Investing based on environmental, social and governance (ESG) factors has gained traction in recent years, reflecting the interest of many investors in deploying their capital to positively influence corporate practices and drive sustainable change.
At Neuberger Berman, we believe ESG investing requires a nuanced understanding of company practices, combined with expertise in discerning responsible corporate behavior in the context of financial and industry considerations. As long-term-oriented asset managers, we believe that ESG should be executed actively, through an in-depth understanding of businesses, triangulation of information from multiple sources and thoughtful forward-looking engagement with company managements.
In our view, it is challenging to implement ESG investing through common mechanical and rules-based approaches based solely on third-party inputs or ESG indices. While a useful aggregation tool and/or starting point, third-party ESG ratings have significant shortcomings, in our view, and may give investors a suboptimal and often imprecise outcome relative to their goals. In this Insights paper, we explain why.
ESG Gains Momentum With Passive Leading the Charge
Over the last several years, the pace of new ESG fund launches has accelerated, along with assets under management (AUM). However, U.S. passive ESG fund AUM has grown at nearly four times the rate of active sustainable fund AUM. Passive now represents roughly 40% of total sustainable fund AUM, and passive sustainable funds saw record inflows in 2020 (+$37 billion), driven by Large Blend (+$17 billion)1.
Among passive ESG funds, ETFs seem to be where assets are moving. Let’s consider the top 10 ESG ETFs with the most assets under management.
Top-10 ESG ETFs and Their Characteristics
|Name||Fund Size ($ Bn)||Ticker||Rebalance frequency||Reference Benchmark||T/O rate||# Holdings||Expense ratio||Third party data source|
|iShares ESG Aware MSCI USA ETF||22.00||ESGU||Quarterly||MSCI USA Extended ESG Focus Index||38%||331||0.15%||MSCI|
|iShares MSCI USA SRI ETF USD Acc||8.33||SUAS||Annual||MSCI USA SRI Select Reduced Fossil Fuel Index||N/A||132||0.20%||MSCI|
|Vanguard ESG US Stock ETF||5.29||ESGV||Quarterly||FTSE US All Cap Choice Index||1.9%||1540||0.12%||FTSE|
|iShares MSCI USA ESG Scrn ETF USD Acc||4.32||SASU||Quarterly||MSCI USA ESG Screened Index||N/A||588||0.18||MSCI|
|Amundi IS Amundi MSCI USA SRI ETF DR||4.23||USRI||Quarterly||MSCI USA SRI filtered Ex Fossil Fuel Index||Very low||134||0.18%||MSCI|
|iShares ESG MSCI USA Leaders ETF||3.93||SUSL||N/A||MSCI USA Extended ESG Leaders Index||9%||275||0.10%||MSCI|
|iShares MSCI USA ESG Enh ETF USD Dist||3.84||OM3L||Quarterly||MSCI USA ESG Enhanced Focus Index||Very low||585||0.07%||MSCI|
|Xtrackers MSCI USA ESG Leaders Eq ETF||3.76||USSG||Tri-annual||MSCI USA ESG Leaders Index||Very low||277||0.10%||MSCI|
|iShares MSCI USA ESG Select ETF||3.74||SUSA||Quarterly||MSCI USA Extended ESG Select Index||27%||194||0.25%||MSCI|
|Xtrackers MSCI USA ESG ETF 1C||3.64||XZMU||Quarterly||MSCI USA Low Carbon SRI Leaders NTR Index - USD||Very low||217||0.15%||MSCI|
Source: FactSet, October 2021.
Based on the table above, we draw some key observations:
- Nine out of the top 10 ESG ETFs listed are based on a single data provider, MSCI.
- Six out of the top 10 ESG ETFs are sold by the same provider, BlackRock.
- The average holdings overlap across these ETFs is 95%.2
- The number of holdings ranges from 132 to 1,540.
In addition, all of these ETFs are rules-based passive funds that are constructed from reference ESG benchmarks offered by third-party ESG data providers. Products relying on third-party ESG data are removed from the underlying assessment of risks and opportunities, the source of analysis, criteria and/or assumptions made in giving a company a certain score. Along these lines, passive ESG strategies are farthest removed, given the sheer scale of holdings as well as a lack of in-depth understanding of the business fundamentals. This is also reflected in the number of companies that are owned by passive ESG strategies, which leads to the question of where the line is drawn with respect to being truly sustainable.
ESG benchmarks seek to deliver returns similar to their parent benchmark, while keeping within a tracking error constraint. This is done by tilting away from companies with lower ESG scores and toward companies with higher ESG scores. An optimizer is then run to figure out the optimal weights of holdings that would deliver results closest to the parent benchmark. In other words, given the nature of their goal, ESG benchmarks and passive ESG products do not allow for the degree of freedom that actively managed strategies may have in distinguishing leaders from laggards and selectively investing in companies that align with intended ESG outcomes or excluding those that don’t. As such, ESG benchmarks largely represent universal ownership rather than best-in-class ESG investing.
For example, the MSCI USA ESG Extended Focus Index targets a tracking error of 0.5% and holds most of the top names as MSCI USA. MSCI USA Extended ESG Select follows a similar construction process to the Focus Index, but with a slightly higher tracking error target of 1.8%. This causes the index to stray further than the standard MSCI USA names toward higher-scored ESG names, although the resulting index does maintain most of the names in the MSCI USA.
Some ESG benchmarks instead start from a universe of highly rated ESG names, and then they seek to construct a benchmark with similar sector weights as the cap-weighted equivalent. The most popular of these is the MSCI USA ESG Leaders Index, which ends up with similar tracking error to the ESG Select Index but only holds highly rated ESG names. This index also excludes most companies in controversial business activities such as alcohol, gambling, tobacco, nuclear power, conventional weapons, nuclear weapons, controversial weapons and firearms.
The constraints that are inherent to a passive ESG product can take the investor further away from their intended outcomes while giving them the seeming comfort of having an ESG-labeled product. Furthermore, the construction of passive ESG products, including ESG benchmarks and ESG ETFs linked to them, are further exacerbated by the flaws in the underlying data, which is discussed next.
The Data Conundrum
In our view, the quality of data (or lack thereof) that passive ESG products rely on makes them problematic. Here’s why.
The vast majority of ESG data today is unaudited. It is largely based on self-disclosure by companies, which is nonstandard and ranges in level of transparency from company to company. While there have been some attempts in the industry to even the playing field, such as the CDP (formerly the Carbon Disclosure Project) and Sustainability Accounting Standards Board (SASB) (which take very different approaches), disclosure is generally far from standardized. For example, only about 25% of U.S.-listed companies disclose all the requested material ESG data,3 and disclosure rates are even lower in small-cap equities and emerging markets equities.
In addition to its lack of standardization, the disclosure also tends to be backward-looking. Companies voluntarily respond to questionnaires from the data providers, typically once a year, using prior-financial-year data. As such, at any point in time, the resulting ESG ratings based on disclosures today could have a 12- to 18-month lag. As active managers, engaging directly with company managements, we come across situations where the ratings and their corresponding explanations may not reflect more recent changes and adjustments in corporate practices.
ESG data providers have their own methodologies to take in all the information disclosed by companies and fill in the missing gaps. This process is typically based on top-down assumptions, by sector and then peer group. These approaches can result in unintended consequences by sometimes missing out on material nuances that are specific to a company’s business model, regional exposure and business mix (examples in the table below). Bottom-up fundamentals-based approaches can allow for a deep-dive, nuanced analysis of material ESG issues.
Sample Third-Party Ratings and Underlying Concerns
|Issuer||Third-party Rating||Third-party Rating Issue|
|Paint company||BBB||Top-down assumptions and stale information: Did not take business mix differences into account in peer relative evaluation of company, in addition to focusing on a 100-year-old issue that is no longer relevant to current mix.|
|Regional bank||BB||Stale information: Time lag (~6 months) in updating governance assessment based on changes made by company.|
|E-commerce company||BB||Generic assumptions: Regarding reduced employee morale simply based on acquisitions by the company.|
|Auto parts company||BB||Stale information: Changes made by new company management not considered in assessment.|
|Energy company||BB||Lack of disclosure: By company, resulting in lower ratings by third party around concerns related to environmental liabilities.|
|Airline company||BB||Generic assumptions: Concerns around labor management issues suggesting operational risks while engagement with company suggested otherwise.|
|Med tech company||BB||Generic assumptions: Regarding product liabilities concerns relative to other healthcare peers.|
|Industrials company||BBB||Missed information: Concerns around lack of risk management experience in audit committee because of scraping bios, instead of a detailed review of skills and experience matrix provided by company in its proxy report.|
|Industrials company||BBB||Lack of disclosure: Relatively new public company with limited disclosure on sustainability issues, which doesn’t necessarily imply poor business practices.|
|E-commerce company||BBB||Generic assumptions: Concerns around risks to privacy and data security even though company had not had any breaches and not collecting any sensitive customer information.|
Source: Neuberger Berman. Represent examples of issues identified in the process of our company research and due diligence.
The difference in methodologies discussed above results in very different outcomes, as evidenced by the magnitude of difference between the minimum and maximum pairwise correlations of six data providers, according to an academic study published on PRI.i Unlike credit ratings, where there is a high level of correlation between ratings providers based on the credit risk underlying a fixed income security, ESG ratings do not offer such confidence on the underlying assessments of material ESG issues.
Average, minimum and maximum correlations across providers
Source: PRI Academic Blog: ESG rating disagreement and stock returns, March 2020.
More recently, leading sustainability standard-setting organizations have moved to integrate the disparate sustainability reporting frameworks into a comprehensive reporting system that would include both financial and sustainability disclosure.4 In addition, evolving data science capabilities offer much progress and opportunity for gleaning additional ESG insights in investment research. However, the evolving frameworks and the enhanced tools to parse ESG data still require nuanced human judgment and deep domain expertise to thoughtfully analyze ESG issues in the context of bottom-up business analysis.
Active Engagement Levers
Another major challenge with passive ESG product is a lack of “skin in the game.” Because passive products deliver universal ownership, engagement efforts often hinge on leverage, specifically, passive shops seek to follow a top-down thematic engagement approach, citing the power of the scale of their assets. However, some say these efforts have no “teeth.”
Passive managers tend to have centralized resources focused on thematic engagement, often generalists that are focused on thematic issues and removed from company-specific nuances. Even with sector specialists, given the scale of holdings, it is practically impossible to have resources dedicated to understanding company-specific nuances. As such, engagement by passive managers can best be described as “a mile wide and an inch deep.”
In contrast, active managers often not only have the skin in the game, but can conduct engage in deep shareholder engagement while occupying a unique position to influence corporate practices in the context of long-term business success; analysts who cover a focused group of companies are more familiar with the company’s value drivers. As such, active managers can target analysis and assessment of ESG issues that are financially material, and thus may be better equipped to make a business case in the context of broader financial and industry considerations, that is more likely to resonate with management than simply a top-down thematic “push.”
With the increased momentum in asset management to commit to net-zero emissions by 2050, we believe engagement is a key lever for investors to help enable this transition. Active managers are likely to play a prominent role in leading nuanced and company-specific engagements with businesses, for all the reasons noted above.
We firmly believe that the most effective approaches to ESG investing are active in nature. In our view, ESG investing requires nuanced domain expertise to discern material issues and the ability to influence corporate behavior, as well as enhanced disclosure related to them. An active investing approach that combines human judgment with shareholder engagement has the potential to deliver better ESG outcomes than passive approaches.
Engagement at Neuberger Berman
The Neuberger Berman research department and portfolio management teams host on average more than 2,000 one-on-one meetings per year with company management teams in-person at our offices and via conference calls, in addition to a similar number of outside meetings and on-site company visits. These meetings provide an opportunity to communicate views and concerns directly to company managements. This engagement is often supplemented with formal written communication with management teams and boards of directors on identified areas of concern and recommended courses of action. Portfolio management teams may also seek governance change through shareholder proposals, proxy contests and other measures of shareholder activism if a company’s responsiveness is deemed inadequate. Another important way in which we exercise engagement is through voting proxies on behalf of our advisory clients for whom we have voting authority. We do this to fulfill our fiduciary responsibility to protect our clients’ best interests and as an important component of our approach to creating shareholder value. In addition, to provide transparency to other investors and clients, we pre-disclose our stance on proxy votes ahead of the annual meeting, as part of our NB Votes initiative.