Pay for Long-Term Performance (US)

2010 Draft Policy for Comment

 

Pay for Long-term Performance Alignment (US)

Background and Overview

Pay for performance is a key tenet of sound corporate governance. From a shareholders’ perspective, performance is predominantly gauged by the company’s stock performance. Even if financial or operational measures are utilized, the achievement of these measures should ultimately translate into superior shareholder returns in the long-term. With the recent financial crisis, investors are looking for pay aligned with longer term performance. The 2009 institutional policy survey results indicated that five years is the appropriate time horizon for assessing long-term performance at a company.

Key Changes Under Consideration

Under current RiskMetrics policy, a company is identified as having a potential pay-for-performance disconnect if its one-year and three-year total shareholder returns are in the bottom half of its industry group (i.e., four-digit GICS - Global Industry Classification Group), and its CEO (with at least two years’ tenure) had an increase in total direct compensation from the prior year. RiskMetrics then more closely scrutinizes the company’s compensation to determine whether non-performance-based pay is responsible for the increase, in which case RiskMetrics may recommend votes against a management "say on pay" ("MSOP") proposal and/or compensation committee members. If more than half of the increase in total direct compensation is attributable to non-performance-based equity compensation and there is an equity plan on the ballot, RiskMetrics may recommend a vote against the equity plan in which the CEO participates. This pay-for-performance analysis involves a careful examination of the situation and does not result in an automatic withhold/against recommendation on compensation committee members and/or a recommendation against the equity plan proposals.

We are proposing two changes to this pay-for-performance analysis:

(1) A company may be also identified as having a potential pay-for-performance disconnect if it has unchanged or marginally decreasing CEO pay in conjunction with below-industry-median 1- and 3-year TSR.

(2) When further analyzing companies with a potential disconnect, RiskMetrics will assess the alignment of CEO’s total direct compensation and total shareholder return over a period of at least five years.

The overall approach remains substantially the same as current policy: in cases of potential disconnect, RMG will further analyze the Compensation Discussion & Analysis (“CD&A”) to better understand the various pay elements and whether they create or reinforce shareholder alignment. The most recent year-over-year increase or decrease in pay remains a key consideration, but there will be additional assessment on the long term trend of CEO total compensation relative to shareholder return. Our Pay for Performance review under the Executive Compensation Evaluation policy continues to be a CASE-BY-CASE assessment. See the sample chart below for an illustration of Pay for Long-Term Performance Alignment.

CEO Pay is defined as the sum of base salary, bonus, non-equity awards, present value of stock options and stock awards, target value of performance awards, change in pension value and deferred compensation and all other compensation for FY 2006 and onwards. Prior to FY 2006 (before new SEC disclosure rules), CEO pay is defined as the sum of base salary, bonus, stock awards, long-term incentive payouts (LTIPs), present value of stock options, other compensation and all other compensation. TSR is indexed based on 2004 = 100.

According to RMG policies, standard stock options and time-based stock awards are not considered performance pay. The general economic or sector conditions may elevate a company’s stock price without the company exhibiting superior performance.

Intent and Impact

Strong majorities of both issuer and investor respondents to RiskMetrics 2009 Policy Survey indicated a desire to consider a 5-year time horizon in assessing long-term performance at a company. The proposed changes will allow RiskMetrics to focus on the long-term pay alignment at a company, rather than solely on the year-over-year changes in compensation relative to total shareholder return. At the same time, broadening our identification of potential pay-for-performance disconnects to include companies with small decreases to CEO pay recognizes that there is no bright line test.

We anticipate recommending a slightly higher number of withhold/against vote recommendations in cases of: (i) marginal CEO pay decreases and an overall misalignment in CEO pay and performance with regard to total shareholder return, (ii) CEO pay increases attributed to non-performance pay, or (iii) CEO pay increases attributed to performance-based compensation that is lacking in disclosure in terms of the performance measure and goals (hurdle rates).

Request for Comment

Please feel free to add any additional information or comments on the proposed policy changes. In addition, RiskMetrics is specifically seeking feedback on the following:

Is five-year pay alignment a useful factor to weigh in assessing pay for performance?

What additional factors, if any, should be included an examination of longer-term pay alignment?

Submit a comment View submitted comments