Why Agencies Should Better Use Data To Make Pay, Leave Decisions

The federal government is not taking advantage of its payroll data to better inform decisions on pay, leave, telework and other workforce benefits, according to a new report.

Federal agencies submit personnel data to the Office of Personnel Management, which it then places into its Enterprise Human Resources Integration system. Information related to pay, leave, work activities and various flexibilities is not widely distributed, according to the Government Accountability Office, robbing agencies and academics alike from properly analyzing staffing issues and other decisions in support of their missions. The report found the database also maintains weaknesses in its data, including errors, incomplete fields and ineffective monitoring.

More than 250 field agencies report 300 data elements to EHRI, which OPM launched in 2009. The agency has effectively cleaned and distributed data related specifically to personnel, such as the number of employees at each component, separations and other bits of information. That information is available publicly on OPM’s FedScope, which can support statistical analyses. OPM also can share information with think tanks and academics and has developed a “suite of analytical tools” for agencies for analysis and forecasting.

OPM has no equivalent functionality for payroll data, GAO said. The information is “intended to provide a centralized, standardized, and comprehensive source of pay and leave related data across the federal government.” It is not fully taking advantage of the database, the auditors said, without making it more widely distributed.

“Because the EHRI payroll database has potential to be used for accountability, research, and data-driven human resource management and policy decision making, making it available would support OPM’s strategic and open data goals,” GAO wrote.

Researchers have issued numerous studies on federal employee payroll information, but have been forced to rely on incomplete information. GAO noted examples in which researchers depended on federal employees to self-report their salaries or only looked at one agency to draw broader conclusions.

“Studies relying on information about a small subset of the federal workforce may not provide reliable insights about overall federal human capital trends or policy effects,” GAO said.

Full and accurate data is especially important when comparing federal employees to their private sector counterparts, the auditors noted, a frequent point of contention among various stakeholders and academics. Individual agencies have demonstrated the benefits of analyzing payroll data, such as the Veterans Affairs Department identifying wrongful incentive benefits and other agencies rooting out improper payments. GAO said it could better audit official time and administrative leave with an improved EHRI system, while agencies could identify pay disparities among various demographics, analyze retention statistics with telework data and better monitor leave use.

GAO also found OPM managed “limited quality checks” on the data in EHRI, though the HR agency said most of those issues stem from the agencies at which they originate and would require stakeholder buy-in to improve. The auditors still blamed OPM for its lack of data monitoring, saying the agency’s insufficient oversight could compound data quality issues. 

“Unless OPM takes steps to correct these internal control weaknesses,” GAO said, “it will be unable to fully leverage these data to meet its mission and allow others to make full use of these data for their research needs.”

OPM said it had plans to “update EHRI security protocols, payroll documentation, testing for reliability issues and data standards,” but GAO said the agency “has not documented these plans or created a schedule to implement them.”