This is an overreaction. In the professions, superior performance is evident only over an extended period of time, and should be so appraised. A given year’s compensation decision should be seen as one of a series, not as a game to be won or lost each year, with no view to past or future. Rapid changes in anyone’s compensation, except where there is a clear consensus that they are merited, are extremely disruptive – as are, at the other extreme, fine distinctions made on the basis of small differences in a single year’s performance.

One way to balance past and present is to think of compensation awards as a smoothed, moving average of performance, updated every year as new evidence comes in. In order to accomplish this, a firm may choose to make compensation decisions on the basis of statistics covering a three-year period rather than a single year’s time. Each year the three-year cumulative totals for each partner are updated by adding in the most recent year’s results and dropping those from four years ago. Thus, a good year will continue to show up in the data for an extended period of time, but not forever. Similarly, the impact of one good for bad) year will be dampened when added in with four average years. If performance improves each year, the partner will have a gradually rising share. If it consistently declines, so will his share. (Incidentally, this device thus deals naturally with the gradually retiring partner.)

Excerpted from ‘Managing the Professional Service Firm’ by David Maister, pages 267 to 268