Essentials for avoiding overcompensating: executive pay practices and disclosure have come a long way, but many conscientious boards are still designing packages that are out of proportion to the results delivered.

• Author: Swinford, David N.
• Position: Compensation

You don't get better performance just by paying people more money--so why do companies throw money at executives for less-than-stellar performance? Historically, a major factor was the closeness between top management and boards of directors, which made many directors reluctant to say "no" to requests for more money or new programs.

[ILLUSTRATION OMITTED]

That day is over, but even many conscientious boards still find themselves providing payouts that are out of proportion to the results delivered. Where are they going wrong in the design and administration of pay-for-performance programs? We'll look at some of the most common pitfalls facing companies and suggest some remedies.

1 Don't be a slave to the data. The talent market is not Wal-Mart, it's Sotheby's--a place where value to the beholder is determined by more than price alone. Unfortunately, both management and boards tend to rely too heavily on compensation data in setting pay levels. The value of a particular executive's work and performance should be assessed relative to the company, the environment and the team--essentially, the level and form of compensation needed to attract, retain and motivate that executive in that particular situation.

Rather than relying on competitive practices alone, pay judgments must take into account the individual executive's experience, skill, performance and contribution; the company's performance relative to its peers/market; whether the peer job's organizational structure and role requirements are truly comparable; and how executive pay decisions affect employee relations overall.

After all, market "median" is simply the halfway point in the data--it's not an absolute standard. It's naive to assume that if you pay executives at the 75th percentile, they will deliver 75th percentile performance. It may be both necessary and appropriate to pay a long-time, high-performing executive at or above the 75th percentile to retain her services, but to pay another executive 8 percent less than the market median due to his experience and contribution.

2 Compare peer performance. The selection and use of peer group compensation data is fraught with peril. To ensure apples-to-apples analysis, competitors should be picked on the basis of multiple, clearly defined criteria: industry, size, complexity, business model/life cycle, performance and/or other critical scope. That became a major issue for the then not-for-profit New York Stock Exchange, where the board was severely criticized for including the biggest Wall Street firms in its peer group.

A closely related problem is the failure to compare a company's actual performance against that of its peer group companies, so that performance expectations and goals can be calibrated against the competition. Actual performance should be analyzed over a three- to five-year period (and total shareholder return, known as TSR, over a 10-year period). That makes it more likely that higher-than-median payouts are provided only for performance that is...