The Firm Characteristics Associated with the Voluntary Issuance of the New Style Audit Report in the Netherlands An Empirical Investigation Fabian Meijs September 2014 This thesis examines the firm characteristics associated with the voluntary disclosure of the new style audit report over 2013 in the Netherlands. This country was selected, as the new style audit report is already mandatory over 2013 in the UK, while in other countries firms do not yet engage in publishing their audit report ‘new style’. The
140
Embed
The Firm Characteristics Associated with the Voluntary ... Web viewThe Firm Characteristics Associated with the ... What are the firm characteristics associated with the voluntary
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The Firm Characteristics Associated with the Voluntary Issuance of the New Style Audit Report in the Netherlands
An Empirical Investigation
Fabian Meijs
September 2014
This thesis examines the firm characteristics associated with the voluntary disclosure of the new style
audit report over 2013 in the Netherlands. This country was selected, as the new style audit report is
already mandatory over 2013 in the UK, while in other countries firms do not yet engage in publishing
their audit report ‘new style’. The research sample consists of 111 Dutch firms, of which 30 voluntary
issued a new style audit report. The research, which is conducted by means of a logistic regression
analysis, contains a wide range of independent variables. These variables were selected based on
theories applicable to voluntary disclosure and a review of previous literature on voluntary disclosure
determinants. It was found that listing age, ownership dispersion and having a Big Four auditor are
significantly positively associated with the voluntary disclosure of the new style audit report. On the
other hand, competitive pressure and being cross-listed exhibit a significantly negative relation with
the issuance of the new style audit report. Industry-related significantly positive associations with
disclosure of the new style audit report were found for firms in the sectors Industrials and Consumer
Services.
Table of contents
Page
Table of contents 2
1. Introduction to the thesis 4
1.1 Introduction to the subject 4
1.2 Research question 5
1.3 Relevance 6
1.4 Methodology 7
1.5 Structure 8
2. Institutional background 9
2.1 The new style audit report 9
2.1.1 ISA 701 – Communicating key audit matters 10
2.1.2 Materiality and an overview of the scope of the audit 11
2.1.3 ISA 570 – Going concern 12
2.2 Voluntary disclosure 13
2.2.1 Agency theory 14
2.2.2 Positive accounting theory 15
2.2.3 Signaling theory 18
2.2.4 Capital need theory 19
2.2.5 Legitimacy theory 19
2.2.6 Institutional theory 19
2.2.7 Litigation cost theory 20
2.3 Measuring voluntary disclosure 21
2.4 Content analysis 22
2.5 Research models in voluntary disclosure literature 23
on the issues to be covered. This would lead to more professional skepticism, hence resulting in a
higher audit quality. Fourth, communication between auditor and management might improve, as
there will be more dialogue about the reported key audit matters.
The most important mandatory additions to the new style audit report in the UK are paragraphs
about the key audit matters/areas of particular focus, materiality issues and an overview of the
scope of the audit, and the firm’s going concern. The proposed IAASB requirements do not
9
include the paragraph about materiality and the scope of the audit (PwC, 2014b). As a result,
some voluntary issuers in the Netherlands do not report on all three subjects, as there are some
firms that do not include a paragraph about the materiality and the scope of the audit. However,
because many Dutch firms include all three new additions in their new style audit report, each
new style audit report regulation is discussed in detail.
2.1.1 ISA 701 – Communicating key audit matters
According to the definition provided by the IAASB (2013a), key audit matters are “those matters
that, in the auditor’s professional judgment, were of most significance in the audit of the
financial statements of the current period.”
The IAASB (2013b) mentions three circumstances under which the auditor should report a key
audit matter:
“1) Areas identified as significant risks in accordance with ISA 315 (assessing risk of material
misstatements) or involving significant auditor judgment.
2) Areas in which the auditor encountered significant difficulty during the audit, including with
respect to obtaining sufficient appropriate audit evidence.
3) Circumstances that required significant modification of the auditor’s planned approach to the
audit, including as a result of the identification of a significant deficiency in internal control.”
(IAASB, 2013b)
(Cursive text added)
In a study of PwC (2014a), it was found that the most recurring key audit matters for Dutch firms
are the acquisition or sale of firm activities, tax issues and the valuation of goodwill. Other
examples of key audit matters are the valuation of real estate, derivatives, pensions and the risk
of fraud. On average, four key audit matters are included in an audit report.
2.1.2 Materiality and an overview of the scope of the audit
10
Materiality is a concept applied in the audit process of a firm. When assessing the fairness of a
firm’s financial statements, auditors collect evidence in order to verify that the line item amounts
in the financial statements correspond to reality. When an auditor finds out that a line item in the
financial statements significantly differs from reality, so that it influences the decision-making of
the user, the difference is considered material. For each line item, a so-called materiality
threshold is constructed. This materiality threshold is usually set at 3-5% of the total amount of
the line item. When the difference exceeds the predetermined materiality threshold, an adverse
opinion might be issued by the auditor, which implies that a firm did not fairly present its
financial statements.
The new style audit report contains information about a firm’s materiality level and the scope of
the audit. In the UK, paragraphs on these subjects are mandatory. However, since the new style
audit report is still voluntary outside the UK and this paragraph is not a proposed requirement by
the IAASB, about half of the Dutch firms publishing a new style audit report does not provide
information about the materiality levels and the scope of the audit (PwC, 2014a). According to
PwC (2014a), a reason for this abstention might be that auditors have trouble to define the
concept of materiality and report on the corresponding choices that come along with the
determination of the materiality level of an audit. Moreover, auditors might find it difficult to
explain their approach and discuss the structure of the audit, issues that are normally defined in
the new paragraph about the scope of the audit. In this research however, firms that have
published their audit report mostly in accordance with the new style audit report regulations but
only lack information on materiality levels and the scope of the audit àre considered new style
audit report adopters. This is done, because the remainder of their audit report is similar to the
new style audit report regulations proposed by the IAASB.
In the materiality paragraph of the new style audit report, first the concept of materiality is
explained. Second, the auditor reports on the materiality level applied during the audit. This level
is expressed both in percentages and in absolute amounts. By providing absolute numbers and
percentages, the degree of transparency about materiality has increased considerably.
11
The new style audit report paragraph on the overview of the scope of the audit mainly explains
how an audit is performed. Since there are no clear IAASB guidelines for the content of this
paragraph, the enclosed information differs between firms. Certain factors that are discussed in
this paragraph are the overall audit approach, the collection of evidence and the type of work
performed. Usually, the amount of reporting components for which a full scope audit was
required is given as well. Furthermore, it sometimes provides information on which parts of the
firm were exactly audited.
By inspecting new style audit reports, it could be determined that in contrast to the remainder of
the text, the last sentence of this paragraph is normally fixed: “This gave us sufficient evidence
we needed for our opinion on the financial statements as a whole.”
2.1.3 ISA 570 – Going concern
This ISA already existed, meaning that this is an amended proposal. The IAASB does not only
want auditors to evaluate a firm’s going concern, but also to report their findings in the audit
report. That is the core of this amended regulation.
A firm is considered a going concern when there is substantial doubt that the firm will remain in
business for the foreseeable future. Under IFRS, the foreseeable future is defined as the twelve
months following the audited reporting period (IFRS, 2013). So in case a firm’s reporting period
ends at year-end 2013, the auditor has to assess whether the firm is still in business as at year-end
2014. This period is the minimum required ‘look-forward’ period. However, auditors have the
flexibility to extend this period and assess a firm’s operating continuance over a longer time
frame (Laudato, 2012). According to the still existing (‘old’) audit report regulations, auditors
only have to report on going concern in case the audited firm is considered a going concern.
However, in line with the new audit report regulation (ISA 570), auditors have to include a
separate paragraph on a firm’s going concern, no matter their financial state.
In the new style audit report paragraph about going concern, the auditor first states that the firm’s
financial statements have been prepared using the going concern basis of accounting. Second, if
12
necessary, the auditor explains in which cases the going concern basis of accounting is
considered to be inappropriate. Third, there is a sentence about the appropriateness of the going
concern basis of accounting for this firm. In case there is no doubt about going concern, the
second part of the paragraph starts with the statement that the auditor has not discovered any
material uncertainties that “may cast significant doubt on the firm’s ability to continue as a going
concern” (p. 15). The paragraph concludes by explaining that no guarantees can be given with
respect to the firm’s going concern (IAASB, 2013b).
2.2 Voluntary disclosure
An important question to ask is how voluntary disclosure can be defined. Meek, Roberts & Gray
(1995) define it as “free choices on the part of company managements to provide accounting and
other information deemed relevant to the decision needs of users of their annual reports.” The
two most vital concepts of voluntary disclosure are embedded in this definition. First of all, it is a
free choice. Voluntary disclosure as defined by the FASB (2001) emphasizes this, as they
explain that a voluntary disclosure item is “not explicitly required” by a governing body.
Moreover, the objective of voluntary disclosure in the annual report is to provide stakeholders
reading this report with relevant information. With this additional information, users should be
better able to evaluate a firm. Based on that, they can make a more legitimate decision whether
or not, for example, to invest in the firm.
It is essential to distinguish between mandatory and voluntary disclosure. Mandatory disclosure
is required by a legislative body, while voluntarily disclosing information is a firm’s own choice.
However, as Hassan & Marston (2010) point out, voluntary disclosure can also be recommended
by an authoritative institute. This is also more or less the case for the new style audit report, as
the IAASB is planning on implementing this report on short notice. However, they still give
firms an opportunity to ‘voluntarily’ get acquainted with the new style audit report before it
becomes mandatory.
The 2014 issuance of a new style audit report by Dutch firms is an example of voluntary
disclosure. Since publishing a new style audit report is still non-mandatory in the Netherlands,
13
firms disclose new style audit reports strictly out of free will. Of course, there are certain
motivations behind the choice of voluntarily disclosing information to the public. These
voluntary disclosure theories are discussed in this subsection and linked to the new style audit
report.
Shehata (2014) identifies four theories that are applicable to voluntary disclosure: the agency
theory, signaling theory, capital need theory and legitimacy theory. Moreover, the positive
accounting theory, institutional theory and litigation cost theory are discussed.
2.2.1 Agency theory
This is the most often applied theory in articles about the determinants of voluntary disclosure.
The agency theory is a normative theory, meaning that the assumptions are value-based. The
agency theory is not about imposing, but about advising and prescribing. Normative theories
describe what ought to be. This is in contrast to positive theories, which are often based on
factual statements and deal with what is (Friedman, 1953).
The agency theory assumes that there is an imaginary contract between the managers of a firm
and its shareholders. The managers are labeled as the ‘agents’. They perform their tasks and run
the firm on behalf of the ‘principal’, the firm’s shareholders (Jensen & Meckling, 1976). So
according to this theory, managers are expected to act according to the will of the shareholders.
However, the agency theory points out that this is often not the case, as the two groups might
have different interests. This is called the ‘principal-agent problem’.
The principal-agent problem consists of two central problems: self-interest and information
asymmetry. When the interests of management are not aligned with the interests of the
shareholders, management might follow a strategy in order to maximize its own wealth at the
cost of the returns of the shareholders. Information asymmetry exists when managers have more
information than shareholders. This way, it is difficult for the shareholders to ensure that
management is acting in line with the shareholders’ interests.
14
In order to reduce management’s self-interest and information asymmetry, agency costs have to
be made. Agency costs are composed of monitoring costs, bonding costs and residual loss. The
former is the cost of shareholders monitoring managers’ activities. Bonding costs are paid by the
agents to ensure the principal that their actions will not negatively affect shareholder value.
Residual loss is the difference between management’s choices and the decisions that would have
led to shareholder welfare maximization (Shehata, 2014).
Moreover, there are costs of risks and rewards. Suppose that the interests of managers and
shareholders are perfectly aligned. In that case, managers make decisions with the purpose to
maximize shareholder value. Many firms apply an incentive system for managers. When
shareholder value rises, manager compensation increases. So when managers’ decisions turn out
to be profitable, there are costs of reward for the firm, as manager compensation increases.
However, there is always the risk that management makes the wrong choices. In that case,
shareholder value declines. This is called the cost of risks.
Voluntary disclosure is a way to reduce information asymmetry and subsequently reduce agency
costs, since it benefits shareholders with inside information.
2.2.2 Positive accounting theory
The positive accounting theory is based on the agency theory. The positive accounting theory
can be seen as a subset of the agency theory which is focused on accounting. This theory aims to
explain and predict why managers choose certain accounting policies, and why accounting
policies differ across firms. The original positive accounting theory as developed by Watts &
Zimmerman (1978) was founded upon three main attributes: the bonus plan hypothesis, the debt
covenant hypothesis and the political cost hypothesis.
The bonus plan hypothesis assumes that managers only take into account their own wealth when
making (accounting) decisions. To give an example, according to this hypothesis, managers
choose to disclose a new style audit report with the aim to enhance their reputation on the
15
managerial labor market. This serves their own interest, because chances increase to get a higher-
paid job.
The debt covenant hypothesis states that managers try to enhance profits in such a way, so that
the firm does not violate its debt covenants by exceeding predetermined borrowing limits. When
a firm exceeds its borrowing limits, it has to pay off its debt on an accelerated basis. This is
disadvantageous for the firm, because it might cause liquidity problems. Furthermore, in case of
a debt covenant violation, the bank may wish to receive a higher interest rate on the debt the next
time they do business with the firm. Translating this to the new style audit report, managers may
issue such a report with the objective to lower information asymmetry and improve transparency.
Higher transparency results in less uncertainty among the creditors of a firm, which is associated
with a lower cost of capital.
According to the political cost hypothesis, managers tend to report lower accounting profits as
they want to reduce their firm’s visibility to the government. Because governments are inclined
to place stricter regulations on highly profitable industries, firms in those industries try to
downplay their accounting profits. However, the political cost hypothesis is not only about
government pressure. Also environmental organizations might be troublesome for firms.
Especially firms with high emission and large public visibility are likely to be pinpointed by
environmental organizations. Take Royall Dutch Shell as an example. Their oil winning
activities in Nigeria were subject to large-scale protests of environmental (and non-
environmental) organizations. Understating accounting profits is a way to reduce public
visibility, and, consequently, attention of environmental organizations. Converting the political
cost hypothesis to voluntary disclosure practices, firms might report more on environmental
issues. Openly providing information on, for example, CO2 emissions, diminishes the chance
that the government investigates a firm’s environmental practices. Because of that, the
government is less likely to impose stricter environmental regulations. Environmental
organizations are probably less likely to attack a firm, as they have more information about a
firm’s environmental practices. Moreover, when disclosing environmental information, firms
often provide solutions to reduce their emission. This further reduces the chance that
environmental organizations revolt against a firm.
16
Related to the positive accounting theory, Graham, Harvey & Rajgopal (2005) examined the
reasons managers engage in voluntary disclosure based on a survey among more than 300
executives. In their theoretical background, they provide five arguments in favor of voluntary
reporting: information asymmetry, increased analyst coverage, stock price motivations, stock
compensation and management talent signaling. Information asymmetry was already discussed
as part of the agency theory, while management talent signaling can be linked to the bonus plan
hypothesis, as this motivation only has the objective to increase managerial wealth. The
management talent signaling principle states that managers would report voluntarily in order to
signal their character to the managerial labor market.
According to Graham et al. (2005), another argument to choose for voluntary disclosure is to
attain more analyst coverage. The reasoning behind this thought is that more firm-specific
information becomes available as a consequence of voluntary reporting. This makes it more
attractive for analysts to follow a firm, as it becomes easier to make predictions about a firm’s
future performance. Voluntary disclosure can also be explained by stock price motivations. It is
argued by the authors that voluntary disclosure can be used to increase share price, and, if
necessary, to distract investors’ attention from poor financial performance. The stock
compensation theory assumes that managers want to reduce contracting costs by providing
voluntary information. Related to stock compensation, managers would be concerned that new
employees might ask a risk premium in case they feel that managers have an information
advantage. It is argued that voluntary disclosure can mitigate this problem, as the information
advantage might disappear (Graham et al., 2005).
In conducting their research, the authors found three reasons that were most significant when
asking the question why managers choose to report information voluntarily. Voluntarily
communicating information namely “promotes a reputation for transparent/accurate reporting,
reduces the ‘information risk’ that investors assign to the stock, and provides important
information to investors that is not included in mandatory financial disclosures” (Graham et al.,
2005).
17
2.2.3 Signaling theory
This theory assumes that firms voluntarily disclose information to the outside world in order to
show that they are better than their competitors. They want to signal their superiority to the
market by disclosing additional, positive information about the firm. These disclosures make it
easier for the firms to attract capital. Moreover, it is good for the firm’s reputation (Shehata,
2014; Verrecchia, 1983).
As for how the signaling theory can explain the voluntary disclosure of the new style audit
report, there seems to be no real association regarding the content of the new style audit report.
The new style audit report only includes objective, neutral information. It does not contain any
positive information which is favorable to the firm. The paragraphs about the areas of focus and
materiality are included to objectively inform users of the annual report about the audit. Only the
extra paragraph about going concern, which (almost always) concludes that a firm’s use of the
going concern basis of accounting is appropriate, might be favorable to the firm.
However, the choice to voluntarily issue a new style audit report signals to the market that a firm
is innovative in its reporting choices. So the signaling theory can be used in explaining the mere
fact that a firm issues a new style audit report, but there seems to be no link between the
signaling theory and the content of the new style audit report.
2.2.4 Capital need theory
The capital need theory argues that firms voluntarily disclose information with the objective to
attract more capital at a lower cost. This theory resembles the debt covenant hypothesis. It is
generally believed and has often been showed that more (voluntary) firm disclosures lead to a
lower cost of capital (Diamond & Verrecchia, 1991; Cheynel, 2013; Petrova, Georgakopoulos,
Sotiropoulos & Vasileiou, 2012).
18
The FASB (2001) explains that a firm’s cost of capital includes a premium for investor
uncertainty. It is motivated that voluntary disclosure results in a decrease in investor uncertainty.
Hence, the premium on the cost of capital is lowered, which also decreases the cost of capital in
general.
2.2.5 Legitimacy theory
According to the legitimacy theory, the voluntary reporting decisions of a firm depend on
management’s views on what society considers to be appropriate for the firm. The connection
between firm and society relies on the notion of a social contract. This is a fictitious contract
which consists of the community expectations about the behavior of a firm. The legitimacy
theory believes that in case a firm diverges from its social contract, it loses its ground to continue
as an organization (Suchman, 1995).
Relating this information to the new style audit report, the legitimacy theory argues that a firm
takes the wishes of society into consideration when making the decision whether or not to issue
the new style audit report. So according to the legitimacy theory, the chance that firms engage in
voluntary disclosure is bigger when society expects them to do so.
2.2.6 Institutional theory
Another widespread theory applied in explaining voluntary disclosures is the institutional theory.
This theory illustrates how mechanisms for obtaining and preserving legitimacy become
institutionalized. The institutional theory consists of two main elements, of which one is
applicable to this study: isomorphism. The ‘institutional model of isomorphic change’ was
developed by DiMaggio and Powell (1983). The authors identified three forms of isomorphism:
coercive, normative and mimetic isomorphism. Coercive isomorphism implies that firms
voluntarily disclose information in response to stakeholder pressures of, for example,
governmental organizations. Normative isomorphism expects firms to voluntarily report in order
to follow group norms and values. Mimetic isomorphism implies that firms only engage in
19
voluntary disclosure in order to copy the disclosure practices of other firms in their industry.
They do this out of uncertainty about their own organization. Mimicking others is seen as a safe
solution.
The institutional theory is different from most other previously discussed theories in that it
provides more than one explanation for the choice of voluntary disclosure. This theory actually
consists of multiple layers. Coercive and mimetic isomorphism might be applicable to the new
style audit report. As stated earlier, the IAASB attempts to make the new style audit report
mandatory from 2015. Firms might feel pressure to practice with the new style audit report in
2014, before it becomes obligatory one year later. Mimetic isomorphism could also be associated
with voluntary disclosure. To examine this, it could be tested whether there are significant
differences in voluntary reporting among industries. As shown in the next section, this relation
has also been tested in previous studies.
2.2.7 Litigation cost theory
Palepu & Healy (2001) add another theory which can explain the choice for voluntary reporting.
However, there are two sides to this theory. On the one hand, it can be argued that managers
engage more in voluntary reporting, as an inadequate disclosure policy might result in
shareholder litigation. On the other hand, managers might also restrain from voluntary reporting
as a result of the threat of shareholder litigation. This is especially the case when providing
forward-looking information. Managers might face shareholder litigations when the projected
voluntary information turns out not to be correct.
For the voluntary disclosure of the new style audit report, the first reasoning seems to be more
appropriate, as there is no forward-looking information involved. However, when a firm’s
auditor does not report on the areas of risk, which is one of the additions to the new style audit
report and could be useful in predicting the continuance of a firm, shareholder litigations might
follow.
20
The theories discussed in this section will be used to motivate the hypotheses. The aim is to
explain the predicted direction of the hypotheses by looking at both the previous literature and
the relevant theories. Furthermore, when analyzing the regression results in Section 6, the
discussed theories can be utilized to explain the outcomes.
2.3 Measuring voluntary disclosure
Often, articles not only examine whether a voluntary disclosure item is included in the Annual
Report, but they also evaluate the quality of the information. Beattie, McInnes & Fearnley (2004)
list numerous ways to determine disclosure ratings. Five methods are discussed in detail:
subjective ratings, disclosure index studies, a thematic content analysis, readability studies and a
linguistic analysis.
When authors employ subjective ratings, it usually means that they use disclosure ratings
constructed by analysts. Analyst ratings measure the informativeness of the disclosed
information. Disclosure index studies is “a partial form of content analysis where the items to be
studied are specified ex ante” (Beattie et al., 2004). Most studies analyzing the determinants of
voluntary disclosure employ this technique, which is based on the analysis of the Annual Reports
of firms. In the case of these studies, the pre-specified items are voluntary disclosure items. By
examining these voluntary disclosure items, they can assign a voluntary disclosure rating to each
firm.
The remaining three rating techniques are textual analyses. In a thematic content analysis the
whole text to be studied is analyzed, sentence per sentence, and is evaluated based on
predetermined criteria. For example, does the sentence imply good news or bad news? Does it
describe the past, the present or the future? Those kinds of issues are investigated in a thematic
content analysis. This technique is applied especially to compare different industries in order to
determine which disclosures are useful and represent good practice for which industries.
Readability studies assess how hard it is to read a text by examining the difficulty of the text and
the amount of complex words and sentences included in the text. A linguistic analysis resembles
a thematic content analysis in that the whole text is examined. However, the difference is that
21
instead of the content, the linguistic analysis looks more at the structure, connectivity and
communicative value of the text (Beattie et al., 2004).
2.4 Content analysis
In order to distinguish new style audit report adopters from non-adopters, a content analysis is
performed. The content analysis performed in this study exhibits most resemblances with the
disclosure index studies method. Namely, only a part of the Annual Report is evaluated (the
audit report) and the items to be studied are specified ex ante. The audit report of each firm in the
sample is evaluated based on the three main attributes of the new style audit report, which are
paragraphs about 1) the key audit matters/areas of particular focus, 2) materiality issues and an
overview of the scope of the audit, and 3) the firm’s going concern. When at least two out of
three paragraphs are included in a firm’s 2013 audit report, the firm is considered to be a new
style audit report adopter. The inclusion of these paragraphs can be recognized by reading the
audit reports carefully. Almost all firms clearly head their paragraphs, making it easy to assess
whether the paragraphs characterizing for the new style audit report are included. However, in
contrast to the disclosure index studies method, this study does not rate the information in the
audit report on quality. It only looks at whether certain information is included in the audit
report, in order to distinguish between new style audit reports and ‘old’ style audit reports.
Authors of studies that construct a voluntary disclosure index to measure the quality of voluntary
disclosure often have to read a firm’s Annual Report in detail in order to assign a voluntary
disclosure score to each firm. The advantage of the method this study takes is that it is quite easy
to tell whether a firm engaged in voluntary disclosure, as only the audit reports of the sampled
firms have to be studied.
2.5 Research models in voluntary disclosure literature
The research models applied in previous literature can be used to construct the research model of
this study. Therefore, this sub-section provides a short review of the research techniques used in
previous literature about the determinants of voluntary disclosure.
22
The articles which will be discussed in Section 3 apply a wide range of research models. The
three research models that are most often applied in these studies are the Ordinary Least Squares
(OLS), logistic and multiple regression models. Other research techniques included a.o. the
Poisson, linear, binomial, probit and panel data regression model. For an overview per article,
see Table 7. Five articles used a dummy variable as their dependent variable in the research
model (i.e. André, Walton & Yang, 2012; Gassen & Sellhorn, 2006; Zunker, 2011; Sheu, Liu &
Yang, 2008; and Spiegel & Yamori, 2003). This method is also applied in this study. The former
three aforementioned studies performed a logistic regression, while the latter two applied an OLS
regression. This empirical part of this study is performed by running a logistic regression, as this
research technique is constructed specifically for the purpose of regressions of which the
dependent variable is binary (i.e. it can only have two values). The choice for a logistic
regression model will be discussed in detail in Section 5.
The research design of this study will be modeled after André et al. (2012), as out of the three
articles applying a logistic regression with a dummy dependent variable, this article is the only
one solely focusing on one issue, the determinants of IFRS adoption. Zunker (2011) does not
only examine the determinants of voluntary employee-related disclosures, but she also measures
the firm characteristics related to the quality and quantity of voluntary disclosure. Besides the
determinants of IFRS adoption, Gassen & Sellhorn (2006) measure the consequences of IFRS
adoption by comparing the post-IFRS differences between IFRS adopters and non-adopters in,
for example, the bid-ask spread and stock price volatility. Both studies employ multiple
regression models for different purposes. This is in contrast to André et al. (2012), whose
research design is quite basic. This makes the research design of this article the easiest and
therefore the most applicable research design to model this study after.
3. Literature review on empirical research
This section discusses previous literature about the determinants of voluntary disclosure. The
information obtained in this section is used in order to develop the hypotheses and the research
model. Since there are no studies in this field that are specifically audit-related, all recent articles
about the characteristics explaining voluntary disclosure are considered. Mainly research about
23
the voluntary disclosure of financial information is taken into account. However, at the end of
this section, also studies related to the voluntary disclosure of non-financial, environmental
information and the early adoption of IFRS are reviewed. The order in which the articles are
discussed is based on the geographical area the articles focus on. First, studies with data of firms
on the American continent are discussed. After that, articles founded upon data from firms in
respectively Europe, Africa, East Asia and Oceania, and the Middle East are observed. All
relationships and associations discussed in this section are statistically significant.
3.1 Determinants of voluntary disclosure
In his study among 198 U.S. firms in the period 1993-2003, Premuroso (2008) examined the
determinants of the voluntary disclosure on initial outsourcing. It was found that leverage, the
total cost ratio (operating expenses over net sales) and the return on assets (ROA) were
positively associated with this voluntary disclosure item. Examining 570 U.S. firms, Zhu &
Gong (2013) found that economic performance was negatively associated with the voluntary
disclosure of executive compensation. El-Gazzar, Fornaro & Jacob (2006) also investigated
American firms. In their research, containing a sample ranging from 1996 to 2000 including 500
firms, they examined the voluntary disclosure of the report of management’s responsibilities,
which was non-mandatory at the time. The percentage of independent audit committee members,
the percentage of voting shares owned by institutional owners and the frequency of audit
committee meetings, new public debt issues and new equity issues were positively related to
voluntary disclosure. On the other hand, financial statement restatements, the percentage of
voting shares owned by management and the average interest rate on debt were found to be
negatively related to voluntary disclosure. Studying 68 Brazilian firms with executive stock
option (ESO) plans in 2007, Schiehll, Terra & Victor (2013) found that a firm’s Board size and
the presence of a compensation committee are positively related to the voluntary disclosure of
ESOs. Moreover, firms audited by a Big Four firm were found to have higher voluntary
disclosure. The authors also showed that family-controlled businesses tend to disclose less
information about ESO plans.
24
Coebergh (2011) and Braam & Borghans (2010) examined voluntary disclosure in the Dutch
market. In his research containing 399 firm-year observations over the period 2003-2008,
Coebergh (2011) found that firms with a foreign exchange listing and a high listing age are
inclined to voluntarily disclose more about corporate strategy. However, a high return on equity
(ROE) is associated with less corporate strategy disclosure. Furthermore, it was found that the
choice of disclosing voluntary information also depends on the firm’s industry. Last, the author
found that firms listed on the AEX have a higher tendency towards voluntary disclosure than
firms listed on the AMX and the ASCX, which are the second and third tier stock indices in the
Netherlands. Braam & Borghans (2010) studied the factors associated with the voluntary
disclosure of financial and non-financial performance measures, based on a sample of 149 firms
in the year 2004. In case a person associated with one firm sits on the Board of Directors of
another firm, that firm’s voluntary disclosure is higher. Moreover, disclosure tends to be higher
when other businesses, to which the firm is related via their external auditor, disclose similar
performance measures. This implies that firms tend to mimic other organizations when it comes
to voluntary disclosure, as indicated by the institutional theory.
In his study among 161 Swiss firms in 1991, Raffournier (1995) explored a positive association
between a firm’s size and international diversification and the extent of voluntary disclosures in
its annual report, measured by a disclosure index. Kateb (2012) examined the determinants of
voluntary disclosure of structural capital information in France in 2006. Applying a sample of 55
firms, it was found that firm size is positively associated with voluntary disclosure of structural
capital information, while the author found a negative relation with managerial ownership
concentration and leverage. Alves, Rodrigues & Canadas (2011) studied voluntary disclosure of
140 firms in Spain and Portugal in 2007. For each firm, a voluntary disclosure index was
measured, based on the presence of voluntary disclosure items in their annual report. The authors
found that a high proportion of the board’s remuneration that is not fixed is related to more
voluntary disclosure. In addition, firm size, growth opportunities, economic performance,
organizational performance, board compensation and shareholder ownership are positively
related to voluntary disclosure. On the other hand, a high bid-ask spread and the presence of a
large shareholder have a negative association with the level of voluntary disclosure. Oxelheim &
Thorsheim (2012) investigated the association between firm characteristics and the voluntary
25
disclosure of macroeconomic effects on corporate performance in Europe. In their sample, the
authors included 100 firm-year observations over the period 2000-2009. Cross-listing, corporate
governance strength and leverage are positively related to the voluntary disclosure of
macroeconomic effects on corporate performance. Also firms in industries with a high threat of
entry are associated with a higher level of voluntary disclosure. However, capital intensity,
which is measured as a firm’s PP&E scaled by total assets, has a negative association with the
level of voluntary disclosure.
Chakroun & Matoussi (2012), Dhouibi & Mamoghli (2013) and Kolsi (2012) all examined the
determinants of voluntary disclosure of Tunisian firms, measured by voluntary disclosure
indices. The research of Chakroun & Matoussi (2012) included 144 observations over the period
2003-2008. The authors found that firms with a high degree of regulatory reform, managerial and
institutional ownership, a large board size, a combination of the functions ‘general manager’ and
‘board chairman’, high indebtedness and a high firm age tend to disclose more voluntary items in
their annual report. Businesses in industries with an intense competition on the market for goods
and services that are family-controlled and have a high degree of board independence and
ownership concentration are associated with a low level of voluntary disclosure. The research
sample of Dhouibi & Mamoghli (2013) consists of 10 banks in the period 2000-2011. Foreign
ownership and firm size have a positive relation with voluntary disclosure, while board size,
blockholder ownership and state ownership are negatively related to voluntary disclosure.
Kolsi’s (2012) study included 52 observations from 2009 and 2010. Firms audited by a Big Four
auditor are, just like firms with a high leverage and ROA, inclined to disclose more voluntary
information. Furthermore, firms in the financial sector are associated with higher voluntary
disclosure than firms in other sectors. Barako (2007), whose research focused on 43 Kenyan
firms in the period 1992-2001, also found that firms audited by a Big Four member tend to
disclose more voluntary information. Just like the three Tunisian studies, Barako (2007)
measured voluntary disclosure on the basis of a voluntary disclosure index. Audit committee,
foreign and institutional ownership, firm size and ROE are positively related to voluntary
disclosure as well.
26
The articles of Wang, Sewon & Claiborne (2008) and Lan, Wang & Zhang (2013) also measured
voluntary disclosure by employing a voluntary disclosure index for each firm. Both studies
focused on the Chinese market. The research of Wang, Sewon & Claiborne (2008) included 109
observations from 2005. They explored that firms having a high ROE and a large degree of state
ownership and foreign ownership are associated with more voluntary disclosure. Also firms
having a Big Four auditor tend to disclose more voluntary information. In their study containing
1,066 observations from 2006, Lan, Wang & Zhang (2013) found that firm size, leverage and
assets-in-place are positively related to voluntary disclosure. However, it was found that firms
audited by a Big Four auditor disclose less voluntary information, which is in contrast to the
findings of previously discussed studies (Schiehll, Terra & Victor, 2013; Kolsi, 2012; Barako,
2007; Wang, Sewon & Claiborne, 2008).
Spiegel & Yamori (2006) examined the determinants of the voluntary disclosure of non-
performing loans in Japan. Using a sample of 814 observations from 1996 and 1997, they found
that firms with a large size and high competitive pressure are more inclined to voluntarily
disclose information about non-performing loans. On the other hand, bad loan problems and
leverage are negatively associated with the voluntary disclosure of this item. The study of Sheu,
Liu & Yang (2008) included 3,841 observations of Taiwanese firms in the period 1998-2005. The authors investigated the factors that explain the voluntary disclosure of directors’ compensation. The sum of the directors’
compensation, board size, diversified ownership and managerial ownership are positively related
to the voluntary disclosure of this item. Government ownership and native institutional
ownership were found to be negatively related to the voluntary disclosure of directors’
compensations. Besides a microeconomic perspective, Ho (2009) also takes a macroeconomic
point of view in examining the factors influencing voluntary disclosure. Applying a sample of
Malaysian firms, measuring firms’ voluntary disclosure indices in 1996, 2001 and 2006, the
author found that voluntary disclosure has increased over time. Also the occurrence of global
corporate scandals has a positive relation with voluntary disclosure. Furthermore, external
regulatory pressures, the strength of the corporate governance structure, ownership concentration
and firm size are positively associated with voluntary disclosure. Vu (2012), whose research
focused on 252 firms in the Vietnamese market over the year 2009, also found a positive relation
27
between corporate governance strength and voluntary disclosure. Moreover, size, profitability
and listing duration are associated with higher disclosure. Also a firm’s industry and auditor are
explanatory factors in voluntary disclosure. Vu (2012) found that organizations audited by a Big
Four firm are more inclined to voluntarily disclose information. State ownership and managerial
ownership are negatively related to voluntary disclosure. Zunker (2011) investigated the
determinants of the voluntary disclosure of employee-related information in Australia. Applying
a sample 970 observations from 2004, she found that firms with high past and current economic
performance, a large size and more adverse publicity tend to voluntarily disclose more
employee-related information.
The research sample in the study of Hossain & Reaz (2007) included 38 Indian banking firms in
the period 2002-2003. The authors measured firms’ voluntary disclosure levels by means of
constructing voluntary disclosure indices, which are based on firms’ annual reports. It was found
that firm size and assets-in-place are positively related to voluntary disclosure. Sehar, Bilal &
Tufail (2013) examined the determinants of voluntary disclosure also by means of composing
voluntary disclosure indices. Their sample consisted of 372 Pakistani firm observations from
2012. Profitability, firm size, firm age and a Big Four auditor were all associated with higher
voluntary disclosure, while leverage was found to be negatively related to the disclosure of
voluntary information. The study of Almutawaa (2013), who examined 206 firms from Kuwait
over the period 2005-2008, also determined the voluntary disclosure level of a firm by looking at
firms’ voluntary disclosure indices. Firms with a high degree of government ownership tend to
exert more voluntary disclosure. Firms that are cross-listed and large-sized are also associated
with higher voluntary disclosure. On the other hand, the authors found that cross-directorships, a
large board size, role duality and firm growth are related to less voluntary disclosure. Hossain &
Hammami (2009) investigated the firm characteristics associated with voluntary disclosure on
the basis of a sample of 25 Qatari firms in 2007. To measure voluntary disclosure, they
composed a disclosure checklist, which examines 44 voluntary items in firms’ annual reports. It
was found that firm age, total assets and assets-in-place are positively connected to voluntary
disclosure. Moreover, the higher the complexity of a firm, which is determined as the firm’s
number of subsidiaries, the higher the voluntary disclosure tends to be.
28
3.2 Environmental voluntary disclosure indicators
So far, primarily determinants of financial voluntary disclosure items were discussed. Although
new style audit reports are mostly about financial information, research about firm characteristics
related to the disclosure of non-financial, environmental information will also shortly be
reviewed. This way, it can be detected whether the determinants of disclosing non-financial
information radically differ from those associated with the voluntary disclosure of financial
information.
Gamerschlag, Möller, Verbeeten (2011) examined the determinants of voluntary CSR disclosure
by using a CSR disclosure index for each firm. The research sample consisted of 470 firm-year
observations of German firms over the years 2005-2008. The authors found that a firm’s size,
visibility, profitability (return on invested capital) and shareholder structure (freefloat in
percentage of shares) are positively related to the voluntary disclosure of CSR information.
Sukcharoensin (2012) studied the same relationship, also applying a CSR disclosure index, but
now for Thai firms. Applying a sample of 50 firms, the author found that firms with a high
corporate governance rating and a large degree of public ownership and ownership dispersion
tend to disclose more CSR information. On the other hand, financial leverage is negatively
related to voluntary CSR disclosure. Using a sample consisting of 300 firm-year observations
from the period 2009-2011, Borghei-Ghomi & Leung (2013) investigated the firm factors
associated with the voluntary disclosure of greenhouse gas emission (GHG) information in
Australia. It was found that firm size, ownership concentration and leverage are positively
related to the disclosure of GHG information. In addition, the larger the proportion of non-
executive directors on a firm’s board, the higher the voluntary disclosure of GHG information
tends to be. Also, cross-listed firms are more inclined to report on GHG than single-listed firms.
Prado-Lorenzo, Rodríguez-Domínguez, Gallego-Álvarez & García-Sánchez (2009) examined the
determinants of GHG disclosure based on a sample of 101 firms worldwide in the year 2005.
They found that firms that are large-sized and have a high market-to-book ratio disclose more
information on GHG. A firm’s industry is an essential factor in explaining the choice for GHG
disclosure as well. Besides airlines, also corporations doing business in chemicals, forest and
paper products, metals, mining, motor vehicles and utilities display a higher amount of GHG
29
disclosure than firms in other sectors. On the other hand, firms in the aerospace and defense
industry report less GHG information. Also firms with a high ROE tend to report less on GHG.
As can be seen, determinants of voluntary disclosure are largely the same when comparing non-
financial disclosures with financial voluntary disclosures.
3.3 Determinants of early IFRS adoption
In 2005, the new accounting framework IFRS (International Financial Reporting Standards)
became mandatory for listed firms in European Union countries. Many firms anticipated on this
regulation by preparing their financial statements in accordance with IFRS before it became
mandatory. The issuance of the new style audit report is similar to the early adoption of IFRS,
because in both cases firms anticipated on the upcoming regulation by voluntarily adopting the
new standard. Therefore, also studies about the determinants of early IFRS adoption are
discussed.
Gassen & Sellhorn (2006) examined the determinants of IFRS adoption between 1998 and 2004
for German firms. The authors found that firm size, international exposure, ownership dispersion
and recent IPOs are positively related to early IFRS adoption. Furthermore, it was showed that
firms that are also listed in the U.S. were more inclined to adopt IFRS. André, Walton & Yang
(2012) investigated the determinants of voluntary IFRS adoption for non-listed UK firms in
2009. In contrast to listed firms in the European Union, IFRS is still non-mandatory for non-
listed firms. By applying a sample of 8,417 firms, the authors found that firm size, leverage,
internationality and having a Big Four auditor are positively related to voluntary IFRS adoption.
Kolsi & Zehri (2013) examined both microeconomic and macroeconomic determinants of
voluntary IFRS adoption. In their sample, they included 700 firm-year observations from 74
developing countries worldwide. Half of those countries adopted IFRS in 2008. Countries with
an Anglo-Saxon culture, a common law system, a good educational system and a high degree of
foreign operations and economic growth are more inclined to adopt IFRS than other countries.
Of the microeconomic determinants, firm size and having a Big Four auditor were found to be
positively related to IFRS adoption. From the discussed articles, it can be derived that the
determinants of IFRS adoption are similar to the explanatory factors for voluntary disclosure.
30
3.4 Conclusion
Concluded, is can be stated that certain firm characteristics are often significantly related to
voluntary disclosure. In many studies, firm size and profitability, which is often measured as the
ROA or ROE of a firm, are found to be significantly positively associated with voluntary
disclosure. Moreover, a large majority of the articles provided evidence that firms with a Big
Four auditor disclose more voluntary information than those audited by non-Big Four firms. In
addition, cross-listed firms tend to be more active in voluntary disclosures than single-listed
firms. Other recurring (at least two times) significantly positive relations have been found for
firm age, listing age, corporate governance strength and assets-in-place. Also industry type
frequently is a factor in explaining the choice for voluntary disclosure. Evidence is mixed
concerning the relation of competitive pressure, leverage, role duality and board size with
voluntary disclosure. No recurring significantly negative associations could be derived from the
studied articles.
Many studies also included variables related to stock ownership in their research. Overall,
institutional ownership and foreign ownership are positively associated with voluntary
disclosure. There is mixed evidence on the relation of voluntary disclosure with managerial and
state ownership, while ownership concentration is negatively related to voluntary disclosure in a
large majority of the articles. In addition, some studies found evidence that family-controlled
businesses are less willing to disclose voluntary information.
Table 13 (see appendix) is constructed with the purpose to show the number of significant
relations for each of the most recurring voluntary disclosure determinants. The numbers are
constructed based on all articles discussed in this section. In the next section, the hypotheses are
developed on the basis of the information acquired in this section. Table 13 is helpful in
recognizing which variables have exhibited most significant associations with voluntary
disclosure in prior studies.
Two tables summarizing the articles discussed in this section are provided below.
31
Table 1: Summary of articles about the determinants of voluntary disclosure
Authors Geographical area
Sample Research method
Voluntary disclosure item
Positive association
No association
Negative association
Premuroso (2008)
U.S. 1993-2003;198 firms
Binary logistic regression
Initial outsourcing
Level of debt, total cost ratio, ROA
Zhu & Gong (2013)
U.S. 570 firms Regression analysis
Executive compensation
Economic performance
El-Gazzar, Fornaro & Jacob (2006)
U.S. 1996-2000; 500 firms
Logistic regression; OLS regression
Report of management’s responsibilities
Ratio of independent to total audit committee members, frequency of audit committee meetings, new public debt issues and new equity issues, institutional ownership
ROA, debt/equity ratio
Financial statement restatements, management ownership, interest rate on debt
Schiehll, Terra & Victor (2013)
Brazil 2007;68 firms with ESO plans
OLS regression
Executive stock options
Board size, presence compensation committee, Big 4 auditor
Proportion of independent directors, CEO duality, ownership concentration
Alves (2011) Spain, Portugal (i.e. the Iberian Peninsula)
2007;140 firms
Multiple regression analysis
Voluntary disclosure index
Management incentives (= proportionof the board’s remuneration that is not fixed),size, growth opportunities, economic performance, organizational performance, board compensation
Managerial ownership, government ownership, board independence, board size, existence of monitoring structures, board expertise, leverage, ownership concentration, turnover ratio
Bid-ask spread, large shareholder
Chakroun & Matoussi (2012)
Tunisia 2003-2008; 144 observations
Multiple regression analysis
Voluntary disclosure index
Regulatory reform, board size, managerial ownership, role duality, institutional ownership, leverage, firm age
Size, Big Four auditor)
Board independence, ownership concentration, family-controlled, competition on the market
Dhouibi & Tunisia 2000-2011; Linear- Voluntary Foreign Number of Board size,
33
Mamoghli (2013)
10 banks multiple regression
disclosure index
ownership, size independent directors, role duality, Big 4 auditor
blockholder ownership, state ownership
Kolsi (2012) Tunisia 2009-2010; 52 firms
Multiple regression analysis
Voluntary disclosure index
Leverage, Big Four auditor, ROA, financial sector
Large shareholder, size
Barako (2007) Kenya 1992-2001; 43 firms
Pooled OLS regression with panel-corrected standard errors
Variable N Min Max Mean St. dev. Median Skewness Kurtosis*NEW 111 0 1 0.2703 0.4461 0 1.0346 -0.9296REV (x 1,000,000) 111 0.069 57,665 3,541 8,220 508.5 0.9150 18.2009LEV 111 0.01 1.27 0.5731 0.2430 0.56 -0.0583 -0.0459BIG_4 111 0 1 0.8739 0.3335 1 -2.2523 3.0729CROSS 111 0 1 0.6486 0.4796 1 -0.6228 -1.6122BOARD_SIZE 111 1 9 2.8108 1.5107 2 1.1821 1.7114AGE 111 1 131 25.559 21.509 18 2.1107 5.9381MRKT_CONC 111 0.075 0.94 0.3721 0.1836 0.358 0.9478 0.6545FREEFLOAT 111 0.01 1 0.6759 0.2830 0.75 -0.6987 -0.5466* Stata assumes a perfect normal distribution at a kurtosis of +3. Because the reference normal distribution often has a kurtosis of zero, the kurtosis is adjusted by -3 (DeCarlo, 1997).
The descriptive statistics of all variables, except the industry variables, are provided in Table 4.
For each variable, the number of observations (N), the minimum and maximum value, the mean,
standard deviation, median, skewness and kurtosis are given. Skewness is the degree to which
the different observations of a variable point into one direction, applying the perspective of a
normal distribution. Suppose that 80 out of 111 observations are on the left side of the null-point
of the normal distribution. In that case, the skewness of the variable is negative. However, when
80 out of 111 observations are on the right side of the null-line, the skewness is positive. Balanda
& MacGillivray (1988) define kurtosis as “the location- and scale-free movement of probability
mass from the shoulders of a distribution into its centre and tails.” A positive kurtosis means that
there is an even distribution of observations of a variable, which corresponds to a sharp peak.
The higher the value of the kurtosis, the sharper this peak. In case of a negative kurtosis, many
observations are concentrated toward the mean, which implies a flat top. A kurtosis of zero,
meaning that it is neither negative nor positive, corresponds to a perfect normal distribution. For
a graphical illustration of the different degrees of kurtosis, see Figure 1 on the next page.
Figure 1: Levels of kurtosis
54
(Source: Signal Trading Group, 2014)
As can be seen in Table 4, 27.0% of firms have issued a new style audit report. This percentage
corresponds to 30 out of 111 firms. Almost two-thirds of the sampled firms are cross-listed
(64.9%), and an even larger majority is audited by a Big Four firm (87.4%).
What is notable is the broad range of values for listing age (1 to 131), although the mean is quite
normal, with a listing age of 25.6 years. Another remarkability is that the maximum value of the
variable LEV is 1.27, meaning that there is a firm with more liabilities than assets, which is the
result of a negative equity.
Table 5: Frequency of sector variablesVariables Frequency
One of the most notable positive correlations (0.2891) exists between REV and BOARD_SIZE. It can be theorized that larger firms have more resources available for the
appointment of a large board. Big firms might also be in need of a larger board, as there are
probably more issues to discuss and decisions to be made.
The variable REV also exhibits a high correlation coefficient with CROSS (0.2714), which
implies that large firms are more prone to list their shares on another stock exchange. Also this
correlation coefficient is perfectly understandable. When a firm grows larger, it might consider
issuing shares on the local stock exchange. The same procedure holds when a firm is already
listed. When a firm continues to grow, it starts thinking about expanding and attracting more
capital, by, for example, listing on a foreign stock exchange.
The most notable positive correlation (0.4025) can be found between the variables BIG_4 and CROSS. Firms with a Big Four auditor are generally larger than firms with a non-Big Four
auditor, which is also implied by the correlation coefficient between REV and BIG_4 (0.1591).
The theory behind this positive correlation is the same as for the correlation between CROSS and REV, which was discussed above. Larger firms, which have a higher probability of having a
Big Four auditor, are more likely to issue shares on a foreign stock exchange in order to attract
new capital.
Besides the industry-related correlations, the most negative correlation (-0.1243) could be found
between AGE and LEV. An explanation could be that start-up firms make large capital
investments. The returns on these investments are often earned years later. Therefore, these firms
might have a higher leverage ratio, as the start-up investments are recorded as liabilities on their
balance sheet.
6.2 Results of the statistical tests
In this sub-section, the results obtained from the statistical tests are explained and discussed.
Table 8 provides the results of the Breusch-Pagan/Cook-Weisberg test for heteroskedasticity.
59
When prob>chi2 is higher than 0.05, the variance of the error term is constant, implying that
there is homoscedasticity. However, when prob>chi2 is lower than 0.05, there is
heteroskedasticity. In this case, the probability equals 0.5282. There is no heteroskedasticity.
Table 8: Breusch-Pagan/Cook-Weisberg test for heteroskedasticity
Prob > chi2 = 0.5282 chi2(13) = 11.99
SECTOR_INDUSTRIALS SECTOR_CONSGOODS SECTOR_CONSSERVICES SECTOR_FINANCIALS Variables: REV LEV BIG_4 CROSS BOARD_SIZE AGE MRKT_CONC FREEFLOAT SECTOR_MATERIALS Ho: Constant varianceBreusch-Pagan / Cook-Weisberg test for heteroskedasticity
Table 9 provides the test for multicollinearity. Normally, a threshold of 10 is applied when
analyzing whether multicollinearity exists among the independent variables (e.g. Hair Jr. et al.,
1995; Kennedy, 2003). When the mean VIF is higher than 10, substantial multicollinearity
exists. As can be seen, the mean VIF of the independent variables is 4.01. Because the mean VIF
is lower than 10, it can be concluded that there is no multicollinearity among the independent
variables.
Table 9: Variance Inflation Factor (VIF) test for multicollinearity
The Shapiro-Wilk W test, provided in Table 10, tests whether the sample population is normally
distributed. The sample population of the large majority of the variables is normally distributed,
60
as their value of prob>z is lower than 0.05. Only the sample populations of LEV and CROSS are non-normally distributed. Overall, it is safe to say that the sample population is normally
distributed, as almost none of the variables exhibits non-normality of the distribution of the
sample population.
Table 10: Shapiro-Wilk W test for the normality of the data distribution
Log likelihood = -50.191422 Pseudo R2 = 0.2251 Prob > chi2 = 0.0037 LR chi2(12) = 29.16Logistic regression Number of obs = 111
This thesis distinguishes between three significance levels: 1%, 5% and 10%. The 1% and 5%
significance levels are widely applied and recognized in academic research. Results with a 10%
significance level have to be taken with caution, as the 10% significance level is less commonly
used in academic studies. However, it has been decided also to recognize the results significant
at 10% as being significant, instead of labeling them as insignificant. First of all, quite some
results in this study are not significant at the 5% level but are significant at the 10% level.
63
Recognizing the 10% significance level gives the readers of this thesis more detailed information
about the strength of the relation between the dependent and the independent variables. Besides,
it seems inappropriate to label variables with a p-value of 0.06 the same as variables with a p-
value of 0.90, i.e. both insignificant.
H1: Firm size is positively associated with the voluntary disclosure of the new style audit report.
As can be seen in Table 11, firm size (REV) is not significantly related to the voluntary
disclosure of the new style audit report. In contrast to what was hypothesized, firm size even
exhibits a slightly negative relation with the dependent variable. This means that H1 is rejected.
There is no significant relation between the two variables.
H2: There is a positive association between leverage and the voluntary disclosure of the new
style audit report.
There is also no significant association between leverage (LEV) and NEW, the dependent
variable. On average, highly leveraged firms are slightly less likely to issue a new style audit
report than firms with a lower leverage ratio, although this relation is non-significant. As a result,
H2 is rejected.
H3: There is a positive association between Big Four-audited firms and the voluntary disclosure
of the new style audit report.
Conform H3, there is a significantly positive association between having a Big Four auditor
(BIG_4) and the issuance of the new style audit report. This relation holds at a significance level
of 10%. It is possible to interpret the regression parameter by computing the exponential function
of the parameter. That results in the odds ratio. The exponential function of the regression
coefficient of BIG_4 (2.056) is 7.81. So the odds ratio shows that firms with a Big Four auditor
are 7.81 times more likely to issue a new style audit report than firms with a non-Big Four
auditor.
H4: There is a negative association between cross-listed firms and the voluntary disclosure of
the new style audit report.
64
There is a significantly negative association between CROSS and the voluntary disclosure of
the new style audit report at a significance level of 5%. The probability that a cross-listed firm
issues a new style audit report is 4.55 times lower compared to a single-listed firm. This number
represents the odds ratio. In case of a negative regression coefficient, 1 has to be divided by the
exponential function of the regression coefficient of CROSS (-1.514), which is 0.22. This result
is conform H4, but in contrast to what theory says about the relation between being cross-listed
and voluntary reporting.
H5: Board size is positively associated with the voluntary disclosure of the new style audit
report.BOARD_SIZE does not exhibit a significant relationship with the dependent variable, which
means that H5 is rejected. Although a positive association was hypothesized, this result is not
that surprising, as previous studies were mostly inconclusive about the relation between board
size and voluntary disclosure. With an extra board member, it becomes only 1.04 times less
likely that a firm engages in issuing a new style audit report. As previously mentioned, this ratio
is calculated by dividing 1 by the exponential function of the parameter coefficient.
H6: Age is positively associated with the voluntary disclosure of the new style audit report.
At a significance level of 10%, there is a positive relation between the listing age of a firm
(AGE) and the voluntary disclosure of the new style audit report. This is in accordance with the
direction specified by H6, so H6 is confirmed.
H7: Competitive pressure is negatively associated with the voluntary disclosure of the new style
audit report.
Competitive pressure is significantly negatively associated with the voluntary disclosure of the
new style audit report, as there is a positive relation between MRKT_CONC and NEW at a
significance level of 10%. This is conform H7, so as a result, this hypothesis is confirmed. From
this result, it can be concluded that firms in markets with lower competitive pressure are more
likely to issue a new style audit report.
65
H8: Ownership dispersion is positively associated with the voluntary disclosure of the new style
audit report.
It is found that there is significantly positive association between ownership dispersion and the
voluntary disclosure of the new style audit report, as the proxy for ownership dispersion
(FREEFLOAT) exhibits a positive relation with NEW at a significance level of 1%.
Therefore, H8 can be confirmed.
H9: Associations exist between the voluntary disclosure of the new style audit and the industry of
firms.
There are six sector variables, of which five are included in the regression. SECTOR_TECHNOLOGY is excluded from the regression and therefore used as the
benchmark, which is common practice when examining industry effects. From the regression
results, it can be derived that firms in the industrials and consumer services sector (i.e. SECTOR_INDUSTRIALS and SECTOR_CONSSERVICES) issue significantly more
new style audit reports than firms in the technology sector. These results hold at a 5%
significance level. For the other sectors, no significant associations can be found. Therefore, H9
is confirmed, as two significant industry effects have been found.
Concluded, BIG_4, AGE and MRKT_CONC are significantly positively associated with the
dependent variable at a 10% level. FREEFLOAT is significantly positively related to the
issuance of the new style audit report at a 1% significance level. On the other hand, there is a
significantly negative association between CROSS and NEW at a significance level of 5%. At
a significance level of 10%, there is also a significantly negative relation between competitive
pressure and the voluntary disclosure of the new audit report. Furthermore, at a 5% level, it was
found that firms in the sectors ‘Industrials’ and ‘Consumer Services’ disclose significantly more
new style audit reports than firms in the sector ‘Technology’, which served as the benchmark.
Summarized, H1, H2 and H5 are rejected, meaning that the results show no significant
association in the expected direction for these hypotheses. However, H3, H4, H6, H7, H8 and H9
are confirmed, as for these hypotheses there is a significant relationship in the pre-specified
direction.
6.4 Analysis
66
Big Four auditor
There is a significantly positive relation between having a Big Four auditor and the voluntary
disclosure of the new style audit report. Previous empirical research provides overwhelming
evidence of a strong positive relation between voluntary disclosure and having a Big Four
auditor, but the relation between the disclosure of the new style audit report and having a Big
Four auditor is only significant at 10%. A reason for this could be that, although they consult
with and get advice from their auditor, the firms themselves still have the final word about
whether or not issuing a new style audit report. Still, although at 10%, there is a significantly
positive association between issuing the new style audit report and having a Big Four auditor. It
is an indication that Big Four audited firms, in cooperation with their auditor, spark innovative
reporting, as these firms and their auditors seem to be the driving force behind the new style
audit report.
Cross-listing
The significantly negative relation for cross-listed firms is the result of the inclusion of non-
Dutch firms in the sample, which are almost per definition cross-listed, engaging less in
voluntarily disclosing the new style audit report. As explained in Section 4, non-Dutch firms are
probably less aware of the new style audit report and its advantages. This result proves that this
thought is likely to be the explanation for the negative relation between CROSS and NEW. So
although the studies in Section 3 only found significantly positive associations between a cross-
listing and voluntary disclosure, this result is therefore not unexpected. An alternative
explanation can be given applying the theory of Graham et al. (2005). They stated that attaining
more analyst coverage is one of the main reasons for firms to engage in voluntary disclosure. It
seems safe to assume that cross-listed firms have more analyst coverage, as, because of their
cross-listing, they have more international exposure in a variety of countries. Non-cross-listed
firms are more likely than cross-listed firms to have a lack of analyst coverage. Therefore, non-
cross-listed firms might be more eager to voluntarily issue a new style audit report, in order to
attract more analyst coverage.
Firm age
67
The significantly positive relation for the variable AGE strengthens the thought that the market
expects a higher level of voluntary disclosure from older firms, as theorized by Coebergh (2011).
As explained in Section 2, Graham et al. (2005) found that one of three main reasons managers
engage in voluntary disclosure is to increase or maintain the reputation of their firm. This result
implies that older firms take more responsibility in taking care of their reputation by responding
to the needs of society, which corresponds to the legitimacy theory. The positive association is in
line with previous empirical research, as the discussed studies in Section 3 found three
significantly positive associations against zero significantly negative associations between
voluntary disclosure and firm age.
Competitive pressure
There is a significantly negative relation between competitive pressure and voluntary issuing the
new style audit report, although only at a 10% significance level. Evidence of previous literature
on the relation between competitive pressure and voluntary disclosure was mixed, so it is not
unexpected that the significance level of this relation is only 10%. This result can be explained
by the proprietary cost theory, which theorizes that firms in industries with high entry barriers
are more affected when there is a newcomer, putting pressure on their profits. As a result, they
are more willing to satisfy their stakeholders, by, for example, providing the new style audit
report. An alternative theory explained in Section 4, arguing that firms in competitive markets
experience higher stakeholder pressure to disclose voluntary information in order to earn or keep
a competitive advantage, is not supported by this result.
Ownership dispersion
The most significantly positive relation exists between ownership dispersion and the voluntary
issuance of the new style audit report, at a significance level of 1%. Although a positive
association was hypothesized, it is quite surprising that this relationship is this strong, as
previous literature on the relation between ownership dispersion was quite mixed; three studies
found a significantly positive result, while two studies obtained a significantly negative result.
This association can be explained by the theory of Chau & Gray (2002), who state firms with a
dispersed ownership structure have more shareholders, therefore facing more public pressure to
issue voluntary information. Again, there are common grounds with the legitimacy theory.
68
Another explanation for this positive relationship can be provided by the agency theory. A high
degree of ownership dispersion often implies that there is a low amount of institutional
ownership. Institutional owners have normally more inside information than the rest of
shareholders do, because they own a larger part of the firm’s share capital. So when ownership is
dispersed, there is likely to be more information asymmetry between shareholders and
management. Information asymmetry is one of the two main pillars of the agency theory.
Because of this higher degree of information asymmetry, management might feel more need to
voluntarily disclose the new style audit report when ownership is dispersed. This type of
innovative reporting might also make management win the trust of their shareholders, thereby
reducing monitoring costs, and subsequently agency costs, in the future.
Insignificant relations with the dependent variable were found for the independent variables REV, LEV and BOARD_SIZE. Firm size, the degree of leverage and the size of the board are
not associated with the voluntary disclosure of the new style audit report.
Firm size
It was theorized that large firms have more resources and feel more pressure to issue voluntary
information, because of their higher public visibility. Large firms are also more likely to issue
new shares by means of a Seasoned Equity Offering (SEO) in order to attract more capital.
Disclosing the new style audit report would lead to less uncertainty about firm performance by
capital providers. However, no evidence for these theories was found when it concerns the
relation between firm size and the issuance of the new style audit report. An explanation for this
result could be that large firms have many other channels to communicate with the public.
Instead of communicating information via the new style audit report, big firms might, more so
than small firms, provide useful voluntary information to their investors via communication
channels like conference calls and press meetings. Another reason for this insignificant relation
is provided by the positive accounting theory, and more specifically by the bonus plan
hypothesis. Recalling the information from Section 2, this hypothesis assumes that managers
engage in voluntary disclosure out of self-interest, because they want to increase their own
compensation. According to Schaefer (1998), CEOs of large firms have a lower proportion of
stock-based compensation relative to total salary than CEOs of smaller firms. Elaborating on this
69
thought, it could be stated that it is more important for CEOs of small firms rather than large
firms that the share price of their firm’s stock increases. One way to do so is by pleasing their
shareholders by, for example, voluntarily disclosing the new style audit report. An innovative
reporting strategy might have a positive effect on the share price, which would increase
compensation especially for CEOs of small firms, who therefore might have more incentives to
issue the new style audit report. Relating the negative association between firm size and the
proportion of stock-based compensation with the bonus plan hypothesis might explain why there
is an insignificant relation between firm size and the voluntary disclosure of the new style audit
report. Still, the insignificant relation between firm size and the voluntary disclosure of the new
style audit report is unexpected. Namely, the discussed articles in Section 3 found sixteen
significantly positive associations against zero significantly negative associations between firm
size and voluntary disclosure.
Leverage
The positive direction of the hypothesis for the relation between leverage and the disclosure of
the new style audit report was mainly based on the agency theory. Higher leverage would lead to
more monitoring of management by the shareholders, hence resulting in more pressure for
management to show their good intentions, by for example disclosing the new style audit report.
An argument in favor a negative association between leverage and voluntary disclosure was
provided by the signaling theory. Firms with low leverage might be more willing to disclose
voluntary information in order to signal their superior performance. However, from the results, it
can be concluded that both theories were not able to explain the relation between leverage and
the disclosure of the new style audit report. This result is not that surprising, as the previous
studies discussed in Section 3 found mixed results on the relation between leverage and
voluntary disclosure; six found a significantly positive relationship, while four times a
significantly negative association was found.
Board size
The expected positive relation between board size and the issuance of the new style audit report
could be explained by the monitoring role of the Supervisory Board. A large Supervisory Board
is associated with a large Board of Directors. Hence, there are more resources to monitor the
70
board, and to evaluate individual directors on their performance. Most directors might have
incentive packages, meaning that they are rewarded for good performance and punished for bad
performance. Issuing an innovative reporting concept like the new style audit report in the
Annual Report, might lead to a better evaluation from Supervisory Board. However, from the
results it can be concluded this this is not the case. A reason for this insignificant relation might
be that the Supervisory Board normally is very close with the Board of Directors, as the
members of the Supervisory Board are appointed by the Board of Directors. This could be
detrimental for the monitoring function of the Supervisory Board. As with leverage, based on the
articles in Section 3, there was no overwhelming evidence of the hypothesized positive relation
between board size and voluntary disclosure. As three studies found a significantly positive
relation and two found a significantly negative relation, this insignificant association might not
come as a total surprise.
Industry
The fact that there are significant industry effects for the sectors Industrials and Consumer
Services supports the institutional theory, and more specifically the notice of mimetic
isomorphism. This theory relies on the notion that firms tend to mimic the practices of
comparable firms, for example firms in their own industry. Since certain industries display a
higher level of issuing the new style audit report than others, this theory seems applicable. Since
previous studies have mostly used other industry classification schemes to test for significant
industry effects, it is hard to compare the industry results of this study to previous studies.
However, it seems quite surprising that the sector Technology is least associated with the
voluntary disclosure of the new style audit report. Firms in the technology business are often
known to make their money by coming up with innovative concepts and products, so one would
think that these firms would embrace an innovative reporting concept like the new style audit
report. However, from the results it seems like the innovativeness of firms in the technology
sector does not hold for their reporting practices.
6.5 Robustness test
71
Table 12 represents the robust logistic regression. According to Shevlyakov & Vilchevski
(2001), robustness is the “stability of statistical inference under the variations of the accepted
distribution models.” That is, the robust logistic regression tests whether the results still hold
when there are small deviations from normality. The exact scenario under which this robust
logistic regression is ran is auto-generated by Stata.
As can be seen in Table 12, the coefficients have not changed severely. This indicates that there
is a true causal relationship between the significant independent variables and the dependent
variables. All except two variables are still significant at the level of the initial logistic
regression. In the robust logistic regression, MRKT_CONC is not significant at the 10% level
anymore, while the sector variable SECTOR_CONSSERVICES is now significant at the
10% level instead of the 5% level. However, because the changes in p-values are only very slight
and the regression coefficients have changed minimally, it can be concluded that the initial
logistic regression is representative for this research.
The next section presents the conclusions of this study. In this section, the results will be
summarized and the limitations of the research will be discussed. In addition, opportunities for
further research are brought forward.
7. Summary and conclusion
This section provides a summary of the thesis, the main results, and recommendations for further
research.
The objective of this thesis was to examine the firms characteristics associated with the voluntary
disclosure of the new style audit report. Therefore, in the introduction, the following main
research question was constructed:
What are the firm characteristics associated with the voluntary issuance of the new style audit
report in the Dutch stock market?
Based on the research performed in this thesis, the answer to the main research questions is as
follows: ownership dispersion, firm age and having a Big Four auditor are the firm
characteristics significantly positively associated with the voluntary issuance of the new style
audit report in the Dutch stock market, while competitive pressure and a cross-listing are
significantly negatively related to the voluntary disclosure of the new style audit report.
Furthermore, the sectors Industrials and Consumer Services display a significantly positive
relation with the disclosure of the new style audit report.
Summary
73
After introducing the subject of this thesis in Section 1, in Section 2, the attributes of the new
style audit report were explained. In contrast to the old format audit report, the new style audit
report usually includes paragraphs on 1) key audit matters/areas of focus, 2) materiality and an
overview of the scope of the audit, and 3) a firm’s going concern. When at least two out of three
paragraphs were included into a firm’s audit report, the firm was considered to be a new style
audit report adopter. Furthermore, Section 2 discussed theories that could be related to voluntary
disclosure practices. These included the agency theory, signaling theory, capital need theory,
legitimacy theory, positive accounting theory, institutional theory and litigation cost theory.
Section 3 provided a review of empirical studies focusing on the factors associated with
voluntary disclosure. Size, profitability, a Big Four Auditor and leverage were the characteristics
found to be most often significantly associated with voluntary disclosure. The former three
factors pointed towards a positive association, while evidence on the relation between leverage
and voluntary disclosure was mainly mixed. Out of these four ‘top variables’, only profitability
was not included in the regression model, because of the substantial yearly changes in
profitability ratio for many (small) firms in the research sample. The other variables included
into the research model yielded either positive or mixed results in previous research. The
evidence on the relation of voluntary disclosure with being cross-listing (4 positive, 0 negative)
and firm age (3 to 0) was convincingly positive, while the factors board size, ownership
dispersion (both 3 to 2) and competitive pressure (1 to 1) yielded mixed evidence in prior
research.
In Section 4, the hypotheses were constructed. In order to test the hypotheses, a logistic
regression model was constructed in Section 5, measuring the relation between the issuance of
the new style audit report and the hypothesized firm characteristics. Besides the research model,
also the research sample, variables, descriptive statistics and statistical tests were discussed in
Section 5.
Results
Section 6 provided the main results of this study. First of all, a correlation matrix was
constructed. No high correlation coefficients (-0.5 > r > 0.5) between the variables could be
74
detected, which is generally a good sign. After that, several statistical tests on the regression
model were performed. These included tests for multicollinearity, heteroskedasticity and the
normality of the data distribution, and a graph depicting estimates of the distribution of the
residuals. Again, no severe problems were detected, indicating that the regression model is well-
fitted. The results of the logistic regression showed that there are significantly positive
associations between the disclosure of the new style audit report and listing age, having a Big
Four auditor (both at a 10% significance level), and ownership dispersion (at 1%). Significantly
negative relations with the issuance of the new style audit report were found for competitive
pressure (10%) and being cross-listed (5%). Insignificant associations with the disclosure of the
new style firms in the sectors Consumer Services and Industrials are significantly positively
related to the disclosure of the new style audit report. As a result, three out of nine hypotheses
had to be rejected, while six out of nine hypotheses were confirmed by the regression results.
Analysis
In the analysis part of Section 6, it was attempted to link the results to existing theories, mostly
discussed in Section 2. Especially the legitimacy theory can be linked to the voluntary disclosure
of the new style audit report. The significantly positive associations found for the factors firm
age and ownership dispersion can be explained by this theory. According to the legitimacy
theory, the public expects rather from older than younger firms that they engage in voluntary
reporting. Older firms feel more pressure to comply with the public’s expectations, for example
by issuing the new style audit report. Also firms with a dispersed ownership structure are likely
to face more public pressure when it concerns voluntary disclosure, because they have a larger
amount of shareholders. The association with ownership dispersion, however, can also be
explained by the agency theory. High ownership dispersion implies little institutional ownership.
Institutional owners often have more inside information about the firm, as they normally own a
large part of the firm’s shares. When ownership is dispersed, it means that shareholders have less
inside information. This results in a higher information asymmetry between management and
shareholders, so there is more need for management to regularly inform their shareholders about
the firm. This can be done by voluntary disclosure, and more specifically by issuing the new
style audit report. The significantly negative association of competitive pressure with the
issuance of the new style audit report was substantiated by the proprietary cost theory. According
75
to this theory, the profits of firms in high entry barrier industries, which are associated with little
competitive pressure, are more hurt by the entrance of new firms into the industry. Therefore, for
these firms it is more important to please their stakeholders. One way to this is by providing
more voluntary disclosures. It was analyzed that the insignificant relation between firm size and
the voluntary disclosure of the new style audit report might have to do with the bonus plan
hypothesis of the positive accounting theory. Previous research has provided evidence that the
total compensation of CEOs of small firms is composed of a relatively higher proportion of
stock-based compensation compared to CEOs of large firms. CEOs of small firms might think
that they can satisfy their shareholders by disclosing the new style audit report, which, in turn,
might increase the price of their stock. Subsequently, this would increase the CEO’s
compensation. The significant industry effects are in line with the institutional theory of mimetic
isomorphism, which argues that similar firms tend to copy other firm’s practices.
Comparison with previous research
The majority of the significant results obtained in this thesis is conform previous empirical
studies examining the factors related to voluntary disclosure, discussed in Section 3. The
significantly positive associations for having a Big Four auditor and firm age were supported by
previous empirical evidence. For leverage and board size, previous empirical evidence was
mixed, although slightly leaning towards a positive relation, so it was not that surprising that
both variables were found to be insignificantly related to the issuance of the new style audit
report. However, previous literature on competitive pressure and ownership dispersion was also
mixed, but this time, a significant association was found for both variables. The significantly
negative association between being cross-listed and the issuance of the new style audit report
was in contrast to what previous research said. This result could be explained by the theory that
non-Dutch firms, which are almost by definition cross-listed in this study, are less aware of the
new style audit report. The insignificant association for firm size was surprising looking at
previous research, which found overwhelming evidence of a positive association with voluntary
disclosure.
Recommendations
76
This thesis examines whether there are significant associations between firm characteristics and
the disclosure of the new style audit report, but it does not tell whether there is causality between
the variables, as it was not the objective of this study to detect causal connections. However, it
cannot be ruled out that there might be causal relationships between some significant variables
and the voluntary disclosure of the new style audit report. Assuming there is causality, a few
recommendations to certain parties of interest are listed. First of all, it might be interesting for
standard-setters that this thesis provides evidence that Big Four audited firms are currently more
engaged with the new style audit report than firms audited by a non-Big Four firms. It is a sign
that Big Four auditors are already quite familiar with the new style audit report, while most non-
Big Four auditors are not. Before making the new style audit report obligatory, it might be wise
for the IAASB to provide training and guidance on the new style audit report to non-Big Four
audit firms. It is of highest importance to get non-Big Four auditors acquainted with the
regulations of the new style audit report, in order to preserve audit report quality in the future.
Moreover, as the new style audit report is an innovative concept, it is important for investors to
know which types of firms tend to engage in or restrain from groundbreaking initiatives like the
new style audit report. Innovativeness in reporting can be linked to innovativeness in business.
As innovative firms are often believed to have good growth prospects, these firms are attractive
targets for investors.
Limitations
There are some limitations to this research. For example, the size of the research sample is quite
small, as it only includes 111 observations. Normally, this reduces the power of the results and
makes it more difficult to obtain significant statistics. Luckily, this research has provided quite
some significant results, but it is a factor to consider when the sample size is small. Furthermore,
the data about firms’ competitive pressure were drawn from the U.S. market. Although the U.S.
data probably exhibit the reality on the European market, it might influence the results.
Further research
Because the new style audit report is still a brand new concept, there are plenty of opportunities
for further research on this topic. It would be interesting to examine the differences between the
future performances of voluntary new style audit report adopters compared to non-adopters.
77
Think about figures like revenue growth, cost of capital and profitability growth. Moreover, in
case the IAASB delays obliging the new style audit report beyond 2015, the firm characteristics
associated with the voluntary issuance of the new style audit report can be also researched for
other countries. Now that many Dutch firms have introduced the new style audit report in 2014,
it seems likely that non-Dutch firms will follow this example in the coming years.
References
Almutawaa, A. M. A. E. (2013). Perceptions of Corporate Annual Reports’ Users Toward Accounting Information and Voluntary Disclosure and Its Determinants: The Case of Kuwait. Working Paper Series, Durham University.
Alves, H. S. A. (2011). Corporate Governance Determinants of Voluntary Disclosure and Its Effects on Information Asymmetry: An Analysis for Iberian Peninsula Listed Companies. Working Paper Series, University of Coimbra.
André, P., Walton, P. J., & Yang, D. (2012). Voluntary Adoption of IFRS: A Study of Determinants for UK Unlisted Firms. Comptabilités et Innovation, 33ème congrès de l'AFC.
Arya, A., & Mittendorf, B. (2005). Using Disclosure to Influence Herd Behavior and Alter Competition. Journal of Accounting and Economics, 40(1-3), 231-246.
Balanda, K. P., & MacGillivray, H. L. (1988). Kurtosis: A Critical Review. The American Statistician, 42(2), 111-119.
Barako, D. G. (2007). Determinants of Voluntary Disclosures in Kenyan Companies Annual Reports. African Journal of Business Management, 1(5), 113-128.
Beattie, V., McInnes, W., & Fearnley, S. (2004). A Methodology for Analysing and Evaluating Narratives in Annual Reports: A Comprehensive Descriptive Profile and Metrics for Disclosure Quality Attributes. Accounting Forum, 28(3), 205-236.
Beck, N. (2011). Is OLS with a Binary Dependent Variable Really OK?: Estimating (Mostly) TSCS Models with Binary Dependent Variables and Fixed Effects. Working Paper Series, New York University.
Berger, A. N., Demirgüç-Kunt, A., Levine, R., & Haubrach, J. G. (2004). Bank Concentration and Competition: An Evolution in the Making. Journal of Money, Credit, and Banking, 36(3), 433-451.
Bikker, J. A., & Haaf, K. (2002). Competition, Concentration and their Relationship: An Empirical Analysis of the Banking Industry. Journal of Banking & Finance, 26(11), 2191-2214.
78
Borghei-Ghomi, Z., & Leung, P. (2013). An Empirical Analysis of the Determinants of Greenhouse Gas Voluntary Disclosure in Australia. Accounting and Finance Research, 2(1), 110-127.
Braam, G., & Borghans, L. (2010). Board and Auditor Interlocks and Voluntary Disclosure in Annual Reports. Working Paper Series, Radboud University Nijmegen.
CBI. (2014). Electronics and Electrical Engineering > Competition. Retrieved on June 24, 2014, from http://www.cbi.eu/marketintel_platform/electronics-and-electrical-engineering-/177469/competitiveness
Cetorelli, N. (1999). Competitive Analysis in Banking: Appraisal of the Methodologies. Economic Perspectives, 23(1), 2-15.
Chakroun, R., & Matoussi, H. (2012). Determinants of the Extent of Voluntary Disclosure in the Annual Reports of the Tunisian Firms. Accounting and Management Information Systems, 11(3), 335-370.
Chau, G. K., & Gray, S. J. (2002). Ownership Structure and Corporate Voluntary Disclosure in Hong Kong andSingapore. The International Journal of Accounting, 37(2), 247-265.
Cheynel, E. (2013). A Theory of Voluntary Disclosure and Cost of Capital. Review of Accounting Studies, 18(4), 987-1020.
Chun, H., Kim, J. W., Morck, R., & Yeung, B. (2008). Creative Destruction and Firm-SpecificPerformance Heterogeneity. Journal of Financial Economics, 89(1), 109-135.
Citro, F. (2013). Disclosure Level Evaluation and Disclosure Determinant Analysis: A Literature Review. ScieConf, 1(1), 198-203.
Coebergh, P. H. (2011). Voluntary Disclosure of Corporate Strategy: Determinants and Outcomes – An Empirical Study Into the Risks and Payoffs of Communicating Corporate Strategy. Working Paper Series, University of Applied Sciences Leiden.
Consumer Postal Council. (2014). Postal Freedom Index: Netherlands. Retrieved on June 24, 2014, from http://www.postalconsumers.org/postal_freedom_index/Netherlands_-_TNT_Post.shtml
Cooke, T. E. (1989). An Empirical Study of Financial Disclosure by Swedish Companies. New York: Garland Publishing.
Crossman, A. (2014). Logistic Regression: An Overview. Retrieved on July 7, 2014, from http://sociology.about.com/od/Statistics/a/Logistic-Regression.htm#
Daboo, J. (2013). Audit Reports that Make a Difference. Retrieved on May 9, 2014, from http://www.kpmg.com/uk/en/issuesandinsights/articlespublications/pages/audit-reports-that-make-difference.aspx
Dahawy, K. (2009). Company Characteristics and Disclosure Level: The Egyptian Story. International Research Journal of Finance and Economics, 34, 194-208.
DeCarlo, L. T. (1997). On the Meaning and Use of Kurtosis. Psychological Methods, 2(3), 292-307.
79
Demsetz, H. (1968). The Cost of Transacting. Quarterly Journal of Economics, 82(1), 33-53.
Dhouibi, R., & Mamoghli, C. (2013). Determinants of Voluntary Disclosure in Tunisian Bank’s Reports. Research Journal of Finance and Accounting, 4(5), 80-94.
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, Liquidity, and the Cost of Capital. The Journal of Finance, 46(4), 1325-1359.
DiMaggio, P. J., & Powell, W. W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review, 48(2), 147-160.
Domencich, T. A., & McFadden, D. L. (1975). Urban Travel Demand: A Behavioral Analysis. Amsterdam: North-Holland Publishing Company.El-Gazzar, S. M., Fornaro, J. M., & Jacob, R. A. (2006). An Examination of the Determinants and Contents of Corporate Voluntary Disclosure of Management’s Responsibilities for Financial Reporting. Journal of Accounting, Auditing & Finance, 23(1), 95-114.
Euronext. (2014). ICB Classification. Retrieved on May 28, 2014, from https://europeanequities.nyx.com/nl/icb
Fama, E. F., & French, K. R. (2004). New lists: Fundamentals and Survival Rates. Journal of Financial Economics, 73(2), 229-269.
Fama, E. F., & Jensen, M. C. (1983). Separation of Ownership and Control. Journal of Law and Economics, 26(2), 301-325.
FASB. (2001). Improving Business Reporting: Insights into Enhancing Voluntary Disclosures. Retrieved on May 5, 2014, from http://www.fasb.org/cs/BlobServer?blobkey=id&blobwhere=1175819611134&blobheader=application%2Fpdf&blobcol=urldata&blobtable=MungoBlobs
Financial Reporting Council (2013). FRC Issues Revised Auditing Standard: Making Auditors’ Work More Transparent to Investors. Retrieved on March 4, 2014, from https://www.frc.org.uk/News-and-Events/FRC-Press/Press/2013/June/FRC-issues-revised-auditing-standard-Making-audito.aspx
Financial Times Lexicon. (2014). Definition of Free Float. Retrieved on June 24, 2014, from http://lexicon.ft.com/term?term=free-float
Friedman, M. (1953). The Methodology of Positive Economics. Essays in Positive Economics. Chicago: University of Chicago Press, 3-43.
Gamerschlag, R., Möller, K., & Verbeeten, F. (2011). Determinants of Voluntary CSR Disclosure: Empirical Evidence From Germany. Review of Managerial Science, 5(2-3), 233-262.
Gassen, J., & Sellhorn, T. (2006). Applying IFRS in Germany: Determinants and Consequences. Betriebswirtschaftliche Forschung und Praxis, 58(4), 1-38.
Golden, B. R., & Zajac, E. J. (2001). When will Boards Influence Strategy: Inclination Times Power Equals Strategic Change. Strategic Management Journal, 22, 1087-1111.
80
Goodstein, J., Gautam, K., & Boeker, W. (1994). The Effects of Board Size and Diversity onStrategic Change. Strategic Management Journal, 15, 241-250.
Graham, J. R., Harvey, C. R., & Rajgopal, S. (2005). The Economic Implications of Corporate Financial Reporting. Journal of Accounting and Economics, 40(1-3), 3-73.
Hair, J. F., Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995). Multivariate Data Analysis, 3rd edition. New York: Macmillan Publishers.
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 and Economics, 31(1-3), 405-440.
Ho, P. L. (2009). Determinants of Voluntary Disclosure by Malaysian Listed Companies Over Time. Working Paper Series, Curtin University.
Hossain, M., & Raez, M. (2007). The Determinants and Characteristics of Voluntary Disclosure by Indian Banking Companies. Corporate Social Responsibility and Environmental Management, 14(5), 274-288.
Hossain, M., & Hammami, H. (2009). Voluntary Disclosure in the Annual Reports of an Emerging Country: The Case of Qatar. Advances in Accounting, Incorporating Advances in International Accounting, 25(2), 255-265.
IAASB. (2013a). Proposed International Standard on Auditing (ISA) 701: Communicating Key Audit Matters in the Independent Auditor’s Report. Retrieved on March 28, 2014, from http://www.ifac.org/sites/default/files/publications/files/Proposed%20ISA%20701%20(Revised)-final.pdf
IAASB. (2013b). Reporting on Audited Financial Statements: Proposed New and Revised International Standards on Auditing (ISAs). Retrieved on March 28, 2014, from https://www.ifac.org/sites/default/files/publications/files/Complete%20ED,%20Reporting%20on%20Audited%20Financial%20Statements.pdf
IAASB. (2014). About IAASB. Retrieved on April 1, 2014, from http://www.ifac.org/auditing-assurance/about-iaasb
IFRS (2013). Staff Paper IASB Meeting: Time Frame for an Assessment of Going Concern. Retrieved on June 11, 2014, from http://www.ifrs.org/Meetings/MeetingDocs/IASB/2013/March/03B-Proposed%20narrow-focus%20amendment%20to%20IAS%201.pdf
Institute for Digital Research and Education. (2014a). Lesson 3: Logistic Regression Diagnostics. Retrieved on July 7, 2014, from http://www.ats.ucla.edu/stat/stata/webbooks/logistic/chapter3/statalog3.htm
Institute for Digital Research and Education. (2014b). Stata Data Analysis Examples: Logistic Regression. Retrieved on July 22, 2014, from http://www.ats.ucla.edu/stat/stata/dae/logit.htm
Institute for Digital Research and Education. (2014c). Stata Annotated Output: Ordered Logistic Regression. Retrieved on July 8, 2014, from http://www.ats.ucla.edu/stat/stata/output/stata_ologit_output.htm
Jensen, M., & Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, AgencyCosts, and Ownership Structure. Journal of Financial Economics, 3(4), 305-360.
81
Kateb, I. (2012). An Analysis of the Determinants of Voluntary Structural Capital Disclosure by Listed French Companies. International Journal of Business and Management, 7(11), 95-110.
Kennedy, P. (2003). A Guide To Econometrics, 5th edition. Cambridge, MA: MIT Press.
Khlifi, F. & Bouri, A. (2010). Corporate Disclosure and Firm Characteristics: A Puzzling Relationship.Journal of Accounting, Business & Management, 17(1), 62-89.
Kolsi, M. C. (2012). The Determinants of Corporate Voluntary Disclosure: Evidence From the Tunisian Capital Market. The IUP Journal of Accounting Research & Audit Practices, 11(4), 49-68.
Kolsi, M. C., & Zehri, F. (2013). The Determinants of IAS/IFRS Adoption by Emergent Countries. Working Paper Series, Emirates College of Technology.
Lan, Y., Wang, L., & Zhang, X. (2013). Determinants and Features of Voluntary Disclosures in the Chinese Stock Market. China Journal of Accounting Research, 6(4), 265-285.
Laudato, M. (2012, October 26). Going Concern. ACCA. Retrieved on June 11, 2014, from http://www.accaglobal.com/gb/en/member/cpd/auditing/cpd-articles/concern-article.html
Long, J. S., & Ervin, L. H. (2000). Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model. The American Statistician, 54(3), 217-224.
Lu, X., & White, H. (2014). Robustness Checks and Robustness Tests in Applied Economics. Journal of Econometrics, 178(1), 194-206.
McFadden, D. L. (1977). Quantitative Methods for Analyzing Travel Behaviour of Individuals: Some Recent Developments. Published in Behavioral Travel Modelling (Hensher, D, & Stopher, P. (eds.)), 279-318. Croom Helm, 1979.
Meek, G. K., Roberts, C. B., & Gray, S. J. (1995). Factors Influencing Voluntary Annual Report Disclosures by U.S., U.K. and Continental European Multinational Corporations. Journal of International Business Studies, 26(3), 555–572.
Oxelheim, L., & Thorsheim, M. (2012). Market Determinants of Voluntary Disclosure of Macroeconomic Effects on Corporate Performance. Working Paper Series, Lund University.
Park, H. M. (2009). Regression Models for Binary Dependent Variables Using Stata, SAS, R, LIMDEP, and SPSS. Working Paper Series, Indiana University.
Pástor, L., & Veronesi, P. (2003). Stock Valuation and Learning about Profitability. Journal ofFinance, 58(5), 1749-1790.
Petrova, E., Georgakopoulos, G., Sotiropoulos, I., & Vasileiou, K. Z. (2012). Relationship Between Cost of Equity Capital and Voluntary Corporate Disclosures. International Journal of Economics and Finance, 4(3), 83-96.
Piersma, J. (2014, March 3). Accountant Geeft Pijnpunten Jaarrekening Aan in Zijn Verklaring. Het Financieele Dagblad, 11-12.
82
Pohlmann, J. T., & Leitner, D. W. (2003). A Comparison of Ordinary Least Squares and Logistic Regression. Ohio Journal of Science, 103(5), 118-125.
Prado-Lorenzo, J. M., Rodríguez-Domínguez, L., Gallego-Álvarez, I., & García-Sánchez, I. M. (2009). Factors Influencing the Disclosure of Greenhouse Gas Emissions in Companies World-Wide. Management Decision, 47(7), 1133-1157.
Premuroso, R. F. (2008). An Analysis of Voluntary Annual Report Disclosures of Outsourcing: Determinants and Firm Performance. Working Paper Series, Florida Atlantic University.
PwC. (2014a). Klare Taal! Benchmark Controleverklaring ‘Nieuwe Stijl’ onder Nederlandse Beursfondsen. Retrieved on March 10, 2014, from http://www.pwc.nl/nl/assets/documents/pwc-klare-taal.pdf
PwC. (2014b). World Watch: Plain Speaking! – Field-Testing the New Auditor’s Report in Holland. Retrieved in May 9, 2014, from http://www.pwc.com/gx/en/audit-services/corporate-reporting/publications/world-watch/articles/netherlands-plain-speaking.jhtml
Raffournier, B. (1995). The Determinants of Voluntary Financial Disclosure by Swiss Listed Companies. European Accounting Review, 4(2), 261-280.
Roberts, R. W. (1992). Determinants of Corporate Social Responsibility Disclosure: An Application of Stakeholder Theory. Accounting Organizations and Society, 17(6), 595-612.
Schaefer, S. (1998). The Dependence of Pay-Performance Sensitivity on the Size of the Firm. The Review of Economics and Statistics, 80(3), 436-443.
Schiehll, E., Terra, P. R. S., & Victor, F. G. (2013). Determinants of Voluntary Executive Stock Option Disclosure in Brazil. Journal of Management & Governance, 17(2), 331-361.
Sehar, N. U., Bilal, B. & Tufail, S. (2013). Determinants of Voluntary Disclosure in Annual Report: A Case Study of Pakistan. Management and Administrative Sciences Review, 2(2), 181-195.
Shehata, N. F. (2014). Theories and Determinants of Voluntary Disclosure. Accounting and Finance Research, 3(1), 18-26.
Sheu, H. J., Liu, C. L., & Yang, F. J. (2008). The Determinants of Voluntary Disclosure of Directors’ Compensation: Empirical Evidence From an Emerging Market. Working Paper Series, National Chiao Tung University.
Shevlyakov, G. L., & Vilchevski, N. O. (2001). Robustness in Data Analysis: Criteria and Methods. Boston: VSP International Science Publishers.
Shumway, T. (2001). Forecasting Bankruptcy More Accurately: A Simple Hazard Model. Journal of Business, 74(1), 101-124.
Signal Trading Group. (2014). Kurtosis Breakdown: All Futures Markets. Retrieved on July 2, 2014, from http://www.signalfinancialgroup.com/Markets/KurtosisData.php
83
Smith, M. (2011). Research Methods in Accounting, 2nd edition. London: SAGE Publications Ltd.
Spiegel, M. M., & Yamori, N. (2006). Determinants of Voluntary Bank Disclosure: Evidence From Japanese Shinkin Banks. Japan's Great Stagnation: Financial and Monetary Policy Lessons for Advanced Economies, 103-127. Cambridge: MIT Press.
Studenmund, A. H. (2010). Using Econometrics: A Practical Guide, 6th edition (Pearson Series in Economics). Upper Saddle River, NJ: Prentice Hall.
Suchman, M. C. (1995). Managing Legitimacy: Strategic and Institutional Approaches. The Academy of Management Review, 20(3), 571-610.
Sukcharoensin, S. (2012). The Determinants of Voluntary CSR Disclosure of Thai Listed Firms. International Proceedings of Economics Development & Research, 46(12), 61-65. Thomson One. (2014). Corporate Governance and Ownership Summary. Data retrieved on May 17, 2014, from http://banker.thomsonib.com/
United States Census Bureau. (2007). Concentration Ratios. Retrieved on June 7, 2014, from http://www.census.gov/econ/concentration.html
Verrecchia, R. E. (1983). Discretionary Disclosure. Journal of Accounting and Economics, 5(3), 179–194.
Vu, K. B. A. H. (2012). Determinants of Voluntary Disclosure for Vietnamese Listed Firms, 2nd edition. Perth: Curtin University.
Wang, K., Sewon, O., & Claiborne, M. C. (2008). Determinants and Consequences of Voluntary Disclosure in an Emerging Market: Evidence From China. Journal of International Accounting, Auditing and Taxation, 17(1), 14-30.
Watts, R. L., & Zimmerman, J. L. (1978). Towards a Positive Theory of the Determination of Accounting Standards. The Accounting Review, 53(1), 112–134.
X-Rates. (2013). Historic Exchange Rates (Euro). Retrieved on May 28, 2014, from http://www.x-rates.com/historical/?date=2013-12-31
Yermack, D. (1996). Higher Market Valuation of Companies with a Small Board of Directors. Journal of Financial Economics, 40(2), 185-211.
Zhu, D., & Gong, J. (2013). Determinants of Voluntary Disclosure of Realized or Realizable Executive Compensation. Working Paper Series, California State University at Fullerton.
Zunker, T. (2011). Determinants of the Voluntary Disclosure of Employee Information in Annual Reports: An Application of Stakeholder Theory, 1st edition. Gold Coast: Bond University.
84
Appendix
Table 13: The number of significant associations (articles Section 3)Determinant of voluntary disclosure
Number of significantly positive associations
Number of significantly negative associations Total
Table 14: List of sampled firms per indexAEX AMX ASCX Other listed firms Non-listed firmsAegon Aalberts AMG Aareal ABN AMROAhold Accell Acomo Accsys PGGMAkzo Nobel Air France-KLM Ballast Nedam AjaxArcelorMittal Aperam BE Semiconductor AND InternationalASML Arcadis Beter Bed Atrium EUR Real
Pharming TKH Group Macintosh Value8 New Sources Energy
USG People SnowWorld Aareal RoodMicrotecVopak Atrium EUR Real Estate TIE KinetixAMG Bever Holding UNIT4Ballast Nedam Boussard & GavaudanDPA Brookfield Asset
ManagementGrontmij EurocastleHeijmans HAL Trust UnitHES Beheer KardanKendrion Leo Capital GrowthNedap Source GroupNeways Van LanschotOranjewoud Volta Finance
87
Accsys Yatra CapitalBatenburg Techniek
ABN AMRO
Hydratec PGGMRoto Smeets Group
* Examples of the main activities of firms in each sector Basic Materials: Oil equipment and services, specialty chemicals, biotechnology. Industrials: Heavy construction, industrial transportation, industrial machinery, electronic equipment. Consumer goods: Leisure goods, food producers, furnishings. Consumer services: Telecommunications, media, airlines. Financials: Banks, asset management, investment services. Technology: Hardware and equipment, software and computer services, semiconductors.