ESG and SBTi

This project was created for an assignment in Asset Pricing and Climate Mitigation at KTH - Royal Institute of Technology. The purpose is to investigate whether companies adhering to high ESG-standards or those with commited and approved SBTi (Science-Based Target Intiative) targets (green firms) outperform stocks with lower standards (brown firms).

ESG

In the file ESG.R stocks are sorted based on ESG-quartiles (putting the 25% of stocks with highest ESG-scores in top portfolio, next 25% in second best and so forth). Since ESG-scores are relative to companies in the same sector, these portfolio should not in theory be overwighted in specific industries. These four portfolios are then compared to an equally-weighted (EQW) portfolio where all stocks in the dataset are equally-weighted independent of other factors (such as ESG-score, industry and market cap).

Portfolio performance for ESG-quartile portfolios (2011-2022)

In absolute performance there seems to be an inverse relationship between ESG-score and financial performance with the worst ESG-portfolio performing best. ESG-scores neither seems to decrease risk (as measured by $1\%$-VaR and volatility $\sigma$) and the inverse relationship therefore holds true also on a risk-adjusted basis (Sharpe Ratio).

SBTi

In the file SBTi.R another variable indicating a company's sustainability maturity is used, whether they have approved SBTi-tarrgets or not. Since it is a volountary for companies to have SBTi-targets or not, and since they require companies to have realistic and approved plan to achieve net-zero it is a good indication of their sustainability work.

To test this, two portfolios are created, one containing SBTi-stocks and one with non-SBTi-stocks. When a company in the dataset have their SBTi-targets approved they are moved from non-SBTi portfolio to SBTi-portfolio. With a requirement of minimum 5 stocks in the SBTi-portfolio, the portfolio can first be formed in September 2019. From then until end of 2022, the performance of the two portfolios is almost identical. Performing a regression on excess returns (controlled for risk using CAPM) we can see that SBTi have low or no performance on financial performance for the stocks in the dataset.

Portfolio performance SBTi and non-SBTi portfolios (2019-2022)

Source

Data is retrieved from Eikon and filtered based on three conditions:

1. Available price data from 2011 to 2022
2. Avaliable ESG-data from 2010 to 2021 (since lagged ESG-scores are used to form portfolios for coming year)
3. Headquarted in the US (to avoid market effects and for risk-free yields and benchmark returns to be valid)