Barack Obama is now President of the United States of America. By all accounts, he is a bright and thoughtful man who wants to examine all perspectives on an issue before charting a course. Many Federal employees (and Federal employee unions) are looking to the new president and his appointed leadership in hopes that recent pay-for-performance (PFP) initiatives will be discontinued.
This is the second in a series of three articles. In the first article, I listed several items the incoming Obama administration should consider when deciding whether to expand or contract pay-for-performance initiatives that proliferated in Federal agencies under the outgoing Bush regime. This article and the one that follows will expand on each of those factors.
The issues to be considered in this article are:
- Distinguishing between the pay-banding aspects of PFP and the appraisal processes upon which they rely. Pay-bands and salary increases are the easier part of PFP. Much tougher is a realistic and unbiased evaluation system. Despite decades of experience with individualized evaluations, the competence of Federal supervisors and managers in the area of performance evaluations is open to debate.
- Worrying about inflated ratings often results in a leadership that is more concerned with rating distributions or bell-shaped curves than with individual and collective achievements. This can prove incompatible with (supposedly) objective appraisal criteria.
- The costs of administering the PFP programs that are now established in Federal agencies are not clear… at least to this author. Many officials believe that such costs exceed any benefits accruing from merit-based pay.
- In calculating merit compensation, agencies may substitute one-time bonuses (formerly known to Feds as "awards") for ongoing salary increases (the equivalent of "steps" under the General Schedule). This can have a negative effect on an agency’s most talented and motivated employees.
The two P’s in PFP
Merit (or variable) pay systems present Federal leadership with two fundamental components:
- Pay banding – establishing and using broad spectra of possible salaries for broad categories of jobs. Thus an employee can see the lowest and highest potential compensation for the job category in which s/he is placed.
- Performance appraisal –comparing an individual employee’s achievement over the course of a year to pre-established standards or objectives. Much like school, standards are established at the start of a given year and employees rated at the conclusion.
Pay-for-performance does not apply to the president, senators, congress members, or cabinet secretaries. They do not get compensated for their successes, although the question of their continued service (re-election) may hang in the balance. Moreover, political appointees to the Executive Branch tend to inflate their own achievements and underplay their shortcomings. Consider Donald Rumsfeld – former Secretary of Defense, the godfather of NSPS. Did he achieve to written ("SMART") objectives? Would his evaluation of his own tenure and those of his immediate subordinates comport with the strictures of the PFP system he brought into existence?
Within the Federal rank-and-file, performance evaluations have had scant success. Inflated, subjective ratings led many agencies to abandon five-tiered appraisal systems in the 1990s in favor of "pass/fail" alternatives. These two-tiered constructions rated virtually no one, as more than 99% of an agency’s workforce received any meaningful news concerning their job performance. As those same agencies are returning to five-tiered schemes, supervisors and managers know nothing more about appraisals than in decades past.
Since I entered the civil service in the mid-1970s, supervisors and managers have seldom been accountable for maintaining good performance documentation. When they have kept good data (qualitative and/or quantitative), pressures from higher levels to limit the number of high outcomes have sometimes trumped efforts to rate people based on merit alone. If PFP is to succeed in years ahead, commitment from this new administration to a realistic performance management system needs to be broader and deeper than we have known in the past.
What’s the worst that can happen?
As in school, a primary reason for having appraisal systems is to weed out those who are failing. Certainly, a program titled National Security Personnel System (NSPS) should concern itself with removing defense employees who cannot carry their weight in the post-9/11 era. But that isn’t necessarily so. As was true with earlier incarnations, most of today’s Federal PFP systems are more concerned with rewards than consequences.
Unlike appraisals under the General Schedule and Federal Wage System, PFP plans often fail to prescribe definite personnel actions for civil servants evaluated as failing. Most FedSmith readers are familiar with systems that mandate major personnel actions (demotion and removal) for those who fail in just one "critical element". Nothing I’ve found in the NSPS regulations (please correct me if I’m wrong), for example, speaks to firing employees whose performance is unacceptable. Instead, these constructs presume that stagnated salaries and the possibility of sending employees to a lower pay band will bring about the desired result.
According to the Federal Times, 0.2% of DoD employees were rated at the lowest of 5 levels. This statistic (one of every 500 employees) seems alarmingly low to those of us who have worked in DoD agencies. [see When Acceptable People Perform Unacceptably and Dealing With the Unacceptable Employee] What may be more disturbing than the statistic is that the department failed to tell us what happened to those people. How many of that .2% are still working in DoD? Are supervisors willing to make repeated unacceptable assessments of those who remain both employed and incompetent? Isn’t that asking a lot?
Performance appraisal works best when both the carrots and sticks look credible to the workforce. While being denied an annual pay increase may be humiliating, if not financially threatening, it may prove bearable when compared to resignation and/or unemployment. If President Obama favors PFP for Feds, he would do well to consider how his branch’s worst employees will be effectively weeded out of the civil service. He will be lauded by the many and reviled by the few for doing so.
Do "Valued Performers" feel valued?
In my experience, status and exclusivity commonly stem from a willingness to rule out the majority. For those who believe in PFP, it only works if just a few employees are rated (and, therefore, compensated) at higher levels – while the many continue on without such distinction. If stellar performers are to be rewarded (and therefore motivated) by substantial pay-outs, there must be a large population of mediocre employees to support them by accepting much more modest ratings and pay increases.
In my school experience, a "C" was the middle rating in a five-level system. Under NSPS, this middle level (3) is termed "Valued Performer". More than half of DoD employees rated under PFP a year ago were bestowed this designation. Similar experiences can be found in DHS, FAA, and other agencies employing PFP.
These modest assessments seem to be backfiring. Employees rated as "valued" are too-often feeling under-valued. Leaders who believe that "A’s" should be conferred upon a few stellar performers may find such a practice misplaced in 21st century Federal agencies. Anecdotal evidence (see comments posted to the previous article in this series) indicates that ensuring a large percentage of employees are rated at the median level is proving as destructive as it is motivational.
Obama administration officials should consider the efficacy of desired rating dispersions (bell-shaped curves) – especially those that exist in the face of rhetoric to the contrary. If large numbers of employees achieve to the standards/objectives management held out to them, is it right to deny them higher ratings in order to maintain an appropriate dispersion of ratings?
We’re winning… if you don’t count the losses
PFP may actually cost much more than it’s worth. In confidence, HR specialists and management officials describe endless hours of paperwork and meetings that are required by these systems. When asked if that time and effort has been well spent, few answer affirmatively. Given the economics of the United States government in this new century, no program should be cost-ineffective without a good reason.
Serious studies of Federal PFP programs (at GAO, DHS, DoD, FAA, etc.) have not seriously explored the price tags associated with running them. These experiments have added layers of complexity to both pay-setting and performance evaluations.
When colleges determine which students make "dean’s list" and which graduate "magna cum laude", we hope such efforts motivate students to increased learning and scholarship. We also presume that the college administration used objective data to arrive at that conclusion. In many PFP environments, however, objective data does not definitively determine who is worthy of which rating. As a result, self-appraisals, written justifications, ongoing salesmanship, and debating endurance are often required to determine who should receive what grade. All that time (running over a 3-month period in many models) adds up.
Calculating the overhead costs associated with PFP will go a long way toward determining their projected value to the Executive Branch. Until the Obama administration has good estimates of these expenses, well-informed decisions regarding PFP cannot be made.
PFP depends on what you mean by "pay"
When Federal employees get raises via "within grade" or "within band" increases, those salary bumps continue into the future and also serve to increase retirement benefits. Under some Federal PFP plans, however, the pay-out to those with the highest performance ratings may be in the form of bonuses rather than (or in combination with) raises. This practice proves most common as our government’s best and most experienced employees reach toward the top of their pay bands.
While one-time bonus checks are often more substantial under Merit Pay, they are in no way equivalent to a similarly valued salary increase. Top performers are smart enough to recognize this. Many complain that their reward is of less value than those of others who received similar amounts in the form of salary increases.
While pay pool management no doubt has reasons for employing such strategies, it may not serve the interests of the incoming administration to have top achievers questioning the form in which they’ve been paid for their superior performance. After all, if the biggest winners are complaining, what value is accruing from these experiments?
Stay tuned for the next and final installment
Many will see the analysis above as a call to end PFP. As a self-employed trainer, however, I work in a similar milieu. Rather, I believe that such management schemes are complex and challenging – especially in government. If President Obama and his appointees don’t consider these difficulties, then expansion (or continuation) of PFP may prove a long-term embarrassment – as was Jimmy Carter’s Merit Pay initiative.
More discussion will come in the last article in this series. My hope in writing for FedSmith is to provoke thoughtful dialogue among Feds concerned with labor and employee relations. In the meantime, many thanks go to those FedSmith readers who posted insightful, sincere, and cogent comments to the first in this series.