If they have objective measures and are compensating based on them, there's no reason not to tell people what the formula is. You can still hide what you measur

If they have objective measures and are compensating based on them, there's no reason not to tell people what the formula is. You can still hide what you measure people at (Bob's only A+ at LINQ, vs. A+++). Without specifying how measurements translate into yearly compensation they're just waving their hands around and saying how scientific they are without actually being so.