When the Defense Department started fashioning a new pay-for-performance system for its civilian employees five years ago, “fairness” was the watchword.
But the first large-scale payout of performance-based pay raises and bonuses in January was riddled with inequalities, a Federal Times analysis has found. Specifically:
* White employees received higher average performance ratings, salary increases and bonuses in January than employees of other races and ethnicities. When compared to Asians and Native Hawaiians and Pacific Islanders, whites received on average almost a full percentage point more in their total payout of raises and bonuses.
* Employees working at Defense agencies — such as the Defense Finance and Accounting Service, Tricare, the Office of the Inspector General and the Office of the Secretary of Defense — earned higher performance ratings and payouts overall than did their civilian counterparts in the three military service branches. When compared with their counterparts in the Air Force, Navy and Marine Corps, employees at Defense Department agencies earned more than a full percentage point more on average in their total payout.
* Raises and bonuses appeared inconsistent at times with their corresponding performance ratings. For instance, employees who were 40 years old or older and those under 40 received closely similar performance ratings overall, yet those who were 40 or older received larger bonuses and total payouts. That could be, perhaps, because older employees more often hold higher-ranking jobs that give them more responsibility and earn them more attention from their superiors. In another example, Asian-American employees received slightly higher performance ratings overall than did blacks, yet they received significantly smaller pay raises and bonuses. And women received larger total payouts overall than did their male counterparts even though they both received the same performance ratings on average. Most experts interviewed by Federal Times say it’s too early to judge whether NSPS is discriminatory or otherwise faulty. But many agree the apparent inequalities cause concern, and they say Defense needs to closely watch these trends in coming years. If the inequalities continue, experts say, NSPS must be revised to correct them.
“It’s not time to cry that it’s a crisis,” said Darryl Perkinson, president of the Federal Managers Association. “But there are some areas of concern raised in these numbers, and we need to examine this thing.”
Pentagon promises review
The performance-based pay system, called the National Security Personnel System (NSPS), was launched two years ago and covered only 11,000 employees at first. Since then, it has expanded quickly. The performance evaluations and raises and bonuses handed out in January were the first large-scale test of the new system, covering 102,000 employees at agencies, bases and commands across the Defense Department.
Since then, Defense has added thousands more at agencies such as the Air Force Materiel Command, Army Corps of Engineers, Defense Finance and Accounting Service and the Army Special Operations Command. NSPS now covers about 180,000 employees and the department hopes to eventually bring 650,000 employees under the program.
Federal Times obtained the individual performance ratings, salary increases, bonuses and total payouts for each of the 102,239 civilian employees compensated under NSPS in fiscal 2007 in response to a Freedom of Information Act request. Defense also provided the employees’ gender, age range, race, ethnicity, location and the Defense component in which they worked. To preserve employees’ privacy, Defense did not provide their names or identify the specific organizations where they worked.
In an interview, NSPS program executive officer Brad Bunn said the department does not yet know what caused the inequalities in ratings and compensation.
“We’ve asked our components to do a deeper dive into the information to look for these issues,” Bunn said. “Are there differences in ratings distribution that we need to pay attention to? Are funding levels different?”
Bunn said differences in how pay raises are budgeted may account for some of the differences in compensation. NSPS rules allow each organization to create its own “pay pool” from which to fund pay raises and bonuses. Organizations can add more or less money, depending on what their budgets allow. So, for example, if an organization has a relatively small pay pool and a relatively high percentage of Asians, then that might explain why Asians might appear to be getting inequitable compensation.
Under NSPS, organizations are supposed to issue performance-based bonuses comparable to bonuses awarded in previous years. And since those varied across the department before NSPS was created, Bunn said that may also have caused some of the pay differences.
Bunn also believes that the Navy and Marine Corps’ relatively low payouts were skewed by the midyear transfer of nearly 7,000 Navy civilian employees from the Space and Naval Warfare Systems Command into NSPS. (Employees of the Navy and Marine Corps were grouped together in the Pentagon’s statistics.) Those employees, who accounted for a large portion of the Navy/Marine Corps segment, were paid only partially under NSPS and partially under their previous personnel system. The result was that their NSPS payouts were low and brought the overall figures down for the Navy and Marine Corps, Bunn said.
But the racial differences in NSPS payouts and evaluations “makes you scratch your head,” Bunn said.
“There could be a lot of different variables at play here,” he said. “It may be where there are concentrations of these various categories of employees. Maybe they’re in organizations that prorated pay pool funding because they came in later in the year. I don’t have the answers as to why that’s the case.”
Bunn pledged to further investigate, and said his office will examine individual pay pools to look for discrimination. He said his office could revise NSPS’ pay pool funding structure or reform managers’ training programs if systemic problems are uncovered.
But he would not speculate on what the results might say about NSPS’ fairness.
“When I look at it at this level, it creates more questions than answers,” Bunn said. “I don’t think we’re at the point where something needs to be done about it. But [the apparent inequalities are] certainly being looked at as part of our evaluation.”
Union: We told you so
The American Federation of Government Employees — a vocal critic of NSPS — said the results reinforce its concerns about the system.
“We feared this would happen, and these stats seem to put some real weight behind those fears,” Brian DeWyngaert, AFGE chief of staff, said. “These systems can have a discriminatory impact. Whether it’s intended or unintended, it happens nevertheless.”
DeWyngaert was particularly troubled by the statistics showing white people received larger ratings and pay increases.
“That variation speaks loudly, that there probably is a problem in the system,” DeWyngaert said. “It’s the first alarm bell that goes off, when the system looks like it’s having a discriminatory impact on people of color and minorities.”
DeWyngaert said Defense should scrap NSPS and go back to the old General Schedule system of regular step and grade increases.
“You can trust GS because it’s colorblind, and blind to all those categories,” DeWyngaert said. “These [pay-for-performance] systems generally have a lower trust level for employees for this very reason.”
He also said that the differences among components could prompt employees to leave lower-paying organizations, like the Air Force, Navy and Marine Corps, for higher-paying agencies within the department.
“If it continues, after several years the spread could be noticeable as employees move to agencies that do better,” DeWyngaert said. “That can’t be good for Defense’s continuity of mission, if it finds itself competing with itself.”
Carson Eoyang, executive director of the Asian American Government Executives Network, said that while the discrepancies are worrying, they aren’t wide enough to “scream of gross injustice.”
“But the data does not represent a basis to become complacent,” Eoyang said. “Continued vigilance and attention to improving NSPS is highly warranted.”
DeWyngaert and Eoyang say Defense should strive to reform NSPS so that results for different demographic groups are largely the same.
That may not be possible, said Paul Dorf, a pay expert and managing director of the human resources consulting firm Compensation Resources Inc. There will always be differences between demographic groups that may not reflect racial or ethnic bias on the part of the reviewer, he said. Pay-for-performance systems tend to favor older employees nearing retirement and highly trained employees in higher-ranking jobs, Dorf said. That could result in higher ratings and payouts for white people, he said.
“Some of those are behavioral issues; I don’t know if it points to flaws” in the system, Dorf said. “The reality is that you’re never going to have a homogenous thing where everyone is evaluated the same. If it were to work out that way, I would be more suspect that they’re gaming it.”
One group or another is bound to come out on top each year under a performance-based pay system, Dorf said. Defense should monitor its system annually to make sure it doesn’t consistently favor one group, he said.
Bunn said his office will do just that.
“This is one data point, one year’s worth of data, so we don’t have trend information yet,” Bunn said. “But we’re not waiting several years before we start looking.”