Shareholders and shareholder advisors expect that executive compensation programs will demonstrate an alignment between executive pay levels and company performance. In 2017, shareholder views of the pay and performance relationship are likely to be the single biggest driver of shareholder voting on the Say on Pay proposal—and potentially on compensation committee director nominations.
Unfortunately, there is still not a clear standard that shareholders and shareholder advisory firms apply when evaluating pay and performance. In fact, the approaches that the shareholder advisory firms use to assess pay for performance continue to present many challenges, even with potential changes coming. As a result, it is critical for companies to be clear in defining how they view performance and to demonstrate that the pay levels executives receive are consistent with performance results.
The definitions of pay as well as performance used by shareholder advisory groups in their pay for performance assessments are imperfect. Examining these definitions can help provide direction on what a better approach may be and to better address their evaluations of your pay programs.
To date, Institutional Shareholder Services (ISS) and Glass Lewis have used Summary Compensation Table (SCT) compensation as the definition of pay in their quantitative assessments of pay (with slight modifications for valuing stock options in the case of ISS). Because the amounts reported in the SCT are grant values, this approach is fundamentally flawed as the majority of CEO pay (in many cases as much as 75% of total pay) is delivered in equity-based long-term incentives.
Since the point of long-term incentive compensation is to reward executives based on the mid- to long-term performance of the company, it is much more useful to look at the value realized from long-term incentive compensation as it incorporates changes in the stock price from the date of grant to the date of vesting for restricted stock units and performance share units. For performance share units, looking at the value realized also incorporates the number of shares actually earned, which could in many cases range from 0% to 200% of the target shares.
Stock options remain problematic, as the time from grant to exercise can be as long as 10 years, but using the grant value of options is less relevant for a long-term pay for performance assessment. In our view, the most reasonable approach is to use a realizable value for stock options, recognizing the change in the value of the stock price from the grant date to the end of the performance period under assessment. An approach of using the intrinsic value of the stock options or an updated Black-Scholes value is superior to using the grant date value of the stock options.
ISS has historically used total shareholder return (TSR) on an absolute basis for a five-year period and relative to peers over a three-year period in their quantitative assessments. Glass Lewis looks at performance for five measures over a three-year period for their relative assessment: TSR, earnings per share (EPS), change in operating cash flow (with a substitute measure for financial services firms and REITs), return on equity (ROE) and return on assets (ROA). For the 2017 proxy season, ISS incorporated into its qualitative review of pay a financial performance assessment with up to seven measures considered: TSR, ROE, ROA, return on invested capital (ROIC), cash flow from operations, revenue growth and EBITDA growth. The weightings of the measures are customized by industry in recognition that some measures are not applicable to certain industries.
While the move to include financial performance measures into the pay for performance assessment is a positive change, there are two main issues with the approaches of ISS and Glass Lewis. When compensation committees and management teams design incentive plans, they typically select performance measures that best reflect the company’s success in achieving its strategic objectives. These are the measures that should be used to assess performance over a one- to three-year period.
Over the long term (e.g., five to seven years), TSR is likely the single most important measure of how well the company has implemented its strategy. However, overemphasizing TSR in a one-year or three-year assessment of performance is too short a period of time to see if the company’s successful execution on strategy has led to financial results and subsequent shareholder returns. We think that the most effective performance assessment would place the most weight on the measures of performance that the company is actually using in its annual and long-term incentive plans.
Another issue with using financial performance measures to assess performance on a relative basis is making sure that the comparisons reflect the sustainable operating performance of the companies being compared. ISS and Glass Lewis both use the GAAP financial statements to derive the financial performance measures used in their assessments. In one sense, this facilitates “apples to apples” comparisons across companies. However, most external parties (e.g., investment analysts) who conduct financial performance comparisons will look beyond GAAP accounting figures to strip out “special items” that are unique to one year and are not expected to be repeated. Over time and across a large sample of companies, “special items” may balance out; however, when calculating point-to-point growth or single-period performance, “special items” can have a potentially misleading (either positive or negative) impact on financial performance comparisons.
We think that using GAAP accounting figures is a reasonable shortcut in conducting the pay for performance analysis; however, ISS and Glass Lewis should do an in-depth analysis of the financial statements of companies included and should be careful in drawing conclusions if any of the years in the performance assessment included major special items (e.g., gain or loss on sale of the business, asset value write down, legal settlement, change in accounting policy, etc.). Companies may want to consider showing in the proxy statement what the differences are between GAAP performance and the definition of the measures for incentive plan purposes and provide a supporting rationale for why the committee thinks the adjusted performance measures make sense for purposes of the incentive plan design.
The importance of pay for performance relationships in determining Say on Pay votes makes it an issue that management teams and compensation committees cannot ignore. It is important to understand the inherent flaws in the way that shareholder advisory firms assess the pay and performance relationship. With an understanding of these imperfections, it is critical for companies to conduct their own assessments of the pay and performance relationship, including what period of time defines long-term (e.g., three to five years, CEO tenure, etc.) and to demonstrate in their Compensation Disclosure & Analysis that the company does have an alignment between the performance measures in its incentive plans and the pay levels that the company has established. In addition, the company should be able to demonstrate that over the long term, the movement in realized/realizable pay levels is consistent with its TSR performance on an absolute and relative basis.
Eric Hosken is a Partner with Compensation Advisory Partners in New York. He can be reached at eric.hosken@capartners.com.