Do ESG Funds Deliver on Their Promises?

Quinn Curtis* Jill Fisch** Adriana Z. Robertson† December, 2021

Corporations have received growing criticism for contributing to climate change, perpetuating racial and gender inequality, and failing to address other pressing social issues. In response to these concerns, shareholders are increasingly focusing on environmental, social, and corporate governance (ESG) criteria in selecting investments, and asset managers are responding by offering a growing number of ESG mutual funds. The flow of assets into ESG is one of the most dramatic trends in asset management.

But are these funds giving investors what they promise? This question has attracted the attention of regulators, with the Department of Labor and the Securities and Exchange Commission (SEC) both taking steps to rein in ESG funds. The change in administration has created an opportunity to rethink these steps, but the rapid growth and evolution of the market mean regulators are acting without a clear picture of ESG investing.

We fill this gap by offering the most complete empirical overview of ESG mutual funds to date. Combining comprehensive data on mutual funds with proprietary data from the several of the most significant ESG ratings firms, we provide a unique picture of the current ESG environment with an eye to informing regulatory policy. We evaluate a number of criticisms of ESG funds made by academics and policymakers and find them lacking. We find that ESG funds offer their investors increased ESG exposure. They also vote their shares differently from non-ESG funds and are more supportive of ESG principles. Our analysis shows that they do so without increasing costs or reducing returns.

We conclude that ESG funds generally offer investors a differentiated and competitive investment product that is consistent with their labeling. In short, we see no reason to single out ESG funds for special regulation.

Introduction

ESG investing—that is, investing informed by environmental, social, and governance criteria or considerations—is growing explosively.¹ Public attention in the United States and globally has increasingly focused on ESG issues,² and a growing percentage of investors consider green investing “a big priority.”³ In one of his first official acts, President Biden rejoined the Paris Agreement,⁴ and the Securities and Exchange Commission (SEC) has, for the first time, a designated policy advisor to advance ESG issues.⁵

The growing focus on ESG investing is reflected by the rapidly expanding number of mutual funds that purport to consider ESG factors in their investment and voting decisions, as well as a surge in the volume of assets invested in such funds. Morningstar reports that both the number of ESG-focused index funds and the total amount of assets held by such funds have doubled in the past three years.⁶ The COVID-19 pandemic and the disruptions it has caused to financial markets have done nothing to slow this rise.⁷

But do these rapidly growing ESG funds deliver what they promise? Do ESG funds offer portfolios with real investment exposure to ESG goals or has the demand for ESG investing led to overpriced, greenwashed funds that are merely marketed as ESG to chase the latest investment fad?⁸ The answers to these questions have legal implications because mutual funds are extensively regulated by the SEC and the inclusion of mutual funds in retirement plans is regulated by the Department of Labor (DOL) under the Employee Retirement Income Security Act of 1974 (ERISA)⁹. In fact, both the SEC and the DOL have recently turned their attention to ESG investing.¹⁰ For the SEC, the central legal question is whether funds that characterize themselves as focused on ESG deliver on that promise—do they invest and vote differently from other mutual funds?¹¹ For the DOL, the question is whether ESG investing is consistent with the fiduciary duties of retirement plan trustees—do ESG funds deliver sound performance at reasonable cost or do they sacrifice returns to promote social causes?¹² Despite these differing concerns, both the SEC and the DOL view the growth of ESG funds as potentially warranting regulatory intervention. Indeed, the DOL has already intervened, adopting a new rule on November 13, 2020, that may deter 401(k) plans from offering ESG funds.¹³ Although the SEC has not engaged in rulemaking to date, members of the Commission have expressed concerns that asset managers’ current disclosure practices with respect to ESG products are insufficient.¹⁴

Other interventions are on the horizon as well. Asset managers’ reliance on third parties, including index providers and rating agencies, in evaluating the ESG characteristics of portfolio companies has led some to call for greater regulation of those providers.¹⁵ The Biden administration is taking steps to review and potentially replace the DOL rule,¹⁶ and new leadership at the SEC will likely look to expand corporate disclosures to address ESG issues.¹⁷

These changes are taking place amid a rapidly evolving ESG landscape that has outpaced the academic literature. Instead, regulators are acting based on a variety of assumptions about how ESG funds operate,¹⁸ often drawn from small-sample studies or anecdotal reports.¹⁹ Even as regulators move, we know relatively little about the market for ESG funds, the investment strategies these funds use, how the funds vote their proxies, and what the funds cost. At the same time, regulatory attention has focused on ESG funds as presenting concerns distinctive from other mutual funds. But it is unclear that ESG funds, as a category, present unique regulatory issues.²⁰ These are all relevant policy questions that should inform rulemaking.

This Article offers the most complete empirical overview of ESG mutual funds to date. Using market-wide data on fund portfolios, voting, fees, and performance, we specifically target the concerns articulated by the SEC and the DOL. We combine detailed information on mutual funds with four proprietary datasets evaluating company-level ESG performance. Using this unique and comprehensive dataset, we explore the practical differences between ESG and non-ESG funds as well as the differences among ESG funds along four dimensions—portfolio composition, voting behavior, costs, and performance. The first two specifically target the SEC’s concerns, while the latter two relate to those raised by the DOL. Our goal is to provide an overview of the market as it currently stands for the purpose of informing a regulatory push that has the potential to reshape the ESG landscape.

From the SEC’s perspective, the fundamental regulatory question is what investors are getting for their “ESG dollars.” We first confront the question of what ESG funds promise—the information conveyed both by the ESG label and fund disclosure practices. We then ask whether and how these funds deliver on that promise. To answer these questions, we survey the existing market and construct several categories of ESG mutual funds—funds with names that convey an ESG-oriented strategy, funds classified by Morningstar as ESG funds, and funds that purport to consider ESG factors in their investment criteria. We then analyze the portfolio composition and voting behavior of these funds to compare them across multiple dimensions. From the DOL’s perspective, the primary concerns are pecuniary: What, if anything are investors giving up when they invest in ESG funds? These pecuniary costs can be direct (in the form of fees) or indirect (in the form of lower raw or risk-adjusted returns). We engage with the concerns of both regulators by providing evidence about what investors are getting and what they are giving up to get it.

Descriptively, we uncover an evolving landscape of ESG funds. Today’s ESG funds range from single-issue funds that address water conservation or religious values to those that incorporate screening criteria into the construction of a broad-based index. We identify extensive disclosures of fund investment strategies—strategies that differ substantially—as well as the extent to which the fund incorporates ESG considerations into voting and engagement. We find, in short, a market that recognizes that ESG means different things to different investors.

Empirically, we demonstrate that ESG funds behave differently from other funds. We first evaluate portfolio composition. Using data from four separate rating providers, we calculate what we term a fund’s “ESG tilt”—the asset-weighted average of the ESG scores of the fund’s portfolio companies. Funds that identify themselves as ESG funds hold portfolios that represent a significant ESG tilt. In other words, contrary to the SEC’s concern about “greenwashing,” ESG funds deliver on their promise to invest differently from other funds, and their holdings are rated more highly with respect to ESG. Because we incorporate ratings from four different providers, our findings offer reassurance that funds are not “gaming” a specific ESG index.

Second, we examine fund voting behavior. Although ESG mutual funds have been criticized for not casting their portfolio-company votes in accordance with their investment profiles,²¹ we document clear differences between the voting behavior of ESG and non-ESG funds. ESG funds do not automatically support every shareholder proposal related to ESG,²² but they do vote more independently of management compared to other funds when it comes to environmental and social issues. With respect to certain governance issues, such as say on pay, we also find clear differences. In short, ESG funds appear to be considering ESG criteria in voting as well as investment decisions.

Third, we look at what these differences cost investors. To do so, we investigate the expenses associated with ESG funds and the returns offered by these funds. Contrary to the concern articulated by the DOL, we find no evidence that ESG funds cost more than comparable non-ESG funds or that they offer inferior performance during our sample period (either raw or risk adjusted). The results persist despite the inclusion of a battery of control variables intended to ensure that we are making “apples-to-apples” comparisons. While these tests are not intended to establish—nor can they establish—whether ESG funds are a “good” investment, we find no evidence that they perform worse than comparable funds.

A final empirical contribution of this paper is to address the impact of variation in ESG ratings. There is little consensus on what falls within the definition of ESG or how to weigh various ESG considerations.²³ There are over six hundred ESG rating providers, and these providers rely on a range of different data sources and employ a variety of methodologies to analyze that data.²⁴ Commentators have highlighted the fact that these differences frequently lead to different ratings.²⁵ Thus, for example, among automobile manufacturers, Tesla receives a top ESG rating from MSCI and a bottom rating from FTSE Russell.²⁶ Although we do not directly interrogate differences among providers in this paper, we take the unique approach of incorporating ESG rating data from four different and well-known providers—ISS, S&P, Sustainalytics, and TruValue Labs—to measure the ESG orientation of the mutual fund portfolios that we examine. We find that although the providers take very different approaches to measuring ESG, the patterns are remarkably stable across providers.

In sum, we provide new data on the role of ESG in mutual fund investing and its effects. Our goal in this Article is modest. We do not seek to establish that ESG funds are good with respect to any specific benchmark—that they are effective in achieving particular environmental, social, or governance objectives or that they outperform non-ESG funds. Rather, the goal of this Article is to address concerns that ESG funds present distinctive regulatory concerns relative to the mutual fund market as a whole, either because (as the SEC fears) they are not doing what they purport to do or because (as the DOL fears) their economic performance is inferior to non-ESG funds. Either concern, if established, would warrant singling out ESG funds for distinctive regulatory treatment. Our empirical results, however, provide powerful evidence that ESG funds are offering investors something different from traditional funds with respect to both portfolio composition and voting, and that they are doing so without causing investors systematically to sacrifice economic performance.

Our findings describe the current state of the market for ESG funds. We do not purport to evaluate the claims made by ESG funds in the past. It may be that the recent proliferation of ESG products has generated meaningful market discipline. Nonetheless, our findings do not suggest a need for regulatory intervention either to limit investor access to ESG products or to curtail their use by ERISA fiduciaries.

I. The Rise of ESG Mutual Funds

A. The Background of ESG

Interest in ESG stems from increasing public, issuer, and investor attention to the impact of corporate operations on stakeholders and society more broadly. A range of commentators have criticized corporations for prioritizing shareholders at the expense of employees and customers.²⁷ The need to address climate change and other environmental issues has taken on heightened urgency and led to a focus on the role that corporations play in carbon emissions and other environmentally damaging activities.²⁸ Corporations have also faced scrutiny over their role in perpetuating racial and gender discrimination, wealth and wage inequality, and exploitation of disadvantaged groups.²⁹

The ESG movement generally calls for corporations to incorporate these concerns into their business practices. ESG is a rough label for an amalgamation of voices, interest groups, and substantive concerns. Those advocating greater attention to ESG often disagree on the relative importance of the various issues that they identify.³⁰ The appropriate benchmark for corporate behavior ranges from demand that corporations at least consider a broader range of stakeholder and societal interests to an exhortation for corporations to “do no harm.”³¹

The role of ESG in investing continues to evolve. For some years, investing on the basis of ESG considerations was thought to be a preference predicated on ethical, political, religious, or other objectives rather than an investment strategy grounded in financial risk and return.³² Commentators debated whether corporations could do well by doing good, and the data generated in response to this debate was mixed.³³ More recently, an increasing number of scholars and policymakers claim that sustainable or ESG investing is associated with better economic performance.³⁴ Max Schanzenbach and Robert Sitkoff observe that investors may have different reasons for ESG investing and differentiate between ESG investing for moral or ethical reasons (which they term “collateral benefits ESG”) and ESG investing for risk and return benefits (which they call “risk-return ESG”).³⁵

One challenge to analyzing the relationship between ESG and economic performance is the absence of a clear definition of ESG. In a 2020 speech, then-acting SEC chairman Elad Roisman explains that “there is not consensus on what, exactly, ‘ESG’ means.”³⁶ Stavros Gadinis and Amelia Miazad note that ESG’s “wide scope” encompasses a range of issues, from environmental concerns and workplace relationships to “the use of sugar in packaged foods.”³⁷ Both the range of potential issues and the scope of data analysis required to evaluate a corporation’s performance with respect to those issues have fueled the development of an array of private standard-setters that gather ESG data and transform that data into company-specific ratings or rankings.³⁸ Today there are more than six hundred ESG rating organizations and rankings worldwide, and the number continues to grow.³⁹ The sheer multitude of ratings organizations makes any attempt to rank companies “difficult, and more of an art in certain situations than a science.”⁴⁰

Moreover, because organizations vary both in the data that they collect and the methodology that they use to incorporate that data, ESG ratings vary substantially among providers. Some providers rely on questionnaires to collect information from issuers, some review issuers’ public disclosures and filings, and some rely on third-party sources.⁴¹ Commentators have documented substantial variation among ratings and have, as a result, questioned the viability of evaluating an issuer’s ESG accurately.⁴² SEC commissioner Hester Peirce notes that “the different ratings available can vary so widely[] and provide such bizarre results that it is difficult to see how they can effectively guide investment decisions.”⁴³

B. The Growth of ESG Mutual Funds

The challenges associated with defining and measuring ESG have not impeded its growth as an investment strategy. The use by mutual funds of ESG criteria in selecting investments and engaging with portfolio companies is one of the hottest investment trends. Investments in the United States in funds using ESG data have almost doubled over the last four years, from $22.9 trillion in 2016 to over $40 trillion in 2020.⁴⁴ Net flows of assets into ESG funds “in 2020 were more than double the total for 2019 and nearly 10 times more than in 2018.”⁴⁵ Many retail investors express strong preferences for ESG investing, and the mutual fund market is driven largely by those preferences.⁴⁶

ESG investing can incorporate several different strategies. One is investment screening, in which a fund uses ESG data as a component of its investment decisions. Funds may engage in negative or exclusionary screening, in which they exclude certain types of companies—oil, tobacco, and gambling companies are common examples—from their portfolio. Alternatively, funds can engage in positive screening, which involves limiting their portfolios to investments that meet designated ESG criteria. Funds can also incorporate ESG data as part of a more comprehensive analysis of an investment, what some funds term an “integrated” use of ESG criteria. For example, Vanguard’s Global ESG Select Stock Fund describes its ESG strategy as “[r]egularly including ESG factors alongside the traditional investment analysis performed by active fund managers. This strategy doesn’t require the fund to rule out any company, industry, or country simply because it’s involved in a business activity that may be objectionable to some.”⁴⁷

ESG investing’s potential impact on mutual fund investors increases because of the ease of incorporating ESG screening into a passive investment strategy. The amount of money invested through passive or indexed strategies has grown dramatically, fueled both by the low costs of indexing and by studies suggesting that active strategies do not consistently outperform indexed strategies over time.⁴⁸ The extensive number of ESG rating organizations creates a ready tool for an index-based investment product in which the selection of a mutual fund’s portfolio companies is predicated on a rating conferred by an external provider. For example, four of Vanguard’s five current ESG fund offerings are indexed, with their portfolio composition tracking several indices created by FTSE Russell.⁴⁹ Indexed mutual funds and exchange-traded funds (ETFs) enable funds to offer the cost advantages and scalability associated with a passive investment strategy, while enabling fund sponsors to delegate the evaluation of portfolio companies to an index provider.⁵⁰

A second ESG investment strategy is engagement.⁵¹ ESG engagement involves a fund exercising its power as a shareholder in an effort to cause its portfolio companies to perform better on some ESG criteria. Engagement may be limited to how the fund votes the shares of its portfolio companies but can also include more proactive measures such as writing letters, meeting with management, sponsoring shareholder proposals, and initiating litigation.⁵²

A third strategy is impact investing, which targets companies seeking to achieve specific goals that are beneficial to society. Impact investing might direct an investment to a company that produces a clean energy product, such as wind or solar power, or alternatively might finance efforts to convert the manufacturing processes of a traditional company to reduce its environmental impact.⁵³

Notably, the foregoing ESG strategies can be used independently or in combination. For example, although an S&P 500 index fund⁵⁴ does not incorporate ESG considerations into its stock selection process, nothing prevents that fund from engaging on ESG issues. By the same token, a fund that invests according to ESG criteria need not engage on ESG issues. Indeed, media reports have highlighted instances in which ESG funds have not voted differently, even with respect to ESG issues, from non-ESG funds.⁵⁵

How funds communicate the role that ESG plays in their investment strategy is a separate issue. One obvious tool for communicating a fund’s strategy is its name. We constructed our initial sample of ESG funds, for example, by searching for terms such as “sustainable,” “ESG,” and “green” in fund names. Using a fund’s name to communicate the role of ESG is tricky, however.

Consider a fund called “XYZ Green Fund.” What information is conveyed by the fund’s name? One possibility is that the fund invests in sustainable or green industries, such as solar panel and wind turbine manufacturers or makers of electric vehicles. An alternative is that the fund seeks out companies that have environmentally responsible practices relative to their industry peers, motivated by a desire to encourage more environmentally responsible practices or by the expectation that such companies will be better positioned to withstand market and regulatory burdens that may be imposed on environmentally harmful companies in the future. The delivery service UPS runs a large fleet of diesel trucks, hardly a “green” business compared to wind farms. If UPS has converted more of its vehicles to electric than competitors, however, then the XYZ Green Fund might buy more of UPS and less of FedEx. Many of the companies in this fund’s portfolio may not be “green” in the sense used above, but they may be better situated than their competitors to thrive if carbon emissions are restricted.

There is a third possibility too. Perhaps XYZ Green Fund is an impact fund that seeks to make companies greener (either to generate returns or to make the world a better place). That XYZ Green Fund might hold a portfolio of particularly egregious polluters—industrial dinosaurs that have failed to consider their environmental impact at all—and then, through the power of their proxy voting, attempt to induce those companies to improve. The portfolio of such a fund would look anything but green in either of the above senses, but such a strategy might nevertheless be consistent with the “green” name so long as the fund seeks to reform those companies.

Independent of a fund’s strategy for investing and engaging is the challenge of determining what counts as an environmentally responsible company. Is Tesla a green company because it makes electric vehicles, or is it not, because it harvests vast quantities of lithium for its batteries? A wind farm company might be green because it produces electricity with zero emissions, but what if it repeatedly refuses to take straightforward steps to mitigate the impact of its wind farms on wildlife?⁵⁶

What is true of the “E” in ESG is equally true of “S” and “G.” A company might be a leader in addressing workplace inequality but fail to oversee child labor practices in its supply chain. Funds that purport to consider E, S, and G must also consider how to weigh practices across all three categories. Does Facebook’s low carbon footprint outweigh its failures in safeguarding customer privacy or its dual-class voting structure? Of course, this problem is not unique to ESG investing; there are a plethora of different self-described “growth” and “value” funds in the market, and different funds sometimes have very different conceptions of what “growth” and “value” investing strategies mean.⁵⁷ Notwithstanding this, no one reasonably argues that funds should not be able to use the words “growth” or “value” in their names.

Obviously, a fund’s name cannot fully explain its investing strategy. Mutual fund companies are required to provide information beyond fund names, however. They must share information that enables investors to determine if the fund’s conception of ESG matches the investor’s preferences. The SEC’s disclosure requirements for mutual funds take a layered approach.⁵⁸ Mutual funds disclose a minimum amount of information in the summary prospectus, which is typically three to four pages in length.⁵⁹ Additional information is provided in the statutory prospectus and the statement of additional information.⁶⁰ In these documents, funds disclose their investment objectives and how they incorporate ESG criteria. They also disclose whether they follow an index strategy and, if so, the applicable index. In many cases, they also disclose their policies regarding voting or engagement. Furthermore, mutual funds are required to disclose their portfolio holdings on a quarterly basis.⁶¹ Finally, SEC rules adopted in 2003 require mutual funds to disclose their overall voting policies as well as the votes they cast at each of their portfolio companies.⁶²

Commentators have criticized the mutual fund disclosure system on the basis that retail investors rarely read the prospectus or other disclosure documents.⁶³ In recent years, however, internet-based disclosures have become increasingly detailed. In addition to providing links to the mandated disclosure documents, funds generally provide detailed descriptions of their screening and engagement strategies on their websites.⁶⁴ Mutual fund companies are starting to post their voting disclosures on their websites as well. For example, Vanguard provides a tool that enables investors to search by fund for the proxy votes cast at each of the fund’s portfolio companies for the 2020–2021 proxy season.⁶⁵

C. Concerns over ESG Funds

The growth in number and size of ESG funds has led to several concerns. Perhaps the most serious concern is that ESG funds falsely portray themselves as adhering to an ESG investing (or voting) strategy to attract investor money, a practice characterized as “greenwashing.”⁶⁶ Without consistent data for evaluating the sustainability of individual portfolio companies, it is difficult to measure the ESG orientation of a mutual fund or to compare the “greenness” of one fund’s portfolio with that of another. As one commentator observes, the possibility that investors do not understand what they are buying or are misled by false claims of sustainability raises consumer-protection concerns, pointing to a gap in existing law.⁶⁷

Greenwashing is not the only issue. Commentators express concern that ESG funds charge higher fees.⁶⁸ These fees could reflect the higher costs associated with identifying and monitoring investments from an ESG perspective.⁶⁹ It could also be the case that ESG funds are smaller and therefore less able to benefit from economies of scale.⁷⁰ More nefariously, funds could be capitalizing on the demand for ESG products and charging high fees while providing little incremental value to investors.

Another concern is that ESG funds sacrifice performance. Commissioner Roisman, for example, has worried about “the extent to which retail investors understand that some of these funds may be prioritizing environmental or social goals above the fund’s economic returns.”⁷¹ Although early studies provided some evidence for this claim, more recent studies suggest that ESG strategies have evolved and that ESG funds have performed as well as or better than non-ESG funds in recent years.⁷²

A final concern is the variety of ESG funds and the investment strategies they offer. ESG funds may be actively managed or tied to an index. They may focus on a small number of companies or offer broad diversification. They may focus on stock selection or engage actively with their portfolio companies. And they may offer a range of substantive ESG priorities—environmental sustainability, diversity, or ethical and religious values—or take a more generalist approach to ESG. This variation may lead to investor confusion.⁷³ Dana Reiser and Anne Tucker warn that the dizzying array of ESG mutual funds means that investors cannot readily “differentiate between their claims of ESG effort or impact.”⁷⁴ Some commentators have called for increased disclosure mandates, such as an SEC requirement that mutual funds disclose how they “approach ESG and long-term matters generally, including voting and any engagement.”⁷⁵ In a 2020 speech, then-acting SEC chair Allison Herren Lee has warned that greater regulation may be necessary, both to standardize disclosure by ESG fund managers and to “require advisers to maintain and implement policies and procedures governing their approach to ESG investment.”⁷⁶

II. Regulatory Pressure on ESG Funds

The significant growth in ESG investing has begun to attract the attention of regulators. In 2020, both the SEC, which comprehensively regulates mutual funds, and the DOL, which regulates the trillions of dollars saved in employee retirement accounts,⁷⁷ took action motivated by concerns about ESG investing. The SEC sought comments from the public on potential future regulation related to the use of ESG terms in fund names.⁷⁸ In doing so, the SEC raised a number of important issues about what, exactly, ESG funds are selling.⁷⁹ Meanwhile, the DOL adopted a rule creating potential legal risk for retirement plans that include ESG funds.⁸⁰ The DOL rule was adopted over vigorous dissent from much of the asset management industry,⁸¹ although as of early 2021, its future is uncertain.

A. The SEC Names Rule

The SEC has signaled interest in potentially tightening regulation of mutual fund names that suggest ESG investing.⁸² This interest is motivated by concern about greenwashing—that a fund might incorporate labels such as “ESG,” “green,” or “sustainable” to give investors the false impression that the fund offers ESG exposure when it actually invests conventionally.⁸³ In a speech, SEC commissioner Hester Peirce noted that “[i]nvestors are pouring assets into ESG-labelled investment products, and asset managers are churning out new products in response. While the demand for these products is clear, less clear is what exactly these investors are buying.”⁸⁴ In particular, Commissioner Peirce highlighted the risk of “an asset manager who talks the ESG talk, but doesn’t walk the ESG walk.”⁸⁵ Given the rapidly increasing demand for ESG funds, should we be concerned that funds that hold themselves out as pursuing social or environmental goals through their names are not actually delivering on those marketing promises?

The SEC’s Names Rule, Rule 35d-1, was predicated on the fact that mutual fund names are an important source of information to investors.⁸⁶ Under section 35(d) of the Investment Company Act, it is unlawful for a fund to use in its name “any word or words that the Commission finds are materially deceptive or misleading.”⁸⁷ Originally, the SEC policed naming conventions through staff guidance and occasional one-off enforcement actions for particularly misleading funds.⁸⁸ In 1996, Congress amended the Investment Company Act of 1940 to give the SEC explicit rulemaking authority to enforce section 35(d).⁸⁹ As a result, the SEC promulgated the Names Rule in 1997 and adopted it in 2001.⁹⁰

The Names Rule outlines requirements for the use of certain terms in mutual fund names. The most important requirement is for funds whose name suggests a particular type of investment or industry. Under the rule, a fund whose name contains a type of security, an industry, or a geographic area must hold 80% of its portfolio in investments consistent with the designation.⁹¹ Thus, the “XYZ Pharmaceuticals Sector Fund” must hold 80% of its assets in pharmaceutical companies, and the “ABC Bond Fund” must hold 80% bonds.

This much is straightforward, but the Names Rule also excludes terms that describe a fund’s “investment objective, strategies, or policies” from the 80% requirement.⁹² Thus, while a “stock fund” must hold 80% stock, there is no requirement under the Names Rule that a “growth fund” hold 80% of its portfolio in assets with any particular characteristic. For example, “growth” is an investment strategy that has many different connotations. To some, a growth stock is a stock of a company with a low ratio of book value to market value of equity.⁹³ To others, it connotes smaller companies with higher potential returns (often coupled with higher risk).⁹⁴ Given this ambiguity, assessing what counts as a growth stock is far more difficult for a regulator than identifying “stock” in a portfolio. While names including terms like “growth” and “conservative” are subject to the antifraud provisions of the securities laws and can be “deceptive or misleading” within the meaning of section 35(d), the bright-line requirement of the Names Rule does not apply to these terms.

Despite the limitations imposed by the Names Rule, funds have substantial leeway in the names that they choose. Moreover, investors rely heavily on names in selecting mutual funds. Studies have shown that when mutual funds adopt a name that is associated with a hot investment trend or style, investments into the fund increase even if the name change does not reflect any change in the fund’s underlying strategy.⁹⁵

The SEC has not updated the Names Rule in twenty years. In 2020, however, the SEC issued a request for comment on the rule.⁹⁶ Among the issues upon which the SEC requested comment was the application of the Names Rule to ESG funds. The SEC specifically noted potential confusion about whether ESG is an investment type (to which the Names Rule would apply) or an investment strategy (to which it would not).⁹⁷

Lurking behind the naming issue is the genuinely unsettled reality of ESG investing. ESG is a rapidly evolving space with numerous strategies pursuing different goals in different ways, and investors may not understand the role of ESG in a particular fund’s strategy. For that reason, the SEC suggestively asked, “Instead of tying terms such as ‘ESG’ in a fund’s name to any particular investments or investment strategies, should we instead require funds using these terms to explain to investors what they mean by the use of these terms?”⁹⁸ Commissioner Roisman signaled a similar concern in a 2020 speech:

[A]sset managers who want to use these terms to name their funds or advertise their products should be required to explain to investors what they mean. . . . [H]ow do the terms “ESG,” “green,” and “sustainable” relate to a fund’s objectives, constraints, strategies, and the characteristics of its holdings? Are “E,” “S,” and “G” weighted the same when selecting portfolio companies? Does the fund intend to subordinate the goal of achieving economic returns to non-pecuniary goals, and, if so, to what extent?101F⁹⁹

Requiring ESG funds to explain their ESG commitments to investors seems unobjectionable; even absent regulatory change, funds holding themselves out as ESG funds ought to be delivering something different to investors to justify their use of the ESG nomenclature. The portfolios of ESG funds should be distinguishable from non-ESG funds. Similarly, although the voting policies of ESG funds might differ among themselves, we would expect ESG funds collectively to vote differently from non-ESG funds, especially on salient ESG issues.

B. DOL Fiduciary Duties in Retirement Plans

Participant-directed retirement accounts, such as 401(k) plans, are among the largest holders of mutual funds.¹⁰⁰ In a participant-directed plan, the employer provides plan participants (employees) with a menu of investment options, and plan participants decide how to allocate their money among those options.¹⁰¹ Under ERISA and DOL regulations, retirement plans are subject to a complex set of rules regarding investment selection, plan design, and the obligations of employers in interacting with plan assets.¹⁰²

ERISA applies trust law to the management of retirement accounts, with plan sponsors held to stringent fiduciary duties.¹⁰³ Under section 404 of ERISA, fiduciaries for employee benefit plans, including retirement plans, must act prudently to minimize risk to investors, including by diversifying plan assets.¹⁰⁴ Fiduciaries must also act “solely in the interest” of plan participants for the purpose of “providing benefits to participants and their beneficiaries.”¹⁰⁵ These obligations track trust law’s fiduciary duties of loyalty and care.¹⁰⁶

Whether ERISA fiduciaries can properly include ESG funds in retirement plans has been the subject of ongoing debate, one premised largely on the question of whether taking nonpecuniary criteria into account is consistent with ERISA’s mandate.¹⁰⁷ Different presidential administrations have taken different approaches with respect to whether ESG investing is consistent with ERISA fiduciary duties.

Starting in 1994, the DOL addressed the issue of “economically targeted investments” (ETIs), defined as “investments selected for the economic benefit they create apart from their investment return to the employee benefit plan.”¹⁰⁸ The DOL explained that while fiduciaries need to act “solely in the interest of the plan’s participants and beneficiaries and for the exclusive purpose of providing benefits to their participants and beneficiaries,” the “fiduciary standards applicable to ETIs are no different than the standards applicable to plan investments generally.”¹⁰⁹ That meant that as long as the ETI investment was as good as other options available to the plan, it could be prudently chosen. A plan could not, however, accept lower economic returns to pursue collateral benefits.

The DOL’s guidance on what is now called ESG investing turned more negative in 2008.¹¹⁰ In a 2008 interpretive bulletin, the DOL stated that “ERISA’s plain text does not permit fiduciaries to make investment decisions on the basis of any factor other than the economic interest of the plan.”¹¹¹ As a result, a fiduciary that weighed collateral benefits when choosing an investment risked breaching its fiduciary duties to the plan. The sole exception, in view of the DOL, was if after “examin[ing] the level of diversification, degree of liquidity, and the potential risk/return” of prospective investments, the fiduciary deemed “two or more investment alternatives” to be “of equal economic value to a plan,” in which case the fiduciary was permitted to factor in the noneconomic benefits as a tiebreaker.¹¹² The bulletin’s approach reflected the then-prevailing view that considering ESG factors might benefit society generally but would usually come at the expense of returns.¹¹³

In 2015, the DOL withdrew the stricter 2008 guidance and reinstated the 1994 articulation of the standard.¹¹⁴ In so doing, the DOL emphasized that just because an investment is an ETI or ESG investment does not mean that the investment is “inherently suspect or in need of special scrutiny.”¹¹⁵ A “fiduciary may not use plan assets to promote social, environmental, or other public policy causes at the expense of the financial interests of the plan’s participants and beneficiaries,”¹¹⁶ but may consider ESG investments that provide equal economic value without drawing special scrutiny or incurring paperwork obligations.

A field assistance bulletin in 2018 “clarified” the 2015 guidance by offering new—and more negative—guidance on how plans can consider ESG factors.¹¹⁷ The bulletin warned that an ERISA fiduciary’s use of ESG factors must “be appropriate to the relative level of risk and return involved compared to other relevant economic factors.”¹¹⁸ Reflecting the increasingly accepted view that ESG factors can be relevant to risk and return, the 2018 guidance acknowledged that “ESG issues [can] present material business risk[s] or opportunities.”¹¹⁹ The guidance noted that if a fiduciary deems ESG factors to “present material business risk[s] or opportunities,” then those factors “should be considered by a prudent fiduciary along with other relevant economic factors to evaluate the risk and return profiles of alternative investments.”¹²⁰

In 2020, the DOL raised the stakes by engaging in rulemaking rather than subregulatory guidance.¹²¹ Its proposed rule, issued on June 30, 2020, took a decidedly negative view of ESG funds. The DOL wrote in the preamble that “[a]s ESG investing has increased, it has engendered important and substantial questions and inconsistencies, with numerous observers identifying a lack of precision and rigor in the ESG investment marketplace”¹²² and said flatly that “ESG investing raises heightened concerns under ERISA.”¹²³

The DOL pointed to the inconsistencies among ESG ratings,¹²⁴ which it described as “vague,” to argue that “[t]here is no consensus about what constitutes a genuine ESG investment.”¹²⁵ The DOL also pointed to the cost of ESG funds, stating that “ESG funds often come with higher fees[] because additional investigation and monitoring are necessary to assess an investment from an ESG perspective.”¹²⁶ The DOL cited the SEC’s request for comment on the Names Rule as evidence of the “questions and inconsistencies” plaguing the ESG space.¹²⁷

Most importantly, the DOL highlighted the concern that ESG funds might be affirmatively inferior to non-ESG funds from a pecuniary perspective:

[I]n the case of some ESG investment funds being offered to ERISA defined contribution plans, fund managers are representing that the fund is appropriate for ERISA plan investment platforms, while acknowledging in disclosure materials that the fund may perform differently or forgo certain opportunities, or accept different investment risks, in order to pursue the ESG objectives.130F¹²⁸

The DOL’s proposed response to these concerns was to reiterate that a plan may not subordinate risk and return to achieve collateral benefits outside the plan. The proposed rule explicitly stated:

Plan fiduciaries are not permitted to sacrifice investment return or take on additional investment risk to promote non-pecuniary benefits or any other non-pecuniary goals. Environmental, social, corporate governance, or other similarly oriented considerations are pecuniary factors only if they present economic risks or opportunities that qualified investment professionals would treat as material economic considerations under generally accepted investment theories. The weight given to those factors should appropriately reflect a prudent assessment of their impact on risk and return.131F¹²⁹

The rule would also have barred “environmental, social, corporate governance, or similarly oriented”¹³⁰ funds from being used as default options in 401(k) plans, though such funds were still permissible if their inclusion was based only on “objective risk-return criteria, such as benchmarks, expense ratios, fund size, long-term investment returns, volatility measures, investment manager investment philosophy and experience, and mix of asset types” in selecting options for the plan.¹³¹ In short, the DOL’s proposed rule would have subjected ESG investments to heightened scrutiny for potential fiduciary breach.

The proposed rule’s skepticism toward ESG funds was met with withering criticism from most of the asset management industry.¹³² Three months later, on November 13, 2020, the DOL adopted a substantially modified final rule.¹³³ Significantly, the new rule removed all explicit discussion of ESG considerations and dropped the requirement that ESG funds not be used as qualified default investment alternatives.¹³⁴ The revised version still emphasized the need for fiduciaries to base investment decisions on pecuniary factors and, as a result, poses some risk to the ESG investment space.¹³⁵ By focusing investment choice on pecuniary factors, it may be difficult for funds to rely on ESG ratings unless all the criteria underlying those ratings are pecuniary in nature.¹³⁶

On October 14, 2021, the DOL published a proposed rule that attempts to address the uncertainty created by the DOL’s prior ESG rule. The proposed rule amends ERISA’s description of investment prudence duties to include explicit authorization for fiduciaries to consider climate-change-related factors, governance factors, and workforce practices.¹³⁷ The proposed rule, if adopted, is likely to reduce the concerns of ERISA fiduciaries about ESG investment options. On the other hand, plan sponsors might be sufficiently risk averse, particularly in light of the threat of private litigation, to avoid funds that foreground ESG goals. To date, ESG funds are rarely included as an investment option in 401(k) plans.¹³⁸ The DOL’s shifting position reflects an underlying skepticism toward ESG investing based on claims about cost and performance that are empirically testable. In what follows, we test some of the claims that motivate the DOL rule and SEC request for comment.

III. Empirical Analysis

This Part presents our empirical tests of the differences between ESG funds and other mutual funds. We first construct several categories of ESG funds. We then examine the holdings, voting practices, costs, and performance of ESG funds, and compare them with the rest of the mutual fund industry, or what we term non-ESG funds. To summarize our results briefly, we find that ESG funds generally deliver greater ESG exposure in their portfolio allocations than non-ESG funds, that they are more likely than other funds to oppose management in proxy voting (particularly when votes are salient to ESG issues), and that they do not cost more or perform worse than similar non-ESG funds. Overall, our findings suggest that, on average, investors in ESG funds do seem to be getting funds that “walk the walk,” at least to some extent, without any material sacrifice in performance or cost.

Section III.A describes our data. Sections III.B and III.C relate primarily to the SEC’s concerns about ESG funds, namely, whether investors are getting what they think they’re getting when they invest in ESG funds. Section III.B examines the portfolio composition of ESG funds and explores to what extent ESG funds invest in companies with higher ESG ratings. Section III.C turns to fund voting behavior and explores the differences between the voting patterns of ESG and non-ESG funds. As is appropriate in a regime based on antifraud provisions, both concerns center around investor expectations. Section III.D then considers the DOL’s primary concerns about ESG funds. Unlike the SEC, the DOL’s focus is on pecuniary costs and pecuniary performance. We explore fund fees and performance to investigate the extent to which these concerns are warranted.

A. Description of Data

1. Data Sources

The market for ESG products has evolved rapidly, both in the number of such products offered and in their characteristics. Analyses that use data from as recently as five years ago may not accurately reflect current market realities. Our analysis focuses on 2018–2019 because these are the most recent years for which data were available.

Mutual fund performance and fee data, as well as holdings data, come from the Center for Research in Security Prices (CRSP) Survivorship Bias Free Mutual Fund Database, which we obtained through Wharton Research Data Services (WRDS). Mutual fund voting data for 2018 and 2019 comes from ISS’s Voting Analytics Database, also through WRDS.

We used two different methods to identify ESG funds.¹³⁹ First, we screened funds based on their names. We identified all mutual funds whose names contained one or more relevant keywords.¹⁴⁰ We then checked this list by hand to ensure that all the identified funds had an ESG connotation. From this, we dropped funds that focus primarily on asset classes other than equities¹⁴¹ as well as funds for which we lacked data. This yields 204 funds, which we refer to as the “ESG name” funds. Second, we relied on a list of ESG funds compiled by Morningstar obtained in May 2020. This list contained 314 funds, but after dropping 11 funds that we were unable to match with CRSP portfolio numbers and those that focus on asset classes other than equities, we are left with 241 funds that we call the “Morningstar ESG” funds. Because of overlap between the two groups, this results in a total of 303 different funds over the sample period.¹⁴²

In addition to Morningstar ESG funds, Morningstar identifies a group of what it calls “ESG Consideration” funds. These are funds that, according to Morningstar, do not have ESG as a central part of their investment strategy but nevertheless mention ESG as a factor considered by the fund managers in assessing investment options.¹⁴³ These funds do not have names that connote ESG investing, nor do they market themselves as ESG funds. While we do not classify these as “ESG” funds, we repeat many of our analyses in this Part on this set of funds.

As noted above, hundreds of providers collect ESG data and disseminate ESG ratings, and their ratings are only weakly correlated. We are agnostic on whether it is possible to measure an issuer’s ESG in a manner that is objectively correct. To address this concern as well as the possibility that a particular provider’s approach may be idiosyncratic, we incorporate ratings from four leading ESG ratings providers—S&P, Sustainalytics, TruValue Labs, and ISS. Providers also differ in coverage, so some portfolio firms have ratings from some, but not all, of the providers.

While all four ESG ratings that we use involve a comprehensive assessment, they employ significantly different data and methodologies. S&P’s ESG ratings rely heavily on information obtained directly from the company in question. Rated companies are sent a detailed industry-specific questionnaire and scores are based on both these responses and other publicly available data.¹⁴⁴

In contrast, rather than beginning with the company, Sustainalytics’s methodology during our sample period sought comment from the company in question only near the end of its six-step process.¹⁴⁵ Its process begins with information collected from the company’s own public disclosures, which are then supplemented by reporting from NGOs and the media.¹⁴⁶ In steps three and four, the data are analyzed, and the company is compared against its peers.¹⁴⁷ The draft report is prepared and sent to the company for feedback before the report is revised and made public.¹⁴⁸

For its part, TruValue Labs does not incorporate any data that is obtained directly from the company in creating its ratings. Instead, it relies exclusively on third-party data about the company, including reports by regulators, media, analysts, and advocacy groups.¹⁴⁹ TruValue Labs then processes this information using a natural-language machine-learning algorithm to come up with its scores.¹⁵⁰ Together, these two features mean that TruValue Labs is employing both different data and a different analytical approach to constructing its ratings than ISS, Sustainalytics, or S&P. We rely on TruValue Labs’s “insight” score in our primary analysis.

Rather than producing a score from 1 to 100, ISS assigns firms a rating from A+ to D-, which can be mapped to a score from 1 to 4 (similar to a grade point average). The scores are assigned by way of a performance assessment, which draws from a pool of over 800 indicators. Industry plays an important role: the overwhelming majority of the indicators ISS uses (approximately 90 percent) are industry specific, and the rating structure provides different weights to the indicators depending on the industry.¹⁵¹ We use the “overall” rating in our primary analysis.

We use these ratings to construct a fund portfolio’s “ESG tilt.” To do so, we use the ESG rating of the companies that the fund invests in to calculate the average ESG rating of the fund’s portfolio, weighted according to the proportional share that each company represents of the fund’s total portfolio. Because we do not have ESG ratings by all providers for all the companies that each fund invests in, we scale the weights by the proportion of the portfolio for which ratings are available.¹⁵² We refer to this as the “scaled weighted ESG score” or simply the “weighted ESG score.” For robustness, we also use these weighted ESG scores to construct “ESG percentile scores.” These scores reflect a portfolio’s ranking relative to all the portfolios in our sample in a given quarter. Because they capture slightly different things, we use both these ESG tilt scores in our analysis.

2. Sample Construction

Before beginning our quantitative analysis, we downloaded the prospectuses of each of the funds identified using the fund’s name or by Morningstar (either as an ESG fund or as an ESG consideration fund). At least two people—one research assistant and at least one author—then read each prospectus. What we uncovered was a rich and diverse array of funds.¹⁵³ Importantly, we know about this diversity precisely because the funds tell investors about it in their prospectuses.

Each reader independently read and coded the prospectuses for a number of features, including the asset class targeted by the fund (equities, debt, mixed, or other) as well as the fund’s objective, including whether it focused on the environment, social issues, governance, or some combination of these. The readers also noted whether the fund was generic or had a more specific objective. Throughout, readers noted things that stood out about the funds, which provide additional texture to our understanding of this market. For example, in reading the prospectuses, readers noticed a substantial number of religiously motivated funds, which enable investors to invest according to the tenets of a particular religion. Readers also noted some of the more specialized funds that they came across, including funds that incorporated concerns about gender diversity (like the SPDR SSGA Gender Diversity Index ETF) or animal welfare (like the Karner Blue Animal Impact Fund) into their investment strategy.

After completing the initial coding, we took a second pass through the prospectuses to collect information on how the fund described its approach to ESG issues when voting. Here again, we found a substantial amount of diversity: some had centralized proxy voting policies that considered ESG matters to varying degrees, while others had fund-specific voting policies. Some of these policies focused on the impact of ESG issues on firm value, while others explicitly sought to support ESG improvement. Interestingly, some prospectuses explicitly noted the manager’s belief that investors in ESG funds might be particularly interested in proxy voting with respect to ESG matters. For example, while BlackRock Investment Stewardship (BIS) generally expects companies to manage environmental and social risks effectively, the prospectus’s description of the proxy voting guidelines for the family of iShares ESG funds (including the iShares ESG MSCI USA ETF) specifically contemplates “split voting,” where BlackRock would vote differently for different funds.¹⁵⁴ According to the prospectus, this is based on “an assessment that clients invested in the ESG funds may expect more urgent action be taken by the company” and BlackRock’s view that “it is reasonable to expect that clients invested in ESG funds may be less patient with regard to evolution in corporate policies on material E&S matters and therefore wish to send a stronger signal to the company by supporting a shareholder proposal.”¹⁵⁵ This suggests that at least some fund managers are attentive to the particularized expectations of investors in ESG funds and are seeking to meet these expectations.

Finally, we merged our hand-collected information with fund-level data from the CRSP mutual fund database. This allowed us to identify index funds, fund objectives,¹⁵⁶ and other important attributes of the funds in our sample.

In Table 1, we provide a breakdown of the ESG funds that were included in our final sample. In Panel A, we summarize the number of ESG funds included in our final sample, broken down by how the fund entered our sample. In Panel B, we tabulate the number of certain subtypes of ESG funds in the sample. For brevity, we restrict the breakdown in Panel B to just the subtypes of funds that we investigate in the quantitative analyses in the remainder of this Part. Because missing data forces us to omit some funds from some of the analyses below, Table 1 reports only the ESG funds we were able to include.

B. Portfolio Composition

The first thing an investor might reasonably expect from an ESG mutual fund is a portfolio that is tilted toward companies with better ESG characteristics. That is, an ESG fund should hold larger positions in stocks that perform well on ESG metrics relative to non-ESG funds with similar investment objectives. We therefore investigate the extent to which ESG funds invest in portfolio companies with higher ESG ratings. To start, we simply plot the distribution of ESG tilts among ESG funds and compare it to the non-ESG funds in our sample. This approach has two major benefits. First, because we are plotting the entire distribution, it is easy to see what parts of the distribution are driving the difference, if any, between the groups. Second, because it is so simple, no complex statistical analysis is required to produce or interpret the results.

Figure 1 contains histograms using weighted ESG scores from the four different data providers. The shaded histograms represent the distribution of ESG funds; the transparent histograms represent conventional funds. “ESG funds” refer to funds that either identify themselves as ESG by their name or are identified by Morningstar as ESG funds.¹⁵⁷ Panel A includes 259 distinct ESG funds, Panel B contains 279 ESG funds, Panel C contains 243 ESG funds, and Panel D contains 271 ESG funds. In all cases, the non-ESG funds include all funds in the CRSP database (other than those we identify as ESG funds) for which we have enough data to produce a portfolio tilt score. The histograms are constructed using quarterly fund-level data.¹⁵⁸ This means that if a fund appears in the sample for a full year, it will appear four times in the histograms.¹⁵⁹

The consistency across the panels in Figure 1 is striking. Using any of the measures of ESG tilt, we find the same general pattern: ESG funds have portfolios with higher ESG scores, on average, than non-ESG funds. As we would expect with an average, the distributions tend to have a central peak, with bars fanning out on both sides. While this general shape applies to both the ESG and the non-ESG funds in the sample, the ESG distribution is shifted slightly to the right of the non-ESG distribution in all four panels. There is, of course, substantial variation across the panels: the pattern is perhaps more visible in Panel A, which uses Sustainalytics scores, than it is in Panel B, which uses scores from TruValue Labs. Panels C and D, which use scores from ISS and S&P, respectively, appear to be somewhere in between. In Panel C, there is a sharp spike of ESG funds slightly above the mean value, but there is simultaneously a significant mass of non-ESG funds at the very top of the distribution. In Panel D, the distribution of non-ESG funds is double humped, while ESG funds are largely missing from the lower hump. Notwithstanding this variation across providers, the overall pattern is the same.

Certainly, there are some ESG funds with low ESG tilts (represented by the ESG bars toward the left of each of the panels), just as there are some funds that are not classified as ESG funds that have high ESG tilts (represented by the non-ESG bars toward the right of each of the panels). This raises the question whether there is a substantial population of ESG funds that do not exhibit substantial ESG tilt with respect to any of our ESG rating services. Even if the average ESG fund has increased exposure to strong ESG companies, there could be a group of ESG funds that are conventional funds masquerading as ESG funds. The fact that most ESG funds do what they claim to be doing is, after all, cold comfort to an investor who is unlucky enough to invest in one that does not. A significant number of such funds could raise concerns for regulators.

FIGURE 1: ESG TILT OF MUTUAL FUND PORTFOLIOS: WEIGHTED ESG SCORES

To test whether there are a significant number of “fake” ESG funds, we examine the funds that fall below varying percentile cutoffs under all four of the ESG measures. Of course, this alone would not necessarily demonstrate that the fund is not in good faith pursuing an ESG strategy: the fund’s portfolio might score better under a different scoring methodology, or the fund might employ its own idiosyncratic ESG analysis. Alternatively, the fund might be pursuing an impact strategy in which it invests in low-scoring firms and then pressures management to do better.¹⁶⁰ So while a low score under all four of the methodologies that we use might constitute a red flag, it is not dispositive. As it turns out, however, this red flag goes up for very few funds.

Only two ESG funds are in the bottom 20 percent of tilt for all four data providers, and none stays there for the whole sample period. Of course, even two funds might be cause for concern if the funds are promising investors something quite different from what they are delivering. We therefore return to these funds’ prospectuses and websites to investigate how these funds are presenting themselves to investors. Both funds are managed by a fund manager that specializes in impact and ESG investments: one of them is expressly an impact fund, and the other commits only to considering ESG factors in portfolio selection. Given this impact orientation, it is perhaps unsurprising that the funds sometimes invest in companies with low ESG scores.

Doubling the cutoff to 40 percent implicates only eleven ESG funds, none of which remains there for the whole sample period. Even if we raise the cutoff to the 50th percentile, we find that only sixteen funds are consistently below the median, only two of which remain there for the entire sample period. This is a fairly striking result: of the 280 ESG funds included in this analysis, and even with incomplete insight into ESG exposure, only two ESG funds are not above the median ESG tilt at some point during our sample period. This strongly suggests that the appearance of ESG funds in the left tail of the histograms above is an artifact of the different scoring methodologies used across the four data providers rather than a failure to deliver ESG exposure with respect to some identifiable measure.

The heterogeneity of ratings approaches also underscores the general consistency of the histograms in Figure 1. Though each ratings provider measures different firm characteristics in different ways and the correlation among ratings is relatively low, the different ratings consistently show higher ESG exposure for ESG funds when viewed in aggregate.

There are some limitations to examining histograms. First, they do not easily accommodate the use of control variables to account for other features of the mutual funds. Second, this analysis does not allow us to conduct statistical tests to investigate whether observed differences are statistically significant.

The results in Table 2 are again consistent. Using all four ESG ratings, and

This pattern does not hold for the funds Morningstar identifies as ESG consideration funds. To show this, we repeat the analysis from Table 2 using an indicator for the fact that a fund is an ESG consideration fund. The results, presented in Table 3, are highly variable. In columns 2, 4, 7, and 8, the coefficients are positive (although not statistically significant), while in columns 1, 3, 5, and 6, they are negative (but again not statistically significant). In other words, there is no systematic statistical relationship in either direction. Given this, we are unable to make any firm conclusions about the overall ESG characteristics of these funds. One possibility is that there is more variation across funds in this category, making it hard to detect average differences.¹⁶¹

To summarize, the results in Tables 2 and 3 make clear that, on average, an investor in an ESG fund is investing in a portfolio that is more tilted toward companies with high ESG scores than an investor in a non-ESG fund. This holds whether we identify ESG funds by their names or using Morningstar’s classification. At the same time, we find no consistent evidence that funds that are described as “considering” ESG hold portfolios with more ESG characteristics. By and large, then, the evidence suggests that investors in ESG funds are getting something more than “greenwashing,” notwithstanding the concerns of some at the SEC.

Of course, the category of “ESG funds” is extremely broad, and environmental concerns can be qualitatively different from governance concerns. As a result, the fact that ESG funds have, on average, portfolios that are substantially more tilted toward companies with high ESG ratings does not necessarily mean that investors in environmental funds, for example, are getting a fund with a higher environmental tilt. To explore this, we drill down further into environmental funds.¹⁶² We manually identify environmental funds by reading the summary prospectus of each ESG fund. We construct the “E-tilt” of each fund in a manner that is analogous to the ESG-tilt measures discussed above, but rather than using the ESG score of the companies in the fund’s portfolio, we use each provider’s environmental scores.¹⁶³ We then estimate a version of the regressions presented in Table 2, where the dependent variable is the environmental tilt of the fund rather than the ESG tilt, and the independent variable of interest is an indicator variable for the relevant type of environmental funds. While we think that this analysis is informative, we caution that the number of “environmental” funds in these regressions is small, which makes it difficult to draw firm conclusions from the regression analysis that follows. As such, we prefer to interpret it as providing suggestive evidence.

The results are presented in Table 4. While the results in Table 4 are slightly more mixed than those in Table 2, overall they indicate that environmental funds hold portfolios with higher environmental scores. Beginning with Panel A, the coefficients of interest in columns 1–2 and 5–8 are consistently larger (sometimes much larger) than their analogues in Table 2. The coefficients in those columns in Panel B, moreover, are about the same size as their Table 2 analogues. When Sustainalytics, S&P, or ISS scores are used, environmental funds tilt substantially more toward companies with high environmental ratings than comparable nonenvironmental funds.

The biggest difference is in columns 3 and 4. When using environmental scores constructed from TruValue Labs data, environmental funds identified based on name (column 3) have a slightly higher environmental tilt in their portfolios, although this difference is not statistically significant. Using funds identified by Morningstar, we find a null relationship—while the point estimates are negative, the t-statistics are quite small, indicating that the relationship is not distinguishable from zero. We note that TruValue Labs does not produce separate environmental scores; their emphasis is on SASB categories. As a result, our finding is likely due to our inability to construct an accurate environmental rating by identifying and aggregating the relevant SASB categories.

C. ESG Fund Voting Behavior

We turn next to the question of whether ESG funds vote the shares in their portfolio companies differently from non-ESG funds. Investors in ESG funds may expect fund managers to support ESG proposals, which may be opposed by management and conventional mutual funds. The SEC has explicitly suggested that some ESG funds fail to follow their stated proxy voting guidelines.¹⁶⁴ The failure of self-described ESG funds to vote differently from non-ESG funds has been compared to greenwashing, with critics calling out funds for not voting their proxies in accordance with their “values.”¹⁶⁵

We note that there need not be a conceptual connection between an ESG stock selection strategy and an ESG voting strategy. An ESG index fund may select stocks based on an ESG index, but the fund’s investment advisor may apply uniform voting policies to all the funds it manages. Those policies may or may not reflect ESG considerations. Indeed, several of the large asset managers are now incorporating ESG considerations into their voting decisions for both their ESG and non-ESG funds.¹⁶⁶

The likelihood of a connection between stock selection and voting also depends on the nature of a particular fund’s ESG mission. A fund that screens against investing in alcohol or tobacco companies for religious or ethical reasons, for example, might not be expected to vote differently from a non-ESG fund on climate change proposals or executive compensation. A fund that screens investments based on their carbon footprint, however, might also support shareholder proposals that seek to address climate change. On the other hand, a fund that has critically evaluated the environmental performance of its portfolio companies and invested in those companies based on the quality of that performance may not view the demands reflected in a shareholder proposal—demands like additional reporting or an explicit commitment to pro-environmental policies such as carbon neutrality—to be necessary.

Although these concerns make it difficult to evaluate the significance of a fund’s vote on a specific shareholder proposal, it is reasonable to ask whether ESG funds vote their proxies differently from non-ESG funds in general. We examine whether there is a difference in the frequency with which ESG funds vote against management relative to non-ESG funds. There are at least three reasons why we might expect them to do so. First, many ESG funds actively promote their engagement activities. That is, they claim to be seeking to persuade corporations to align their behavior with ESG values.¹⁶⁷ To the extent these engagement activities fail to persuade managers, we would expect such funds to disagree with management about issues with high ESG salience. Second, fund voting behavior might be more salient to investors in ESG funds than it is to investors in conventional mutual funds. ESG funds market themselves as advancing certain social goals, and their investors may expect the funds’ votes to align with those goals. This may lead ESG funds to vote against management more often. Finally, ESG funds might simply be more independent of management because they are operated by companies that are less likely to seek out 401(k) business from their portfolio companies, which is often argued to induce funds to toe the management line.¹⁶⁸

Without attempting to distinguish these causal stories, we investigate this question using the Voting Analytics (VA) database from ISS for 2018–2019.¹⁶⁹ The database contains more than twenty-five million fund-vote observations during those years on issues including director elections, corporate governance changes, say-on-pay votes, and shareholder proposals. For each issue, the database records the management-recommended vote as well as each fund’s actual vote.

Because there is no common identifying variable between VA and other mutual fund databases, we match VA with ESG funds based on names. Not every ESG fund appears in VA, and the name match appears unreliable in some cases, so our sample of 303 ESG funds is reduced to 231 for the purposes of this voting-pattern analysis.

We investigate whether ESG funds vote differently by regressing a variable indicating that the fund voted against management’s recommendation on a variable indicating that the fund is an ESG fund (under either of our definitions). In the first three regressions, models one through three, we use company-year dummy variables to control for the average characteristics of each portfolio company. This allows us to compare ESG funds’ votes with the votes of conventional funds at each individual company. This control is important because of the propensity of ESG funds—documented above—to hold different portfolios from conventional funds. In all regressions, standard errors are clustered at the fund level to address the correlation between votes coming from the same mutual fund.

In addition to the controls described above, in the first three regressions we include an indicator variable that takes the value 1 if the fund is part of an ESG family (more than 50% ESG funds based on the CRSP data) or 0 otherwise. This is important because mutual fund voting has historically been highly correlated at the family level, with many fund families voting in lockstep. By including separate variables to identify ESG funds and funds in ESG families, it is possible to determine whether ESG voting patterns are entirely driven by ESG-specialist fund families.¹⁷⁰

In models four through six, we replace the company-year dummy variables with dummy variables identifying unique combinations of companies and fund families in a particular year. Implicitly, this compares funds from the same family, voting at the same company in a given year. This provides additional robustness against the possibility that ESG fund support for ESG issues is driven solely by ESG-focused families.

Table 5 presents the results. Column 1 examines the relationship between classification as an ESG fund and the propensity to support shareholder proposals over management objections. Notably, when a shareholder proposal is included in the proxy statement, management virtually always recommends that shareholders vote against it.¹⁷¹ The results show that ESG funds are substantially more likely than other funds invested in the same company to oppose management by supporting shareholder proposals.

These results do not seem to be driven just by ESG fund families. The control variable “ESG Family” is positive and significantly associated with increased propensity to vote against management, but the ESG-fund indicator still has independent predictive value. In unreported regressions, the association between being an ESG fund and voting against management goes through even when all votes from funds in ESG families are dropped from the sample.

Column 2 examines the subset of ESG funds we identify as having an explicit environmental focus (“E” funds). Just as there is a tremendous amount of heterogeneity across ESG funds, ESG proposals include a wide variety of issues. One would expect funds focused on environmental issues to be particularly interested in environmental proposals and perhaps less interested in social or governance proposals. Once again, these tests focus on shareholder proposals with ESG salience, but this regression includes controls for funds with an explicit environmental focus and shareholder proposals that raise environmental issues. The results show that “E” funds are statistically no more or less likely than conventional funds to oppose management on shareholder proposals in general. When the shareholder proposals address environmental issues, however, “E” funds are far more likely than other funds to oppose management.

Finally, column 3 steps outside the shareholder-proposal context and looks at fund votes in uncontested director elections. Funds can express displeasure with management by withholding their votes from directors, even when there are no other candidates. At many companies, if a majority of votes are withheld, the director must resign the board seat.¹⁷² While shareholder proposals can have an impact when they are broadly supported, director elections have immediate, concrete consequences. Notably, fund votes against management recommendations in uncontested director elections are relatively rare.¹⁷³ The results in column 3 show that ESG funds vote differently from non-ESG funds in these elections, with ESG funds and funds in ESG families being far more likely to oppose directors. Overall, ESG funds are about twice as likely to withhold votes in an uncontested director election.

Columns 4 through 6 run the same set of regressions but with fixed effects at the firm × fund family × year level. We find that the results are robust to these controls and are not driven by family effects. In sum, we find substantial differences between the voting behavior of ESG and non-ESG funds. Although our results do not speak to the question of whether ESG funds vote against management or in favor of shareholder proposals “enough,” there is compelling evidence that they vote differently from their peers and that a typical ESG fund’s mission involves voting policies as well as stock selection.

D. Performance and Fees

Having established that investors in ESG funds seem to receive, on average, a portfolio tilted toward companies with higher ESG scores and distinctive voting behavior, we next ask what this “costs” them. As discussed, critics of ESG investing, particularly regulators, are concerned that investors in ESG funds may be sacrificing returns or taking on higher costs.¹⁷⁴ For example, the Trump administration’s DOL rule identified increasing risk or sacrificing returns to pursue “nonpecuniary” benefits as a major issue.¹⁷⁵ Empirical evidence about fund cost and performance can give context to the need for regulatory intervention.

We investigate this “cost” in two ways. First, we ask whether the fees charged by ESG funds are higher than those of comparable non-ESG funds. This represents a direct measure of the cost of choosing to invest in an ESG fund. Next, we consider a broader concept of “cost,” which includes both risk and opportunity cost. We therefore ask whether the returns offered by ESG funds differ systematically from those of comparable non-ESG funds. We also adjust these returns for risk and again look for differences between ESG and non-ESG funds.

To assess the fees of ESG funds, we regress expense ratios on our identifiers of ESG funds and present the results in Table 6. Unlike the tilt analyses, in this analysis we use fund class-year level observations, that is, one observation per fund share class per year.¹⁷⁶ We also include a series of additional control variables and fixed effects in the regressions. First, we include objective code × year fixed effects. As in the tilt regressions presented in Tables 2 through 4, this allows us to ensure that we are comparing apples to apples by comparing the expenses of funds with similar investment objectives at the same time. Because index funds are known to have lower fees than actively managed funds, we also control for whether a fund is an index fund.

In addition to these fixed effects, we include three different controls for size, since fund fees are known to vary systematically by size.¹⁷⁷ First, we include a control variable for the total net asset value of all funds managed by the fund manager. This allows us to account for larger asset managers’ ability

to spread costs over more total dollars invested. A larger fund may, for example, have more in-house ESG expertise at the manager level. Second, we control for the total net asset value of the fund by adding up the size of all the fund’s classes. This allows us to account for the fact that the managers of larger funds might be able to spread certain costs—such as portfolio-selection and administrative costs—across all the fund’s classes. Finally, we control for the total net asset value invested in the particular class itself. For all three variables, we use the natural logarithm of the size, as is common in the literature.¹⁷⁸ We cluster standard errors by fund (not class).

The results in columns 1 and 2 show no evidence that ESG funds—identified in either of the two ways—are more expensive, as measured by their expense ratios, than non-ESG funds. This suggests that while investors in these funds are receiving portfolios that have a greater ESG tilt, they are not paying for it through higher expenses on average. Column 3, in contrast, suggests that the ESG consideration funds identified by Morningstar do charge investors more than their competitors. Interestingly, these are funds that, according to the results in Table 3, do not exhibit higher ESG tilt than mutual funds generally.

Next we turn to returns, both raw and risk adjusted. There is a vigorous ongoing debate over the impact of ESG investing on economic returns.¹⁷⁹ Much of this debate has centered around individual companies. Scholars and practitioners have asked whether the stocks of companies with strong ESG scores are likely to outperform the market. By contrast, the performance we are interested in capturing here is not at the individual company level but at the mutual fund level. This is important because the general performance of high- and low-ESG stocks may not be fully informative about mutual fund performance.¹⁸⁰

We do not seek to settle the question of whether ESG investing is an advisable strategy here. ESG encompasses a vast number of potential strategies, and advocates point to benefits that they believe will accrue over a very long time horizon, both of which make empirical evaluation of ESG investing a thorny topic.¹⁸¹ Our goal is much narrower and more modest: to evaluate whether the empirical claims that underlie the DOL’s concerns about the inclusion of ESG funds in 401(k) plans are supported by the evidence. In other words, we look at ESG fund performance over our sample period for evidence suggesting that investors in such funds are bearing short-term costs in terms of reduced performance or increased risk.

To do so, we estimate similar regressions as the expense-ratio regressions presented in Table 6 to examine returns and risk. The results of these regressions are presented in Table 7. We use returns as the dependent variable in columns 1 through 3. In columns 4 through 6, we adjust these returns for risk by computing Sharpe ratios. An investment’s Sharpe ratio, defined as its return divided by its standard deviation, is a common risk-adjusted performance measure.¹⁸² Intuitively, the Sharpe ratio captures the incremental return that an investor receives per unit of risk. A higher Sharpe ratio therefore implies a higher risk-adjusted return. Because return data are available at the monthly level, we use fund class × month level observations and objective code × month fixed effects. As with Table 6, we control for objective codes¹⁸³ and whether the fund is an index fund using fixed effects, and we include the same three controls for size (manager, fund, and class).¹⁸⁴ We cluster the standard errors by fund and month.

The results in Table 7 suggest that investors in ESG funds do not give up returns—either raw or risk adjusted. If anything, investors in these funds might do a bit better, on average, than investors in non-ESG funds. Both returns and Sharpe ratios are higher for funds identified as ESG by their names (columns 1 and 4), and the point estimates are also positive for the funds identified by Morningstar, although the results are not statistically significant. Interestingly, the funds that consider ESG also do better, on average, than non-ESG funds, although the magnitudes are smaller (roughly half as large) than ESG-name funds.

As in the portfolio tilt analysis, where we looked specifically at environmental funds, we repeat our analyses of costs and performance focusing on two categories of funds. First, we investigate the differences, if any, between indexed ESG funds and actively managed funds with respect to fees and performance. Second, we investigate whether there are differences between “generic” ESG funds and specialized funds in terms of costs and performance. For the sake of brevity, we omit the tables showing these results.

We begin by splitting out indexed ESG funds from their actively managed competitors. We then repeat the analyses in Tables 6 and 7, this time including a variable indicating that a particular ESG fund is indexed. Because we are already including a variable to control for whether a fund is an index fund, adding in this new variable effectively allows us to answer the question: Apart from any difference we would expect to see between index funds and actively managed funds, do indexed ESG funds behave differently from actively managed ESG funds, in terms of either expenses or performance?

The answer, with respect to fees, is no. Whether identified by their names or by Morningstar, we find no statistically significant difference between actively managed and indexed ESG funds. Of course, this does not mean that indexed ESG funds charge the same fees as their actively managed counterparts. Instead, it simply means that their fees do not appear to be different from what we would expect, given all the other controls we include, from other index funds.

Turning to the performance analyses, we find that ESG index funds (whether identified by Morningstar or their names) perform slightly better than actively managed ESG funds. This incremental performance boost is statistically significant at the 5% level with respect to raw returns (the analogue to columns 1 and 2 in Table 7) and is marginally significant (i.e., significant at the 10% level) with respect to Sharpe ratios, or risk-adjusted performance. Again, this result is on top of any performance differential we see on average between actively managed funds and index funds.

What about highly specialized ESG funds? These are funds like the Fidelity Women’s Leadership Fund, or the LKCM Aquinas Catholic Equity Fund. While these funds have varied goals, purposes, and strategies, they all sell a product to investors that is much more specific than a “generic” ESG fund. There are a relatively small number of these funds in our sample relative to generic ESG funds, but they have received a disproportionate amount of attention. To study the fees and performance of these funds, we repeat the analysis presented in Tables 6 and 7 a third time, this time including a variable indicating that the fund in question is both an ESG fund (as identified either by its name or Morningstar, as appropriate) and that it is a highly specialized ESG fund. Including this variable allows us to investigate whether these highly specialized funds behave differently, in terms of fees or performance, than generic ESG funds, given all the other included controls.

While the small number of funds in this category suggests caution in interpreting our results, our findings are quite favorable for specialized funds. Beginning with fees, we find that, if anything, these specialized funds have lower expenses than either non-ESG funds or generic ESG funds, although this difference is only statistically significant when we identify funds using the Morningstar list. Turning to performance, we find no statistically significant differences in any of the four specifications (using raw returns or Sharpe ratios, whether ESG funds are identified by their names or with the Morningstar list). While the relatively small number of specialized funds in the analysis¹⁸⁵ means that these results should be interpreted with caution, there is certainly nothing to suggest that there is cause for concern, let alone regulatory intervention.

Overall, the results in this Section indicate that ESG funds, on average, do not cost investors more than comparable funds in terms of higher fees, reduced returns, or diminished risk-adjusted performance. To be clear, these results do not imply that ESG funds are a superior investment to ordinary funds, and the question of whether ESG investing outperforms conventional investment strategies is beyond the scope of this Article.¹⁸⁶ There is simply nothing in our results that suggests that ESG funds are worse than conventional funds when it comes to costs, returns, or risk. Given the DOL’s primary focus in its ERISA rulemaking on prudent decisionmaking, it is hard to see why ESG funds should be singled out for particular scrutiny.

ESG consideration funds present more of a puzzle. ESG consideration funds do not hold themselves out as using ESG as a central part of their investment thesis.¹⁸⁷ Instead, they typically commit only to considering it, often among many other factors, and this is typically only disclosed in their prospectuses rather than highlighted in their marketing materials. Their portfolios show no obvious ESG tilt relative to conventional funds, and their fees are higher, though their performance seems to offset these fees at least during our sample period. ESG consideration funds appear to be a distinct category of fund from “true” ESG funds. On the one hand, the very fact that these funds commit only to “considering” ESG—a commitment that may be buried in the fine print of the fund’s prospectus—should lower the level of concern these funds might pose to regulators. On the other hand, it remains unclear why these funds make such a commitment at all.¹⁸⁸

IV. Implications for Regulatory Policy

It is not surprising that the rapid expansion of ESG funds and the introduction of a significant number of new products have attracted the attention of regulators. Critics have questioned whether ESG funds live up to their names by providing investors with a different (and more sustainable) portfolio of investments than non-ESG funds.¹⁸⁹ They have chastised ESG funds for failing to vote and engage in a manner that is consistent with their ESG mission.¹⁹⁰ They have criticized asset managers for greenwashing—using an ESG name or describing an investment strategy as an excuse to charge investors higher fees.¹⁹¹ And the Department of Labor, in particular, has worried that fiduciaries will cause their beneficiaries to sacrifice returns by choosing or offering ESG-oriented investments.¹⁹²

However, our findings suggest that a regulatory response specifically targeting ESG funds is unwarranted. Specifically, our results stand in contrast to the criticisms of high costs, reduced performance, and greenwashing and generally point to a market that is working. Importantly, we do not argue that ESG investing is an optimal or even advisable strategy. There is a significant literature on ESG investing as an asset management strategy, and evaluating the long-term prospects of the numerous iterations of ESG is beyond the scope of this Article. Instead, we simply observe that we find no glaring evidence of problems in the ESG space from the perspective of the regulatory mandate of either the SEC or the DOL. The ESG sector of the fund market does not seem to be functioning worse than other parts of the mutual fund industry.

In this Part, we describe our results’ implications for the criticisms of ESG and current regulatory interventions and advance a suggested framework for regulators. We argue that regulators should adopt a presumption against ESG-specific interventions in the absence of clear evidence of ESG-specific problems. If there are issues with transparency around names or problems with fund costs, regulators should begin by questioning whether those issues are unique to ESG funds before making new rules targeting this particular market segment. Our results suggest that given the current state of the market, the answer to that question is generally “no.”

A. The Empirical Picture

As noted above, critics have identified numerous concerns about ESG mutual funds. We interrogate the empirical basis for these concerns and find it lacking. Although our analysis does not speak to the issue of whether ESG funds are a better investment option than non-ESG funds, either from an economic perspective or otherwise, we find that investors in ESG funds largely get what they pay for. ESG funds genuinely offer their investors portfolios and voting policies that differ from those of non-ESG funds and that seem generally aligned with ESG goals as measured by ESG ratings. At least in recent years, these funds do so without higher fees, lower returns, or increased risk. There is no evidence, in other words, that ESG funds are not performing on ESG-specific matters or that they are any worse than the rest of the mutual fund market on matters that are not ESG-specific.

To be fair, our analysis focuses on the past several years, and the market for ESG mutual funds is evolving rapidly. Both the number of ESG investing options and the total quantity of assets invested in ESG funds have exploded. The patterns we document may well be the result of increased competition in the market for ESG funds. They may also be a response to scrutiny by the financial media and regulators as well as the prospect of regulatory action for high fees or false claims. If the ESG market has improved, however, critics and regulators should note that improvement. Putative problems in the market from years ago should not drive today’s regulatory interventions. Focusing on the ESG market as it currently stands, many of the critiques of ESG lack empirical support. This could change as the market evolves, but taking the market as it is, there seems to be little need for action.

The role of third-party information providers in improving the market is notable. Morningstar and ESG ratings providers have constructed extensive disclosure mechanisms well beyond what regulations require.¹⁹³ These evaluations are inputs into our empirics, and our results should provide some comfort that this privately ordered system is providing useful information to investors.¹⁹⁴ While the proliferation of ESG ratings has drawn some criticism, it is clear that the market is responding to investors’ desire for more information about ESG factors without regulators stepping in.¹⁹⁵

Our observations are important because capital-market regulation poses ongoing challenges. Market participants place a premium on innovative new products, and regulators are often slow to identify and respond to market trends. Regulatory responses must be flexible enough to allow new developments while nonetheless protecting against market disruption, misallocation of capital, and outright fraud.

Although it is difficult to extrapolate from a single example, the characteristics we document with respect to today’s ESG funds offer a promising story for capital-market regulation. Existing disclosure requirements, market competition, third-party information providers, and the prospect of enforcement action appear to have offered investors sufficient protection against abuse in connection with the introduction of a new product. The regulatory regime that governed these funds, moreover, was generic: the funds were not subject to specialized, ESG-specific regulation during this time. Importantly, the mutual fund market is largely a retail market. As such, this protection did not depend on intermediation by more sophisticated investors.

B. The Pecuniary Benefits Debate

Much has been made of the possibility that ESG funds pursue social benefits at the cost of returns to investors. The distinction between collateral-benefit ESG and risk-return ESG drawn by Schanzenbach and Sitkoff¹⁹⁶ was part of the motivation for the DOL rule, and SEC commissioner Roisman cited the concern that funds might be sacrificing returns without explicit disclosure to investors.¹⁹⁷

If certain ESG funds are explicitly making decisions that sacrifice returns, that information should be disclosed to investors. Indeed, some funds disclose on their websites that their investment strategy might lead them to sacrifice returns.¹⁹⁸ Having done so, these funds provide fiduciaries with clear and explicit notice that their investment strategy might not be appropriate for an employer-sponsored pension plan under ERISA. There is no need for any sort of ESG-specific rule here: plan sponsors can straightforwardly apply standard fiduciary principles in light of this disclosure and reasonably exclude the fund from a 401(k) menu.

But the vast majority of ESG funds argue that ESG ultimately creates long-term value for investors: “doing well by doing good.” The claim that ESG can act as a hedge against long-term risks like climate change that may be excluded from conventional financial analysis is hard to test empirically, but the question of the long-term benefits of ESG investing would be far more pressing if there was evidence that investors are bearing short-term costs in pursuit of those benefits.

While our analysis clearly relies upon a relatively short evaluation window, we find nothing to contradict this claim. If anything, ESG funds perform a little better than other funds (net of costs) and cost about the same. Of course, these industry-level measurements are not substitutes for evaluating individual funds as suitable investment options. If, however, ESG funds as a category do not seem to be making short-term sacrifices, the case for subjecting them to special scrutiny, as the originally proposed DOL rule sought to do, seems weak even if one is skeptical of the long-term prospects of ESG factors.

The DOL should be conscious of a countervailing risk as well. If including ESG funds in retirement plans carries heightened risk for plan sponsors, such funds may simply be excluded from plan menus. ERISA fiduciary duties are backed by a private right of action, and plaintiffs’ attorneys have enjoyed success in a recent wave of 401(k) lawsuits alleging excessive fees.¹⁹⁹ This has led 401(k) plans to simplify and streamline their menus,²⁰⁰ often dropping high-fee options. Few will lament striking high-cost funds from plan menus,²⁰¹ but our results show that ESG funds offer something different from conventional funds without increased costs. Deterring plan sponsors from offering such funds risks harming investors financially if some version of the ESG investment thesis turns out to be correct. More broadly, many plan participants want options attuned to ESG issues, and offering these options may be a critical ingredient in encouraging younger employees to save for retirement.²⁰²

C. The Puzzle of ESG Consideration Funds

Finally, we are puzzled by the empirical picture of so-called ESG consideration funds. These funds do not foreground ESG values in their name or marketing but claim to consider ESG as part of their overall investing strategy in their prospectuses and other disclosures. These funds are more expensive than other funds (though their short-term performance is also better) and exhibit no discernible ESG tilt. These funds are also more numerous than “real” ESG funds.

Because ESG consideration funds do not aggressively market themselves as ESG funds, concerns about greenwashing or investors being misled are limited. If ESG consideration funds do not ultimately deliver much ESG tilt, they also do not hold themselves out as doing more than weighing ESG factors. On the other hand, if mutual fund complexes are touting their consideration of ESG to avoid challenges to their power to affect corporate decisionmaking or public retribution for the exercise of that power, the accuracy of their claims may be of concern.²⁰³

It is fair to ask, though, what ESG consideration funds are doing. While ESG funds often provide some clarity on their overall ESG strategy, ESG consideration funds leave investors (at least those who are somehow aware of the nature of these funds) guessing as to precisely how ESG factors come into the equation, if they do at all. Ironically, concerns about insufficient disclosure and high costs commonly directed at ESG funds would appear more suitably directed at ESG consideration funds that do not foreground their ESG commitment to investors.

D. The Diversity of ESG Ratings

Our results also shed light on the broader ESG ecosystem. Some critics have pointed to the variety and low correlation of ESG ratings as suggesting that ESG investing lacks discernible content.²⁰⁴ If the key ratings providers cannot agree about what is important or how to measure it, and ESG takes those ratings as an important input, how can ESG work as an investment strategy? If an investment strategy cannot be precisely characterized, what prudent person would choose to pursue it?²⁰⁵

From an investor point of view, however, it seems less important that ESG ratings agree about individual companies than that they have some consistency at the portfolio level. If we can agree on what counts as an ESG fund, then maybe there is less need for precision in determining what counts as an ESG company. This portfolio-level consistency is, in fact, what we find. While ratings are heterogenous, ESG funds tend to have higher ESG tilt across the ratings we measure.

This is consistent with two possibilities. ESG fund managers might be diversifying across ESG ratings in portfolio selection, so that they exhibit ESG tilt regardless of the ratings provider used to evaluate the fund. Alternatively, it may be that ESG fund managers are engaging in their own independent evaluations of companies so that their portfolios exhibit an aggregate commitment to ESG that the various ratings providers successfully measure, despite micro-level disagreement about individual portfolio companies.

Neither of these hypotheses is consistent with greenwashing, or even with a “lazy” form of ESG investing in which fund managers simply delegate portfolio management to ESG rating providers. Instead, it is most consistent with the possibility that fund managers take the information in these ratings into account when making their investing decisions—exactly what ESG funds have been promising.

Some commentators have criticized fund sponsors for taking advantage of a “low-regulation environment” to offer an opaque and “dizzying array of ESG products.”²⁰⁶ But, as with concerns over the heterogeneity of ESG ratings, why should we worry about numerous, divergent ESG funds when our results suggest that nearly all these funds provide a degree of ESG tilt with respect to at least one set of ESG ratings? There is no consensus about what ESG investing means or precisely how to do it, but ESG funds are doing something measurable and consistent with broad ESG ratings. To demand more precision is to hold the rapidly evolving ESG market to a standard that simply is not applied to other types of funds. There is not great consternation over disagreement about what a “growth” fund is, what counts as a “growth” stock, or that different “growth” funds might take materially different approaches. Why should ESG be singled out?

E. An ESG-Neutral Agenda for Regulators

Our results suggest that the market for ESG mutual funds is functioning reasonably well. Regulators should be responsive to that reality. It may be that the ESG fund market of, say, five years ago, which consisted of a much smaller number of funds, lacked sufficient competition and transparency to enable effective decisionmaking. But with ESG entering the mainstream, regulators are faced with a more mature, competitive market and ever-improving disclosure practices in which investors seem to be getting the information they need.

In our view, the most productive approach regulators can take when it comes to ESG funds is to adopt a presumptive stance of ESG neutrality. Many of the critiques of ESG funds might be made with equal force against other types of funds. When faced with a critique of ESG funds, regulators should ask first whether there is an empirical basis for singling out ESG funds or if the purported ESG issue is one that affects the entire fund market.

Notably, this is more or less the approach that the Department of Labor ultimately took in its rule on financial considerations in asset selection for retirement plans. The initial draft of the rule evinced considerable skepticism toward ESG strategies and emphasized that such funds could only be included in plans if fiduciaries conducted sufficient diligence to establish that such funds would ultimately generate an optimal trade-off of risk and return for investors.²⁰⁷ The DOL’s preface to the proposed rule cited many of the criticisms of ESG outlined above and was read by many in the industry to create a presumption against the use of ESG factors in retirement plans.²⁰⁸ This singling out of ESG sparked significant criticism from asset managers who viewed ESG as integral to portfolio management.

In the final version of the rule, the DOL all but abandoned explicit mention of ESG and instead focused on the types of diligence that prudent fiduciaries should conduct before selecting an investment option, regardless of the strategy.²⁰⁹ While the DOL emphasized the need to produce financial benefits through asset appreciation—rather than benefits to other constituencies—in the context of an ERISA plan, the ultimate rule did not explicitly target fiduciaries’ use of ESG factors.²¹⁰ Despite the rule’s ostensible neutrality, the DOL made its skepticism of ESG investing clear in the preamble, which warned that ESG investing “raises heightened concerns under ERISA.”²¹¹ Similarly, the SEC’s risk alert for ESG investing suggested that ESG products and services pose distinctive risks.²¹²

In our view, neutrality rather than special scrutiny is the correct approach. Take the Names Rule as an example. While including ESG in a fund name provides investors with limited information about a fund’s approach, the same is true of many other terms that are commonly used in fund names. “Growth,” “capital preservation,” and “blue-chip” all connote strategies in broad terms but are hardly concrete. It is plausible that investors would be served by increased disclosure around the use of these kinds of terms, and revising the twenty-year-old Names Rule may make sense. The vagueness of ESG names seems no worse to us, however, than other types of names suggesting investment strategies.

Conclusion

ESG investing is more important than ever, and ESG mutual funds are the primary mechanism by which ordinary people can engage in ESG investing. Regulators, academics, and the financial media have raised concerns, however, about the growth of ESG mutual funds and the substantial asset flows into such funds. These concerns include that labeling a fund with the term “ESG” is potentially misleading, that asset managers may be exploiting investor demand to charge excessive fees, that an orientation toward ESG sacrifices economic value, and that ESG funds do not vote their proxies in accordance with their values. These concerns have already resulted in new DOL regulation as well as the prospect of further regulatory intervention by the SEC. The prospect of regulation targeted specifically at ESG mutual funds will likely make it more difficult for mutual fund sponsors to provide these investment options and for employers to include ESG mutual funds in their 401(k) plans.

We collected data on ESG funds and provided a framework for interrogating these concerns. Our empirical results provide no justification for regulatory invention. Simply put, our analysis reveals that ESG funds do not currently present distinctive concerns from either an investor-protection or a capital-markets perspective. Funds that market themselves as employing an ESG investment strategy invest and vote differently from funds that do not purport to do so. ESG funds do not appear to be charging investors higher fees or sacrificing returns relative to their traditional counterparts. Our findings suggest caution in curbing the marketing of ESG products or limiting their use by ERISA fiduciaries.

* Professor, University of Virginia School of Law.

** Saul A. Fox Distinguished Professor of Business Law, University of Pennsylvania Carey Law School.

† Honourable Justice Frank Iacobucci Chair in Capital Markets Regulation and Associate Professor of Law and Finance, University of Toronto Faculty of Law. We are grateful to ISS, S&P, and TruValue for sharing their ESG ratings with us. We thank participants at our pop-up virtual workshop, the American Law and Economics Association Annual Meeting, our presentation at the Investment Company Institute, the Goethe University DFG Center for Advanced Studies Foundations of Law and Finance (LawFin) Seminar, and the Corporate Law Academic Webinar Series for their many helpful comments and suggestions. Audrey Youn, Logan Hale, and Shannon Brooks provided superb research assistance.