Shareholder voting has undergone a remarkable transformation over the past few decades. Institutional ownership of shares was once negligible; now, it predominates. This is important because individual investors are generally rationally apathetic when it comes to shareholder voting: value potentially gained through voting is outweighed by the burden of determining how to vote and actually casting that vote. By contrast, institutional investors possess economies of scale, and so regularly vote billions of shares each year on thousands of ballot items for the thousands of companies in which they invest.

For example, an investor purchasing a share of an S&P 500 index mutual fund would likely have no interest in how each proxy is voted for each of the securities in each of the companies held by that fund. Indeed, it would defeat the purpose of selecting such a low-maintenance, lost-cost investment alternative. And so it is left to the investment adviser to the index fund to vote on the investor’s behalf. This enhanced reliance on the investment adviser to act on behalf of investors inevitably results in a classic agency problem: how do we make sure that the investment adviser is voting those shares in the investor’s best interest, and not the adviser’s?

The Rise of Proxy Advisory Firms

The Commission took up this very issue in a rulemaking in 2003, putting in place disclosures to inform investors how their funds’ advisers are voting, as well as outlining clear steps that advisers must undertake to ensure that they vote shares in the best interest of their clients. But every regulatory intervention carries with it the risk of unintended consequences. And the 2003 release has since proved that to be true—to the point where the costs of the unintended consequences now arguably dwarf those benefits originally sought to be achieved. How exactly did this happen?

Proxy Voting by Investment Advisers

In the 2003 release, the SEC took on one specific manifestation of the general agency problem discussed above: that an adviser could have a conflict of interest when voting a client’s securities on matters that affect the adviser’s own interests (e.g., if the adviser is voting shares in a company whose pension the adviser also manages). To remedy this issue, the release stated that an investment adviser’s fiduciary duty to its clients requires the adviser to adopt policies and procedures reasonably designed to ensure that it votes its clients’ proxies in the best interest of those clients. Further, the Commission noted that “an adviser could demonstrate that the vote was not a product of a conflict of interest if it voted client securities, in accordance with a pre-determined policy, based upon the recommendations of an independent third party.” From these statements, two specific unintended consequences arose.

First, some investment advisers interpreted this rule as requiring them to vote every share every time. This seemed, perhaps, to be the natural outgrowth of the Department of Labor’s 1988 “Avon Letter,” which stated that “the fiduciary act of managing plan assets which are shares of corporate stock would include the voting of proxies appurtenant to those shares of stock.” As a result, investment advisers with investment authority over ERISA plan assets—and thus regulated by the Department of Labor as well as the SEC—were already required to cast a vote on every matter. Reading the SEC’s 2003 rule, some advisers may have assumed that the Commission intended to codify that result for all investment advisers.

A requirement to vote every share on every vote, however, gives rise to a significant economic burden for investment advisers who may own only relatively small holdings in a large number of companies. For example, one study found that “most institutional investor holdings are relatively small portions of each firm’s total securities. For example, in our sample … the mean (median) holding of an individual stock by institutional investors is 0.3% (0.03 %).” Given that institutional investors hold stock in hundreds or thousands of companies (for example, TIAA‐CREF holds stock in 7,000 companies), institutional investors—particularly the smaller ones—may not be able to invest in the costly research needed to ensure that they cast each vote in the best interest of their clients. The logical answer is to outsource the research function to a third party, who could do the needed research and sell voting recommendations back to investment advisers for a fee: a proxy advisory firm. While these firms already existed, the 2003 rule gave advisers new economic incentives to use them.

Second, proxy advisory firms noticed the suggestion in the 2003 rule that soliciting the views of an independent third party could overcome an adviser’s conflict of interest. In 2004, a proxy advisory firm requested—and received—“no-action” relief from the SEC staff that significantly expanded investment advisers’ incentive to use these firms. Specifically, the staff advised Institutional Shareholder Services (“ISS”) that “[A]n investment adviser that votes client proxies in accordance with a pre-determined policy based on the recommendations of an independent third party will not necessarily breach its fiduciary duty of loyalty to its clients even though the recommendations may be consistent with the adviser’s own interests. In essence, the recommendations of a third party who is in fact independent of an investment adviser may cleanse the vote of the adviser’s conflict.” Thus, rotely relying on the advice from the proxy advisory firm became a cheap litigation insurance policy: for the price of purchasing the proxy advisory firm’s recommendations, an investment adviser could ward off potential litigation over its conflicts of interest.

Finally, in a second 2004 no-action letter to Egan‐Jones, the staff affirmed that a key aspect of some proxy advisory firms’ business model—selling corporate governance consulting services to companies—“generally would not affect the firm’s independence from an investment adviser.” This determination is somewhat incredible, as it places the proxy advisory firm in the position of telling investment advisers how to vote proxies on corporate governance matters that had been the subject of the proxy advisory firm’s consulting services—a seemingly obvious, and insurmountable, conflict of interest.

In sum, the 2003 release and the 2004 no-action letters set the stage for proxy advisory firms to wield the power of the proxy, through investment adviser firms that had economic, regulatory, and liability incentives to rotely rely on the proxy advisory firms’ recommendations and through the SEC staff’s assurances that this arrangement was just fine, despite the obvious conflicts of interest involved throughout. But it would take some additional developments for proxy advisory firms to attain the dominant voice in American corporate governance that they have today.

Subsequent Developments

Since 2003–2004, some features of the SEC regulatory regime have acted to deepen investment advisers’ reliance on proxy advisory firms. First, the quantity of company disclosures has increased significantly over the past few years. For example, the SEC in 2006 adopted revisions to the proxy and periodic reporting rules to require extensive new disclosures about “executive and director compensation, related person transactions, director independence and other corporate governance matters and security ownership of officers and directors.” The new rule generated reams of new disclosures that were long, complex, and focused on regulatory compliance rather than telling the company’s compensation story. The sheer volume of information that an investment adviser would have to review in order to make a fully-informed voting decision is difficult even to organize, much less to read and digest.

Second, the average number of items on which investors are asked to vote has also been on the rise. This trend is attributable at least in part to the Dodd‐Frank twin advisory votes on executive compensation: a vote for how often to approve executive pay (“say-on-frequency”), and a vote to in fact approve (or disapprove) that pay (“say-on-pay”). We have also seen a continued increase in shareholder proposals that SEC rules generally compel companies to include in the proxy to be voted on, which in turn reflects increased activism around shareholder voting.

As a result, the economic imperative to use proxy advisory firms that the vote-every-share-every-time interpretation of the 2003 rulemaking created has only deepened over time. At the same time, serious questions emerged, particularly in the corporate community, about the power being wielded by proxy advisory firms in making their recommendations. These recommendations are of course provided contractually to investment advisers; proxy advisory firms have no fiduciary duty to shareholders, nor do they have any interest or stake in the companies that are the subject of the recommendations.

In particular, corporate observers raised two key questions about proxy advisory firms: are their recommendations infected by conflicts of interest, and even assuming they are not, do they have the capacity to produce accurate, transparent, and useful recommendations?

With regard to the former question, as alluded to in the Egan-Jones no-action letter, proxy advisory firms may have other, complementary lines of business. For example, in addition to selling vote recommendations to institutional investors (along with voting platforms, data aggregation, and other auxiliary services), they may also sell consulting services to companies that want to ensure that they have structured their governance and other proxy votes so as to avoid “no” recommendations from the proxy advisory firms. The sale of voting recommendations to institutional investors creates a risk that proxy advisory firms, in formulating their core voting recommendations, will be influenced by some of their largest customers (e.g., union or municipal pension funds) to recommend a voting position that would benefit them. The sale of consulting services to companies creates a risk that proxy advisory firms would be lenient in formulating voting recommendations for companies that are their clients and harsh in crafting the recommendations for those companies that have refused to retain their services.

With regard to the latter question, proxy advisory firms themselves face the same difficulties as institutional investors faced before they determined to outsource their voting: how does one formulate timely, high-quality recommendations for thousands of votes at thousands of companies based on millions of pages of data—all while competing on price with other firms? To put it charitably, they just do the best they can. But their best often is simply not good enough: proxy advisory firms publish some recommendations that are based on clear, material mistakes of fact. Moreover, they base some recommendations on a cookie-cutter approach to governance—i.e., in favor of all proposals of a certain type, like de-staggering boards or removing poison pills, even if there is a sound basis for challenging the assumption that an otherwise beneficial governance reform might not be appropriate for a given company. As one academic article has argued:

Unfortunately companies have little access to proxy advisory firms in order either to correct a mistake of fact, or to explain why a generic corporate governance recommendation is the wrong result in the specific instance: letting companies appeal to the advisory firm is time-consuming and expensive, neither of which is consistent with the proxy advisory firm’s business model. As a result, while the companies that also hire a proxy advisory firm for its corporate consulting service may have some minimal degree of access (e.g., by being provided an opportunity to make limited comments on draft reports), smaller companies that are not clients generally are not afforded any such rights.

Advisers that rely rotely on the proxy advisory firm’s recommendations also tend not to afford companies an opportunity to tell their story. This is unsurprising: if the advisers wanted to make contextualized decisions about casting each vote, they would not have outsourced their vote in the first place. But it is also supremely ironic: a company that may want to engage in good faith with its shareholders may find that it has no meaningful opportunity to do so. This trend is deeply troubling to me. If an investment adviser is approached by a company with information indicating that the basis on which the adviser is casting its vote is fundamentally flawed, is it really consistent with the investment adviser’s fiduciary duties for the adviser to simply ignore that information? I think the rote reliance on proxy advisory firms has caused investment advisers to lose the forest for the trees: they are so focused on checking the compliance boxes to absolve conflicts of interest under our rules that they forget that they still have a broader fiduciary duty to investors to cast votes in the investors’ best interest. That fiduciary duty, I believe, cannot be satisfied through rote reliance on proxy advisory firms.

Regulatory Response

First Steps

These issues have been on the SEC’s radar for some time now, most notably when they were raised in the 2010 Concept Release on the U.S. Proxy System (the “Proxy Plumbing” release). This release outlined the conflict-of-interest and low-quality voting recommendation issues addressed above, and it requested comment on a long list of potential regulatory solutions. I raised this issue in a number of speeches in 2013 and 2014, and the Commission in December 2013 held a roundtable to examine key questions about the influence of proxy advisers on institutional investors, the lack of competition in this market, the lack of transparency in the proxy advisory firms’ vote recommendation process and, significantly, the obvious conflicts of interest when proxy advisory firms provide advisory services to issuers while making voting recommendations to investors. A wide range of other parties, including Congress, academia, public interest groups, the media, and a national securities exchange, have also been calling for reforms.

There has also been substantial interest and work regarding the role of proxy advisers on the international front. Recently, the European Commission introduced legislation to address the accuracy and reliability of proxy advisers’ analysis as well as their conflicts of interest. If adopted by the EU’s legislature, Article 3i (entitled “Transparency of proxy advisors”) would require proxy advisors to publicly disclose certain information in relation to the preparation of their recommendations, including the sources of information, total staff involved, and other meaningful data points. It would also require that member states ensure that proxy advisers identify and disclose without undue delay any actual or potential conflicts of interest or business relationships that may influence their recommendations and what they have done to eliminate or mitigate such actual of potential conflicts. While I may not often find myself in a position of agreeing with the European Commission, here I believe their proposal takes an incredible step forward and one that I commend them for promoting.

Staff Legal Bulletin No. 20

After the concept release and the roundtable, which provided a wealth of information and perspectives, the SEC staff on June 30th moved toward addressing some of the serious issues. The Division of Investment Management and the Division of Corporation Finance released Staff Legal Bulletin No. 20 (“SLB 20”), providing much-needed guidance and clarification as to the duties and obligations of proxy advisers, and to the duties and obligations of investment advisers that make use of proxy advisers’ services.

This guidance is a good initial step in addressing the serious deficiencies currently plaguing the proxy advisory process. In particular, it does three important things worth highlighting.

First, it clarifies the widespread misconception discussed above that the Commission’s 2003 release mandates that investment advisers cast a ballot for each and every vote. The guidance makes clear that this interpretation is wrong. Rather, an investment adviser and its client have significant flexibility in determining how the investment adviser should vote on the client’s behalf. The investment adviser and client can agree that votes will be cast always, sometimes (e.g., only on certain key issues), or never. They similarly can agree that votes will be cast in lockstep with another party (e.g., management, or a large institutional investor). Advisers could agree with investors in a mutual fund managed by the adviser that the adviser would only vote shares in companies representing more than a certain threshold percentage of the fund’s assets—and refrain from voting smaller holdings, vote them with management, or vote them some other way. While possibilities may not be endless, there is room for much more creativity than exists today.

Second, SLB 20 cautions against misguided reliance on the two 2004 staff no-action letters, which have been widely misinterpreted as permitting investment advisers to abdicate essentially all of their voting responsibilities to proxy advisers without a second thought. The guidance makes clear that investment advisers have a continuing duty to monitor the activities of their proxy advisers, including whether, among other things, the proxy advisory firm has the capacity to “ensure that its proxy voting recommendations are based on current and accurate information.” I have heard from many companies that proxy advisory firms sometimes produce recommendations based on materially false or inaccurate information, but they are unable to have the proxy advisory firm even acknowledge these claims, much less review them and determine whether to revise its recommendation in light of the corrected information.

While I encourage companies to attempt to work with proxy advisers, I also believe it is important for companies to bring this type of misconduct by proxy advisers to the attention of their institutional shareholders. As explained in the new guidance, investment advisers are required to take reasonable steps to investigate errors. Repeated instances of proxy advisers failing to correct recommendations they based on materially inaccurate information should cause investment advisers to question whether the proxy adviser can be relied upon. Separate and apart from the guidance they receive, I believe investment advisers’ broader fiduciary duty should compel them to review the corrected information provided by the company and consider it when determining how ultimately to cast their votes.

Third, SLB 20 makes clear that a proxy advisory firm must disclose to recipients of voting recommendations any significant relationship the proxy advisory firm has with a company or security holder proponent. This critical disclosure must clearly and adequately describe the nature and scope of the relationship, and boilerplate will not suffice.

Further Interventions?

While these reforms are much-needed, I am concerned that the guidance does not go far enough. SLB 20 provides some incremental duties and suggests ways that individual entities could structure their advisory relationship so as to reduce reliance on proxy advisory firms, but it has become clear to me that, over the past decade, the investment adviser industry has become far too entrenched in its reliance on these firms, and there is therefore a risk that the firms will not take full advantage of the new guidance to reduce that reliance.

I therefore intend to closely monitor how these reforms are being executed and whether they are solving the current significant problems in this space. In fact, if a company does experience difficulties in getting the proxy advisory firm to respond to the company’s concerns about the accuracy of the information on which the recommendation is based, and does therefore follow my suggestion to reach out directly to its institutional investors, I would encourage the company also to provide a copy of its shareholder communications directly to my office. I would be very interested to learn which complaints are being disregarded by proxy advisory firms and institutional investors. In addition, I believe SLB 20 should diminish the number of these complaints over time, and I will be very interested to discover whether this is in fact the case.

Finally, while I appreciate the important steps that are being taken above, I believe that the release of SLB 20 still may not fully address the fact that our rules have accorded to proxy advisors a special and privileged role in our securities laws—a role similar to that of nationally recognized statistical ratings organizations (“NRSRO”) before the financial crisis. I intend to continue to seek structural changes that will address this dangerous overreliance.

For example, the Commission could replace the two staff no-action letters with Commission-level guidance. Such guidance would seek to ensure that institutional shareholders are complying with the original intent of the 2003 rule and effectively carrying out their fiduciary duties. Commission guidance clarifying to institutional investors that they need to take responsibility for their voting decisions rather than engaging in rote reliance on proxy advisory firm recommendations would go a long way toward mitigating the concerns arising from the outsized and potentially conflicted role of proxy advisory firms.

In addition, as I have stated in the past, I believe that the Commission should fundamentally review the role and regulation of proxy advisory firms and explore possible reforms, including, but not limited to, requiring them to follow a universal code of conduct, ensuring that their recommendations are designed to increase shareholder value, increasing the transparency of their methods, ensuring that conflicts of interest are dealt with appropriately, and increasing their overall accountability. I do not believe that the Commission should be in the business of comprehensively regulating proxy advisory firms—as we’ve seen from the 2006 NRSRO rule, such regulation often is simply ineffective—but there may be additional steps that we can take to promote transparency and best practices.

In Sum

To be clear, I realize that proxy advisers can provide important information to institutional investors and others. But that business model should be able to stand or fall on its own merits—i.e., based on the usefulness of the information provided to the marketplace. The SEC’s rulebook should not accord proxy advisory firms a special, privileged role—or, if that privilege cannot be completely stripped away, proxy advisory firms should be subject to increased oversight and accountability commensurate with their role.