Federal employees are still smarting over President Obama's two-year pay freeze, but for some Republicans a mere freeze is not enough. House Speaker John Boehner, R-Ohio, promises to eliminate tens of billions of dollars from the budget, and federal workers will not be immune.

Meanwhile, the Office of Personnel Management argues that feds actually deserve a raise, not a pay reduction. OPM's 2010 annual report says federal employees earn less than their private sector counterparts, noting the pay gap grew from 22 percent in 2009 to 24 percent in 2010. This is surprising, given that federal employees received a 2 percent pay increase from 2009 to 2010 while economywide wages fell by 1.5 percent during that period. Nevertheless, public sector unions cite this data to dismiss claims of federal overcompensation.

Is the overpaid federal worker really just a myth? Not according to academic research. Economists have studied federal pay since the 1970s, and their methods and conclusions differ markedly from those of the government. Economists use statistical techniques that account for differences in workers' age, education, experience, gender, race, marital status and other characteristics.

Those studies generally have found a federal pay premium in the range of 10 percent to 20 percent, according to the 1999 Handbook of Labor Economics. A private sector worker earning $50,000 per year, for example, might receive $55,000 to $60,000 per year as a federal employee. The largest premiums are for lower-skilled employees, with smaller benefits as education increases. Interestingly, foreign studies also have found pay premiums for their government employees, suggesting government's weaker budget constraints allow public sector pay to rise above market levels.

Using the Census Bureau's 2009 Current Population Survey, the authors calculated an average federal pay premium of 12 percent over comparable private workers. Other studies tackle the issue from different angles, such as following the same workers over several years. Economists have demonstrated that private workers who switch to federal employment enjoy a substantial boost in wages.

In addition, feds quit their jobs at much lower rates than private sector workers, implying that civil service positions offer better compensation, job security and benefits. These retention rates persist even with the federal retirement program's shift away from a defined benefit pension structure, which was believed to account for low quit rates.

Why is this research so inconsistent with claims that federal workers are underpaid? Because economists compare similar workers, while OPM looks at similar jobs. This seemingly minor distinction between personnel and positions actually is important.

To estimate pay gaps, OPM surveys nonfederal positions, assigning each job a grade level based on its description and level of responsibility. A partner in a law firm might be classified as a GS-13, for example, while a junior clerical worker might be a GS-8. Compared to private jobs at the same assigned grade level, federal jobs seem to pay less.

One problem with this method is subjectivity: How can we be sure a particular private sector job is equivalent to a GS-9 rather than, say, a GS-8? And even if two jobs' responsibilities seem similar, how do we account for differences in job security, benefits, flexibility and myriad other factors that affect salary demands?

And there is a larger problem. According to the Congressional Budget Office, federal workers tend to be less educated and experienced than private workers at the same occupational level because the government hires people at higher grades and promotes them faster. A senior accountant at a federal agency, for example, might qualify only as a junior accountant in the private sector. This is why federal jobs seem to pay less, even while federal workers are paid more.

The federal pay system requires fundamental reform, starting with objective analysis from independent economists. Excessive salaries might be only a small part of the government's budgetary shortfall, but their existence implies government is not serious about fiscal belt-tightening.

Andrew G. Biggs is a resident scholar at the American Enterprise Institute, and Jason Richwine is a senior policy analyst at the Heritage Foundation.