Understanding our Compensation Analysis

Below is an overview of Glass Lewis’ approach to analyzing compensation proposals in the United States, with separate sections covering  say-on-pay analysis, pay-for-performance analysis, and our analysis of equity-based compensation plans.

Say on Pay Analysis

Glass Lewis’ approach to say-on-pay has two main components: (i) a qualitative assessment of the structure of a company’s compensation program and the accompanying disclosure; and (ii) a quantitative assessment reflected in our pay-for-performance grade. As a result of this approach, a poor grade in our pay-for-performance analysis will not automatically result in a negative recommendation, and a favorable grade does not guarantee a positive recommendation.

Our quantitative approach is derived from the Glass Lewis pay-for-performance model, explained in our Pay-for-Performance Analysis. The relationship between relative executive compensation and relative performance is the basis of the pay-for-performance model. Our model evaluates the compensation of the top five executives against the compensation of the top five officers at peer companies, then compares the company’s performance with those same peers. Through these analyses, we are able to evaluate whether the company’s executives have been paid in line with the company’s relative performance.

In considering the qualitative merits of a compensation program, we review industry, company size, maturity, financial position, historical pay practices and any other relevant internal and external factors. We generally highlight any compensation-related decisions or features we believe may be detrimental to shareholders’ interests, as well as any important information that has not been clearly provided.

Our review of a company’s practices also takes into consideration the compensation committee’s response to previous say-on-pay votes and the level of shareholder support. When a company receives low support for its say-on-pay proposal, we believe the compensation committee should provide some level of response to shareholders’ concerns, including engaging with large shareholders to identify the concerns driving the opposition. Shareholders should also expect adequate disclosure of any such engagement and any resulting feedback or changes being made to address outstanding concerns.

Our say-on-pay analysis includes two additional views of compensation for the chief executive officer that may differ from a company’s statutory disclosure of compensation. One figure is realizable pay; the other is a breakdown of CEO compensation granted but not necessarily earned for the year in review, presented in the CEO Compensation Breakdown table. This table excludes changes in pension value and non-qualified deferred compensation earnings (“NQDCE”). Neither of these figures is used in our pay-for-performance model.

Say on Pay FAQs

Q: How do pay-for-performance grades affect Glass Lewis’ say-on-pay recommendations?
A: The pay-for-performance analysis provides a quantitative view of a company’s pay and performance alignment, which is considered alongside other qualitative factors, including the company’s compensation structure, the compensation-related decisions made in the past year and the company’s operations.

Q: If a company receives an “F” in the pay-for-performance model, will Glass Lewis automatically recommend against the company’s say-on-pay proposal?
A: No. A company that receives a failing pay-for-performance grade will not automatically receive an “against” recommendation on its say-on-pay proposal. Likewise, a company that receives a passing grade will not automatically receive a “for” recommendation. Our approach to analyzing advisory votes on executive compensation is based on both a quantitative and qualitative assessment of the company’s compensation practices.

Q: Why are the figures in the CEO Compensation Breakdown table different from the Summary Compensation Table figures?
A: The CEO Compensation Breakdown table reflects compensation granted but not necessarily earned in the year in review. It also excludes changes in pension value and non-qualified deferred compensation earnings (“NQDCE”). When there is a significant discrepancy between the figures in the two tables, it is often due to: (i) differences in when long-term cash is accounted for; (ii) substantial changes in pension value or NQDCE; or (iii) companies granting long-term incentives for the year in review following the fiscal year end.

Q: How is realizable pay calculated?
A: Glass Lewis Realizable Pay is calculated over a three-year period and includes: actual salary received; actual incentive cash granted and earned; the intrinsic value of time-vesting equity granted; the intrinsic value of performance-based equity granted and earned; and all other compensation paid.

Q: How does Glass Lewis assess low levels of say-on-pay support in past years?
A: When a company receives low support for its say-on-pay proposal (generally less than 75%), we believe at a minimum the company should provide some level of disclosure regarding the company’s response to shareholder opposition; such disclosure, which often includes a discussion of engagement meetings and feedback received, should be accompanied by relevant changes and/or rationale intended to address outstanding concerns.

Q: How does Glass Lewis assess an award’s time-vesting period?
A: We view an award’s time-vesting period to be the time an individual must wait before an entire award is fully vested.

Q: When does Glass Lewis consider an award to be performance-based?
A: We consider an award to be performance-based if it is earned and/or vests based on the attainment of previously established goals. The process by which achievement is determined must be established substantially before the close of the measurement window. If a set of minimum goals is not achieved, the award should be subject to forfeiture.

Q: When does Glass Lewis consider a provision to be “single-trigger”?
A: Awards that vest automatically upon a change in control are considered to have single-trigger vesting. If vesting may be accelerated, but is not the automatic course of action, we do not consider the vesting provisions to be fully single-trigger.  We consider awards to have double-trigger vesting when they vest upon termination of employment following a change of control.

Q: What is Glass Lewis’ view on recoupment policies in the United States?
A: We believe that companies should have recoupment policies that are consistent with those contemplated under the Dodd-Frank act. Recoupment provisions should allow for the recovery of awards for all senior executives, which, at a minimum, includes a company’s named executive officers. Awards should be subject to recoupment as a result of financial restatement, noncompliance with applicable rules, negligence and other forms of misconduct.

Pay-for-Performance Analysis

Glass Lewis’ view on executive compensation is based on the premise that management’s primary duty is to maximize shareholder value and the performance of the company. Compensation practices should align management’s interests with those of shareholders. Thus, executive compensation should be closely tied to company and stock performance.

We recognize that many of the factors that affect a company’s performance will also affect the rest of the industry. Therefore, we believe executive compensation should be closely tied not to absolute or overall performance but rather to the company’s track record of performance relative to its peers. Management should be especially rewarded for directing the company in a manner that outperforms its peers.

Our model evaluates five indicators of stock pay-for-performance-model 419x367performance and business performance, depending on the company and its industry. For the majority of sectors, these five metrics are: (i) total shareholder return; (ii) EPS growth; (iii) change in operating cash flow*; (iv) return on equity; and (v) return on assets.

* Change in operating cash flow is replaced with: (i) tangible book value per share growth for companies in the Banks, Diversified Financials and Insurance sectors; and (ii) growth in funds from operations for REITs, with the exception of Mortgage and Specialized REITs.

This relationship between relative executive compensation and relative performance is the basis of Glass Lewis’ proprietary pay-for-performance model. Our model evaluates the compensation of the top five executives by benchmarking that compensation against the compensation of the top five officers at peer companies. The model then compares the company’s performance to that of those same peers. Through these analyses, we can evaluate whether the company’s executives have been paid in line with the company’s relative performance.

Peer group selection is a critical and highly scrutinized segment of executive compensation analysis today.

Glass Lewis utilizes the Equilar peer group as a fundamental component of its proprietary pay-for-performance model. Equilar uses a method of peer group development based directly on market data and social analytics.

Equilar Peer Selection Process

Equilar has developed a methodology for creating peer groups that solves the one-size-fits-all dilemma. Based on the disclosed peer groups of the company, this flexible process creates the most logical peer groups by using the disclosed peers of the entire Russell 3000 and S&P/TSX Composite Indices.

Using analytics and algorithms proven in thepeers 279x381 social networking space, Equilar Insight generates an interconnected network of peer companies consisting of “who you know” and “who knows you.” Equilar Market Peers evaluates the strength of these relationships (one-way vs. reciprocal connections) and creates a list of the strongest connections.

By bringing together peer group disclosure for thousands of public companies and proven analytics from the social networking space, we believe this approach best models market choices. The results are pivotal for assessing and developing more accurate and sound peer benchmarking groups.

SEC regulations require companies to list firms compared for executive compensation benchmarking. Equilar uses this disclosed company relationship information to build a peer network.

The network consists of companies and their disclosed peer connections. The Equilar algorithm extracts this information to identify the strength of relationships between two companies. The stronger the correlation, the higher the peer is ranked.

By looking directly at market data, this approach avoids the limitations of arbitrary financial cut-offs or discrete industry groupings and better represents the complex relationships that exist in a competitive marketplace.

Equilar updates its market-based peers in January and July. Companies wishing to update their peers can do so at Equilar.com. The market-based peer data is based on publicly-disclosed information, as well as information provided to Equilar via its portal during its open submission periods. Glass Lewis may exclude certain market-based peers from a company’s pay for performance analysis if the peer falls into one or more of the following categories:

  • Has less than two years of trading history from the most recent fiscal year end;
  • Has been privatized, acquired or delisted;
  • Is a non-U.S. company or foreign private issuer;
  • Does not have two full years of compensation data that aligns with the years that it has been publicly traded;
  • Has changed the company’s fiscal year end, impacting the consistency of the financials used to calculate growth rates;
  • Has experienced M&A transactions that would impact the consistency of the financials used to calculate growth rates.

Grading Pay-for-Performance

The Glass Lewis model calculates a weighted-averagepay-performance-grades 238x222 executive compensation percentile and a weighted-average performance percentile. These two percentile rankings are compared to determine how closely the compensation tracks the relative performance of the company.

The companies with the largest “gap” can be identified as companies that have done a poor job of linking compensation with performance. Based on the magnitude of the gap, each company is assigned a school-letter grade: “A”, “B”, “C”, “D” or “F”.

Pay for Performance FAQs

Q: In which markets does Glass Lewis use a pay-for-performance model?
A: The pay-for-performance model and methodology outlined here is used for the United States and Canadian markets. We maintain a separate pay-for-performance model for the Australian market, based on different methodologies.

Q: Why doesn’t Glass Lewis use pay-for-performance model in other countries?
A: Since our pay-for-performance model requires consistent, standardized and market-wide disclosure of executive compensation, multi-year financial and compensation data, and a publicly disclosed and comparable peer group, we are only able to use our model for companies in the U.S. and Canada.

Q: How does the pay-for-performance model work?
A: The pay-for-performance model measures the company’s weighted average executive compensation percentile rank for its top five executive officers against the company’s weighted average performance percentile rank within a group of 15 peer companies. The model covers three years of pay and performance, but is more heavily weighted toward the year in review.

Q: Where do the peers used on the pay-for-performance page come from?
A: The peer groups are generated by Equilar based on a company’s self-disclosed peer group and the strength of connection between peer companies (i.e. one-way vs. reciprocal connections). The top 15 peers are used in our pay-for-performance analysis.

Q: When does Equilar update its market peers?
A: Equilar updates its market-based peers twice yearly in January and July.

Q: If the peers listed for my company are not accurate, whom should I contact?
A: Since Glass Lewis does not alter the peer group selection we receive from Equilar, we recommend contacting Equilar directly at info@equilar.com

Q: Which metrics does Glass Lewis use in determining pay-for-performance alignment?
A: Our model evaluates five indicators of shareholder wealth and business performance: total shareholder return, earnings per share growth, change in operating cash flow, return on equity and return on assets. Change in operating cash flow is replaced with: (i) tangible book value per share growth for companies in the Banks, Diversified Financials and Insurance sectors; and (ii) growth in funds from operations for REITs, with the exception of Mortgage and Specialized REITs.

Q: What timeframes are company performance measures based on?
A: Performance measures, except ROA and ROE, are based on the weighted average of annualized 1, 2, and 3 year data. ROA and ROE are calculated over one year.

Q: How are the pay-for-performance metrics weighted?
A: Glass Lewis does not disclose the weightings.

Q: How does Glass Lewis calculate compensation figures for a given year?
A: We capture the sum of all cash and equity compensation paid to the five most highly paid NEOs, including the CEO, in their roles as continuing executives, net of severances and forfeitures. We perform our own stock and option valuations and exclude any changes in pension value.

Q: How does the model treat mid-year CEO changes?
A: If a company changes CEOs in the year in review, compensation paid to the outgoing and incoming executive is pro-rated for time served in that role and aggregated as compensation paid for the position of CEO for the year.

Q: How do mergers or acquisitions affect the model’s analysis?
A: We may exclude a company’s pay-for-performance analysis or growth rate calculation if there are M&A transactions that would impact the consistency of the financials used to calculate growth rates.

Q: How are compensation data for Canadian peer companies treated?
A: For Canadian peers, equity awards are normalized using the grant date exchange rate and cash compensation data is normalized using the fiscal year average exchange rate.

Equity-Based Compensation Analysis

Glass Lewis evaluates equity-based compensation plans based on criteria we believe are key to equity value creation. Our model seeks to determine whether the company’s granting practices are excessive when compared to a peer group.

The analyses can be classified into three categories: program size and granting pattern; program cost; and program features. Our analysts also consider relevant specific situations before making a recommendation on a case-by-case basis.

Program Size and Granting Pattern

The program size analysis generally focus on two questions: Does the company already have a sufficient number of shares available under its existing plans? If the proposed plan were approved, approximately how many years would the company be able to grant awards without returning to shareholders for approval of more shares? Generally, we believe that companies should only seek new shares when needed and that shareholders have the right to review equity compensation programs roughly every three years.

We also measure the company’s pace of historical grants to see if it is excessive. We look at the full-share equivalent dilution during the past three years to measure dilution. Full-share equivalent grants are calculated as [(net options granted / full-value grant multiplier) + net full-value awards granted].

Additionally, we measure the level of overhang at the company and compare this dilutive measure to peer companies. We believe programs should not be excessively dilutive on a whole and provide a peer comparison to identify company overhang that is significantly above that of a peer group.

Program Cost

Glass Lewis measures the annual cost of a company’s equity compensation in two ways before comparing it with peer companies.

Our “projected cost” is an estimate of the grant-date fair value of awards for the coming year based on the company’s historical granting patterns. Generally, our estimate is a weighted average calculation of the company’s gross grants for the past three years. The analyst then determines whether the weighted average should be adjusted due to any company-specific circumstances.

Glass Lewis performs its own valuation to determine the value of stock options using the Black-Scholes model, along with standardized methodologies, to derive some of the input variables for all companies in our model. Using these standardized calculations ensures that the valuation can be compared on an equalized basis across peer companies. For full value awards, we use the average of the company’s closing share price over the last four quarters.

We also measure the cost of the plan as disclosed by the company; we refer to this as a company’s “expensed cost.” This is the company’s recognized amount of stock-based compensation expense for the most recently completed fiscal year in its statement of cash flows.

The projected cost and expensed cost are then measured as a percentage of a company’s operating metrics (operating cash flows and revenues) and enterprise value. As with our pay-for-performance model, operating cash flow is replaced with: (i) tangible book value for companies in the Banks, Diversified Financials and Insurance sectors; and (ii) funds from operations for REITs, with the exception of Mortgage and Specialized REITs. Our projected cost is also measured on a per-employee basis. A plan will “fail” one or more of these criteria when the plan’s cost is significantly above the average of a peer group.

Program Features

We also consider these factors:

  • Does the plan permit the administrator to reprice, exchange or buyout underwater options?
  • Do interested parties administer the plan?
  • Does the plan have an evergreen provision?
  • Does the plan have a reload provision?
  • Does an excessive proportion of awards (greater than 70%) go to a company’s top executives?
  • Does the company provide loans to employees to exercise options?
  • Has the company engaged in repricing or an option exchange in the past 3 years?
  • Does the plan have a single-trigger change of control provision?
  • Does the plan contain an “inverse” multiplier or fungible share reserve that counts options as less than one award under the plan?

Equity-Based Compensation FAQs

Q: Why does the number of outstanding shares differ between the Company Profile page and the analysis of the equity plan?
A: The figure on the Company Profile page is as of the “Closing Price” date listed on that page. The analysis of the equity plan proposal is as of the company’s fiscal year end date.

Q: How does Glass Lewis calculate “potential dilution based on shares requested”?
A: Potential dilution = shares requested / (shares outstanding at FYE + shares requested)

Q: How does Glass Lewis calculate “total potential dilution” or “overhang”?
A: Total potential dilution (overhang) = (shares requested + all awards outstanding + awards available for future issuance) / shares outstanding at FYE

Q: How does Glass Lewis calculate its “3-year average burn rate”?
A: This is a simple three-year average of a company’s gross annual dilution, measured for each year as: (Options granted + FV awards granted)/ (Shares outstanding at FYE)

Q: How does Glass Lewis incorporate a company’s burn rate and total potential dilution into its equity plan analysis?
A: As described in more detail in “Equity Compensation Analysis,” our analysis does not include a burn rate comparison or an absolute limit on total potential dilution;  the display of “burn rate” and “run rate” in our analyses is for informational purposes and does not affect the scoring in our quantitative model.

Rather, we measure a company’s dilution using a “pace of historical grants” test. This measure of dilution (which we refer to as “full-share equivalent grants”) is calculated as [((net options granted / full-value grant multiplier) + net full-value awards granted) / shares outstanding at fiscal year end].

While we do not set strict limits on total potential dilution (often referred to as “overhang”), we capture whether a proposed share increase is excessive through our “program size” analyses.

Q: Are there any plan features or company practices that will result in an automatic recommendation against an equity compensation plan proposal?
A: While we review plans on a case-by-case basis, plans that contain the explicit authority to reprice, exchange or buyout underwater options without shareholder approval will almost always lead us to recommend against the plan. Other features that will likely trigger an “against” recommendation include evergreen or reload provisions, as well as “fungible share reserves” that count options granted as less than one share under the plan. We refer to this feature as an “inverse” full-value award multiplier.