annual report IC disclosure in The Netherlands, France and Germany. Journal of Intellectual Capital, 6(1), 89104.

Vergauwen, P., Bollen, L., Oirbans, E. (2007). Intellectual capital disclosure and intangible

SC

value drivers: an empirical study. Management Decision, 45(7), 11631180.

Weber, M. (1978). Economy and society. Berkeley: University of California.

AC
C

EP

TE
D

M
AN
U

Weber, R. P. (1990). Basic content analysis. Newbury Park: Sage Publications.

ACCEPTED MANUSCRIPT

37

Denmark

Description of the storytelling in ICS

Ordez de
Pablos (2002)

13

Description of the archetypal role of Scandinavian

firms for firms worldwide when setting up ICS

Ordez de
Pablos (2003)
Ordez de
Pablos (2005)
Mouritsen et al.
(2008)

Austria, Denmark,
India, Israel,
Spain, Sweden
Spain

India

Denmark

RI
PT

17

M
AN
U

Mouritsen et al.
(2001c)

INC considered
Indicators in the renewal and development focus,
such as investment in product development or in
process improvement
Innovation efforts as part of ICS storytelling as an
instrument for the achievement of organisational
goals
Investment in product and process development,
number of new services/products, etc.

SC

Table 1
Previous descriptive studies of intellectual capital disclosure in ICS (listed in chronological order)
Study
n
Country
Overall results
Mouritsen et al. 1
Sweden
Description of the content and goals of intellectual
(2001a)
capital supplements at Skandia

AC
C

EP

TE
D

Describes the archetypal role of the Skandia Navigator for ICS in Spain
Describes the predominance of the narrative form in
Indian ICS and the absence of indicators
Description of the content of ICS at Maxon Telecom
and the relationship between different knowledge
resources and the organisational goals to develop
innovation

Investment in product and process development,

number of new services/products, etc.
No specific information provided on INC
Innovation is an organisational goal described by a
network of knowledge narrative, management challenges, initiatives and related indicators

ACCEPTED MANUSCRIPT

38

Semantic validity

Structural validity
Functional validity

Correlative validity

Predictive validity

the degree to which a sample of texts accurately

represents the population
the degree to which the analytical categories accurately describe the meanings and uses in the chosen
context
the degree to which the analytical construct models
the network of relations in the chosen context
the degree to which the analytical categories are
useful over time and in many empirical situations
the extent to which the results correlate with variables that measure the same construct (convergent
validity) and the extent to which correlation is absent with variables measuring different phenomena
(discriminant validity)
the degree to which anticipated observations actually occur

SC

high intraobserver agreement tested on a sample of items 12 months after first coding (Krippendorffs alpha=0.971, indicating high stability)
high intercoder reliability measured on a sample of items taken from the 51 ICS by investigator triangulation with three experienced content analysts (Krippendorffs alpha=0.917, indicating high reproducibility)
intensive coder training with experienced content analysts

literature review of content analyses employing the category system used; adjustment to the
INCD setting in a discursive process by three experienced content analysts and the formulation of coding rules
innovations are major drivers of economic growth, foster employment, address social challenges; thus, INCD is important to a wide range of stakeholders

M
AN
U

Empirical validity
Sampling validity

TE
D

Social validity

the correspondence between the definitions of the

categories and the constructs intended to be measured
the addressing of important social issues, contribution to public debates

EP

Validity
Face validity

the degree to which the analysis conforms to a predefined standard

AC
C

Accuracy

Procedure or measure taken in our study

RI
PT

Table 2
Reliability and validity of the content analysis procedure
Definition
Reliability
Stability
the degree to which the analysis is unchanged over
time
Reproducibility
the degree to which the analysis can be replicated by
different analysts working under varying conditions

an ICS population cannot be identified; our study identifies 51 online available ICS by forprofit firms based on five major European disclosure guidelines, thus exceeding the sample
size of previous studies of ICS (see Table 1)
development of a coding manual on the basis of a training sample of ICS by three experienced content analysts; rejection of computer-aided content analysis

INCDQUANT and INCDQUAL are constructs already validated by Botosan (1997) and
Boesso (2007) as instruments suitable for measuring the quantity and quality of disclosure
the category system of intellectual capital (internal, external, human capital) has been employed since Guthrie et al. (1999) by numerous studies for different countries, industries, and
firm sizes
we identify alternative measures of INCDQUAL by standardising the measure used in the
paper by the number of pages, the number of words, and the number of characters in ICS and
find that these alternative measures all correlate significantly (p<0.01) with our INCDQUAL
measure (Pearsons correlation coefficients range between 0.33 and 0.54), indicating convergent validity for our disclosure quality measure
could only be tested with further ICS that are currently unavailable; according to Weber
(1990), this is why predictive validity is seldom tested by content analysts

Note: The definitions reported in the table are taken from Krippendorff (2004) and Weber (1990).

ACCEPTED MANUSCRIPT

39

Wholesale,
retail and
other
services

Austrian
Research
Centers

Disclosure guidelines
Danish ICS German
Modelo
Guidelines Federal
Intelect
Ministry of
Economics
and Technology

RI
PT

Industries
ProfesElectricity Financial
sional,
and con- and insurtechnical struction ance
and health
activities
service

--

16

Medium-sized

--

17

Small

--

11

--

--

Micro

--

--

24

10

Nordic
Harmonized
Knowledge Indicators

16

--

11

--

17

15

--

--

--

15

--

--

--

--

51

25

18

51

M
AN
U

Large

SC

Information and
communication

TE
D

Table 3
Sample description
Size
Manufacturing

AC
C

EP

Note: The industry nomenclature is oriented on the NACE Rev. 2 classification (European Commission, 2008b). Manufacturing, information and communication as well as
professional, technical, and health services are considered to be high R&D intensity industries, based on the results by the Community Innovation Survey (European Commission, 2008a). Firm size is measured based on employee headcount following the thresholds proposed by European Commission (2005): large: 250 employees; medium-sized:
50 and < 250 employees; small: 10 and < 50 employees; micro: <10 employees. The median of headcounts across all 51 sample firms is 89 employees.

ACCEPTED MANUSCRIPT

40

where
INCDQUANTij=instances of the disclosure of INC in category i for firm j.

AC
C

EP

TE
D

Internal (structural)
539 (36.2 %)
External (relational) 331 (22.2 %)
Human (employee)
617 (41.6 %)
Disclosure in terms of type of disclosure
(no. of items and % of total items)
Type
Quantitative
379
(25.5 %)
Qualitative
1,108
(74.5 %)
Nature
Financial
38
(2.6 %)
Non-financial
1,449
(97.4 %)
Time-reference
Historical
1,313
(88.3 %)
Forward-looking
174
(11.7 %)
Notes: Items refer to the instances of the disclosure of INC. The distinction between high and low R&D intensity industries is informed
by the results of the Community Innovation Survey (European Commission, 2008a).

RI
PT

Low R&D
(n=9)
4.44
3.33
9.89

SC

High R&D
(n=42)
11.88
7.17
12.57

Table 5
Descriptive statistics for INCD indexes
Index
INCDQUANT
INCDQUAL
Mean
29.16
141.02
Median
21.00
107.00
S.D.
24.89
119.26
Kurtosis
2.34
2.23
Skewness
1.47
1.44
Min
1.00
5.00
Max
109.00
527.00
Notes:
The table reports on descriptive statistics for innovation disclosure quantity (INCDQUANT) and quality (INCDQUAL). The indexes are computed as follows.

M
AN
U

Table 4
Descriptive results of the content analysis
Total number of pages
1,183
Average pages per firm
23.20
Total number of disclosed items on INC
1,487
Average items per firm
29.16
Disclosure by capital category
Average number of INCD
(no. of items and % of total items)
by category and sector

where
Kj=number of total text units of the ICS of firm j;
typek j= 0 if the text unit k does not provide any item on INC; 1 if the text
unit provides a qualitative item; 2 if the text unit provides a quantitative
item;
naturek j= 0 if the text unit k does not provide any item on INC; 1 if the
text unit provides a financial item; 2 if the text unit provides a nonfinancial item;
outlookk j= 0 if the text unit k does not provide any item on INC; 1 if the
text unit provides an historical item; 2 if the text units provides a forward-looking item.

ACCEPTED MANUSCRIPT

41
Table 6
Bivariate results for INCDQUANT and INCDQUAL
Variable

Groups

Mean INCDQUANT

S.D. INCDQUANT

Statistics
INCDQUANT

Mean INCDQUAL

S.D.
INCDQUAL

H1

IND

IND_RD
IND_NON_RD

42
9

31.62
17.67

25.83
16.09

t-statistic 1.550
(0.064) *

153.12
84.55

123.90
76.56

t-statistic 1.589
(0.059) *

H2

SIZE

SMALL
LARGE

26
25

23.50
35.04

16.11
30.74

t-statistic 1.688
(0.098) *

114.11
169.00

77.99
147.29

t-statistic 1.672
(0.101) n.s.

H3

REGION

GER_SP
NORDIC
SOUTH_EUR

26
23
2

31.92
23.26
61.00

18.24
26.04
67.88

F-statistic 2.608
(0.084) *

154.19
113.17
290.00

87.13
124.73
335.17

Fw-statistic 0.848
(0.520) n.s.

H4

DISC_GUID

GFMEL
PIIP
OTHERS

25
18
8

31.92
17.22
47.38

18.62
13.44
44.87

Fw-statistic 5.416
(0.016) **

153.92
83.89
229.25

88.91
63.95
216.02

Fw -statistic 5.422
(0.016) **

M
AN
U

SC

RI
PT

Hyp.

Statistics
INCDQUAL

AC
C

EP

TE
D

Notes:
The table reports the relationship between independent variables and INCDQUANT as well as INCDQUAL (as defined in section 4.1) tested using t-tests (one-sided for
directional hypotheses on IND, two-sided for non-directional hypotheses on SIZE) in the case of two groups and ANOVA and F-tests in the case of more than two groups
(Fw-statistic=Welchs test, employed for non-homogeneous variances across groups according the Levene test); p-values in brackets; * p<0.1, ** p<0.05, *** p<0.01.
n.s.=not significant.
Independent variables are: IND (IND_RD include firms belonging to a high R&D intensity industry according the Community Innovation Survey, IND_NON_RD include
firms belonging to a high R&D intensity industry according the Community Innovation Survey), SIZE (number of employees in the disclosure year, SMALL and LARGE
based on the median split of the sample, median=89), REGION (region of registered office, GER_SP=German speaking countries, i.e. Germany and Austria; NORDIC=
Nordic countries, i.e. Denmark, Finland, Iceland, Norway and Sweden; SOUTH_EUR: southern European, i.e. Spain), DISC_GUID (disclosure guideline: GFMEL=German
Federal Ministry of Economics and Technology; PIIP= Nordic Harmonized Knowledge Indicators: Putting IC into Practice; OTHERS: Danish ICS pilot Project; Austrian
Research Centers Model; Modelo Intelect Spain).

ACCEPTED MANUSCRIPT

42
Appendix 1
European intellectual capital disclosure guidelines (in chronological order)

Intellectual capital classification

Financial focus; Renewal and development focus; Customer focus; Human
focus; Process focus

Intangible Assets Monitor (Sveiby, 1997)

External capital; Internal capital; Competences

Modelo Intelect (Euroforum, 1998)

Human capital; Structural capital; Relationship capital

Dutch Ministry of Economics (Backhuijs, Holterman, Oudman, Overgoor and Zijlstra, 1999)

Human capital; Customer capital; Process capital; Innovation capital

Austrian Research Centers (2000)

Human capital; Structural capital; Relationship capital; Core research projects; Financial and intangible outcome

M
AN
U

SC

RI
PT

Skandia Navigator (Edvinsson et al., 1997)

Meritum (2000)

Human capital; Structural capital; Relational capital

Danish Ministry of Science, Technology and Innovation (2003)

Customers/users; Employees; Processes; Technology

Human capital; Structural capital; Relational capital

Schmalenbach working group Accounting and Reporting of Intangible Assets (2005)

Innovation capital; Human capital; Customer capital; Supplier capital; Investor capital; Process capital; Location capital

German Federal Ministry of Economics and Technology (2008)

Human capital; Structural capital; Relationship capital

TE
D

Nordika (Nordic Industrial Fund, 2001); Frame (Nordic Industrial Fund, 2003); Nordic Harmonized Knowledge Indicators; and the follow-up project Putting IC into Practice
(Thorleifsdottir et al., 2006)

AC
C

EP

Note: None of the guidelines above is restricted or specific for one country only. Nevertheless, national institutions like ministries or state-owned research centres started initiatives to develop, sponsor and spread guidelines and the use of ICS. Thus these guidelines have the name of national institutions. However, all guidelines above from their
content and structure can be used worldwide. While the other guidelines publish their own recommendations, Nordika, Frame, Nordic Harmonized Knowledge Indicators and
the follow-up project Putting IC into Practice attempt to summarise pre-existing intellectual capital disclosure guidelines. The guidelines listed were detected on the basis of
an extensive literature review and internet research for European guidelines aimed at encouraging the delivery of a consolidated numeric and narrative report for external disclosure purposes by profit-oriented firms. Guidelines issued outside Europe (e.g. those by the Japanese Ministry of Economy Trade and Industry (2005) or by the Australian
Society for Knowledge Economics (2005) were excluded as they are not expected to affect intellectual capital disclosure in European firms. Since the research design is restricted for the sake of better comparability to only for-profit firms, guidelines that address only disclosure by non-profit entities (e.g. those by the Austrian Ministry of Science
and Research, 2002) were also excluded. Approaches aimed at internal reporting but not designed for external disclosure (e.g. the Balanced Scorecard by Kaplan and Norton,
1996) were not considered.

ACCEPTED MANUSCRIPT

43

2.2 Distribution channels

2.3 Firm reputation
2.4 Business collaborations

RI
PT

Legal rights, i.e. major patents and copyrights, that protect the firms innovations
General beliefs about the role of innovations for the firm, usually formulated into vision or mission statements
Set of common behaviours and attitudes adopted by a firm in order to encourage innovation
Functional structure, procedures and organisational practices used by the firm for innovation efforts
Network of all communication channels used within a firm to facilitate innovation
Resource commitment of the firm for innovation efforts

SC

External (relational) capital category

2.1 Customers and market

Definition / key concepts

M
AN
U

Appendix 2
Categories and definitions of INCD
Category and subcategory name
Internal (structural) capital category
1.1 Intellectual property
1.2 Management philosophy
1.3 Corporate culture
1.4 Management processes
1.5 Information and networking system
1.6 Infrastructure

Customer involvement and customer reactions to innovations, innovations developed with customers, as well as the financial
impact of innovations in terms of income, returns, market share and savings
Process involving the marketing and distribution of the firms innovations
The firms credit due to innovation efforts
Relationships between the firm and individuals or organisations that are characterised by mutual cooperation on innovation
efforts

EP

Workforce involved in innovation efforts

Education level considered to be important in order to generate successful innovations
Actions taken to impart the skills considered to be important for successful innovations
Competencies, capabilities and experience considered to be important for successful innovations
The results of the innovation process generated by employees in terms of new ideas, new products, new services, new processes
and related scientific publications

AC
C

3.1 Employees
3.2 Education
3.3 Training
3.4 Work-related knowledge
3.5 Innovativeness of employees

TE
D

Human (employee) capital category