2. Making sense of the environmental pillar in ESG investing

As market participants show greater awareness and concern that climate transition may present material and non-material risks to companies and a wider range of stakeholders; environmental, social and governance (ESG) products are increasingly being used as a tool to assess the alignment of companies with low carbon economies and to identify financially material sustainability issues. To meet growing demand for these products, investment funds and ESG rating providers are working to integrate metrics aligned with environmental impact, climate risk mitigation, and strategies toward renewable energy. These efforts can be broadly considered the environmental (E) pillar of ESG ratings and investing.

At the same time, two important developments are occurring within sustainable finance that relate to the purpose of ESG. First, there is a sharp growth of institutional investors using ESG approaches with the aim to enhance the long-term value of their investments, as measured by superior risk-adjusted financial returns. Incorporating risks from climate change and stranded assets as a result of the climate transition is increasingly recognised as a central element to this assessment. Second, there is a growing commitment by institutional investors – with financial or impact objectives -- to strengthen the climate resilience of economies by disincentivising carbon emissions. While ESG assessment methodologies of major ratings providers and global investors appear to strive to incorporate both perspectives, questions remain as to whether this is being achieved in practice. Chapter 1 of this Outlook assesses the alignment of ESG investing with superior performance of risk-adjusted returns, finding little evidence of outperformance over the past decade. Building on this analysis, this chapter focuses on the environmental pillar of ESG investing and assesses the extent to which E score methodologies in their current form align with the expectations of the investors and stakeholders that use them.

The extent to which E scoring and investing reflect the environmental impact, carbon footprint and resource use of these investments is critical to enable market participants to make informed decisions relating to a low carbon transition. This can include a strategic re-orientation towards renewables, climate-related risk management and adaptation,1 as well as operational processes to improve water use, waste management and impact on biodiversity. In its current form, the E pillar includes metrics across all of these areas, with limited scope to distinguish between those that require strategic re-orientation in the long-term and those that require easily implementable operational changes in the short-term. In addition, as market participants seek to understand their exposure to potential risks, the E score itself does not prioritise carbon footprint or intensity within the range of metrics that comprise the E score, so may be of limited value in protecting portfolios from climate transition risks such as stranded assets. This means that E scoring may not yet be suitable for use by investors wishing to use it as a proxy for judging the carbon footprint of companies in line with low-carbon economies.2

To varying degrees, institutional investors and central banks are using ESG considerations, and in turn E scores, to rebalance their portfolios for long-term value, and in some cases as a proxy for a greening of the financial system. This raises two important points for consideration. First, for those wishing to invest in products with long-term risk adjusted returns, the E scores should have a transparent methodology to help investors recognise the extent to which these scores can be used to contribute to long-term value. Second, for investors wishing to reduce carbon exposure, the environmental pillar should be clearly discernible as to the extent to which it aligns with risk management or carbon exposures. Establishing consistent E scoring metrics and methodologies within ESG ratings will be an important step to contribute to these efforts. This should include clear and publicly available information on metrics with guidance on the underlying methodology. This should go hand in hand with greater clarity on required E pillar reporting by companies; establishing core metrics with associated metadata for measurement would be a valuable first step. While ESG rating providers tend to have similar metric categories for E scoring, there are wide differences in both the number and choice of quantitative metrics and the way in which metrics are calculated and weighed. Within this, methodologies that score companies positively based on superior climate risk management and renewables transition plans, despite poor performance on emissions and other metrics, can be valuable, but may merit further scrutiny. At the very least, disparities in underlying metrics should be made available to market participants to reduce green-washing and any misconception that environmental scoring is consistent with a low carbon framework.

This chapter assesses the extent to which E scores align with carbon emissions and core metrics that negatively affect the environment, to examine whether E scoring and reporting effectively serve markets and investors that are using ESG investing in part as a tool to make portfolios more resilient to physical and climate transition risks. This point of inquiry will have implications for the approaches of rating providers, institutional investors, central banks and asset managers who wish to use it as a proxy to green the financial system. Section 2.1 outlines related efforts by central banks, financial regulators and governments. Section 2.2 includes empirical analysis of how E scores may contribute to efforts to tilt portfolios toward low-carbon investments, and the extent to which these E scores may influence the emission composition of high-ESG portfolios. Given limited publicly available information in this area, the analysis focuses on the three main rating providers for which information has been shared. Given the abundant availability of their data, this assessment is intended to reflect more broadly on industry practices, rather than to scrutinise their practices. Beyond this, alternative investment products tailored to low carbon activities are also discussed, to illustrate that a range of products exists to align investments with low-carbon transitions. Section 2.3 reflects on implications following the outbreak of the COVID-19 pandemic, including potential opportunities to support a sustainable economic recovery. In addition, policy considerations to improve the E of ESG scoring and reporting going forward are provided.

The key findings in this chapter suggest that E scores often do not align with current carbon emissions exposures, and can be difficult to interpret due to the multitude of diverse metrics on environmental factors. Some of the ESG rating providers analysed exhibit a low correlation between the E score and the overall ESG score. While this is not unexpected, given that the E score is distinct and ESG scores include social and governance metrics, it confirms that investing in high-scoring ESG portfolios does not necessarily equate to a positive environmental impact or greening of the financial system. The analysis also shows that for some ESG rating providers, high E scores positively correlate with high carbon emissions.

ESG scoring and reporting has the potential to unlock a significant amount of information on the management and resilience of companies in line with sustainability, including environmental and physical climate risks when pursuing long-term value creation. It could also represent an important market based mechanism to help investors make decisions on long-term carbon prices and climate transition risks implied by climate mitigation policies and the Paris Agreement.3 Institutional investors, including central banks, use ESG ratings, and possibly E scores, as tools to green their portfolios and in turn the financial system. This means that more effort is needed to ensure that E scoring provides the information that market participants need to design and implement effective strategies to address environmental factors and the transition to low carbon economies.

Looking ahead, greater consistency on the high-level purpose of the E of ESG investing could provide stakeholders and investors with a framework to assess the level of environmental4 and financial materiality5 for aspects of the E pillar. This is ever more pressing in the context of COVID-19 as governments channel significant resources into investments that will inevitably impact the path of the climate transition.

Many investors around the world, including a number of central banks, are paying greater attention to sustainable activities in their investment portfolios and risk assessment practices with the aim of addressing physical and transition risks as a result of climate change. In turn, ESG investing is seen as an important tool to reduce information asymmetries using non-financial reports, which could be utilised to help facilitate greening of portfolios, and in aggregate, the financial system. It represents an important market-based mechanism to help investors make decisions based on environmental aspects of firms, including emissions, resource use and how the company is forward looking in terms of climate innovation. In this regard the E pillar can support investors seeking to also incorporate long-term expectations on carbon prices and transition risks implied by the Paris Agreement (see Box 2.2). As a step to realign the financial system toward low-carbon economies, governments and other public institutions are increasingly seeking ways to incorporate environmental and climate related risks into investments. Initiatives such as the OECD’s Centre and Forum on Green Finance and Investment6 and the Network for Greening the Financial System (NGFS)7 have been established to support better alignment with low-carbon economies and enhance activities such as investment approaches and risk management, incorporating climate transition and its subsequent physical risks (NGFS, 2019a; OECD, 2017).

Governments have the primary responsibility for developing policy frameworks to address climate risks, yet central banks and financial market participants are playing an important role in supporting this by mainstreaming ESG considerations and identifying potential risks to market actors and the financial system. Separately, market participants and businesses are struggling to accurately estimate and price the financial impact of future climate-related risks. To improve these tools, additional collective leadership and globally coordinated action to improve the E pillar may be warranted to strengthen practices that support transparency and market integrity. This section outlines the steps undertaken by these actors to integrate ESG criteria into their roles and responsibilities.

The legal mandates of central banks and financial regulators vary throughout the OECD membership, but they typically include responsibility for price stability, financial stability and the safety and soundness of financial institutions. The primary responsibility for ensuring the success of the Paris Agreement rests with governments, yet it is increasingly considered as within the responsibility of central banks and supervisors to deliver on their substantial role in addressing climate-related risks to financial stability within the remit of their mandates. For example, understanding how structural changes affect the financial system and the economy is core to fulfilling these responsibilities. These risks might have persistent impacts on macroeconomic and financial variables (for instance, growth, productivity, food and energy prices, inflation expectations and insurance costs) that are fundamental to achieving central banks’ monetary policy mandates.

Also, physical and transition risks that can have system-wide impacts on financial stability and might adversely affect macroeconomic conditions. Physical impacts include the economic cost and financial losses resulting from the increasing severity and frequency of extreme climate change-related weather events (such as heat waves, landslides, floods, wildfires and storms) as well as longer term progressive shifts of the climate (such as changes in precipitation, extreme weather variability, ocean acidification, and rising sea levels and average temperatures). Transition impacts relate to the process of adjustment towards a low-carbon economy (reaching ‘net zero’). The process of reducing emissions is likely to have a significant impact on all sectors of the economy affecting financial asset values.

This financial impact of climate related risk is likely to take two forms: i) the physical risks (which are the most immediate and visible) caused by the proliferation of extreme weather events such as floods, storms, wildfires and rising water levels, and the subsequent damages that insurance firms, for instance, will have to cover;8 and ii) the transition risks associated with public authority measures or private initiatives to support the move towards a low carbon production model. Both of these factors will have an inevitable impact on valuations, demand and output, effecting financial stability, which could in turn feed back into the real economy through market losses or credit tightening (OECD, 2017).

In addition to the activities of central banks (see Box 2.1), a number of financial regulators have established work on reporting for climate-related information as well as integration of environmental risks. These are largely focused on ESG disclosure by companies,9 with guidance emerging for financials and intermediaries. Looking ahead, two considerations in this area should remain a priority. First, investors and markets need accurate and comparable information on companies’ environmental and climate-related performance and activities in order to assess to what extent this fits with investment strategies they have in place today. Second, and perhaps more difficult, investors and markets also need information on how companies are assessing and responding to risks posed to their operations in the medium and long-term as a result of carbon transition. Independent reporting frameworks are being developed, yet these vary significantly and cater to a range of stakeholders. They feed into the range of metrics used by major ESG rating providers, and often include qualitative guidance on environmental and climate related issues (OECD, 2020b). Section 2.2 further addresses these considerations.

In addition to efforts by central banks and financial regulators, there is a growing concern among both governments and businesses regarding the fundamental role of carbon pricing in this transition. While carbon pricing policies remain at an early stage of implementation, a number of OECD countries have increased their average effective carbon tax rates in the real sector, yet these efforts remain limited and in some cases are only targeted to a handful of industries (OECD, 2019c). As a result, financial markets may not be efficiently incorporating future prices of fossil fuels (see Box 2.2).

Countries are increasingly planning to implement carbon pricing policies, yet a carbon price gap remains,10 suggesting that many governments continue to under-price externalities from carbon emissions. In turn, this has an inevitable impact on the ability of markets to price risks associated with carbon-intensive activities and transition to low-carbon alternatives. Ultimately, addressing the mispricing of carbon and effectively taking into consideration the misprice risk associated with carbon-intensive activities will depend on the extent to which governments and financial market actors can accurately estimate the true financial cost of a transition to low-carbon economies.

Two important developments have occurred in sustainable finance that relate to the purpose of the E pillar. First, there is a sharp growth in institutional investors using ESG approaches with the aim to enhance the long-term value of their investments, as measured by superior risk-adjusted financial returns. In this respect, incorporating risks from physical risks and stranded assets as a result of the climate transition is increasingly recognised as a central element to this assessment.11 Second, there is a growing commitment by institutional investors – with financial or impact objectives – to strengthen the climate resilience of economies by disincentivising carbon emissions.12 While ESG assessment methodologies of major rating providers and global investors appear to strive to incorporate both perspectives, questions remain as to whether this is being achieved in practice.

To meet growing demand for sustainably branded finance products to support transitions to low-carbon economies, ESG rating providers and investment funds are striving to integrate metrics aligned with environmental resilience, climate risk mitigation, and strategies toward renewable energy. This section assesses the extent to which E scores within ESG ratings are currently suitable as a tool to guide efforts to shift portfolios to low-carbon investments, and how these E scores may influence the emission composition of high-ESG portfolios. Data from the three main rating providers (Bloomberg, MSCI, and Thomson Reuters) are used to assess the alignment of E scores with environmental metrics such as carbon emissions and waste. In addition, a comparison of portfolios weighed toward issuers with high-ESG scores and non-weighted traditional portfolios is undertaken to assess their emissions content.

Notably, this section finds that for some providers, high E scores positively correlate with high emissions, suggesting that the content of the E score is not always synonymous with low emissions or an alignment with a low-carbon trajectory. These findings raise questions as to the benefit of ESG portfolios for investors that wish to prioritise low-carbon investments or near-term carbon transition objectives. In particular, the E score in its current form is not an effective tool to differentiate between companies’ activities related to outputs that affect the environment, climate risk mitigation to improve risk-adjusted returns, and medium-term strategies to align portfolios with lower-carbon activities. Scores tend to incorporate each of these factors, but the inputs and weighting are not transparent. By contrast, other types of investment products, such as those tailored to climate transitions, provide ample opportunities for investors to rebalance portfolios away from companies with carbon-intensive outputs or supply chains. However, highly tailored low-carbon or carbon-transition portfolios may have asset compositions and risk characteristics that stray widely from standard market benchmarks that are most commonly used by institutional investors, so risks need to be balanced against climate and long-term return objectives.

Financial and official sector participants, including central banks, institutional, and retail investors, have shown an increasing interest in ESG investing as a tool to support a greening of the financial system. Yet, there remains some ambiguity about the extent to which such investors are willing and able to invest in low-carbon portfolios that deviate significantly from broad-based market indices. Despite a growing interest by many countries’ policymakers in seeing investment utilised to facilitate a climate transition through low-carbon portfolios, the question remains as to whether the criteria in ESG investing13 are sufficient to isolate and address the obstacles to this transition, or to accelerate the transition to sustainable assets and support financial markets in absorbing potential transition risks. In this respect, if ESG investing does not divert material financial resources away from those companies that currently contribute the highest levels of emissions, it raises questions about the efficacy of ESG investing to achieve green transitions and mitigate portfolio risks. Evidence of this, however, remains to be seen, and analysis in this section suggests that E scores can be difficult to interpret and do not always reflect strong performance on individual, environmental and climate related metrics.

Importantly, individual E scores do not align with overall ESG ratings for all rating providers, suggesting that high ESG scores cannot necessarily be used as a proxy for environmental performance. Depending on the rating provider and methodology, however, there is scope to use ESG ratings to align with a focus on environmental standards, in the event that investors carefully choose ESG rating providers or develop in-house rating methodologies, or in the event that transparency as to the methodology used for performance in line with E, S and G improves. Figure 2.1 illustrates that for two of the three rating providers analysed, the R squared is high, denoting a correlation between the overall ESG ratings and the E score. For a third provider, however, a low R squared signals a low correlation between a high ESG rating and a high E score. Therefore, a better understanding of the methodologies used by ESG rating providers, in particular to generate E scores, is important to understand the methodological aspects that drive these findings.

Environmental and financial factors that are material to a company can differ depending on the industry and place in the global value chain. Financial materiality refers to sustainability topics that potentially have a significant financial impact on a company’s performance, while environmental materiality considers the factors that will have a long-term impact on the society and future generations. In some cases, factors that are environmentally material can also be financial material, for example metrics used to measure environmental impact can also be important in measuring transition risks, and may change over time. Nevertheless, there are a number of environmental metrics that could be considered core in assessing any company’s impact on the environment that may not be considered financially material today. These are, overall CO2 emissions, total waste production, total energy use and total water withdrawals (see Table 2.1).

The challenges investors face relate to how sub-metrics that go into calculating these metrics may differ, ranging from the information each company reports to how rating providers aggregate this information. The methodological practices that are then used to aggregate sub-metrics and environmental metrics to define the E score are important. This section addresses these concerns by analysing the alignment between E scores and individual core environmental metrics, as well as assessing the underlying methodology used to define the E score.

The overall level of CO2 emissions is higher in companies with high E scores for two of the three rating providers assessed (see Figure 2.2). This means that for two providers, higher scoring companies generate higher gross output of CO2 emissions. This includes those stemming from the direct burning of fossil fuels and indirect carbon dioxide emitted during consumption of solid and liquid gas, gas fuels and gas flaring.14 The third ESG provider shows the inverse trend, with a reduction in average CO2 emissions observed as the E score increases.

Similarly, rating providers that show alignment between high CO2 emissions and high E scores, typically exhibit an increase in both the scope 1 and scope 2 CO2 emissions when E scores are high (see Figure 2.3, Panel A and Panel B).15 This suggests that higher CO2 emissions for high scoring companies is not driven by an irregularity in the measurement of a certain type of emissions. In addition, there is an indication that when direct emissions are high, indirect emissions from production may also be high, without a clear differentiation in the rating segment of companies with these characteristics.

Accounting for revenue, it is possible to measure intensity of CO2 emissions by unit of output (see Figure 2.4). This can reduce the bias associated with higher emissions of larger companies within high ESG ratings, and can be used as a proxy for the extent of greening by issuers, and in turn for portfolios. Using this measure, one provider still exhibits a notably higher level of emissions by higher E-scoring companies. To have a fuller picture, it will be important to measure over time the rate of change of revenue adjusted CO2 emissions (see Box 2.3), which currently none of the rating providers analysed measure.

Total waste produced and total water withdrawals are also greater in higher E scoring companies across at least two out of three rating providers analysed (see Figure 2.5). Total waste produced refers to both hazardous and non-hazardous waste produced by companies, and can include plastic waste that contaminates waterways and ecosystems. Separately, treatment and disposal of waste can also generate carbon dioxide and equivalent emissions. Higher E scores for two providers appear to correlate with higher waste produced for two of the three rating providers analysed. On the other hand, higher water withdrawals16 are seen for higher E scores across all rating providers analysed.

Interestingly, total energy use for higher E score companies is lower for two out of three providers analysed (see Figure 2.6). This is measured by gigajoules of energy use, accounting for renewable energy. Without additional information, it is difficult to decipher exactly the extent to which renewable energy is adopted, yet a reduction in non-renewable energy use for higher E scoring companies could indicate that high E score companies are also increasingly adopting renewable energy options in their production and operations.

Therefore, if investors aim to use these E scores as a proxy for environmental performance and/or to create a portfolio that aligns with lower-carbon activities, this analysis raises questions as to the usefulness of E scores in achieving this. The findings of this analysis indicate that there is little alignment between ESG ratings, E scores and individual environmental metrics. In order for markets to use E scores effectively, investors will be required to understand and choose a rating provider whose E score components and weightings fits their criteria. This would require further analysis of the extent to which underlying metrics of the E score of the rating provider they use align with their view of environmental materiality.

Transparent, accurate and comparable ESG data are critical for effective investment analysis and decision-making. In the context of the environmental pillar, it is equally important that investors have reliable information in order to prepare their portfolio for future risks that may arise from the carbon transition, and to facilitate decisions that deliver risk-adjusted returns on investment. The analysis conducted in this section suggests that when looking at the top performing companies by environmental pillar scores across leading ESG rating providers, outcomes at the metric level do not always represent a positive effect on carbon emissions and the environment. Notably, the E score captures a range of information and metrics beyond carbon emissions that may drive this trend. This is valuable in that it provides information on activities that may require environmental risk management. This said however, for investors wanting ESG portfolios to align with low-carbon economies, the E pillar may not be an appropriate proxy.

Given the result of the analysis performed, the indication is that criteria other than carbon emissions tend to drive E score rating methodologies through the divergence of environmental performance for various tranches of scores across prominent ESG rating providers. This section elaborates on this by outlining the underlying methodologies of E pillar scores and discusses differences in the scope and measurement of metrics across these providers. As such, it explores to what extent differences in environmental pillar scores by rating providers are the result of variations in methodology, or of an underlying difference in the rationale of what constitutes good environmental perfowrmance, for example an emphasis on company management versus outright emissions and environmental outcomes. Initial findings suggest that it is a combination of both. Beyond this, wide differences in the number and choice of quantitative metrics, as well as the way in which the metrics are calculated and weighed, can impact the overall E score, making the comparison of ratings between providers difficult, even when referring to the same company (see Figure 2.7). Similarly, supplementary analysis in the form of subjective or qualitative research by rating providers can also impact the overall rating, with the impact of different metric calculations and weights on the overall E score being considerable.

Despite similarities in categories of metrics, the difference in the number of metrics used and measurement criteria contributes to considerable rating inconsistencies. Preliminary research indicates that these differences may be the result of a combination of the following:

  • Differences in the scope of metrics, namely those outside of the core;

  • Difference in the measurement (i.e. absolute vs. relative) or input indicators to measure metrics;

  • Difference in weight.

Practices such as estimation in the case of missing data may also exacerbate differences between providers. The number of input metrics used by rating providers varies significantly, with Provider 1 using a selection of around 115 metrics,17 Provider 2 uses around 26 and Provider 1 around 27 metrics on environment (see Table 2.2).18

Thomson Reuters and Bloomberg adopt an outright metric value measurement approach (for example, direct CO2 equivalent emissions measured in either parts-per-million or units of micromol mol-1), whereas MSCI adopts a binary approach for all metrics (for example, three-year trend of average carbon emissions intensity measured by either -1, 0, or 1), which could include the calculation of several sub-metrics in order to decipher the overall metric score (revert back to Figure 2.7). These metrics are then used to create composite sub-scores across key issues, or rather categories used by rating providers. This quantitative element of the rating is complemented by additional analysis (in some cases qualitative analysis) of company management strategies or insights (in the case of MSCI), which can have a notable impact on the rating. The logic for this is to also assess how well a company may be preparing to deal with environmental and carbon transition risks and opportunities in future, yet the approach and methodology used for this portion of the rating can differ, and often relies on subjective judgement.

The types of environmental metrics included in each of the E rating frameworks appear to vary widely, with a multitude of possible measurement methodologies as well as underlying rationales for the inclusion of metrics. To illustrate this, metrics can be grouped as falling somewhere along the input-output-outcome-process chain (see Figure 2.8). Production-related metrics such as those measuring energy consumption or water withdrawals tend to be inputs. Emissions metrics, including CO2 and GHG emissions by source, regardless of whether they are expressed in unit value or as a share of revenue, tend to represent outputs. Outcome focused metrics can include those that look at impact such as ecological and biodiversity. Process metrics can include binary metrics or descriptions of policies and risk management practices as discussed above, including for example, information on board oversight related for climate related risks.

The logic used in this chain can also be applied to frameworks such as those set out by the Task Force on Climate-related Financial Disclosures (TCFD) which recognises the importance of: (a) metrics on climate-related risks associated with water, energy, land use, and waste management (inputs and outputs); (b) greenhouse gas emissions using the scope 1 (direct emissions), 2 (indirect emissions from direct production), and 3 (indirect emissions from activities along the value chain) definitions (outputs and outcomes), and; (c) company management processes, anticipated regulatory requirement, market constraints or other goals (TCFD, 2017b).

In addition to the difference in metrics and their associated measurement, the source and type of underlying data used differs among providers. Bloomberg collects public ESG information disclosed by companies through their sustainability reports, annual reports and websites. In some cases, Bloomberg also integrates metrics from third party rating agencies, such as Sustainalytics, the ISS quality score, and the Carbon Disclosure Project (CDP) score. Thomson Reuters also collects reported data from the public domain, including company sustainability reports, annual reports and third party providers. MSCI uses a multitude of sources, including company disclosure, third-party databases, as well as direct co-operation with companies to develop qualitative and binary metrics.

The difference in the scope of metrics, related measurement of input data, and difference in weighting used by Bloomberg, MSCI and Thomson Reuters, can account for differences in E rating scores and similarly the lack of correlation between the scope 1 and 3 emissions metrics (refer back to Figure 2.2 and 2.3) and tranches of E scores. In addition, not all metrics align well with what is financially material today (i.e. already included in company reporting or board committee activities), which may be more commonly found in the process group of metrics, nor with environmental impact and carbon footprint (i.e. output and outcome groups of metrics), nor with the longer-term transition to renewables.

Issues discussed within this section give rise to a number of challenges that may limit the usefulness of E scores, and ESG ratings more broadly, for environmental considerations in investment approaches, and may necessitate greater scrutiny by market participants using these scores. Key issues include:

  • Limited comparability: While diversity of analytical approaches may be welcome, the current state of approaches has resulted in the limited comparability of E scores across major providers.

  • Lack of transparency: Disclosure of all relevant elements of the rating methodology is vital for investors. This should include information on the main metrics used, as well as clarity on when supplementary judgements or estimations are used to arrive at the resulting E score. The majority of this information is labelled as proprietary with little made available publicly. In addition it is difficult to find information regarding weighting of various categories or underlying sub-metrics, making it even harder to understand the focus of different elements of E, S, and G within methodologies.

  • Selection bias: Larger companies have the resources to implement and communicate E pillar related strategies from renewable energy programmes, internal carbon pricing and risk management. Limited reporting on these strategies within smaller companies may impact overall E scores (OECD, 2020c).

  • Limited scope within metrics: The use of binary indicators can limit the value of metrics to measure environmental performance or carbon emissions. In addition, these can be misleading. For example, in the measurement of a company’s ability to deal with environmental scandals, rating providers may assess more favourably a company that has weathered a number of scandals compared to those that have had no scandals.

  • Subjectivity judgements: The use of qualitative, subjective questionnaires or interviews with companies, as well as research conducted in collaboration with rating providers have a significant impact on E scores and in some cases call into question the consistency of assessments (both within the agency and across key rating providers) that are highly judgement-based to contribute to the E rating process.

Interest by investors in ESG compliant investments has also spurred demand in more targeted sustainable indices and funds. This section seeks to understand how these funds and indices are built and to what extent their framework, and particularly sustainability integration, may differ from the parent benchmarks that they are derived from. This also includes an assessment of the difference in overall CO2 emissions between standard index and fund constituents compared to those in the targeted ESG index or fund. Alternative investment products tailored to the low-carbon transition are also outlined to see what benefit these may bring to investors who wish to invest in low-carbon activities.

The number of sustainable indices and funds has increased significantly in the past several years, with many large indices having ESG counterparts. The analysis in this section focuses on the S&P 500, STOXX 600 and their ESG counterpart index,19 as well as MSCI World, MSCI ESG leaders and MSCI ESG screened counterpart funds.20 Analyses of their methodologies are undertaken to understand the variation of levels of CO2 emissions between the parent index or fund and its related ESG index or fund to understand if high ESG portfolios have a lower carbon footprint than traditional funds with similar constituents. This analysis reveals a stronger presence of exclusionary screening, which removes companies based on predefined criteria,21 such as the exclusion of specific industries (i.e. tobacco); rather than titling or integration of ESG scores22 to build the ESG counterpart indices or funds. An assessment of the presence of companies from the most polluting industries is undertaken,23 to understand if and how these companies are removed or limited in ESG counterpart indices and funds, indicating a portfolio rebalancing towards companies with a lower carbon footprint.

The findings of this section highlight two issues. First, similarly to ESG ratings, ESG indexes do not currently represent low-carbon investment option for investors. Second, inconsistencies across indices in terms of the extent to which ESG indices and funds reflect stronger E performance, highlight the considerable amount of analyses required to determine which ESG investment products align with these objectives.

As such, investors may find it difficult and time consuming to extrapolate information on ESG indices and funds to the extent metrics and ratings vary, and sub-metrics that unlock more specific information may not be readily available. By contrast, other types of investment products, such as those tailored to climate transitions, provide ample opportunities for investors to rebalance portfolios away from companies with carbon-intensive outputs or supply chains. However, highly tailored low-carbon or carbon-transition portfolios may have asset compositions and risk characteristics that stray widely from standard market benchmarks that are most commonly used by institutional investors.

Notably, in some cases, the weight of high polluting industries, such as the energy sector, increased in the ESG index (with little change in the constituent companies). For example, this is the case for the S&P 500 ESG and STOXX 600 ESG indices. Focusing on S&P 500 and its ESG counterpart (S&P 500 ESG) there is little difference in terms of the largest holdings and industry representation. A more detailed comparison of the STOXX 600 index and the STOXX ESG-X index, shows that the ESG focused index does little to improve the sustainability of the parent index (see Table 2.3). The integration methodology appears to remove 20 companies (from the original 600) with little difference in the characteristics of constituents in terms of sector and ESG criteria. For example, in the energy sector, Total, BP and Royal Dutch Shell remain the largest companies for both the STOXX 600 and STOXX ESG-X index. Despite the overall level of emissions and perceived impact on the environment, these companies exhibit high E pillar scores for some rating providers, at or above 90 out of 100 for Refinitiv and between 77 to 92 out of 100 for Bloomberg. This is in part due to transition policies enacted by companies and targets for lowering the carbon footprint, which often make up a large part of the E score. In contrast, MSCI World funds and ESG counterpart indices show clear differences in the constituents included (see Table 2.4),24 with the MSCI ESG leader fund, for example, including companies based on their MSCI ESG ratings, through a best-in-class approach.

Understanding what impact differences in the composition of parent indices and funds and their ESG counterparts have on the environmental performance of these indices and funds is important. Focusing on CO2 emissions, and taking into consideration a weighting for market capitalisation when relevant, this section compares aggregate CO2 emissions between parent and ESG indices and funds, as well as analysing tilting methodology (that takes into consideration market capitalisation weighting before and after ESG integration).

CO2 emissions for the S&P ESG index appear lower than the S&P 500 on face value, yet when tilting is applied by sector, CO2 emissions in some sectors appear higher in the S&P ESG compared to the parent S&P 500. This could be due to exclusion of low polluters (relative to the index average) from some sectors, resulting in an increase in the weighting of the sector in the index (see Table 2.5).

In parallel, a number of alternative investment products tailored to low carbon transition have emerged, and may bring benefits to investors who wish to invest in instruments with a focus on environmental outcomes or low-carbon activities. These include green bonds (see Box 2.4) and specialised climate funds, such as PIMCO Climate Bond fund, the Blackrock iShares Global Green Bond ETF, and the Invesco WilderHill Clean Energy ETF. These funds focus on investing in green bonds as well as bonds from issuers showing innovative approaches to environmental sustainability. The bonds in which the investments are made are certified to be green by third parties or by institutional investors internally, thereby providing greater assurance that the investment mechanisms are supportive of the transition to low-carbon economies. 25 In line with this, PIMCO and iShares funds invest in bonds of companies and countries specifically issued to back green projects, while the Invesco fund invests in equities that are involved in renewable and clean energy. These may increase with the implementation of the European Commission’s climate benchmarks in 2019.

As the portfolio composition would deviate more from the traditional indices, the returns and variance might differ from a standard portfolio, and therefore may not align with existing investment strategies of portfolio managers. However, for those that have flexibility of portfolio composition within their strategies, existing metrics and investment products allow investors to construct portfolios that appear to align with transitions to low-carbon economies. These types of investments might, for example, be benchmarked to specialised climate funds, such as the Bloomberg Barclays MSCI Green Bond Index which tracks the global market for fixed income instruments that are certifiably invested in green projects.

In sum, ESG approaches provide benefits in unlocking a significant amount of information about corporate practices that affect the environment. Nevertheless investors should not draw generalised conclusions that the E pillar is synonymous with corporate behaviours that are better for the environment. In this regard more work is needed to assess the consistency of metrics, their significance and how they are integrated along with other metrics related to environmental risks and opportunities to derive the E pillar score.

This chapter acknowledges that the E pillar has been instrumental in combining a wide range of information that draws new attention to environmental factors. E scores include valuable information on outputs such as emissions and waste; climate change scenarios and risk management; and strategies to transition to renewable energy. This information is of growing importance as a number of public sector authorities, from central banks to financial regulators, contemplate how to incorporate climate risk and environmental resilience into their policymaking and activities. Furthermore, climate-related disclosures are of critical importance to institutional investors trying to meet the demand by asset owners for products that align with sustainability in terms of risk-adjusted returns and/or with the climate transition.

In addition to the health and economic crisis caused by COVID-19, the pandemic has also brought to light the materiality of ESG-related risks and the deep linkages between businesses and their stakeholders across the value chain. For investors, ESG approaches can help unlock a significant amount of information about the management of and preparedness for sustainability risks, including environmental and climate risks, when pursuing long-term value. However, fragmented ESG ratings do not deliver the transparent, accurate and comparable data investors need to manage climate risks. For policymakers, ESG investing and ratings are seen as important tools to green the financial sector. They represent important market based mechanisms to help investors make decisions based on long-term carbon prices and transition risks implied by the Paris Agreement. Clearly governments cannot depend on E scores in their current form to align financial and corporate value with carbon targets in the current market environment.

In an economic recovery context, ESG risk management could play a strong role by ensuring that material sustainability risks are managed and properly priced, so that business do not sacrifice long-term value even as they struggle to meet immediate challenges. This is particularly important for countries committed to a green recovery, which if implemented accurately, could allow private finance to properly price and efficiently allocate capital to green projects. Looking forward, the pandemic has underlined the importance of managing financial and non-financial risks, including ESG risks, throughout company operations and supply chains. Addressing these risks is an important element to long-term resilience against future shocks, both in financial markets and the economy as a whole (see Box 2.5).

As an increasing number of investors look to invest in environment and climate transition products, a more standardised and comparable approach across E rating providers may support more sustainable capital realignment away from carbon intensive economic activities. This chapter highlights where there is a lack of consistency across rating providers and identifies areas that may give rise to constraints. Also, further consideration may be given to the extent to which corporate reporting frameworks consider financial materiality and separately environmental materiality in the environmental pillar and the extent to which it is aligned with the carbon transition. Greater clarity from corporate reporting frameworks on their approach as to what constitutes material information in the context of environment would also be beneficial.

Today, the lack of comparability between E pillar ratings illustrates a fragmentation that may limit the benefits of E scores as a tool to assess environmental impact. Similarly, opacity in the measurement of longer-term management and strategy metrics limit the E pillar’s use as a tool to assess companies’ conformity to the carbon transition, which in turn, raises concerns as to the extent to which financial market participants can rely on E scores as a key tool to manage environmental and carbon transition risks. Therefore, for the E pillar to be most useful to investors with differing motivations, methodologies to generate E pillar scores will need to contain metrics that measure financial materiality and distinct aspects of environmental materiality in a mutually exclusive and transparent manner, where possible, so that investors have no doubt as to what is driving the E pillar score.

In line with this, the following considerations should be addressed to improve the relevance of the E pillar:

  • Greater consistency on the high-level purpose of the E pillar could provide stakeholders and investors with a framework to assess the level of environmental and financial materiality for aspects of the E pillar. This would help untangle the various metrics and information that serve disparate stakeholders, which may undermine the value of underlying metrics and information. This should also include clear boundaries as to which areas of the E pillar are relevant to greening the financial system, and which focus on other objectives, such as long-term financial value.

  • Improved transparency on the methodologies used to generate E scores and sustainable indices is vital to make the E pillar fit for purpose. Ultimately, this should include clear and publicly available information on metrics with guidance on the measurement of supplementary analysis. This should go hand in hand with greater clarity on required E pillar reporting by companies; establishing core metrics with associated metadata for measurement would be a valuable first step.

  • Stronger coordination is required to define the role that the E pillar should play in helping financial market participants navigate the low carbon transition. Notably, the E pillar may have an important role to play as an informational vector supporting the assessment of companies’ exposure to physical and transition risks, as well as support broader government policies to meet climate mitigation targets. The extent to which these two approaches support each other can be instrumental in creating clarity for companies, investors and in turn markets.

ESG approaches have the potential to unlock a significant amount of information on the management and resilience of companies in line with sustainability, including physical climate risks and other environmental risks when pursuing long-term value creation. It could also represent an important market based mechanism to help investors make decisions based on long term carbon prices and climate transition risks implied by climate mitigation policies in response to the Paris Agreement. Looking ahead, more needs to be done to ensure that the E pillar provides investors with the insight needed to help both long-term value and a greening of the financial system. Irrespective of investor needs, steps to improve transparency and market integrity will ensure markets are efficient and resilient to best contribute to a transition and deliver long-term value. This is ever more pressing in the current COVID-19 environment, in which governments are channeling a significant amount of resources to investments that will inevitably impact the economic path of the recovery.


BIS (2019), Green bonds: the reserve management perspective, Bank of International Settlements Quarterly Review, https://www.bis.org/publ/qtrpdf/r_qt1909f.htm

Bloomberg (2020), ESG Stock Resilience Is Paving the Way for a Surge in Popularity, https://www.bloomberg.com/news/articles/2020-03-31/esg-stock-resilance-is-paving-the-way-for-a-surge-in-popularity

Climate Bond Initiatives (2019), Greening the financial system Tilting the playing field The role of central banks, https://www.climatebonds.net/files/reports/cbi-greening-the-financial-sytem-20191016.pdf

ECB (2020), Eurosystem reply to the European Commission’s public consultations on the Renewed Sustainable Finance Strategy and the revision of the Non-Financial Reporting Directive, https://www.ecb.europa.eu/pub/pdf/other/ecb.eurosystemreplyeuropeancommissionpubliconsultations_20200608~cf01a984aa.en.pdf

Greenhouse Gas Protocol (2020), Technical Guidance on Scope 1, 2 and 3 Emissions, https://ghgprotocol.org/scope_2_guidance

GSIA (2019), Global Sustainable Investment Review 2018, Global Sustainable Investment Alliance, http://www.gsi-alliance.org/wp-content/uploads/2019/03/GSIR_Review2018.3.28.pdf

IMF (2009), Global Financial Stability Report: Responding to the Financial Crisis and Measuring Systemic Risks, https://www.imf.org/en/Publications/GFSR/Issues/2016/12/31/Global-Financial-Stability-Report-April-2009-Responding-to-the-Financial-Crisis-and-22583

IOSCO (2019), Final report, Sustainable finance in emerging markets and the role of securities regulators, www.iosco.org/news/pdf/IOSCONEWS534.pdf

I4CE (2018), Carbon pricing across the world: how to efficiently spend growing revenues?, Climate Brief N°55, https://www.i4ce.org/wp-core/wp-content/uploads/2018/10/20181106_PC55_Carbon-Revenues_vENG.pdf

I4CE (2019), Global Carbon Account 2019, https://www.i4ce.org/wp-core/wp-content/uploads/2019/05/i4ce-PrixCarbon-VA.pdf

NGFS (2019a), A call for action Climate change as a source of financial risk, Network for Greening the Financial System First comprehensive report, https://www.banque-france.fr/sites/default/files/media/2019/04/17/ngfs_first_comprehensive_report_-_17042019_0.pdf

NGFS (2019b), A sustainable and responsible investment guide for central banks’ portfolio management, https://www.ngfs.net/sites/default/files/medias/documents/ngfs-a-sustainable-and-responsible-investment-guide.pdf

OECD (2020a), Sovereign Borrowing Outlook for OECD Countries, https://www.oecd.org/finance/Sovereign-Borrowing-Outlook-in-OECD-Countries-2020.pdf

OECD (2020b), ESG Investing: E Scoring and Reporting, www.oecd.org/finance/esg-investing-e-scoring-and-reporting.pdf.

OECD (2020c), ESG Investing: Practices, Progress and Challenges, www.oecd.org/finance/esg-investing-practices-progress-and-challenges.pdf.

OECD (2019a), Government at a Glance 2019, OECD Publishing, Paris, https://doi.org/10.1787/8ccf5c38-en.

OECD (2019b), Sovereign Borrowing Outlook for OECD Countries, http://www.oecd.org/finance/Sovereign-Borrowing-Outlook-in-OECD-Countries-2019.pdf

OECD (2019c), Taxing Energy Use 2019: Using Taxes for Climate Action, OECD Publishing, Paris, https://doi.org/10.1787/058ca239-en.

OECD (2018), Effective Carbon Rates 2018: Pricing Carbon Emissions Through Taxes and Emissions Trading, OECD Publishing, Paris. https://doi.org/10.1787/9789264305304-en

OECD (2017), Investing in Climate, Investing in Growth, OECD Publishing, Paris. http://dx.doi.org/10.1787/9789264273528-en.

OECD (2015), Aligning Policies for a Low-carbon Economy, OECD Publishing, Paris, https://doi.org/10.1787/9789264233294-en.

Refinitiv (2019), ESG Carbon Data & Estimate Models, https://www.refinitiv.com/content/dam/marketing/en_us/documents/fact-sheets/esg-carbon-data-estimate-models-fact-sheet.pdf.

S&P (2020), The ESG Lens on COVID-19, Part 1, https://www.spglobal.com/ratings/en/research/ articles/200420-the-esg-lens-on-covid-19-part-1-11444298.

S&P (2019), S&P 500 ESG Index, https://us.spindices.com/indices/equity/sp-500-esg-index-usd.

STOXX (2019), STOXX Europe 600 ESG-X https://www.stoxx.com/index-details?symbol=SXXPESGX.

TCFD (2017a), Recommendations of the Task Force on Climate-related Financial Disclosures, https://www.fsb-tcfd.org/wp-content/uploads/2017/06/FINAL-2017-TCFD-Report-11052018.pdf.

TCFD (2017b), Implementing the Recommendations of the Task Force on Climate-related Financial Disclosures, https://www.fsb-tcfd.org/wp-content/uploads/2017/12/FINAL-TCFD-Annex-Amended-121517.pdf.

World Bank (2019), State and Trends of Carbon Pricing, http://documents.worldbank.org/curated/en/191801559846379845/pdf/State-and-Trends-of-Carbon-Pricing-2019.pdf.


← 1. Climate related risks include either physical risks, which occur from exposures to more frequent and more devastating disasters caused by natural hazards, and; transition risks that arise from uncertainties surrounding the timing and speed of the transition to a low-carbon economy.

← 2. A low-carbon economy is an economic output based on low carbon power sources that therefore has a minimal output of greenhouse gas emissions, with specific reference to carbon dioxide. More information on policy alignment needed to achieve the transition to a low carbon economy can be found in OECD (2015).

← 3. The Paris Agreement is the first universal climate agreement. It was adopted on 12 December 2015 at the 21st Conference of the Parties (COP21) of the United Nations Framework Convention on Climate Change (UNFCCC). The Agreement lays out the foundation for meaningful long-term action, providing transparency and review mechanisms that should allow countries to assess and adjust the scale of their efforts. In doing this, it aims to strengthen the global response to climate change over the course of the 21st century by: i) “holding the increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognising that this would significantly reduce the risks and impacts of climate change”; ii) “increasing the ability to adapt to the adverse impacts of climate change and foster climate resilience and low greenhouse gas emissions development, in a manner that does not threaten food production”; and iii) “making finance flows consistent with a pathway towards low greenhouse gas emissions and climate-resilient development” (Article 1, UNFCCC, 2015).

← 4. There is no single definition of environmental materiality. For the purpose of this paper, a working definition, adapted from Global Reporting Initiative, is: those topics that have a direct or indirect impact on an organisation’s ability to create, preserve or erode environmental value for itself, its stakeholders and society at large. (see GRI G4 Guidelines).

← 5. Materiality as a concept defines why and how certain issues are relevant for a company or a business sector, this can refer to issues that are financially material (i.e. the approach taken by SASB) or issues that are material from the perspective of a range of stakeholders (i.e. the approach of GRI), for example by representing a material risk or importance for the business model of a company. Materiality is applied in a wide variety of contexts, including in accounting, reporting, business model, financial, legal, risk and more recently, environmental, social, and governance issues. In this context, there appears to be two major perspectives on materiality. The first is with respect to financial materiality, centering on the impact the environment and climate change – both physical and transition risks – could have on a company. This provides a basis for financial materiality in the context of risk-adjusted returns for a company. The other is a stakeholder-oriented concept of materiality, which emphasises the impact a company has on the wider environment and society.

← 6. In addition, OECD recommendations on due diligence for responsible business conduct include guidance to help integrate risks from transitions into operations and business activities.

← 7. In 2017, nine central banks and supervisors established the Network of Central Banks and Supervisors for Greening the Financial System (NGFS). Driving this, was a view that “climate-related risks are a source of financial risk, [and are] therefore within the mandates of central banks and supervisors to ensure the financial system is resilient to these risks.” (NGFS, 2019). Since 2017, the network has grown to 69 international Members and 15 Observers as of 24 July 2020. The NGFS aims to accelerate the work of central banks and supervisors on climate and environmental risk and on scaling up green finance.

← 8. For example, the number of extreme temperature events has more than tripled since the 1980s (Munich Re, 2018).

← 9. For example, financial regulators, through the International Organisation of Securities Commissions (IOSCO), have established a sustainable finance network. The objective of this work is to introduce and improve standards for ESG disclosures recommended by regulators (IOSCO, 2019).

← 10. The OECD estimates that the carbon pricing gap was 76.5% in 2018, representing EUR 30 per tonne of CO2. By the current rate of gap narrowing, carbon prices will only meet real costs in 2095 (OECD, 2018).

← 11. See the Taskforce for Climate-related Financial Disclosure (2017), Final Report: Recommendations on Climate-related Financial Disclosures.

← 12. One example is the Climate 100+ initiative, which is an investor initiative launched in 2017. More than 450 investors have joined the initiative, with over USD 40 trillion in assets collectively under management (AUM), with the aim to engage companies to: curb emissions, improve governance and strengthen climate-related financial disclosures.

← 13. Environmental, social and governance (ESG) investment refers to a set of standards for a company’s operations that investors use to screen potential investments. ESG investing is therefore not a clearly defined investment strategy, but a broad criteria that can be applied to any kind of investment strategy that takes into account issues within the E, S or G. Please refer to Chapters 1 and 4 for more information.

← 14. In line with the Task Force on Climate-related Financial Disclosures’ (TCFD’s) scope 1 (direct emissions), 2 (indirect emissions from direct production).

← 15. Due to lack of available data for scope 3 emissions, scope 1 and 2 emissions have been used for the analysis.

← 16. Water withdrawals refer to all freshwater usage, excluding evaporation losses from storage basins. This can include water used for agricultural irrigation or industrial use such as through cooling for thermoelectric plants.

← 17. Number of metrics estimated using publicly available information provided by ESG rating providers.

← 18. Bloomberg (2019), MSCI (2019) and Thomson Reuters (2017) have been used to as the publically available source of the frameworks and categories used by rating providers. Individual metrics as sourced in the annex, and referred to throughout this chapter have been shared with the Secretariat by the rating provider (not for publication), or downloaded from the Bloomberg Terminal and Thomson Reuters Refinitiv platforms.

← 19. S&P 500 ESG and STOXX 600 ESG-X.

← 20. MSCI World ESG Leaders and MSCI World ESG Screened.

← 21. Exclusionary screening in ESG occurs when specific assets are screened out of an index or portfolio for value reasons, for example excluding any company from the coal or oil production industry (regardless of their ESG score), resulting in divestment.

← 22. Tilting occurs when an index or portfolio provider uses data from ESG rating agencies to rank the ESG scores of the companies within their parent product, to create an ESG counterpart. For example, by tilting the portfolio towards high ESG firms by excluding those with the ESG scores below a certain threshold.

← 23. Industries are grouped by economic sector using Refinitiv data. Economic sectors have been selected according to the most polluting industries using International Energy Organisation CO2 emissions data.

← 24. In terms of number of companies included, the MSCI World fund includes 1 207 companies, with the MSCI ESG screened includes 1 540 companies, and the MSCI ESG leaders includes 847.

← 25. In this regard, the certification of the projects does not necessarily mean that benefits like CO2 emission reductions have been quantified.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.