Acadlore takes over the publication of JAFAS from 2023 Vol. 9, No. 4. The preceding volumes were published under a CC BY license by the previous owner, and displayed here as agreed between Acadlore and the owner.
Board of Directors and Corporate Social Responsibility Reporting of Quoted Companies in Nigeria
Abstract:
Purpose: Despite significance of boards of directors, very few studies have examined their influences. This paper investigates the effect of board on corporate social responsibility reporting of quoted companies in Nigeria. Methodology: Although, the study is empirical, the research design is correlational in nature. Panel data of 329 observations was extracted from 47 firms over 7 years (2013-2019). The data was tested for descriptive analysis (mean, number of observations, standard deviations, minimum and maximum means, normality, multicollinearity, and heteroskedasticity. Findings: Results reveal that board size, diligence and independence failed to show significant effects. However, board gender diversity, firm size and leverage show significant effect. It also shows that social disclosure index is superior to corporate social responsibility disclosure model. Originality/Value: The paper concludes that female directors, firm size and debt are the key elements of corporate social responsibility reporting of quoted firms in Nigeria. Therefore, shareholders should appoint more women on the board. Management should increase firm size by acquiring additional assets and take advantage of cheap debt opportunities in the capital market.
1. Introduction
Corporate social responsibility reporting (CSRR) is both good for the host communities and firms’ bottom lines. The motivation of this paper is CSRR has become of interest both to policy makers and the community leaders in recent times. Hence, there is need to investigate further the role of CSR in business. Corporations have realized that it is good for business to support or help the environment in which they do operate. Nigeria represents an interesting case as the country is largely led by the government sector, with very little contributions coming from the organized private sector. There is need, therefore to understand the contributions of the organized private sector and to investigate whether it is good for business to be engaged in it.
CSR is a desirable goal well cherished by every stakeholders, including shareholders, creditors, employees, host communities, government at all levels and others. Perhaps, it is also true that some shareholders will prefer to receive higher dividends and capital gains than to part with returns on their investments. It is in the self- enlightened interest of the greater majority of stakeholders that corporations give back to the society that has allowed them to flourish. It is also in the interest of generations to come that we leave the world better than we met it.
Corporate social responsibility has received its own share of academic and policy controversies. While Carnegie (1889) in his treatise, the Gospel of Wealth, describes the responsibility of the rich to give back to society, the like of Friedman (1962), in his book, Capitalism and Freedom, discusses the limit of the role of economic capitalism in liberal society. Nevertheless, the world has moved greatly away to the middle of compromise and has accepted some form of responsibilities for modern corporations to bear part of the burden of providing for the society.
It is instructive at this point to identify and understand factors that influence corporate social responsibility disclosure: board size, meeting, gender, independence, firm size and leverage. This study is useful to our understanding of concept, theory and empirics, and corporate governance mechanism and social responsibility. This paper fills in empirical, theoretical, conceptual, methodological and demographical gaps that exist in Nigeria. The study contributes to the body of knowledge that is available, clarify differences in understanding of theories, methods, and demographics. For example, researchers, students, managers, creditors, suppliers, community leaders, government at various tiers, regulatory authorities, and tax authorities may benefit from the paper.
2. Literature Review
This study is predicted on the agency theory, which is a concept used to explain the association between principals (shareholders) and managers (agents) in order to resolve conflict. In the case, shareholders are the principal owners and managers are the agents. Xian et al. (2014) found board size to be positive and significant. Similarly, Martin and Herrero (2018) examined board of directors: Composition and effect on the performance of the firm they found board size to be significant. Also, Guerrero- Villagas et al. (2018) board attributes and CSRR: Meta-analysis and found that board size to significant. Isa and Sabo (2015) found that board size is significant and positive. Naseem et al. (2019) also examined the impact of board characteristics on corporate social responsibility disclosure and found board size to be significant. Riyadh et al. (2019) also examined the effect of corporate social responsibility disclosure and board characteristics on corporate performance and found board size to be significant. Bello and Abdul-Manaf (2017) found it to be positive and significant. Abduh and Algeely (2015) investigated Islamic banks and concluded that board size had negative influence. Bulto (2011) and Jensen (1993) agreed that board size had negative influence. Therefore, in null form, this study predicts that:
H0a: Board size has no significant effect.
H1a: Board size has significant effect.
Board diligence is the number of times a meeting takes place. Naseem et al. (2019) found board meeting to be significant. Harywaman et al. (2020) examined board meeting, loss and corporate social responsibility disclosure in Indonesia and found that meeting is negative but significant. Ahmad et al. (2017) examined board meeting frequency and corporate social responsibility disclosure: Evidence from Malaysia and found that board meeting is not significant. Bello and Abdul-Manaf (2017) examined board governance mechanisms and sustainability disclosure: A moderating role of intellectual capital and found it to be insignificant. Pantamee (2014) found it to be significant and negative. Therefore, in null form, this study opines that:
H0b: Board diligence has no significant effect.
H1b : Board diligence has significant effect.
Isa and Sabo (2015) found that board gender is significant. Naseem et al. (2019) found board gender to be significant. Riyadh et al. (2019) also found board gender to be significant. Anazonwu et al. (2018) found board gender to be significant. Also, Yaroson and Giwa (2016) found it to be significant. Awodiran and Kareem (2019) found it significant. Abubakar (2016) found gender to be positive and significant.
Bello and Abdul-Manaf (2017) found it to be positive and significant. However, in null form, this study suggests that:
H0c: Board gender has no significant effect.
H1c: Board gender has significant effect.
Board independence is the number of independent members on the board. Riyadh et al. (2019) found board independence not significant. Bansal et al. (2018) examined board independence and corporate social responsibility disclosure: The mediating role of the presence of family ownership and found that board independence is significant. Khan et al. (2013) examined impact of corporate governance on corporate social responsibility disclosures: Evidence from an emerging economy and found board independence to be significant and positive. However, Ebimobowei (2017) examined corporate governance and corporate social responsibility of listed companies in Nigeria and found board independence not significant. Anazonwu et al. (2018) also examined corporate board and found independence to be significant. Awodiran and Kareem (2019) found board independence to be negative and significant. Abubakar (2016) found independence of the board to be positive and significant. Bello and Abdul-Manaf (2017) found it to be positive and significant.
H0d: Board independence has no significant effect.
H1d: Board independence has significant effect.
3. Research Methodology
This section deals with methodology of the paper. Two models were adapted and tested based on Ahmed (2016) in his treatise corporate social responsibility disclosure and financial performance of quoted deposit money banks in Nigeria:
Whereas:
sdix = Social disclosure index, measured by (Ahmed, 2016).
csrr = Corporate social responsibility disclosure, measured by its dummy, that is, presence of it 1, 0 otherwise (Ahmed, 2016).
bsiz = Board size (Martin & Herrero, 2018; Xian et al., 2014; Yahaya & Tijjani, 2020).
bdil = Number of meetings (Harywaman et al., 2020; Naseem et al., 2019; Yahaya & Bilyaminu, 2020).
bgdr = Board female members (Isa & Sabo, 2015; Naseem et al., 2019; Riyadh et al., 2019; Yahaya & Bilyaminu, 2020; Yahaya & Awen, 2020).
Bind = Board independence is the number of independent directors divided by total directors (Bansal et al., 2018; Ebimobowei, 2017; Khan et al., 2013).
fsiz = Firm size (Eriki & Osifo, 2015; Sulub, 2014). levg = (Gadzo & Asiamah, 2018; Sulub, 2014).
i = Firm word t = Year script α = Constant
β1-6 = Beta coefficients
ε = Error term.
4. Findings and Discussion
Variable | Observation | Arithmetic Mean | Standard Deviation | Minimum | Maximum |
sdix | 316 | .682 | .196 | 0 | 1 |
csrr | 316 | .487 | .501 | 0 | 1 |
bsiz | 316 | 9.177 | 2.902 | 0 | 19 |
bdil | 316 | 4.111 | 2.101 | 0 | 19 |
bgdr | 316 | 8.352 | 10.063 | 0 | 40 |
bind | 316 | .619 | .1726 | 0 | .95 |
fsiz | 316 | 1.714 | .849 | -.449 | 3.775 |
levg | 316 | .846 | 2.017 | 0 | 17.977 |
As indicated in Table 1, the number of observations is very high 316; board size has the highest arithmetic mean 9.177 with standard deviation of 2.902 and minimum mean and maximum mean of 0 and 19, respectively. This is followed by board gender of 8.352 with standard deviation of 10.063 and minimum mean and maximum mean of 0 and 40, respectively. This followed by board diligence with mean of 4.111 with standard deviation of 2.101 The least is csrr of .487 with standard deviation of .501.
Variable | Obs | QH | QH* | P-value |
sdix | 316 | 0.995 | 0.091 | < 0.0001 |
csrr | 316 | 0.631 | 6.559 | < 0.0001 |
bsiz | 316 | 1.011 | -0.197 | > 0.200 |
bdil | 316 | 0.901 | 1.755 | < 0.0001 |
bgdr | 316 | 0.877 | 2.181 | < 0.0001 |
bind | 316 | 1.011 | -0.188 | > 0.200 |
fsiz | 316 | 1.000 | -0.003 | >0.058 |
levg | 316 | 0.452 | 9.751 | < 0.0001 |
From Table 2, board size, board independence and firm size are insignificant at 5%, meaning that they are normally distributed. The rests such as corporate social responsbility reporting, board diligence, board gender and leverage are not normally distributed.
| sdix | csrr | bsiz | bdil | bgdr | bind | fsiz | levg |
sdiz | 1.000 |
|
|
|
|
|
|
|
csrr | 0.684* .000 | 1.000 |
|
|
|
|
|
|
bsiz | 0.252* .000 | 0.209* 0.000 | 1.000 |
|
|
|
|
|
bdil | 0.258* .000 | 0.235* .000 | 0.150* 0.008 | 1.000 |
|
|
|
|
bgdr | 0.093 0.100 | 0.155* 0.006 | 0.096 0.089 | 0.192* 0.001 | 1.000 |
|
|
|
bind | 0.081 0.147 | 0.092 0.102 | 0.395* .000 | 0.032 0.566 | 0.046 0.416 | 1.000 |
|
|
fsiz | 0.424* .000 | 0.304* .000 | 0.537* .000 | 0.297* .000 | 0.150* 0.008 | 0.089 .114 | 1.000 |
|
levg | -0.478* 0.000 | -0.139* 0.013 | -0.132* 0.019 | -0.269* 0.000 | -0.109 0.053 | -0.098 0.084 | - 0.325* 0.000 | 1.000 |
Table 3 results show absence of multicollinearity among the figures; while results in Table 4 confirms this statement.
Variable | Variance IF | 1/Variance Inflation Factor |
Bsiz | 1.71 | 0.585 |
Fsiz | 1.67 | 0.599 |
Bind | 1.23 | 0.813 |
Levg | 1.18 | 0.849 |
Bdil | 1.16 | 0.860 |
Bgdr | 1.05 | 0.951 |
Mean VIF | 1.33 |
|
From Table 4, there is no multicollinearity among the explanatory variables.
Source | chi2 | df | p |
Heteroskedasticity | 49.88 | 27 | 0.005 |
Skewness | 17.10 | 6 | 0.009 |
Kurtosis | 5.54 | 1 | 0.019 |
Total | 72.53 | 34 | 0.000 |
csrr | |||
Source | chi2 | Df | p |
Heteroskedasticity | 62.21 | 27 | 0.000 |
Skewness | 36.03 | 6 | 0.000 |
Kurtosis | 97.37 | 1 | 0.000 |
Total | 195.61 | 34 | 0.000 |
Table 5 shows that both models have heteroskedasticity problems with p-value less than .05. Therefore, Table 6 is robust because of this problem.
sdix | Odds Ratio | Robust Std. Err. | z | P>z | [95% Conf. | Interval] |
bsiz | .727 | .144 | -1.61 | 0.108 | .493 | 1.072 |
bdil | .832 | .116 | -1.32 | 0.186 | .634 | 1.092 |
bgdr | .930 | .031 | -2.17 | 0.030 | .870 | .993 |
bind | .485 | .759 | -0.46 | 0.644 | .023 | 10.440 |
fsiz | 4.595 | 3.701 | 1.89 | 0.058 | .948 | 22.278 |
levg | .455 | .088 | -4.07 | 0.000 | .312 | .666 |
_cons | 1481.759 | 2739.854 | 3.95 | 0.000 | 39.5238 | 55551.56 |
Wald chi2(6) = 63.98 | ||||||
Prob > chi2 = 0.0000 | ||||||
Log pseudo likelihood = -25.231358 Pseudo R2 = 0.5055 |
From Table 6, it shows that board gender and leverage are significant and have positive effects on csrr. The research fails to find board size, meeting (diligence), independence and fsiz significant.
Variable | Observation | Arithmetic Mean | Standard Deviation | Minimum | Maximum |
sdix | 316 | .682 | .196 | 0 | 1 |
csrr | 316 | .487 | .501 | 0 | 1 |
bsiz | 316 | 9.177 | 2.902 | 0 | 19 |
bdil | 316 | 4.111 | 2.101 | 0 | 19 |
bgdr | 316 | 8.352 | 10.063 | 0 | 40 |
bind | 316 | .619 | .1726 | 0 | .95 |
fsiz | 316 | 1.714 | .849 | -.449 | 3.775 |
levg | 316 | .846 | 2.017 | 0 | 17.977 |
As indicated in Table 7, the number of observations is very high 316; board size has the highest arithmetic mean 9.177 with standard deviation of 2.902 and minimum mean and maximum mean of 0 and 19, respectively. This is followed by board gender of 8.352 with standard deviation of 10.063 and minimum mean and maximum mean of 0 and 40, respectively. This followed by board diligence with mean of 4.111 with standard deviation of 2.101 The least is csrr of .487 with standard deviation of .501.
Variable | Obs | QH | QH* | P-value |
sdix | 316 | 0.995 | 0.091 | < 0.0001 |
csrr | 316 | 0.631 | 6.559 | < 0.0001 |
bsiz | 316 | 1.011 | -0.197 | > 0.200 |
bdil | 316 | 0.901 | 1.755 | < 0.0001 |
bgdr | 316 | 0.877 | 2.181 | < 0.0001 |
bind | 316 | 1.011 | -0.188 | > 0.200 |
fsiz | 316 | 1.000 | -0.003 | >0.058 |
levg | 316 | 0.452 | 9.751 | < 0.0001 |
From Table 8, bsiz, bind and fsiz are insignificant at 5%, meaning that they are normally distributed. The rests are not normally distributed.
| sdix | Csrr | bsiz | bdil | bgdr | bind | fsiz | levg |
sdiz | 1.000 |
|
|
|
|
|
|
|
csrr | 0.684* .000 | 1.000 |
|
|
|
|
|
|
bsiz | 0.252* .000 | 0.209* 0.000 | 1.000 |
|
|
|
|
|
bdil | 0.258* .000 | 0.235* .000 | 0.150* 0.008 | 1.000 |
|
|
|
|
bgdr | 0.093 0.100 | 0.155* 0.006 | 0.096 0.089 | 0.192* 0.001 | 1.000 |
|
|
|
bind | 0.081 0.147 | 0.092 0.102 | 0.395* .000 | 0.032 0.566 | 0.046 0.416 | 1.000 |
|
|
fsiz | 0.424* .000 | 0.304* .000 | 0.537* .000 | 0.297* .000 | 0.150* 0.008 | 0.089 .114 | 1.000 |
|
levg | -0.478* 0.000 | -0.139* 0.013 | -0.132* 0.019 | -0.269* 0.000 | -0.109 0.053 | -0.098 0.084 | - 0.325* 0.000 | 1.000 |
Table 9 shows absence of multicollinearity. Table 10 confirms this.
Variable | Variance IF | 1/Variance Inflation Factor |
Bsiz | 1.71 | 0.585 |
Fsiz | 1.67 | 0.599 |
Bind | 1.23 | 0.813 |
Levg | 1.18 | 0.849 |
Bdil | 1.16 | 0.860 |
Bgdr | 1.05 | 0.951 |
Mean VIF | 1.33 |
|
From Table 4, there is no multicollinearity among the explanatory variables.
sdix | |||
Source | chi2 | df | p |
Heteroskedasticity | 49.88 | 27 | 0.005 |
Skewness | 17.10 | 6 | 0.009 |
Kurtosis | 5.54 | 1 | 0.019 |
Total | 72.53 | 34 | 0.000 |
csrr | |||
Source | chi2 | df | p |
Heteroskedasticity | 62.21 | 27 | 0.000 |
Skewness | 36.03 | 6 | 0.000 |
Kurtosis | 97.37 | 1 | 0.000 |
Total | 195.61 | 34 | 0.000 |
Table 11 shows that both models have heteroskedasticity problems with p-value less than .05. Therefore, Table 12 is robust because of this problem.
sdix | Odds Ratio | Robust Std. Err. | z | P>z | [95% Conf. | Interval] |
bsiz | .727 | .144 | -1.61 | 0.108 | .493 | 1.072 |
bdil | .832 | .116 | -1.32 | 0.186 | .634 | 1.092 |
bgdr | .930 | .031 | -2.17 | 0.030 | .870 | .993 |
bind | .485 | .759 | -0.46 | 0.644 | .023 | 10.440 |
fsiz | 4.595 | 3.701 | 1.89 | 0.058 | .948 | 22.278 |
levg | .455 | .088 | -4.07 | 0.000 | .312 | .666 |
_cons | 1481.759 | 2739.854 | 3.95 | 0.000 | 39.5238 | 55551.56 |
Wald chi2(6) = 63.98 | ||||||
Prob > chi2 = 0.0000 | ||||||
Log pseudo likelihood = -25.231358 Pseudo R2 = 0.5055 |
From Table 12, it shows that board gender and leverage are significant and have positive effects on csrr. The research fails to find board size, meeting (diligence), independence and fsiz significant.
5. Conclusions
We add to the level of knowledge by examining board gender and gearing among the factors affecting csrr in Nigeria. The results show that more female board of directors and enhanced leverage has important contributions to csrr. Two things are clear from the results of the study: the appointment of more female board members and the use of leverage by firms instead of more equity. This study addresses the issue of gender and issue of reducing dilution of control of ownership.
In terms of board size, a comparative analysis of our results and empirical findings suggest that our work is consistent with findings of Xtian et al. (2014), Martin and Herrero (2018), Guerro-Villagas et al. (2018), Isa and Sabo (2015) and Naseem et al. (2019). However, inconsistent with findings from Abduh and Ageely (2015), Bulto (2011) and Jensen (1993). In terms of diligence (meetings), we are consistent with the findings of Naseem et al. (2019), and inconsistent with the work of Harywaman et al. (2020). In terms of gender, we are consistent with the findings of Isa and Sabo (2015), Naseem et al. (2019), Riyadh et al. (2019), Anazonwu et al. (2018), Yaroson and Giwa (2016), Bello and Abdul-Manaf (2017), Awodiran and Kareem (2019) and Abubakar (2016). Our findings on independence are consistent with findings of Abubakar (2016), Bello and Abdul-Manaf (2017), Bansal et al. (2018), Khan et al. (2013) and Anazonwu et al. (2018). However, the findings are inconsistent with work of Ebimobowei (2017) and Awodiran and Kareem (2019).
We appreciate that the paper is limited to Nigeria and may not apply to elsewhere or everywhere, since Nigeria is a developing country. Corporate social responsibility has not taken sufficient roots among corporations in Nigeria, like it had taken firm roots in developed economies of USA, UK and EU, therefore, caution must be observed. Nonetheless, the findings have policy and performance improvement implications. We suggest that managers of corporations should take measures to increase female directors in the board, increase firm size and take advantage of cheap leverage since both are significant.