Government jobs are not the same as nongovernment jobs. We should not expect government jobs to have the same salaries as nongovernment jobs because we do not expect different jobs to have the same salaries. So if we compare the salaries of government jobs to the salaries of nongovernment jobs, we need to adjust for the differences in the jobs.
Suppose there is a government worker who has a master of arts degree in government studies, conducts studies on how the government can solve social problems, and is paid $70,000 dollars per year. Suppose there is a nongovernment worker who dropped out of school, is an experienced computer programmer, and is paid $70,000 dollars a year. After adjusting for the differences in their jobs, which is paid more?
Since the government worker has more education than the nongovernment worker, the government worker is worth twice as much as the nongovernment worker. The government worker should be paid twice as much as the nongovernment worker. The government worker is underpaid.
On the other hand, the government worker writes reports that no one ever reads, and makes recommendations which no one ever follows. The nongovernment worker writes programs which make computerized devices work better. The government worker has no useful or productive skills and accomplishes nothing, while the nongovernment worker possesses valueable skills and accomplishes a lot. The nongovernment worker should be paid ten times as much as the government worker. The government worker is overpaid.
Notice that the fudge factor which adjusts for the differences between government jobs and nongovernment jobs is so large that the data is irrelevant.
Most people prefer a higher paying job to a lower paying job. Therefore, most government workers probably chose to become government workers because the available government jobs paid better than the available nongovernment jobs. Likewise, most nongovernment workers probably chose to become nongovernment workers because the available nongovernment jobs paid better than the available government jobs. Thus most government workers are probably earning more than they could earn in a nongoverment job, and most nongovernment workers are probably earning more than they could earn in a government job.
If we assume the free market is perfectly efficient, then the correct salary for every worker is whatever that worker can earn in the free market, in a nongovernment job. Since government workers are paid more than they could earn in nongovernment jobs, government workers must be overpaid. It is impossible for government workers to be underpaid. Any data which shows otherwise must be incorrect.
Thus it is logical for anyone who truly believes the free market is totally perfectly efficient to reject any data which shows that government workers are underpaid. Logical, but maybe not correct.
People who oppose the free market say the free market is unfair, because the greedy corporate CEOs are overpaid, while they underpay all their workers. If we assume that the government is more fair than the free market, then the salaries of government workers must be correct, while the salaries of nongovernment workers are too low. If that is true, then why don't more nongovernment workers quit their slave labor nongovernment jobs and get better government jobs?
But people who oppose the free market also say that government workers are underpaid, that government workers are paid less than equivalent nongovernment workers. So they simultaneously believe that corporations underpay their workers more than the government, and that the government underpays its workers by more than corporations. They say that the government pay workers more fairly than corporations, and at the same time they want government salaries to be based on corporate salaries.