Board Independence and Corporate Social Responsibility Disclosure: The Mediating Role of the Presence of Family Ownership
Round 1
Reviewer 1 Report
Review of “Board Independence and Corporate Social Responsibility Disclosure: The Mediating Role of the Presence of Family Ownership”
The authors of this paper ask whether the quality of a firm’s CSR disclosures, as measured by their comparability and utility, are affected by the percentage of independent directors on its board and whether it is a family firm. They find that the percentage of independent directors is negatively related to the quality of the CSR disclosures but that the effect is eliminated if the firm is family-controlled.
I have several comments and concerns.
A general concern is that the authors’ hypotheses are not supported by careful and well-developed arguments.
1. The paper draws heavily on Garcia-Sanchez and Martinez-Ferrero (2018) and Martinez-Ferrero, Rodriguez-Ariza and Garcia-Sanchez (2016), and, as such, these papers should be better integrated into the literature review and development of the hypotheses. For example, the authors’ H1 has already been examined and extended in Garcia-Sanchez and Martinez-Ferrero (2018). Specifically, Garcia-Sanchez and Martinez-Ferrero argue that because of reputational and career concerns, independent directors, who do not have deep knowledge of the internal workings of the firm, discourage CSR reporting unless CSR performance is particularly good. The authors ignore the underlined condition, but it is important that it be considered given the evidence in Garcia-Sanchez and Martinez-Ferrero (2018). I suggest dropping H1 but consider CSR performance, using Garcia-Sanchez and Martinez-Ferrero’s paper as a guide, in studying H2a, the authors’ main hypothesis. (More in point 2 below.)
2. The rationale underlying H2a that links CSR disclosure, independent directors and family firms, should be more closely tied to the arguments and results in Martinez-Ferrero, Rodriguez-Ariza and Garcia-Sanchez’s (2016) study of the use of CSR in family firms. This is particularly important in light of the discussion in the previous point that CSR activity affects how independent board members view CSR disclosure. More specifically, if there is a relation between family firm status and the use of CSR as shown in prior work, would that in turn affect how independent directors in family firms view CSR disclosure?
3. H2b is not well integrated into the paper, but could be if the underlying rationale were further developed. As a starting point, the authors should consider how the arguments and results in Garcia-Sanchez, Rodriguez-Dominguez and Frias-Aceitino (2015) relate to family versus non-family controlled firms. Further, are family firms more common in one of the two legal systems (common law or civil law)?
I also have suggestions for improving the empirical work and discussion.
1. It would be helpful to see a break-down of the sample by country. Is the sample dominated by a few countries? Is there balance between common law and civil law countries?
2. Where did the authors gather the information used to determine family firm status? Providing descriptive statistics separately for the family and non-family firms would be helpful.
3. I suggest providing separate tables for the descriptive statistics and Pearson correlations instead of combining them in Table 2. Doing so would allow the authors to provide means, medians, Q1 and Q3 for each variable.
4. The authors need to address possible endogeneity concerns and provide additional tests so that the reader can determine their significance.
Author Response
S.No | Reviewer’s comment | Response and Change Reference |
1. | The paper draws heavily on Garcia-Sanchez and Martinez-Ferrero (2018) and Martinez-Ferrero, Rodriguez-Ariza and Garcia-Sanchez (2016), and, as such, these papers should be better integrated into the literature review and development of the hypotheses. For example, the authors’ H1 has already been examined and extended in Garcia-Sanchez and Martinez-Ferrero (2018). Specifically, Garcia-Sanchez and Martinez-Ferrero argue that because of reputational and career concerns, independent directors, who do not have deep knowledge of the internal workings of the firm, discourage CSR reporting unless CSR performance is particularly good. The authors ignore the underlined condition, but it is important that it be considered given the evidence in Garcia-Sanchez and Martinez-Ferrero (2018). I suggest dropping H1 but consider CSR performance, using Garcia-Sanchez and Martinez-Ferrero’s paper as a guide, in studying H2a, the authors’ main hypothesis | This paper aims to extend the findings of Garcia-Sanchez and Martinez-Ferrero (2018) which shows that greater social and environment performance is related positively to CSR disclosure and it is more reliable. In this case, the information reduces independent director reputation risk. In this paper we extend the literature by analysing the role of family firm in reducing independent director reputation risk associated with receiving misleading information as family ownership in the firm control the management discretion and bring down the information asymmetry for independent director. Hence in this paper our scope is limited to analyze the moderation effect of family ownership on independent directors and CSR disclosure and we are not considering the role of CSR performance which was documented earlier.
|
2. | The rationale underlying H2a that links CSR disclosure, independent directors and family firms, should be more closely tied to the arguments and results in Martinez-Ferrero, Rodriguez-Ariza and Garcia-Sanchez’s (2016) study of the use of CSR in family firms. This is particularly important in light of the discussion in the previous point that CSR activity affects how independent board members view CSR disclosure. More specifically, if there is a relation between family firm status and the use of CSR as shown in prior work, would that in turn affect how independent directors in family firms view CSR disclosure? | Hypothesis H2a has been updated along with the argument and result of Martinez-Ferrero et al. (2016).
Change: Explanation added in the Hypothesis development section, Line 200-208; 225-227.
|
3. | H2b is not well integrated into the paper, but could be if the underlying rationale were further developed. As a starting point, the authors should consider how the arguments and results in Garcia-Sanchez, Rodriguez-Dominguez and Frias-Aceitino (2015) relate to family versus non-family controlled firms. Further, are family firms more common in one of the two legal systems (common law or civil law)? | Hypothesis H2b has been updated along with the argument and result of Garcia-Sanchez et al. (2015).
Change: Explanation added in the Hypothesis development section, Line 234-239; 251-255.
|
4. | It would be helpful to see a break-down of the sample by country. Is the sample dominated by a few countries? Is there balance between common law and civil law countries? | We have reported the new Table which provides break-down of the sample by country
Change: Table 4, Page 12
Yes we have balanced observation among the common and civil law countries, We have around 479 observation of common law countries and 593 observation of Civil law countries
|
5. | Where did the authors gather the information used to determine family firm status? Providing descriptive statistics separately for the family and non-family firms would be helpful. | We manually compile the data of family ownership by looking in to the ownership structure of the firm. Later this information is combined with the Thomson database information.
Change : Explanation added in the methodology section, Line 334
We have reported the new Table which provides the descriptive statistics separately for the family and non-family firms
Change: Table 5, Page 13
|
6. | I suggest providing separate tables for the descriptive statistics and Pearson correlations instead of combining them in Table 2. Doing so would allow the authors to provide means, medians, Q1 and Q3 for each variable | We have separated the Table of descriptive statistics and Pearson correlations and reported means, medians, Q1 and Q3 for each variable.
Change: Table 2 and Table 3, Page No 10 and 11.
|
7. | The authors need to address possible endogeneity concerns and provide additional tests so that the reader can determine their significance. | Similar to previous literature (Uribe-Bohorquez et al. (2018); Choi et al. (2001)) we use the variable Independent one period t-1 which helps us in avoiding the problem of endogeneity |
Round 2
Reviewer 1 Report
Thank you for your sincere efforts in addressing my concerns.