Do Climate-Related Exclusions Have an Effect on Portfolio Risk and Diversification? A Contribution to the Article 9 Funds Controversy

Do Climate-Related Exclusions Have an Effect on Portfolio Risk and Diversification? A Contribution to the Article 9 Funds ControversyWhitepaper | May 2025

Abstract

Despite regulatory efforts to enhance consistency, the increase in funds claiming to be sustainable has led to polemics about some of the funds’ stock holdings having a negative impact on climate change. In contrast to the existing literature which focuses on short term performance, we examine the impact of these controversial stocks on the risk profile of 161 funds with a ‘sustainable investment objective’. We show that excluding these stocks with a naive reallocation technique has a limited effect, which can be further reduced via an optimization procedure. These results suggest that holding stocks with a negative contribution to climate change currently has no justification from a risk perspective.

Key takeaways:

Out of 161 funds reporting a ‘sustainable investment objective’ in accordance with the European sustainable financial disclosure regulation (article 9), 50 contain stocks of companies involved in coal, oil and gas or aviation. These stocks represent on average 2.8% of the funds’ invested capital. Excluding these stocks with a naive (pro rata) reallocation leads to an average tracking error of 0.53% with the original portfolio.
Sector deviations occur mainly in the ‘Energy’ and ‘Utility’ sectors. In terms of fundamentals, exclusions increase exposure to higher ‘quality’ stocks, whereas exposure to ‘investment’ and ‘value’ stocks is slightly reduced. Excluding these stocks with an optimized reallocation technique further reduces the average tracking error to 0.42%, while also reducing fundamental differences between the original and optimized portfolio. These results suggest that excluding climate controversial stocks has overall a limited impact on the risk profiles of the funds.

Introduction

With the increase in funds claiming to be ‘responsible’, ‘sustainable’ or ‘impact’, the regulators are clarifying the scope of these funds in order to improve transparency and prevent ‘financial greenwashing’. As part of its Action Plan for Sustainable Finance, the European Union (EU) now imposes, through its Sustainable Finance Disclosure Regulation (SFDR), reporting requirement that depend on the ambition of the funds: funds that claim to have a ‘sustainable investment objective’ must comply with ‘Article 9’ requirements, while funds that highlight environmental or social characteristics as part of a broader investment strategy must comply with ‘Article 8’ requirements. These two categories of reporting requirements have promptly been considered by the market as ‘labels’ associated with the funds, Article 9 funds being the most ambitious in terms of extra-financial impact. However, the reporting requirements of the SFDR leaves room for interpretation of the definition of a ‘sustainable’ investment, as well as flexibility with respect to the levers to be used to achieve the ‘sustainable investment objective’: exclusions, reallocation, shareholder engagement.

The flexibility left to managers in the implementation of impact strategies is at the root of a significant controversy concerning funds considered as ‘Article 9’. In November 2022, two investigative journalism platforms – Follow the money and Investico – published the Great Green Investment Investigation (GGII) in collaboration with a dozen European media1. Their work revealed that nearly half of the 838 European ‘Article 9’ funds had invested in companies that are involved in the fossil fuels and aviation sectors and that the journalists identified as significant contributors to global warming due to their lack of ‘a credible climate strategy’. Following this investigation and other controversies, 40% of the initial ‘Article 9’ funds – many of these being passive funds tracking EU Climate Transition and Paris-Aligned benchmarks – where voluntarily downgraded to ‘Article 8’ by the fund managers at the end of 2022 according to Morningstar.

Several reasons may explain the holding of these controversial stocks in sustainable portfolios: data discrepancies in the assessment of their environmental impact, a willingness to engage with the company as a shareholder with the objective of improving their practice, or a strategy of gradual exclusion. Whatever the reason, a question that arises sooner or later for the asset manager is to know what the impact of a possible exclusion on the performance and risk profile of the fund would be. There is no consensus on the existing literature whether the integration of environmental, social, and governance (ESG) criteria in investment funds affects the financial performance of the funds (Friede, Busch and Bassen 2015). One of the reasons for this lack of consensus is the broad spectrum of practices that fall under ‘integration’ of ESG criteria. Among these different practices, several studies have investigated the effect of the exclusion of stocks, with mixed results (e.g., Khajenouri and Schmidt 2021; Capelle‐Blancard and Monjon 2014). Even when focusing on the exclusion of fossil fuels, the effect on funds’ performance remains uncertain: Henriques and Sadorsky (2018) find negative effects while Trinks et al. (2018) find no difference and Hunt and Weber (2019) find positive effects. Our study aims to contribute to this literature by focusing on the effects of exclusion on risks rather than short-term performance. The risk profile of a fund is responsible for its long-term performance patterns, and thus it is a primary concern for managers. We therefore seek to answer the following question: what would be the impact of excluding stocks associated with climate change on the risk profile of funds that have a ‘sustainable investment objective’ (‘Article 9’) funds?

To do so, we study a sample of 161 funds considered as ‘Article 9’ as of July 2023 and study the effects of excluding stocks involved in coal, fossil fuels and aviation on their tracking error, their sector allocation, and their exposure to the five financial risk factors developed by Fama and French (2015).

We find that of the 161 funds studied, 50 funds contain companies that are involved in at least one of the three climate-related controversial activities and that these companies represent on average 2.82% of the funds’ portfolios. Overall, the exclusion of these stocks has a small effect on the funds risk profile. The tracking error between the initial portfolio and the portfolio without the controversial stocks is on average 0.53% (the median is 0.32%) when the exclusion is performed using a naive (pro-rata), and 0.43% (the median is 0.23%) when the reallocation is optimized. Similarly, when we look at the tracking error relative to a regional benchmark, the changes remain small: on average the initial tracking error is 4.74% (before exclusion), and for 50% of the funds, the increase is less than 0.03% with the naive reallocation. The relative sector deviations that we observe are also small, and they affect mainly the ‘Energy’ and ‘Utility’ sectors. However, we find that excluding controversial stocks does affect exposures to traditional risk factors when performed naively. For instance, we observe a slight increase in average exposure to the ‘Quality’ factor (from 0.11 to 0.13), whereas exposure to the ‘Investment’ and ‘Value’ factors is reduced (from -0.05 to -0.07). As for the tracking error, the optimization method also mitigates the effect of exclusions on the exposure to factors.

These results suggest that maintaining the current risk profile does not constitute a ground for holding these environmentally controversial stocks, as they could be excluded without significantly altering it. To preserve the credibility of funds with a ‘sustainable investment objective’, particularly in the eyes of retail investors, it therefore seems essential to improve transparency regarding the holding of ‘controversial’ assets and the reasons for such holdings.