1
The relationship between board gender diversity and firm performance
International evidence
Master thesis
MSc in International Financial Management
08-06-2018
Bassel Alhaj Darwish S3131017
2
Keywords: Board gender diversity, Corporate governance, country legal origin, Firm performance, women on boards.
Abstract
3 Table of Contents
1. Introduction ...5
2. Theoretical framework and hypothesis. ...8
2.1 Board gender diversity. ...8
2.2 Board gender diversity and firm performance ... 11
2.3 Corporate governance and the board gender diversity ... 15
2.4 Country legal origin and board gender diversity ... 18
3. Methodology and Data Collection ... 21
3.1 Variables ... 21
3.2 Sample and data collection ... 24
3.3 Methodology ... 28
4. Empirical results ... 29
5. Robustness check. ... 34
6. Discussion and Conclusion ... 38
7. Managerial implications ... 39
8. Limitations and avenues for Further research... 40
9. References... 42
4 Table of figures:
Figure 1 Countries that have adopted quotas……….…….……….………...…….11
Figure 2 Corporate governance and the balance sheet model ………..…...…….17
Table of tables: Table 1 Gender quota legislation by country ………....10
Table 2 Literature summary ………..………....14
Table 3 Variables description ………..….…... 23
Table 4 Summary statistics ………..……... 26
Table 5 Correlation matrix ………... 27
Table 6 Regression Results regardless of moderating effects……….. 30
Table 7 Regression Results, firm and country moderating effects….……….. 31
Table 8 Regression Results with and without quota countries………...……….. 33
Table 9 Regression Results using board gender diversity dummy variable………...….. 35
5
1. Introduction
The attainment of board gender diversity has become a salient issue for recent academic studies since it is being considered a value-driver strategy and an important corporate governance mechanism (Bebchuk & Weisbach, 2010). The positive impact of board gender diversity on firm performance translates into more females on top management positions or on the board of directors increasing corporate efficiency and profitability.
About three-quarters of corporate board members of larger American firms are men. Women hold only 16.9 % of the seats on fortune 500 boards. Only 11.9 % of boards in Russel 3000 companies are females (Rhode & Packel, 2014). The interest in appointing women in top management positions has been increasing in the literature since Catalyst (2004) has found that companies with more women on their boards achieve better financial results. Board gender diversity is therefore considered as a part of firms’ corporate governance.1 Institutional investors are paying more attention to the governance of listed companies, in pursuing the financial and social goals of stakeholders (Francoeur et al., 2008). In addition, the attention in corporate governance has been quickly growing inside and outside the academia, this interest in corporate governance has been interdisciplinary with more work being undertaken by researchers from economics, finance, law, management, and accounting.2 In spite of the multiple academic investigations, the results remain mixed. The existence of females on boards reportedly causes more focused strategic decision-making process, and fewer board conflicts (Rhode & Packel, 2014).
This thesis examines the term of board gender diversity by testing, in an international sample, if female presence in the board affects the financial performance of the company. The impact of board gender diversity on the firm performance has been studied by many researchers in the existing literature, because of the accelerating importance of board diversification. I build on the agency and stakeholder theoretical framework to develop my main hypothesis relating to board
1 Corporate governance can be defined as referring to the way in which suppliers of finance assure themselves return on their investment (Shleifer & Vishny, 1997). The term corporate governance can have different aspects, like ownership and shareholders activism, directors, executives and their compensation, comparative corporate governance, controlling shareholders, the political economy of corporate governance, and cross-border investments in global capital markets (Bebchuk & Weisbach, 2010).
6
gender diversity in boardrooms (Donaldson & Preston, 1995). However, Agency theory perspective cannot explain gender difference, which is based on the difference of board tasks on the firm value, because the gender of corporate director does not matter for the performance of board tasks (Nielsen & Huse, 2010). Whereas “stakeholders theory” or “stakeholder thinking” has appeared to be a new description to understand and cure three interconnected business issues. Firstly, the issue of understanding how value is created. Secondly, the issue of connecting ethics and capitalism, and thirdly, the issue of helping managers think and behave about management in the same framework that the first two issues are addressed. Certainly, boards of directors make significant decisions that can influence the employees, stakeholders, the community, and the organization itself.
Boards of directors appoint CEO’s to act to the advantage of the company. To fully perform this mission, organizations need good corporate decision-making process, which requires the ability to know and consider variant points of view. Those points of view come from people who have dissimilar experience, perspectives, and backgrounds. Women on boards can affect the firm's path through points of strengths like the diversity of thought, competitive advantage, stakeholder representation, and availability of essential skills.
This thesis expands this view to suggest that females on board may self-execute compliance with women gender role stereotype when it comes to the firm valuation issues, in order to balance the tension between a managerial and gender role stereotype which lower their performance. The aim of this thesis is to investigate the relationship between board gender diversity and firm performance with two moderating effects. Specifically, it tests for the effect of corporate governance and country legal origin. While many previous studies have examined the board gender diversity effects on performance at the national level, in this thesis I test the impact of the board gender diversity on firm performance at the international level.
7
The findings of this thesis contribute to the understanding of the role of moderating factors on the relationship between board gender diversity and firm performance, which can result in a better firm performance on a corporate level. Moreover, this perspective gives insights on the link between gender diversity on boards, firm corporate governance, country law origin, and firm financial performance.
What distinguishes this thesis from previous works is that it includes both corporate governance on the firm level, and country legal origin on the country level as moderators. Many studies have examined the link between board gender diversity and firm performance on the country level using different moderating factors, but these researchers have mixed findings explaining the link between board gender diversity and firm performance.
Furthermore, I investigate female board representative effects on a panel data of international corporations, whereas the past studies have investigated this effect on the domestic level. Despite that it is challenging methodology, panel data analysis with firms or country fixed effects allow me to control for non-observable variables, such as cultural and behavioral factors, differences in management practices among companies. It allows for more heightened capacity in modeling the complexity of human behavior than a single cross-section or time series data and factors that change over time but not among entities. In addition, panel data analysis helps to overcome the problem of data availability, by means of controlling the impact of omitted variables, which may hinder the achieving of correct results. (Hsiao, 2007).
8
examining the relationship between the legal system and firm performance over the time frame between 2002 and 2016.
The empirical results of this thesis show that board gender diversity is positively associated with firm performance on the international level. Tobin’s Q as a dependent variable shows a significant increase when more women are on boards. Furthermore, firm corporate governance score shows a positive and significant association with board gender diversity. But I couldn’t prove that it positively moderates the relationship between board gender diversity and firm value. In addition, contrary to my hypothesis, common-law origin countries have a negative effect on the relationship between board gender diversity and firm value, so it negatively moderates this relationship. Lastly, I performed a test on the countries with female quotas, and countries without female quotas or before they have a quota. Results show a negative effect of quotas on firm performance. A detailed discussion and explanation will follow.
The remainder of this thesis is structured as follows. Section 2 will discuss the theoretical framework and hypotheses, especially boards gender diversity and firm performance. Section 3 presents the methodology in performing the thesis and data collection process. Section 4 presents results of the empirical analysis. In section 5 I perform the robustness check. Section 6 will discuss the results and concludes the thesis. Section 7 declares the managerial implications of this thesis. Limitations and future research are presented in section 8.
2. Theoretical framework and hypothesis.
2.1 Board gender diversity.
Over the last decades, the international finance literature has studied the relationship between the difference in corporate governance and its consequences on corporate performance. Despite discordant results, everyone agrees that senior executives reflect organization’s actions. More specifically, business outcomes are driven by top management's characteristics and functioning.
idea-9
generating creativity, consequently, it is linked to more innovative firms. Rodriguez-Dominguez et al. (2009) show how gender heterogeneous boards achieve higher performance in control and monitoring activities. This diversification, allows scholars to derive divergent views to which it follows greater independence of the board of directors compared to the executive director’s division.
Wood & Eagly (2009) argue that beliefs can act as personal characters (embedded in individual gender identities) or social norms (embedded in other’s expectations). Galaskiewicz (1991) contends that such cultural norms and beliefs, mindsets, cognitive frames, worldviews, are essential determinants of the way that managers run their corporations. However, most studies on diversity and its consequences focus on demographic diversity or the observable diversity. Some researchers like Watson et al. (1993) suggest that board diversity leads to creativity, innovation, and greater knowledge base, which becomes a competitive advantage.
According to Gul et al. (2011), Peterson & Philpot (2007), Adams & Ferreira (2009), who investigated on the effects of proper gender diversification at the board level, a greater percentage of female administrators increase the monitoring roles at the higher layer of the organization. This improvement in the control functions leads to a better protection of the shareholders’ interests. And Erhardt et al. (2003) found a positive association between board diversity and financial firm performance in 127 large U.S companies. To conclude, decision-making’s key actors make better investment choices because of the presence of different background figures of the board members, who bring different perspectives to be taken into consideration.
As reported by Smith et al. (2006) the low presence of women on boards is considered as a significant weakness in the organization’s strategy. In fact, female gender qualified directors positively affect the firm performance. Even though in the 80s and 90s, organizations reported an increase in size at the board size, increasing the rate and the probability of women presence, because of the glass ceiling, there are still relevant barriers that decrease the male/female ratio at the top management (Akpinar, 2013).
10
2014). Norway was the pioneer by requiring publicly traded companies with corporate boards of nine or more board members to have at least 40% of female directors by 2008. The Netherlands and Spain have lately followed suit with regulation demanding firms have at least 40% female directors by 2015. Finland requires having 40% of women on the boards. Italy and Belgium require one-third of executives to be women. In 2017, France has adopted 40% quota. Germany, UK, and Sweden are negotiating similar legislation. In emerging economies countries like India and the United Arab Emirates, require certain firms to have females on their boards.
In contrast, a study on Norwegian companies found a negative effect of quotas on the performance (Adams & Ferreira, 2009). Concurrently, another study on Scandinavian companies has found no link between the board gender diversity and the firm performance (Bøhren & Strøm, 2007). Another study performed by Mukherjee (2017) conclude that board gender diversity does not have a negative effect on firm performance.
In Table 1, I summarize the countries who adopted a gender quota legislation in its firms, and quota implementation scope in addition to the year of implementation.
Table 1 Gender quota legislations by country.
Country Quota Companies Passage
date
Compliance date
AUSTRALIA - Non-public employers with at least 100 employees 2012 2014-2015
BELGIUM 33% WBD3 Publicly traded and SOEs 2011 2011-2012
BRAZIL4 40% WBD Publicly traded, SOEs, and financial institutions
2010 2022
CANADA At least 40% of each gender
Publicly traded, SOEs, and financial institutions
2014 2020
DENMARK Increase share of underrepresented gender
Large publicly traded, large private, and SOEs with
at least 50 employees 2006 2011
FINLAND 40% WBD SOEs 2005 2005
FRANCE 40% WBD Publicly traded or non-listed companies that have > 500 employees or revenues > €50 million. 2011 2017
ICELAND 40% of each gender
Publicly traded, private limited, and SOEs with at
least 50 employees. 2010 2013
INDIA At least one-woman director Publicly traded and every other public company 2013 2015
ISRAEL 50% WBD SOEs 2007 2010
ITALY 33% of underrepresented gender Publicly traded 2011 2012
KENYA 33% of each gender SOEs 2011 2013
3 “WBD” stands for Women Board Directors and “SOE” stands for State-Owned Enterprise. 4 Brazil, Canada and European Union have adopted women legislations to be applied after year 2016.
11
NETHERLANDS 30% WBD Publicly traded (250+ employees) 2011 2013
NORWAY 40% WBD Publicly traded and SOEs 2003 2006-2008
QUEBEC,
CANADA 50% WBD SOEs 2006 2011
SPAIN 40% of each gender Publicly traded companies with 250+ employees 2007 2015 EUROPEAN
UNION (EU)
40% non-executive WBD on
supervisory boards Publicly traded 2012 2020
Source: http://www.catalyst.org/
Figure 1 countries that have adopted quotas.
2.2 Board gender diversity and firm performance
12
decision-making process relates to different aspects such as gender diversity, board size, the portion of directors chosen by the employees, and age dispersion (Bøhren & Strøm, 2007).
The examples provided by Smith et al. (2006) intricate the discussion summarizing several of the positive theoretical arguments for gender diversity. First of all, female directors may better recognize special market circumstances than male directors. This better understanding may increase both the quality and creativity to board decision making. Secondly, appointing a woman to a particular executive position could increase the external talent pool for board members. Thirdly, a more gender diverse board may produce a better public picture of the organization, which improves firm performance. Moreover, career development of females in lower positions could be positively influenced by female top managers, this enlarges the internal pool of female candidates for top positions, which boost firm performance both directly and indirectly5.
Other studies have found a positive relationship between board gender diversity and several measures of firm performance. Erhardt et al. (2003) found a significant and positive link between gender and minority representation on boards, for both return on investment (ROI), and return on assets (ROA). In a similar way, a study for Adams & Ferreira (2004) in 1066 publicly traded companies found a positive and significant association between the part of female directors and financial performance, when they used Tobin’s Q as firm performance measurement. The same study has found no association between board gender diversity and Return on Assets (ROA).
Bonn (2004) found a positive association between the ratio of women directors and book-to-market ratio in Australian firms. Furthermore, researchers in other countries found a positive association between board gender diversity and financial firm performance based on different measures (Rhode & Packel, 2014). Campbell et al. (2008) proved a significant positive link in Spanish firms between board gender composition and Tobin’s Q. The same results proved by Nguyen & faff (2007) in an Australian sample. In China, Liu et al. (2014) document a significant and positive relationship between female executives and firm performance.
5In a study of Fortune 500 firms, Daily et al. (1999) conclude that women have made significant progress in terms of
13
Despite some studies claimed that board gender diversity leads to improved financial performance, such correlations do not demonstrate causation. And causal correlations are extremely difficult to prove. It could be that better firm performance leads to more women on board rather than the opposite. So that more successful companies may attract women and minority applicants for board positions. The reason may be that bigger and better performing corporations may have resources to dedicate to following the diversity and may face pressure to raise female board representation. In addition, both better performance and greater board diversity could be caused by some third factor (Rhode & Packel, 2014).
On the other hand, board gender diversity may have a negative effect on form effectiveness, through carrying more costs to the organization. Anconal & Caldwell (1992) discussed that any possible increase in firm performance may not be big enough to offset coordination costs of diverse top management teams; in addition to the difficulty of such coordination. Furthermore, the likelihood of reaching consensus may be smaller in heterogeneous boards, which slow down the decision-making process. The major result will be weaker decision-making entity, which possibly will significantly hinder firm’s competitive performance (Hambrick et al., 1996). Another study held by Jensen (1993) discussed the decreasing efficiency in controlling too big boards. And thus, separation of the ownership and control leads to more agency problems on the boards. In the setting of adhocracy culture, which is categorized by an external orientation and emphasis on individuality and competition, it appeared that gender diversity has a significantly negative impact on performance. Similarly, Frink et al. (2003) in a study of top 1500 US public firms, established an inverted U-shape relationship between gender diversity and firm performance. The findings of Dezso & Ross (2008) on US corporations reveal that appointing women CEOs had no impact on firm performance, but it is positively correlated with women involvement below CEO level.
14 Table 2 Board gender diversity literature summary.
Literature summary table of board gender diversity effect on firm market value Country
sample Observed effect Quota Author (s) Year Remarks
-- Positive -- Rhode & Packel 2014
Australia Positive No Bonn 2004
Australia Positive No Nguyen & Faff 2007
China Positive No Liu et al 2014
Denmark Positive No Bøhren & Strøm 2006 Norwegian sample
Denmark Positive No Smith et al 2006
Denmark No Link No Rose 2007
India Not Negative Yes Mukherjee 2017
International Negative -- Anh & Khanh 2017
Malaysia No-link No Adnan et al 2007
Malaysia Negative No Abdulla & Ismail 2013 Norway Negative Yes Adams & Ferreira 2009
Norway Negative Yes Kenneth et al 2013
Scandinavia No Link Yes Bøhren & Strøm 2007 Scandinavian Sample
Spain Positive No Campbell et al 2008
Spain Positive Yes Reguera et al 2015
US Negative No Hambrick et al 1996
US Positive No Daily et al 1999 Fortune 500
US Inverted U shape effect No Frink et al 2003 Top 1500 Firms
US Positive No Carter et al 2003 Fortune 1000
US Positive No Erhardt et al 2003
US Positive No Carter et al 2007 Fortune 500
US Positive No Peterson & Philpot 2007 Fortune 500
US No Link No Dezso & Ross 2008
US Positive No Adams & Ferreira 2004
US Positive No Gul et al 2011 Stock prices
US Positive No Brieger et al. 2017
Note: Quota denotes if the country complied with a gender quota legislation in the year of study.
15
Furthermore, Bøhren & Strøm (2007) found negative significant effects of board gender diversity on firm performance. Their study included a sample of Norwegian non-financial listed corporations. Moreover, other studies like Hambrick et al. (1996) suggest that diversity can be a disadvantage for the companies in terms of group performance.
To conclude, it seems that there is equivocal evidence about diversity effects of firm performance. From one side, it enhances performance by increasing the decision- making capacity, on the other side, it deducts from group performance by increasing conflicts and communication difficulties between board members. And negatively affects firm performance.
Based on the discussion above, I expect to find a positive relation between board gender diversity and firm performance, as documented in most recent studies performed with US, Australian, and Canadian data. Thus, I hypothesize that:
H1: Board gender diversity has a positive influence on firm performance.
2.3 Corporate governance and the board gender diversity
In the last three decades, the term corporate governance has developed noticeably as an independent field of study (Gillan, 2006). Which yielded a large number of definitions for corporate governance (Ntim, 2018). The Cadbury committee (Cadbury, 1992, p.15) defines corporate governance as the” …system by which companies are directed and controlled”. While the scope of corporate governance has been greatly expanded to be the combination of different disciplines including ethics, finance, management, law, accounting, politics, and organizational behavior, no universal definition was formulated to date (Mallin, 2002).
16
instruments – both internal and external - can be utilized to enhance firm value and/or performance for the joint benefits of shareholders and other possible stakeholders.6
Lie & Zaiats (2017) mention that firms adopt various governance provisions beyond the minimum threshold required for firms in a country. Such governance improvements exhibit a positive significant relationship with firm value, as shareholder rights relate positively to firm value. Further, Bebchuk et al. (2009) illustrate a significant positive association between corporate governance and firm performance, measured by Tobin’s Q or abnormal stock returns.7 La Porta et al. (1999) argue the different origins of laws and its consequences on corporate governance. In countries with low investor protection, profits can be easily stolen by insiders and consequently, outside parties are wary when financing a firm in this setting. On the other hand, when a firm is located in a good investor protection country, insiders are only able to over-reward themselves, assign relatives in management, and engage in inefficient projects.
The results of Klapper & Love (2004) propose that firms can partly compensate for ineffective laws and enforcement through good corporate governance, which provides protection for investors. They also show that firm-level performance and governance are lower in weak legal environment countries. And they suggest that policymakers should make a high priority to improve the legal system. Ngo et al. (2018) show that foreign shareholders tend to invest in firms that adopt stricter corporate governance to compensate for the existence of weaker investor guard and weaker governance instruments. These shareholders rely on governance instruments with which they are familiar to avoid expropriation risk for minority shareholders and to avoid the information asymmetry and the lingual and cultural barriers.
La Porta et al. (1999) discussed that there are common elements to explain how investors, are legally protected from expropriation by managers and controlling shareholders of corporations.
6 For example, Ntim (2018) and Lie & Zaiats (2018) explain that existing literature validates that corporate governance
is associated with high firm value and that separating cash flows and voting rights can maximize the firm value. In addition, the firm value is associated positively with cash flow rights of the largest shareholder but it is negatively associated with an instance of deviation between voting and cash flow rights. A huge literature also documented the connotation between firm value and corporate governance (Li & Zaiats, 2018).
17
These country elements like, financial systems, the pace of new security systems, corporate ownership structures, dividend policies, the efficiency of investment allocation, which can be explained by the laws succeed in protecting outside investors. The guard of shareholders and creditors by legal systems is essential to clarify the shapes of corporate finance in dissimilar countries. The US firm’s corporate governance index reveals that firms with stronger shareholders rights are more likely to make acquisitions have a higher market valuation, and better operating performance (Klapper & Love, 2004). In addition, higher corporate governance is associated with higher Tobin’s Q value (market valuation of the firm). NGO et al. (2018) found the firm value increase in association with stricter corporate governance principles, especially in terms of the independent board of supervisors, and the appointment of directors with multiple director seats.
As stated above, corporate governance influence firm performance through internal and external channels (Gillan, 2006). Concurrently, it affects the relationship between board gender diversity and firm performance indirectly through internal channels. Specifically, this mechanism functions through the activity of the board of directors. Since the board lies at the peak of internal control systems, it decides on major projects, investments, assets, and the general resource allocation. Moreover, it has the role of screening and advising the management and can hire, fire and reward the top management team. As Shleifer & Vishny (1997) state, capital providers ensure that they will get a return on their investments by imposing high levels of corporate governance. In other words, they tend to invest in organizations of higher corporate governance level, which has a positive effect on firm value through increasing the market value of the company.8 Figure 1 illustrates corporate governance and the balance sheet model of the firm.
Figure 2 illustrates corporate governance and the balance sheet model of the firm (Gillan, 2006).
18
One of the important dimensions of corporate governance is to impose a high level of board gender diversity (REF). As capital providers look to strengthen the governance of the firm, they are therefore should impose higher proportions of board gender diversity, together with additional prerequisites that are likely to improve the performance of the board. The higher levels of diversity within the board’s structure would, therefore, more easily lead to higher performance. Thus, I expect that the association between board gender diversity and firm value will be positively affected by the level of organization corporate governance. I, therefore, hypothesize that:
H2a: The level of corporate governance will strengthen the relationship between board gender diversity and firm performance.
H2b: The level of corporate governance will weaken the relationship between board gender diversity and firm performance.
2.4 Country legal origin and board gender diversity
A vital necessity in the design of a legal system is the guard of law enforces against coercion by litigants through either violence or bribes. The need for protection and control of the law enforces by the government increases when the risk of coercion increase. Even with substantial lawful evolution, these primary design choices have persisted for centuries (mainly because France remains less peaceful than England) and may clarify many differences between common and civil law societies with respect to both the structure of legal systems and the observed social and economic outcomes (Glaeser & Shleifer, 2002).
The developing trend of literature highlights that differences in legal origins across countries lead to a significant difference in laws that protect investors. The findings also explain that cross-country differences in ownership structure, dividend payout, cost of external finance and availability, and market valuations, stemmed from the differences in cross-country differences in laws and their enforcement effects (Klapper & Love, 2004).
19
The civil law relies more on professional judges, legal codes, and written records, whereas common law relies more on lay judges, broader legal principles, and verbal arguments.
Due to the recent research, there are significant differences between common law and civil law countries in the difference between economic and political conditions. La Porta et al. (1999) argue that French civil law countries experience weightier rules, more corrupt, less efficient governments, and weaker property rights protection. Moreover, they form less political freedom than do the states of common law origin. In addition, common law countries appear to be financially advanced than civil law states (La Porta et al., 1997). Because countries institutional framework evolves over time, Li et al. (2018) claim that a country’s legal tradition is a major driver of financial development, this applies to investor protection rights and investor protection laws. Since the work of La Porta et al. (1997) legal rights and legal origins have been considered to be close substitutes. Li et al. (2018) considered two factors, Anti director rights (ADRs) and creditor rights. Creditor rights protect debtholders from insider expropriation when the firm liquidates, whereas ADR ’s protect shareholders from insider expropriation. Dividend payout policy is a good example of their effects on firm financial policies. Firms which are located in Strong (weak) shareholder (creditor) rights are more (less) likely to pay dividends (La Porta et al., 2000; Brockman & Unlu,2009). Furthermore, La Porta et al. (1997) provide one of the earliest explanations for the role of creditor and shareholder rights in financial development. And that legal origins, the rule of law, and ADR, (Scandinavian and French) are significant factors in explaining the market capitalization.
20
less ex-ante limitations on management performance, and prefer shareholder protection (Liang & Renneboog, 2017).
Common law countries are usually characterized by strong shareholder protection, good accounting standards, lower ownership concentration, and lower private benefits of control (La Porta et al., 1999). Whereas in other law countries (civil law countries), the weaker legal guard paid to shareholders leads to higher probability of minority shareholders wealth being expropriated by controlling shareholders. In civil law countries, firms tend to give more attention to wider social concerns and implement stakeholder orientation because state involvement in economic life via rules and regulations is higher (La Porta et al., 2008; Liang & Renneboog, 2017).
The protection of outside investors differs systematically across legal origins. For instance, in common law countries, both shareholders and creditors are highly protected. On another hand, they are not in French civil law countries. German civil law counters come in the middle. Compared to French civil law countries, German civil law countries have a stronger protection of creditors. The same applies to Scandinavian origin countries (La Porta et al., 1999).
Generally speaking, the proposition that some countries guard all their outside investors better than other countries describe the difference between legal origins, and not that some countries protect creditors and some protect shareholders (La Porta et al., 1999). La Porta et al. (1997) given that markets in countries with higher level of development are expected to be more efficient and there is sufficient legal support to protect shareholders’ rights (Li & Oian, 2013).
In countries where legal system pays more protection to shareholder rights, shareholders ability to exert influence on firm’s strategic decision is greater. In this context, shareholders will influence the board composition, through forceful actions to have more women on boards to maximize their interests. Shareholders’ interests are maximizing their equity value, which leads to an increase in firm’s value. Furthermore, stronger legal guard of shareholder rights and developed accounting standards cultivate the growth of capital markets.
21
Based on the differences between civil law and common law counties, I expect that country legal origin, has a moderating effect on the relationship between board gender diversity and firm performance. Thus, I hypothesize the following.
H3a: Countries with a common-law legal origin will strengthen the relationship between board gender diversity and firm performance.
H3b: Countries with a common-law legal origin will weaken the relationship between board gender diversity and firm performance.
3. Methodology and Data Collection
This section presents the thesis method and sample of the thesis. The design of this quantitative thesis is based on a collection of secondary data. To test the hypothesis, I use panel data OLS regression. The time-frame of my investigation covers the period from 2002 to 2016. 3.1 Variables
Firm performance
22
The following formula illustrates the way in calculating the market value of the firm.
𝑇𝑄 =
𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑒𝑞𝑢𝑖𝑡𝑦+𝑏𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑡𝑜𝑡𝑎𝑙 𝑑𝑒𝑏𝑡𝑠𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠 (1)
Board gender diversity
Brieger et al. (2017) have concluded that a board of directors is more diverse when the percentage of women increases. But board gender diversity can be measured by two ways. Firstly: the percentage of women on boards. Secondly: using dummy variables, indicates (1) if there are female directors on the board, or (0) if not. In this thesis, I use the first way for the empirical analysis, and the second way to perform robustness check.
Control variables influencing firm performance
Board size has been reviewed by Hermalin & Weisbach (2003) who revealed the existence of a mainly negative correlation between board size and the firm performance. Another issue related to agency problem between owners and managers is the share of independent directors, which has an influence on firm performance. So, I will include both of it as a control variable.
23 Firm-level moderator variable
Corporate governance is corporate governance proxy for firm-level moderator,𝐶𝐺𝑖𝑡 is corporate governance variable of firm i at time t.
Country-level moderator variable
Country-level variable in this thesis is country law origin. I divide the whole sample into two groups. Common-law origin countries, and civil-law origin counties. Based on this classification, the proxy for country law origin is dummy variable takes the value (1) if the organization headquarter located in common-law origin country and (0) otherwise. I named this variable 𝐶𝐿𝑂𝑗. I used the classification of Djankov et al. (2007) to address common and non-common law countries. In addition, I used a gender quota legislations country dummy variable to capture the effect of quota on performance 𝑄𝑢𝑜𝑡𝑎𝑗.
Table 3: Variables description
Variable Description
Tobin’s Q Tobin’s Q which is the proxy for the firm performance.
Board gender diversity It is the board gender diversity calculated as the percentage of female presence in the BoD
from ASSETS4.
Board independence It is the board independence control variable for the board which is the proportion of independent directors to the total number of board directors.
Board size Board size. Measured as the total number of managers on board.
Chairman ex CEO
Dummy variable takes the value of (1) if the chairman held the CEO position in the company prior to becoming chairman.
Firm size Company size which is measured by the natural logarithm of the total assets.
Firm age Firm age at the beginning of each year. Calculated as the difference between fund year and
the current year (Year- found year).
Return on assets Return on total assets of an individual firm. It is measured as EBITDA over assets of a firm.
Growth ratio This variable represents the opportunities for growth for a firm. It is measured as the ratio of
capital expenditures to assets of a firm.
Cash to assets ratio Cash to Assets ratio is used to measure the capacity of a firm to generate cash.
Debt to assets ratio Debt to assets determines the operations from external funding. It is the ratio of long-term
24
Tangibility ratio Tangibility Ratio of net fixed assets to total assets. The higher the ratio, the less liquid the firm. Tangibility = (Net fixed Assets/ Total Assets) *100
Country law origin Country law origin dummy variable. Takes the value of (1) if the firm located in a
common-law origin country, and (0) otherwise. Corporate governance score
Corporate governance score, the corporate governance pillar measures a company's systems and processes, which ensure that its board members and executives act in the best interests of its long-term shareholders.
Board gender diversity *corporate governance score
This is the interaction term between board gender diversity and the proxy for corporate governance firm level.
Board gender diversity * country law origin
This is the interaction term between the board gender diversity and the dummy for country law origin.
Board gender diversity *board independence
This is the interaction term between board gender diversity and the proxy for board independence.
Board gender diversity *board
size This is the interaction term between board gender diversity and the proxy for board size.
Board gender diversity *
chairman ex CEO This is the interaction term between board gender diversity and the variable chairman ex CEO.
Quota Dummy variable takes the value of (1), if the country has passed a Gender quota in the studied
year, and (0) otherwise.
Board gender diversity*Quota The interaction term between board gender diversity and the dummy for country quota. Board gender diversity dummy Dummy variable takes the value of (1) if at least 1 female is on board, and (0) otherwise.
ɛ Error term.
3.2 Sample and data collection
25
I include publicly-listed firms since they are more likely to disclose information. Firstly, I collected the list of firms from ASSET4. Therefore, information on gender diversity, board size, board independence, company size, and company ROA have been downloaded from DataStream. Table 4 shows the descriptive statistics for the variables among study sample. In this table, I have disclosed the mean, median, maximum, minimum, standard deviation, and a number of observations. The initial summary statistics referred to unusual values for ROA and TQ variables. To correct this issue, I winsorized the data at 5% level.
My dependent variable Tobin’s Q has an average of 1.518 and a standard deviation of 1.103. The maximum value is 4.584 and minimum 0.350. The minimum Board gender diversity value (0%) recorded in Hong Kong, China. Whereas the maximum values recorded in Norway (66.67%) and The United States (75%). The average board consists of approximately 10 managers, of whom 10% are female, and 56.5% are independent directors. The largest board in the sample consists of 33 managers, and the smallest one consists of 1 manager. In average, the Chairman was a previous CEO for less than 0.3% of the total sample with a maximum of 1%.
26 Table 4 Summary statistics of variables
Mean Median Maximum Minimum Std. Dev. Observations
Tobin’s Q 1.518 0.714 4.584 0.350 1.103 24467
Board gender diversity (%) 0.107 0.1000 0.750 0.000 0.109 24467
Board size 10.001 10.000 33.000 1.000 3.327 24467
Chairman ex CEO 0.283 0.000 1.000 0.000 0.450 24467
Board independence (%) 0.565 0.600 1.00 0.000 0.279 24467
Firm age 50.755 37.000 312.000 0.000 41.489 24467
Firm size 14.615 15.259 20.497 7.025 1.460 24467
Cash to assets ratio 0.156 0.123 2.461 0.000 0.166 24467
Debt to assets ratio 0.238 0.229 0.866 0.000 0.188 24467
Tangibility ratio 0.315 0.257 0.793 0.123 0.240 24467
Return on assets ratio 0.124 0.123 0.492 -0.079 0.094 24467
Growth ratio 0.084 0.047 1.154 -0.009 0.061 24467
Corporate governance score (%) 0.532 0.632 0.987 0.112 0.291 24467
Tobin’s Q, ROA values are winsorized at 5% level.
Most of the firms in my sample originate from developed countries. Consequently, corporate governance score was quite high on the level of firms. The average corporate governance score is approximately 54% with the highest value of 98.78 % in Canada and the minimum value of 1.12 % in Japan.
27 Table 5 Correlation matrix
Notes. N= 24467. All correlations 0.10 or greater are significant at p <00 05, 0.12 or greater are significant at p <00 01, and 0.16 or greater are significant at p <00 001 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Tobin’s Q 1.000 Size 1.000 -0.167 Return on assets 1.000 0.027 -0.027 Tangibility 1.000 -0.007 0.020 -0.017 Board independence 1.000 -0.001 0.036 -0.003 -0.014
Debt to assets ratio 1.000 0.006 0.028 -0.096 0.168 0.056
Cash to sales ratio 1.000 -0.018 0.001 -0.001 -0.031 -0.052 0.032
Country law origin 1.000 0.017 -0.018 0.522 -0.007 0.019 -0.219 0.059
Chairman ex CEO 1.000 0.091 -0.013 0.007 0.129 0.001 0.020 0.108 -0.024
Corporate governance score 1.000 0.093 0.567 -0.005 0.008 0.631 0.000 0.033 0.074 -0.078
Growth 1.000 0.024 -0.039 0.078 0.013 0.029 0.030 -0.004 0.051 -0.097 0.153
28
3
.3 MethodologyEstablishing causality is one of the serious issues in studying the relationship between board gender diversity and firm performance. Specifically, finding a correlation between the two does not imply causality. This means that the causality link can be in both directions. Either board gender diversity implies a higher firm performance, or high-performing organizations tend to have more diversified boards. This complicated relationship can imply endogeneity of both variables, firm performance and board gender diversity. To be able to interpret results correctly, I have one-year lag in the regression model (t-1). In addition, I will apply industry and year fixed effects to the regression model.
The first moderator is corporate governance score in organizations. Based on past literature, the level of corporate governance will influence the relationship between board gender diversity and firm performance.
To represent my country moderator. I will use a dummy country variable. Common law-based countries will get a score of (1), and firms in civil-law countries with a (0). Since countries subject to the common law are documented to foster the creation of shareholder value, I expect that firms which are based in common-law countries will have a positive effect on board gender diversity and firm performance because they are more stakeholder-oriented.
The following model includes all variables presented in my hypotheses deriving Tobin’s Q model:
TQ it = b0 + b1 Board gender diversity it-1 + b2 Corporate governance it-1 + b3 Board gender diversity it-1* Corporate governance it-1 + b4 Country law origin j + b5 Board gender diversity it-1 * Country law origin j + b6 Firm size it-1 +b7Firm age it-1+ b8 Board size it-1 +b9 Board independence it-1 + b10 Chairman ex CEOit-1 + b11 Return on assetsit-1 + b12 growthit-1 + b13 Cash to assets it-1 + b14 Debt to assets it-1 + b15 Tangibility it-1 + b16 Quota j +b17 Board gender diversityit-1 *Quota j
+ ᶓit
29
Where i are the firms, t index the time, in this case, the years. ᶓ is the error term which is independent and identically distributed.
4. Empirical results
I used Stata 15 to estimate correlation results by using OLS and by applying industry and year fixed effects to the panel data. Endogeneity problem is one of the major issues in firm performance research. Part of this problem could be that OLS models cannot control for unobserved but fixed firm heterogeneity (Conyon & He, 2017). To solve this problem, I used one-year lag in the variables. And this could help me to identify the influence of board gender diversity and take endogeneity into account at the same time. One reason of this endogeneity is that the presence of females on boards could be linked with the industry-fixed effects. In Accordance with this, I applied industry and year fixed effects for all regression models.
Table 6 presents baseline estimations of the relation between board gender diversity and firm performance, the determinants of Tobin’s Q are presented in 6 models.In all OLS models, I find a positive correlation between firm performance and the percentage of women on boards. The results are significant at 1% for all models.
In the first model, I test the role of board gender diversity on Tobin’s Q, regardless of country-level moderating variables. I find a positive significant link between women on boards and firm value. I also find that corporate governance proxies, corporate governance score, board size, board independence and chairman ex CEO have a significant positive effect on firm performance. That is firm performance increases when corporate governance increase.
30
Table 6 Regression results on Tobin’s Q regardless of moderating effects
Model1 Model2 Model3 Model4 Model5 Model6
Constant 2.068*** 1.668*** 1.697*** 2.083*** 2.099*** 1.877*** [0.343] [0.371] [0.369] [0.307] [0.307] [0.335] Board gender diversity 0.439*** 0.479*** 0.499*** 0.370*** 0.398*** 0.382***
[0.054] [0.054] [0.055] [0.052] [0.053] [0.053] Board independence 0.271*** 0.268*** 0.158*** [0.019] [0.019] [0.022] Board size 0.009*** 0.009*** 0.009*** [0.002] [0.002] [0.002] Chairman ex CEO 0.095*** 0.092*** 0.087*** [0.011] [0.011] [0.011] Firm size -0.163*** -0.151*** -0.152*** -0.177*** -0.178*** -0.170*** [0.004] [0.004] [0.004] [0.005] [0.005] [0.005] Firm age -0.070*** -0.056*** -0.056*** -0.059*** -0.059*** -0.053*** [0.006] [0.006] [0.006] [0.007] [0.007] [0.007] Return on assets 4.687*** 4.650*** 4.648*** 4.642*** 4.640*** 4.622*** [0.081] [0.081] [0.081] [0.082] [0.082] [0.082] Growth -0.000*** -0.000*** -0.000*** -0.000*** -0.000*** -0.000*** [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] Cash to assets 1.931*** 1.921*** 1.915*** 1.914*** 1.906*** 1.916*** [0.058] [0.057] [0.057] [0.059] [0.059] [0.059] Debt to assets 0.305*** 0.302*** 0.301*** 0.257*** 0.255*** 0.256*** [0.037] [0.037] [0.037] [0.038] [0.038] [0.038] Tangibility -0.152*** -0.161*** -0.163*** -0.148*** -0.151*** -0.152*** [0.030] [0.030] [0.030] [0.031] [0.031] [0.031] Corporate governance 0.221*** 0.038* 0.043* [0.018] [0.022] [0.022]
Country law origin 0.195*** 0.188*** 0.132***
[0.012] [0.013] [0.013]
Quota -0.063** -0.084*** -0.044*
[0.026] [0.025] [0.026]
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
R-squared 0.444 0.448 0.449 0.447 0.447 0.449 Observations 26189 26189 26189 24467 24467 24467
F-Statistic 0.000 0.000 0.000 0.000 0.000 0.000
31
The empirical results of the moderating role of Corporate governance, country law origin, are illustrated in Table 7. I examined the difference in performance between countries that have a quota on one side, and countries that don’t have a quota or before they pass a female quota legislation. I created a quota dummy variable (Quota) takes the value of (1) for the years that quota has implemented, and the value of (0) in the years when there is no quota, and for the countries that have not passed a quota legislation.
In all models, I find a positive and significant influence on board gender diversity on firm performance. I examined the role of board gender diversity in the presence of corporate governance moderating effects. We see that board gender diversity has a positive effect on firm value. But the moderating effect of corporate governance on the relationship between board gender diversity and firm performance seems to be significantly negative. That is, corporate governance has a supplementary effect on board gender diversity on firm performance. Origin of law has a significant positive effect on firm performance, this impact turns to significant and negative effect when it is combined with board gender diversity. In other words, country law origin has a negative moderating effect on the relationship between board gender diversity and firm performance. That is, firms have better performance if they are located in countries of common law origin. But firm performance decreases when firms are located in common law country and it has more females on boards.
As mentioned above, I examined quota effects on firm performance as a difference in firm value before and after the quota. The results are summarized in models 1, 2, 4 and 5. We notice that quota adoption has a significant negative impact on firm performance. In other words, the countries that adopted quotas have endorsed a decrease of their performance compared to the countries that have not adopted a female quota or compared to the same country before quota adoption.
Table 7 Firm and country level moderating effects.
32 Board independence 0.300*** 0.154*** 0.153*** [0.024] [0.022] [0.022] Board size 0.009*** 0.008*** 0.009*** [0.002] [0.002] [0.002] Chairman ex CEO 0.095*** 0.087*** 0.075*** [0.011] [0.011] [0.016] Firm size -0.164*** -0.151*** -0.177*** -0.170*** -0.170*** [0.004] [0.004] [0.005] [0.005] [0.005] Firm age -0.068*** -0.055*** -0.058*** -0.052*** -0.052*** [0.006] [0.006] [0.007] [0.007] [0.007] Return on assets 4.677*** 4.622*** 4.640*** 4.620*** 4.620*** [0.081] [0.080] [0.082] [0.082] [0.082] Growth -0.000*** -0.000*** -0.000*** -0.000*** -0.000*** [0.000] [0.000] [0.000] [0.000] [0.000] Cash to assets 1.919*** 1.920*** 1.915*** 1.917*** 1.917*** [0.058] [0.057] [0.059] [0.059] [0.059] Debt to assets 0.301*** 0.303*** 0.258*** 0.257*** 0.256*** [0.037] [0.037] [0.038] [0.038] [0.038] Tangibility -0.151*** -0.164*** -0.145*** -0.148*** -0.149*** [0.030] [0.030] [0.031] [0.031] [0.031] Corporate governance 0.268*** [0.022]
Country law origin 0.233*** 0.152*** 0.155***
[0.014] [0.017] [0.016]
Quota -0.129*** -0.073*** -0.051* -0.052**
[0.025] [0.026] [0.026] [0.026]
Board gender diversity *Corporate
governance -0.636***
[0.190]
Board gender diversity *country law
origin -0.325*** -0.192** -0.221**
33 Board gender diversity*board
independence -0.302*
[0.161]
board gender diversity*board size 0.015 [0.013] Board gender diversity*chairman ex
CEO
0.114 [0.101]
Year FE Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes
R-squared 0.445 0.448 0.447 0.449 0.449
Observations 26189 26490 24467 24467 24467
F-Statistic 0.000 0.000 0.000 0.000 0.000
Note: This table report results from regressing Tobin’s Q (TQ) in order to test the three hypotheses. *, **, *** denote P-value statistical significance at 1%, 5%, and 10% respectively.
In table 8, I run the regression on two samples, in model 3, I run the regression for countries that have adopted a quota. Regression for countries that have not adopted quota summarized in model 4 9.
In all models, I find a positive and significant influence of female CEO’s on firm performance. In model 1 we see that country law origin has a positive and significant effect on firm performance. In the same models, I find a negative effect of quota on firm performance, and this is consistent with the results I find in the previous models. the same result applies in common law countries and corporate governance score.
Table 8 regression results with and without quota countries
Model1 Model1 2 Model 3 Model 4
Constant 1.698*** 2.075*** 1.714*** 1.423***
[0.370] [0.352] [0.380] [0.335] Board gender diversity 0.488*** 0.708*** 0.640*** 0.698
[0.056] [0.047] [0.134] [0.709] Firm size -0.153*** -0.161*** -0.154*** -0.074***
[0.004] [0.004] [0.004] [0.020]
Firm age -0.056*** -0.074*** -0.054*** -0.019
34 [0.006] [0.006] [0.006] [0.034] Return on assets 4.648*** 4.703*** 4.522*** 4.598*** [0.081] [0.080] [0.084] [0.358] Growth -0.000** -0.000*** 0.892*** 0.001*** [0.000] [0.000] [0.153] [0.000] Cash to assets 1.915*** 1.897*** 1.952*** 1.204*** [0.057] [0.057] [0.058] [0.264] Debt to assets 0.301*** 0.316*** 0.307*** 0.006 [0.037] [0.037] [0.038] [0.169] Tangibility -0.163*** -0.174*** -0.291*** 0.182 [0.030] [0.030] [0.036] [0.181] Corporate governance 0.043* 0.068** -0.682*** [0.022] [0.028] [0.211]
Country law origin 0.189*** 0.208*** 0.16
[0.013] [0.018] [0.100]
Quota -0.084*
[0.047]
Board gender diversity *Corporate governance
-0.066 -0.433 [0.217] [1.025] Board gender diversity *country
law origin
-0.212* -0.395 [0.111] [0.422]
Board gender diversity*Quota 0.122
[0.196]
Year FE Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes
R-squared 0.449 0.44 0.456 0.455
Observations 26189 26490 24880 1309
F-Statistic 0.000 0.000 0.000 0.000
Note: This table report results from regressing Tobin’s Q (TQ) in order to test the three hypotheses. *, **, *** denote P-value statistical significance at 1%, 5%, and 10% respectively.
5. Robustness check.
35
Additionally, the interaction effects of board gender diversity, corporate governance, and country law origin are still the same and it is consistent with my prior findings. That is, I didn’t observe significant changes in results when I use board gender diversity as a percentage of a female in the board, and when I use board gender diversity as a dummy variable.
Table 9 Regression analysis using Board gender diversity as a dummy variable.
Model 16 Model 17 Model 18 Model 19 Model 20
Constant 2.812*** 2.902*** 2.787*** 2.702*** 2.649*** [0.154] [0.138] [0.151] [0.152] [0.150] Board gender diversity dummy 0.161*** 0.128*** 0.126*** 0.006 0.068***
[0.023] [0.015] [0.023] [0.041] [0.018] Board independence 0.258*** 0.115*** 0.116*** [0.030] [0.024] [0.024] Board size 0.006*** 0.003 0.006*** [0.002] [0.002] [0.002] Chairman ex CEO 0.083*** 0.074*** 0.077*** [0.012] [0.012] [0.020] Firm size -0.160*** -0.172*** -0.168*** -0.161*** -0.161*** [0.004] [0.004] [0.005] [0.005] [0.005] Firm age -0.071*** -0.061*** -0.061*** -0.056*** -0.055*** [0.007] [0.006] [0.007] [0.007] [0.007] Return on assets 4.550*** 4.234*** 4.549*** 4.533*** 4.533*** [0.086] [0.082] [0.088] [0.088] [0.088] Growth -0.000*** -0.001*** -0.000*** -0.000*** -0.000*** [0.000] [0.000] [0.000] [0.000] [0.000] Cash to assets 1.893*** 2.014*** 1.879*** 1.877*** 1.879*** [0.061] [0.057] [0.063] [0.063] [0.063] Debt to assets 0.257*** 0.273*** 0.211*** 0.211*** 0.211*** [0.039] [0.037] [0.040] [0.040] [0.040] Tangibility -0.121*** -0.129*** -0.110*** -0.119*** -0.120*** [0.032] [0.030] [0.033] [0.033] [0.033] Corporate governance 0.232*** [0.029]
Country law origin 0.200*** 0.122*** 0.131***
[0.016] [0.021] [0.020]
Quota -0.109*** -0.041 -0.012 -0.015
[0.028] [0.029] [0.030] [0.030] Board gender diversity *Corporate governance -0.076*
[0.039]
36
[0.039]
Board gender diversity*board size 0.007**
[0.003]
Board gender diversity*chairman ex CEO -0.003
[0.025]
Year FE Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes
R-squared 0.434 0.436 0.435 0.438 0.438
Observations 23975 26624 21941 21941 21941
F-Statistic 0.000 0.000 0.000 0.000 0.000
Note: This table report results from regressing Tobin’s Q (TQ). *, **, *** denote P-value statistical significance at 1%, 5%, and 10% respectively.
Furthermore, I performed another analysis using the same sample but with dropping of the observations of US, UK, and Australia. The United States has 31.47% of the whole sample. The United Kingdom has 8.7%, and 9.21 % of my sample stemmed from Australia. These 3 countries have 49.38% of the total number of observations. Thus, it may have driven the results. In other words, there could be other common variables in the sample stemmed from these countries which have an effect on the results. In addition, each of the three countries is common law origin countries. The results of the interaction effect between board gender diversity and common law origin may change after dropping observations of these countries.
Table 10 reports regression results using independent variables on Tobin’s Q after dropping sample observations of the mentioned countries.
Obviously, we notice that the main coefficients have the same results before and after dropping the mentioned observations. The coefficient of board gender diversity, corporate governance and country law origin, still positively related to firm performance in the rest countries. All results are significant at 1% level. the interaction effects of board gender diversity with corporate governance and country law origin, still significantly negative, but it has less moderating effects on Tobin’s Q.
37
Table 10 Regression analysis excluding observations of US, UK, Australia.
Model 21 Model 22 Model 23 Model 24 Model 25
Constant 1.863*** 1.551*** 2.016*** 1.715*** 1.717*** [0.148] [0.144] [0.151] [0.155] [0.153] Board gender diversity 1.146*** 0.611*** 0.812*** 0.409** 0.413*** [0.152] [0.072] [0.119] [0.200] [0.086] Board independence 0.138*** -0.02 -0.024 [0.031] [0.027] [0.027] Board size 0.005*** 0.004** 0.005*** [0.002] [0.002] [0.002] Chairman ex CEO 0.065*** 0.067*** 0.019 [0.016] [0.016] [0.019] Firm size -0.180*** -0.178*** -0.194*** -0.190*** -0.190*** [0.006] [0.006] [0.007] [0.007] [0.007] Firm age -0.060*** -0.051*** -0.060*** -0.052*** -0.052*** [0.009] [0.009] [0.010] [0.010] [0.010] Return on assets 4.818*** 4.782*** 4.918*** 4.914*** 4.910*** [0.129] [0.129] [0.136] [0.136] [0.136] Growth 0.907*** 0.845*** 0.872*** 0.819*** 0.820*** [0.228] [0.227] [0.250] [0.250] [0.249] Cash to assets 1.375*** 1.403*** 1.341*** 1.363*** 1.362*** [0.086] [0.085] [0.093] [0.093] [0.093] Debt to assets 0.293*** 0.316*** 0.186*** 0.205*** 0.205*** [0.057] [0.055] [0.064] [0.063] [0.063] Tangibility -0.184*** -0.216*** -0.167*** -0.179*** -0.178*** [0.051] [0.051] [0.055] [0.055] [0.055] Corporate governance 0.181*** [0.031]
Country law origin 0.177*** 0.166*** 0.165***
[0.020] [0.022] [0.022]
Quota 0.054* 0.058** 0.082*** 0.087***
[0.029] [0.029] [0.030] [0.030] Board gender diversity *Corporate
governance
-0.000***
[0.000]
Board gender diversity *country law origin -0.007*** -0.006*** -0.006*** [0.001] [0.001] [0.001] Board gender diversity*board
independence
-0.940*** [0.191] Board gender diversity*board size 0.013 [0.015]
Board gender diversity*chairman ex CEO 0.550***
[0.140]
Year FE Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes
38
Observations 13153 13339 11521 11521 11521
F-Statistic 0.000 0.000 0.000 0.000 0.000
Note: This table report results from regressing Tobin’s Q (TQ). *, **, *** denote P-value statistical significance at 1%, 5%, and 10% respectively.
6.
Discussion and Conclusion
This thesis has provided international evidence on the relationship between board gender diversity and firm performance. Showing that gender relation to board diversity is characterized by the potential of tougher monitoring of the CEO, and by higher alignment with shareholders’ interests. Boards are seen as vital to overcoming agency problems, which may arise between managers and shareholders. My results suggest that female directors could add value by bringing new ideas and different perspectives to the board. The results for Tobin’s Q in all models are consistent with the positive relation between board gender diversity and firm performance exposed in past studies. The coefficient of diversity is positive and significant at 1% level. To address the problem of omitted variables, I add industry and year fixed effects in all models. All the outcome results show that board gender diversity is positively associated with firm performance. Thus, the formulated hypotheses H1 is confirmed.
39
In the framework of governance regressions, it is difficult to find valid instruments (Adams & Ferreira, 2009). Because the variables that are most correlated with endogenous variable are some other governance variables that should be included in performance regressions. I performed analysis using other variables such board size, board independence and chairman ex CEO. I find that board independence negatively moderates the relationship between board gender diversity and firm performance. Which is consistent with my previous finding of corporate governance score. I find that diversity has a positive influence on performance in firms that have weak governance. In firms with stronger governance, enforcing gender quotas in the boards may ultimately diminish shareholders value through over monitoring led by greater gender diversity. I couldn’t prove hypothesis H2a, thus, I confirm hypothesis H2b, that corporate governance negatively moderates the positive link between board gender diversity and firm value.
Lastly, I found a positive impact of common law origin countries on Tobin’s Q. These results are significant in all models, except model 4 in table 7. firms that are located in common-law origin countries have better performance than countries located in civil law origin countries. The enforcement of law and the better protection for all stakeholders in common-law countries proved to have significant positive effects on shareholders’ value.
In table 7 models 2, 4, and 5 and table 8 model 3, we notice a negative significant effect of the interaction effect of country law origin and board gender diversity on firm performance. Which means that country law origin negatively moderates the relationship between board gender diversity and firm value when firms are located in common law origin countries. The mixed evidence on the impact of board gender diversity and its moderating effects is consistent with the previous academic literature in this field. That is, contrary to my hypothesis H3a, I can’t confirm that countries with a common-law legal origin will strengthen the positive relationship between board gender diversity and firm performance. Thus, I confirm hypothesis H3b.
7. Managerial implications
40
2013). My conclusion is that when women are on boards, they are better to be able to influence organizations performance. This thesis shows that there may be strategic reasons to appoint women on boards, which enables firms to perform better.
However, in this thesis, I uncovered evidence of the double-edged nature of board gender diversity. Despite the interaction effects I found, the relationship between board gender diversity and firm performance remains significantly positive. My correlations and regression coefficients between the moderating effects and the relationship are small and negative. The negative moderating effect of board gender diversity and corporate governance variables which explained above gives a signal to governments who consider passing gender quotas.
More generally, my results show that women on boards have an essential and value relevant influence on firm’s value. Nevertheless, this evidence doesn’t provide support for quota-based policy initiatives in highly governed firms. There is no evidence that such legislation would improve firm performance on average. The same applies for common-law countries. Where suggestions for legislation enforcing gender quotas for females on boards should not be motivated for the reasons other than the improvements in governance and better performance.
8.
Limitations and avenues for Further research
In this thesis, I have included a sample of 4125 international organizations, which operate in 43 around the world. The basic line of the thesis is to investigate board gender diversity on Tobin’s Q, and covered the period from 2002 until 2016. However, I haven’t taken the date of the appointment of a female board member in consideration. Which may limit the full view of female board member impact on firm performance.
41
Later studies can use other models for the empirical analysis. One plausible method is the 2SLS regressions method. This method tries the regression analysis in both ways. The impact of board gender diversity on firm value, and the impact of firm value on board gender diversity. Another solution to overcome this problem is by performing the empirical analysis using ROA as a firm performance measurement.
