Page 207 - Bruce Ellig - The Complete Guide to Executive Compensation (2007)

Chapter 5. Salary 193

Company D
Survey Actual Pay % Survey

CEO $880 $900 102
Other four highest 2,500 2,100 84

Total $3,380 $3,000 89
Table 5-16. Aggregating pay, CEO and other four highest-paid executives (in thousands)

Breaking this question down, the problem is whether to project the survey data to the
date of intended adjustment and, if so, how much further the projection should be extended.
For example, assume the survey used for determining job grading has an effective date of
March; it is now September, and the individual pay increase using these new ranges will take
place in January (and will not be reviewed again until the following January).
It seems logical to first project the survey community to January 1. If it is assumed
that the survey community will be increasing pay at an annual rate of 10 percent, then it is log-
ical to add 7.5 percent to the reported rate (i.e., 10 percent times 3/4 year). Assume it is also
assumed that pay increases next year for the survey community will be increasing at a 10 per-
cent annual rate. There are essentially three choices: (1) Make no additional projection, recog-
nizing that pay will begin to slip competitively shortly after January 1 and will be trailing 10
percent by the following January; (2) increase the survey data an additional 10 percent (on top
of the 7.5 percent adjustment), thereby placing the company theoretically 10 percent ahead of
the survey on the date of adjustment, recognizing that this will deteriorate and by next January
will approximate community pay; or (3) take a compromise position between “no increase” and
“full increase” (e.g., use 5 percent, recognizing that this will place the company 5 percent ahead
of the community on the date of adjustment, equal to the community average at midyear, and
5 percent behind the community by the following January). It is a matter of company philoso-
phy regarding competitive position in the marketplace as to which of these positions to take.
Actions based on projections need to be adjusted (plus or minus) when the actual num-
bers are known. If the estimate proved too high, the next adjustment should reflect this over-
estimate. For example, assume the estimate a year earlier had been for a 6 percent market
growth, but the actual increase in pay proved to be 5 percent. The following year’s projec-
tion would be reduced by 1 percent to offset the incorrect adjustment. The same would apply
in reverse. If the estimate was 6 percent but it turned out to be 7 percent, then 1 percent
would be added to the estimated market growth. This type of adjustment means the data is
never off more than one year.

STRUCTURAL ADJUSTMENTS

While most companies use survey data to adjust their pay structures, they differ as to how
often they make those adjustments. Companies that review everyone in the same period usu-
ally adjust annually, while those that use individual review dates, such as anniversary date of
employment, may adjust on a quarterly or biannual basis. If a company does not update its
structure at least annually, it will face pressure to increase grades in response to pay increases
in the marketplace.