RiskMetrics 2010 Policy Updates (Compensation) Summary

RiskMetrics issued its 2010 Policy Updates on November 19.  The interesting thing is that there aren’t that any significant policy changes for 2010.  I view the 2010 policy updates as more of a refinement of the existing policies than a dramatic evolution.  That being said, the trouble as many folks know, is often in the details.  Interestingly, RiskMetrics issued its Frequently Asked Question on U.S. Compensation on the same day, and, unfortunately, some of the responses to the questions posed are significant changes from prior understandings of how the RiskMetrics’ policies operated.

First, let’s review the U.S. Corporate Governance Policy 2010 updates for Compensation. The 2010 Policy Updates will apply to all shareholder meetings on or after February 1, 2010. RiskMetrics is combining three of its existing policies into a new framework to provide a more integrated and holistic Executive Compensation Evaluation Policy. This new policy will combine RiskMetrics’ prior considerations regarding (1) pay-for-performance, (2) pay practices, and (3) board responsiveness and communication on compensation issues. Along with this re-structuring RiskMetrics will re-order the Compensation Section of its Voting Manual into four policy areas:

  • Executive Pay Evaluation,
  • Equity-Based and Other Incentive Plans,
  • Director Compensation, and
  • Shareholder Proposals.

Additionally, RiskMetrics is making a few tweaks to the existing policies that will underlie the Executive Pay Evaluation policy. These key changes are as follows:

  • Pay-for-Performance – RiskMetrics will now consider the alignment of a CEO’s total direct compensation (TDC) and total shareholder return (TSR) over a period of at least five years. As is currently the case, the first portion of this policy looks at a company’s one- and three-year TSR versus its four digit GICS industry group’s medians. If a company’s one- and three-year TSR bothare below its industry group median, then RiskMetrics will now look to see if the CEO served in that position for at least two consecutive fiscal years, and, if so, whether his/her TDC is aligned with the company’s TSR over time, including both recent (1- and 3-year periods) as well as long-term (at least 5 years) periods. Furthermore, RiskMetrics will consider the mix of performance-based compensation to total compensation. But RiskMetrics did not indicate how this consideration will be weighed or factor into its analysis. RiskMetrics’ draft policies indicated that it is no longer simply a matter of ensuring that a CEO’s pay has not increased in order to avoid application of the policy. Companies now must consider cases where the CEO’s compensation remained the same or even decreased slightly if the company’s one- and three-year TSR were both below its industry group medians.  RiskMetrics has not offered much guidance on what level of decrease will avoid application of the policy.
  • Problematic Pay Practices – RiskMetrics will now re-order how it will express displeasure with a company having problematic pay practices. If a company maintains a problematic pay practice(s), RiskMetrics will generally: (1) first vote AGAINST a management say on pay (MSOP) proposal, (2) then AGAINST/WITHHOLD on compensation committee members in (i) egregious situations, or (ii) when no MSOP item is on the proxy, or (iii) when the board has failed to respond to concerns raised in prior MSOP evaluations; and/or (3) AGAINST an equity-based incentive plan proposal if excessive non-performance-based equity awards are the major contributor to a pay-for-performance misalignment.

RiskMetrics has indicated that certain adverse practices carry more weight on a stand-alone basis in its evaluation, and may result in negative recommendations on a stand-alone basis, i.e., “problematic pay practices,” which include:

  • Multi-year guarantees for salary increases, non-performance based bonuses, and equity compensation;
  • Including additional years of unworked service that result in significant additional benefits, without sufficient justification, or including long-term equity awards in the pensions calculation;
  • Perquisites for former and/or retired executives, and extraordinary relocation benefits (including home buyouts) for current executives;
  • Change-in-control payments exceeding three times (3x) base salary and target bonus;
  • Change-in-control payments without job loss or substantial diminution of duties (“Single Triggers”);
  • New or materially amended agreements that provide for “modified single triggers” (under which an executive may voluntarily leave for any reason and still receive the change-in-control severance package);
  • New or materially amended agreements that provide for an excise tax gross-up (including “modified gross-ups”);
  • Tax reimbursements related to executive perquisites or other payments such as personal use of corporate aircraft, executive life insurance, bonus, etc.;
  • Dividends or dividend equivalents paid on unvested performance shares or units;
  • Executives using company stock in hedging activities, such as “cashless” collars, forward sales, equity swaps or other similar arrangements; or
  • Repricing or replacing of underwater stock options/stock appreciation rights without prior shareholder approval (including cash buyouts and voluntary surrender/subsequent regrant of underwater options).

[see the FAQ section below for information on “minor” problematic pay practices.]

RiskMetrics announced that it will also assess company policies and practices related to compensation that could incentivize excessive risk-taking, including:

  • Guaranteed bonuses;
  • A single performance metric used for short- and long-term plans;
  • Lucrative severance packages;
  • High pay opportunities relative to industry peers;
  • Disproportionate supplemental pensions; or
  • Mega annual equity grants that provide unlimited upside with no downside risk.

RiskMetrics has indicated that some factors that potentially mitigate the impact of risk incentives are rigorous claw-back provisions and robust stock ownership/holding guidelines.

Equity-Based and Other Incentive Plans

In the area of equity-based and other incentive plans, RiskMetrics has tweaked its policies regarding volatility and the stock price used in its Shareholder Value Transfer (SVT) and Burn Rate analyses. Last year in the face of the steep market decline, RiskMetrics announced that for the December 1, 2008, March 1, June 1, and September 1, 2009 quarterly download dates it would use the 400-day volatility and the 90-day average stock price for the SVT and burn rate policies. For the December 1, 2009 and subsequent quarterly data download dates (which means this will apply to companies having their next shareholder meeting on or after February 1, 2010), RiskMetrics announced it will revert to using the 200-day volatility and 200-day average stock price.  Given the volatility in the stock markets last fall and again this year, most companies will be evaluated using a significantly higher volatility.  This will cause companies’ outstanding stock options, even those that are underwater, to carry a higher valuation, as well as any shares available for grant and any newly requested shares grantable as stock options.  This higher valuation for stock options will limit the size of acceptable share proposals for most companies in 2010 compared to what they would have been able to achieve using the 400-day volatility and 90-day stock prices.

RiskMetrics also updated its Burn Rate Table for 2010, which is detailed on page 28 of the U.S. Corporate Governance Policy, 2010 Updates (November 19, 2009).

RiskMetrics’ Compensation FAQs

RiskMetrics also released its 2010 Corporate Governance Policy Updates and Process, Frequently Asked Questions on U.S. Compensation (November 19, 2009). As mentioned previously, the FAQs are where some real changes got introduced for 2010.

The first group of questions (designated with a number, i.e. Q1.#, in the PDF version only) all involve the newly introduced Executive Compensation Evaluation policy. These questions and answers (Q&As) give some guidance on how RiskMetrics will apply this new policy analysis. 

In Q1.4, RiskMetrics lists additional problematic pay practices that may receive a withhold/against vote recommendation or cautionary language upon a case-by-case analysis. This list expands the list consolidated into the new Executive Compensation Evaluation policy.  The way I read it, none of these practices standing alone generally would be sufficient to trigger a withhold/against vote recommendation, unlike the practices detailed in the 2010 Policy Updates which could.  Of course, in cases where a company has a history of problematic practices or does something RiskMetrics views as an egregious compensation practice, I’d imagine that RiskMetrics could apply the against/withhold vote recommendation if even a single one of these more “minor” problematic practices was present.

Q1.7 addresses how companies might be able to overcome an against/withhold vote recommendation on their compensation committee or a negative recommendation for their MSOP or equity plan proposal on the basis of a CEO pay for performance disconnect. RiskMetrics indicates that any action to try and overcome such a vote recommendation must be tailored according to the underlying issues identified in RiskMetrics’ analysis, and should be a recommitmet/commitment to performance. Examples of the types of underlying issues RiskMetrics could identify and how they might be overcome in a performance commitment are:

  • The primary source of pay increase is due to time-vested equity awards, a renewed commitment could be for a company to commit to making a substantial portion (at least 50 percent of the shares awarded) of equity awards to named executives officers performance-based.
  • The primary source of pay increase is due to discretionary bonus, a renewed commitment could be to award only performance-based bonuses.

RiskMetrics also indicates that companies making such commitments should be sure to disclose the performance measures and goals, including at least the following:

  • the measure(s) used (and rationale for the selections);
  • the goal(s) that were set for each metric and the target (and, if relevant, threshold and maximum) payout level(s) set for each NEO;
  • the reason each goal was determined to be appropriate for incentive pay purposes (including the expected difficulty of attaining each goal);
  • the actual results achieved with respect to each goal; and
  • the resulting award (or award portion) paid to the NEO with respect to each goal.

RiskMetrics clarifies that the actual performance results and the resulting award do not need to be disclosed until the performance cycle is complete

Qs2.3-2.8 deal with how RiskMetrics will apply its stock option carve-out policy.  Basically, prior to this guidance, this policy was generally thought to provide that if stock options have been outstanding for more than 6 years and are in-the-money, RiskMetrics might exclude such stock options from its SVT analysis for a new share proposal if certain additinal information was disclosed in the proxy proposal concerning these and other outstanding stock options. Among other things, these Q&As place the following constraints on companies ability to use the stock-option carve out exemption:

  • Companies must have sustained positive stock performance, i.e., positive 5-year TSR as well as positive year over year performance for the past five fiscal years at the time of the analysis. Exceptions may be made if TSR was negative for the first two years and then strongly positive for the remaining three years, but vested grants that were underwater for a substantial time during the five years are not eligible for the carve-out.
  • Companies must have high overhang costs attributable to in-the-money options outstanding for more than six years. High overhang cost means that the sum of outstanding stock options and stock awards and remaining shares available under existing equity plans should exceed or approach a company’s specific allowable cap, and that outstanding stock options and stock awards must be a significant driver of the high overhang, i.e., in the range of 75 to 100 percent of the total overhang. I’m assuming this means that the SVT attributed to the outstanding stock options and stock awards must be 75% to 100% of a company’s allowable cap.
  • The dilution caused by any new share request must be reasonable. However, RiskMetrics has not offered any guidance on what would be considered a reasonable amount of dilution in this context.

When a company wants to try and utilize the stock-option carve-out policy exception, it should keep in mind that this is applied by RiskMetrics on a case-by-case basis.  RiskMetrics has also indicated that simply stating the number of stock options that have been outstanding for more than six years and are in-the-money is insufficient.  Instead, companies should disclose much more information about such options as well as additional information about all the other outstanding stock options in order to enable RiskMetrics to assess whether and how the policy might be applied. As an example of the type of information that it is looking for, RiskMetrics cited Myriad Genetics’ DEFA 14A filed October 28, 2009.

In analyzing whether the stock option carve-out policy should be applied, RiskMetrics will look at a company’s concentration ratio for equity awards, i.e., the total equity grants to the top five executives divided by total equity grants to employees and directors.  Concentration ratios greater than 50 percent to NEOs could concern RiskMetrics. Finally, RiskMetrics heads off further questions and requests for exceptions for application of the policy for companies that may not have experienced positive TSR lately. RiskMetrics believes that strong performing companies have experienced significant market rebound and should reflect that the stock price decline is temporary.

In Q2.9, RiskMetrics addresses a timing issue some companies may have with compensation decisions and RiskMetrics’ Pay for Performance policy. Specifically, if a company makes equity grants near the beginning of a year based on an evaluation of the company’s and/or an executive’s performance in the immediately preceding year, such equity grant information then appears in the following years’ proxy (because the proxy disclosure rules dictate that only equity awards granted in the year covered by the proxy are to be included). The question some companies have faced is whether RiskMetrics will take into account the timing og such equity awards and make adjustments to the top executives’ total compensation when conducting its pay-for-performance analysis? RiskMetrics may consider the timing of equity awards at the beginning of a fiscal year if complete disclosure and discussion is made in the proxy statement about such awards. But, RiskMetrics cautions that such additional information needs to be provided in a shareholder-friendly manner, such as through an alternative Summary Compensation Table that reflects compensation payable based on performance achieved for the year and including details of the equity awards granted after the end of the fiscal year covered by the proxy tables.

Conclusion

As you can probably tell, there are still a few details to be worked out with these new policies and revisions.  Hopefully RiskMetrics will issue some additional guidance (or more FAQs) addressing these ambiguous issues to help give issuers more certainty about how RiskMetrics will apply its policies going forward.

The revised 2010 compensation policies coupled with the FAQs for 2010 generally will mean more companies will run afoul of these policies, fewer companies will be able to take advantage of the small exceptions that do exist, and companies overall will see lower share authorizations that pass the SVT model (if they do at all). Finally, these trends coupled with the decline in equity grants during 2009 will most likely lead to lower SVT allowable caps for the 2011 proxy season. Of course, RiskMetrics may also introduce new requirements for the 2011 proxy season, which could further exacerbate the difficulties companies face in requesting shareholder approval of additional shares for equity compensation plans in 2011.

[print_link]

Print Friendly, PDF & Email

RiskMetrics’ Draft 2010 Policy Updates

Last week RiskMetrics (RMG) released its Draft 2010 Governance Policy Updates.  These can be found at:

http://www.riskmetrics.com/policy/2010comment

RMG is taking comments on these draft policy updates through November 11, 2009.

The draft US policy updates include possible revisions to the following policies:

  • Adoption or Renewal of Non-Shareholder Approved Pills
  • Director Independence
  • Long-Term Pay for Performance Alignment
  • Pay Riskiness

Adoption or Renewal of Non-Shareholder Approved Pills
RMG is suggesting the following changes to this policy:

  1. Clarify that short-term pills (those with a term of 12 months or less) adopted without shareholder approval will cause RMG to consider the recommendation for director elections on a CASE-BY-CASE basis taking int account certain specified factors. Additionally, any long-term pills (those with a term of more than 12 months) adopted or renewed without shareholder approval will cause RMG to recommend a WITHHOLD/AGAINST vote on the full board of directors (except new nominees, who will be considered on a CASE-BY-CASE basis).
  2. RMG will review companies that adopt long-term pills at least once every three years and may recommend (or continue to recommend) that shareholders vote AGAINST or WITHHOLD votes from the entire board if the pill is still maintained.
  3. Clarify that if a board makes a material, adverse change to an existing poison pill without shareholder approval, RMG will recommend an AGAINST or WITTHOLD vote for the directors.

Director Independence

  1. RMG proposes to bifurcate the materiality test it applies (currently applies the NASDAQ-based materiality test) on transactional relationships, so that companies that follow NYSE listing standards will be subject to the NYSE-based test of the greater of $1 million or 2% of the recipient’s gross annual revenues, and that non-NYSE listed companies will continue to be subject to the NASDAQ-based test of the greater of $200,000 or 5% of the recipient’s gross annual revenues.
  2. RMG is proposing to define “professional services” as “advisory in nature, generally involving access to sensitive company information or to strategic decision-making, and typically have a commission or fee-based payment strusture.”

Pay for Long-Term Performance Alignment
RMG is proposing to slightly modify its Pay for Performance Policy that is applied to equity compensation plan proposals to determine if there is a disconnect between CEO pay and company performance (measured by 1- and 3-year TSR compared to the company’s 4-digit GICS industry group’s medians):

  1. RMG wants to modify the policy so that a company can be 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. Furthermore, when further analyzing companies after finding a potential disconnect exists (i.e., the company’s 1- and 3-year TSRs are below their industry group’s medians), RMG want to assess the alignment of the CEO’s total direct compensation and total shareholder return over a period of at least 5 years.

Comment: If these policy changes are implemented, many more companies will be found to have a disconnect between their pay and performance. It is not clear what would constitute “marginally decreasing” CEO pay, so it is a bit difficult to offer specific comments.  On a relative basis, one might think that this could refer to some type of de minimis decline in pay, but without guidance as to where the line will be, it is impossible to get an idea of how this policy change could impact companies.  If, for example, RMG decides that a decline of 15% or less is “marginally decreasing” I’d imagine many companies could be pulled into a further review under this policy.

As for exapnding the period used to see if there has been a disconnect between CEO pay and company performance to at least 5 years, I think this may prove troublesome for some companies. What would RMG do if the CEO had not been in that position for the past 5 years?  Would the policy not apply? Or, would RMG use whatever data is available? I’d argue that in such a case the policy should not be applied since the CEO has not been there for the performance period. Based on how this policy change is described, it appears that the initial screen would still be 1- and 3-year TSR, and you’d only get to look at 5 year performance if the first screen was failed.

Pay Riskiness
RMG is proposing to address pay riskiness in its policies (not currently done).  RMG is proposing the inclusion of an assessment of company pay practices that may incentivize inappropriate risk-taking under the overall executive compensation evaluation of “problematic pay practices” policy. RMG indicates that such practices could include, but not limited to, guaranteed bonuses, single performance metric used for short- and long-term plans, high severance packages or high pay opportunities relative to industry peers, mega annual equity grants and disproportionate level of supplemental pensions. Additionally, RMG will take into account mitigating factors, such as rigorous claw-back provisions and robust stock ownership/holding guidelines.

Comment: The pay riskiness comments are somewhat unsettling given the early nature of risk assessments as applied to compensation generally.  I don’t know that there is sufficient data or generally acceptable practices as far as riskiness of pay is concerned.  Additionally, while RMG has indicated some mitigating factors it will take into account, it is unclear what practical steps companies will be able to undertake to avoid being branded as risky payers. If concern of being stuck with that moniker is wide-spread, we might end up in a situation where compensation practices that serve the attainment of legitimate corporate strategic goals could be discarded entirely due to the perception of such practices.  Such a rush to judgment may not be warranted and could hurt companies (and their shareholders) trying to craft compensation policies and programs that serve their corporate strategic goals. Assuming RMG includes this as a narrative piece of its proxy analysis and this does not control any vote recommendation, the policy would be more palatable, but RMG will need to work on developing a sound and well-explained and supported rationale for labeling certain pay practices as risky and should provide for exceptions when such practices can and should be undertaken by companies in the pursuit of their goals.

Open Issues

What About the Tax Gross-Up Policy?
As you may recall, this past proxy season, RMG announced a revision to its poor pay practices policy that indicated that if a company adopts a change-in-control or severance arrangement that provides an excise tax gross-up, RMG would consider that a poor pay practice and could recommend against the CEO, Compensation Committee, or the entire board.  RMG applied this policy to a number of companies in 2009 that adopted or renewed contracts or arrangements that contained excise tax gross-ups.  Some companies eliminated these provisions entirely to avoid the policy while other companies committed to not adopting any new agreements with such provisions.

However, I’ve heard from several folks that RMG may be issuing a clarification of this policy for 2010 that will indicate that the policy only applies to the election of directors in the year after a company adopts, renews or modifies an agreement or arrangement containing an excise tax gross-up provision, and that in the subsequent years, the policy will not be applied to the election of directors unless the company subsequently adopts, amends or renews an agreement or arrangement with such a provision.  If this is the case, then the policy is a bit more manageable and companies’ board might decidethat from a business-perspective they want their executives to have an excise tax gross-up and that they’ll take their lumps in the following year’s elections, but thereafter would not have to worry about the policy until another agreement or arrangement with such a provision is adopted, amended or renewed. I don’t know for certain that this policy will be released.  But if it is, it will definitely provide something for companies to contemplate when trying to figure out a response to WITHHOLD/AGAINST director vote recommendations from RMG for violation of the poor pay practices policy due to an excise tax gross-up.

What about companies that reincorporated to Switzerland?
Will such companies, even though primarily traded on US exchanges and markets, still be subject to RMG’s International Proxy Voting Guidelines? RMG Research changed to this policy early this year, and a number of folks – issuers and institutions — were a bit upset.  Consequently, the final 2010 Corporate Governance Updates might address the Switzerland reincorporation issue.  RMG left that possibility open in the last FAQ it posted about the topic back in April 2009:
http://www.riskmetrics.com/policy/2009_Switzerland_Reincorp

[print_link]

Print Friendly, PDF & Email

Disturbing Trends from RiskMetrics’ Model

Well, I’ve now seen enough RiskMetrics (RMG) models after the 9/1/2009 quarterly lock-in date that I can tell you that there is trouble brewing.

First, the valuation of stock options (and SARs) under RMG’s binomial option pricing model has climbed dramatically in relation to the stock price.  Just what exactly do I mean by that??? Well, what we’re talking about is the economic value assigned by the RMG model to shares available for grant as stock options under new or existing and continuing plans.  Primarily due to the surge in volatility both in late 2008 and then again through 2009 as the market regained 10,000, the value generated by the binomial model has grown. Volatility is one of the key inputs into the binomial model and as it increases, so does the value of the stock option.

OK, so how bad are things looking? Well, based on the companies I’ve worked with so far, I have yet to see a company where one full value award (an award other than a stock option or SAR that is settled with shares) equaled 2 or more stock options.  Put another way, in all the companies I’ve seen so far, less than 2 stock options equal the value of one full value award (which is set at the 3 month closing stock price). Note that last year, seeing this equation at just 2 stock options equal to 1 full value award was almost unheard of, the vast majority of companies’ had 2+ stock options equal 1 full value award.

In practical terms, as the value of stock options has grown (on a relative basis compared to the stock price) the number of new shares that companies can get approved under the RMG model has dropped. So many companies are in a real share squeeze under the RMG model, that I think RMG will be forced to make some major concessions again this year and tweak the operation of its policies in order to avoid the loss of credibility.

Last year, as you’ll recall, as a result of the precipitous stock price decline of many companies, RMG revised its methodology for determining volatility and stock price:

  • Volatility – prior to 12/1/08 RMG used the 200-day volatility, annualized in its model. After 12/1/08, RMG used the 400-day volatility, annualized in the model.  This change helped by generally lowering volatility (but volatility still increased compared to that measured on 9/1/08).
  • Stock Price – prior to 12/1/08, RMG used the 200-day average closing stock price. After 12/1/08, RMG used the 3-month closing average stock price. This generally caused companies to have lower market values under the model, and helped lower the valuation of stock options and full value awards alike.

I’ve heard a rumour that RMG might be looking at pushing out the period of time it uses to measure performance from 3 years to 5 years.  It is unclear which of the many performance measurements this would impact, but it could include: the 3-year TSR performance used within each4-digit GICS industry group to help determine the top quartile performers upon which RMG bases its regression formulas for determining companies’ allowable caps; it could be the performance measured in the pay for performance policy (1- and 3-year TSR compared to 4-digit GICS industry median); or something entirely different. Apparently RMG received feedback from the survey it conducted this past summer that indicated a number of its clients regarded 5 years as the proper measure of long-term performance.

Even if this change does not get made, RMG still must determine what to do concerning the stock price and volatility.  Last year when RMG modified the way it calculated these figures, it indicated that it would revisit them in a year and determine what it would do, i.e., keep the new methodology, return to the old methodology, or, perhaps, do some further modifying of things (the last is only my speculation at this point).  We’ll have to wait until RMG releases its policy updates (hopefully the week before Thanksgiving this year) to see just exactly what will change, but I bet that the methodology for determining these and other figures must change or RMG could be in a position where its model would otherwise recommend against the majority of companies, instead of just the 30% or so of proposals that it tries to target for failing the model. So stay tuned, as new developments come in, I’ll blog about them here.

Second, the allowable caps being generated by the model when coupled with the high amounts of overhang at many companies are causing much pain when it comes to seeking additional shares.  In many cases companies could not even pass with the shares they currently have available.  In other words, most companies are looking at not being able to get any additional shares and have the RiskMetrics model approve of the share request. I have dealt with some that the model would permit to get additional shares, but for the most part, the amounts that pass the model are down from prior years. Specifically, in my experience, the 15% simple dilution threshold often was far below the shares that the RMG model would approve.  Now the reverse is true.  This puts companies in a precarious position – they’re unable to ask shareholders to approve shares with a proposal that will receive a FOR vote recommendation from RMG.  What to do? There are a number of strategies to consider.  I will discuss several of the more significant ones at the upcoming NASPP conference in San Francisco with a panel of other experts in our presentation, Top Tips to Ensure Shareholder Approval of Your Stock Plan (on November 10 at 2:00 pm local). For more information about the NASPP conference, please visit: www.naspp.com

Next week I’ll take a look at some of the plan features that can cause problems with RMG, as well as some of the other policies you should be aware of for the 2010 proxy season.

[print_link]

Print Friendly, PDF & Email

Launching 9/8/2009!

Thanks for visiting!

Only a few more days until I officially launch this blog.  Until then, feel free to wander around, but keep in mind that things are still “under construction.” Also, please take a look at my presentation on implementing new equity compensation plans and plan amendments (link below):

View more presentations from EHauder.

If you have any suggestions on how this site could be useful to folks interested in equity compensation plans, please let me know.

[print_link]

Best regards,
Ed Hauder

Print Friendly, PDF & Email