GENDER DIVERSITY IN ITALIAN LISTED COMPANIES: FEMALE DIRECTORS’ ROLES AND CORPORATE PERFORMANCE

How to cite this paper: Tenuta, P., & Cambrea, D. R. (2023). Gender diversity in Italian listed companies: Female directors’ roles and corporate performance. Business Performance Review, 1 (2), 8–19. https://doi.org/10.22495/bprv1i2p1


INTRODUCTION
Over the last ten years, the topic of corporate governance, especially the structure of the board of directors, has become more important and has aroused the interest of many academics because they recognized it as an important value for the market (Esposito De Falco, 2014).
Recently, following the regulatory interventions that have introduced gender quotas into the boardroom of listed companies, academic and independent researchers have been focusing their interest on the relationship between gender diversity and performance (Liu et al., 2014;Iacoviello et al., 2015).
Equal opportunities in top management is a subject that has become increasingly important at the global level, causing the issuing of laws and voluntary initiatives to reduce the gender gap in companies.
Considering the low level of participation in voluntary measures, many countries have adopted legislative measures to destroy the -glass ceiling‖ 1 in corporate governance, some with a sanctioning system (Norway, France, Italy, Belgium, and Germany), others without introducing any coercive restrictions (the Netherlands and Spain).
Italian Law 120/2011 (the so-called Golfo-Mosca Law) obligates listed companies to reserve a predetermined percentage of board seats to women.Its validity is ten years, a deadline within which women's representation on the board and in top management should be introduced into the companies' culture.
Numerous studies in the literature examine empirically the effect of women's presence in the boardroom on the performance of the company (Campopiano et al., 2017).However, the results are very contrasting.Also, many studies analysing the Italian context focus their interest on the mere presence of women on the board and refer to a limited period, not allowing the real early effect of the above-mentioned regulatory initiative to emerge.This research work aims to analyse the relationship between gender quotas and the performance of Italian listed companies in the period from 2006 to 2015.The period considered is new in the context of Italian studies because it allows evaluating both the period before the introduction of the law and the first three years of its application.In addition, we investigate not only the percentage of women directors on the board but also the different roles that women directors can fulfil on the board, such as independent and executive.
Whether it is true that with Law 120/2011 (Golfo-Mosca Law) Italy has seen an increase in the number of women on the board of directors; it is also true that the presence of women alone may not be enough to improve business performance and to appreciate their quality.Thus, we want to test whether the different results in the literature can also be accounted for by the personal characteristics of the women.
Our study contributes to the literature on board gender diversity and firm performance in several ways.First, we extend the literature by providing the first empirical evidence on board gender diversity and firm performance from Italy, which aims to investigate the effect in two different periods, before and after the introduction of Law 120/2011 (Golfo-Mosca Law).Second, we disentangle the main and general effect of female directors into the executive effect and the monitoring effect, showing that in the Italian context, the monitoring role outweighs the executive effect.Lastly, we provide empirical evidence that the effect of female directors on firm performance is contingent on different firm size.
The paper is organized as follows.Section 2 describes the main characteristics of the Italian context.Section 3 reviews the main literature on gender diversity, focusing on the relationship between female directors and firm performance, and develops the research hypotheses.Section 4 illustrates the empirical method and the econometric model.Section 5 describes the results of the empirical analysis, and finally, Section 6 presents the conclusions.

THE ITALIAN CONTEXT
Italy is a -civil law‖ country, and it is well known in the literature that -civil‖ countries are characterized by lower investor protection (La Porta et al., 2000).Therefore, over the last few years, the authorities in charge of the supervision of listed companies and the institutions that manage the financial markets have been suggesting improving corporate governance practices, in line with international best practices provisions.
Specifically, the Italian context provides an interesting institutional setting to examine the effect of female directors on firm performance.Indeed, apart from being characterized by weak legal protection of minority investors (Belcredi & Enriques, 2014), inefficient law enforcement (Volpin, 2002) high private benefit (Sancetta et al., 2018) and high ownership concentration (Lepore et al., 2018), the Italian government has approved a law on gender quotas, which forces the Italian listed companies to respect a minimum percentage of the less represented gender (Bianco et al., 2015).Thanks to the new regulation, the corporate board of listed companies must reserve at least one-fifth of the seats in 2012 and one-third from 2015 for women.
As stated previously, in recent years the researchers of corporate governance have focused their attention on the analysis of the structure of the board (Minichilli, 2014;Rubino et al., 2017;Cucari, 2019).Meanwhile, there was the alternation of various versions of the Corporate Governance Code (Codice di Autodisciplina)the code that recommends the best practices of Italian listed firms (Borsa Italiana, 2011).These innovations allow us to compare some aspects of governance already widely investigated, such as the number and heterogeneity of members but most importantly allow us to examine new features of corporate governance, such as the protection of minority shareholders, through the presence of independent directors, and female directors.Table 1 presents the evolution of the main characteristics of the board of directors, which occurred during the last decade in Italy.The descriptive statistics refer to the sample object of the analysis in the current research.Concerning the size of the board, there were small changes from 2006 to 2015.Differently, the presence of chief executive officer (CEO) duality and the participation of independent and busy directors has considerably changed.Specifically, CEO duality has decreased by 4% compared to 2006.Independent and busy directors show an increase of 5% and a reduction of 11% respectively, confirming the fact that the audit bodies aspire to have a governance structure more open to minority shareholders, closely related to independent directors, and with directors less otherwise committed, so they can devote more time to the firms.At the same time, there is a substantial increase in female participation on the board of directors.In 2012, the presence of women on boards of Italian listed companies grew by 6% in 2006 to 11.2%.In 2015, the last year of observation, this percentage rose almost to 27%.The number of female directors is four times larger than in 2006.

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
Gender diversity on the board and in top management is attracting the attention of policymakers in the Italian socio-economic system (Cambrea et al., 2018).Legislative actions increasing the female presence on the boards of directors led researchers to verify whether the gender quotas have caused a positive or negative effect upon the economic and financial conditions of the companies.
As a result, research contributions on the subject are growing vertiginously, giving rise to In light of agency theory, one of the main tasks of the board of directors is to monitor top management.For this purpose, the diversity of the board could be a useful tool to use to minimize potential agency problems (Erhardt et al., 2003).Indeed, it seems that women have an impact like that of independent directors (Adams & Ferreira, 2009).At the same time, they appear to monitor more severely on the activity of top management, to have greater involvement than the directors do in decision-making and to have better conformity with the interests of shareholders (Adams & Ferreira, 2009).Also, female directors are more present at the board of directors meetings (Adams & Ferreira, 2004), contributing not only to solving the problems arising from the absences of the directors but above all to improving the efficiency of the board through participation in the decision process.
According to resource dependence theory (Pfeffer & Salancik, 1978), board members represent resources for all businesses and work to create value for all shareholders, not just for the majority.From this perspective, female directors are considered particularly valuable in improving the management of an enterprise.They have different personal and relational skills than the men, which allow them to contribute to having a heterogeneous and comprehensive board.Also, as Huse and Grethe Solberg (2006) suggest that female directors appear to be more prepared than male counterparts and appear to be very interactive, which makes them less dependent on management.To understand what benefits, the women directors bring to the boardroom Hillman et al. (2002) have highlighted that most of them come from non-business occupations, so they are skilled in a variety of areas, such as marketing, public relations, and law.Moreover, they have more wisdom and diligence than men and have an excellent ability to make alliances with the most influential actors (Huse & Grethe Solberg, 2006), have a better capacity to relate to the external environment and to join their second board faster than the male directors (Hillman et al., 2002).Finally, they invest more in research and development (R&D) by promoting business innovation and send positive business signals to the public regarding the company's ethical behaviour (Terjesen et al., 2016).
The analysis of the literature shows the presence of conflicting results.Generally, the thesis emerges that companies characterized by board diversity are distinguished by an effective and efficient board of directors, which can help to avoid opportunistic behaviours from managers and to increase the company value.
In light of the literature review, we develop the following research hypotheses.

Female directors and firm performance
After the entry into force of Law 120/2011, which aims to increase the number of women in corporate governance and to make them more involved in decision-making processes within companies, many researchers have focused their attention on the analysis of the consequences on company performance.Many empirical researchers are studying the relationship between female directors and performances in Italy (Amore & Garofalo, 2016; Amore et al., 2014;Bronzetti et al., 2010).
Despite the numerous research on the topic, the empirical results do not allow a clear definition of the effects of gender quotas on the value of companies.On the one hand, some studies emphasize the benefits that female directors can bring to enterprises (Carter et  Especially, the latter empirical analysis, employing a sample of Norwegian companies, shows that imposing a quota has forced many firms to appoint female directors who in some cases did not have any experience. However, despite it being possible to identify in the literature studies with conflicting results, the recent meta-analysis by Post and Byron (2015) highlights the existence of a positive relationship between women on the board of directors and business performances.Women are perceived as a precious resource for businesses, where they are a source of unique skills and different points of view within the board, so they contribute to improving the quality of decision-making.
Conflicting results come to light from the literature that does not allow for defining the sign of the relationship.Therefore, it is assumed that the presence of female directors can influence business performance in both ways: H1: The presence of female directors affects firm performance.

Independent female directors and firm performance
The concept of independence finds its origin in the agency theory, whose supporters believe that the task of independent directors is to control the work of top management to protect the interests of shareholders and to avoid possible conflicts that might reduce the value of the companies.Many studies acknowledge that when the board of directors is characterized by gender diversity, there is more independence (Adams & Ferreira, 2009;Terjesen et al., 2016).
The presence of women in the boardroom should increase the profits of a company, both through a reduction in agency problems and by a different ability to observe internal issues (Rubino et al., 2017).However, in some cases, more independence can reduce the performance of companies.As shown by Adams and Ferreira (2009), boards with many women could engage in overmonitoring activities and could ultimately decrease shareholder value in firms with strong governance.
Contrary to these results, Terjesen et al. (2016) suggest that the independence of the board has a positive effect on the value of the enterprise when the board is more diversified in gender.On the contrary, when there are few or no female directors, the results show that the presence of independent directors is damaging to company performance.The same opinion is taken by Bøhren and Staubo (2016), who have seen an increase in the independence of the boardroom because of a mandatory gender quota.
Although we believe that female independent directors may affect firm performance, given all contrasting opinions, the sign of the expected relationship between the presence of independent female directors and firm performance cannot be hypothesized a priori.Therefore, we leave the following generic provision: H2: The presence of independent female directors affects firm performance.

Executive female directors and firm performance
Several studies show that women directors do not cover executive positions on the board (Lückerath-Rovers, 2013; Smith et al., 2006).Ahern and Dittmar (2012) clarify that newly appointed female directors have a greater probability of assuming non-executive positions than male directors.Also, they consider that women have significantly less CEO experience and are more likely to be employed as non-executive managers.
However, Smith et al. (2006) show, despite the low percentage of executive board members in their sample, that women in top management positions tend to have a positive impact on company value.Also, Liu et al. (2014), comparing the impact of female independent and executive directors on business performance, highlight the greater effect of the latter.When the CEO is a woman, in addition to managing companies in another way, they are perceived differently by financial markets (Jalbert et al., 2013).Faccio et al. (2016), studying the relationship between the gender of the CEO and the assumption of business risks, reveal that women might reduce corporate risk-taking after they become CEOs.In addition, firms run by female CEOs have lower leverage, less volatile earnings, and a higher chance of survival than otherwise similar firms run by male CEOs.
In the light of empirical evidence, it is assumed that a greater presence of female executive directors can improve firm performance and, therefore, we formulate: H3: The presence of female executive directors positively affects firm performance.

RESEARCH METHODOLOGY
To test the effect of female directors on firm performance, the following empirical model is applied.
(1) After conducting the Hausman test, we opted for fixed-effects over the random-effects model.All regressions included year-fixed effects, which capture the influence of aggregate (time-series) trends and any variation in the outcome that happens over time and that is not attributed to your other explanatory variables.
The dependent variable used in the study is the ROA (return on assets), which is a proxy of firm performance, and it is the result of the ratio between total operating income and total assets (Amore & Garofalo, 2016).However, in order to increase the robustness of our findings, we also report our main analyses using an alternative accounting measure of performance ROE (return on equity), computed as pretax income to common equity (Ararat et al., 2015); and a market-based proxy of performance Tobin's Q, which is calculated as the sum of the market value of equity, the book value of shortterm debt, and the book value of long-term debt, scaled by total assets (Belkhir et al., 2014;Pinkowitz et al., 2006).
The main independent variable Female variables refers to the three types of proxies described in the hypotheses: Female directors, Female independent directors, and Female executive directors.Female variables are continuous variables and are computed as the percentage of female directors over the total number of the board of directors members and as the percentage of independent and female executive directors, respectively (Liu et al., 2014).
Based on prior studies that examine the relationship between female directors and firm performance, and to check the firm-specific effects, we introduced into our analysis several control variables (Adams & Ferreira, 2009;Ahern & Dittmar, 2012).Firm size is measured as the logarithm of total assets.Cash holdings are the amount of liquidity in the firm.It is calculated as the availability of cash and cash equivalents to total assets.Debt is calculated as the long total debt divided by the total assets of the firm.The Cash flow is derived from the ratio of cash flow from operations to total sales.Growth opportunity takes into account the firm's growth investment opportunities and is measured by the rate of sales growth.Capex is the ratio of capital expenditures to total assets.Cash flow volatility is a proxy for measuring the uncertainty of the cash flows generated from operations.It is the company's mean standard deviation of cash flows over the past ten years divided by total assets.We also considered some board variables, which might affect firm performance.Board size is the number of members who sit on the board of directors.CEO duality is a dummy variable, which takes value one if the CEO also covers the position of chairman of the board, and zero otherwise.Male independent directors are identified by the ratio of male independent directors on the board.To mitigate the effect of outliers, we winsorize observations at the 1st and 99th percentiles.The variables used are described in detail in the Appendix.

Sample
The hypotheses are tested on a sample of industrial firms listed on the Italian stock exchange in Milan and included in Datastream for the period 2006-2015 (10 years).The timeline allows for capturing the impact of the introduction of Law 120/2011 on the corporate governance of listed companies.We exclude banks and other financial institutions because their budgets are influenced by exogenous factors (Rubino et al., 2017).From the initial sample of 1,871 firm-year observations, we excluded 192 firm-year observations with insufficient governance data and 276 firm-year observations with insufficient financial data.In addition, as our dependent variable is a leading variable, which allows us to mitigate potential endogeneity issues but leads to a reduction of the total observations, the final sample consists of 1,285 observations and 190 firms.The data on the presence of women on the board of directors were collected manually by referring to the annual reports on corporate governance of the individual firms, available on their official websites and the website of the Italian Stock Exchange.Table 2 presents the sample distribution across the Borsa Italian Industry Classification.The industries with the largest representation include Industrial goods & services (22.31%), personal & household goods (11.26%), technology (10.62%), utilities (9.34%), and media (8.20%).Concerning explanatory variables, the results show that the percentage of women on the board is about 12% and the median is 11%.Regarding the characteristics of female directors, the results show that 5.2% of the members of the board of directors are independent females, whereas almost 3% of directors are identified as female executive directors.Regarding the board variables, the results show that, on average, the CEO duality condition is present in almost 27% of the cases.The average percentage of male independent directors on the boards is 31.2%.The average number of directors on the board is comprised of nine and ten members.

Figure 1. Percentage of female directors on Italian listed companies
Source: Authors' elaboration on board of directors data included in the dataset.
Figure 1 shows that although the number of women has grown increasingly over the years, not all different types of directors have grown proportionally.Only the percentage of independent women follows the growing trend of the total percentage of female directors.On the contrary, the proportion of executive women has remained constant over the years.We conclude that thanks to the application of Law 120/2011 (Golfo-Mosca Law) the percentage of women on Italian boards has experienced an upward trend in recent history, but not uniformly distributed across the role.
Table 4 shows the correlations among all independent variables.In general, problems of correlations due to multicollinearity are negligible, as obtained from the correlation matrix and variance inflation factor (VIF) test, not shown for reasons of brevity.Female directors Female Independent Female Executive In general, the estimated coefficients of the control variables are in line with the indications from previous studies.Firm performance is positively correlated to cash holdings, bets, cash flows, and growth opportunities.On the contrary, we find a negative relationship between performance and both firm size and cash flow volatility.

Hypothesis testing
Concerning the hypotheses previously formulated, H1 is confirmed because the coefficient of the variable female directors is statistically significant (β = 0.0646, p < 0.05).The empirical results show that the presence of women on the board has a positive effect on business performance.By examining female directors who are independent of shareholder control, the empirical findings show the positive and statistical significance of the female independent directors (β = 0.0631, p < 0.05).Differently, from the results shown by Adams and Ferreira (2009), an increase in the gender quota, which is not related to shares holding in the enterprise, leads to better firm performance.Thus, H2 is supported.Our findings are consistent with the empirical research, which attribute the positive result to the monitoring of female directors on management performance.Overmonitoring of the top management could improve board efficiency and, consequently, financial performance.Finally, regarding H3, which aims to test the impact of female executive directors on firm performance, the results of our regression are not statistically significant (β = -0.0595,p > 0.10).It seems that the presence of women directors in the management of the firm does not improve firm performance.Companies do not benefit from the different capabilities of females and new managerial perspectives, which could allow for enhancing firm strategies and increasing firm performance.
Despite our main empirical finding showing that female directors positively affect firm performance, it is possible that the relationship could be affected by the period under investigation.
Consequently, to estimate the influence of female directors on firm performance before and after the introduction of the Golfo-Mosca Law, we split the entire period based on the year 2012, which is the year the Golfo-Mosca Law entered into force.Column 3 and column 4 of Table 5 show the results related to the effectiveness of female directors in determining firm performance before and after the Law.
The empirical results show that after the entry into force of the Pink Quota Law, the participation of women on the boards is not able to influence the company's performance.Both coefficients of the variables female independent and executive are not statistically significant.Diversely, a higher proportion of independent women in the pre-law period proves to be able to improve firm performance.The difference in results in the two different periods is very important.Provocatively, it could be linked to the fact that imposing the selection of female directors has determined choice inefficiencies within the companies, which have found themselves facing the request without being able to draw on a specialized and ready human resources market.

Robustness test
As a robustness test, we repeat our baseline empirical analysis using two alternative measures of corporate performance: accounting and marketbased measures respectively.The empirical findings are presented in the following Table 6.In columns 1 to 4 of Table 6, we regress our empirical model considering ROE as the dependent variable.ROE is computed as pretax income to common equity (Ararat et al., 2015).The results are unchanged with our main models presented in Table 5.The coefficient of the variable female directors is positive and statistically significant (β = 0.769, p < 0.05).The same results are obtained by looking at the different roles of female directors on the board.In fact, only the variable of female independent directors is statistically significant (β = 0.795, p < 0.10).Differently, the coefficient of female executive directors is not statistically significant (β = -0.567,p > 0.10).Referring to the effect of independent and executive female directors before and after the Golfo-Mosca Law, the empirical results found using ROE as a dependent variable are identical to those obtained using ROA.
The results are unchanged, corroborating the reliability of our findings.
Diversely, in columns 5 to 8 of Table 6, we run the empirical analyses employing a market measure of performance: Tobin's Q.Given the difficulties in estimating replacement costs, we follow common practice and use the market-to-book (M/B) ratio as a proxy for Tobin's Q.It is calculated as the sum of the market value of equity, the book value of short-term debt, and the book value of long-term debt scaled by total assets (Belkhir et al., 2014;Pinkowitz et al., 2006).In this case, the findings differ from our results from Table 5.Using a market-based measure of performance, the results regarding the entire period of analysis are not supported.The variables of female directors and female independent directors are no longer statistically significant.The only identical result, in terms of empirical findings, is that the coefficient of the variable female independent directors, before the introduction of the Golfo-Mosca Law, remains positive and statistically significant (β = 0.714, p < 0.  36).Therefore, given that in Italy, more than in other European countries, the stock market is only of marginal importance and few firms are listed, we believe it is better to consider accounting-based measures performance as proxies to measure firm value.
Although the results show a positive effect of female directors and female independent directors on performance, it is possible that the relationship could be affected by other factors able to change the sign and intensity of the relationship.Therefore, to assess how potential moderators may affect the role of female directors, further analyses are conducted.Specifically, we test the role of the firm size as a moderator able to influence the conditional effect of female directors on firm performance.To verify the differences between small and large firms, the interactions between female variables and a continuous variable are included in the model.
Table 7 above shows the empirical evidence about the relationship between female variables and performance influenced by firm size.
The empirical results show that the impact of female directors on firm performance is contingent on the size of the company.In particular, the findings regarding the entire period of analysis, displayed in columns 1 and 2 of Table 7, show that all the coefficients of the interaction terms between female variables and size are statistically significant and have different signs compared to the sign of the main effect investigated.Specifically, looking at the impact of female directors and female independent directors, their positive effect on firm performance decreases as the size of the firm increases.Conversely, the negative effect of female executive directors on corporate performance is positively moderated by firm size.
Referring to the moderating effect of the size in the relationship between female independent directors and female executive directors on corporate performance before and after the introduction of the Golfo-Mosca Law, as the interaction terms between female independent directors and size are negative and statistically significant, the previous empirical findings are confirmed in the case of female independent directors.
Diversely, the interaction terms between female executive directors and size are not statistically significant in both sub-periods.Consequently, it does not seem that the size of the company can affect the main effect of a female with executive roles on firm performance.

CONCLUSION
This research work contributes significantly to exploring the effects of the mandatory introduction of gender diversity into the board of directors of listed companies in Italy.After testing the mere relationship between female directors and business performances, the aim of this article has been to test whether this relationship may depend on other parameters, such as the role of directors on the board.The empirical analysis, conducted on a sample of 190 listed companies regarding the years 2006-2015, highlighted a positive and significant relationship between female directors and business performance.
Regarding the role of female directors on the board, it is possible to highlight discordant empirical results.Only independent women directors affect company performance positively.Regarding women with independent qualifications, our findings show that gender-diversified boards are more severe in control activity.Consequently, excessive monitoring of firms with strong governance can affect the performance of the board positively, leading to benefits for the firms.
The empirical evidence emerging from the study must be interpreted taking into consideration the limitation deriving from the use of a sample of exclusively Italian companies, for which the results do not allow a direct comparison with those of other studies which are based on contexts different from the Italian one or which carry out analyzes on multiple countries.Future research could evaluate how the evolutions relating to the presence of women on the board, not only in purely quantitative but mainly qualitative terms, will influence the functioning of the board and the corporate value.Considering that the legislation will remain in force after 2022, the natural continuation of the work could be to focus on the analysis of the effects of gender diversity on performance in light of more updated and definitive data.In the present paper, this type of evaluation presents difficulties.In fact, the Golfo-Mosca Law introduced compliance with the pink quotas in 2012, but only starting from the first renewal of the board of directors.Therefore, the companies that had renewed the board in 2013 were not required to respect the percentage of 20% of pink quotas in 2012.Consequently, any analyses in the two sub-periods, 2006-2011 pre-law and 2012-2015 post-law, would have given false results.
This work allows interesting implications to be drawn for institutional actors, who have strongly supported gender diversity both in business and top management.Descriptive statistics show that the percentage of women on corporate boards has grown strongly in recent years.However, Law 120/2011 obliges companies to appoint women.At the same time, it was expected that the different categories of directors could grow uniformly, which did not happen.The conclusions allow us to say that it is not enough to impose the presence of women on corporate governance to have good management results.It would be advisable for female directors to be considered for all the different roles that exist in corporate governance, and not only for the role of independent directors but especially it is necessary to consider the specific features of the institutional context of reference and the experiences of female directors to be included in corporate governance.
the one hand, research finds a positive effect of female directors on business performance (Conyon & He, 2017; Liu et al., 2014; Terjesen et al., 2016).On the other hand, some studies show negative results (Adams & Ferreira, 2009; Ahern & Dittmar, 2012) and highlight the disadvantages associated with the presence of female directors in the boardroom.Finally, some empirical research does not reveal a relationship between women and performance (Lückerath-Rovers, 2013; Marinova et al., 2016; Rose, 2007).
Columns 1 and 2 and columns 5 and 6 show the results considering the entire period under investigation: 2006 to 2015.Columns 3 and 4 and columns 7 and 8 report the empirical results in the two sub-periods: before and after the Golfo-Mosca law.The dependent variables are return on equity (ROE) (columns 1 to 4) and Tobin's Q (columns 5 to 8).Temporal dummies are included in the model.In parentheses is the p-value.(*), (**) and (***) indicate the statistical significance of each coefficient at a level of 10%, 5%, and 1%, respectively.
05).Despite the empirical findings are robust to different accounting measures of performance, the econometric results are partially confirmed by employing a market-based proxy (Tobin's Q).As Tang (2017) underlines: -It should be noted that as different measures of firm performance (e.g., accounting vs. market-based measures) capture different dimensions, it is unrealistic to expect them to lead convergent results (Combs, Crook, & Shook, 2005; Richard, Devinney, Yip, & Johnson, 2009).What measure is most appropriate and thus should be chosen should be based on the conceptual arguments underlying the hypotheses (Combs, Crook, & Shook, 2005; Richard, Devinney, Yip, & Johnson, 2009)‖ (p.

Table 1 .
Evolution of board structure in the period 2006-2015

Table 2 .
Sample composition by Borsa Italian Industry Classification

Table 3
presents the main descriptive statistics of all variables for our 1,285 firm-year observations.

Table 4 .
Correlation matrix Note: (*) indicate the statistical significance of each coefficient to a level of 0.05.

Table 5
shows the results of the analysis that examines the role of female directors in determining firm performance in the total sample of companies.

Table 5 .
Empirical analysis of the relationship between female directors and firm performance The table shows the empirical findings of the relationship between female directors and firm performance.Columns 1 and 2 show the results considering the entire period under investigation: 2006 to 2015.Columns 3 and 4 report the empirical results in the two sub-periods: before and after the Golfo-Mosca Law.The dependent variable is a performance measure (ROA).Temporal dummies are included in the model.In parentheses is the p-value.(*), (**) and (***) indicate the statistical significance of each coefficient at a level of 10%, 5%, and 1%, respectively.

Table 6 .
Relationship between female directors and firm performance employing different measures of performance (ROE and Tobin's Q) The table shows the empirical findings of the relationship between female directors and firm performance.

Table 7 .
Relationship between female directors and firm performance according to different firm sizeThe table shows the empirical findings of the moderating role of firm size on the relationship between female directors and firm performance.Columns 1 and 2 show the results considering the entire period under investigation: 2006 to 2015.Columns 3 and 4 report the empirical results in the two sub-periods: before and after the Golfo-Mosca Law.The dependent variable is a performance measure (ROA).Temporal dummies are included in the model.In parentheses is the p-value.(*), (**) and (***) indicate the statistical significance of each coefficient at a level of 10%, 5%, and 1%, respectively.