• The prevailing paradigm for voting institutional investor portfolio shares consists of delegating the function to corporate governance specialists (housed either internally or externally at proxy advisory firms), except, in the case of active investment managers, where votes with perceived economic significance are often decided with input by portfolio managers.  
  • To make this “outsourcing” paradigm economically viable in light of the thousands of votes required to be cast each proxy season, corporate governance specialists have automated their voting systems by adopting one-size-fits-all voting policies and metrics, which are applied to all companies without regard to their particular circumstances.  
  • The one-size-fits-all voting policy paradigm has been criticized on a number of bases, including the lack of convincing empirical evidence that the voting policies have any appreciable positive effect on corporate performance, the paradigm’s rigidity and its empowerment of ISS and Glass Lewis as arbiters of corporate governance issues.  
  • The 2011 Say on Pay voting season illustrates the worst faults of the one-size-fits-all paradigm and has set the stage for ISS to become the veritable dictator of executive compensation policies and practices for all public US companies.  
  • We believe there is an alternative institutional voting paradigm that would avoid the problems inherent in the prevailing voting model. In 2008, the Department of Labor revisited the ERISA fiduciary requirements relating to voting portfolio shares and made clear there is no absolute duty to vote all portfolio shares on all matters; rather ERISA fiduciaries should first conduct a cost benefit analysis with regard to exercise of the shareholder franchise and vote only if the cost of doing so is outweighed by the value created by the vote.  
  • This elaboration of the fiduciary standard makes clear that institutional investors could well conclude that a voting policy of following management’s voting recommendations in all cases (except where an investor affirmatively concluded a contrary vote would better serve shareholder value creation) would be a far better and lower cost solution, compared to the current corporate governance dominated paradigm.
    • The new paradigm would align institutional investor votes with portfolio investment policy. To put it most simply, if a stock is in an actively managed portfolio, presumptively the investment manager has confidence in the company’s management. If the manager lacks that confidence, it should be exiting the position, not holding it in the hopes that some day corporate governance reform will make a positive difference.  
    • It would take quantitative institutional investors out of an area in which they don’t belong of making subjective voting decisions for portfolio companies selected on the basis of a quantitative model, not subjective investment judgment.
  • It would permit elimination of most, if not all, of the corporate governance infrastructure and associated costs, including fees for proxy advisory services.
  • It would relieve public companies of the costs, burdens and distractions of the current institutional voting paradigm, which are significant and largely ignored by the corporate governance community.  

The Prevailing One-Size-Fits-All Voting Policy Paradigm

A year ago, we published a Corporate Governance Commentary titled “Future of Institutional Share Voting: Three Paradigms.” We began by observing that the prevailing paradigm for institutional investors voting of portfolio company shares is to delegate all but the most obvious economically related voting decisions to either an internal or external corporate governance team that is largely, or all too often totally, separate from the investment policy decision making team—in effect, a parallel universe of voting decision makers. Because of the huge number of voting decisions facing institutional investors, the prevailing corporate governance methodology for deciding how to vote portfolio shares is to apply formulaic voting policies that push all portfolio companies, no matter how different their particular circumstances, through a uniform one-size-fits-all voting mold.

We also noted in our Three Paradigms Commentary, and in our subsequent Corporate Governance Commentary titled “Proxy Advisory Business: Apotheosis or Apogee,”2 that the prevailing one-size-fits-all voting policy paradigm serves several important goals of the institutional investor community.  

  • It is perceived to meet the fiduciary duty obligations of institutional investors with respect to voting portfolio shares—which is commonly assumed to require that all shares be voted on all matters.  
  • It deals with the cost implications of voting all portfolio shares on all matters by making the decisions as automated as possible and by reducing and simplifying the amount of informational input needed to apply the appropriate voting policy.  
  • Its relative cost effectiveness allows institutional investors to spend as little money as practicable on voting decisions to which they attach little or no economic or financial importance.3  

On the other hand, we observed in our Three Paradigms Commentary and our Apotheosis or Apogee Commentary that a number of considerations militated against continued use of the currently prevailing voting policy paradigm, including:  

  • The lack of convincing empirical evidence that the voting policies espoused by the corporate governance community create economic value.  
  • The failure of the voting policy model to make any allowances for the vastly different circumstances at the seven thousand or so listed US portfolio companies, not to mention the additional thousands of less actively traded companies.  
  • The growing concern on the part of public companies and their advisors with the power of the two proxy advisory firms and their lack of accountability for their voting recommendations.  
  • A 2008 interpretation by the Department of Labor (DOL) of the implementation of the fiduciary duty of ERISA advisors to vote portfolio shares, and the possibility of a similar interpretation by the SEC.4

Implications of 2011 Say on Pay Advisory Votes: Proxy Advisors Rule

This year saw the first proxy season in which there was a universal requirement for a Say on Pay advisory vote at all public companies with a meaningful market capitalization.

The 2011 Say on Pay advisory vote experience makes clear that the executive compensation standards used by ISS, and to a lesser extent those of Glass Lewis, will very shortly become dominant in the corporate universe.5  

  • The difference between receiving a favorable recommendation from ISS and an unfavorable one, on average, was a swing of approximately 25 percent of votes cast. Glass Lewis’ recommendations seemed to produce a 5 percent swing in votes cast.  
  • Companies receiving a negative proxy advisory recommendation from ISS averaged less than a 70 percent positive shareholder vote, compared to those receiving positive proxy advisory recommendations which routinely scored 90 percent or higher positive shareholder votes.  
  • It is in the cards that the proxy advisory firms will begin basing withhold vote recommendations at subsequent director elections for companies with a lower than 70 percent vote on Say on Pay (perhaps after a year to allow for company adherence to the proxy advisor’s executive pay policies).6  

In sum, the over-riding lesson of the 2011 Say on Pay season is that companies have two practical choices in dealing with Say on Pay votes in the future.  

  • The first is to try harder to explain to their investors why a board’s executive pay policies that run afoul of a proxy advisor’s model nevertheless are appropriate in the company’s particular circumstances. The expectation would be that, by focusing on clarity and conciseness of presentation, investors would “get it” and opt out of the tyranny of a one-size-fits-all voting policy and accompanying executive compensation metrics of a proxy advisory firm’s devising. The goal of the effort would be to achieve a sufficient positive vote from shareholders to avoid the stigma of a “vote of no confidence” or a “yellow card/red card” in the emerging ISS parlance.  
  • The other choice, of course, would be for a board to tailor its executive compensation programs to ISS metrics, to game the system so to speak.7 Compensation committees and boards taking this tack will be adhering to the time honored and too often effective principle of “going along to get along.” In the view of these boards, if you “can’t beat the system, you might as well join it.”  

Neither of these choices is satisfying on a theoretical level. They illustrate the irreconcilable dilemma of trying to squeeze the variety and complexity of thousands of companies’ circumstances and pay policies into the rigid mold of a one-size-fits-all governance model driven largely by ideological principles and lowest cost methodologies.  

More important, on a practical level, is the probability that, over time, most boards will pick the far easier and non-controversial route of tailoring compensation policies to ISS metrics, rather than the higher visibility, higher cost, higher risk route of trying to convince their shareholders that ISS “got it wrong.” The end result almost certainly is going to be the practical hegemony over pay policies and practices by ISS and, to a lesser extent, Glass Lewis. As so many predicted when Say on Pay was being debated, the outcome of mandatory Say on Pay advisory votes will be the ascendency of the proxy advisory firms’ executive compensation models, whether or not the proxy advisors have any expertise or knowledge about executive compensation, whether or not their executive compensation metrics are well founded conceptually and fairly and accurately applied in practice and whether or not those metrics are at least more often than not applicable to specific companies facing specific issues in terms of management retention, management incentives and shareholder value creation.  

In sum, Say on Pay advisory voting demonstrates the strengths and weaknesses of the one-size-fits-all voting policies paradigm. On a superficial level it works, it is far less expensive than a paradigm that would require specific company situations to be taken into account and it enhances the power and prestige of the activist corporate governance community. On the other hand, the paradigm clearly imposes economic costs on institutional investors, forces portfolio companies to live under the tyranny of its one-size-fits-all voting policies and saddles corporate America with an increasing number of corporate governance and pay policies that lack a convincing connection to the creation of shareholder value.

Two Alternate Paradigms

It is, of course, one thing to point out weaknesses in the prevailing paradigm of one-size-fits-all voting policies. It is quite another to suggest a theoretically more defensible but nevertheless pragmatic substitute. We advanced two alternate paradigms in our Three Paradigm Commentary.

  • Our first alternate paradigm posited a major revision of the conventional US model of shareholder democracy — one that would solve the problem of too many votes on too many issues of too little economic consequence by vastly streamlining the voting system — including reducing the number and frequency of director votes, eliminating all 14a-8 and other advisory votes and moving to biennial or triennial shareholder meetings. The obvious difficulty with this paradigm is its very radical reformulation of our hoary tradition of annual shareholder meetings and its abrupt about-face from widely held and popular notions of corporate governance, such as annual elections for all directors and other policies intended to increase director “accountability” to shareholders.  
  • The second alternate paradigm we advanced would be to force the convergence of the parallel universes of investment decision making and voting decision making at the institutional investor level and to restructure the proxy advisory function into a utility for all shareholders (retail and institutional) that would base its recommendations on a case-by-case, rather than a one-size-fits-all, analysis. The obvious difficulty with this paradigm is that it would cost far more than the current voting policy paradigm, a cost investors clearly would not want to bear unless pressured into it by a far more rigorous application of fiduciary duty standards than those currently prevailing in the industry or some other regulatory intervention.  

Is There a Practical “Fourth” Paradigm?

As we have frequently noted, the current system for voting portfolio securities by application of uniform voting policies to the tens of thousands of votes cast each year during proxy season works for three simple reasons.  

  • It is perceived as successfully addressing the commonly understood fiduciary duty of institutional investors to vote all of their portfolio securities on all matters.  
  • It does so at a relatively low cost, because it is based on one-size-fits-all voting policies and simplistic metrics.  
  • It answers the drive for power and prestige of the separate parallel universe inhabited by corporate governance specialists who effectively run the voting mechanics for most institutional investors, a parallel universe which, in the absence of the current system, would have far less reason to exist.8

Cutting this Gordion knot is not easy, as our prior attempts in our Three Paradigms Corporate Governance Commentary illustrated. There may, however, be another approach. We commented at length in our Apotheosis or Apogee Corporate Governance Commentary on a 2008 Department of Labor interpretation of the ERISA prudent man standard, as applied to voting of portfolio securities by ERISA funds.9 The interpretation rejects the commonly understood fiduciary requirement that an ERISA fiduciary has a duty to vote all portfolio shares on all matters. Rather, the DOL interpretation makes clear that an ERISA fiduciary’s first determination must be a cost-benefit analysis with regard to whether more value would be created by voting than by not voting, after taking into consideration the actual cost of voting, including the systemic costs of maintaining a vote determining process and, presumably, the costs imposed on portfolio companies to deal with corporate governance activism.10 Only if the fiduciary is satisfied that voting is economically preferable to not voting, does the issue become how to vote.

The 2008 DOL interpretation is a potential game-changer for institutional voting paradigms, because it makes clear that, if available, adoption of a simpler and cheaper voting paradigm would be justified, indeed required, for most institutional investors. Only those which affirmatively determined, through a process meeting the fiduciary duty of due care, that the prevailing one-size-fits-all voting paradigm added more shareholder value at its portfolio companies than its all-in economic cost would be entitled to adopt a one-size-fits-all voting policy paradigm.11 All other institutional voters would have no justification as fiduciaries to burden plan beneficiaries and portfolio companies with the cost of the prevailing voting paradigm.12  

This more nuanced analysis of the fiduciary duty of institutional investors with regard to portfolio share voting strongly suggests, if it does not demand, adoption of a different voting paradigm that would better serve the fiduciary duty imposed on institutional investors by ERISA and by the Investment Advisers Act — namely, simply not to vote portfolio shares except where the institutional manager affirmatively believed it would improve its investors’ all-in economic returns by voting. Under this paradigm, the default position would be not to vote portfolio shares. Obvious exceptions would be M&A transactions, restructurings and recapitalizations, change of control contests and the like.

The problem with this articulation of a new voting paradigm is that not voting portfolio shares could have a significant negative economic consequence. Not voting would make the conduct of many shareholder meetings difficult, and sometimes impossible, because of its adverse effect on quorum requirements. It is hard to see how institutional investors could reasonably conclude that inability to conduct business at shareholder meetings because of failed quorums is good for their portfolio values.13  

There is a way around the failed quorum conundrum. Shareholder value creation would remain the key determinant of whether and how to vote portfolio shares. However, rather than not voting on ballot issues the manager did not think had economic significance, the investment manager would adopt a simple, lower cost default principle of voting in favor of management so long as it believed the company was performing well.  

A default voting position of following management’s voting recommendations would allow institutional investors to disband much, if not all, of their internal corporate governance function, eliminate proxy advisory fees and contract voting execution mechanics to the cheapest provider. It would likewise allow companies to disband that part of their corporate governance infrastructure that is required to service the “needs” of the institutional corporate governance community, including the new favorite of a 5th Analyst Call, other forms of corporate governance “engagement” and mini-proxy contests to counter negative vote recommendations from the proxy advisory firms.

Under our proposed Fourth Paradigm, if a company is not performing well in the view of a manager’s investment decision makers because of governance failures, the principal issue would naturally be whether to sell the stock and move on, or whether to hold the investment in anticipation either of management fixing the problem or some group of investors banding together to use the corporate franchise to fix the problem.14 In any case, it would be appropriate for the portfolio manager to vote against management’s recommendations on ballot issues relevant to the perceived governance failures; this would be an instance where a contrary vote would be justified as a means of improving over-all investor returns.  

Our suggested Fourth Paradigm could be viewed as nothing more than re-introducing the fabled “Wall Street Walk,” pursuant to which investment managers commonly voted with their “feet” rather than with their ballot. Ironically, it was in this very context that corporate governance activism was born in the late 1980s. State and local pension funds and union pension funds pointed out that the bulk or all of their investments were managed quantitatively, largely by indexing strategies. Quantitatively managed funds, of course, can’t simply adopt the “Wall Street Walk.” So the notion was put forward that instead of walking the walk, public pension funds and unions should talk the talk of corporate governance reform and by improving corporate governance improve portfolio company performance and, to mix the metaphor, thereby benefit from the rising tide that would lift all of the portfolio company boats.  

The problem then and now, of course, is that there is a lack of consistent empirical evidence about what constitutes good corporate governance, as well as whether such good corporate governance does in fact lead to better company performance. While a number of scholarly articles purport to find positive correlations between a specific corporate governance practice and corporate economic performance, many others find either no correlation or a negative correlation. Moreover, there is a big difference between statistical correlation and demonstration of cause and effect. The lack of persuasive evidence of the economic benefit of so-called good corporate governance practices is the Achilles heel of the corporate governance movement and of the one-size-fits-all voting policy paradigm.15  

Where, then, do quantitatively managed portfolios fit into a voting paradigm of defaulting to management’s voting recommendations, except where the investment manager sees a positive connection between the vote and creation of sufficient shareholder value to justify the cost of analysis and implementation? The answer is that quantitatively managed funds should adopt the same voting paradigm of defaulting to management’s recommendations.  

  • First, it is cheaper than the one-size-fits-all voting policy paradigm because it eliminates the need for any internal corporate governance function and any outside proxy advisory recommendations services. This, in itself, would benefit the economics of a quantitatively managed fund.  
  • Second, the very point of a quantitative investment style is that it is not about governance principles, or management compensation, or any other of the things that corporate governance specialists care about, just as it is not about stock picking or any other active manager’s subjective investment style. Nor is a quantitative investment policy about creating a rising tide to lift all boats even if there were convincing evidence that currently advocated concepts of good corporate governance did create positive shareholder returns on the whole.  
  • Finally, quantitative investment managers and corporate governance specialists don’t have the training and expertise to evaluate the effect of any given corporate governance policy on industry in general, much less on any particular company. They don’t have the tools to understand or measure whether a corporate governance policy does raise or lower the tide, nor whether the boats in their portfolios are moored in the tidal basin they are trying to control.

If a quantitatively managed fund is unwilling to live by its own internal logic and metrics in votes with obvious economic consequences (mergers, recapitalizations, change of control contests and the like) — that is, if they are unwilling to eat their own cooking when something tasting better appears in their kitchen — so be it. Exceptions for these types of votes may not meet the internal investing model of a quantitatively managed fund, but are sufficiently rare as not to be troublesome.  

Moreover, dealing with such exceptions does not require maintenance of a corporate governance staff. These situations are not part of the corporate governance universe and should not be dealt with by its inhabitants. The fact that quantitative investment mangers are willing to ignore their investing principles in occasional cases of economic clarity does not justify creation of a corporate governance infrastructure within quantitatively managed funds or outside of the funds on a fee for service basis that comes out of beneficiaries’ pockets. In sum, our suggested Fourth Paradigm for institutional voting is as appropriate for quantitatively managed funds as for actively managed funds.


The prevailing institutional investor share voting paradigm has pluses and minuses. Its pluses lie chiefly in its relatively low cost, which allows institutional investors to vote thousands of corporate ballots each proxy season without causing push-back by those whose money is being managed. Its minuses include its necessarily cookie cutter type approach to portfolio share voting, the tyranny it creates in terms of imposing uniform (and empirically questionable) corporate governance policies and executive compensation policies on all public companies, and the largely ignored costs it imposes on thousands of public companies. Notwithstanding the minuses, in the absence of a competing paradigm that has both a reasoned and reasonable basis and is practical to implement, the one-size-fits-all voting policy paradigm will continue to prevail.  

A thoughtful reevaluation of an investment adviser’s fiduciary duties to its clients leads to an alternative institutional share voting paradigm, consisting of a default voting policy of voting in accordance with management’s recommendations (except in situations in which the money manager affirmatively believes a contrary vote would result in creation of greater shareholder value after taking into account its direct and indirect costs to shareholders).

This paradigm would recognize that it makes no economic sense for active money managers to try to correct a company’s governance problems through exercise of the voting franchise and reformation of the company’s governance practices and that quantitative money managers have neither a theoretical basis to utilize the voting franchise in an effort to improve a company’s economic performance, nor the expertise to do so. The Fourth Paradigm would also be far better than the one-size-fits-all voting policy model because it would be easier to administer and far cheaper to maintain, thus benefitting all clients of all investment advisers directly and indirectly through lower corporate costs. It would permit shareholder meetings to function without quorum worries. It would protect shareholders in situations where a contrary vote would produce greater economic benefit. And it would satisfy the fiduciary duty of investment advisers with respect to portfolio share voting under both ERISA and the Investment Advisers Act of 1940.

The only negative of our proposed Fourth Paradigm is that it would put most corporate governance activists out of work. This may be good or bad depending on one’s ideology. But it certainly is not a reason to burden institutional investors and their clients with the expense of maintaining a parallel universe for corporate governance activists, nor a reason to burden public companies with the expense and distraction of dealing with this separate universe which creates an additional economic charge to shareholders.