I received my fancy booklet containing the WorldatWork 2009-2010 Salary Budget Survey in the mail Wednesday. Our editor, Ann Bares has already discussed the initial findings about the budgeted increases in this post on her Compensation Force blog. I want to discuss the Pay for Performance section in the Executive Summary.
I think that if we’re going to call merit increases “merit”, then pay should be differentiated by performance; but I don’t know that we’re always doing such a good job of actually differentiating. According to the survey, employees performing in the middle received an average 3.1% increase in 2008, employees performing at the top received an average 4.5% increase, and low performing employees who received a raise got 2% on average. A 4.5% increase is nothing to sneeze at, but it’s less than 1.5 times higher than the middle-performers, and is only 2.25 times higher than the low performers.
Over the years I’ve come to believe that in most cases, employees who regularly and continually aren’t meeting performance expectations shouldn’t get increases. Chances are pretty good they shouldn’t be working for the company anymore either, but that’s another post. Also, withholding increases from employees who aren’t making an appropriate contribution to the business could potentially fund the increases for higher performers.
I don’t think it’s always the size of the increase that we give to top performers (which is expected to drop to 3.4% in 2009 according to the survey), as much as it is the delineation of the increase. If we tell our top performers that we value their exceptional contributions, we should walk the talk with an increase that backs that up. It doesn’t have to be huge, it just has to be different from the increase we give those who provide a regular contribution. Here’s a quick (oversimplified for illustration) example of how this could work:
In this situation you could give both of your top performers an additional 2% increase without changing the overall budget or taking anything away from your middle performers who make their own contribution. You can now tell your top performers that the increase they’re getting is more than 2 times higher than what your middle performers will receive. By putting your money where your mouth is; you’ll likely see even higher productivity from your top performers, at least for awhile. I still believe that incentive pay is generally a better way to pay for performance, but if we say we pay for performance in our merit increases, then we’d better make sure our top performers see more money than others in the organization.
Darcy Dees works as the Compensation Manager for Rock Bottom Restaurants, Inc., headquartered in Louisville, CO. She has been working in Compensation for over 5 years now and recently attained her Certified Compensation Professional (CCP) designation. She spends what little free time she has hiking and reading.

Amen to your thoughts on making meaningful differences in pay if there are meaningful differences in performance!
May I add another possible pay feature to the equation? With total salary increase budgets becoming so small, some have envisioned a greater mix of lump sum payments (aka, variable pay) to the equation. With lump sum payments in the equation you can make even greater pay differentiation for different performance levels. And there is no compounding of salary costs with lump sum payments.
The July 2009 issue of Workspan Magazine has a good article on this topic called "The Past, Present and Future of Variable Pay" (http://www.worldatwork.org/waw/adimLink?id=33404). I tend to agree with the author's prediction of a continued trend away from fixed cost increases to more variable pay adjustments.
It would be a simple matter to add lump sums into employers' merit matrices. Further, you could provide performance-based lump sum percentages to those who have salaries at or above their salary range - after all, they need to be reinforced for their contributions also.
Posted by: Paul Weatherhead | 08/14/2009 at 10:06 AM
Paul - Thanks for the comment and for keeping the discussion going.
I agree that variable pay is usually a better vehicle for rewarding performance than base pay increases. It's easier to create a differential in pay, and it doesn't become a guaranteed annuity payment if performance results decrease over time.
Posted by: Darcy Dees | 08/17/2009 at 11:05 AM