This paper examines and evaluates the performance of an ESG factor constructed using publicly available data of stocks on the Copenhagen Nasdaq OMX in the period 2010-2020.
Firstly, An ESG score is constructed as the sum of a firm’s environmental, social, and governance score. The environmental score is calculated as the negated normalised value of a firm’s total scope 1, 2, and 3 greenhouse gas emissions with data originating from the CDP. The social score is calculated as the sum of a firm’s normalised productivity and staff compensation with data retrieved from Compustat. The governance score is calculated as the negated sum of a firm’s operating accruals, investing accruals, and working capital accruals normalised using total assets with data retrieved from Compustat. Using the ESG score, all firms are ranked 0-10. Ten is given to the firms with the highest ESG and 0 is given to the worst ESG performing firms. The firms are then divided into quintiles so that the ESG factor can be calculated as the best minus worst ESG quintile. The results show that the ESG factor has a superior excess annual return and lower annual total risk compared to all other quintiles. The ESG factor portfolio also has the lowest value-at-risk and highest Sharpe ratio.
Using the value-weighted ESG factor portfolio in combination with five different asset pricing models to explain the cross-section of returns of all stocks on the Copenhagen Nasdaq OMX yields an ESG factor that is never significant. The alpha is always significantly different from 0. The ESG factor is never significant, and there is, therefore, no evidence that an ESG factor is currently priced into listed firms on the Copenhagen Nasdaq OMX.
Next five different asset pricing models are used to explain the returns of both the equal-weighted and value-weighted ESG factor portfolio. It is concluded that the constructed social factor drives the returns of the constructed ESG factor. The environmental and governance factors do not contain significant excess returns and, therefore, do not seem to add much to the excess return of the ESG. Furthermore, the significance of the returns of both the constructed ESG factor and the MSCI ESG factor can be decreased by controlling for momentum. The loss of significance when controlling for momentum indicates that the returns of ESG factors can, at least in part, be explained by momentum.
Finally, various risk measures of the firms on the Copenhagen Nasdaq OMX is predicted using the E, S, and G as key explanatory variables while also accounting for firm characteristics and quality profile. The findings show that the risk of firms can be predicted using, mainly, the social score of a firm, but also the governance score provides explanatory power. The social score significantly predicts a negative effect on the total risk and stock-specific risk. The governance score significantly predicts a positive effect on the total risk.
The environmental score significantly predicts a negative effect on the beta. Overall the results show compelling evidence towards the opportunities of ESG factor investing, but the lack of consistency among ESG relevant data makes the construction of an ESG factor troublesome.
|Educations||MSc in Finance and Investments, (Graduate Programme) Final Thesis|
|Number of pages||82|