With Pay-for-Performance, It's the Performance That Really Counts

As the 2012 proxy season gets underway and companies scramble to craft executive pay packages that will avoid a “no” vote from shareholders, evidence is mounting that investors don't care much about all that effort.

According to one large activist investor and data from a recent survey, investors may be voting against pay plans during say-on-pay votes at companies that have bombed in the last year, regardless of the details of their executive pay plans. In other words, they aren't linking pay with performance; they're just lashing out at poor performance, period.

Tim Smith, senior vice president at Walden Asset Management, says some investors lack the ability to research a company thoroughly and to understand any long-term effort to create value that might be going on. They see poor performance and vote against the pay packages, without considering the details of those packages—including whether the pay is intended to revive the company's fortunes.

“They might look at a stock price … and say ‘I'm not doing that well and yet the top manager is going home with a pay package.' They might, get frustrated, vote no, and go home,” Smith says.

A recent survey from benefits consulting firm Towers Watson suggests the practice could be fairly common. According to the survey, 9 percent of companies considered to have relatively low-paying compensation plans still couldn't muster even 70 percent support among shareholders. At companies with mid-range compensation plans, 19 percent of them failed to hit the 70-percent threshold. All that implies that shareholders were unhappy with the stock performance at these companies and voted against the pay plans, even though the CEOs weren't paid all that much.

What does matter to shareholders as they ponder say-on-pay votes? Return on investment, by far. The Towers Watson survey found that poor-performing companies were more than three times as likely to receive less than 70 percent shareholder support on compensation packages, compared to their peers with medium or good performance. (The study was based on 2011 proxy voting numbers and pay data from public filings for 728 companies between 2008 and 2010.)

“The connection of low performance with low say-on-pay support wasn't a surprise,” says Todd Lippincott, leader of Towers Watson's executive compensation consulting business for the Americas. “But the extent to which high pay opportunities translated to such a significant reduction in support of pay programs—the magnitude of that impact—surprised us.”

Lippincott says high pay and poor performance are the two main drivers of a no vote, but performance is the stronger factor in the decision.

Smith says shareholders are increasingly checking to make sure pay is commensurate with company performance. “As you look at other indicators of a company, like long-term value or just that the company's not doing terribly well but the CEO is laughing all the way to the bank, pay-for-performance criteria comes in,” he says. “Overwhelmingly, it's the criteria investors across the spectrum use to evaluate whether they'll vote for the pay plan.”

Hence, even though shareholder votes on pay are non-binding, companies that flunk the votes would still do well to amend their plans and shore up investor support. (Not that amending the plans will guarantee more shareholder support.) According to the Towers Watson study, 78 percent of companies that cut pay for 2010 received acceptable shareholder support (above 70 percent), compared to 74 percent of those that kept pay at higher levels.

About 40 companies failed their say-on-pay votes last year, but Lippincott says many are improving this proxy season. “Our early returns suggest there's continued moderation in pay, so companies that had a lower say-on-pay vote last year are much less likely to have an increase on CEO pay for this cycle,” Lippincott says.

“As you look at other indicators of a company, like long-term value or just that the company's not doing terribly well but the CEO is laughing all the way to the bank, pay for performance criteria comes in.”

—Tim Smith,

SVP and Director, ESG Shareholder Engagement,

Walden Asset Management

Examples

Hewlett-Packard is one such company that won over shareholders. The company flunked its 2011 say-on-pay vote (a support level of only 48 percent), but spent the last year reaching out to large investor groups on the issue. HP then adjusted its executive compensation range to the market median, rather than in the 75th to 90th percentile where it was previously. This year's shareholder support then jumped to 77 percent. (Although, firing unpopular CEO Leo Apotheker and replacing him with Meg Whitman no doubt helped.)

“During fiscal 2011, we engaged in discussion with institutional investors and proxy advisory firms to gather feedback about [Hewlett-Packard's] executive compensation program to address the concerns of stockholders,” HP spokesman Michael Thacker said.

The HP board's compensation committee restructured incentive compensation programs to limit the use of discretion, and redesigned the incentive plan to reinforce links between pay and performance. The company also eliminated tax gross-ups (always unpopular with investors) for some officers, amended the executive severance plan to fall in line with the market, and increased talent management efforts to focus on the development of internal talent rather than pay large signing bonuses to lure outside executives from elsewhere.

CEO PAY & SHAREHOLDER SUPPORT

The excerpt below is from Towers Watson's analysis of 2011 say-on-pay voting results for 728 mid-sized and large companies:

Towers Watson found that almost one out of three companies (32%) with high CEO pay opportunities (in the top quartile) received low say-on-pay shareholder support (below 70%) in last year's proxy season, compared to only one out of five companies (19%) with CEO pay opportunity at or near the median (see figure below). Similarly, companies that performed poorly (bottom one-third in total shareholder return [TSR]) were more than three times as likely (34%) to receive less than 70% shareholder support for say on pay, regardless of their pay levels, than were companies with top levels of TSR (10%). The findings are based on a Towers Watson review of pay data of 728 companies from publicly available proxy filings from 2008 ? 2010 and a review of shareholder voting results in 2011.

“While say-on-pay votes primarily reflect absolute levels of pay for companies with high pay levels, they can become say-on-performance votes when companies do poorly in generating shareholder returns,” said Todd Lippincott, leader of Towers Watson's executive compensation consulting business for the Americas. “The strong connection among say-on-pay outcomes, executive pay opportunities and shareholder returns indicates that other efforts like eliminating certain pay practices, such as change-in-control tax gross-ups, may have a limited impact on say-on-pay voting results. Based on our research, it appears that companies that target high pay opportunities run a much greater risk of unacceptable voting outcomes than companies that target median pay levels.”

The study found companies with high CEO pay opportunities in 2008 and 2009 received similar levels of shareholder support, regardless of whether they changed pay levels for 2010. Slightly more than three-quarters (78%) of companies that lowered pay for 2010 received acceptable shareholder support levels (more than 70%), compared to just less than three-fourths (74%) of those that kept pay at high levels. However, companies with median pay levels for 2008 and 2009 that increased their CEO pay opportunities to high levels during 2010 saw reduced shareholder support for their say-on-pay votes, with just less than two-thirds of those having done so receiving acceptable shareholder support.

“It will be interesting to see how say-on-pay voting plays out in the 2012 proxy season, given that many companies had strong operating results last year, while their share prices and shareholder returns were flat,” said Lippincott. “Clearly, our research confirms that shareholder votes are strongly influenced both by the sheer size of the pay opportunity and the sensitivity of pay to performance, which many investors equate with shareholder returns.”

Pay opportunity

Median pay opportunity

Currently Use

Percent of companies

receiving less than 70%

shareholder support for

2011 say-on-pay votes

High

$6.5 million

32%

Medium

$4.1 million

19%

Low

$2.0 million

9%

Source

Towers Watson.

Charles Elson, business professor at the University of Delaware, says companies routinely spend time tweaking their plans following a failed vote. “Hopefully a lot of companies anticipate the no votes and come up with plans to get a lot more yes votes. You need to work with your constituents and come up with a plan, so by the time you present it it's already been vetted,” Elson says.

Like Hewlett-Packard, Beazer Homes USA saw similar success the second time around.

The Atlanta-based company was one of the first to fail its say-on-pay vote last year, securing only 46 percent shareholder support, but it made changes this year to secure 95 percent support at its February meeting.

“In response [to the 2011 vote], our board of directors … completed an in-depth review of our pay practices. As part of this review, management contacted several major stockholders to better understand the reasons behind the votes against the compensation for our [named executive officers],” the company stated in a filing to the Securities and Exchange Commission.

According to the report, Beazer Homes brought its base salary down from the median or high pay target to the bottom quartile of the 2012 peer group, aligning CEO Allan Merrill's pay in the 25th percentile. His base salary is $900,000. Other executive officers fall between the 25th and 50th peer group percentile. It also states that Beazer will not approve any bonuses where the performance fails to meet certain predetermined levels, and the company established a long-term performance-based incentive plan for executives.

Beazer Holmes also adopted a policy to claw back compensation from its officers in select circumstances and retained new compensation consultants with in-depth knowledge of current best practices and peer group standards.

As the proxy season continues to unfold, Lippincott says say-on-pay is a blunt tool, and shareholders might vote against a plan for any number of reasons, regardless of pay or performance.

“Statistically we know low performance is the most significant, and high pay opportunities are there, but there could be other issues that we see. Some red flags are associated with potentially generous termination provisions, significant use of perquisites, or a unique business where a company hasn't distinguished itself in the handling of certain issues. There could be a variety of issues,” Lippincott says.

The perfect scenario, Lippincott says, is a high-performing company that sets CEO pay exactly at the median for his peers. Even so, high-performing companies with high pay still win strong levels of shareholder support—which also supports the theory that investors are more concerned with performance than any particulars about CEO pay.

“The overall big message is shareholders are strongly supportive of executive pay programs,” Lippincott says. “However, if you are a low-performing company, then they'll show a lack of support or opposition through say-on-pay.”