GAO: Pay study accuracy is in the eye of the beholder


There’s been a slew of reports issued over the last three years comparing federal employees’ pay and benefits to private sector workers, and they’ve all come to some radically differing conclusions. Which one is right?

Everyone, and no one, according to a new report from the Government Accountability Office posted online Monday.

GAO spent the last year digging into six reports on federal pay, and concluded that they “varied because they used different approaches, methods and data.” For example:

  • The Congressional Budget Office, American Enterprise Institute, and Heritage Foundation all used a so-called “human capital approach” to compare federal and private data. This means they took into account personal attributes (such as education and job experience), demographic attributes (like race and gender), and other factors such as occupation, locality, and firm size. All three of these studies found feds were paid anywhere from 2 percent to 22 percent higher than private sector employees.
  • The President’s Pay Agent and the Project on Government Oversight used a “job-to-job” approach that, like the human capital approach, takes occupation and job level into account. They did not factor in personal attributes. The Pay Agent concluded feds are paid 24 percent lower than nonfederal workers (including state and local employees) and POGO found feds are paid 20 percent higher.
  • The Cato Institute conducted a “trend analysis” that did not control for any attributes and looked only at broader pay trends over time. Cato found feds’ pay is 58 percent higher than private sector employees pay.

Unsurprisingly, the authors of each study told GAO that their method was the most accurate way to compare pay. For example, the researchers who used the human capital approach said their method can best measure whether the government is overpaying or underpaying its employees, compared to what they would earn if they quit their federal jobs and looked for work in the private sector. But even the human capital researchers conceded that their method shouldn’t be used to set an individual’s rate of pay, since demographic attributes such as race and gender aren’t work related. (Just imagine the lawsuits if the government started paying certain employees less because other members of their race also earned less.)

GAO did not say that any of the studies were incorrect or faulty. But it did conclude that because they are so different, “comparing their results to help inform pay decisions is potentially problematic.” GAO also said that “given the different approaches of the selected studies, their findings should not be taken in isolation as the answer to how federal pay and total compensation compares to other sectors.”

Long story short, GAO’s conclusions don’t leave us any closer to settling the nagging pay gap question.

GAO also reiterated the Pay Agent’s “serious concerns” about the current method for setting pay under the General Schedule, which requires a single, across-the-board adjustment for all employees in each locality. This means that everybody — accountants, clerks, engineers, etc. — gets the same raise, even if the market for clerks grew softer that year while engineers became harder to find. The Pay Agent has called for the government to re-examine the method it uses to determine the federal-private pay gap.

And it notes that while President Obama called on Congress last September to create a commission to reform federal compensation and other personnel matters, “such a commission has not yet been established.”


About Author


  1. Brenda S Brown on

    You cannot group everyone in the same catagory. Each job requires to be checked out to the private sector. You cannot take a group of employees and mass mingle their pay to equal what you want it to.
    The Feds pay right now per job per employee is way below the private sector pay. This I know just for my pay grade and job. I was making over $50,000.00 in the private sector and now make less than $50,000.00 in the Federal pay. So each individual job needs to be scrutinized with out grouping oter pay grades in a mass bunch.

  2. I’d love to know who all these studies that say feds are over paid are comparing too. My field is also well under private sector pay for simiar jobs. So why do I stay? Because I’m in that trapped zone — too young to retire and too old to start over.

  3. I retired from Fed Gov a couple of years ago. Feds are way overpaid as a whole but less for the competent. My observations were that approx 20 to 30% were carrying an excess load counteracting the 20 to 30% that were providing negative support for the mission with the remaining 40 to 60% providing the minimum output required for the job description.

  4. An accurate comparison will never be achieved given the diversity of jobs in the federal work force. As best as I have observed, some jobs, notably the lower grades involving labor not requiring a college degree pay more based on the fact that these jobs have a defined pension and 401K while comparable work in the private sector does not.

    Professional jobs like engineers, scientists, lawyers and doctors in the federal sector have total compensation decidely lower than the private sector.

    Federal benefits like health care, a defined pension and 401K as well as sick and annual leave are generally uniform across job series and grades. This is where the disparities arise in comparing federal versus non federal employment compensation. Large companies are not hamstrung with this. The same company can offer health care and a 401K to a scientist, but not to the janitor or secretary.

    Does congress really want to give that kind of leeway to federal agencies? What are the consequences therein?

Leave A Reply