5. Results and Discussion
5.2 Sub-‐hypotheses
5.2.3 Sub-‐hypothesis 3: Operational risk
N=31 T-‐test Wilcoxon’s
signed-‐rank test
Day T-‐value Mean Std. p-‐value
-‐1 -‐1,469 -‐0,472% 1,788% 0,038*
0 -‐,288 -‐0,105% 2,020% 0,126
1 -‐1,708 -‐0,453% 1,478% 0,057
2 ,582 0,086% 0,825% 0,248
3 -‐,871 -‐0,192% 1,225% 0,176
CAR event -‐1,789 -‐1,135% 3,533% 0,027*
*** p<0.001, ** p<0.01, * p<0.05
Table 5.10. Top performing high-risk companies, 2013
When looking at abnormal returns for each day in the event period as well as cumulative abnormal returns over the whole event window for the top performing companies in operationally high-risk industries, the results differ between the days.
However, none of the days or CAR shows a significant result for the parametric test.
For the non-parametric test on the other hand, day -1 is significant at a 5 % significance level, as is the CAR for the whole event period. Over the whole period, as well as for each day separately except for one, the impact is negative. Hence, from the results, a clear trend of negative returns is observed, which implies that a high performance in terms of environmental activities not necessarily is awarded among environmentally high-risk companies’ stock returns.
N=31 T-‐test Wilcoxon’s signed-‐rank test
Day T-‐value Mean Std. p-‐value
-‐1 -‐,486 -‐0,326% 3,733% 0,557
0 -‐1,341 -‐0,556% 2,308% 0,337
1 -‐,694 -‐0,220% 1,764% 0,367
2 -‐1,966* -‐1,049% 2,972% 0,164
3 -‐2,842** -‐1,622% 3,178% 0,003**
CAR event -‐2,679** -‐3,549% 7,376% 0,009**
*** p<0.001, ** p<0.01, * p<0.05
Table 5.11. Bottom performing high-risk companies, 2013
For the bottom performing companies, the results show an even more negative trend.
In this case, all days show a negative impact, in which day 2, 3 and the whole event window-CAR are significant at a 5 % and 1 % level, respectively, for the parametric test. For the non-parametric test, day 3 and the whole event window are significant at a 1 % level. Hence, the null hypothesis can be rejected at a 1 % level for the whole event. These results imply that companies that are considered environmentally risky as they are operating in high-risk industries, and perform badly in terms of environmental activities, are punished through a negative effect on stock return.
When comparing the impact from the CSR ranking release on the top- and bottom performing companies respectively, a negative trend is observed for both segments.
For almost all days, the abnormal returns are negative and the same is shown for the whole event period. This is particularly observable for the bottom performing companies where the null hypothesis should be rejected at a 1 % level for the whole event period, implying that CSR engagement, or in this case poor CSR engagement, in fact has an impact on the investors’ perception of the firms. Companies who are operationally risky and do not put enough efforts into CSR activities are clearly punished.
5.2.4 Sub-hypothesis 4: Company Size as Defined by Market Capitalisation
N=30 T-‐test Wilcoxon’s
signed-‐rank test
Day T-‐value Mean Std. p-‐value
-‐1 -‐2,341** -‐0,313% 0,732% 0,016*
0 -‐,832 -‐0,083% 0,557% 0,422
1 -‐,161 -‐0,020% 0,691% 0,922
2 ,097 0,013% 0,757% 0,583
3 ,247 0,043% 0,969% 0,769
CAR event -‐,963 -‐0,280% 1,595% 0,229
*** p<0.001, ** p<0.01, * p<0.05
Table 5.12. Top performing large cap companies, 2013
The fourth sub-hypothesis shows significant results for the first day of the event period for the top-performing large cap companies. The impact on stock returns are significantly negative at a 1 % significance level when using the parametric test, while the results are significantly negative at a 5 % level for the non-parametric test.
The direction of impact is varying between the days, but over the whole event period the impact is slightly negative, although non-significant.
As the test shows, only the first day of the event period, i.e. day -1, is significant and the overall direction is varying between the days. This could be regarded as contradicting to theories about large companies and the importance of their responsibility-taking as a corporate citizen. However, the results might simply be explained by an efficient market and investors’ awareness, indicating that the CSR information already is incorporated in the stock price. Another potential explanation to the insignificant results could be a lack of interest for CSR and Folksam’s CSR ranking report among investors. This will be further elaborated on in the Analysis chapter.
N=29 T-‐test Wilcoxon’s signed-‐rank test
Day T-‐value Mean Std. p-‐value
-‐1 -‐1,384 -‐0,266% 1,034% 0,056
0 1,076 0,308% 1,539% 0,567
1 ,874 0,111% 0,681% 0,417
2 ,168 0,025% 0,808% 0,940
3 -‐,313 -‐0,056% 0,963% 0,837
CAR event ,271 0,121% 2,417% 0,673
*** p<0.001, ** p<0.01, * p<0.05
Table 5.13. Bottom performing large cap companies, 2013
The results for the bottom performing large cap companies are similar to the top performing companies, however with no significant results at all. The overall trend appears to be positive on the event day and the following two days, while negative on the first and last day. The total effect on the stock price measured by CAR is slightly positive, however not significant.
The large companies segmented into bottom performers are not proven to have any significant abnormal returns, but the overall direction is slightly positive. These results can potentially be explained by the fact that their ranking as “bottom-performers” implies that they are doing less or reporting less on their CSR work, compared to the top-performers, rather than “doing bad”. Another potential explanation is that investors are already aware of the companies’ lack of CSR engagement, or that they simply do not care about it and therefore do not incorporate it in their investment decisions. Moreover, as can be seen in Appendix 9.4, most companies segmented into the bottom-group of these large companies in fact do not have “bad” rankings. The appendix shows that the actual bottom-performers, all companies included, often are found in small- or sometimes mid-cap segments. This will also be discussed further under Analysis. As a result, what is considered to be
“bad performing” among the large cap companies may not be perceived as being bad at all among investors.
To conclude, the release of Folksam’s CSR ranking report does not have any substantial impact on large cap companies, no matter their ranking score. However, worth mentioning is that the number of observations for both tests under this
hypothesis does not fulfil the requirements of the normality assumption of 30 observations for the parametric test.