Integrating Text-Mining and Sustainability Balanced Scorecard Methods to Examine the Relationship between CEO Messages of Homepages and Firm Value: Emphasis on Fashion Companies in South Korea
Round 1
Reviewer 1 Report
I enjoyed reading the paper. The latter addresses an important issue.
Below, I will discuss some issues that authors should overcome to develop further this paper.
Issue 1: The theoretical arguments could be strengthened. In this respect, it is not clear how the paper advances the literature on a conceptual level.
To strengthen their theoretical framework, authors could rely on the stakeholder approach to the resource-based theory (Barney 2018) to build a strong theoretical framework that may explain a positive relationship between firm value and CEO messages in connection with corporate social responsibility (CSR). Based on this new approach, CEOs can have access to critical resources that many stakeholders can provide. Authors can also rely on the conflict resolution hypothesis based on the stakeholder theory (Harjoto and Jo, 2011; Jo and Harjoto, 2012) and the legitimacy theory to develop additional strong arguments in favor of a positive association between firm value and CSR.
In the literature, ROE and ROA a used as proxies of accounting measures of firm financial performance and Tobin’s Q and MB as proxies of market measures of firm financial performance. Hence, it is not clear on a theoretical basis why ROE and ROA are considered as proxies of actual financial performance and Q and MB as proxies of future financial performance. In the literature, it is accounting measures versus market measures of financial performance (not measures of actual versus future financial performance). Theoretically, firm value à t = 0 can be expressed as following:
Stock prices at (t) = f( BV (t) + expectations at (t) about future earnings)
In other words, firm value at time (t=0) is determined by the firm’s book value at time (t=0) and market expectations at time (t=0) about future earnings (t+i). Hence, from a fundamental perspective, it is possible that Q at (t =0) contains information posted on the company homepage even if such information should increase firm future earnings. Hence, Tobin’s Q should contain all the public information (future and present) known about firm perspectives. From a fundamental perspective, there should be no difference between H1 and H2. Authors should address this issue.
Issue 2:
Authors should also highlight more their contributions relative to the existing literature. it is not clear whether the paper is the first analysis that examines the relationship between CEO messages of Homepage and firm value. If not, authors should explain how their study contributes to the existing literature.
Issue 3:
Because CSR engagement can be endogenously determined, for robustness, authors should also conduct an endogeneity correction procedure when social responsibility is used as an explanatory variable. As suggested by Harjoto & Jo (2011), without considering endogenous treatment effects in which better quality firms tend to have strong CSR engagement, the association between CSR and firm value will be overstated or falsely attributed. Furthermore, it may also be possible that firms, engaging in CSR activities, deliver higher returns/financial performance to investors. In this case, an OLS estimation will produce biased parameters because CSR is correlated with the error term. Authors should address the endogeneity concern by using some of the well-known econometric approaches that help address endogeneity (e.g., Instrumental variable approach).
Issue 4: The number of observations in the study is two low (42 observations). We propose to use panel data estimations to increase the number of observations and strengthen the empirical findings. Authors should also include the number of observations (N) in tables 5-14
Author Response
Thank you for your sincere comments on this paper. Thanks to your careful and sharp comments, the quality of the paper could be improved. Thanks once again.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Reviewer report on a manuscript titled ‘Integrating Text-Mining and Sustainability Balanced Scorecard Methods to Examine the Relationship between CEO Messages of Homepage and Firm Value: Emphasis on Fashion companies in South Korea’ that was submitted for publication consideration in SAMPJ.
Overall, I applaud the authors' effort to study a CEO's message on the balanced scorecard. The authors appear to have put a fair amount of time in this paper. However I have some important concerns about the paper's potential to be published, listed below:
The paper's contribution is very unclear at this stage. It is very hard to decipher what the authors do that is new compared to prior work.
Why South Korea? There is insufficient justification for setting this study in South Korea. Are there institutional or other contextual factors favoring this setting for the study?
Why the fashion industry? There is unconvincing argumentation about the selection of this industry. Why not open it up to other industries and expand the sample size?
Why study only a single year (2020)? This constrains the sample size but also the econometric techniques to be used. Perhaps constructing a multi-year panel data set and exploiting intertemporal changes would be more fruitful.
Endogeneity is definitely an issue. This is an association study. we do not know if the CEOs of more profitable firms choose to talk more about the balanced scorecard or the other way around. More broadly, we do not know the factors that lead certain CEOs to self-select into talking more about sustainability balanced scorecard and others not to.
To interpret insignificant results as evidence that it takes a long time for the CEO's message to affect firm value is pure conjecture and seems far-fetched.
The writing needs improvement—both in the use of language and organization.
The references should be in the text. The current use of footnotes that simply list citations with not a single word about what any of these papers makes for a very difficult reading.
There is no direct mention of a single paper in the introduction—what past work did and how it relates to the current paper. Surely this paper does not exist in a vacuum.
The absence of discussion of prior work in the text continues in the hypotheses development section. Please see how past work is discussed and linked to a current study by its authors (e.g., in the ssrn platform).
The hypotheses are poorly developed. There is very little to differentiate between the effect of the CEO's message on current vs future performance.
In describing the data selection process, the financial industries are described to be removed. I thought only fashion firms are included, and thus ALL other industries are removed.
the study uses a very small sample size (n=42). The statistical insignificance is unsurprising. The models are trying to estimate ten or so variable coefficients. There are limited degrees of freedom to this end making for unreliable coefficient estimates.
The measures used to distinguish between future and current performance (Mb assets and mb equity vs roa and roe) are not convincing. Please see past research on the topic.
Variable GRW appears to have extreme values.
Does current performance affect future performance and should thus be a control variable in the model?
In the hypothesis 1 regressions all variables are insignificant except the loss dummy (negative relation). However, this is definitely because the loss dummy is partly the same variable on both sides of equation—it is not surprising that it is negative and clouds the non-results.
In the tests of hypothesis 2 no variable is significant—yet the model F-test is significant --why is that? (this could be due to the small no of observations and possible multicollinearity)
Given concerns about the empirics, the study's conclusions are not convincing.
Overall, the paper requires major improvement in several dimensions.
Author Response
Thank you for your sincere comments on this paper. Thanks to your careful and sharp comments, the quality of the paper could be improved. Thanks once again.
Author Response File: Author Response.pdf
Reviewer 3 Report
When reviewing scientific papers for publication, I usually start with a general overview in terms of a structure, abstract, literature review, methodology, findings of the research, discussion, conclusions, as well as limitations of the study.
This research is highly innovative and has a special perspective, which is worthy of reference and learning by subsequent researchers. And the structure of the paper is clear and consistent with accepted standards.
To improve the quality of the paper I would suggest to:
1.The first sentence is "This research aims to investigate the relationship between CEO messages and corporate value on the website of Fashion companies. ". It just came out of the blue. This sentence should have an antecedent.
2.I think this article does not have a good background introduction and literature discussion.
3.How to determine the adequacy or representative number of samples.
4.Research scope and limitations need to be supplemented
Author Response
Thank you for your sincere comments on this paper. Thanks to your careful and sharp comments, the quality of the paper could be improved. Thanks once again.
Author Response File: Author Response.pdf
Reviewer 4 Report
The article is interesting and well-described.
I don't think there is the need to add so many references to support this comment: " Many previous studies have utilized ROA and ROE as proxy variables for current firm value, and Tobin's Q and market value to book value ratio as proxy variables for future corporate value [35, 36, 37, 38, 39, 40, 41, 42, 43]." Adding to the fact that some of them are self-citations.
References need to be organized.
Author Response
Thanks to your good comments on the paper, the quality of the paper could be improved. All of our researchers thank you very much for your hard work.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Authors have addressed adequately issues 2 and 3. I suggest that they explain the study contribution in the introduction.
On the other hand, the conceptual development is still weak in the new version of the manuscript. Authors weren’t able to offer a strong theoretical framework based on the stakeholder resource-based theory (Barney 2018), stakeholder theory (e.g., conflict resolution hypothesis developed by Harjoto and Jo, 2011; Jo and Harjoto, 2012), and the legitimacy theory (Deegan et al. 2002; Ghozali and Chariri, 2007).
Authors have mentioned that: " Many previous studies have used ROA and ROE as proxy variables for current corporate value, and Tobin's Q and market value to book value ratio as proxy variables for future corporate value (Park et al., 2016; Nazir and Afza, 2018; Na et al., 2019; Benamraoui et al., 2019; Fatal, 2019; and et al. And the reason why this study used two measures for current and future corporate values is to present more robust and objective empirical analysis results (Kautonen et al., 2015; Marin et al., 2018) ". However, it is still not clear on a theoretical basis why ROA and ROE are good proxies for current firm value and Tobin’s Q and market value to book value are good proxies for future firm value. As suggested in the first report, Q could be considered as a proxy for all public information (present and future). In addition, one can argue that ROA and ROE are accounting measures that proxy only for past information (weak form efficiency). Hence, authors should clarify why ROA and ROE are considered as good proxies for current corporate value, and Tobin's Q and market value to book value ratio as good proxy variables for future corporate value.
Author Response
Thanks to your good comments on the paper, the quality of the paper could be improved. All of our researchers thank you very much for your hard work.
Author Response File: Author Response.pdf
Reviewer 2 Report
I reviewed the earlier version of this paper two weeks ago expressing a number of concerns that would have required extensive revisions. The revised version made an attempt at addressing some comments but major concerns about the unconvincing contribution, small sample and method issues remain.
The authors should have taken more care at addressing these comments.
Author Response
Thanks to your good comments on the paper, the quality of the paper could be improved. All of our researchers thank you very much for your hard work.
Author Response File: Author Response.pdf
Reviewer 3 Report
I think the authors answered my question. Good job.
Author Response
Thanks to your good comments on the paper, the quality of the paper could be improved. All of our researchers thank you very much for your hard work.
Author Response File: Author Response.pdf
