RiskMetrics’ Introduces New Governance Measurement for Proxy Voting Reports

Margaret E. Tahyar is a partner and member of the New York Financial Institutions Group at Davis Polk & Wardwell LLP. This post is based on a Davis Polk General Counsel Update by Ning Chiu and William M. Kelly. Another Davis Polk General Counsel Update, available here, includes information and advice on preparing for the new Governance Risk Indicators. A detailed critique of the CGQ system and its methodology can be found in a discussion paper issued by the HLS Program on Corporate Governance, The Elusive Quest for Global Governance Standards, by Lucian Bebchuk and Assaf Hamdani, which was described on the Forum here.

RiskMetrics Group has recently overhauled its core corporate governance yardstick. Highlights for U.S. companies:

  • The Corporate Governance Quotient (CGQ), which for the past several years has ranked companies, both within their industry and on a broader basis, according to their overall adherence to RMG’s notions of governance best practices, is being discontinued as of June 2010.
  • RMG is adopting a new approach as of March 2010 called Governance Risk Indicators (GRId), which will be applied to all of the 6,400 U.S. companies that it reviews.  Under the GRId system governance practices will be grouped into four headings — Board Structure, Shareholder Rights, Compensation and Audit — and a color-coded risk assessment — High, Medium or Low Concern — will be applied to each category for each company.  These assessments will be made on an absolute rather than a relative basis.
  • The implementation timetable will affect companies with annual meetings after late-March 2010, as companies will start seeing the GRId appear on their proxy research reports beginning in early March for those meetings.
  • RMG is also abandoning its practice of “certifying” director education programs.

The GRId methodology uses a set of between 60 and 80 questions for each company.  For example, the Board Structure category includes individual assessments of director independence, board committee composition, director election voting standards, and board leadership structure.  Each practice is scored to “increase concern”, “reduce concern” or “has no impact.”  The scores are then weighted and compiled to create the overall score for the category.  RMG plans to update the data quarterly and as companies disclose public filings.

Companies with annual meetings after late-March should be on the lookout for the RMG alert notifying them of when they can review their data.  It will be important to review this closely, as RMG sometimes makes mistakes in the midst of trying to handle thousands of proxy voting recommendations around the same time.

What do these changes mean?

  • These are for the most part changes in packaging and presentation rather than in substance.  RMG has not changed its views as to what constitutes best practices, and at least in the example RMG provided, the GRId indicators used in the evaluation largely follow the corresponding CGQ factors.  In other words, the new system replicates the core failing of its predecessor, based as it is on the notion that there is a single set of “best practices” that should apply to all companies.
  • The changes are at least in part a reaction to criticisms that RMG’s practices have been opaque and that elements of its business model, such as selling governance services to companies that it rates, have an unattractive aroma.  The GRId assessment is supposed to be “fully transparent,” with a forthcoming technical document available for anyone to use to replicate RMG’s analysis.  RMG says that the GRId methodology is now aligned with its voting policy, whereas CGQs had previously been somewhat separate.  It is unclear what this will mean in practice.
  • The death of the simplistic and formulaic CGQ is overdue, and the separate evaluation of major categories should make it easier to understand where RMG’s concerns are centered for particular companies.  We think it may also magnify the apparent impact of particular practices, especially in the Board Structure and Shareholder Rights categories.  Under the CGQ system, for example, practices like a classified board or the inability of shareholders to call a meeting or to act by written consent might simply lower your CGQ by a few percentage points, depending on the practices of other companies.  Under the new system, these types of practices, combined with others, might result in a rating of High Concern and thereby increase pressure on the board.
  • Moving from a relative to an absolute system does simplify things.  Under the CGQ system companies were grouped with others based on two S&P classifications—size and industry—and a company’s scores could decrease merely because others changed their practices.  The GRId system will at least eliminate this “moving target” problem and make it easier for companies to understand the implications of potential actions.
Both comments and trackbacks are currently closed.

One Trackback

  1. By Let’s Be Absolutely Clear | Grahall, LLC on Monday, April 12, 2010 at 9:28 pm

    […] March 21, 2010 on the Harvard Law School Forum on Corporate Governance and Financial Regulation (“RiskMetrics’ Introduces New Governance Measurement for Proxy Voting Reports”).  In her blog, Tahyar said: “RiskMetrics Group (RMG) has recently overhauled its core […]

  • Subscribe or Follow

  • Supported By:

  • Program on Corporate Governance Advisory Board

  • Programs Faculty & Senior Fellows