Office of Personnel Management Director Linda Springer and Defense Department Chief Human Capital Officer David S.C. Chu spoke recently about pay-for-performance at the Excellence in Government conference on Monday, May 12, 2008. 

OPM Director Springer tried to put pay reform into perspective.  "What we have is antiquated and we are ill-served. ...  The pay-for-performance effort began even before the Bush administration took up the issue.  ... Will the effort to continue this reform go on? In my judgment it will, because if it doesn't, we will lose the battle of recruiting talent on this issue. You can have all the tradition of public service and pins and plaques and awards you want, but if you go to the grocery store and put your pins on the belt, they'll probably call security."

CHCO Chu insisted that despite the highly visible objections to the Defense Department's new National Security Personnel System, it would remain in place in the next administration.  "I think the most important signal was given by Congress last year.  It could have decided to reverse NSPS. It did not. It did change the union bargaining system. That was a contentious point. But it did not change the underlying nature of the system itself."

Many members of Congress - both Republican and Democrat - were part of the initiative to establish pay-for-performance in the Senior Executive Service.  From what I have read of the regulations and OPM guidance, they tried to allow flexibility in design for the different agencies, but assure a handful of core principles.  I can't imagine a new President or Congress abandoning those bipartisan design principles.  Can you?     

I had a couple of informal conversations with Ron Sanders (who was then at OPM) when the 2004 regs were working though the approval process.  The conceptual model resembles best practice from industry.  There is nothing in the regs to criticize.

 However, when I see 88% of the covered executives rated at the two highest performance levels, its a red flag.  Of course they may all be great people -- I trust they are -- but these are very difficult jobs.  And the purpose of performance management is to define what is expected foer the job and at year end to assess how well the individual perfomed in light of the expectations.  At the executive level, the focus should be job-specific -- its not a beauty pageant where one executive is compared with others.  Each executive should be assessed on how well he or she met or exceeded the expectations for the job.

That makes the 88% suspect -- and the credibility question undermines confidence in the process.  Its not surprising a high percentafge of executives are satisfied with their rating!

Compared with industry practice, one difference stands out.  Industry emphasizes the notion of 'stretch' goals.  Those are goals that are difficult and challenging.  The idea of course is to stretch the individual's performance.  But with stretch goals, it is OK to miss a goal by a small margin. That's expected with dificutl goals.  Industry would not live with 88% rated at the higher levels.  The policy idea of a so-called 'forced distribution' which was adopted for several years by a number of leading corporations permitted only 20% to be rated at the 'exceeds' level.  This is not suggest government should try a forced distribution policy -- I think its a bad idea -- but the contrast in thinking is striking.

I appreciate that failing to achieve a goal has political implications.  That's a difficult problem to solve.  But there should be an incentive to exceed goals.  Industry celebrates and rewards executives who exceed their goals.

The UK senior civil service pay and performance system appears to be working much better.

I see Ron Sanders was the featured resource in the Washington Post's article about a new pay-for-performance system in the government intelligence agencies. http://www.washingtonpost.com/wp-dyn/content/article/2008/05/18/AR2008051801749.html

Glad to see agencies are proceeding with PFP reforms despite the recent criticisms of SES PFP, and despite the fact that a new President and possibly new agency leadership is coming next year.   

It looks like the pay for performance initiaitve in the Intelligence agencies is halfway home.  The final test comes a few weeks after pay increases are granted when employees have a chance to react to what happened.  On paper the GAO system looked exceptionally solid but we now know there were a couple of missteps.  The rollout process and the communications are probably more important than the actual program design.  Ron Sanders led the transformation initiative at IRS several years ago and I have to guess he learned a few lessons.

Pay for performance is great in theory, but generally is ineffective in practice.  The problem is that incentives work too well.  Specifically, they motivate employees to focus excessively on doing what they need to do to gain rewards, sometimes at the expense of doing other things that would help the organization.   Employees tend to get frustrated by factors out of their control (process failure, slow materiel delivery, weak supporting cast, etc.) and become frustrated when they do not attain the highest incentives.  Inconsistent management standards and unrealistic expectations may also hamper performance achievement, thereby creating more disatisfaction than performance.  Pay for performance works best for occupations in which the employee maintains nearly sole control of their production (ie produce 100 widgets, receive $100). 

Log Dog,

Thanks for your recent comments on the string of messages about pay-for-performance in the federal government.  You have raised many legitimate concerns, and I'm sure that many in government share your concerns.

Nonetheless, I think there is at least one federal entity that has made good use of pay-for-performance - the US Postal Service.  They have had over 12 years experience with the concept for their white collar employees, and have had significant and sustained organizational results over this period.  Also, the employee associations representing supervisors and postmasters have been active participants in the design and implementation of the program. 

Since pay-for-performance is a concept that has received continued support by both political parties, I think we can expect to see it continue to expand and evolve in the federal government into the future.  The key is to find out what works within each federal entity, and build on that success.

Again, thanks for sharing your concerns.