by Joseph E. Bachelder
This article is reprinted with permission from the
September 27, 2012
edition of the
New York Law Journal.
© 2012 NLP IP Company.
Targeting Total Shareholder Return Versus Creating Long-Term Value
It would seem quite obvious that long-term incentives should be designed to encourage long-term growth in value of an enterprise. Today's column discusses whether governance of long-term pay for executives is losing some of its focus on that objective. In particular, it examines the current emphasis on Total Shareholder Return (TSR).
TSR measures stock growth or decline (plus dividends) between two dates (sometimes trading day averages ending on corresponding dates are used instead). For example, on Dec. 31, 2012, the closing market price for a public company's stock could be measured against its closing price on Dec. 31, 2011. That, together with dividends, indicates the TSR for that year.
Compensation surveys indicate that TSR is a predominant metric (at least one survey indicates the number one metric) for performance-based long-term incentive plans.1 TSR also has been at the center of say-on-pay votes for the 2011 and 2012 proxy seasons. It has been the predominant metric of proxy advisors such as Institutional Shareholder Services (ISS) in recommending whether shareholders should vote for or against the executive pay practices at particular companies. Both in performance metrics under long-term incentive plans and in say-on-pay recommendations by proxy advisors TSR typically is expressed in terms of the TSR at the company in question relative to the TSR at a group of comparable companies.
There are numerous problems in the current widespread use of TSR as a performance target under long-term performance plans.
1. TSR at a particular company is not something within the control of a CEO or other executives at that company. CEOs cannot forecast their company's stock price three years—or even one year—from today.
2. The use of TSR as a relative measure of performance in setting a future performance target compounds the problem just noted.2 It compares a particular company's future TSR performance with that of a group of other companies or even a broad-based index like the S&P 500. If a CEO cannot control the future stock price of his or her own company's stock, it is difficult to see how that executive's performance is enhanced by comparing the company's stock price levels with price levels for other companies three years in the future. (See below for use of TSR as a modifier of a primary target rather than itself being a primary target.)
The point made in the preceding paragraph is discussed in a recent article by Todd Sirras and Barry Sullivan of the Semler Brossy Consulting Group.3 They discuss the frequency of disconnect that often occurs between financial performance and shareholder returns at any particular point in time "in some cases due to exogenous factors, and in others due to increasing investor expectations. These disconnects can become compounded when measuring relative market performance, because the factors and variables for one company extend to all companies within the comparator set."
The article explains the compounding effect of the disconnect when relative TSR is used:
A three-year cycle is the most common measurement period used in long-term incentive plans. It is clear then that relative TSR-based incentives can become divorced during the measurement period from the core financial and operational performance that executives manage directly. For this reason, relative TSR-based incentives are often perceived as something outside of the executive team's control—"manna from heaven" when they actually do pay off—rather than a reward earned for specific achievements. In good years from a TSR standpoint, management can be rewarded even if the underlying financial performance or strategic achievements are not as strong, and similarly management can receive little or no reward if stock prices are down sharply—or even just down relative to their peers—despite above average financial performance.
3. Targeting stock price at a single point in time (even if measured over 10 or 20 trading days) is a questionable way to motivate performance designed to increase the long-term value of an enterprise. It is generally acknowledged that "short termism" was a major contributor to the problem that erupted into the financial crisis in 2008. Stock prices in public markets are the ultimate example of short-term behavior.
In an interesting paper, Professor David Larcker of Stanford Business School has made some perceptive comments on the difficulties in determining whether a CEO's compensation is too high or too low.4 According to that paper:
The only way to make [such] a determination is to measure whether the total compensation received is commensurate with the value of services rendered. To do this, stakeholders need to understand 1) the value drivers of the organization, 2) the impact that the executive has on these value drivers, and 3) the percentage of value created that should be appropriately offered as compensation for performance. While this is simple in theory, it is exceptionally difficult in practice.5
Looking at Larcker's first point, what are the "value drivers" of a business enterprise? They vary, of course, from company to company. They include financial and non-financial criteria. In the thinking of the author of this column there are two very important characteristics of these criteria: (a) They are internally generated and applied to the business conducted by the enterprise. (b) They develop long-term enterprise value.
TSR is not itself a long-term value driver. (Professor Larcker's article does not get into this point and the view expressed here is the author's own.) TSR is a result. As already discussed, it records a company's stock price on a specific date, or dates, compared to that company's stock price on an earlier date, or dates.
Examples of long-term financial value generators based on financial performance are growth in revenues, growth in earnings, and return on invested capital. Non-financial metrics include product research and development, human resource development and succession planning.6 The specific value generators most appropriate to a business enterprise at a particular point in time obviously vary considerably.
In a recent federal district court case involving a complaint brought against directors and certain officers of Navigant Consulting Inc., a Delaware corporation, for breach of fiduciary duty in approving "excessive executive compensation," the judge dismissed the complaint and recognized the importance of not ignoring factors, set forth in Navigant's proxy statement, that drive enterprise value.7 The complaint was brought after a negative say-on-pay vote by Navigant's shareholders. The court noted that "the complaint…focuses solely on the company's performance and shareholder return, which is only part of the equation." The opinion goes on to point out, among other things:
[T]he Proxy upon which Plaintiff relies further directs the Compensation Committee to look at additional factors in determining its compensation recommendations. The Proxy specifically describes other performance aspects that the Compensation Committee considered for 2010, including "net income and earnings per share," "strategic investment in core growth practice areas; senior level recruitment; and management's timely and effective response to changes in the competitive landscape." (Id. 50; R. 34-3, Proxy Statement at 16.) Additionally, it is the Compensation Committee's practice to consider "individual performance in the area of the company over which [the executive] has direct responsibility" and "his or her individual contributions to the company's financial and strategic performance for the year in question." (Id. 34; R. 34-3, Proxy Statement at 15-16.). Yet, the Complaint does not contain any allegations regarding these other factors. Because Plaintiff's allegations do not contain any facts regarding any of the other factors that the Committee considered in determining executive compensation, she has not established that Navigant violated company policy in awarding the executive compensation at issue in her Complaint.8
There are many cases in which CEOs have received large equity awards at a time when TSR was poor or even negative. One example is the award by Apple Computer Inc. (now Apple Inc.) to Steve Jobs in March 2003 of approximately $75 million of restricted stock in exchange for Jobs' agreeing to the cancellation of stock options totaling almost 30 million shares, all of which were under water. On March 19, 2003, the date on which the restricted stock was granted, Apple's one-year TSR was a negative 40 percent. Its annualized three-year TSR for the period ending on that date was a negative 38 percent. Obviously, the Board of Directors saw the value-generating potential of Steve Jobs—something that turned out to produce immensely more value than the cost of the restricted stock grant exchanged for the underwater stock options.
Appropriate Use of TSR
TSR can serve as a modifier of a primary target. To illustrate, assume a performance share award has a pre-set cumulative earnings per share as the target. The award might provide that notwithstanding achievement of the earnings per share target the award might be modified if the company's TSR relative to that of a selected group of companies falls below a certain level. The use of TSR as a modifier protects shareholders if achievement of a target such as growth in earnings is not reflected in the company's share price.
The use of TSR as a "lid" reflects the third point in Larcker's paper: determination of the portion of value created for shareholders during a particular period of time that should be provided to executives for their services during that period. If, notwithstanding achievement of targets that generate long-term value in the enterprise, the public market does not reflect, in relative TSR, the achievement of those targets, perhaps a limit on the payout would be appropriate. An alternative to forfeiture or reduction of the award would be to defer realization of the award subject to TSR returning to a satisfactory level.
ISS's Use of TSR
ISS has used TSR as a metric based on stock price data attributable to the past—meaning periods ending before the awards being evaluated by ISS have been earned out. Put another way, ISS is looking backward at where TSR has been in prior periods in determining whether grants of long-term incentive awards to be earned out in future are appropriate.9 This practice creates two problems. The first problem is that it compares apples and oranges: (a) historic stock price levels and (b) awards of long-term incentives to be earned out by results accomplished over a period of years in the future.
The anomalies of this comparison are compounded when relative TSR, on which ISS bases much of its analysis, introduces past TSRs at one or two dozen other companies being compared to the company under scrutiny. The second problem, discussed earlier in the column, is that evaluation of long-term incentives in terms of snapshot stock price comparisons like TSR is to place variable, short-term features of stock prices ahead of long-term value generators such as earnings, return on capital and others noted above.
TSR is being misused as a target in long-term incentive plans and as a metric for determining whether shareholders should approve or disapprove an executive pay program for purposes of say-on-pay votes. Long-term value drivers, both financial and non-financial, should be the primary targets in long-term incentive awards. TSR, if correctly applied (no apples and oranges) may be an appropriate "after the fact" measure in determining whether the amount of long-term pay otherwise earned out based on targets that generate long-term value should be modified if shareholder value during a given period has fallen short of expectations.
1. See "Measuring Long-Term Performance in 2011: An Analysis of the Prevalence and Types of Metrics in S&P 500 Long-Term Incentive Plans," published by Equilar (June 2012), at p. 3. Document available at http://www.equilar.com/knowledge-network/research-articles/2012/pdf/equilar-ltip-metrics-article.pdf.
2. For recent data reflecting use of TSR on both a relative and absolute basis, see "The 2011 Top 250—Long-Term Incentive Grant Practices for Executives," published by Frederic W. Cook & Co. Inc. (October 2011), at p. 13. Of the 250 companies using TSR, 93 percent used TSR on a relative basis and not on an absolute basis. Another 5 percent used TSR on both an absolute and relative basis. Document available at http://www.fwcook.com/alert_letters/2011_Top_250_Long-Term_Incentive_Grant_Practices_for_Executives.pdf.
3. "The Problem With Relative Total Shareholder Returns," published by WorldatWork (May 2012). Document available at http://www.semlerbrossy.com/wp-content/uploads/2012/05/The-Problem-with-rTSR.pdf.
4. Larcker's paper appears under the heading of Stanford Graduate School of Business, CGRP-23, Jan. 24, 2012. It notes that "Professor David F. Larcker, Usman Liaqat, and Brian Tayan prepared this material as the basis for discussion." The paper is available at http://www.gsb.stanford.edu/cgrp/research/closer_look.html.
5. Id. at p. 2.
6. Larcker and Christopher D. Ittner, professor at the Wharton School of the University of Pennsylvania, authored an article, "Coming up Short on Nonfinancial Performance Measurement," Harvard Business Review (Nov. 2003). Document available at http://hbr.org/2003/11/coming-up-short-on-nonfinancial-performance-measurement/ar/1.
7. Gordon v. Goodyear, No. 1:12-CV-00369, 2012 U.S. Dist. LEXIS 97623, 2012 WL 2885695 (N.D. Ill. July 13, 2012).
8. Id. at *35-36.
9. See NYLJ columns Dec. 1, 2011, and March 23, 2012.