1
Voluntary purchase of audit reviews? The role of national
culture
ABSTRACT:This study investigates if the decision to voluntarily purchase an audit review of quarterly financial statements (review), is determined by national culture. I argue that a review is purchased as a result of the increased demand for assurance stemming from how national culture affects perceptions of agency problems. Drawing on a sample of 2,975 firm-year observations for 1,232 non-financial listed firms from 22 countries, during the period 2004-2014, I find that national culture acts as a determinant of purchasing a review. Specifically, I find that collectivistic cultures are more likely to purchase a review, as agency problems are increased resulting from corruption and nepotism issues. Moreover, I find that in high uncertainty-avoidant cultures the reviews are less prevalent, given substituting mechanisms. Lastly, I find that the demand for reviews is larger in short-term oriented cultures, as the focus on short-term gains leads to agency problems. This study is the first cross-country in this field, and contributes to the voluntary literature by showing that national culture represents a determinant of the voluntary decision to purchase a review.
Keywords: quarterly financial statements, audit review, national culture
Author: Ria de Jong
S-number: S2565900
Address: Terpwei 3a, 9114 AB Driezum
Telephone: 06-30443980
E-mail: h.l.de.jong.1@student.rug.nl
Word count: 10,025
2
Voluntary purchase of audit reviews? The role of national culture
I. INTRODUCTIONThe demand for financial reporting and disclosure arises due to information asymmetry and agency conflicts between managers and firm stakeholders (Jensen and Meckling, 1976; Watts and Zimmerman, 1983; Healy and Palepu, 2001). One possibility to add credibility to financial statements prepared by management is to hire external auditors (Minnis, 2011). External auditors get increasing responsibilities due to tighter audit requirements following
major accounting scandals.1 Currently, external auditors are more involved in the verification
of financial statements relative the state of play before the downturn. This also applies regarding to quarterly financial statements ((International Standard on Review Engagements (ISRE) 2410, 2010)). In the United States (US, hereafter) the Securities and Exchange Commission (SEC) mandated in 2000 that quarterly financial statements of listed companies have to be
reviewed by an external auditor on a timely basis.2 In other countries firms can voluntarily
choose to purchase an audit review of quarterly financial statements (review), because in countries such as the Netherlands, Germany, Canada and the United Kingdom (UK, hereafter) reviews are not mandated. As explained by Boritz and Liu (2006), this decision should depend on benefits outweighing the costs of performing a review. However, it is not clear if the benefits outweigh the costs, given extant empirical evidence. In the literature several researchers have found mixed evidence, especially with regard to the benefits (Ettredge, Simon, Smith, and Stone, 2000; Manry, Tiras, and Wheatley, 2003; Bédard and Courteau, 2015; Kajüter, Klassmann, and Nienhaus, 2016). A potential reason for the mixed findings of previous research is that the authors did not consider the variability in country characteristics for the decision to purchase a review (Filip, 2016).
For example, a factor which is not investigated is the influence of national culture. According to Hofstede (2001) people in different cultures do not think, feel, and act the same, which has consequences for their beliefs, attitudes, and skills. In turn, these differences can also
play a role in making financial decisions (Aggarwal, Faccio, Guedhami, and Kwok, 2016).3
1 Examples of major accounting scandals are the scandals as Enron and Worldcom, which are part of the top 10
worst accounting scandals of all time (http://www.accounting-degree.org/scandals/).
2 In the US the review of quarterly financial statements is called a timely review. Before the introduction of the
interim review requirement by the SEC (1999), companies in the US had to perform a retrospective review and could choose to perform a timely review. This retrospective review differed from a timely review as an auditor performed a retrospective review at each year-end audit, while a timely review was performed at each quarter-end audit.
3 According to Aggarwal et al. (2016) perceptions, preferences, and behaviours are influenced by culture, and
3
Consequently, national culture directly influences decisions and it shapes decisions and information costs through institutions. In the context of this study, national culture is particularly relevant, since it can affect perceptions of agency problems and further impact the
demand for assurance.4 Therefore, national culture is likely to explain why companies in certain
countries are more likely to choose to purchase a review. This paper empirically tests this expectation in an international setting. Specifically, it answers the question to what degree the decision to purchase a review is influenced by dimensions of national culture.
Prior research assessing the decision to purchase a review, focuses on the costs and benefits of this decision. With regard to the costs it is found that a review is associated with additional audit costs (Hoehn, 2013; Bédard and Courteau, 2015). With regard to the benefits it is found that a review is associated with fewer fourth quarter adjustments (Ettredge et al., 2000), a simultaneous association between quarterly returns and earnings (Manry et al., 2003), a signal that earnings quality has been improved (Bandyopadhyay, Boritz and Liu, 2007; Kajüter et al., 2016), and a lower cost of private debt (Porumb and Karaibrahimoglu, 2017). However, there is not documented an association between purchasing a review and earnings quality (Bédard and Courteau, 2015; Kajüter et al., 2016). Although this provides interesting insights, all studies have in common that the sample consists of only one country and my paper comes to extend the literature in this concern. Relative to previous research I focus on an international sample and assess the impact of national culture as a determinant of purchasing a review.
Looking at multiple dimensions of culture, I first expect to find an association between the degree of uncertainty avoidance and the decision to purchase a review. Uncertainty avoidance is defined as the extent of feeling either comfortable or uncomfortable in unstructured situations (Hofstede and Bond, 1988). The unstructured situation in light of this paper is that the quarterly financial statements are not being reviewed. People in an uncertainty-avoidant culture want to increase the comfortableness to deal with this situation, what could be achieved by purchasing a review. From the viewpoint of managers, purchasing a review could decrease comfortableness as a review may be accompanied by negative consequences resulting from disclosing implausible information. Moreover, Carrington and Catasús (2007) state that the degree of audit work to finalise an audit assignment is associated with the degree of comfort of the audit team members. Therefore, it is likely that auditors put more effort when performing a review in an uncertainty-avoidant culture, relative to an uncertainty-accepting culture. The
4
additional efforts increase the risk for managers that a review may be accompanied by potential negative consequences. In turn, this tendency of the management to avoid purchasing a review increases the uncomfortableness of stakeholders as this avoidance increases information-asymmetry. Therefore, it is likely that stakeholders demand to purchase a review to increase the level of comfort regarding the reliability of financial statements.
However, it is also possible that a higher degree of uncertainty avoidance is associated with a lower demand for assurance. This lies in the fact that Hofstede (2001) notes that uncertainty-avoidant cultures require additional mechanisms to cope with uncertainty. It is possible therefore that there exist substitutes for a review, and consequently the purchase of a review has no additional value for high uncertainty-avoidant cultures (Hooghiemstra, Hermes, and Emanuels, 2015). Based on these arguments, both a positive as well as a negative association is possible. My findings suggest that there exists a significantly negative association between uncertainty avoidance and the decision to purchase a review. In other words, I find that in high uncertainty-avoidant cultures, there exist substitutes for a review and therefore purchasing a review has no value for these cultures. Regarding the level of disclosure on internal controls, a similar association is found where additional bonding mechanisms substitute for the level of disclosure on internal controls (Hooghiemstra et al., 2015).
Second, I expect to find a positive association between the degree of individualism and the decision to purchase a review. Individualism is defined as the degree of integration of individuals into groups (Hofstede and Bond, 1988). As individualistic managers want to pursue their own goals as a consequence of not being integrated in a group, this can increase agency costs and can be a reason to purchase assurance. However, it can also be attractive for managers to purchase a review and build a reputation for engaging in credible reporting (Graham, Harvey, and Rajgopal, 2005). Therefore, I expect that it is more likely that a review is purchased when a country is characterized as individualistic. Contrary to my expectation, I find a significantly negative association. This means that collectivistic societies are more likely to purchase a review. According to Kyriacou (2016) this is not unlikely, as he argues that in-group favouritism leads to issues as corruption and nepotism which exacerbates agency problems.
Third, I expect to find a positive association between short-term oriented cultures and the decision to purchase a review. When a short-term orientation prevails, the managers’ focus is on the present and past. In this situation it could be attractive to sacrifice profitability in the long term in favour of profits in the short term. These short-term gains are at the expense of longer-term value creation, which constitutes an agency problem as this is not in line with the goals of most principals who prefer a long-term orientation (Jensen and Murphy, 1990; Jensen,
5
2008; Martin, Wiseman, and Gomez-Mejia, 2016). Therefore, it is likely that agency problems are more severe in a short-term oriented culture than in a long-term oriented culture. As a consequence, the demand for the purchase of a review is increased. I therefore expect to find a positive association between short-term oriented cultures and the decision to purchase a review. I find corroborating evidence for this expectation, as I find a significantly negative association between long-term oriented cultures and the decision to purchase a review.
In order to test if cultural dimensions are associated with the voluntary decision to purchase a review, I draw on a sample which comprises 2,975 firm-year observations for 1,232 non-financial listed firms from 22 countries, during the period 2004-2014 in a voluntary setting. This setting is chosen as it is most times required to issue quarterly reports during this period, however it is voluntary to purchase a review of quarterly financial statements in these countries. Financial listed firms are excluded because these firms are subject to special regulation (Bédard and Courteau, 2015; Kajüter et al., 2016), as well as firms that have no review information or ambiguous information, and firms that have other missing needed information. In line with Hooghiemstra et al. (2015), I use three measures from Hofstede (2001) to investigate my research question. To examine the dependent variable, I rely on hand-collected panel data and for the main independent variables I rely on research of Hofstede (2001) and research of Hofstede, Hofstede, and Minkov (2010). Subsequently, the model is tested by means of a PROBIT regression.
This research contributes to the literature that discusses the costs and benefits of purchasing a review (Ettredge et al., 2000; Manry et al., 2003; Hoehn, 2013; Bédard and Courteau, 2015; Kajüter et al., 2016; Porumb and Karaibrahimoglu, 2017). While previous researchers use a sample drawn of one country, I am the first to my knowledge to investigate the association between national culture and the decision to voluntarily purchase a review. This cross-country study adds to the accounting academic literature, as it shows that a determinant of voluntarily purchasing a review is formed by national culture. The findings in my paper could also explain the mixed findings found in previous research, by suggesting that a clear understanding of the decision to purchase a review could be obtained by assessing the impact of national culture. Moreover, this study responds to the call of Filip (2016) to study the voluntary decision to purchase a review in an international context. Furthermore, the findings in my paper contribute to the literature regarding national culture. While previous literature has investigated various topics in relation with national culture (Aggarwal et al., 2016; Beugelsdijk, Kostova, and Roth, 2017), the association between reviews and national culture has not yet been investigated. Additionally, these findings are of interest to regulators. A critique of
6
regulators is that empirical evidence on the effects of reviews is scarce (Bédard and Courteau, 2015). This research provides regulators with empirical evidence which shows that they need to be aware of the effects of national culture in an international setting. These insights can contribute to the debate to mandate reviews or not.
The remainder of the paper is structured as follows. Section II outlines the institutional background. Subsequently, section III outlines the theoretical background and section IV reflects the development of the hypotheses. Next, the research design will be introduced in section V. In section VI the results will be presented, and section VII concludes.
II. INSTITUTIONAL BACKGROUND
The legal systems of countries are not the same, which is also true with regard of financial reporting requirements. A leading country in this respect is the United States. After the introduction of the SEC, the filing of annual reports was mandated in 1934 and the requirement of semi-annual reporting and quarterly reporting followed respectively in 1955 and 1970 (Butler, Kraft, and Weiss, 2007). More recently, the US also has mandated the review of these quarterly reports (SEC, 1999). Although firms located in the US have no freedom of choice, companies in other countries are not always required to issue quarterly reports and/or to purchase a review of these reports.
This is for example shown by the lack of consensus in Europe. The Transparency Directive (2004/109/EC) required that issuers of securities traded on regulated markets within Europe had to disclose financial information on a yearly, a half-yearly and a quarterly basis (European Commission, 2013). However as described by the European Commission (2013), the quarterly reporting requirement has been abolished in 2013. The reasons for this abolishment were to reduce the administrative burden accompanied with quarterly reporting and to encourage long-term investments. Listed companies in Europe can voluntary choose to disclose quarterly financial statements. When assessing the Asian setting, there is also a lack of consensus about the requirement of quarterly reporting. In countries like Singapore, Malaysia and Thailand quarterly reporting is mandated for the larger listed companies, but in countries like Jakarta and Hong Kong quarterly reporting is not mandated (Campbell-Noë, 2015).
Next to the lack of consensus amongst regulators about whether to mandate the issuance of quarterly reports or not, there is also no consensus amongst these regulators about the requirement to purchase a review of these quarterly reports. Where a review is mandatory in the US and Australia, these reviews are not mandated in other countries like Germany, The
7
Netherlands, and the United Kingdom.5 However, some firms in these latter countries do
purchase a review, even if they are not required to do so. This is an interesting issue and therefore a cross-country study is needed to investigate why there is a lack of consensus about mandating or voluntarily purchasing a review.
Characteristics of the review
To investigate the lack of consensus about mandating a review or not, consideration should be given to the fact that a review differs from a full year end-audit in several respects as described by ISRE 2410 (2010). Quarterly financial information can be defined as financial information that is prepared and presented in accordance with an applicable framework for financial reporting, which consists of a complete or a condensed set of financial statements for a period that is shorter than the financial year of the entity (ISRE 2410, 2010). So, quarterly information is reported on a more frequent basis as full year end-audit information. Underlying this is the fact that there is made use of more estimates and quarterly reports contain forward-looking statements as opposed to full year end-audits. This implies that these statements involve risk and uncertainty, as many factors can cause that corrections have to be made to these statements as also stated by firms in their quarterly reports (Philips, 2016).
Moreover, the objectives of a full year end-audit and a review differ. Where the auditor expresses an audit opinion in the form of a positive statement in case of full year end-audits (reasonable assurance), the auditor expresses a conclusion in the form of a negative statement in case of a review (limited assurance). Underlying this difference in the level of assurance given, is that the scope of a review is less than a full year end-audit. As the objective of a review is not to corroborate evidence concerning significant accounting matters which relates to the quarterly financial information, a review does not include such tasks as inspection, observation and/or confirmation. Evidence for a review is mainly gathered by performing inquires, and performing analytical and other review procedures. Because a review is less in scope, a review does not provide sufficient evidence to conclude that the financial statements are free from material errors as done with full year-end audits. The auditor therefore does not express an audit opinion, but expresses a conclusion on the financial statements. As a consequence, Kajüter et al. (2016) state that a review can be seen as a device that verifies the plausibility of the reported numbers in the quarterly financial statements.
5 According to Bédard and Courteau (2015) a review is also voluntary in Canada, but Canada has considered to
8
Although reviews require less assurance work than full year end-audits, it is likely that a review increases monitoring. This lies in the fact that a review enables the auditor to detect and signal potential financial reporting misstatements in a timely manner (Ettredge et al., 2000). This relevant and timely information can be communicated to the managers and the audit committee to solve potential reporting misstatements. So, a review is likely to be used as a monitoring device, which is not possible with a full year end-audit.
III. THEORETICAL BACKGROUND
There is a lack of consensus amongst regulators about mandating or not mandating reviews. Because purchasing a review is a voluntary decision in some countries, it is possible to investigate what drives a firm to voluntarily purchase a review. The starting point is that firms decide to purchase a review when the firm associates this review with more benefits than costs (Kajüter et al., 2016). That is why researchers have investigated the associated costs and benefits of the decision to purchase a review, most times in a voluntary setting.
Kajüter et al. (2016) give some insights into the costs for the firm concerning the purchase of a review. First, a review is associated with non-trivial out-of-pocket costs. The reason for this is that audit firms provide additional services. A study of Bédard and Courteau (2015) has tried to quantify the additional audit fee costs of a review. It is found that the additional audit costs are approximately 18 percent based on a Canadian sample of listed firms. Another research performed by Hoehn (2013) also tried to quantify these costs and found for a German sample an increase of approximately 15 percent. Next to these costs, a review can increase the reporting lag because it takes time to give assurance. This in turn can act as a bad signal to investors as supported by Kajüter et al. (2016). Additionally, the reporting flexibility of management is compromised because the auditor is involved with the financial statements for the entire year (Kajüter et al., 2016).
With regard to the benefits, findings are not that clear-cut. An often cited benefit of reviews is the potentially enhanced credibility of the disclosed information. As explained by Kajüter et al. (2016) this enhanced reliability can help to reduce agency problems (as well external as internal agency problems). Early research concerning this statement by Ettredge et al. (2000) and Manry et al. (2003) is concerned with a comparison between retrospective and timely reviews. Ettredge et al. (2000) find that companies that perform a timely review have fewer adjustments in the fourth quarter as compared with companies that perform a retrospective review. Manry et al. (2003) investigate the association between quarterly returns and earnings and find that this association is predominantly contemporaneous for timely
9
reviews, and not so much for retrospective reviews. Both papers have drawn on a sample of firms located in the US, a country which is characterised by heavy regulation. That is why more recent research has looked to a voluntary setting for the decision to purchase a review. Bédard and Courteau (2015) investigate if the quality improves when purchasing a review. These researchers find no association between purchasing a review and earnings quality, based on different measures of abnormal accruals. Kajüter et al. (2016) attribute this to the fact that earnings quality is an indirect consequence of reviews. These authors therefore rely on a measure which is a more direct consequence of a review. Specifically, Kajüter et al. (2016) rely on capital market consequences to investigate if there are investor reactions taking place upon the release of quarterly financial statements for a German sample. In other words, it indicates if quarterly financial statements have information content. However, Kajüter et al. (2016) also do not find an association between a review and earnings quality as measured by assurance value. As opposed to assurance value, the researchers do find an association with regard to signalling value. This indicates that earnings quality does not have to improve per se, but that users expect the information to be of higher quality as a consequence of this signal (Bandyopadhyay et al., 2007).
The underpinning theory: agency theory
A theory which can explain that some firms voluntarily purchase a review, even if they are not required to do so is the agency theory. This theory is centred around the existence of two agency problems, which could influence the decision of firms to purchase a review (i.e.
assurance).6
The first agency problem referred to is that of the principal-agent problem, based on the notion of risk-sharing (Eisenhardt, 1989). To achieve like-minded goals between principals and agents, the principal (the owner or owners) delegate(s) some decision making authority to agents (the managers) to perform some service on their behalf, which can be defined as a contract according to Jensen and Meckling (1976). These authors describe this as an agency relationship. The goal of this delegation is that agents perform in line with the best interests of the principal. However, when both parties maximize utility it is likely that the agent will not always perform in line with the best interests of the principal (Jensen and Meckling, 1976). As
6 The existence of agency problems could influence the decision of firms to purchase a review, as assurance is a
resource to reduce agency problems (Simnett et al., 2009). The article of Kajüter et al. (2016) relies on a similar line of reasoning, as illustrated by the fact that the benefits of a review are interpreted as an alleviation of agency problems.
10
a consequence of this divergence, the expected agency costs based on the initial contract will change (Fama, 1980). From the viewpoint of the principal the contract is now accompanied by more risks, which constitutes the principal-agent problem.
The second agency problem referred to, monitoring the behaviour of agents, is related to the first. To align the interests of both parties, principals make use of incentives (Jensen and Meckling, 1976). Fama and Jensen (1983) suggest that when agents have equity in the firm, their interests are more aligned with the interests of principals. But when there exists a perceived inequity, the result is that agents will engage in self-interested behaviour. Such behaviour creates information asymmetries when the principal is not able to monitor this behaviour properly. Therefore, the second agency problem is that of monitoring the behaviour of agents. It can be concluded that the two problems are linked, because the divergence of interests creates information asymmetries, and this makes it harder for principals to monitor the behaviour of the agent. That is why agency theory is inextricable linked to information asymmetry.
It is more likely that public companies will resolve their information asymmetry between insiders and outsiders by means of public financial statements, relative to private companies (Ball and Shivakumar, 2005). According to these authors private companies are more likely to use an ‘insider access’ model to resolve information asymmetry, and therefore specific policies such as taxation drive their financial reporting. As opposed to private companies, public companies use financial statements as a resource of contracting activities and use it in both primary as secondary equity transactions. So, the financial statements are likely to be of higher quality in public companies as opposed to private companies.
National culture in previous research – theoretical advances
Over the past 30 years, culture has been conceptualized in multiple ways (Glynn, Giorgi, and Lockwood, 2012; Giorgi, Lockwood, and Glynn, 2015). As suggested by Giorgi et al. (2015) these conceptualizations can be divided into five models; values, stories, frames, toolkits, and categories. These authors define each model and describe what the dominant focus points and who the associated authors are. An example and often cited definition is the one of Hofstede (2001), which is part of the first model. Hofstede (2001) defines culture as ‘‘the collective programming of the mind that distinguishes the members of one group or category of people from another’’. What Hofstede (2001) means with the mind is ‘‘thinking, feeling, and acting, with consequences for beliefs, attitudes, and skills’’. For the purpose of this study, I will use the definition of Aggarwal et al. (2016) which is based on the definition of Hofstede (2001).
11
Aggarwal et al. (2016) define culture as ‘‘an enduring set of beliefs or values that influence individuals’ perceptions, preferences, decisions, and behaviours’’.
IV. HYPOTHESES DEVELOPMENT
As the main benefit of a review is considered to be an alleviation of agency problems (Kajüter et al., 2016), national culture could affect the decision to purchase a review. This lies in the fact that the enduring set of beliefs or values that define a culture, can have an impact on the perception of agency problems and consequently drive a decision (Hooghiemstra et al., 2015; Aggarwal et al., 2016). When agency problems are perceived as less strong, it is more likely that the benefits of purchasing a review will not outweigh the costs and vice versa.
Based on this reasoning, I investigate if there exists an association between national culture and the decision to purchase a review. National culture is measured by three cultural dimensions; (1) uncertainty avoidance, (2) individualism, and (3) long- or short-term orientation. These dimensions are based on Hofstede (2001), and as indicated by Hooghiemstra et al. (2015) the first two dimensions are the most studied ones. Moreover, these dimensions are considered as having an impact on the degree of exacerbation and attenuation of agency problems. I expect that this also holds for the long- or short-term orientation dimension as this dimension can influence the perception of agency problems (Jensen and Murphy, 1990; Jensen, 2008).
The first dimension that is discussed is uncertainty avoidance. Uncertainty avoidance is defined as the extent of feeling either comfortable or uncomfortable in unstructured situations, where unstructured situations are defined as ‘‘novel, unknown, surprising, or different from usual’’ situations (Hofstede and Bond, 1988). People in uncertainty-avoidant cultures want to minimize the chance of experiencing such a situation, as opposed to people who accept the
uncertainty accompanying the unstructured situation.7
When quarterly financial statements are left unaudited, this can also be seen as an unstructured situation. A possible solution to deal with the unstructured situation is to purchase a review of the quarterly financial statements. Although a review has not the same
7 To minimize the chance of experiencing uncomfortableness in an unstructured situation, people in an
uncertainty-avoidant culture are very careful with regard to changes, try to avoid risks, and try to adhere to strict laws and rules (Hofstede and Bond, 1988). Moreover, people in an uncertainty-avoidant culture are feeling uncomfortable with conflict and competition in pursuit of predictability and stability (Hofstede, 1980). Contrary to uncertainty-avoidant cultures, people in uncertainty-accepting cultures are more comfortable in unstructured situations. Therefore, these people have less problems with changes and risks, and want to minimize laws and rules. Moreover, these people are more likely to accept competition and are more comfortable with conflicts (Hofstede, 1980).
12
characteristics as a full year end-audit, it is possible to provide a conclusion about the plausibility of the reported numbers (ISRE 2410, 2010; Kajüter et al., 2016). This additional
assurance is probably seen as a decrease in comfortableness from the viewpoint of managers.8
Moreover, Carrington and Catasús (2007) state that auditors are looking for a state of comfort
when giving an audit opinion.9 So, the degree of audit work to finalise an audit assignment is
associated with the degree of comfort of the audit team members. Audit team members operating in uncertainty-avoidant cultures are likely to invest more effort in the pursuit of feeling comfortable enough with the reported numbers to provide an audit opinion. These additional efforts increase the risk for managers that a review may be accompanied by potential negative consequences. Therefore, it is likely that managers have a tendency to avoid purchasing a review. However, stakeholders see this tendency as a situation of increased information-asymmetry and therefore look for a manner to reduce the uncomfortableness with the unaudited quarterly financial statements. It is therefore likely that stakeholders demand to purchase a review to increase their comfortableness.
However, there also exists another line of reasoning based on Hooghiemstra et al. (2015). It is also possible that the demand to purchase a review by stakeholders is less, as a consequence of additional mechanisms that substitute for the assurance provided by reviews. As Hofstede (2001) states that people in high uncertainty-avoidant cultures require that firms implement additional mechanisms to deal with uncertainty, it is likely that a review has no additional value for these societies as other mechanisms already are in place to cope with uncertainty accompanying the quarterly financial statements. Based on the aforementioned arguments, a higher degree of uncertainty avoidance can be associated with both a higher as a lower demand for assurance. Therefore, I formulate the following non directional hypothesis:
H1: There exists an association between uncertainty-avoidant cultures and the decision to purchase a review
The second dimension that is discussed is that of individualism (versus collectivism). This dimension can be described as the degree of integration of individuals into groups
8 Additional assurance can be seen as a decrease in comfortableness by managers, as a review can reveal that the
disclosed information is not plausible (Kajüter et al., 2016). When information is implausible managers can face potential negative consequences such as lawsuits or situations of conflict, which is at the expense of a predictable and stable situation.
9 A state of comfort means that audit members are comfortable enough with the reported numbers to give an audit
opinion (Carrington and Catasús, 2007). This state can also be described as a situation where sufficient efforts have been invested that makes it possible to give an audit opinion according to Carrington and Catasús (2007).
13
(Hofstede and Bond, 1988). In an individualistic culture, where there is a low integration into groups, the prominent view is that the person is centralized at the expense of society (Oyserman,
Coon, and Kemmelmeier, 2002).10 This is in contrast to collectivistic cultures, where
individuals are more integrated into groups and the in-group interests are centralized (Hofstede
and Bond, 1988; Schwartz, 1990).11
It is possible that there exists an association between the degree to which individuals are integrated into groups and the decision to purchase a review. This could be possible as in an individualistic culture, managers want to pursue personal goals. However, the pursuit of personal goals puts a damper on the agency relationship between the manager(s) and the principal(s). The principals can have doubts about the question if agents act in the best interests of these principals when agents try to achieve their own goals. This in turn can heighten agency costs and can be a reason to purchase assurance for firms. However, the achievement of personal goals by managers can also be a reason for aligning the interests of the principals and agents. Purchasing assurance can also show that a manager indeed discloses credible information (Graham et al., 2005). This in turn will heighten the reputation of the manager as a credible discloser (Hooghiemstra et al., 2015), and can be seen as an achievement of a better status,
which is at the heart of individualism.12 In short, this indicates that it is likely that individualism
is associated with a higher demand for the purchase of a review, by the stakeholders or by the managers themselves. Based on these arguments, I formulate the following hypothesis:
H2: There exists a positive association between individualism and the decision to purchase a review
10 As explained by Hofstede and Bond (1988), in a society where individuals are not that integrated into groups
there is a preference for a loosely-knit social framework. This means that people only look at themselves and their immediate family. Moreover, in individualistic societies personal autonomy and self-fulfilment is valued and people base their identity on their personal accomplishments (Hofstede, 1980). This is also reflected in the definition of Schwartz (1990), where individualistic societies are characterized by narrow primary groups and secondary social relations which are negotiated, based on specific obligations and expectations focused on the achievement or modification of one’s status.
11 In a culture where individuals are more integrated into groups there is a preference for a tightly-knit social
framework. These societies can be referred to as collectivistic and differ from individualistic societies, because in these societies in-group members continue to take care of the individuals in exchange for unquestioning loyalty (Hofstede and Bond, 1988). Schwartz (1990) describes it as a characterization of extended primary groups, in which individuals have diffuse mutual obligations and expectations, which are largely based on their enduring ascribed statuses. As opposed to the centralization of the person, in these societies the in-group interests are centralized.
12 In contrast, in collectivistic cultures the focus is on common goals and cohesion. Therefore, it is likely that
agency problems are in this situation not that strong and therefore the demand by stakeholders for purchasing a review is not that strong. As a consequence, it is likely to be less attractive for firms to purchase assurance as this is most profitable when there are substantial agency costs.
14
The third dimension discussed is that of long-term orientation versus short-term
orientation, which is based on the Chinese Value Survey.13 The basic values comprising this
survey can be assigned to two categories (Hofstede and Bond, 1988; Hofstede and Minkov, 2010). The first category includes the values; (1) persistence, (2) thrift, (3) ordering relationships by status and observing this order, and (4) having a sense of shame. The second category includes the following values; (1) reciprocation of greetings, favours and gifts, (2) respect for tradition, (3) protecting one’s face, and (4) personal steadiness and stability (Hofstede and Bond, 1988; Hofstede and Minkov, 2010). As the values in the first category can be associated with an orientation toward the future and the values in the second category can be associated with an orientation toward the present and past, Hofstede (1991) introduced a new dimension referred to as long-term versus short-term orientation. This dimension has been developed by the years (Hofstede, 1991; 1994; 2001). As noted by Bearden, Money, and Nevins (2006), a short-term or long-term orientation is now more used as a distinction between the
tendency to focus on the ‘here and now’, versus a more holistic view of the future and the past.14
A short- or long-term orientation can have consequences for the relation between agents and principals. When managers’ focus is on the here and now, it can be attractive for managers to sacrifice profitability in the long term in favour of profits in the short term, which is motivated by private gains and temporal preferences. This phenomenon, referred to as short-termism, constitutes another agency problem (Jensen and Murphy, 1990; Jensen, 2008). This is because a preference for short-term gains, is at the expense of longer-term value creation and that is not
in line with the goals of most principals (Martin et al., 2016).15 So, in countries which can be
characterized by a short-term outlook, it is likely that agency problems are more severe than in long-term oriented countries. The demand by stakeholders for a review is larger in these countries and as a consequence it is more likely that firms purchase a review when a short-term orientation prevails than when a long-term orientation prevails. Therefore, I formulate the following hypothesis:
13 Uncertainty avoidance and individualism versus collectivism, together with power distance and masculinity
versus femininity, were part of the originally developed cultural dimensions by Hofstede (1983; 2001). However, there was a concern that the questionnaires used for these results were biased as stated by Hofstede (1983). Respondents in non-Western countries were invited to answer Western questions, but it was found that the way people think could be culturally constrained (Hofstede, 1983; Hofstede and Minkov, 2010). To account for this bias, Bond had asked Asian colleagues to define a list of basic values. Guided by this list Bond designed a questionnaire which was Asian oriented, named the Chinese Value Survey.
14 For example, this distinction is reflected in recent research of Wang, Rieger, and Hens (2016). These researchers
consider a long-term orientation as the extent to which a society has a mentality that is dynamic and future-oriented , measured by thrift compared to respect for tradition (Wang et al., 2016).
15 According to Bolton, Scheinkman, and Xiong (2008), one exception to this argument is that some shareholders
15
H3: There exists a positive association between short-term oriented societies and the decision to purchase a review
V. RESEARCH DESIGN
Sample
The disclosure practices of firms throughout the world differ according to specific country settings. This is illustrated by the fact that a review is not mandated in many countries around the world. However, some firms choose to voluntarily purchase a review (Boritz and Liu, 2006). Therefore it is interesting to investigate if cultural dimensions act as a determinant of the voluntary decision to purchase a review.
I use the Compustat database to select listed firms in countries where the purchase of a review is a voluntary decision for the period 2004-2014. Also, I use the Dealscan database to select companies that have syndicated loans. I choose to do this in order to avoid selecting firms
with heterogeneous incentives to purchase a review.16 I focus on this period as a lot of countries
do require quarterly reporting, but not a review of these reports. The situation in Europe is an example of this, where it was required to issue quarterly reports in the period 2004-2013 (European Commission, 2013). Moreover, a larger sample as a consequence of the cross-country nature, offers the opportunity to study the hypothesized interactions in a larger window, and it increases the power of the empirical tests. Next, I hand-collect information about the question if firms purchase a review. I first searched for the company’s Website, looking for the ‘investor relations’ and ‘public disclosures’ sections to find the quarterly reports. If there was no quarterly report available, I have used the annual report to collect information about a purchased review. To give an opinion on the existence of a purchased review, I used keywords as ‘review’, ‘interim’, ‘quarterly’, ‘audited’, and ‘verified’. In most cases there was a statement from the management or the board if the quarterly reports had been reviewed or not. As a result, the initial sample is comprised of 4,562 firm-year observations. In advance, financial listed firms are already excluded as these firms are subject to special regulation (Bédard and Courteau, 2015; Kajüter et al., 2016). Furthermore, observations are excluded if there was no review
information or ambiguous review information available.17 Also, observations are excluded if
16 According to Porumb and Karaibrahimoglu (2017), a significant incentive to purchase a review is represented
by the existence of private debt. Therefore, in order to focus on a homogeneous sample of banks in the international setting, I eliminate all listed firms without private debt in the Dealscan database.
17 Ambiguous review information was present in some reports, as it was stated that quarterly reports were subject
to a limited review at the side-lines. Further information was missing, so it is not sure if these reviews are similar to a full interim review.
16
the needed information was missing. I refer to Table 1 for a detailed description of the sample selection. Therefore, the final sample consists of 2,975 firm-year observations from 1,231
unique firms.From this final sample, 343 (11,5%) observations have a voluntary review, and
the remaining 2,632 (88,5%) observations have no voluntary review. This distribution is not in line with previous research, as the review firms percentage is respectively 58,6% (Boritz and Liu, 2006), 59% (Bédard and Courteau, 2015), and 36,1% (Porumb and Karaibrahimoglu, 2017) in previous research. However, all these researchers have used a sample based on Canadian firms. As a consequence of the international setting the percentage is likely lower.
< INSERT TABLE 1: Sample selection >
Model
To test if cultural dimensions are associated with the decision to voluntarily purchase a
review, I run the following PROBIT regression:18
REVIEW = α + β1 CULTURE + β2 SIZE + β3 IRTA + β4 MB + β5 LEV + β6 ROA + (1)
β7 BIG4 + β8 YEAR + β9 IND + β10 GDP + ε
< INSERT TABLE 2: Variable definitions >
The dependent variable in this model is constituted by REVIEW. This variable indicates if the quarterly financial statements are reviewed by an auditor or not, measured by a dummy variable coded as respectively 1 or 0. The use of a dummy variable for purchasing a review or not, is in line with previous research (Boritz and Liu, 2006; Bédard and Courteau, 2015; Kajüter et al., 2016; Porumb and Karaibrahimoglu, 2017).
Subsequently, my main variable of interest is constituted by CULTURE. This variable includes the three cultural dimensions used in this paper to measure national culture. The first two cultural dimensions are uncertainty avoidance (UAI) and individualism (IDV). The degree of uncertainty avoidance and the degree of individualism in a country is measured by using the work of Hofstede (2001). This is in line with previous research, that has investigated the association between national culture and voluntary reporting (Han, Kang, Salter, and Yoo, 2010; Hooghiemstra et al., 2015). Although the use of Hofstede’s (2001) measures has also received criticism, it is still a very often used framework today (Kirkman, Lowe, and Gibson,
17
2006; Beugelsdijk, Maseland, and Van Hoorn, 2015; Beugelsdijk et al., 2017). Therefore, I measure uncertainty avoidance and individualism by the framework of Hofstede (2001). I expect that the coefficient for uncertainty-avoidant cultures is significant, but this coefficient can be both positive as negative, as my first hypothesis is a non-directional hypothesis. With regard to individualism, I expect to find a positive and significant coefficient, as my second hypothesis predicts a positive association between individualism and the decision to voluntarily purchase a review.
The third cultural dimension is the existence of a long-term orientation (LTO). To measure this variable I use the work of Hofstede et al. (2010). These researchers have linked the original values from the Chinese Value Survey to areas covered in the World Values Survey (WVS) database. The outcome of this alignment is a quantitative score for long-term orientation for a total of 93 countries. As this study can be characterized as a quantitative study, the study of Hofstede et al. (2010) is a useful source to measure long-term orientation. Other approaches to measure long-term orientation are a survey approach (Wang et al., 2016), or an approach by use of a proxy (Martin et al., 2016). As a consequence of the nature of this study, these approaches are not or less applicable. Therefore, I measure long-term orientation by the framework of Hofstede et al. (2010). I expect to find a negative and significant coefficient for long-term orientation in a country, as my third hypothesis predicts that a short-term orientation is positively associated with the decision to purchase a review.
Additionally to the main variables, I control for several firm-level and country-level variables as previous research suggests that these variables are associated with the decision to purchase a review. First, I control for firm size (SIZE) and the ratio of inventory and receivables to total assets (IRTA). These two variables define the complexity of a firm, and evidence of Boritz and Liu (2006) shows that this complexity is positively associated with the decision to purchase a review. This research also shows that there exists a negative association between growth opportunities and the decision to purchase a review. I control therefore for this possibility, by including the market-to-book ratio (MB). Third, agency costs can be influenced by the external and internal financial power of firms as noted by Kajüter et al. (2016). Higher leverage increases the risk that investor’s money will be lost, and therefore the demand for additional monitoring will be higher (Jensen and Meckling, 1976). To account for this, I include the variable leverage (LEV). However, firms with a higher internal financial power face less agency costs. Also, when firms are more profitable it can heighten their internal financing power. Therefore, I also include profitability measured by return on assets (ROA). Moreover, it is proven that Big4 auditors perform audit services of higher audit quality than non-Big4
18
auditors (DeAngelo, 1981; Francis, 2004; Ashbaugh-Skaife, Collins, and Kinney, 2007). It is therefore more likely that a review will be purchased when the firm is audited by a Big4 audit firm, as by a non-Big4 audit firm. To control for this possibility, I include a dummy variable that takes a value of 1 if the firm is audited by a Big4 audit firm (BIG4) and 0 if a firm is audited by a non-Big4 audit firm. In addition, I include year (YEAR) and industry (IND) dummies to control for time-series and cross-sectional differences in the decision to purchase a review. The industry variable is based on the five-sector classification model by Fama and French. This dummy takes the value of 1 if the firm is belonging to that sector, and 0 otherwise. In this research, the hold-out group is the ‘other’ category.
Next to these firm-level variables I control for the level of economic development, which is a country-level variable, as economic development might affect the decision to purchase a review. According to Ball (2001) an adequate infrastructure is needed for an ‘‘economically efficient system of public financial reporting and disclosure’’. Poor economic development hinders such an infrastructure and even if there does exist an adequate infrastructure, it can still be too costly to support credible external verification of financial information when economic development is poor. So, the incentives for firms to voluntarily purchase assurance are lower when economic development is poor. I include therefore the log of gross domestic product per capita (GDP) in line with Gaio (2010) to control for this possibility. To deal with potential outliers, I have winsorized all continuous firm-level and country-level variables at the 1 and 99 percentile.
VI. RESULTS
Descriptive statistics
I provide information regarding my dependent variable in Table 3, Panel A, at the country level. This table shows that the full sample consists of 2,975 observations, partitioned in a no review sample of 2,632 observations, and a review sample of 343 observations. This means that approximately 12 percent out of the full sample purchases a review. Subsequently, I provide information pertaining to firm- and country-level determinants in Table 3, Panel B. This table shows that these firms have been audited in most cases by a Big4 audit firm (95,5 percent). Also, the sampled firms belong in most cases to the consumer, manufacturing or other than hi-tech and health care industries.
19
In Table 4, I provide information pertaining to Pearson correlation coefficients. This table shows that the correlation coefficients are not higher than 0.6, which indicates that there are no possible multicollinearity issues. However, the correlation coefficient between the log of gross domestic product per capita and individualism is close to this maximum (r = .538). This is in line with previous research, as Hooghiemstra et al. (2015) find a similar correlation coefficient (r = .54).
< INSERT TABLE 4: Correlation matrix >
Regression results
To test for an association between the decision to purchase a review and various cultural dimensions, controlling for other explanatory variables, I perform a PROBIT regression. The results of these tests are presented in Table 5. In columns (1) to (3) the results for each cultural dimension, controlling for all other explanatory variables, are presented. Column (1) shows that there exists a significantly negative association between the degree of uncertainty avoidance and the decision to purchase a review (β = -.011, p < .01). In other words, the results suggest that in high uncertainty-avoidant cultures there is a lower demand for a review. This finding suggests that there are substitutes available for reviews in high uncertainty-avoidant cultures, which makes purchasing a review not valuable. Moreover, in column (2) I find a significantly negative association between individualism and the decision to purchase a review (β = -.011, p < .01). This finding is not in line with my hypothesis, as Hypothesis 2 predicts a significantly positive coefficient. Furthermore, in column (3) I find a significantly negative association between long-term orientation and the decision to purchase a review (β = -.007, p < .01). This is in line with Hypothesis 3, which predicts a significantly positive coefficient between short-term oriented cultures and the decision to purchase a review.
In columns (4) to (6), the effect of two combined cultural dimensions is assessed. These tests reinforce the results reported in the first three columns, as almost all dimensions are significant at the 1 percent level and remain negative. The only exception is the result found in column (5), where long-term orientation is significant at the 10 percent level (β = -.004, p < .10).
In column (7) all variables are included in the PROBIT regression. This model specification corroborates the evidence that there exist a significantly negative association between the cultural dimensions and the decision to purchase a review. In the full model there is a significantly negative association between uncertainty avoidance and the decision to
20
purchase a review, however the significance is less than in the other columns (β = -.005, p < .05). Moreover, in the full model there exists a significantly negative association between respectively individualism and long-term orientation and the decision to purchase a review (β = -.011, p < .01; β = -.010, p < .01). Moreover, the full model has the highest explanatory power as it has the highest pseudo R2 (R2 = .07).
With regard to the firm-level and country-level variables, these variables remain relatively stable. An exception is the result regarding BIG4, as the associations are quite mixed. However, in the full model a positive association is found which is in line with previous research. Further, the associations found with regard to SIZE, LEV, ROA, and GDP are in line with previous research (Boritz and Liu, 2006; Bédard and Courteau, 2015; Kajüter et al., 2016; Porumb and Karaibrahimoglu, 2017). Surprisingly, complexity measured by means of the ratio of inventories and receivables to total assets is found to be negative, but in most cases not significant. Also, the market to book ratio is surprisingly significantly positive at the 1 percent level, which is in line with the findings of Porumb and Karaibrahimoglu (2017).
< INSERT TABLE 5: PROBIT regression >
Sensitivity checks
To address potential sensitivity issues, I re-estimate my initial model using an alternative measure for national culture and an additional control variable that can influence the decision to purchase a review. In Table 6 and 7 the results are shown.
First, it is possible that the initial results are the result of using Hofstede’s (2001) measures for national culture. Therefore I re-estimate the model by using the GLOBE cultural dimensions studied by House, Hanges, Javidan, Dorfman, and Gupta (2004). The nine cultural dimensions that underlie the GLOBE study, have similarities with the ones of Hofstede (2001). Both have an uncertainty avoidance measure, defined in the GLOBE study as the ‘‘degree of reliance by a society, organization, or group on social norms, rules, and procedures to alleviate the unpredictability of future events’’ (G_UAI). Moreover, Hofstede’s (2001) individualism/ collectivism measure is measured in the GLOBE study by two dimensions. The first dimension, in-group collectivism (G_IGC), is defined as ‘‘the degree of expressing loyalty, pride, and cohesiveness by individuals in their organizations or families’’ (House et al., 2004). The second dimension, institutional collectivism (G_INSC), is defined by House et al. (2004) as ‘‘the degree of encouragement and rewarding of collective distribution of resources and action, by organizational and societal institutional practices’’. Lastly, Hofstede’s (2001) long-term
21
orientation is similar to GLOBE’s future orientation dimension (G_FO). This dimension is defined as ‘‘the degree of encouragement and rewarding of future-oriented behaviours, by a collective’’.
In Table 6, I provide the results of this sensitivity check. These results corroborate my initial results for two out of my three hypotheses. With regard to uncertainty avoidance, again a significantly negative association is found. Also, the results regarding the two collectivism cultural dimensions are in line with my initial results. This seems contradicting, but Hofstede (2001) measures individualism, whereas the GLOBE study measures collectivism. So, a significantly positive association is in line with the initial found significantly negative association based on Hofstede’s (2001) measure of individualism. Contradicting to my initial results is the positive association found in columns 2, 5, and 7, between future orientation and the decision to purchase a review. This suggests that a review is purchased by countries characterized by a long-term orientation, instead of a short-term orientation.
< INSERT TABLE 6: Sensitivity check using alternative national culture measure >
Second, it is possible that other variables influence the decision to purchase a review. It is possible that the level of investor protection affects the decision to purchase a review. This lies in the fact that the degree of agency problems varies with investor protection (Hooghiemstra et al., 2015). In low investor protection countries, agency problems will be exacerbated as shareholders have less power to influence managers. In such a situation it would be beneficial to voluntarily purchase a review. This is already evidenced in light of sustainability reporting. Herda, Taylor, and Winterbotham (2014) show that managers in low investor protection countries are more likely to purchase sustainability assurance, which is used as a substitute monitoring mechanism. Therefore, I include the anti-director rights index (ADRI) from
Djankov, La Porta, Lopez-de-Silanes, and Shleifer (2008) as additional variable.19 This index
can take a value from 0 to 6, where 0 represents the weakest anti-director rights and 6 the strongest.
In Table 7 the results are provided, for both the full model based on Hofstede’s (2001) cultural dimensions and for the GLOBE’s study cultural dimensions (House et al., 2004). These results corroborate my initial findings almost completely, as all coefficients are in line with the initial coefficients except the one of uncertainty avoidance using Hofstede (2001). Moreover,
19 This index is based on the original anti-director rights index developed by La Porta, Lopez-de-Silanes, Shleifer,
22
in this model there exists a negative association between future orientation and the purchase of a review in line with my initial results. However, these associations are both not significant.
< INSERT TABLE 7: Sensitivity check using an additional variable >
VII. CONCLUSION
In this study I analyse if national culture acts as a determinant of the decision to purchase a review. This possibility arises since the demand for assurance is a function of perceptions regarding agency problems which in turn are affected by national culture. Drawing on a sample of 2,975 firm-year observations for 1,232 non-financial unique listed firms from 22 countries, during the period 2004-2014, I find that national culture acts as a determinant of purchasing a review. Specifically, I find that uncertainty avoidance is negatively associated with the decision to purchase a review. This finding suggests that in societies characterized by high uncertainty avoidance, the review has no additional value for reducing agency costs, since these cultures have substituting mechanisms in place. Moreover, I find no support for a positive association between an individualistic society and the decision to purchase a review. Rather, I find that collectivistic societies are more likely to purchase a review. A possible explanation for this finding is that in-group favouritism actually increases agency problems. According to Kyriacou (2016), in-group favouritism leads to issues as corruption and nepotism in a public sphere. Corruption and nepotism is not in line with the interests of principals and makes it harder to monitor agents, and therefore increases agency problems and the demand for assurance. Furthermore, I find that there exists a positive association between short-term orientation and the decision to purchase a review. This finding suggests that the focus on short-term gains is accompanied with the demand for assurance in the form of purchasing a review. These results continue to be largely supported when altering the model, by using an alternative measure for national culture and adding an additional explanatory variable.
The findings of this study have several implications for academics, regulators, firms, and firms’ stakeholders. I extend the extant literature that discusses the costs and benefits of voluntarily purchasing a review (Ettredge et al., 2000; Manry et al., 2003; Hoehn, 2013; Bédard and Courteau, 2015; Kajüter et al., 2016; Porumb and Karaibrahimoglu, 2017), by documenting that a determinant of the voluntary decision to purchase a review is constituted by national culture. As the first cross-country study in the field of the decision to voluntarily purchase a review, this finding suggests that the mixed results of previous research can be due to differences in national culture. Given that national culture acts as a determinant of the voluntary
23
decision to purchase a review is also of interest to regulators. This finding provides regulators empirical evidence on the effects of reviews, and could aid in the debate to mandate reviews or not (Bédard and Courteau, 2015). This study shows that regulators need to be aware that firms are more likely to voluntarily purchase a review when these firms operate in societies that can be characterized as collectivistic or short-term oriented, and less likely to purchase a review when societies can be characterized as high uncertainty-avoidant. These insights are also of interest to firms, as the association between national culture and the voluntary decision to purchase a review indicates that firms need to be aware of the effect that national culture has on the demand for assurance by stakeholders. For example, when choosing a country to set up a business, assurance costs can be higher in a country that is characterized as collectivistic or short-term oriented. Lastly, the insights provided by this study are also of interest to firms’ stakeholders as this study shows that agency problems are larger in particular cultures, but that firms are more likely to purchase a review to resolve these agency problems.
Finally, I provide some limitations of this study and some potentially fruitful areas for future research. First, the distribution of the review and no review sample is not in line with previous research. Where this study draws on a sample where approximately 12% of the observations purchase a review, other researchers have drawn on a more even distributed sample (Boritz and Liu, 2006; Bédard and Courteau, 2015; Porumb and Karaibrahimoglu, 2017). However, previous researchers have drawn on a sample out of one country. Therefore, the lower percentage of review firms could be explained by the cross-country nature of this study. As this is the first cross-country study in this area, future research is needed to provide more in depth information regarding the decision to purchase a review and more research can show if a distribution of approximately 12% is a common distribution.
Second, this study has investigated if cultural dimensions can explain why a review is purchased by some firms, whilst other firms do not purchase voluntarily a review. Although I investigated three cultural dimensions and investigated if the results are similar when using an alternative measure for national culture, not all cultural dimensions are considered and also there exist other alternative measures for national culture. Future research could deepen this understanding, by investigating for example Hofstede’s (2001) power distance and masculinity/femininity cultural dimensions. Also, an alternative measure like the one of Schwartz (1990) could be used.
Third and lastly, this study is based on the reasoning that a review is purchased when this review is accompanied by an alleviation of agency problems. However, this study does not
24
directly investigates if a review results in an alleviation of agency problems. Future research could investigate the before and after review situation in light of agency costs.
25
REFERENCES
Aggarwal, R., Faccio, M., Guedhami, O., and Kwok, C.Y. (2016). Culture and finance: An introduction. Journal of Corporate Finance, 41(10), 466-474.
Ashbaugh-Skaife, H., Collins, D.W., and Kinney, W.R., Jr (2007). The discovery and reporting of internal control deficiencies prior to SOX-mandated audits. Journal of Accounting and Economics, 44(1-2), 166-192.
Ball, R. (2001). Infrastructure requirements for an economically efficient system of public financial reporting and disclosure. Brookings-Wharton Papers on Financial Services, 127-182.
Ball, R., and Shivakumar, L. (2005). Earnings quality in UK private firms: comparative loss recognition timeliness. Journal of Accounting and Economics, 39(1), 83-128.
Bandyopadhyay, S.P., Boritz, J.E., and Liu, G. (2007). Voluntary assurance on interim financial statements and earnings quality. CAAA 2008 Annual Conference Paper. Available at:
https://ssrn.com/abstract=1080619, retrieved December 21, 2016.
Bearden, W.O., Money, R.B., and Nevins, J.L. (2006). A measure of long-term orientation: Development and validation. Journal of the Academy of Marketing Science, 34(3), 456-467.
Bédard, J., and Courteau, L. (2015). Benefits and costs of auditor’s assurance: Evidence from the review of quarterly financial statements. Contemporary Accounting Research, 32(1), 308-335.
Beugelsdijk, S., Kostova, T., and Roth, K. (2017). An overview of Hofstede-inspired country-level culture research in international business since 2006. Journal of International Business Studies, 48(1), 30-47.
Beugelsdijk, S., Maseland, R., and Van Hoorn, A. (2015). Are scores on Hofstede’s dimensions of national culture stable over time? A cohort analysis. Global Strategy Journal, 5(3), 223-240.
Bolton, P., Scheinkman, J., and Xiong, W. (2006). Executive compensation and short-termist behaviour in speculative markets. Review of Economic Studies, 73(3), 577-610.
Boritz, J.E., and Liu, G. (2006). Why do firms voluntarily have interim financial statements
reviewed by auditors? Available at:
https://www.researchgate.net/profile/J_Boritz/publication/228340831_Why_Do_Firms _Voluntarily_Have_Interim_Financial_Statements_Reviewed_by_Auditors/links/0046
