perfor-mance, the coefficient of SPS is big, and positive, indicating that a firm’s high social performance transcends into a higher ROA. This may be caused by a corresponding low employee turnover ratio and high employee morale. Nevertheless, the ability to analyze and interpret our findings in relation to H.2.2 is limited.
Overall, considering the lack of significant findings, we find investors as less in-terested in the particular ESG components’ performance of publicly listed companies in our sample. Instead, considering the significant findings of H.1.1, they seem to closely inspect the overall ESGP score as a proxy of ESG performance. This is con-sistent with Refinitiv’s assessment (2020), that the overall ESG score is the main focus for investors. The negative individual e↵ects, albeit statistically non-significant, are consistent with the overall ESG performance score.
8.3 RQ3 – E↵ect of ESG Scores on Financial
sustainability proxies on financial performance in fact di↵er over time. The cause for this development could be the increased levels of regulatory pressure in Europe and public awareness surrounding ESG matters. As a result, investors’ sentiment may have shifted over the years, putting greater emphasis on good ESG performance of publicly listed corporations. We can confirm this for the ESG proxies of ESGP, EPS and SPS. Even more so, we find a significantly positive relationship of ESGP and logTobin’sQ for the year 2019. Albeit exhibiting a relatively small marginal e↵ect, this corresponds to a sign-reversal from our overall finding of the main regression displayed insection 8.1. This trend is consistent for the pillar scores of EPS and SPS as well.
Conversely, when testing for accounting performance, via ROA, only the ESG disclosure (ESGD) score exhibits a statistically significant change in coefficient over time (versus the base year). Consequently, the relationship turns from a negative to a positive one when comparing the base year to the interaction terms of the years 2014-2019. Consequently, we can reject the null-hypothesis that time has no e↵ect on the relationship between lagESGD and ROA. It therefore appears that increased transparency on ESG matters yields a positive e↵ect on a firm’s profitability. This sig-nificant change in e↵ect of the ESGD-ROA relationship, starting from the year 2014, coincides with the introduction of the non-financial reporting directive 2014/95/EU.
Said directive put extensive demands on corporate ESG reporting, and simultane-ously manifested the importance of ESG reporting for other stakeholders. Following the regulatory lead of the EU, other stakeholders seem to put increased focus on a company’s ESG disclosure, mirroring the observed e↵ect on accounting performance.
For instance, a greater appreciation for a reduction in information asymmetry be-tween firm’s stakeholders (e.g. banks) and a reduced perception of risk seem to gain additional importance. As a result interest payments may be lowered, having a posi-tive e↵ect on ROA. Additionally, a corporation’s generated income may increase due to an increase in ESG reporting, with more sustainability-conscious consumers being better able to assess a company’s sustainability e↵orts.
As previous research on the changing e↵ect of ESG performance and financial per-formance is limited, we are unable to make any remarks on comparisons to extended literature. We can conclude that albeit being on average negative for the whole sample period of 2010-2019, the e↵ect of lagESGP, lagEPS and lagSPS on Tobin’s Q exhibit an increasing coefficient trend over time, culminating in a positive relationship in the most recent years of our sample period. Similarly, when investigating the e↵ect on
accounting performance, the sustainability proxy of ESG disclosure follows a similar pattern. We expect this trend to gain further momentum in the future, with an in-creasing number of investors incorporating ESG performance, and other stakeholders including corporate transparency, into their decision making.
Lastly, a remark on the statistically insignificant ESG pillar scores of section 8.2.
The positive development of pillars scores over time might explain the insignificant findings displayed insection 8.3. Consequently, for both the EPS and the SPS e↵ect on logTobin’sQ over time, the overall e↵ect turns positive in recent years (already in 2016 for the SPS). Thus, this may somewhat distort the findings made in the previous section, resulting in a non-significant finding overall.
This concludes our analysis and discussion section. In the following section we display a number of robustness tests, which we deem crucial in testing the validity of our data.
Robustness Checks
We run several robustness tests to check the reliability of our aforementioned findings.
Firstly, we check the findings in light of the diverse set of panel regressions applied in the extended literature. Thus, we test whether a di↵erent conclusion in the model specification section would have altered our findings. Secondly, we test the impact of our self-induced survivorship bias. Consequently, regressions for each CSR proxy and FINP will be run without deleting those companies, which fail to record 4-consecutive year observations. Thirdly, we check whether our negatively significant finding for ESGP on market performance is robust to being ran for each industry separately.
This is done to ensure that the omitted industry-dummy variable is not deterring our made findings. Lastly, we run the regressions across each di↵erent country to see whether choosing an individual country would have significantly altered our results.