This thesis combines passive asset management with exchange-traded funds (ETFs) to socially responsible investing (SRI) as they both are current hot trends in the financial markets. Mutually exclusive, both the ETFs and SRI are gaining market significance in asset size. This thesis examines the two trends as a combined investment vehicle. This part of the thesis will discuss the results from the empirical part and link them to the theoretical part. It will provide answers to whether the socially responsible ETFs perform consistently and can these investment vehicles generate alpha for the investors over the passive counterpart. Furthermore, it will provide answers that what are the strategies (Inclusion or Exclusion) and attributes (ESG and product-related) that were driving the results.
The empirical part investigates if the SRI ETFs generate better alpha consistently over the passive counterpart consisting of the S&P500 ETFs portfolio. Exploiting a unique survivorship-free data set of 121 passive U.S. equity SRI ETFs from the time period of January 2010 to December 2020. The results from the Carhart four-factor model imply that on the full sample period from January 2010 to December 2020, the passive S&P500 ETFs are performing better than the passive SRI ETFs. The SRI ETFs are generating a negative annual alpha of 7.06%, while the S&P500 ETFs generate negative annual alpha of 1.86%. These results are significant for both of the sample groups. These results are similar to Meziani (2014), who provides evidence that the SRI ETFs are not performing as well as their benchmark index prior to 2015. However, narrowing down the sample period to January 2015 to December 2020, the SRI ETFs start to perform better than the S&P500 ETFs. The SRI ETF portfolio yields 0.08% annual alpha while the S&P500 ETF portfolio yields annual alpha of negative 2.32%. The SRI ETF portfolio loses its significance in this time period. Furthermore, limiting the sample period to account for the last two years (January 2019 to December 2020), the SRI ETF portfolio yields 5.33%
annual alpha while the benchmark portfolio is yielding negative 2.60% annual alpha.
These results stay significant correspondingly at 10% level and 5% level. Thus, the difference portfolio increased to 6.74% in favor of the SRI ETFs.
This thesis provides evidence that passively managed socially responsible ETFs are not consistently losing to the passive ETF counterpart that does no SRI screening. The results are also consistent even after controlling for the fees, which are higher for the SRI ETF portfolio than for the S&P500 ETF portfolio that does no screening. Therefore, as the previous evidence finds out, socially responsible investing can generate better alpha than a conventional counterpart. Furthermore, this thesis is a significant contribution to academia considering SRI ETFs. The previous research of passive SRI funds mainly focuses on mutual funds and founds no consistent evidence for underperformance or overperformance (e.g., Bauer, Koedijk & Otten, 2005; Renneboog et al., 2008; Chang, Nelson & White, 2012; Nofsinger & Varma, 2014). Therefore, we can think that the results are in line with the previous studies on mutual funds that SRI is not consistently losing to its counterparts, and it can generate better alpha depending on the time period and data sample used.
On the other hand, the results are statistically significant at the full sample period favoring the S&P500 ETF portfolio, which overperforms the SRI ETF portfolio. The results are also statistically significant at the last sample period consisting of the two last year’s where the SRI ETFs overperform the S&P500 ETFs. The economic significance of these results is also moderate since on the full sample period, the return difference was 6.58%
annually in favor of the S&P500 benchmark portfolio, while the last two years, the return difference was 6.74% in favor of the SRI ETF portfolio. Therefore, there is an economically huge return difference between the two portfolios.
As demonstrated in the theoretical part, the ETFs and SRI are both relatively new financial inventions. They have both had significant growth after the financial crisis of 2009. The unique data sample in this thesis also provides evidence that most of the SRI ETFs are issued after the year 2015. Therefore, the sample period from 2010 to 2015 demonstrates the beginning of the SRI ETF industry when there were only 24 SRI ETFs available for investors. The markets and instruments for SRI ETFs are developing more after the year 2015. This can explain the results from the regressions. The younger the industry for SRI ETFs is, the poorer the instruments are performing. While the industry
develops after the year 2015, the SRI ETFs start to perform better when the screens and methods develop with the products.
The second objective is to examine what are the differences between screening strategies (Inclusion and Exclusion) by SRI ETFs and what are the attributes (ESG and product-related) that create the value in SRI ETFs and drive the performance results.
Previous evidence points out that the different screening criteria’s and ESG dimensions might have a different effect on financial performance. For example, Derwall et al. (2005) find firm-specific abnormal returns on environmentally clean firms, Edmans (2011) and Derwall et al. (2011) on firms with high employee satisfaction, and Bebchuck et al. (2009) on firms with good corporate governance, and Kempf and Osthoff (2007) on firms with good environmental performance. Further, Humphrey and Tan (2014) argue that using exclusion can result in increased risk and lower returns, and Nofsinger and Varma (2014) show overperformance for funds that use positive screening in the ESG dimension.
The empirical part of this thesis divides the sample group of SRI ETFs into different categories regarding their screening style and ESG attribute. Inclusion strategy (positive screening) demonstrates significant negative alpha of 6.31% annually for the full sample period of January 2010 to December 2020. Furthermore, Environmental Inclusion demonstrates significant negative annual alpha of 5.61% in the first sample period.
Narrowing down the sample to the last two years, the same categories now exhibit huge positive abnormal returns. The inclusion strategy yields annual alpha of 7.81%, while the Environmental Inclusion yields annual alpha of 13.80%, being statistically significant at a 1% level. Therefore, Environmental Inclusion drives the results statistically significantly as well as economically significantly. These results are similar to Kempf and Osthoff (2007), who demonstrate overperformance for firms with good environmental attributes, and to Nofsinger and Varma (2014), who demonstrate overperformance for positive screening.
This thesis and its findings offer answers for investors considering socially responsible investments, ETFs, and passive asset management, individually and as a combined
investment instrument. The results imply that investors considering passive U.S. equity SRI ETFs, the investors are not consistently losing to a passive counterpart that does no screening for SRI. The latter part of the research period, from January 2015 to December 2020, demonstrates that the investors have the potential to gain value from the SRI ETFs as the industry develops. Furthermore, the results imply that it is the SRI ETFs that incorporate Inclusion (positive screening) and further the Environmental Inclusion as a strategy that drives the abnormal returns. In other words, this thesis provides evidence that the investors do not consistently lose to counterparts by choosing an SRI ETF.
As a final conclusion, the financial performance and sustainable attributes do not exclude each other when investing with ETFs. Therefore, one can do great while doing good in the investment landscape, and the ETFs offer a transparent and cost-efficient vehicle to practice this. This thesis gives reference to investors that they can be satisfied with the financial performance when choosing ETFs incorporating Environmental Inclusion.
The results cover only the U.S. equity ETF market and therefore extending the results to outside the U.S. is misleading. However, as the U.S. is the leader and market-dominant in the field of ETFs and SRI, it offers the widest research in the field. Other limitation is the available data for SRI ETFs. This thesis focuses on the most recent eleven years (January 2010 to December 2020). As the SRI ETF industry develops more and more time-series data will come available to improve the results of the performance axis. Still to mention, the data should be survivorship bias free. However, collecting the data manually from many different databases, there is a possibility that some funds are left out of the scope of this thesis. The problem is that there is no encompassing way to gather all information of the available SRI ETFs. Similar to Nofsinger and Varma (2014), many researches are carried out to obtain the most suitable dataset available.
The future research of passive SRI ETFs could further narrow down the ETFs to purely passive ETFs by using the Active share presented by Cremers and Petajistö (2009). As the SRI ETF industry grows, more data will come available and eases the trouble of manually
collecting the data from holdings and strategies. In addition, as time passes, much more time-series data of the SRI ETF performance will come available, and furthermore, more products will come available.
Future research could also focus on evaluating the socially responsible ETFs more precisely, for example, by measuring the sustainable rating providers empirically to choose the right corporations acting in a socially responsible way. Also, a need for consolidating the ratings as united is needed like Dorfleitner et al. (2015) demonstrates that different ratings are affecting the results. To answer what is actually socially responsible still persists. Gladly, initiatives like the EU Taxonomy have the potential to harmonize the industry. Based on this thesis, future research could also investigate how much investors actually appreciate the differences between the non-financial attributes and the financial performance.