Are Pentagon Civilians Really Behind the Pentagon’s Money Woes?
Justifying the last point, the analysts wrote to the Pentagon‘s civilian leaders and lawmakers on Capitol Hill that:
From 2001 to 2012, the active duty military grew by just 3.4 percent. Yet over the same timeframe the number of civilian defense employees grew by 17 percent, an increase five times greater than the armed forces.
Sounds bad, right? Sounds like the Defense Department’s civilian workforce is out of control? Wrong.
The problem with the analysts’ letter is the baseline they use. They picked the period when the Pentagon’s civilian workforce was the smallest it has been since at least 1981. The number of Pentagon civilians hovered around 1.1 million throughout the 1980s.
During the 1990s, the Defense Department’s civilian workforce shrunk from 1 million in 1990 to 649,000 in 2002, according to White House data. Those 350,000 civilians represented the bulk of federal civilian jobs lost during the 1990s, thanks largely to the Federal Workforce Restructuring Act of 1994, but also to reduced Pentagon expenditures as the Cold War drew to a close.
Many of the civilians shed in the 1990s were acquisition professionals—not just contracting officers, but engineers and scientists—that oversaw contracts and would call out contractors when they were pulling a fast one on the government.
Much of the hiring in the 2000s—a relatively modest increase of 128,000 civilian employees—was to rebuild the workforce gaps created during the 1990s that became an issue as the U.S. fought two wars in Afghanistan, Iraq, and elsewhere as U.S. military spending nearly doubled.
In fact, what really grew during this time and — especially into the 2000s — was service contract spending. Instead of hiring as many federal civilians as the Pentagon would have liked, arbitrary workforce caps and a White House that ideologically preferred contractors—whether they cost less or actually performed better—led to extraordinary growth in Pentagon contract spending. It was during this time that more money in contract spending began to be spent on services rather than on goods like tanks, planes, bullets, food and fuel.
It got so out of hand that contractors were getting hired to oversee contractors. The Pentagon, to put it bluntly, was losing control of its affairs. Plus it was costing a ton of money, and still is.
Don’t take it from me. Here’s a Pentagon chart comparing spending on civilians, uniformed military personnel, and service contractors.
The graphic says service contractors are “Increasingly Unaffordable” and that “the savings are here.” (There are some differences between the White House numbers for DoD civilian staffing and the chart’s; part of that may be explained by the fact that the White House used “Full Time Equivalents” rather the actual number of positions; some employees are likely part time.)
In fiscal year 2010, about 17,000 Pentagon positions were shifted from contractors to federal civilians. At least 50% of those positions were moved to save money, according to a Pentagon report. Others were changed because they were inherently governmental or closely associated with inherently governmental functions (in other words, they should not have been contracted out in the first place). This short-lived “insourcing” initiative boosted the number of federal civilians, but was done so, in part, to save money.
Many of these defense analysts should know better. Cutting federal civilians could lead to those functions simply being handed over to more expensive contractors. To really save money while maintaining an effective military, the total force mix has to be analyzed as a single entity.
Oddly, the 25 analysts conclude that cutting civilians is a solution while acknowledging that “it is unclear if that growth [in Pentagon civilians] was appropriately matched to the changing needs of a downsizing military and shifting strategy.” Without assessing the service-contractor part of the workforce—which is not mentioned in their letter—it’s hard to conclude that Defense Department civilians are really the problem.