Governance disclosure quality and market valuation of firms in UK and Germany

This study develops a “ comply or explain ” index which captures compliance and quality of explanations given for non-compliance with the corporate governance codes in UK and Germany. In particular, we explain, how compliance and quality of explanations provided in non-compliance disclosures, and various other internal corporate governance mechanisms, affect the market valuation of firms in the two countries. A dynamic generalised method of moments (GMM) estimator is employed as the research technique for our analysis, which enabled us to control for the potential effects of endogeneity in our models. The findings of our content analysis suggest that firms exhibit significant differences in compliance, board independence and ownership structure in both countries. The “ comply or explain ” index is positively associated with the market valuation of UK firms suggesting that compliance and quality governance disclosure are value relevant in the UK. Institutional blockholders' ownership is, however, negatively associated with the market value of firms, which raises questions about the monitoring role of institutional shareholders in both countries. We argue that both compliance and explanations given for non-compliance are equally important, as long as valid reasons and justifications for non-compliance are provided by the reporting companies. These findings thus imply that the “ comply or explain ” principle is working well and that UK and German companies could benefit from the flexibility offered by this principle. With respect to the role of board size, board independence, ownership structure, and


| INTRODUCTION
In the aftermath of the 2007-2009 global financial crisis corporate governance regulation has received considerable attention.The severe economic and social impact of the crisis and related corporate scandals resulted in major changes in the governance structure and regulations in most countries around the world.For instance, the Walker (2009) review of corporate governance mechanisms in the UK and the Dodd-Frank Wall Street Reform and Consumer Protection Act (2010) in the US recommended substantial changes to the corporate governance mechanisms of financial firms.Since the 2007-2009 financial crisis was attributed to the weak corporate governance practices at major financial institutions (Kirkpatrick, 2009;Conyon et al., 2011), the associated reviews specifically focused on financial firms.This highlights the perceived importance of firm-level corporate governance mechanisms for organisational performance and protecting the interests of stakeholders.
A considerable amount of literature has therefore examined the impact of corporate governance mechanisms on firm performance.For example, Ntim, Lindop, and Thomas (2013) suggest that improving risk reporting as well as risk management practices produces positive effects on the quality of risk disclosure/management practices in South African companies.Similarly, Tan and Liu (2016) document that, as compared to several other factors, governance structure variables can better explain the long-term idiosyncratic volatility of Australian firms.Furthermore, Farag and Mallin (2017) report that board diversity reduces banks' vulnerability to the effects of finical crisis in Europe.Therefore, in relation to a firm's operations and long-term success all these findings pinpoint the important role of corporate governance mechanisms in different countries.
Although conceptually and theoretically the link between corporate governance and performance is clear, the overall findings of existing governance-performance literature are inconclusive.As such there are studies which report a strong positive relationship (see for example, Durnev & Kim, 2005;Klapper & Love, 2004), while others document a negative relationship (see for example, Bauer, Guenster, & Otten, 2004); or no relationship (see for example, Schultz, Tan, & Walsh, 2010;Wintoki, Linck, & Netter, 2012) between corporate governance and firms' performance.In the context of the 2007-2009 financial crisis, a recent study by Gupta, Krishnamurti, and Tourani-Rad (2013) document that well-governed companies do not necessarily out-perform poorly governed companies.However, an emerging stream of research in this area argues that the majority of the index-based studies have focused only on the compliance aspect of "comply or explain" codes.As the governance systems based on the 'comply or explain' principle have two important elements, "comply" and "explain", emphasising on just one aspect that is, compliance is not sufficient as the "explain" aspect of such systems is equally if not more important than the other (Shrives & Brennan, 2017).The existence of inconclusive evidence in this area, in general, and existence of limited evidence on the explain element of the governance codes in particular, call for further investigations.
Considering all of the above arguments, this study contributes to this line of literature and examines the relationship between the quality of explanations provided by companies when they choose non-compliance and the market valuation of firms across two different corporate governance systems, namely, the UK and Germany.As suggested by Van Essen, Engelen, and Carney (2013), corporate governance is a context-specific issue, and as a result, the impact of corporate governance mechanisms could be different in different countries.In terms of the legal and financial systems, corporate governance regulations, institutional settings, and capital market regulations, there are substantial differences between Germany and the UK.However, the corporate governance systems in the two countries are based on the "comply or explain" principle.Therefore, there may be different implications of firm-level corporate governance mechanisms on the market value of firms in the two countries.In line with this, this study investigates whether or not the relationship between the quality of explanations and firms' market value is different in UK and Germany.In other words, the study sheds light on how corporate governance disclosures are perceived by stakeholders in the two governance systems which are based on the same principle of "comply or explain" but have some key institutional differences.
This paper therefore makes significant contributions to the existing corporate governance literature.Unlike previous index-based studies which only cover the compliance aspect, we develop a "comply or explain" index, that takes into account the level of compliance as well as the quality of reported explanations for non-compliance with the prescribed corporate governance codes.We argue that focusing only on the compliance aspect of a "comply or explain" principle would undermine the second important pillar of a "comply or explain" system of corporate governancethe explanations reported for non-compliance with corporate governance codes.A recent version of the UK Corporate Governance Code (2018) requires each listed company to justify and explain clearly the reasons for non-compliance in case a company has implemented alternative governance practices, other than those prescribed in the Code.The Code states: "Explanations should set out the background, provide a clear rationale for the action the company is taking, and explain the impact that the action has had.Where a departure from a Provision is intended to be limited in time, the explanation should indicate when the company expects to conform to the Provision.Explanations are a positive opportunity to communicate, not an onerous obligation".(Financial Reporting Council, 2018, p. 2).
This clearly highlights the importance attached to explanations by the regulator in case of non-compliance.The current study, therefore, focuses on the quality of explanations provided by companies and its association with the market valuation of firms.Using a sample of 120 firms from the UK and Germany for the period 2007-2011, the results of this study reveal that compliance or providing good quality explanations is positively associated with the market value (Tobin's Q) of firms in the UK and Germany.In terms of individual governance mechanisms, board independence is positively associated with the market valuation of German firms, while this relationship is significantly negative for the UK firms.We find that noninstitutional blockholders' ownership has a positive impact on the performance of German firms.However, institutional blockholders' ownership is negatively associated with the market valuation of firms in both countries, which raises questions about the monitoring role of institutional shareholders in both countries.This finding contradicts the assumptions of Shleifer and Vishny (1986), who argue that blockholders exercise significant influence over their investee companies.It is therefore argued that institutional investors invest in a number of companies and hence they may not be effectively monitoring their investee companies as compared to noninstitutional investors, whose investment is highly concentrated in a few firms.
The rest of the paper is structured as follows.Section 2 provides justifications for choosing the UK-German context.Section 3 presents a review of the relevant literature and develops the research hypotheses.Section 4 contains discussions on the research methodology, data and econometric specification of the models.Section 5 discusses the results of our empirical analyses.Section 6 concludes the paper by presenting a short summary of the overall contributions and pinpoints the potential implications.The final section also contains a brief note on the research limitations and specifies avenues for future research.

| WHY THE UNITED KINGDOM AND GERMANY?
The publication of the Cadbury Report (1992) in the UK could be considered as the basis for various corporate governance reforms around the world over the last three decades.Unlike the USA and the UK, where corporate laws and governance codes included significant protection for investors, German regulators and policy makers seemed less concerned about the protection of shareholders until the late 1990s (Cromme, 2005).However, German companies listed on foreign stock exchanges were required to fulfil the listing requirements in overseas markets (Baums, 2003).In response to the growing local and international pressures calling for additional corporate governance reforms, the first German Corporate Governance Code was published in February 2002, exactly a decade after the publication of The Cadbury Report.In addition, there are several key differences between the UK and Germany.
The first key difference between the UK and Germany relates to the purpose of the corporate governance codes in these countries.In the UK the purpose of the Governance Code is explained by the regulator as "Good corporate governance should contribute to better company performance by helping a board discharge its duties in the best interests of shareholders; if it is ignored, the consequence may well be vulnerability or poor performance.Good governance should facilitate efficient, effective and entrepreneurial management that can deliver shareholder value over the longer term" (Financial Reporting Council, 2008, p.1). 1 German corporate governance code on the other hand, explains the purpose of the code as "…to promote the trust of international and national investors, customers, employees and the general public in the management and supervision of listed German stock corporations" (The German Corporate Governance Code, 2008, p.1). 2 These quotations from the two governance codes indicate that in the UK the focus is on protecting the interests of shareholders while in Germany the code focuses on protecting the interests of a wide range of stakeholders.Similarly, there are also differences in terms of legal systems, capital markets, ownership structures, code formation and developments (Kaufmann & Valderrama, 2008;Seidl, Sanderson, & Roberts, 2013).
In terms of similarities between the UK and Germany, the codes of corporate governance in both countries are based on a "comply or explain" principle.This means that listed companies are required to report if they have complied with the recommended corporate governance provisions and in the case of noncompliance, they are required to provide explanations.The empirical debate about the effectiveness of a unitary board system in the UK versus a two-tier board system in Germany (Davies, 2000), and the effectiveness of a common law system versus a civil law system (La Porta, Lopezde-Silanes, Shleifer, & Vishny, 1998) provides us with an opportunity to examine the effectiveness of corporate governance regulations in a cross-country setting.
A comparative study could therefore better explain the differences of and implications for national corporate governance regulations.Prior studies have examined the governance-performance relationship in the context of the UK (Dahya, McConnell, & Travlos, 2002;Weir & Laing, 2000;Weir, Laing, & McKnight, 2002), USA (Brown & Caylor, 2006;Gompers, Ishii, & Metrick, 2003), Germany (Drobetz, Schillhofer, & Zimmermann, 2004;Goncharov, Werner, & Zimmermann, 2006), Europe (Bauer et al., 2004), and emerging markets (Durnev & Kim, 2005;Klapper & Love, 2004).However, considering the inconclusive international evidence in this area, we contribute to the existing international corporate governance literature in a cross-country setting in the context of the UK and Germany.The next section covers an overview of the existing literature and develops the research hypotheses on the association between internal corporate governance mechanisms and market valuation of firms.

| Developing the theoretical framework
In the existing literature, the relationship between corporate governance and various firm level outcomes is explained with several theoretical perspectives.For instance, agency, stakeholder, stewardship and resource dependence theories have been used extensively to investigate how firm level governance mechanisms affect performance.In this regard, Kumar and Zattoni (2015) recommend the use of multiple theoretical perspectives, particularly when investigating governance issues in different corporate governance systems.Agency theory focuses on issues arising from the separation of ownership and control in publicly listed companies (Jensen & Meckling, 1976).The main argument is that the agency costs arising from the separation of ownership and control cannot be dealt with only through contracts (Hart, 1995).
As managers have control over the free cash flows of a firm, corporate governance mechanisms are needed to monitor managers so that they are not risk-averse and self-serving, to make them accountable and to make sure that the free cash flows are either returned to shareholders, or re-invested for the long-term success of the organisation (Shleifer & Vishny, 1997).The increased monitoring and accountability brought about by strong governance mechanisms, it is assumed, will lead to the efficient use of resources and will be translated into improved profitability and long-term success (Jensen, 1986).The agency theory also assumes that strong internal corporate governance mechanisms will lead to a decrease in agency costs, that could result in lower premiums being charged by the providers of finance, and a lower cost of capital for the firm (Drobetz et al., 2004).
Existing literature that adopts agency theory, proposes a positive association between better alignment of managerial incentives and shareholders' interests (Jensen & Meckling, 1976;Shleifer & Vishny, 1997).Given its perceived importance for monitoring and protecting the interests of shareholders, researchers have extensively examined how the composition of board of directors (see for example, Adams, Hermalin, & Weisbach, 2010;Hermalin & Weisbach, 2003), board committees and structures (Klein, 1998;Liao, Luo, & Tang, 2015), and compliance with a prescribed code of corporate governance (for a review see, Bozec & Bozec, 2012) affect corporate performance.
From the perspective of stakeholder theory companies are not only responsible to their shareholders but also to meet the needs of their various stakeholders such as, customers, employees, general public including nongovernmental organizations, government and local community (Freeman, 1984;Freeman, Wicks, & Parmar, 2004).Therefore, from this theoretical perspective corporate governance mechanisms are needed to protect the interest of a wide group of stakeholders.This is in contrast to the narrow focus on shareholders adopted in agency theory.The managerial and ethical branches of stakeholder theory provide alternative explanations for the responsibilities of management and the information that will be disclosed by companies.The managerial branch of stakeholder theory postulates that corporate governance mechanisms and reporting are driven by the needs of the most powerful stakeholder of an organisation (Deegan, 2002).Therefore, if the most powerful stakeholder demands certain governance mechanisms or information, managers are more likely to implement those mechanisms and provide such information.However, the ethical branch explains that from an ethical perspective, corporate governance mechanisms and reporting by firms would aim to satisfy the needs of all their stakeholders and companies behave in this way out of necessity to treat everyone fairly (Deegan, 2002).
As mentioned earlier, the UK Corporate Governance Code is developed with a view to protect the interests of shareholders while the German Corporate Governance Code focuses on the needs of multiple stakeholders.Therefore, agency theory may be more appropriate in the UK context while Stakeholder theory provides a better theoretical lens for Germany.However, agency theory is also applicable in the context of Germany but the nature of agency problems might be different.For instance, Faccio and Lang (2002) provide evidence that in Germany only 10% firms are widely held while in the UK the figure is 63%.Similarly, the authors also document that in Germany family ownership is 64% while in the UK it is only 23%.These ownership patterns have been shown to give rise to different types of agency problems between controlling shareholders and minority shareholders (Bebchuk & Weisbach, 2010;Claessens, Djankov, & Lang, 2000).It is therefore argued that agency and stakeholder theories are providing the appropriate frameworks to analyse the relationship between explanations provided by firms and performance in the context of this study.

| Compliance with corporate governance codes and firm performance
As discussed above, theoretically strong corporate governance mechanisms will be associated with good firm performance.In this regard, governance prescriptions provided in corporate governance codes are considered as proxies for strong corporate governance mechanisms.Most prior studies in this area have therefore largely focused on the level of compliance with corporate governance codes by either developing a corporate governance index for each firm (e.g., Beiner, Drobetz, Schmid, & Zimmermann, 2006;Hooghiemstra, 2012) or by using commercially available corporate governance indices (Aggarwal, Erel, Stulz, & Williamson, 2010;Bauer et al., 2004;Bauwhede & Willekens, 2008;Gompers et al., 2003;Klapper & Love, 2004).However, existing evidence on the relationship between a firm's compliance with corporate governance codes and performance is mixed.Studies havereported three types of findings: (i) a positive relationship (Ammann, Oesch, & Schmid, 2011;Chhaochharia & Laeven, 2009;Klapper & Love, 2004;Dharmapala & Khanna, 2013)); (ii) a negative relationship (Bauer et al., 2004); or (iii) no relationship (Schultz et al., 2010;Wintoki et al., 2012).
There are different explanations in the existing literature about the mixed findings.For instance, Chhaochharia and Grinstein (2007) argue that the impact of compliance/non-compliance may not be uniform, and that it depends on the size of the firm.For a sample of US listed companies, the authors document that large firms that are non-compliant earn positive abnormal returns but small firms that are non-compliant earn negative abnormal returns.This points in the direction that perhaps investors take into account a multitude of factors when using compliance or non-compliance related disclosures in their decision making.Similarly, Van Essen, Engelen, & Carney, (2013) argue that the relationship between governance and performance is context specific hence studies conducted in different contexts will report different results.Other studies argue that the mixed results may be attributed to the use of methods that fail to control for endogeneity and related issues (see for example, Wintoki et al., 2012).
This study adds to this stream of literature and provides evidence on the impact of the nature of explanations provided by firms in two similar yet different contexts.It is argued that in addition to communicating their level of compliance it is also important how noncompliance is justified and communicated.Furthermore, it is also argued that the compliance or non-compliance indices in prior studies have entirely focused on one aspect, that is, compliance with a "comply or explain" principle.However, the explain element of a "comply or explain" principle has largely been ignored in the existing governance literature.The quality of governance disclosure is important because full disclosure could lead to informed decision making which will also help in enhancing transparency in listed companies (Wu & Bowe, 2012).In line with the index-based studies in the governance literature (e.g., Aggarwal et al., 2010;Bauer et al., 2004;Bauwhede & Willekens, 2008;Beiner et al., 2006;Gompers et al., 2003;Hooghiemstra, 2012;Klapper & Love, 2004;Liu, Padgett, & Varotto, 2017), we develop a "comply or explain" index which captures not only the level of compliance with corporate governance codes but also the quality of explanations provided by companies when they do not comply with the corporate governance codes.
As the level of compliance has gradually improved, the attention of regulators in the UK and rest of the Europe has shifted to the quality of the reported explanations for non-compliance with the corporate governance codes (European Commission, 2012;Financial Reporting Council, 2012a).The EU regulations 2009 also suggest that the existing "comply or explain" systems need to be strengthened, and should not be abandoned (European Commission, 2009).It is also evident from the existing literature that only a few index-based studies has focused on the "explain" element of the "comply or explain" principle.In line with this Rose (2016) shows that Danish firms which either comply or provide plausible explanations for their non-compliance have higher return on assets (ROA) and return on equity (ROE).Similarly, Shrives and Brenan (2017) focus specifically on the explanations provided for their non-compliance by a sample of FTSE 100 firms.They document that firms exploit the flexibility and attempt to mislead the readers by adopting different strategies, such as providing ambiguous explanations, communicating that non-compliance is not a big issue, and justifying non-compliance by focusing on the end results.
While reviewing the literature on the effectiveness of compliance in the UK, Dedman (2002) provides evidence suggesting that compliance enhances board oversight, reduces earnings manipulation and is associated with better disciplining of executives.However, MacNeil & Li (2006) investigate a sample of FTSE 100 companies which were persistently non-compliant with the UK corporate governance code.The authors argue that shareholders of such firms tolerate non-compliance and ignore the nature of explanations as long as such firms are profitable.Furthermore, Luo and Salterio (2014) develop a three-point scale (index) for a sample of Canadian firms to measure compliance and the quality of corporate governance disclosure (in the case of non-compliance).Their index assigns a value of zero for noncompliance or non-disclosure, '1' for compliance, and '2' for explanations reporting alternative governance arrangements.The index thus captures: (a) full compliance; (b) non-compliance with no explanations for non-compliance; and (c) non-compliance with reasons for non-compliance reported by companies.They report a significantly positive relationship between the quality of corporate governance disclosure and the market value of firms.More recently, Honisberg (2019), documents that in the case of "comply or explain" governance systems, providing explanations reduce misreporting at hedge funds and leads to better internal control and improved performance.
In order to assess the quality of explanations provided for non-compliance, Arcot and Bruno (2011) develop a corporate governance and disclosure index.They find that firms which are fully compliant or, alternatively, in the case of non-compliance, which fully explain their deviations for non-compliance have a better operating performance.Similarly, with application of a corporate governance index for Dutch companies which assigns a lower score to uninformative explanations and higher score to firm-specific detailed explanations reported for non-compliance, Hooghiemstra (2012) argues that firms having concentrated ownership structures are less likely to have disclosure related issues, and these firms therefore approach the disclosure related requirements for only symbolic reasons.In line with this, the present study focuses on Germany where concentrated ownership is more common and compares it with the UK where ownership of publicly held firms is more widespread.Therefore, it provides us with an opportunity to investigate how the quality of explanations provided in these two different contexts affect firm value.
The above discussion indicates that only a few indexbased studies have focused on the disclosure aspect of the "comply or explain" principle.Furthermore, there is no study which has focused on the quality of explanations provided in a cross-country context.As discussed earlier, corporate governance systems in the UK and Germany are both based on the principle of "comply or explain" but the objectives of both these codes are quite different.Similarly, there are key differences between the two countries in terms of ownership structures and legal systems.Therefore, this study adds to the existing literature by providing evidence whether the quality of explanations provided in two different contexts affect firm value differently.We argue that the existing "comply or explain" principle explicitly offers firms the flexibility to either comply or provide an explanation for their noncompliance.Therefore, non-compliant firms are not necessarily badly governed as long as non-compliance is justified and communicated to the relevant stakeholders.Based on this, the quality of explanations provided by firms could be considered as important and relevant information that would affect market values of firms both in the UK and Germany.Based on agency and stakeholder theories, firms that appear to have implemented mechanisms to protect the interests of their stakeholders (potential investors and shareholders are some of the major stakeholders) in both countries should outperform their counterparts with weak governance mechanisms.We therefore develop the following hypothesis: H 1 : High quality corporate governance disclosure is positively associated with the market valuation of firms in the UK and Germany.

| Board size
The size of board of directors is another important determinant of firms' governance mechanisms that affects performance.In the majority of cases the board size of UK companies varies from 7 to 17 directors (Guest, 2009, p. 32).In contrast, German boards are relatively larger, with board size of 15 to 30 directors (Du Plessis et al., 2012).The larger board size in the case of Germany is because of the two tier-board governance structure as opposed to unitary boards in the UK.In Germany the non-executive directors (NEDs) sit on the supervisory board and include representatives from a number of stakeholders such as, employees, financial institutions and government.In terms of the effect of board size, Lipton and Lorsch (1992) argue that when board size increases above ten directors, then it creates additional costs for the organisation in the form of a slow decision making process.Neither the UK nor the German corporate governance codes recommend any specific board size for listed companies.However, the German Codetermination Act requires that at least one third to half of board members should be representatives of employees on the supervisory boards of companies having more than 500 or over 2,000 employees respectively (Commission of the German Corporate Governance Code, 2015, p. 1).The German system of corporate governance has also put a greater emphasis on employee empowerment, which is often known as a stakeholderbased system of corporate governance.Based on these two contrasted perspectives on the role of board of directors, this study uses agency theory and stakeholder theories to investigate the impact of board size on firm valuation.
While looking at the findings of the existing literature on board size and performance, there is again inconsistent evidence on this issue.Some studies show a significantly positive association between board size and firm performance, which supports the assumptions of agency theory that larger boards enhance monitoring and accountability.In addition to this, resource dependence theory also predicts that larger boards provide a link between the organisation and its external environment (see for example, Beiner et al., 2006;Kiel & Nicholson, 2003).However, larger boards have been shown to be associated with increased agency costs in the existing literature.For instance, examining the impact of board size on the performance of 450 firms across three countries from market-based systems (UK, USA, and Canada) and seven countries from the relationship-based system (Germany, Belgium, Spain, France, Italy, The Netherlands, and Switzerland), De Andres, Azofra, and Lopez (2005) show a significantly negative relationship between larger board size and firm performance across the two different corporate governance systems.Similarly, using a large sample of 2,746 UK companies, Guest (2009) finds a negative relationship between board size and the market value of firms and shows consistent results with the earlier findings of Yermack (1996).More recently, Yamori, Harimaya, and Tomimura (2017) document that in cooperative banks, board size is negatively associated with efficiency.Based on the above discussions we develop and test the following research hypothesis: H 2 : There is a negative relationship between board size and the market valuation of firms in UK and Germany.

| Board independence
In the context of corporate governance literature, board independence has been regarded as a key internal governance mechanism for firms.The appointment of independent non-executive directors on boards has therefore been widely acknowledged in different corporate governance codes around the world.Initially, the Cadbury Report (1992) called for the appointment of at least three independent NEDs on corporate boards.Over the last few decades, the proportion of independent NEDs on corporate boards has increased to at least half of the board size for large and at least two for small UK companies (The UK Corporate Governance Code, 2016).
In Germany, firms with more than 500 and 2000 employees are legally required to appoint one third to one half of their employees' representatives to the supervisory boards, whereas the remaining members are appointed by the shareholders.The supervisory board, which comprises NEDs, is exclusively responsible for monitoring and advising the management board (Davies, 2000).Section B3.3. of the UK Corporate Governance Code (2016) imposes regulatory restrictions on executive directors and requires that "the board should not agree to a full time executive director taking on more than one non-executive directorship in a FTSE 100 company nor the chairmanship of such a company".Section 5.4.5 of the German Corporate Governance Code (2015) stipulates that "members of the management board of a listed company shall not accept more than a total of three supervisory board mandates in non-group listed companies or in supervisory bodies of non-group companies which make similar requirements".It is thus evident that detailed coverage on the presence of NEDs in the governance codes of UK and Germany specify the importance given to the presence of NEDs on boards for firms' operations and performance in both countries.
There is evidence in the existing literature which documents a positive relationship between the presence of a larger proportion of independent NEDs on boards and the financial performance of firms (e.g., Aggarwal et al., 2010;Zahra & Pearce, 1989).Similarly, Li, Lu, Mittoo, and Zhang (2015) show a positive relationship between board independence and performance of Chinese listed companies, and pinpoint the importance of board independence in firms with concentrated ownership structure.For a cross country sample, Hu, Li, Taboada, and Zhang (2020) document that board reforms that improve monitoring and the oversight of management reduce crash risk significantly.From an agency theory perspective, it is expected that the proportion of NEDs would signal improved monitoring and therefore will be positively associated with the financial performance of companies.In the context of Germany, a higher proportion of NEDs would signal that such boards are better equipped to protect the interest of a wide range of stakeholders.Thus, in line with the recommendations of corporate governance codes in the UK and Germany, and based on the assumptions of agency and stakeholder theories, the following research hypothesis is formed: H 3 : Existence of independent non-executive directors on boards is positively associated with the market valuation of UK and German companies.

| Number of board meetings
Another key element of an effective corporate board structure that may enhance the performance of firms is related to board meetings (measured by the total number of board meetings during the year).In relation to board meetings, Lipton and Lorsch (1992) argue that time spent on board meetings could be regarded as a resource provided to the organisation, and increasing board activities enhances the monitoring and control function exercised by the board of directors.In line with this, Vafeas (1999) emphasises that firms can improve their internal control mechanisms by increasing their board activities (e.g., frequency of board meetings).In the context of the 2007 financial crisis, Brick and Chidambaran (2010) have reported a positive relationship between board meetings and firm performance.They argue that external market pressures from investors and regulators may significantly influence a firm's management to increase its board activities, which positively influences firms' performance.Similarly, in relation to financial institutions, Salim, Arjomandi, and Seufert (2016) document a positive relationship between board meetings and the performance of Australian banks.Although use of the number of board meetings is only a proxy indicator of board activities and may appear as a rather simplistic representation of a board's activities, however, its main advantage is that this information is accessible and comparable.Thus, consistent with prior research of Brick and Chidambaran (2010), and bearing in mind the recommendations of the UK Corporate Governance Code (2016) and the German Corporate Governance Code (2015) we develop the following research hypothesis: H 4 : There is a positive relationship between board activity and the market valuation of UK and German companies.

| Gearing
The presence of a higher debt to equity proportion (gearing) in the statement of financial position is another important control mechanism that can restrain managers from diverting free cash flows to low-return projects.In this regard, Jensen (1986) argues that managers are likely to invest free cash flows in low-return projects instead of distributing it to the shareholders, and gearing can be used as an alternative corporate governance mechanism for controlling such activities.A firm's debt financing provides a signal to the market that its managers will be monitored by the creditors and that managers will be willing to distribute free cash flows to the shareholders (Beiner et al., 2006).
In the context of Germany, where banks play a significant role in their corporate governance structure, Agarwal & Elston (2001, p. 226) argue that "[b]ank-influenced firms should enjoy increased access to capital through easier access to bank debt or preferential terms on loans.In addition, bank involvement with a firm serves as a signal to outside investors and causes a certification effect, which makes it easier for firms to attract additional equity".The huge role played by the German financial institutions consequently put Germany at the forefront of "bank-based" systems of corporate governance (Luintel, Khan, Leon-Gonzalez, & Li, 2016).However, empirical evidence on the relationship between gearing and firm performance is inconclusive.Some studies show a negative relationship between gearing and firm performance and confirm the presence of a conflict of interests between shareholders and creditors in highly geared companies (see for example, Agarwal & Elston, 2001;Bauwhede, 2009;Francis, Hasan, & Wu, 2012).Other studies report a positive relationship between debt financing and firm performance and show that debt holders require additional internal control mechanisms (through debt agreements) beyond those implemented by the firm (see for example, Beiner et al., 2006;Goncharov et al., 2006).In the context of the UK, McKnight and Weir (2009) document a positive relationship between debt financing and firms' performance.Therefore, based on the free cash flow hypothesis of Jensen (1986), and consistent with prior empirical research (Beiner et al., 2006;Goncharov et al., 2006;Gorton & Schmid, 2000), we develop the following hypothesis: H 5 : There is a positive relationship between gearing and the market valuation of UK and German companies.

| External blockholders
The presence of external blockholders in the ownership structure exert a significant influence on the corporate governance practices of firms.In this regard, Shleifer and Vishny (1986) argue that, although concentrated ownership structure can partly mitigate the free-rider issues in large organisations, it may also expropriate the rights of minority shareholders.While some studies show a positive impact of institutional blockholders' ownership on firm performance (Gorton & Schmid, 2000;Lehmann & Weigand, 2000), other studies report a positive relationship between non-institutional blockholders' ownership and the performance of firms (see for example, Andres, 2008).It is therefore evident that existing governance literature on the relationship between institutional and non-institutional blockholder ownership and firm performance provides mixed evidence.
Figure 1 below presents a conceptual framework of this research, where the main differences between the two jurisdictions are clearly highlighted.research has either used an ordinary least squares regression (see for example, Gompers et al., 2003;Klapper & Love, 2004) or a fixed-effects model (see for example, Ammann et al., 2011;Chhaochharia & Laeven, 2009;Yermack, 1996).However, the findings reported in these studies should be "interpreted with caution" because the econometric models used in these studies fail to control for different kinds of endogeneitya situation when the causality may run from performance to governance (Schultz et al., 2010, p. 146).
In line with the above arguments, Wintoki et al. (2012) identify three sources of endogeneity, namely: (a) unobserved heterogeneity; (b) simultaneity or reverse causation; and (c) dynamic endogeneity.Unobserved heterogeneity arises when the relationship between governance and performance is affected by unobservable factors (for instance, firm-specific characteristics).Simultaneity or reverse causation arises when governance and performance affect each other simultaneously.For instance, compliance with a corporate governance code may enhance a firm's performance and valuation, but it is also possible that firms with a higher market valuation may choose strong corporate governance mechanisms (Durnev & Kim, 2005).Moreover, dynamic endogeneity arises when a firm's past/current performance affects the current/future governance structure of a firm (Wintoki et al., 2012, p. 582).For example, poor corporate performance may cause changes in the governance structure (e.g., removal of one or more directors from the board by shareholders) of a firm.
While conducting this investigation, this study considers all the above mentioned econometric issues and by following Schultz et al. (2010) and Wintoki et al. (2012), the following dynamic generalised method of moments (GMM) model estimation is proposed: In Equation 1, P it stands for firm valuation; P i, t − 1 is a one period lag operator (previous year performance); Gβ it represents corporate governance variables; Xn it represents control variables; μ it is firm-specific fixed effects; and ε it represents the error term.Table 1 provides definitions of variables.Our main explanatory variable is the "comply or explain" index.Consistent with Arcot and Bruno (2011) and Hooghiemstra (2012), we developed a corporate governance index, which assigns a lower score to uninformative explanations reported by non-compliant firms and vice versa.The index assigns a highest score of 5 to a firm fully compliant with all provisions of the UK and German Governance Codes.Each non-compliant firm in the sample receives a score from 1 to 5 for each category of explanation reported for noncompliance, and the score considers the relative informativeness of each category of explanation.For instance, a lower score of 1 is assigned when a firm reports "no explanation" for non-compliance; a score of 2 when a firm provides "generic" or "standard explanations" or assurance of future compliance; a score of 3 for an explanation offering alternative corporate An index which assigns a value of 1 to 5 to each firm and takes into account a firm's compliance, non-compliance and the explanations reported for non-compliance with the corporate governance codes.

Board size
The total number of directors serving on the board at the end of the year.

Board independence
The ratio of NEDs to total board members at the time of reporting.

Number of board meetings
The total number of board meetings during a year.

Gearing
The ratio of a firm's total debt to the book value of its total assets.

Institutional blockholders (%)
The percentage of equity (5% and above) owned by financial institutions, insurance companies, pension funds and unit trusts.

Non-institutional blockholders (%)
The percentage of equity (5% and above) owned by all other external shareholders (excluding institutional shareholders).

Firm size
The natural logarithm of the book value of a firm's assets.
Firm-specific risk (beta) A measure of firm's riskiness.The beta factor is derived by performing a least squares regression between adjusted prices of the stock and the corresponding Datastream market index (values taken from Datastream).
Foreign listing A dummy variable that takes a value of 1 if a firm is cross-listed and zero otherwise.

R&D expenditure
Research and development expenditure divided by sales.
Tobin's Q Total assets + market value of equity -Total common equity -Deferred taxes/Total assets.
governance mechanisms; a score of 4 for partial noncompliance over a temporary period of time; and a score of 5 for high quality detailed and firm-specific explanation.
While examining the differences between compliant and non-compliant UK companies, Arcot and Bruno (2011, p. 12) noted that: "..there should be no difference between a compliant company and a non-compliant company that deviates from standards for good and valid reasons that are fully disclosed".In fact, ignoring any aspect of the "comply or explain" principle would result in inaccurate generalisation about the effectiveness of the "comply or explain" principle.Similarly, in relation to compliance and noncompliance in Dutch firms, Hooghiemstra (2012), argues that firms with concentrated ownership structures are less likely to have disclosure related issues, and that such firms only implement the disclosure requirements for symbolic reasons only.Therefore, consistent with the existing literature (see for example, Hooghiemstra, 2012), we use the following approach to develop the index: The "comply or explain" index = 1(no explanations) + 2(generic or standard explanations + assurance of future compliance) + 3(description of alternative practice) + 4(partial non-compliance) + 5(firm-specific detailed explanations)/Total number of explanations reported by a firm.
Through the application of content analysis approach we examined 600 corporate governance reports for a sample of 120 UK and German companies, over the period between 2007 and 2011. 3 Content analysis is widely used in the accounting and finance literature to measure the quality of corporate governance disclosures (see for example, Beck, Campbell, & Shrives, 2010).Analysing 600 sample reports for 120 firms in two countries, between 2007-2011, we identified instances of compliance/noncompliance in the corporate governance reports and developed a "comply or explain" index for each firm.A higher score on the "comply or explain" index shows a higher level of compliance with the respective corporate governance codes as well as a higher quality of corporate governance disclosure (in terms of the quality of reported explanations).We explain our index construction procedures for a hypothetical company.For instance, if company A is non-compliant with three provisions of the corporate governance code but provides 'no explanation' for each deviation, the "comply or explain" index will be: In order to operationalise the development of "Comply or explain" index, the first author manually read a sample of 600 corporate governance reports for 120 companies from both countries.Corporate governance codes in both countries have different requirements, and different numbers of provisions, and this was also one of the reasons for doing a separate regression analysis in subsequent section.The 2006 version of the governance code in the UK has a total number of 50 provisions while the German Corporate Governance Code of 2006 has a total number of 71 requirements.As the number of provisions are different, it is expected that the level of compliance will be different and therefore a simple compliance/noncompliance index will be not be useful for comparability reasons.Instead we use the weighted average index developed by Hooghiemstra (2012).Several steps were used to develop the "Comply or explain" weighted average index.First, each corporate governance report was carefully read to determine whether a company is fully compliant or there are instances of compliance.In the second step, information was collected about the number of provisions with which the company has reported noncompliance.The corporate governance codes in both countries require explanations in response to non-compliance.The denominator in our "Comply or explain" index includes the total number of explanations reported by a non-compliant company.The next step was to look individually into those explanations reported in response to non-compliance and categorise them based on the degree of informative of those explanations.In Table 2 below we develop a taxonomy of those explanations based on their degree of informativeness.An uninformative explanation in response to non-compliance gets a lower score and an informative explanation receives a higher score on the "Comply or explain" index.It is possible that one company has several instances of non-compliance and different types of explanations are reported for each instance of non-compliance.In the above example, a company has reported a total number of three explanations and all those three explanations are classified as 'no explanations for non-compliance (see Table 2 below for definitions of each category in our "comply or explain" index).The coding scheme for each category in Table 2 is developed in line with the prior research [see for example, Hooghiemstra, 2012).
We also include control variables such as firm size, foreign listing, firm-specific risk, and research and development expenditure (R&D).We use the natural logarithm of the book value of a firm's assets to proxy for firm size (Aggarwal et al., 2010;Chhaochharia & Laeven, 2009;Sarhan, Ntim, & Al-Najjar, 2019).We also control for a firm's foreign listing.In line with prior research (e.g., Beiner et al., 2006;Schultz et al., 2010), we use firm-specific beta as a proxy for firm-specific risk.R&D has been reported by the majority of companies in the UK and Germany.The inclusion of R&D expenditure is captures a firm's growth opportunities.Consistent with the governance-performance research (e.g., Aggarwal et al., 2010;Ammann et al., 2011;Bauer et al., 2004;Beiner et al., 2006;Chhaochharia & Laeven, 2009;Gompers et al., 2003;Klapper & Love, 2004;Yermack, 1996), we use Tobin's Q as a proxy for a firm's market valuation.Tobin's Q is calculated using the following formula: (Total assets + Market value of equity -Total common equity -Deferred taxes)/Total assets.
The choice of our Tobin's Q measure is consistent with the well-established governance-performance work of Aggarwal et al. (2010), Gompers et al. (2003), Bhagat and Bolton (2008), Bebchuk, Cohen, and Ferrell (2009).There is no doubt that there are more "sophisticated" approaches used in prior research to calculate Tobin's Q.However, owing to the cross-country nature of our dataset we applied a more simplistic, and representative measure of the market valuation, for which the data was available for most of the sample companies in both countries.The above Tobin's Q measure exclude deferred taxes which resulted in a more refined and accurate measure of a firm valuation.As deferred taxes are taken out it has further diluted the values in the numerator, and as a result we have generally lower Tobin's Q values for most of our sample companies.

| Data and sample
The data sample (see Tables 3 and 4) includes 60 nonfinancial firms from MDAX 50 and DAX 30 in Germany for the period ending 2007-2011.MDAX 50 and DAX 30 are the indices for the largest 80 companies in Germany.For the purpose of comparability, 60 non-financial firms in the UK were selected based on their size and industrial classification of their corresponding German counterparts.In line with other studies in the area, financial firms and utilities are excluded from our sample as they have to comply with additional regulatory requirements which may affect their performance.
Owing to the lack of availability of governance data for small-sized German companies, the scope of the sample is limited to only large companies.The compliance statements of small size German companies are (a) not available, and (b) if they are available, they are in The total number of explanations reported by a firm, in response to noncompliance with different provisions of the Code.This measure is used in the denominator in our "comply or explain" index.

Note:
The above taxonomy is developed based on the prior work of Hooghiemstra (2012).
German, and hence it was not possible for the research team to translate those.Owing to the accessibility and readability issues relating to the compliance statements for German companies, we only focused on medium size (MDAX) and large size (DAX30) German companies as these annual reports, compliance statements and corporate communication documents are available in English.
Corporate governance and financial data were extracted from Datastream (now Thomson Eikon), whereas blockholders' ownership data were collected from Thomson One.All 600 corporate governance reports were downloaded and a manual mechanistic content analysis was carried out to develop a "comply or explain" index.Table 5 presents a size based comparison of our sample firms, which indicate that the "size differences" of UK and German firms in our sample are statistically insignificant.Following the matching sample criteria used in the literature (see, Peasnell, Pope, & Young, 2001), the total assets of a comparable UK matched firm were chosen in the range of ±25% of the total assets of a corresponding sample German firm.In Table 5, we also used other size-based measures (number of equity shares, number of employees) when choosing corresponding matching firms in the UK, and mean values for all these relevant size based measures are insignificant.We did not use exactly the same number of firms from the same industry in each country because of the unavailability of the same number of firms in the chosen industries.
A detailed discussion on the results of our empirical analyses is presented in the following section.

| Univariate analysis
Table 6 presents descriptive statistics and outcome of the univariate analysis of this study.The "comply or explain" index is significantly higher for the UK as compared to the German companies, indicating a higher level of compliance for the UK firms.In terms of board size, the results show that German companies have significantly larger boards where the maximum value for the board size is recorded as 22.One reason for a larger board size could be the two-tier board structure, whereas another reason is the German codetermination law, which requires that one half of the supervisory boards of listed companies should be comprised of employees' representatives (Du Plessis et al., 2012).German companies also have a significantly higher percentage of non-executive directors on their boards as compared to their UK counterparts (UK = 45.703;Germany = 83.910).The UK companies have a significantly higher number of board meetings, with the average number of annual board meetings for the German and UK firms being 5.923 and 8.793, respectively.German companies have a significantly higher gearing ratio (Germany = 0.254; UK = 0.223), which supports the argument that German banks contribute significantly to the German corporate governance system (Gorton & Schmid, 2000).
The UK companies have a significantly higher percentage of institutional blockholders' ownership (UK = 14.153%;Germany = 10.130%), while German companies have a significantly higher percentage of noninstitutional blockholders' ownership as compared to UK firms (UK = 8.468%; Germany = 38.456%).One of the key differences between corporate governance systems in the UK and Germany is the ownership structure of companies.The empirical literature also shows that German companies have a highly concentrated ownership structure (Goergen, Manjon, & Renneboog, 2008), and that non-institutional blockholders play an important role in the German corporate governance system.
In terms of firm size and firm-specific risk we did not find any significant differences between the UK and German companies.In addition, the UK companies have a significantly higher R&D expenditure as compared to German companies (UK = 0.044; Germany = 0.023) which pinpoint the significantly high level of innovation, spending on new product development, and related initiatives in UK organisations as compared to their counterparts.Table 7 below also shows the percentage of shares owned by German corporations in the sample German companies.Table 8 reports the average values of the "comply or explain" index for sample industries in both countries.In Germany, construction and materials companies have the highest "comply or explain" index, followed by automobile and parts, while in the UK, companies in the personal and household goods, and food and beverages industry have the highest score on our "comply or explain" index.
In Table 9 we use a non-parametric test to compare the median values of firms across these two countries.We used a two-sample Wilcoxon rank-sum test, and the Mann-Whitney p-values are reported in Table 9.The results from the non-parametric test reveals significant difference in board structures (board independence of companies in the UK and Germany, with German companies having higher percentage of NEDs in their boardrooms).
We now discuss the outcome of our regression analyses in section below.

| Regression analysis
As the most commonly used method of estimation, we employ ordinary least squares (OLS) regression analysis as the first technique for estimating our models.The results of our OLS estimations are reported in Table 10.In order to check the consistency of OLS regression results we follow Wintoki et al. (2012), and apply the Durbin-Wu-Hausman test for endogeneity to our models.
As reported in Table 11, the results of the Durbin-Wu-Hausman test turns out to be significant which imply that the OLS estimations are inconsistent owing to the presence of endogeneity problems.
We therefore control for minimising the impact of endogeneity and carried out a revised analysis using the dynamic generalised method of moments (GMM) estimation.The output of our GMM estimation is reported in Table 12.
The results in in Table 12 suggest that the "comply or explain" index is significantly positively associated with the Tobin's Q for the UK firms.This is consistent with the findings of Arcot and Bruno (2011) which show a positive association between governance compliance and firm performance in the UK.This would mean that compliance with the recommendations of the corporate governance codes (or non-compliance when valid justification is provided) has a positive effect on the market valuation of firms (see for example, Beiner et al., 2006;Gompers et al., 2003;Klapper & Love, 2004, Honisberg, 2019).The findings also support the idea of self-regulation through the applications of a 'comply or explain' principle in the UK which is working well in its existing state.
As the index captures both the level of compliance and quality of explanations a positive relationship will imply two things: first that compliance has a positive impact; and second that good quality explanation means that a firm has adopted good alternative governance practices which the capital markets regard as an as alternative to governance compliance.This is consistent with Luo and Salterio (2014) who show a positive impact of corporate governance compliance and a disclosure index 4 on the market value of Canadian firms.For German firms, this relationship is also positive but statistically insignificant.We therefore find only partial support for the contention that compliance or justified in accordance with explanation with the German Corporate Governance Code positively affect the market valuation of German companies.These findings indicate that external disclosure quality is more value relevant in a dispersed ownership system, the UK.However, in the context of Germany, the representation of employees and shareholders on the corporate boards of German companies bridge the information asymmetry gap between firms and their stakeholders.This is because most of the stakeholders will have some input in the governance mechanisms of German companies.Therefore, the importance of compliance or quality of disclosure might be perceived as less important by various stakeholders.
The results show a significantly negative relationship between board size and the financial performance of UK firms.The negative relationship between board size and Tobin's Q is in line with the results of existing literature (Yamori et al., 2017;Guest, 2009).For the German sample, we did not find any significant impact of board size on the market valuation of firms.The results of our univariate analysis indicate that board size of German firms is significantly larger than UK firms.However, the characteristics of boards are quite different in the UK from Germany.In the German context larger boards could signal a better representation from the key stakeholders of an organisation, such as employees, banks, and shareholders which is expected to produce a positive impact on the valuation of companies.However, as reported by Guest (2009), the existence of larger boards in the UK firms could indicate more agency costs and a negative relationship is therefore observed between board size and firm value in UK companies.
The relationship between board independence and Tobin's Q is shown as significantly negative.This is consistent with the findings reported by Weir and Laing (2000) for the UK firms, and Agrawal and Knoeber (1996), and Francis et al. (2012) for US firms Similarly, while examining the causes and impact of the 2007-2008 financial crisis the effectiveness of nonexecutive directors on corporate boards has also been recently questioned by researchers.For example, Aebi, Sabato, and Schmid (2012) show a negative relationship between the presence of a high percentage of NEDs on corporate boards and the financial performance of US firms.In a recent study, Moursli (2020) documents that board independence negatively affects market valuation of Swedish firms and this negative relationship can be explained by the busyness of such directors.Interestingly, for German firms, we find a significantly positive relationship between board independence and the market valuation of firms.Comparing the UK and German boards, it is evident that German board structures are perhaps more complex with a two-tier system and a wide range representation on boards from employees, banks and shareholders.This positive relationship between the percentage of NEDs on boards and the market valuation of firms supports the assumptions of agency theory and resource dependence theory.It also indicates that the presence of NEDs on German firms' corporate boards represent the key stakeholders and their role is more effective in monitoring and advising the board of directors.This finding implies that the market may perceive the role of non-executive directors differently in different countries, and consequently, corporate governance mechanisms may not yield the same results in different corporate governance systems (Van Essen et al., 2013).This finding would question the true independence of the corporate boards in UK firms because a truly independent board is theoretically expected to reduce agency costs and show a positive association with firm performance.We find a significantly positive relationship between the number of board meetings and the market valuation of both the UK and German firms.This finding is consistent with Brick and Chidambaran (2010) which show a positive relationship between the number of board meetings and organisational performance.The findings indicate that an increase in board activities (measured by the number of board meetings) enhances a firm's monitoring and control functions and is therefore regarded as significant in firms' operations and financial performance Moreover, our results show that gearing has a significantly positive impact on the market valuation of the UK and German companies.This finding is consistent with the findings reported by Goncharov et al. (2006) for a sample of German firms, and, McKnight and Weir (2009) and Dahya et al. (2002) for a sample of UK firms.The significantly positive impact of gearing on the firm market value indicates that capital markets in both the UK and Germany consider gearing as an important strategic factor in organisational corporate governance systems.
The results for the blockholders' ownership show that institutional blockholders' ownership has a negative impact on the market valuation of UK and German firms.This is also consistent with the findings of Gugler et al. (2008) where they compare the Anglo-Saxon and relationship-based corporate governance systems.In the context of the UK, Mura (2007) has also reported a significantly negative relationship between institutional blockholders' ownership and the market valuation of UK firms.More recently, regulators in the UK have raised concerns over the monitoring role of the UK institutional shareholders.In order to address this issue, the Financial Reporting Council issued The UK Stewardship Code for the institutional investors in 2012 (Financial Reporting Council, 2012b).The above-mentioned negative relationship between institutional blockholders' ownership and the market valuation of firms suggests that the capital markets in the UK and Germany do not consider institutional investors as effective monitors.
Interestingly, non-institutional blockholders' ownership has a significantly positive impact on the market valuation of UK and German firms.This finding is consistent with Lehmann and Weigand (2000) for German firms, and Short and Keasey (1999) for the UK sample.This outcome of this finding would imply that non-institutional blockholders (individuals, families, corporations) exert a significant influence in monitoring their investee companies.One possible explanation for different findings for non-institutional and institutional blockholders could be, that non-institutional investors (families', individuals' and others') investment is highly concentrated in a few firms, and because of their higher investment stakes and substantial exposure to risk, they closely monitor the performance of their investees' companies.On the other hand, institutional investors would usually hold shares in many companies so their exposure to an individual firm would be limited which weakens their incentive to monitor and scrutinise individual companies.This could also indicate that institutional investors invest others' money in these companies while non-institutional investors invest their own money.Therefore, the stake of non-institutional investors could be considered much higher as compared to institutional investors.In fact, the UK regulators have recently issued a separate code for institutional investors in the UK which is called the UK Stewardship Code.We therefore argue that our findings about the negative relationship between institutional shareholders' ownership and market valuation of firms confirm the concerns and criticism over the weak monitoring role of institutional investors highlighted in the UK Stewardship Code.
In relation to the control variables we find that firm size has a positive association with the market valuation of UK and German firms.This finding supports the contention that larger firms have a higher market valuation from the capital markets in both countries.Foreign listing has a significantly positive impact on the market valuation of German firms; however, for UK companies, the relationship between foreign listing and the performance of firms is statistically insignificant.Firm-specific risk (beta) has a significantly positive impact on the market valuation (Tobin's Q) of the UK and German firms which is consistent with the findings reported by Beiner et al. (2006).As expected, R&D expenditure has a significantly positive impact on the market valuation of firms for both the UK and Germany companies.This suggests that investment in R&D expenditure is value-relevant and investors are willing to pay a premium for companies with higher spending on R&D.Moreover, our findings are robust when we use an alternative measure for the "comply or explain" index and an additional explanatory variable in our analyses.

| Does the quality of CG compliance/ non-compliance explanations depend on the CG structures?
The primary focus of this research is to understand whether governance mechanisms, compliance, and the quality of governance disclosure have implications for firms' market valuation in the UK and Germany.As a supplementary analysis, we tested the determinants of corporate governance compliance and disclosure.In doing so, we included the "comply or explain" index as the explanatory variable, and firm-level corporate governance attributes as control factors.Instead of using traditional panel data models, we used ordered logistic regressions for our additional analysis, and the results are reported in Table 13.One particular reason for using this approach is that our dependent variable "comply or explain" index has a meaningful order.The "comply or explain" index takes a minimum value of one and maximum value of 5 and hence ordered logistic regression was a most suitable methodological choice for this type of investigation.The results for UK firms indicate that institutional blockholders and debt financing have a significant relationship with the "comply or explain" index which shows the monitoring role of external capital providers.As expected, firm size has a positive relationship with the "comply or explain" index which indicates that larger firms tend to be fully compliant or will provide high quality disclosure, and this is in line with the traditional governance-performance research in comparative international context (Aggarwal et al., 2010;Chhaochharia & Laeven, 2009;Klapper & Love, 2004).In the case of Germany, non-institutional blockholder ownership is negatively related with the "comply or explain" index.In the context of Germany, firms with foreign listing have a positive relationship with the "comply or explain" index which is consistent with the notion that foreign listed firms are subject to additional disclosure requirements in overseas regimes and hence they are more likely to comply with domestic regulations as overseas listing regulations could often be more stringent than host country's regulations, depending on the jurisdiction(s) where company is listed.Contrary to our expectations, we could not find any significant relationship between board attributes and the "comply or explain" index in both countries.

| Robustness tests
Consistent with the previous studies using commercially available ratings (such as Aggarwal et al., 2010;Bauer et al., 2004;Gompers et al., 2003;Klapper & Love, 2004), we also use commercially available ratings (the Datastream Corporate Governance Score) to test whether the results reported under the "comply or explain" index are robust when an alternative measure of the quality of a firm corporate governance is used (see Table 14 for details).According to Datastream "a corporate governance score is a number between 0 and 100 showing how the company performs compared with the entire ASSET4 5 universe based on the 'value' in the related indicator".Sales growth was also included as an additional control variable.Consistent with prior literature (see for example, Chhaochharia & Laeven, 2009;Drobetz et al., 2004;Gompers et al., 2003), sales growth is calculated as: current year's sales minus previous year's sales divided by previous year's sales.It is expected that there will be a positive relationship between sales growth and the performance of firms.
Table 14 shows that our findings are robust when we use additional control variables and an alternative measure for the quality of corporate governance.Except in the UK sample (refer to Model 5), where the impact of board size and board independence are now significant at the 10% level only, as compared to the 1% significance level reported in Table 8.In addition, sales growth is significantly positively associated with Tobin's Q for the UK and German companies.In addition, the post estimation tests, including the Sargan test and the Arellano-Bond test for auto-correlation suggest, that the instruments/ models are valid.

| CONCLUSION AND IMPLICATIONS
Over the last three decades, differences in the corporate governance systems of different countries have been widely explored and debated in the accounting, finance, economics and corporate governance literature.However, very little is known about the implications of these differences for firm performance and actual corporate governance practices in different countries.This study examines the impact of differences in the corporate governance regulations in UK and Germany and examines their impact on firms' governance practices and performance in the two countries We develop a "comply or explain" index which not only captures the level of compliance with the governance regulations but also the quality of explanations given for non-compliance in UK and German organisations.Using a sample of 120 German and UK listed companies and through the application of both the univariate and multivariate analyses, this study contributes to the existing literature in two ways.First it explores whether there are differences in corporate governance mechanisms of companies in the two countries, and then examines how compliance, the quality of explanations provided in non-compliance disclosures, and various other internal corporate governance mechanisms, affect the market valuation of companies in the UK and Germany.
The results of our univariate analysis reveal significant differences in the corporate governance mechanisms, ownership structure, and control procedures of UK and German companies.The differences in corporate governance mechanisms have implications for the market valuation of firms in the two countries.For example, the "comply or explain" index has a significantly positive impact on the market valuation of UK firms.This would mean that fully compliant firms and firms that have implemented strong alternative corporate governance mechanisms have a higher market valuation in the UK.Our governance index captures compliance and also assigns a higher score to non-compliant firms that have implemented their own strong governance mechanisms over and above those mechanisms which are prescribed by the regulators.The "comply or explain" index is, however, insignificantly associated with the market valuation of German firms, which indicates that investors react differently to the non-compliance disclosure in two countries.This may indicate the cultural aspect of the user perception in a particular jurisdiction because firms in Germany are generally expected to comply with the given regulation.
Our findings also highlight that the impact of board size on market valuation is different in UK and Germany.While a negative relationship is shown between board size and the market valuation of UK firms, no significant relationship of this kind is indicated by our findings for German firms.This finding implies that similar (but not identical) corporate governance mechanisms may have different implications for firms in different jurisdictions.We argue that, because of the two-tier board structure and various stakeholder representations on corporate boards of German firms, the negative impact associated with larger boards is mitigated.This finding thus suggests that the negative impact of larger corporate boards depends on the characteristics and types of the directors on the board and the overall board structure.It is also evident from the findings that the board size of German firms is significantly larger than the UK firms due to which the characteristics of corporate boards are quite different in the two countries.It is therefore argued that in case of Germany larger boards could signal more representation from the key stakeholders of an organisation, such as employees, banks and shareholders.In contrast, as the regulatory requirements on board size in the UK is different, larger boards could result in more agency costs.These findings therefore present policy implications for the regulation about board size and structure in the UK and other countries.
In relation to the positive impact of gearing on the market valuation of UK and German firms, our findings support the notion that the level of gearing in firms' capital structure serves as an important corporate governance mechanism in both corporate governance systems.We also find that non-institutional blockholders play a significant monitoring role in the German corporate governance system.Interestingly, and contrary to our expectations, institutional blockholders' ownership has a negative impact on the market valuation of firms in both countries.In relation to this finding we argue that as institutional investors hold shares in many companies, so their attention exposure to an individual firm may be limited which would reduce their incentive to monitor and scrutinise individual companies in their portfolios.The negative relationship between institutional blockholders' ownership and market valuation of firms confirms the concerns and criticism over the weak monitoring role of institutional investors highlighted in the UK Stewardship Code (2016).It is therefore argued that, with respect to the role of institutional investors, our findings have policy implications and present useful insights for governments and regulators in different countries.
As our research focuses on the often-ignored aspect of the "comply or explain" principle, these findings have implications for practitioners and regulators.We argue that both compliance and the explanations given for noncompliance are equally important as long as companies can offer valid reasons and justifications for the noncompliance with the corporate governance regulations.The findings also highlight that the "comply or explain" principle is working well in both the UK and Germany and that companies in both these countries could benefit from the flexibility offered by this principle.We argue that the mixed and inconclusive empirical evidence on the relationship between governance and firm performance in the existing literature would indicate that the governance-performance relationship cannot be examined through the lens of a single and universal theory of corporate governance.We therefore propose a multiple theoretical perspective in this area of research and argue that this approach could be helpful in examining the governance-performance relationship in different corporate governance settings.In fact, investigating the complex nature of the governance-performance relationship through the application of multiple theories and multiple methods may take us closer to developing a more comprehensive theory of corporate governance.Our findings therefore support the free market perspective on corporate governance regulation and present policy implications for the introduction of "comply or explain" based governance principles in different countries.
Despite the significant contributions that this study has made to existing literature, we also acknowledge some of the limitations of this research.First, despite making substantial efforts in the data collection process, we only managed to collect data for 120 companies, and inferences are therefore based on a relatively small sample in both countries.We therefore argue that the context of both the UK and German companies, a larger sample would contribute more to the debate on the relationship between the "comply or explain" principle and the market valuation of UK and German companies.Second, as the governance regulations and compliance requirements in other European countries are different, it would have been better to include other major European countries, such as: France, Italy, Spain, Netherland, Belgium, etc., in the analyses, because inclusion of such countries would certainly add more insights to our understandings and the debate on these issues.Third, other board characteristics, such as directors' education, ethnic background, gender diversity, directors' time with the company, and firm's age would certainly produce useful insights on the relationship between the "comply or explain" principle and the market valuation of firms in both the UK and Germany.However, due to time and resource constraints, coverage of all the above mentioned avenues is out of the scope of this paper, and is therefore left to future research.

ENDNOTES
1 https://www.frc.org.uk/getattachment/56920102-feeb-4da7-84f7-1061840af9f0/Combined-Code-Web-Optimized-June-2008.pdf 2 https://www.dcgk.de/en/code/archive.html 3 Although the end date of the crisis is not universally agreed in any case, Lu and Whidbee (2013) argue that the financial crisis ended in 2011. 4Similar to the "comply or explain" index, the index used by Luo and Salterio (2014) also measures compliance as well as the quality of explanations reported by non-compliant firms.

|
Econometric specificationsgeneralised method of moments model (GMM) Recent research has raised serious concerns about the econometric techniques applied in prior corporate governance studies (see for example, Abdallah, Goergen, & O'Sullivan, 2015; Schultz et al., 2010; Ullah, Akhtar, & Zaefarian, 2018; Wintoki et al., 2012).The governance-performance Internal Corporate Governance Mechanisms • "Comply or explain" index • Board size • Board structure • Number of board meetings • Gearing • Institutional blockholders • Non-institutional blockholders Control Variables • Firm-specific risk (Beta) • Firm size • Foreign listing • Industry dummies • Research and development expenditure Internal Corporate Governance Mechanisms • "Comply or explain" index • Board size • Board structure • Number of board meetings • Gearing • Institutional blockholders • Non-institutional Conceptual framework [Colour figure can be viewed at wileyonlinelibrary.com] Definition of various categories of explanations used in the "comply or explain" index Descriptive statistics and univariate analysis Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijfe.2053 by Test, Wiley Online Library on [10/02/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Note:The values for assets and market capitalisation are reported in a common currency (the euro).Differences in sample mean values are reported using t-statistics.The mean differences between all firm-specific characteristics are statistically insignificant.T A B L E 6Note: T-tests are used to compare the mean values of UK.
T A B L E 7 Percentage of shares owned by German corporations in the sample German companies Mean values of the "Comply or Explain" index across industries in the UK and Germany T A B L E 9 Non-parametric test: Two-sample Wilcoxon rank-sum (Mann-Whitney) test Note: Mann-Whitney p-values are used to compare the median values of UK and German firms.
Definitions of variables are provided in Table 1.This table reports Durbin-Wu-Hausman test statistics (abbreviated as DWH) for each independent/control variable used in the OLS models reported in Table 6 (e.g., Model 1 -Model 2).The null hypothesis states that all regressors (corporate governance mechanisms) are exogenous.The Durbin-Wu-Hausman test was carried out for all independent/control variables.STATA (1999) provides guidelines about how to carry-out a Durbin-Wu-Hausman test in STATA for each individual variable.*** p < 0.01, ** p < 0.05, * p < 0.1 stand for statistical significance at the 1%, 5% and 10% level, respectively.
Only one lag of the dependent variable is included in both models.AR (1) and AR (2) are the Arellano-Bond test statistics for first-order and second-order correlation, under the null hypothesis of no serial auto-correlation.The Sargan test statistic is a test of overidentifying restrictions, under the null hypothesis that all instruments are valid.Standard errors reported in parentheses.Definitions of variables are provided in Table1.*** p < 0.01, ** p < 0.05, * p < 0.1 stand for statistical significance at the 1%, 5% and 10% level, respectively.
Ordered logistic regression: Does the quality of corporate governance compliance/non-compliance explanations depend on the corporate governance structures?The table reports ordered logistic regression results.The dependent variable is the "comply or explain" index, and firm-level corporate governance mechanisms are included as explanatory variables.The coefficient for the UK and German sample are reported separately.