More Evidence That Federal Employees Are Overpaid

Great post by Andrew Biggs of AEI.

"In my AEI working paper on federal employee compensation with Jason Richwine of the Heritage Foundation, we compare the salaries, benefits, and job security of federal employees to that received by private-sector workers with similar earnings-related attributes—that is, similar education, experience, region, race, gender, and so on. These calculations, which are performed using regression analysis, show that federal workers receive salaries around 14 percent higher than similar private-sector workers. The federal pay premium is largest for employees with less education, and increases as workers gain experience. (So, for instance, a less-educated federal employee with long job tenure would get a larger pay premium than a newly hired PhD).

Almost no economist really disagrees with this approach, so much so that studies on federal salaries—after a spurt during the 1970s and 1980s—are pretty infrequent today. In other words, most labor economists seem to consider the pay premium issue more or less settled.

But some people find this kind of statistical analysis unconvincing, probably because they don’t think it’s really possible to control for all the relevant differences between different kinds of workers. While regression analysis can control for whether a person has, say, a bachelor’s or master’s degree, it doesn’t control for the quality of the school attended or the grades the person received. Likewise, maybe federal employees are unusually hard-working or creative, such that they create more value than private employees who look the same on paper. I doubt it, but you can’t prove that it’s impossible.

What these folks want to know is how much the exact same person would be paid in the federal government versus the private sector. And Jason Richwine’s new paper released by Heritage answers that question. Instead of comparing pay for different people at the same point in time, it follows the same people over time as they shift into and out of different jobs. If a given person earns more in a federal job than a private-sector job then we can be pretty sure it’s the job that’s making the difference, since the person himself barely changes over time. (And Jason controls for the limited instances where the person’s characteristics do change, say by getting an additional educational degree or by gaining an extra year of experience.)

What does this analysis show? As Jason states, “Private-sector workers who switch to federal jobs receive an average real wage increase of 9 percent, while private workers who find another private job earn just an additional 1 percent, implying an 8 percent federal premium.” These are the same workers, with the employer and job being the only difference. Similarly, most people who leave federal employment take a pay cut, undercutting the common claim by federal employees that they could earn much more on the outside.

But does this study, which finds an 8 percent average pay premium for people switching to federal jobs, undercut our previous estimate of a 14 percent pay premium? Not at all. Remember that our analysis found that the federal pay premium is smallest in the initial years after a worker has been hired, and Jason’s new paper calculates the pay premium only in the first year of employment. So it’s actually fully supportive of the cross-sectional results we generated.

If there’s a convincing rebuttal to all this from the Office of Personnel Management and the public employee unions, who in the past have pooh-poohed federal-private pay comparisons, I’d like to hear it."