BOARD DIVERSITY AND GENDER COMPOSITION ON CORPORATE SOCIAL RESPONSIBILITY AND FIRM REPUTATION IN INDONESIA

The purpose of this study was to find out the impact of board diversity and gender composition on CSR performance and firm reputation. This study also revealed the role of CSR performance as a mediator in the impact of board diversity, gender composition, and firm reputation. Samples were chosen from companies awarded Indonesia’s Best Wealth Creator Award for year 2010–2012 by SWA Magazine, Indonesia, using purposive sampling. This study showed that CSR performance and gender composition had positive impact on firm reputation but did not significantly impact board diversity.


INTRODUCTION
Firm reputation represents public perception of a firm or company from time to time. There are various company performances which contribute to firm reputation, among others the result of the accounting calculation of profit and risk factors, market share, media visibility, stock ownership, dividend allowances, firm size, and social concern (Fombrun & Shanley, 1990). A company may gain many benefits from the positive reputation it posseses, such as increasing company's attractiveness to prospective employees, increasing job satisfaction and lessening the inclination to shift to other companies (Bear et al., 2010); improving company's image, building company's image when promoting and marketing new products (Dowling, 2006), and positively influencing company's financial performance, institutional investment, and share price (Bear et al., 2010).
Besides through firm reputation, a company can also build a positive reputation by fulfilling its res-ponsibility in the preservation of environment and social justice (Marshall, 2007). Actions which show corporate social responsibility (CSR) can strengthen a company's reputation. Branco & Rondrigues (2006) stated that CSR enabled a company to raise its reputation in the perception of extensive stakeholders which included customers, suppliers, competitors, banks and investors. If a CSR program was well communicated to the public, the program could help raise the company's reputation and credibility (Carter et al., 2002). Bear et al. (2010) carried out a research to find out the impact of directors' resources diversity and the number of female directors in the management board on CSR, which in turn would have an impact on firm reputation. This research is quite appropriate to be performed in Indonesia, where female employees are raising exponentially in public service, and in agricultural and industrial sectors since the dawn of the industrialization era in 1970 (The Jakarta Globe, 2011).
Besides, after the monetary crisis in 1997-1998, the movement for gender equality in various aspects of life including firm management, has raised significantly (ILO, 2007). Firm management board which in former times was exclusively occupied by male employees are now open for female employees. A number of companies determined to leave that atmosphere by allowing female employees to occupy the position of directors. The aim is to give female employees the same opportunity as male employees, by giving them the chance to develop the best work performance, in order to empower themselves and other female employees in the company (The Jakarta Globe, 2011).
Female co-leaderhip in companies also provides greater chance to support and influence the community than exclusively male leadership (Hillman et al., 2003). Having more female directors also makes a company more sensible to its corporate social responsibility, and gives a company more perspectives when formulating its CSR programs (Wang & Coffey, 1992;William, 2003). In general, a management board which includes gender diversity has become an important issue in firm management today (Singh et al., 2008). This is caused by the market drivers which require that investors who are sensible to social responsibility put significant value on firm corporate social responsibility and on gender equality (Grosser & Moon, 2005).
According to ISO 26000, gender equality means similar treatment to man and woman. Division of role according to gender sometimes discriminate not only woman, but also man. ISO 26000 stated that in all cases, gender discrimination limited the potentials of individuals, families, social groups, and the community. There is a positive relationship between gender equality and social and economy development. Gender equality constitutes one of the Millenium Development Goals (MDGs). The purpose of this study was to find out the impact of CSR performance on firm reputation, and also the impact of board diversity and gender composition (in this case the number of female directors), and to find out the role of CSR performance as a mediator in board diversity and firm reputation. Bear et al. (2010) proposed that board diversity and the number of female directors in the management board had a positive impact on CSR performance. Since CSR performance had a positive impact on firm reputation, in his study Bear et al. (2010) surveyed CSR performance as a mediator in the impact of board diversity and firm reputation. Bear et al. (2010) also surveyed CSR performance as a mediator in the impact of female director composition in the management board and firm reputation.

THEORETICAL BASIS AND HYPOTHESES
Further, it was proposed that board diversity and gender composition in the management board had an impact on firm reputation and CSR performance (Galbreath, 2011). A good CSR performance may have a positive impact on firm reputation. Therefore CSR performance is expected to become a mediator in board diversity and firm reputation. The frame of thought of this study can be seen in Figure 1.  The hypotheses observed by this study were adopted from the previous study of Bear et al. (2010). Donaldson & Dunfee (1994) confirmed that every company was required to show acts of responsibility to all stakeholders, not only from inside the company but also from outside the company, which also included responsibility to environment. This was an important factor which sustained the company's continuation. This proposition was supported by the study of Marshall (2007) which found that a company could establish a positive reputation by showing responsibility in the preservation of environment and social justice. In this case, corporate social responsibility could promote firm reputation. Thus, the raise of firm quality and reputation was intrinsically intertwined with environment quality and social responsibility (Galbreath, 2011). This statement became the basis of formulating hypothesis 1 (H 1 ), namely CSR performance had a positive impact on firm reputation. Jehn (1995) was of the opinion that board diversity included ethnicity, nationality, age, and gender, collective experience and individual skill of each director (Bear et al., 2010). Wang & Coffey (1992) proposed that board diversity could be categorized according to stock ownership, namely internal stock ownership and external stock ownership. Internal stock ownership included the directors from inside the company who owned the company's stocks, while external stock ownership included directors from outside the company which might or might not own the company's stock. Hafsi & Turgut (2012) divided board diversity into two large groups, namely structural diversity and demographic diversity. Structural diversity included the size of the management board (the number of directors in a company), director independence, director stock ownership, and board leadership duality. Demographic diversity included director gender, director ethnicity, director age, and director experience. The management board needs a great variety of experienced and competent human resources to evaluate the management, and to assess business strategies and their impacts on CSR (Hillman & Dalzeil, 2003). This proposition became the basis of the formulation of hypothesis 2 (H 2 ), namely board diversity had a positive impact on CSR performance.
The third hypothesis (H 3 ) stated that gender composition among the directors had an impact on CSR performace. This hypothesis was based on the study of Bear et al. (2010), which concluded that gender composition had an impact on CSR performance, and on the study of Hafsi and Turgut (2012) which concluded that the presence of gender in the management board had a significant impact on a company's social performance. According to Betti Alisjahbana in The Jakarta Globe (2011), the presence of female directors was quite necessary in businesses of service.
Besides, a company needed female leadership on account of the patience, accurateness, egalitarianism, cooperativeness, and compassion which were predominantly female characteristics. Supporting the opinion of Betti Alisjahbana, Kassandra Putranto in The Jakarta Globe (2011) stated that invaluable female contributions in business did not only come from females who were in leadership positions or public figures, but also from behind the screen females who could be motivators and inspirators in innovation and performance.
In his study, Bear et al. (2010) concluded that board diversity and gender composition among directors had positive impacts on CSR performance. It had also been stated that CSR performance had an impact on firm reputation. Therefore, CSR performance was expected to become a mediator in the impact of board diversity and firm reputation. This was proposed in hypothesis 4a (H 4a ), namely CSR performance was a mediator in the impact of board diversity and firm reputation.
Directors are the chief group in a company which determines the policy of the company, therefore determining the proper composition of directors is critical and important (Galbreath, 2011). The composition of directors which respects board diversity also influences the directors' critical function in monitoring. In order to attain effective monitoring, directors need "skill, experience, expertise, and knowledge" (Hillman & Dalzeil, 2003). Board diversity can give much help by providing extensive knowledge, management skill, and other supporting skills such as competency in matters of law, banks and insurances, and experiences of working with the community (Hillman et al., 2000).
Further, CSR performance was expected to become a mediator in the impact of directors composition and firm reputation, which was stated in hypothesis 4b (H 4b ), namely CSR performance was a mediator in the impact of the number of female directors and firm reputation. The presence of female directors in the management board is expected to have a positive impact on CSR (Hillman et al., 2000); to raise charity giving (Wang & Coffey, 1992); to enhance work atmosphere (Bernardi et al., 2002) and to raise concern towards environment (Fernandez et al., 2012). William (2003) stated that there was a relationship between the number of female directors and the charitable giving of the company. The larger the number of female directors, the larger the allocation for charitable giving. The presence and number of female directors in a company can be a signal for stakeholders that the company has concern for women and minority groups in community, and for social responsibility (Bear et al., 2010).

RESEARCH METHOD
In this study, the population was taken from 100 companies which obtained the Indonesia's Best Wealth Creator (IBWC) Award from SWA Magazine in year 2010, 2011, and 2012, so the total of the population in three years amounted to 300 companies. The sample acquirement used the non-probability design with the judgment or purposive sampling technique. The criterion of sample acquirement is as follows: 1. Obtained the IBWC Award for three consecutive years since 2010 to 2012. 2. Had yearly financial report which had been audited and recorded from year 2009 to 2011. 3. Had yearly company report which had been audited and recorded from year 2009 to 2011. 4. The yearly company report in point 3 included directors' profiles with directors' photographs, and account of the company's corporate social responsibility.
Through purposive sampling, we obtained respectively 73 companies in year 2010, 2011, and 2012 which met the criterion, so the total samples amounted to 219 companies. The research variables and their measurements were shown in Table 1.
Hypotheses tests in this study used the analysis of parametric model linear regression with multiple linear regression as parameter estimation method, individual parameter significance test (t statistical test), simultaneous significance test (F statistical test), and coefficient determination. Regression analysis was used as a statistical tool which explained the relationship pattern between two or more variables.
The data processing method used in this study was the Statistical Package for Social Sciences (SP-SS), while the phases of multiplied linear regression analysis followed the phases of Bear et al. (2010), namely: 1. Linear regression to find out the impact between independent variable and intervening variable. 2. Linear regression to find out the impact between independent variable and dependent variable. 3. Linear regression to find out the impact of intervening variable on dependent variable. 4. Linear regression to find out the impact of the presence or the absence of intervening variable on the relationship between independent variable and dependent variable. The linear regression phases were made to answer the hypotheses of this study, which was an adoption of the hypotheses of Bear et al. t statistical test showed how far the impact of individual independent variable was in explaining dependent variable. The test was carried out using the significance level of 0.05 (α = 5%). The validity of the hypothesis was determined using the following significance value (probability): 1. If the significance value was > 0.05, the hypothesis was valid (regression coefficient was not significant). This meant that partially, independent variable did not have a significant impact on dependent variable. 2. If the significance value was ≤ 0.05, the hypothesis was not valid (regression coefficient was significant). This meant that partially, independent variable had a significant impact on dependent variable.

RESEARCH RESULTS AND DISCUSSIONS
This study was carried out in companies awarded the Indonesia's Best Wealth Creator Award by SWA Magazine Indonesia in year 2010, 2011, and 2012. The purposive sampling technique generated 73 companies respectively in year 2010, 2011, and 2012 which met the criterion. If the companies were categorized according to the industrial classification of Indonesian Stock Exchange, the larger part of the research sample came from the sector of finance (26.03%), followed by the sector of trade, service, and investment (16.44%), the sector of consumer goods (12.33%), the sector of property, real estate, and building construction (10.96%), the sector of basic industry and chemicals (10.96%), the sector of infrastructure, utilities, and transportation (9.59%), the mining sector (8.22%), the agricultural sector (5.48%), and the sector of miscellaneous industry (4.11%).
In Table 2 the regression test result showed that CSR performance had a positive impact on firm reputation. This test result was consistent with hypothesis 1 (H 1 ) and confirmed the research conclusion of Bear et al. (2010). The positive value obtained in t test showed that if CSRPerfm increased, firm reputation also increased. This proposition was supported by the study of Marshall (2007) which concluded that a company could establish a positive reputation by showing responsibility in the preservation of environment and social justice. This also agreed with the proposition of Galbreath (2011) which stated that firm quality and reputation was intrinsically intertwined with environment quality and social responsibility. Further, Yang et al. (2010), as quoted by Fuzi et al. (2012), proposed that CSR performance in social performance dimension would raise return on assests (ROA), which affirmed the hypothesis that CSR performance had a positive impact on firm reputation.
The next discussion was about Table 3 which showed that directors (board) diversity did not have impact on CSR performance in the dimensions of ROA, ROE, and % TSR, thus the result was not consistent with hypothesis 2 (H 2 ). This finding agreed with the finding of Bear et al. (2010), which did not support hypothesis 2 (H 2 ), either. Bear et al. (2010) found that the raise of directors diversity did not have a direct impact on the raise of CSR performance. Referring to the study of Hafsi & Turgut (2012) which proposed that there were four dimensions of directors demographic diversity, this H 2 result was possible because the measurement dimension for directors diversity was selected only according to directors' professionalism backgrounds, beside race, age, and gender. The professionalism background dimension surveyed did not include much, either, only three dimensions, namely profession, social community, and business skill. As a result, there was not much variation of data.
Then, Table 4 showed that gender composition had a significant impact, but a negative relationship. This was inconsistent with hypothesis 3 (H 3 ) and disagreed with the findings of the study of Bear et al. (2010), which supported hypothesis 3 (H 3 ), namely that gender composition significantly had a positive impact. Bear et al. (2010) concluded that gender composition (in this case the number of female) in the management board had a positive impact on CSR performance. The conclusion of Bear et al.(2010) was determined from data obtained in the USA, a developed country which was stereotyped as having high gender composition, while in countries stereotyped as having low gender composition, there was no impact of gender composition on firm social per-formance (Boulouta, 2012).
Although female participation in business and industry has increased exponentially after the monetary crisis several years ago, South East Asian countries including Indonesia are still categorized as countries with low gender composition. If we consider the result of GMI 2012 Woman on Board Diversity, this can also happen because gender composition, in this case the number of female directors in Indonesian companies, is still about 4.6% and on the average one company only has one female director.
Gender composition in the management board which had a significant impact but a negative relation with CSR performance could happen because in companies with high gender composition, CSR programs were not reported. The companies did not provide integrated report of their CSR programs (Fernandez et al., 2012). This would have an effect when CSR performance was exposed using content analy-sis.
Observing the R 2 value in Table 4, we found that the value of R 2 EnviQlty was higher than the R 2 value of CSRPerfm and SRQlty. This meant that CSR performance in enhancing environment quality was higher than CSR performance in social responsibility. This condition was confirmed by the statement of Coffey &Wang (1998) which said that companies with high percentage of female directors had low donation levels. Boulouta (2012) proposed that this condition might happen because female directors were more interested in CSR programs which focused on improving the bad image of the company because of the environmental impact of the company's business (the controversy which arose in the community because of the environment contamination caused by the company's business, the severance of the commu-nity's access to natural resources and water, the safety of product), rather than on establishing the company's positive image through social activities (charity giving). Table 5 model 1 showed that the regression test result was consistent with hypothesis 4a (H 4a ), namely CSRPerfm acted as a mediator in the impact of diversity on firm reputation variable. This finding disagreed with the finding of the study by Bear et al. (2010), which did not in the least support CSRPerfm as a mediator in the impact of diversity on firm reputation.
The different test result of hypothesis 4a (H 4a ) might happen because the study of Bear et al.
(2010) focused on samples taken only from companies in health care industry, which generated a high value in the calculation of Blau index with low deviation standard. This meant that the data variation was not much. While in this study, the samples were taken not only from companies in health care industry but also from other types of companies. This caused a higher deviation standard in the calculation of observed data, which indicated that there was more data variation. From Table 5 model 2, we found that the result was consistent with hypothesis 4b (H 4b ) and agreed with the finding of the study by Bear et al. (2010), which confirmed the role of CSRPerfm as a mediator in the impact of Nfemale on firm reputation variable. Similar also the conclusion of the study by Hafsi & Turgut (2012) which stated that variables of demographic diversity which had a significant impact on social performance were director gender and director age.
Directors composition which respects directors diversity also influences directors critical function in monitoring. To achieve effective monitoring, directors need "skill, experience, expertise, and knowledge" (Hillman & Dalzeil, 2003). Directors diversity can help to equip these needs by providing extensive knowledge, management skill, other supporting skills which include skills in matters of law, bank, and insurances, and experience of working together with the community (Hillman et al., 2000).
The more extensive the directors diversity, the larger the potency for solving problems, and the more extensive the networking that can be established by the company, which makes it easier for the directors in directing their business environment and in promoting CSR performance (Hillman et al., 2000). The larger the diversity of directors' resources, the larger the potency for understanding and solving problems which can motivate directors to actively and effectively overcome the conditions of their business environment, which in turn will promote positive reputation for CSR (Bear et al., 2010).
Referring to the proposition of Marshall (2007) in his study which took the director of "The body Shop," Anita Roddick, as a sample, the female leadership in "The Body Shop" supported the company's CSR performance in empowering women through community based commerce. The company's CSR policies gave priority to gender equality and directly created their own market and attracted costumers. This raised the company's profit which was returned for promoting CSR programs.

CONCLUSIONS AND SUGGESTIONS
The test result of the hypotheses formulated in this study showed that CSR had a positive impact on firm reputation, CSR performance could become a mediator in the impact of directors diversity on firm reputation, and CSR performance also became a mediator in the impact of gender composition on firm reputation. The findings of this study did not statistically support the hypothesis which stated that directors (board) diversity had a positive impact on CSR performance.
The findings of this study have important implications on directors and investors. For directors, the positive impact of female director composition can promote processes which occur inside the management board such as analyses and decision makings in director level. Beside that, as a reference, CSR can become an intangible asset which builds firm reputation and helps a company to expand its business globally.
For investors, the findings of this study can be used as a reference for investing in a company. Investors who are concerned with a company's social responsibility can make these findings additional references, which show that the raise of gender composition in the management board will raise CSR performance, and indicate the effort to raise the company's performance and reputation. The presence and number of female directors give signal to stakeholders that a company gives proper attention to women and minority groups in the community, and to social respon-sibility (Bear et al., 2010).
In the description of the test results, the determination coefficient value (R 2 ) was R 2 < 0.5 (50%). This meant that directors (board) diversity variable from professionalism background and gender composition, in this case the proportion of female directors in the management board, had less than 50% impact on CSR performance and firm reputation, while the other percentage of the impact was filled by other variables which were not surveyed in this study. Therefore we would like to suggest that another study be carried out which surveys the impact of other variables of directors diversity and gender composition on CSR performance and firm reputation.
Other variables of directors diversity according to Hafsi and Turgut (2012) are structural diversity such as the size of the management board (the number of directors in a company), director independence, director stock ownership, and board leadership duality; and demographic diversity which includes director ethnicity and director age.