Corporate Social Responsibility and Reputation: A Study on Top 100 Companies Operating in India

: The corporate social responsibility (CSR) performance of a company enhances a company’s credibility, thereby influencing the perception of its stakeholders in improving the corporate reputation of a company. This study examines this claim by empirically investigating the CSR performance of India’s top 100 companies and its impact on their reputational status. We have employed the panel-corrected standard error model by controlling companies’ financial performance, size, age, and market risk to analyze the impact of CSR performance on reputation. The result of the analysis is contradictory to the common belief that CSR positively impacts reputation building. Although this study is not novel in nature, it is incremental to the current literature as this study is conducted from an Indian perspective (where no study has been conducted as per our knowledge). This study uses data that are more objective and concrete than those of previous studies, making this study another valuable addition to the extant literature as it is an improvement over the previous study.


INTRODUCTION
In the new era of globalization and the increasing demands of the stakeholders, the companies must put extra effort to meet the needs of the hour. Demands are not limited to the main products or services directly offered in the market but have extended to the social behavior of the company. Social behavior is now judged by the stakeholders in the matter of information disclosed by the company in its social reports. Companies now disclose information on CSR performance either due to statutory mandates or to make themselves more visible to their stakeholders (Axjonow et al., 2018). CSR performance in the form of disclosure of information in the social report of companies lead to companies being better identified among its stakeholders, creating greater commitment on their part towards the company which ultimately leads their way to enhanced reputation (Sen et al., 2006;Garg & Gupta, 2021). However, there exist some alternating views where CSR is considered as a wasteful activity of the company resources (Moser & Martin, 2012). CSR is an essential component for reputation building (Johnson et al., 2018). CSR is found to have many benefits like increasing financial performance, employee The significant increase in the importance of the social performance of companies justifies the study and interest in the topic. The choice of Indian companies was made because India is one of the tops among the emerging economies in the world and earlier studies were conducted in developed economies (Axjonow et al., 2018;Johnson et al., 2018;Melo & Garrido-Morgado, 2012;Odriozola & Baraibar-Diez, 2017;Singh & Misra, 2021) giving us the picture that is prevalent there. Also, in recent times CSR have been made mandatory in India which has given birth to the thought to what extent CSR might have contributed towards reputation building. Prior research studies also present a mixed picture of the relationship between CSR and reputation. Therefore, this study although not novel, is incremental to the existing literature in the field from an Indian perspective (where no study has been conducted as per our knowledge).
With this paper, we have tried to address the issues of previous research by capturing CSR performance both from the angles of quality and quantity by conducting a content analysis of the social reports of the companies and scored them against the social parameters set by GRI 4. Reputation is scored from 10 different dimensions namely innovation, use of corporate assets, long-term investment value, financial soundness, quality of management, quality of product and services, ethics and transparency, use of corporate assets, global competitiveness, and people practices, and talent management. The scoring methodology is based on an approach (Baruah & Panda, 2020) that is not only comprehensive to capture the perception of almost all the stakeholders but also relatively more objective in nature. For analyzing the impact of CSR performance on the reputation we have employed Panel Corrected Standard Error (PCSE) model.
Additionally, this study might be of interest and help to practitioners and standard setters as it presents an important empirical contribution to the underlying theories. Also, recommendations for the companies can be obtained from the findings especially as to how and to what extent CSR information may be disclosed so that it aids in the augmentation of reputation. (Brammer & Millington, 2005;Chun, 2005;Fombrun & Shanley, 1990;Kanto et al., 2016;Melo & Garrido-Morgado, 2012;Sánchez & Sotorrío, 2007;Zhang & Schwaiger, 2009); we use reputation which is computed by following the works of (Baruah & Panda, 2020). The score is computed using appropriate proxies for different dimensions as follows (Table 1).

Different authors have used different methods and measures of reputation
The values obtained on these 10 different dimensions are converted into a single composite score as indicated below by creating an equally weighted average which provides us with the measure of corporate reputation of a particular company.
Where CRS = Corporate Reputation Score and V 1 , V 2 , V 3 …V n are the parameters used in the model and n is the number of parameters. CSR performance is captured by "Content Analysis" of the sample companies' Sustainability Reports or Business Responsibility Reports. The extent of CSR performance of companies is measured by constructing a CSR Index based on the Global Reporting Initiative Guidelines 4. CSR index measures the CSR performance of a company. To assess the CSR performance in the CSR index we have employed a four-point summative scale ranging from 0 to 3. We have assigned the code as'0' if the item is not disclosed, '1' if the GRI specified item is partly disclosed in the published report (most of the indicators/items specified in the GRI framework can be classified into sub-items, therefore, disclosing of even 1 sub-item qualifies as partial disclosure), '2' if the item is Indonesian Journal of Sustainability Accounting and Management, 2022, 6(2), 264-277 fully disclosed in descriptive form and '3' for quantitative form (If the disclosed items cannot be expressed in quantitative terms, we have assessed these disclosed items in terms of their preciseness and clarity to assign the code '3' otherwise '2'). CSR index is computed as the ratio of the computed total disclosure score obtained by a company to the maximum possible score (i.e., total number of items included in the index).
Where CSRI = Corporate Social Responsibility Index of company i at t period, e i = computed total disclosure score, E = maximum possible disclosure score. We further carried forward our analysis by checking the individual effect of the four broad parameters set by GRI while judging the impact of social performance on the companies' reputations. The four broad parameters are Labor Practices and Decent Work performance (LPDWP) having 16 indicators, Human Rights performance (HRP) having 12 indicators, Society performance (SP) having 11 indicators, and Product Responsibility performance (PRP) having 9 indicators. For this purpose, the same methodology used to measure CSR is used. We have employed a four-point summative scale ranging from 0 to 3. We have assigned the code as'0' if the item is not disclosed, '1' if the GRI specified item is partly disclosed in the published report (most of the indicators/items specified in the GRI framework can be classified into sub-items, therefore, disclosing of even 1 sub-item qualifies as partial disclosure), '2' if the item is fully disclosed in descriptive form and '3' for quantitative form (If the disclosed items cannot be expressed in quantitative terms, we have assessed these disclosed items in terms of their preciseness and clarity to assign the code '3' otherwise '2'). The Score for each parameter is computed as the ratio of the computed total disclosure score obtained by a company to the maximum possible score (i.e., total number of items included in the index).
Indonesian Journal of Sustainability Accounting and Management, 2022, 6(2), 264-277 The relationship between CSR and CR may be affected by various exogenous variables. Therefore, prior literature has used several control variables. We in this study have controlled four variables: firm size, financial performance, firm age, and market risk. Firm size (FS) is measured as the natural log of market capitalization (Dang et al., 2018). Large firms are more visible than small firms and thus stakeholders tend to remember a large company more easily and readily than the smaller ones (Brammer & Pavelin, 2006;Fombrun & Shanley, 1990). Financial performance (FP) will be captured by the ratio market to book value (Roberts & Dowling, 2002). This ratio helps in capturing the beneficial effects of high performance by companies. Stakeholders may develop a favorable attitude towards financially successful firms than the companies who are financially unsuccessful. A firm's age is the number of years between the date of incorporation of the firm and the date of the study (Zhu et al., 2014). Reputation is an accumulation of stakeholders' perspectives over the years (Baruah & Panda, 2020). The older the firm, the higher the reputation the company may enjoy. On the other hand, this notion may be found contradictory as reputation over the years may be positive, negative, or a combination of both leading to a company enjoying a stagnant reputation. Market risk (MR) is measured by the company's beta coefficient (Fombrun & Shanley, 1990). As stakeholders are risk-averse, firms that offer less risk may show signs of a better reputation. On the contrary, some stakeholders may prefer to invest in high-risk companies as high risk renders high returns. The data for the control variables are collected from the corporate database "Capitaline Plus", "ProwessIQ", and the Indian financial portal "moneycontrol.com".
The top 100 listed companies are selected from BSE (Bombay Stock Exchange, the oldest in India) based on the market capitalization list of 2015. The top 100 companies are selected as they capture around 62% of the whole market share. The study covers a period of 5 years from 2015 to 2020. CSR performance data is collected from the year 2015 onwards and CR data has been collected from 2016 onwards. The sample companies are drawn from BSE 500 based on market capitalization, 2015; this was when the reporting of CSR activities was made mandatory under Section 135 of the Indian Companies Act, 2013 for these top 500 companies. CSR reporting was at its nascent stage where proper reporting and the resources for proper reporting of CSR may not have taken shape in the very first year and further disclosure takes some time to form perception among the readers. We have assumed the long-term effect of CSR performance and the related control variables on the reputation of a company, and accordingly reputation data was collected following a lag of a year.
To address the research questions, we regress corporate reputation (dependent variable) with CSR performance (independent variable) and several control variables, formulating the following regression models: CRi, t = β 0 + β 1 CSRi, t+ β 2 FSi, t+ β 3 FPi, t+ β 4 MRi, t + β 5 FAi, t + ei, t (eqn.1.) CRi, t = β 0 + β 1 PRPi, t+ β 2 FSi, t+ β 3 FPi, t+ β 4 MRi, t + β 5 FAi, t + ei, t (eqn.5.) Before conducting the regression analysis, we ran a series of tests to evaluate the assumptions for regression analysis and its fit. To test the normality of the data Jarque Bera test was conducted and all the results (3.3, 2.9, 6.5, 3.1 and 2.1 in respective order of equations) being greater than .05 indicate that our data is normal. Multicollinearity among variables was assessed by using Variance Inflation Factor (VIF) and the test revealed the mean VIF as 1.33, 1.20, 1.10, 1.11, and 1.13 in respective order of the equations (all being < 10), indicating that multicollinearity is not a concern for the analysis (Johnson et al., 2018). Additionally, to test heteroscedasticity assumption the Breusch-Pagan/Cook-Weisberg test for heteroskedasticity was run. The p-values (0.0096, Indonesian Journal of Sustainability Accounting and Management, 2022, 6(2), 264-277 0.0062, 0.0412, 0.0136, and 0.0026 in respective order of equations) obtained from the test revealed that the variables suffer from heteroscedasticity. Test for autocorrelation (using Wooldridge test of autocorrelation) revealed the existence of autocorrelation. To address these problems, we employed the Panel Corrected Standard Error (PCSE) model. SE estimates of PCSE model are robust to disturbances that are heteroskedastic, contemporaneously cross-sectionally correlated, and autocorrelated of type AR (1). In the econometric model, the reputation of the current year is regressed with the CSR index and control variables of the precedent year. By this lagging, we could evade the potential endogeneity issue (Melo & Garrido-Morgado, 2012) and isolate this study from the theories supporting two-fold proposition (Hillman & Keim, 2001). Table 2 presents the descriptive statistics for all the variables used in this study. The mean value of reputation is recorded at 1.666613 and skewness at .277 demonstrating that the distribution is not far from symmetrical. The same can be observed in the case of firm size. The mean value of CSR is .2534583 signifies that Indian companies on average disclose only 25% of the items specified in the GRI framework. Also, the low average disclosure of LPDWP, HRP, SP, and PRP are 32%, 23%, 21%, and 21% respectively can be seen. The firm performance of Indian companies is much skewed as the observed value is high (3.036). The firm's age ranges from 7-112 years.  Table 3 provides the correlation among variables included in our regression models. A negative correlation is obtained between Corporate Reputation and CSR (r = -0.179) significant at 1% level. A negative significant correlation at 1% level is observed in all the cases of LPDWP, HRP, SP, and PRP and Corporate Reputation (r = -0.175, -0.187, -0.142, and -0.187 respectively). Regarding our control variables, Corporate Reputation and market risk (r= -0.2270) are negatively correlated at a 1% level of significance. However, Corporate Reputation is positively correlated to firm size (r = 0.2350) and firm performance (r = 0.5166) (both at 1% level of significance) but maintains no correlation with firm age (r = 0.0289).   Table 4 provides the result of estimating all the equations using Panel Corrected Standard Error (PCSE) model. For eqn. 1 the estimated coefficient of CSR is statistically significant at a 5% level of significance. Thus, we find that there exists a negative significant relationship between Corporate Reputation and CSR. The result of estimating eqn. 2 revealed coefficient of LPWDP is statistically significant at a 1 % level of significance indicating that a negative significant relationship exists between Corporate Reputation and LPWDP. Estimating eqn. 3 revealed coefficient of HRP is statistically significant at a 1% level of significance. Thus, we can say that there exists a negative significant relationship between Corporate Reputation and HRP. The estimated coefficient of SP is not statistically significant. Thus, we find that no significant relationship exists between Corporate Reputation and SP. The estimated coefficient of PRP is statistically significant at a 1% level of significance inferencing that there is a significant negative relationship between Corporate Reputation and PRP.

RESULTS AND DISCUSSION
Regarding our control variables, we find significance in the relationship of Corporate Reputation with firm size and firm performance (for all the equations). However, market risk and firm age, other explanatory variables, are not statistically significant (for all the equations). The R 2 for all the equations in respective order are 38.53%, 38.93%, 39.36%, 38.26% and 39.13%. Estimates of Rho for all the equations are in the range -1 to 1.
Our findings contrast the prior affirmations and indicate the contradictory, thus not accepting the null hypothesis that there exists a positive relationship between CSR performance and reputation. Additionally, the component parameters of CSR's (as per GRI) LPWDP, HRP, and PR are also found to have a significant negative impact on reputation. Therefore, we draw the inference that CSR performance does not lead to reputation-building or enhancement. These results are consistent with the findings of another recent study (Axjonow et al., 2018;Singh & Misra, 2021). Source: Computed and compiled by the authors using STATA 16 Note: * , ** and *** represent significant levels at 0.10, 0.05 and 0.01 respectively.
This paper investigates the impact of CSR performance on reputation. The result of our analysis shows that CSR performance negatively impacts reputation. The findings of the study support the claims of CSR as a waste of shareholder's money (Friedman, 1970), the wasteful discretionary act of management (Brammer & Millington, 2005), born of altruistic impulse (Matuleviciene & Stravinskiene, 2015) or the desire of selfaggrandizement (Bartkus et al., 2002;Navarro, 1988). Additionally, CSR reports are attempts by the companies to win over its stakeholders (Godfrey, 2005) but the credibility of CSR disclosure is uncertain (Simnett et al., 2009) and information provided in these reports provides a more positive slant (Moser & Martin, 2012). Also, CSR when is used as a communicating instrument, the company may represent them in a self-laudatory manner when reporting its efforts (Hooghiemstra, 2000). Furthermore, the findings also point out that market risk and the age of a firm have no impact on the reputation of a company.
The CSR disclosures made by the companies might need to disclose information in a way that is taken positively by the stakeholders. Stakeholders are likely to pay more attention to the negative information than the positive ones. Although this transparency (be it negative) is said to have a positive impact on reputation building and enhancement (Johnson et al., 2018) but the case may not be always true. A bad impression in the minds of stakeholders has a lasting and devastating impact on the reputation of a company, be it the company itself or a third party disclosing such information.
Indonesian Journal of Sustainability Accounting and Management, 2022, 6(2), [264][265][266][267][268][269][270][271][272][273][274][275][276][277] The CSR Index used to measure CSR performance in this study uses the quality and quantity of disclosure information. Studies have proved that the quality of CSR reports increases the likelihood of a company enjoying a higher corporate reputation (Odriozola & Baraibar-Diez, 2017). As the samples are drawn from BSE 500 based on market capitalization, 2015; this was when the reporting of CSR activities was made mandatory in India under Section 135 of Companies Act, 2013 for these top 500 companies. CSR reporting might be at its nascent stage where proper reporting and the resources for proper reporting of CSR may not have taken shape. This could be a reason that might have rendered to our finding that contradicts the previous literature where CSR is found to have an impact on Reputation (Johnson et al., 2018;Melo & Garrido-Morgado, 2012;Odriozola & Baraibar-Diez, 2017;Zhu et al., 2014) or companies are making these disclosures just for the sake of the mandate put on them under the Act which is evident from studies like (Garg & Gupta, 2020). Another reason may be companies are not able to perform as per those guidelines laid by GRI 4 or Business Responsibility Reporting due to which we find no significant relationship between Corporate Reputation and CSR. This result is considered tenable as the regression analysis evidences it.
Mandating CSR is criticized in India because it lacks proper mechanisms for its enforcement. One of the major problems in effective CSR enforcement is finding credible projects that these companies can support. Bigger charities get flooded with money while the smaller charities struggle to seek funds as they lack the resources and capacity to cope with the company's bureaucratic and operational demands. Another hindrance is a geographical bias where companies tend to fund only those projects that are closer to where they are based. This is resulting in industrialized areas getting preference over the poorer and underdeveloped areas. Various analyses show that the law in its current form is failing to promote healthy CSR initiatives due to its poor enforcement and lack of clear obligations. The legal provisions related to CSR contain vague language which results in a high degree of self-interpretation. Another flaw from which the CSR has to struggle is that the Act doesn't penalize a defaulter and just allows them to walk away with an explanation regarding their failure on CSR activities. This results in high corruption, low levels of public confidence, low development, and weak institutions (Bhardwaj, 2021). Companies in India to avoid paying money for CSR activities, follow poor disclosure standards when it comes to revealing the details of their spending on the CSR initiatives. This was confirmed by a report made by the Institutional Investor Advisory Services. According to the report, 51 companies listed on the Bombay Stock Exchange's Sensex defaulted under this category. The report says that to spend less than what is required on CSR initiatives, the companies are not very forthcoming when it comes to sharing the details of their CSR spending (Rai, 2020).
However, with the amendments that has been made to the Act in 2019 with regards to the meaning of "CSR policy" which is a document that outlines the approach of a company in selecting, implementing, and monitoring CSR activities, making provisions for international organizations, and ongoing multi-year CSR projects. The government has been enabled to set up a 'National Unspent Corporate Social Responsibility Fund' for any unspent CSR budgets. These changes are expected to benefit the current scenario of India to a certain extend.

CONCLUSION
CSR, which is considered beneficial to a company in many ways, is expected to create and enhance reputation. But on the contrary, we find that CSR negatively impacts a firm's reputation. Result of estimating Panel Corrected Standard Error (PCSE) model highlights how the CSR performance of a company impacts reputation in India. It has been observed in our study that CSR performance negatively impacts reputation, making the 274 Baruah & Panda Indonesian Journal of Sustainability Accounting and Management, 2022, 6(2), 264-277 research hypothesis 'CSR performance positively impacts reputation' unacceptable. This empirical examination contributes to the existing literature and is consistent with previous studies proving the contrary to our hypothesis (Axjonow et al., 2018;Bartley, 2007). The explanation for this is CSR is a double-edged sword. CSR may be practiced by a company for benefitting the society at large or may be done as a resource allocation decision to maximize its profits (According to Theory of the firm, Friedman, 1970). CSR may not be perceived as a philanthropic act of the company that enhances reputation rather is an act of profit maximization or just compliance to the mandate put on the company. CSR in India has been reduced to the mere accumulation of projects without creating any social impact. Companies have been seen behaving irresponsibly on this front in addition to their non-compliance with the CSR laws. Companies are using the scarce resources of the economy; therefore, they ought to behave responsibly towards its development. Companies need to realize and comprehend that their activities play an integral role in the development of the nation. However, all this is only possible when contributing towards the development of society comes from one's inner conscience and not because of the fear of getting sanctioned in case of non-compliance. The current study used real-world data extracted from reliable sources like published annual financial statements to compute Corporate Reputation while many earlier types of research used survey data to capture Corporate Reputation (Khan et al., 2013;Siltaoja, 2006) that might lack the objectivity and concreteness. By using these reliable measures for reputation and CSR performance (index based on GRI 4), the result obtained from this study provides validity to the contrary proposition and contributes greatly to the existing literature thereby paving the way to the future body of research. In addition to this the study has brought forward the scenario in a new environment (India) which is an instrumental contribution to the literature in the field. ORCID Loopamudra Baruah https://orcid.org/0000-0003-2435-4973