As you know, Donald Trump won remarkably few policy victories in the first six months of his presidency. The courts have blocked his Muslim ban. Obamacare repeal and replace is on life support in the Senate. Tax reform seems a distant prospect. Funding for a border wall remains hypothetical.
One item on his agenda, however, is moving right along: cutting the size of the federal workforce. You don’t hear as much about this one, in part because Trump himself doesn’t talk much about it. But it’s clearly a priority—one the administration has billed as a way for Trump to make good on his promise to “drain the swamp.” His “Contract with the American Voter,” released a couple of weeks before last November’s election, began with “Six measures to clean up the corruption and special interest collusion in Washington, DC.” Item number two was a hiring freeze “to reduce the federal workforce through attrition.”
Trump instituted the hiring freeze in one of his first acts as president. In March, his administration put out a budget outline that called for a $54 billion increase in defense-related spending offset by major reductions at other agencies—a budget that, if enacted, could result in a net cut of as much as 9 percent of the federal workforce, according to estimates by the chief economist of Moody’s Analytics, Mark Zandi. In April, Trump lifted the hiring freeze, but his budget director, Mick Mulvaney, circulated a memo—titled “Comprehensive Plan for Reforming the Federal Government and Reducing the Federal Civilian Workforce”—asking every government agency to develop a plan by September to cut staff in line with Trump’s budget objectives.
Of course, Congress, not the White House, decides the budget, and some leading Republicans have signaled their discomfort with how far Trump’s proposed cuts would go. Still, GOP lawmakers are likely to sign on to cuts that are plenty deep. Indeed, decimating the civil service is one of the rare policy areas where Steve Bannon and Paul Ryan see pretty much eye to eye. Last year, Republican majorities in the House and Senate agreed on a nonbinding budget blueprint that would cut the nondefense workforce by 10 percent and direct agencies to hire just one employee for every three vacancies. (Barack Obama promised to veto any such spending bills.) The Senate can mandate workforce cuts through the filibuster-proof reconciliation process.
Even without Congress, Trump has a lot of power to force administrative agencies, which are under his control, to shed staff. Congress decides how much money agencies can spend on personnel, but the Office of Management and Budget (OMB) generally gets to tell them how to spend it. In response to the April memo, Veterans Affairs Secretary David Shulkin announced that his agency would leave more than 4,000 vacant jobs unfilled, and Secretary of State Rex Tillerson is reportedly planning to cut 2,300 jobs, or 9 percent of his whole staff, through attrition and buyouts.
The Trump/GOP effort to shrink the civil service plays on a narrative the American people have been hearing for decades: the federal workforce is bloated. As Mulvaney’s guidance put it, there are “too many Federal employees stuck in a system that is not working for the American people.” Press secretary Sean Spicer, announcing Trump’s executive order in January, explained that the hiring freeze “counters the dramatic expansion of the federal workforce in recent years.”
The only problem with this narrative is that it is the exact opposite of the truth. As a share of the U.S. workforce, the federal civil service is actually smaller than at any time since before World War II. In absolute terms, it has been about the same size for half a century. In 1966, there were about 2.1 million executive branch civil servants (not including Postal Service employees). Since then, the country’s population has increased from 196 million to 323 million. The annual gross domestic product, along with annual government spending, more than quadrupled. And the workforce? In 2016, there were still only 2.1 million federal employees.
There’s no rule that says the number of civil servants has to rise in lockstep with the population or the economy. Many federal jobs in the 1960s were clerical positions that computers have made obsolete. But still. In 1966, there was no Environmental Protection Agency, no Department of Homeland Security, no Federal Emergency Management Agency. Medicare and Medicaid had been signed into law just a year earlier. It’s hard to believe that the same number of people we had in 1966 can run such a radically larger government enterprise.
And, in fact, they don’t.
While the number of federal employees has basically flatlined for a half century, the government has ballooned if you include another group in your tally: private contractors. As the size and scope of federal programs grew, but the number of civil servants stayed fixed, that labor had to get done by someone. Congress’s answer has increasingly been to contract with the private sector. So when Trump and the Republicans say they’re going to shrink government by cutting federal workers, do a mental autocorrect. What they’re really saying is, we’re going to be shoveling a lot more money out the door to federal contractors.
This means that the federal government is about to get considerably costlier and more dysfunctional. Despite the claims made about the efficiency of the private sector, hiring for-profit contractors costs taxpayers a lot more money than paying civil servants does—to the tune of hundreds of billions per year. Meanwhile, an understaffed government is one that screws up more often as shorthanded agencies struggle to manage programs, monitor all the contractors, and prevent Hurricane Katrina–type disasters. And the addiction to contractors at the expense of adding federal employees means that these agencies are staffed overwhelmingly by people at or nearing retirement age, with no young talent coming up through the ranks behind them.
Ever since Ronald Reagan said “government is the problem,” Republicans have been happy to outsource—even though it wastes money—because it lets them look tough on “big government” without actually cutting the programs their constituents like. Democrats, meanwhile, have gone along with growing this shadow government as a way to make sure federal programs are administered without opening themselves up to attacks that they are growing the bureaucracy.
In other words, Trump didn’t create this problem. But if he follows through on his threats, he may push it to a breaking point.
No one knows exactly how many contractors the federal government uses at any one time—more on that later—but Paul Light, a public service professor at New York University and the leading expert on what he calls “the true size of government,” puts the number at around 3.7 million as of 2015, down from a much larger total at the height of the Iraq and Afghanistan wars. (That’s in addition to 1.6 million grant-funded nonprofit employees.)
This phenomenon has received some attention in the context of the wars. George W. Bush—followed by Barack Obama—needed more bodies to serve overseas than the military could provide. Rather than reinstate the draft (politically impossible) or entice more volunteers with higher pay (politically risky), the government turned to private contractors. That helped hide the true size of the war effort. News reports on the number of troops left in those countries don’t include the enormous contractor force. As of 2016, per Foreign Policy, there were about three contractors (28,626) for every uniformed soldier (9,800) in Afghanistan. When contractors are killed—and they are, more often than U.S. troops—they aren’t included in the official casualty numbers.
Very few Americans realize that this exact trend—using contractors to hide the true size and cost of a government mission—is not even close to being limited to the military. It’s ubiquitous. There has been a de facto political cap on federal workers since the 1970s, and every administration since at least Jimmy Carter’s has used contractors to get around it. Bill Clinton, who announced that “the era of big government is over,” harvested the end of the Cold War to cut 400,000 federal jobs, mostly in defense. But big government was only over if you ignored the contracting workforce, which kept rising by the hundreds of thousands. The Bush administration took things to another level, as contracting morphed from a way to sneakily expand government to an ideological goal in itself.
The Department of Defense (DOD) employs around 740,000 full-time civilians—and an estimated 700,000 contractors. In 2010, Department of Homeland Security (DHS) officials told Congress they had more contractors (about 200,000) than direct employees (188,000). At the United States Agency for International Development, the ratio of contractors to employees is nine to one.
Government contracts broadly fit into two categories: goods—like fighter jets and office furniture and copy machines—and services. A service contractor is, in essence, someone brought in to do the kind of work that could be done by a federal employee, from cybersecurity to mopping the floors at a VA hospital. Walk into any federal agency office, and you’ll see service contractors and civil servants sitting side by side, doing the same work, indistinguishable other than the employer listed on their badges.
NYU’s Paul Light estimates that there are between 600,000 and 800,000 service contractors; the government spends more on them than it does on the salaries of the entire civil service, which has three times the number of people. Contracting accounted for 40 percent of all discretionary spending in 2015, and service contracts accounted for 60 percent of that (even more at nondefense agencies). And it’s rising. Even the Defense Department spends twice as much on contracts for services as it spends on aircraft, ships, and land vehicles. According to a 2015 Congressional Budget Office (CBO) report, spending on contractors nearly doubled from 2000 to 2012, and the subset “that grew the most in dollar terms was contracts for professional, administrative, and management services”—that is, service contracts.
“If you really want to cut government, you’ve got to cut government programs,” said Don Kettl, a political scientist at the University of Maryland and a prominent expert on government. “If you focus on cutting government bureaucrats, you can weaken the programs, but you’ll increase your reliance on contractors and worsen the ongoing cycle that we’re trying to escape from.”
The equation is simple: if feds go down, contractors go up. Trump has actually proposed spending more money—but not necessarily hiring more employees, other than a small number of immigration enforcement officials—at the Departments of Defense and Homeland Security, two of the biggest, most outsourced government agencies. Contractors are licking their chops.
So what? Are contractors really so terrible?
No. Other than the occasional Erik Prince, the notorious founder of Blackwater, contractors are just people. Many do essential tasks that the government is ill-suited for—like cutting-edge research or product design—or blue-collar work that doesn’t need to be done by a full-time federal employee. The problem isn’t that contractors are bad. One glass of wine is good for you. But if contractors were wine, we’d be at two bottles a day.
One consequence of this binge is cost. One of the most stubborn fantasies about contracting out government work is that it will somehow save money. In fact, it generally costs the government more to hire a service contractor than to hire a federal employee to do the same work. So when politicians promise to cut costs by getting rid of civil servants, they’re only telling half the story. Most likely, those civil servants will be replaced by more expensive contractors.
When the government pays contractors, it’s not just paying their salaries; it’s paying their employers’ overhead and profit margins. A contractor who works for the financial services and consulting company Deloitte, say, and makes $50 an hour, might bill the government at $350 an hour, with Deloitte pocketing that $300 difference. So why do some people, particularly conservatives, argue that hiring contractors instead of employees saves money?
One reason is the belief that competition will force the private sector to cut costs. But contracting isn’t nearly as competitive as advertised. Last year, the DOD spent nearly half its $296 billion contracting budget on projects that had only one bidder. Another belief is that contracts are only temporary: Hire civil servants, and you may be stuck paying them for thirty years. Hire contractors, on the other hand, and you can get rid of them once the project is complete. The trouble is that this doesn’t happen much in practice either. How could it? Contractors are responsible for the ongoing, day-to-day functions of government agencies. As one high-ranking official in the Department of Energy—one of the most contractor-dependent agencies in government—said, “If we execute the program we’re on, it’s going to be as busy or busier than it is now for about ten or fifteen years. So the idea that we’re hiring support-service contractors for six months and then the work is going to subside—it’s not going to subside.”
A younger contractor who recently finished a stint in cybersecurity at an agency within the DHS put it more colorfully. “It’s like watching a drug addict, man,” he said of the feds whose work he supported. “They’re so dependent on other services that, left to their own devices, they would just crumble to the ground.” He offered another analogy: “You get a liver transplant, the liver’s from somebody else. For the rest of your life, you have to take drugs that prevent an adverse reaction from taking place. Contractors are those drugs.”
Okay, so contractors aren’t really so competitive or temporary. But doesn’t the government save money by not paying the excessive salaries and benefits of federal employees? A 2010 Heritage Foundation report with the subtitle “How Americans Are Overtaxed to Overpay the Civil Service” concluded that feds earned 30 percent to 40 percent more in total compensation (wages plus benefits) than people doing the exact same job in the private sector. On that basis, the report recommended increasing the government’s use of contractors.
But the Heritage report had a crucial flaw: it didn’t compare what civil servants make with what the government actually pays private contractors. A restaurant pays less for ingredients than a home cook, but that doesn’t mean it’s cheaper to dine out. Even if a federal bureaucrat makes more money than a private-sector counterpart, that says next to nothing about what it would cost the government to pay a private firm to supply that contract employee.
A 2011 study by the Project on Government Oversight tackled this question. The researchers looked at thirty-five categories of workers where it was possible to compare the average cost of a fed—salary plus benefits—with the cost of a similarly qualified contractor. Using data from the Department of Labor, they found that, benefits included, the average federal employee makes about 20 percent more than a comparable worker—not a contractor, just any old person—in the private sector. But here’s the crucial part: in thirty-three of the thirty-five categories, according to the report, hiring a contractor costs the government nearly twice as much as an employee. (One category was tied, and one was cheaper to contract out: groundskeepers.) In some cases, it’s much more: hiring a contractor to do “claims assistance and examining” costs nearly five times as much as hiring a federal employee. My personal favorite: it costs 2.29 times as much to hire a contractor to do the work of a . . . “contract specialist.” That’s right: the federal government hires private contractors to help hire private contractors, and according to the best available analysis, it pays them more than double what it pays its own employees to do it. (A surreal Government Accountability Office report on the Army’s Contracting Center of Excellence—let the name sink in—found that in 2007, “contractors—who work side by side and perform the same functions as their government counterparts—comprised 42 percent of CCE’s contract specialists,” cost the Army 27 percent more per person, and were inappropriately blurring the lines between civil servant and private-sector employee.)
Do you remember Snackwell cookies? Snackwell was a line of low-fat junk food marketed by Nabisco as a healthier option, back when we still thought the key to losing weight was to eat less fat. Of course, the cookies had even more sugar to compensate. Eating them only made people fatter, since, as we now know, replacing calories from fat with calories from carbohydrates will make you gain weight, not lose it.
Replacing civil servants with contractors to save money is like replacing fat with sugar to lose weight. Outsourcing is the Snackwell cookie of federal government.
Capping the federal workforce wastes money in other ways. Back in 1982, the Government Accountability Office (GAO) found that hiring freezes didn’t succeed in substantially reducing the head count, but they did end up costing the government more money. Freezes and furloughs sound like sensible belt-tightening tactics, but the business of government has to keep going. So a freeze creates backlogs of work that then end up costing more to try to catch up on.
A big part of the government’s job is keeping track of its money, and it needs staff to do that. The Internal Revenue Service is the most egregious example. Since 2010 its funding has been cut by more than 17 percent, and staff by 13,000, including nearly a quarter of enforcement staff. Meanwhile, according to the most recent data, the United States has a $450 billion annual tax gap. The Treasury Department has estimated that every $1 invested in more staff to go after those unpaid taxes would yield $6 in revenue.
So will members of Congress, constantly railing about the need to rein in the deficit, take advantage of this almost magical way to bring in more money without raising taxes? Of course not. Instead, pursuant to legislation passed in 2015, the IRS is about to hire private debt collectors to go after delinquent taxpayers. This has been tried twice before, in 1996 and 2006, and it was a disaster both times. Private debt collectors, it turns out, are a lot less efficient at recovering unpaid taxes than IRS agents. The 2006 experiment ended up costing millions more than the contractors managed to collect—and after the contract ended, IRS agents brought in 62 percent more from the same pool of outstanding payments than the private collectors had.
It’s really hard to overstate how stupid this is. More IRS employees would save the government money, but Congress refuses to hire them, supposedly because it’s worried about money. IRS Commissioner John Koskinen, who has been repeatedly attacked by conservative members of Congress, said this at a congressional hearing in April: “We will do everything we can to make sure this program is effective. Because if it works, that would be fine. If it doesn’t work, I don’t want anyone saying, well, we actually sandbagged it some way or the other.”
Translation: This will be a fiasco, but don’t blame me.
Donald Trump claimed a victory when, after he complained about the price of the F-35 fighter jet, the Defense Department announced that it had agreed on a lower price with the manufacturer, Lockheed Martin, saving the government $728 million. In fact, the price drop had been in the works for years. Meanwhile, the F-35 is a $400 billion boondoggle. When there aren’t enough experienced civil servants supervising contracts, the predictable results are cost overruns, missed deadlines, and even fraud. The acquisition workforce, which is essential for managing these contracts, was cut nearly in half from 1989 to 1999. If Trump really cared about defense contractor costs, he would ask Congress to create more of those jobs. In his book Bring Back the Bureaucrats, the political scientist John DiIulio points out that “between 2010 and 2013, when the DOD not only added about 3,500 personnel to its acquisition workforce but also trained them better than usual for the job, the agency’s on-time contract compliance assessments increased by nearly a third.”
The GAO regularly puts out a list of “high risk” agencies. The GAO’s comptroller general, Gene Dodaro, told the National Academy of Public Administration last year that the office has never removed an agency with more contractors than employees from the list. In a new book, Valuing Bureaucracy, Paul Verkuil writes that Dodaro told him “that the reason these agencies have never left the list is largely due to inadequate contractor management.”
The Centers for Medicare and Medicaid Services employs around 6,000 employees and at least 14,000 contractors. That in-house staff grew by more than a thousand during the Obama administration, but is still laughably tiny compared to its docket. The employees are ultimately responsible for overseeing around $1.1 trillion in federal spending annually—a quarter of the federal budget and about $183 million per employee. The CMS is a perennial inclusion on the GAO’s annual list of “high risk” agencies; with such low staffing, it’s hard to watch out for fraudulent or erroneous Medicare and Medicaid payments.
It isn’t just about money; understaffing and outsourcing lead predictably to government screw-ups. The HealthCare.gov fiasco, for example, occurred on the CMS’s watch. According to NYU’s Paul Light, it “was highly dependent on a poorly coordinated collection of 55 outside vendors [and] delegated to an understaffed, underfunded agency.” A report by the Health and Human Services inspector general attributed the botched website rollout to several factors, including “failing to properly manage its key website development contract.”
In 1996, the Office of Personnel Management turned its background-checking arm, U.S. Information Services, into a private contractor. When hackers stole personal information about millions of people from the OPM in 2014, they did it by hacking the USIS network. The company later paid the Justice Department $30 million to settle an unrelated lawsuit for collecting payment on thousands of background investigations it never performed. The contractor hired to replace USIS, KeyPoint, was promptly hacked as well.
It’s contractors all the way down. USIS, a contractor, handled the security clearance for Edward Snowden, a subcontractor for Booz Allen Hamilton, a contractor, working at the National Security Agency. Earlier this year, another Booz intelligence analyst was indicted for stealing classified material. In a March Washington Post op-ed, author Tim Shorrock wrote that “contractors have been responsible for at least five major security lapses in four years.” But nothing happens to the companies, in part because the cybersecurity industry is dominated by just five corporations.
Perhaps the strangest thing about the contractor workforce is that no one knows exactly how big it is. A 2015 Congressional Budget Office report on the contractor workforce lamented, “Regrettably, CBO is unaware of any comprehensive information about the size of the federal government’s contracted workforce.” Agencies simply don’t have to keep track of the number of contractors they employ. People who try to figure it out, like Paul Light, have to extrapolate from the amount of money agencies spend on contracts and the average salaries of the people who do the jobs being contracted for. As Light put it, “It’s not great data. It’s not even good data. It’s okay data.”
Talk about a swamp. Contracting is a huge, politically powerful industry. Federal law prohibits firms that contract with the government from making campaign contributions—but that law doesn’t apply to corporate political action committees or to corporate officers or shareholders. As a result, it only really affects individual contractors or sole proprietors, who make up an infinitesimal slice of the pie. A 2011 study found that, on average, companies that gave more in campaign contributions subsequently received more contracts. Meanwhile, most regulations on contracting come from the OMB, which doesn’t draw much media attention; many OMB officials leave government to work for contractors.
All this shadiness benefits players on both sides. It’s good for the contractors, because as long as they are a way for Congress to spend money with less accountability, the contracts will keep coming in. Some government officials, meanwhile, like it this way because service contracts are a handy slush fund. Since annual budget allocations are based on what an agency spent in the previous year, everyone in government has a perverse incentive to spend as much as possible. Heading into Q4 and still have a lot of unspent cash? You can’t go above your official personnel allotment, but you can spend some of those leftovers on contractors.
DOD regulations, for instance, specifically exempt service contracts from the reporting requirements in its annual five-year spending projections. And while Congress has required the DOD to report on contracts from the previous year, a 2016 GAO report noted that in its 2014 statement the department simply flouted the law, excluding “up to $100 billion—almost two-thirds—of its estimated spending on contracted services . . . on which it was statutorily required to report.”
You may be wondering where the boundaries are drawn around the work that contractors are allowed to do. The good news is that the Federal Acquisition Regulation (FAR) places two restrictions on swapping in contractors for government employees. The first is the ban on so-called “personal service contracts,” which are defined as contracts that make someone from the private sector a de facto government employee. The second is the prohibition on contractors doing any work that qualifies as “inherently governmental.”
The bad news is that these regulations are a joke. Though technically legally binding, they essentially rely on self-
enforcement. They’re like the “Alarm Will Sound” signs on the emergency doors in the New York City subway. Push the bar—it doesn’t go off.
According to a 2011 GAO report, an internal Army review of commands and headquarters organizations “identified approximately 2,357 contractor FTEs performing inherently governmental functions . . . and 1,877 contractor FTEs providing unauthorized personal services.” (FTE stands for “full-time equivalent,” essentially government jargon for employee.)
No one seems to know exactly what “inherently governmental” means, but contractors clearly do many things that look like exercises of state power. In Iraq and Afghanistan—and other, unofficial war zones all over the world—contractors carry guns, shoot to kill, and interrogate detainees. Domestically, private prison guards exert coercive force over federal prisoners. When the inherently governmental distinction is observed, it’s often in a formalistic way that leads to absurdities and inefficiencies. CIA contractors can pilot drones over Pakistan, but a government employee has to pull the trigger. One civil servant who works in cybersecurity at Homeland Security told me about a contractor who is paid handsomely to work the weekend shift at a 24/7 operations center. Usually nothing happens on weekends. But even if something did, the contractor wouldn’t be able to do anything without going to a federal employee for permission—making his position effectively useless. “He’s getting paid $120,000 literally to just sit there and do nothing,” the civil servant said.
This isn’t new; it’s just likely to get worse if Trump gets the cuts he’s asking for. A 1991 GAO report found that “agency officials stated that the major reasons they use contractors to administer some functions that may be governmental in nature are the lack of authorized federal positions or employees and the lack of experienced federal employees to do the work.” That was before the Clinton administration and Congress cut around 400,000 federal jobs over the next decade.
This mingling of civil servants and the private sector has long prompted hand-wringing about the conceptual implications of outsourcing sovereign powers to for-profit enterprises. But there are also more practical concerns. Jon Michaels, a law professor at UCLA, argues that the contractor workforce gives the executive branch opportunities for “workarounds.” For instance, federal laws inhibit the DHS’s ability to mine Americans’ personal data. But those laws generally don’t cover contractors. So the DHS can hire private data brokers, who can then sell the information back to the government. “DHS thus gets the benefit of more sweeping, intrusive searches than would otherwise be permitted of government officials,” Michaels writes.
Another troubling possibility is using contractors to bypass the career civil servants who might balk at carrying out an administration’s policies. In April, Politico reported that associates of Scott Pruitt, the head of the Environmental Protection Agency and a staunch climate change denier, were urging him to hire private lawyers to draft regulations rolling back Obama-era climate protections. The plan would be a blatant and unprecedented end run around career civil servants. One example of inherently governmental functions under the FAR, by the way, is “determination of agency policy, such as determining the content and application of regulations.”
Here’s another problem that Trump’s plan to cut government jobs will make even worse: the federal workforce is too old.
According to a 2016 GAO report, “In 2014, people 39 years old and younger represented 44.8 percent of the U.S. employed civilian labor force and [only] 29.6 percent of the total civilian federal government workforce.” The problem, the report found, has worsened recently, and caps on federal hiring are a direct cause. “From fiscal years 2008 to 2014, the total number of new federal employees hired decreased by 33 percent, from approximately 164,000 to 110,000 employees per fiscal year. Employees 25 years old and younger have experienced the largest decrease with 58 percent fewer hired in 2014 than in 2008.”
The GAO estimates that 600,000 civil servants—nearly one in three—will be eligible to retire by 2019. Of the IRS’s roughly 80,000 employees, more than half are over fifty, and only about 2,400 are under thirty. According to the Partnership for Public Service, there are nearly five times as many IT professionals over the age of sixty as under the age of thirty across the whole government.
If those 600,000 people all retired in two years, the government would instantly cease to be able to function. Observers have worried for years about this sort of mass exodus of older employees. So far, it hasn’t materialized. What has happened instead is that each attempt to shrink the workforce has just made it older.
There are a few tools available to politicians who want to shrink government. One is attrition: stop hiring, and let the workforce decline naturally as people leave and aren’t replaced. This is very likely the approach the current Congress will take. But trying to shrink government through attrition depends on people leaving in droves, and for the most part, older federal employees haven’t been. Instead, the government ages. It’s simple: if new positions aren’t added, the existing workforce gets older year after year.
Layoffs may be even worse. When an agency wants to lay off employees, it has to take into account length of service before performance rating. This means that younger employees are more likely to get targeted. Even when a certain job held by an older employee is eliminated, that employee gets to “bump” someone lower in the hierarchy and switch into that position. The upshot is that the standard methods for cutting government almost inevitably increase the age problem. In many cases, these older employees are the most experienced and knowledgeable people in government. That’s why the possibility that they will begin retiring en masse is terrifying. By refusing to let the size of the workforce keep pace with the growth of the country, Congress has choked off the flow of younger generations who should be coming through the pipeline.
When these older employees do retire, they too often take their knowledge with them. The lack of middle-tier employees creates holes in expertise. One executive at a major defense contractor told me he notices this all the time when dealing with Defense Department procurement officers. Certain age cohorts just seem to be missing—the casualties of various hiring freezes over the years. Either the officers are old and highly experienced or they’re brand new and clueless. The result—and this is from a contractor’s perspective—is that the government too often writes contracts that are wasteful or doomed or both.
If you’re still not concerned, consider the Social Security Administration. The backlog for appeals over disability benefits is already 1.1 million cases deep (not a typo), thanks largely to understaffing. The typical wait is nearly a year and a half. According to a GAO report, the SSA “could lose nearly 22,500 employees, or nearly one-third of its workforce,” by 2020. Due to recent hiring freezes, the agency hasn’t been able to fill in positions left vacant by retirees. Other agencies, including the VA, face similar appeals backlogs. The GAO report warns that “without a sufficient number of skilled employees, backlogs and wait times could significantly increase and improper payments could grow.”
The idea of getting rid of employees to fix government fits into the time-honored Republican strategy of identifying a real problem, then proposing the exact opposite of the proper solution.
Any serious effort to make the federal government work better needs to start by putting to rest the myth that the bureaucracy is too big. There may well be jobs or departments that don’t need to be there. But if the government itself were too big, it wouldn’t hire legions of contractors to do work that could be done by federal employees. John DiIulio, the author of Bring Back the Bureaucrats, argues for hiring a million more civil servants by 2035.
But repopulating the federal workforce and kicking the contractor habit will require confronting the forces that make hiring contractors instead of feds so attractive in the first place. The most powerful is the politics of small government. But there are also structural incentives to outsourcing that must be wrangled with. We’ve already seen one: secrecy. A baby step toward weaning government off outsourcing would be bringing the data on the contracting workforce to light. DiIulio suggests requiring a “separate act of Congress for each and every sole source deal above a certain dollar value (try $25 million, for starters), and let the public see what ‘the contracting process’ hides.”
The other incentives to contract out come from the shortcomings of federal personnel management. The first is one you’ve probably heard: it’s too hard to fire federal employees. According to a 2013 report by the Partnership for Public Service (PPS), the crux of the problem is the overlapping avenues available for employees to contest firing decisions. Between arbitration, the Merit Systems Protection Board, the Equal Employment Opportunity Commission, and more, appeals can take more than a year. The result is that firings are rare, and federal jobs are too often de facto lifetime employment. Managers wary of hiring someone they then can’t get rid of are drawn to contractors instead.
The answer isn’t to scrap the civil service protections, as many Republicans have long dreamed of doing and federal employees’ unions dread. Due process protections are crucial to protect career civil servants from partisan purges and whistleblower retaliation. And in any case, the problem is not that federal employees have too many rights; it’s that exercising those rights takes too long. The PPS report recommends a simple solution: combining the various review processes into a one-stop shop where all grievances can be considered at once. That would speed up the process and ensure that agencies that staff up can do so without being effectively stuck with everyone they bring in. Another fix that could play well politically is to hire all new federal employees to renewable fixed terms of, say, five or seven years. Most young people today don’t want or expect to work at one place for life anyway.
Surprisingly, in interviews with dozens of people in government, academia, and the contracting workforce, what came up even more than the difficulty of firing federal employees was how hard it is to hire them. That was far and away the most common reason people gave for why an agency would choose to bring in a contractor instead of an employee.
The standard federal hiring process is a nightmare, in part because it is a relic of the pre-internet age. When a department wants to hire someone, it has to post the opening publicly. Today that means posting to USAJobs.com, the portal for nearly every civil service job in every agency across the country. By law, anyone is allowed to apply. It’s very common to receive hundreds of applications for a single entry-level position. Then the human resources department—which is itself usually under-resourced—is required to review every application and sort the applicants into groups based on their qualifications. That takes time. Despite some Obama administration efforts at streamlining, it routinely takes six months or more to hire a new employee.
Then comes the really sticky situation. By law, the federal government must give preference to veterans, especially disabled veterans. It’s a well-intentioned way to reward the people who served in the military, but managers at federal agencies, some of whom are themselves disabled veterans, told me it has drawbacks. When combined with the USAJobs.com system, there is very often a large pool of people tied for “best qualified” applicants based on their basic information. Suppose a department needs to hire three IT specialists. It gets 100 applications, of which twenty fit into “best qualified.” Of those, the three clearly best candidates aren’t veterans, but there are five vets in the pool of twenty. The hiring officer is legally required to choose from the veterans. That means the preference often prevents agencies from hiring the best talent.
But there’s a workaround to all this stuff: hire a service contractor. That appeals not just to managers who need to get their tasks done, but also to political leaders at agencies. After all, if you know you may only have four years to leave your mark, can you really afford to wait a year to hire staff? And if you want the best person for the job, would you really want to risk not hiring the ideal candidate?
Oh, and I haven’t even mentioned the real pain in the ass: security clearances. Even once an applicant receives a job offer, who knows how long it will take to get the necessary clearance; it could be another year. One reason is that—surprise!—there aren’t enough federal employees overseeing the process. So private contractors hire people who already have clearances, then hire them back out at exorbitant rates to agencies that are desperate for manpower and don’t want to wait more than a year to get it.
Fixing the hiring process is a necessary step toward weaning the government off its addiction to contractors. Proof that it’s possible lies in the fact that certain agencies and job categories are exempt from many hiring rules. The Securities Exchange Commission and Consumer Financial Protection Bureau, for instance, are exempted from the standard hiring process and authorized to pay higher salaries in order to compete with the private sector for top legal talent. There is no magic bullet to fix everything about the application process, but the problem is not insurmountable.
The key is finding the political will. As a measure of the power of the political aversion to fixing the federal workforce, consider the 2010 Federal Hiring Improvement Act, which among other reforms would have required agencies to limit hiring times to eighty days. “The bill failed to pass the House,” as the Washington Post editorial page put it, “thanks to lawmakers leery of affiliating themselves with ‘federal hiring’ legislation at a time when government spending is unpopular.”
Given that, it may seem quixotic to argue for an expansion of the federal workforce. After all, presidents and Congresses from both parties have embraced, or at least tolerated, the “government is too big” narrative for half a century. President Obama made some progress on dialing back the pro-contracting drift of the Bush years, but overall didn’t make much of a dent—and he certainly never came out in favor of expanding the civil service. We got to our current position because the GOP is committed to trying to cut the (nondefense) civil service and the Democrats don’t feel any pressure to resist.
Meanwhile, despite the straightforward conservative appeal of saving money by insourcing more government work, Republicans know that bashing civil servants can be very effective politics. Katherine Cramer, in her study of rural Wisconsin voters, The Politics of Resentment, documents an antipathy toward public employees that is hard for someone from D.C. or Los Angeles to fathom. To these voters, public employees are the enemy. They don’t work as hard as rural people, yet they make more money and have better benefits—paid for by taxes. It isn’t that these people believe in small government in some abstract way; they do want government to help them. They just don’t trust it to deliver, so they’d rather at least stick it to the employees who, in their view, are getting rich undeservedly. That’s why Trump’s insistence that he can drain the swamp by firing federal employees has a strange genius to it. To liberals and traditional conservatives, the idea is self-evidently absurd: the president is staffing his government with Wall Street bankers, giving plum jobs to his family members, and spending public money at his own resort properties. But broad segments of the country see lazy, overpaid public employees as the epitome of government corruption.
Even if it’s naive to expect any members of the modern Republican Party to rally behind anything besides more and more cuts, progressives will be making a huge mistake if they continue to go along with the status quo. There is a real and deep sense that government isn’t working. But the right has taken control of the narrative around why it isn’t working. The left needs to take it back.
The problem of a government that is both under-resourced and overly expensive should be obviously bipartisan—and, in saner times, perhaps it would be. But it’s especially acute for Democrats, who today are, strangely, the only party that thinks government is a good thing. If they ever manage to regain power, they will be taking over a government that, after decades of starvation, is teetering on the verge of a catastrophic breakdown: overstretched, too dependent on contractors, lacking young talent, and facing a potential crisis of mass retirement. The more the beast starves, the louder the cries to starve it some more. To make good on any of their policies that rely on government intervention and administration—that is, most of them—future liberal leaders will need to find a way to replenish the federal ranks. In an era in which voter behavior is characterized by deep mistrust of government, it’s malpractice to not have a robust platform of government reform.
The key is to transcend the false choice of big government versus small government. The real choice is between a transparent and relatively competent big government of civil servants and a dysfunctional shadow big government of private contractors. Don’t defend bureaucrats—that’s a snoozer. Attack the contractors. It’s a powerful industry that will put up a fight against anything that threatens its bottom line. But at some point it has to be confronted.
Hell, who knows; maybe Trump will help us snap out of it. Perhaps draconian cuts—and the breakdowns and scandals that will almost inevitably result at hollowed-out agencies—will be just the thing to trigger a moment of reckoning. An enterprising populist candidate could do worse than to point out to the American people who really wins and loses when the functions of government are sold to the lowest bidder.