The Cato Institute has generated big headlines with its new study claiming that federal employees are paid a whopping 78 percent more than private-sector workers. Not only does this drive up federal costs, Cato claims. It’s responsible for “drawing talented people away from higher-valued activities in the private sector.”

Cato’s study is precisely right—except for where it’s dangerously wrong. More important, it misses the fact that these changes flow from precisely the recommendations that Cato has been making for years. Except that the recommendations have led to big problems that are driving up government’s costs.

It’s precisely right because the average wage in the federal government is substantially higher than in the private sector. It’s dangerously wrong because the study completely misses the fact that the average federal job is wildly different than the average private sector job.

As I explore in my forthcoming book, Escaping Jurassic Government: How to Reclaim Government’s Lost Commitment to Competence (Brookings, 2016), the big movement to contract out federal jobs, led by groups like Cato, has had a huge impact on the federal workforce. From 1960 to the present, the average level of federal employees in the General Schedule has risen from 6.73 in 1960 to 10.33 in 2014. That alone accounts for half the difference in the public-private comparison that Cato found.

And why has the GS level risen? The nature of federal jobs has fundamentally changed. As the federal government contracted out more work, the number of clerical jobs in the federal government shrank 78 percent from 1973 to 2014, and the number of blue-collar jobs fell by 61 percent. It’s almost impossible to find a federal custodian, cafeteria worker, security guard, or mechanic because almost all of those jobs have been contracted out. In fact, at the now-shuttered Department of Energy’s Rocky Flats plant, which once manufactured the plutonium triggers for nuclear weapons, security was higher even than the White House—and the most dangerous stuff on the planet was carefully guarded. By Wackenhut.

Meanwhile, the number of administrative employees has grown 128 percent and professional workers are up 78 percent—all within a workforce that is just 2 percent larger. Fewer feds are providing front-line services. Fewer feds are providing support services. More feds are managing contracts and are performing more high-level technical work.

So Cato has the numbers right—average federal wages are higher than in the private sector—but the explanation wrong. Feds are making higher wages because, compared with the private-sector colleagues, there are fewer of them working blue-collar jobs and more of them working in administrative and professional positions.

The big boost in contracting out has led to a vast array of government programs over which the government has only loose control. For example, at the Centers for Medicare and Medicaid Services, which manages some of the federal government’s biggest and most important programs, each employee is responsible, on average, for $144 million in spending. Stop and let that sink in: $144 million, on average, for each employee. This is for a collection of programs that already represents one of the most privatized systems in government. And it’s responsible for big problems, including $75.5 billion in improper payments every year, according to the Government Accountability Office. In fact, one of every eight dollars in the flagship Medicare fee-for-service program is spent improperly.

So, the increase in federal wages comes from following Cato’s own recommendations over the years to privatize the federal government. The increased privatization, in turn, is a source of deeply rooted fraud, waste, abuse, and improper payments that the privatized government isn’t equipped to handle.

If we really thought about the issue, we’d realize that the increase in government salaries is a product of the changing strategies and tactics of the government’s operations. And we’d realize that if we really wanted to rein in government’s costs, we’d hire more feds to improve the management of programs in trouble, as John J. DiIulio argued so powerfully in Bring Back the Bureaucrats and in the pages of the Washington Monthly.

So Cato’s study is a tremendously important contribution to the public debate. Just not in the way Cato intended.

Donald F. Kettl

Donald F. Kettl is a professor in the University of Maryland School of Public Policy, where he served as dean from 2009 to 2014. From 1990 through 2004, he was on the faculty of the University of Wisconsin, Madison, and directed its Robert M. La Follette School of Public Affairs from 1996 to 1999. He also chaired two blue-ribbon commissions for Wisconsin Governor Tommy Thompson, on campaign finance reform (1996-1997 and state-local government finance (1999-2000).