Pay For Performance: Beating “Best Practices”

Widespread criticism of CEO pay packages have spurred directors to engage in a diligent search for best practices. This vigilance is transforming the process of executive compensation design, administration and oversight at many major public companies. But have all these process changes improved the compensation plans?

We conducted empirical research on the way various compensation structures work for or against shareholder value creation. We looked at S&P 500 executive compensation plan data, supplemented by conversations with hundreds of executives and consultants. Against this standard, the evidence indicates that certain practices prove out favorably; some with plausible rationales have questionable value, at best, and some are clearly counter productive.

CEOs must weigh a plan’s costs against the degree to which it creates alignment between managers and owners. They also need to trade off the accountability one gains from objective standards versus the ability of any objective standards to be “gamed” by managers, or the need to encourage long-term thinking versus the requirement to evaluate managers each year or quarter.

No bonus plan is immune from these trade-offs, and no plan can make them optimally in the same way for any two firms, but that doesn’t mean that anything goes when it comes to compensation. Certain practices became widespread simply because someone besides the shareholders stood to gain from their adoption and the board had no reason to doubt their value. It’s time to re evaluate those practices in light of the evidence:

1. Motivation without alignment.

Companies that pay their managers for revenue growth tend to see impressive revenue growth. Companies that pay for return on capital tend to have well-above-average returns on capital. Directors believe their incentive plans are working. But are these plans working for the shareholders? If the focal measure is something other than profit, the answer is no.

The 67 companies in the S&P 500 that used revenue as a main driver of bonuses in 2002 to 2004 grew their sales about 7 percentage points faster than their peers over that period, but underperformed by 3.7 percentage points per year in total shareholder returns. Profitability is related to revenue, but the opposite is not necessarily true. Paying for, and getting, revenue or sales growth is a good example of motivation for managers without alignment with the shareholders.

Paying for any objective measure, including profit, cost-cutting, market share and customer satisfaction, invites “gaming the system.” It achieves the primary, intended effect (i.e., improvement in the result being measured) but also secondary, unintended effects, including manipulation of results, short-term behavior and risky behavior.

Still, paying for profit can be a rewarding strategy for shareholders. The third of S&P 500 companies that pay primarily for profit outperform their peers in total shareholder returns over time by nearly 2 percent per year. The lesson? Paying for objective measures is challenging enough; there is no point in compounding that challenge by trying to pay for a focal metric not closely related to value.

2. Paying for inputs instead of results.

Many companies adopt the best practice of tying incentives directly to a company’s business strategy. In practice, this means paying for multiple financial or non financial metrics, or leading indicators of value creation, such as customer satisfaction, employee safety or diversity.

How well do these plans work? Managers say that multiple measures invariably translate into no priorities, channel short-term behavior in several directions, are often irrelevant shortly after the planning cycle that produced them, and make senior managers less empowered. “I feel like a rat in an elegant maze,” is how one manager described working under the balanced-scorecard bonus plan, a sophisticated version of this concept. “Our compensation plan lends new meaning to €˜who moved the cheese?'” Also, leading indicators don’t always lead to valuable results; paying for them often means little accountability for results that owners would care about.

Companies that base their incentives on a balanced scorecard invariably trail their peers in total shareholder returns. About 15 percent of the S&P 500 have objective, multi- etric incentive plans. On average, these companies under perform their peers by 3.5 percent per year. Whatever their virtues, these “elegant mazes” simply don’t create the value of a plan that aligns executive pay with actual profit results.

3. Alignment without motivation.

 Most boards believe that rewarding managers through stock options is an effective incentive leading to long term shareholder value. Equity ownership, by definition, aligns managers and shareholders. But effective incentive implies a motivation to do something, as opposed to a simple desire to see the share price go up. Most senior executives, right up to the CEO, will tell you that movement in the stock price over several years usually has more to do with exogenous factors than with the actions of management.

Management actions do matter, but they matter indirectly, over a long period of time and in often unpredictable ways, all of which tend to make equity ownership a weak motivator. While the alignment provided by equity ownership does motivate managers to seek out the drivers of shareholder value, equity by itself is inherently unable to provide a sufficiently detailed guide for value-creation to managers.

4. Equity grants that achieve the opposite of alignment.

Fixed-value grants are based on a target level of pay designed to keep the overall compensation package competitive.For example, if competitive considerations mean $100,000 worth of options per year, then that value would determine the number of options granted each year. Annual compensation in equity may look benign, but it’s not. The practical result of fixed-value equity grants is to penalize managers for increases in the stock price and to reward them for drops in the share price.

If the stock price goes up, providing the same value in options as the prior year means granting fewer options. On the downside, as the stock price drops, the number of options climbs. From 1999 to 2004, for example, Calpine’s stock price dropped by two-thirds, requiring the company to triple the rate at which they granted options to managers to remain competitive. As a result, top management eventually accumulated so many additional options that if the stock simply recovered to its 1999 price, the CEO alone would have been ahead by $20 million.

To varying degrees, about two thirds of the S&P 500 grant non performance equity in this manner. There are alternatives to fixed-value grants, such as front-loaded equity (i.e., with long-vesting periods), a more level number of shares over time or indexed options. Research shows that companies with better grant policies out perform the rest.

5. Long-term plans undermining long-term performance.

In an attempt to overcome the devilish short term, long-term problem associated with objective metrics, about a third of the S&P 500 adopted long-term incentive plans or LTIPs. Generally the second or third incentive plan from which a given executive can derive rewards, LTIPs are typically three-year plans linked to multi-year goals, which may or may not be the same goals in the short-term plan.

So, how do separate short- and long-term plans motivate managers and serve the shareholders? Over a period of years, not one of thousands of managers surveyed informally about long-term plans reported behavior reflective of the intended incentive effect. As one former engineer at an aerospace defense firm said, “It’s not like I do the math.”

In the minds of managers, if the LTIP is based on accounting measures, it makes no difference that its targets were set two or three years ago versus last January. If the LTIP is based on three-year shareholder return, managers don’t feel they have any impact on the outcome. What advantage is a plan that provides no additional motivation? According to the S&P 500, there is none. In fact, nearly two-thirds of companies with LTIPs based on three-year cycles did worse than their peers over the last three years.

6. Paying managers to lie.

Perhaps the most common practice in compensation is deriving incentive plan targets from annual budgets. Jack Welch called the budget process “an enervating exercise in minimization” encouraging business teams to negotiate “the minimum number they think they can €˜sell.'” In addition to minimization, budgets discourage stretch goals.

While you can’t run a business without a budget, the use of annual budget goals as compensation targets corrupts the budget process and encourages mediocrity. Yet, often managers, directors and their compensation experts don’t even know there is another way, particularly where the main nexus between HR and finance is through the budget process.

But anyone comfortable with finance can easily infer the market’s (i.e., investors) profit expectations from a combination of the stock price and current operating financial performance. These expectations are, in fact, easier to infer over multiple years than for a single year. For example, the higher a company’s P/E ratio, the higher the market’s expectation of that company’s earnings growth.

Budget-based target setting is so prevalent that it’s difficult to quantify the penalty paid by firms who engage in it. But companies that comfortably set targets based on year-over-year improvement-or expected improvement derived from market data-are among the best performing firms in the S&P 500.

7. Incentives to limit performance.

When Tyco brought on Ed Breen and a new board to clean things up, an early step was to dramatically lower the caps on executive bonuses. The bonus “now caps out at 200 percent of base salary,” noted Breen, “whereas before it was more like 600 percent or 700 percent.” As best practices go, caps on bonuses are almost universal.

It’s easy to see why. If you approach any bonus plan as something o be gamed by management or driven as much by luck as skill, then a cap seems a logical control. It’s a defense against poorly designed or poorly calibrated bonus plans, or plans whose targets were sandbagged by management. If the measures driving management pay don’t make complete sense, you certainly don’t want management maximizing on such metrics. Also, there are certain businesses where the long term trade-offs make current year maximization an unsafe practice, such as in real estate.

Boards are convinced that caps are good for shareholders because they realize they don’t have incentive plans that can safely be uncapped, and they can’t imagine how it could be otherwise. Yet, there are companies that operate without bonus caps.

They overcame the rationale that we shouldn’t reward gaming or good luck by effectively deferring a portion of bonuses arising from exceptional performance with mechanisms that can claw them back if that performance turns out to be unsustainable. The evidence shows that such efforts are well worth it. Companies without caps on their bonuses-and which have taken care of the preconditions for safely removing them-perform far better than their peers.

The Entrepreneurial Model

There is one plan that provides legendary wealth for both managers and shareholders- the entrepreneurial model. It’s straightforward: Management gets a fixed share of the profits generated by the business or a fixed percentage of the value of the business. The personal wealth of a Bill Gates or a Warren Buffett is a direct reflection of the profitability of their businesses. Nobody misunderstands the source of that wealth, and few begrudge it.

Consistent with an entrepreneurial model, the most effective plans appear to be profit-based, objective and stable-often with some subjective element in the distribution, if not the funding, of bonuses. For example, in 1965, Nucor’s then president Ken Iverson instituted a plan that funded a bonus pool based on a share of the profits above a threshold level of profitability (what we would now call an economic profit, or EP, plan).

The plan’s integrity came from three sources: First, a discipline for setting the threshold profit, which was based on capital invested and past performance, provided accountability for new capital, and ensured that man-agement rarely got paid twice for the same profit achievement. Second, Nucor had one plan for senior executives, and it paid out in cash and stock. There was no opportunity to make up with a second plan what they failed to achieve in the first. Third, and perhaps most important, the plan was stable. That fixed share of profitability was there whenever the profits materialized, whether it took one year or 10.

This plan survived for 37 years. As far as Nucor’s managers were concerned, it motivated and rewarded long-term performance. After several years with flat profits, Nucor’s plan paid out handsomely after management took a struggling joist company and turned it into a highly profitable steelmaker. In its fairness, simplicity and integrity, it may be the most successful incentive plan ever created

Elements of Integrity

Many critics argue that paying managers using objective measures induces them to steal from tomorrow or fudge numbers. But this is far more likely with incentive plans subject to change each year. With stable, profit- based plans, there is little point to gunning it for the short term. The integrity of the plan rewards the integrity of management.

Ball Corp., which pays managers based on a fixed share of economic profit (EVA) growth, provides an example of a more modern, corporate version of the entrepreneurial model. This plan has been in place for over a decade now, through ups and downs in their business cycle.

A handful of companies in the S&P 500 provide incentives that properly motivate and align their managers in a way that avoids all seven of the aforementioned pitfalls. hey have entrepreneurial incentives hat are uncomplicated for anagement, despite target-setting nd claw-back mechanisms. Sometimes criticized as unsophisticated and expensive plans, these companies have the ultimate answer to the critics who accuse them of being behind the times, or profligate, or worse: They significantly outperform their peers, on average, by over 5 percentage points per year.

Many people underestimate the unpredictable effects of their complex plans that have typically evolved over many years. They think that having average plans will save them from criticism for average performance, or worse. Comparing widely accepted trends with actual data suggests some guidelines that are far more effective:

•  Pay for profit. Incentive plans based on an objective profit measure, i.e., profit with a specifically defined threshold that accounts for capital.
•  Keep plans simple. You don’t need three. You may not even need two. One plan with the proper target bonuses and a deferral mechanism can fund all your incentives.
•  Use equity as a reward, not as a motivator. Your accounting-based plan can pay out in equity as well as cash. Shareholders benefit from having managers own equity.
•  Avoid fixed-value grants. If you offer equity to supplement salary, offer it in a front-loaded or fixed-share basis, or as indexed options.
• Avoid using planning targets for any executive compensation purposes. Use improvement, peer based or market-derived targets for executive incentive plans.

This last point suggests perhaps the most valuable area of compensation decisions. Currently, most boards are content with being at the top of the budget negotiation. They ask the CEO if that is the best the corporation can do, sign off on targets when they can’t argue otherwise, then certify at the end of the year that the targets were met or not in order to authorize payments, sometimes exercising discretion to protect themselves from poorly designed or calibrated plans. This hardly reinforces compensation integrity.

Directors belong at the intersectionof management and the shareholders, not in the bowels of a planning discussion. Public company boards can independently determine whether the financial results promised by the long-term plans are consistent with observable shareholder expectations. Whenever this is done now, it’s often treated as a finance issue independent of the budget, not an HR issue that can inform incentives.

Ball Corp.: Incentives That Pack a Punch

Most packaging companies pay executive bonuses based on profits or earnings. Ball Corp. goes one better by paying executives for profit growth in excess of their cost of capital, or EVA. Furthermore, Ball pays for EVA improvement over time-not profit versus a budget that is negotiated every year and may ignore new capital. Finally, Ball’s managers are assured that whatever the EVA growth, whenever it materializes, they will get a fixed share of it, without limit. That gives Ball’s shareholders something 95 percent of all other shareholders don’t have-managers’ interests almost perfectly aligned with their own.

That alignment was a comfort to shareholders when the company announced the $600 million acquisition of U.S. Can’s American and Argentinean businesses last February-the kind of deal that generally leaves investors skeptical. The acquisition was EPS accretive, but it was the fact that it would grow EVA that allayed investor skepticism. EVA growth is a more reliable indicator of value creation than EPS accretion, and EVA growth is exactly what Ball’s managers are paid to achieve. The simple fact that Ball’s incentive plan has been in place, essentially unchanged, since 1993 provides much of its discipline and long-term perspective. It leaves little hope of forgiveness in the form of new metrics or renegotiated targets if, say, a major acquisition goes sour. There is plenty of evidence that such a high degree of alignment works. Ball, for example, is one of the best performing packaging companies in the world.