Black Budget in the Red

This tale of government bloat was unspooled in a lengthy Washington Post series, which described a federal leviathan "so large, so unwieldy and so secretive that no one knows how much money it costs, how many people it employs, how many programs exist within it or exactly how many agencies do the same work."

And yet conservatives, especially in the years since the 9/11 attacks, have been reluctant to apply their own insights to the subject of the Post's exposé: the American Intelligence Community (IC). If the Post can discover that government is wasteful, can conservatives begin to think of the IC as one more bundle of government programs, with all the faults to which that breed is prone?

There are, of course, obvious differences between the IC and other agencies: Nobody doubts that the FBI and the NSA serve vital functions. And if $75 billion per year is the price of detecting and preventing plots to murder Americans by the thousands, it would be hard to call it money wasted.

Yet the most compelling conservative arguments for skepticism about runaway government growth have never depended on the worthiness of the goals at which government aims. Rather, conservatives have drawn on the insights of public-choice economics, which predicts that rational bureaucratic actors — often in collusion with profit-seeking firms — will more reliably act to maximize their own power and budgets than seek the general welfare. They have borrowed the insight of Friedrich Hayek — back on the best-seller lists after six decades, thanks to the tireless promotion of Glenn Beck — that there are limits to the volume of dispersed information any centralized authority can effectively manage.

Yet conservative jeremiads against federal pork seldom focus on examples like — to pick one boondoggle that became public — the NSA's Trailblazer. The Science Applications International Corporation, one of the 800-pound gorillas of intelligence contracting, signed a $280 million contract to set up this classified data-mining system in 2002, as reporter Tim Shorrock recounts in his 2008 book, Spies for Hire. NSA veteran William Black, who'd been hired on as a vice president at SAIC "for the sole purpose of soliciting NSA business," returned to his old agency to run the project. More than three years later, having run up a tab of at least $1.2 billion, the system was scrapped. The contract to build the successor system went, of course, to SAIC.

Or consider the controversial program of warrantless wiretapping authorized by Pres. George W. Bush. The political debate over that program — later revealed also to encompass large-scale data mining, perhaps of the sort Trailblazer had been meant for — centered above all on weighty legal questions about the balance between privacy and security interests and the legitimate scope of executive power in wartime.

Yet surely the more obvious question was: Does it work? The only assurance we had that it did came from the very officials tasked with running it — the kind of testimony conservatives rightly greet with an arched eyebrow when it comes from an EPA administrator or a jobs czar.

When the inspectors general for the IC finally produced an unclassified report on the "President's Surveillance Program" in 2009, they concluded that the large majority of the leads generated by the program had no connection to terrorism — corroborating early press reports in which FBI officials complained of being sent on wild-goose chases. "Most IC officials interviewed" by the inspectors general, the report concluded, "had difficulty citing specific instances where PSP reporting had directly contributed to counterterrorism successes." The classified version of the report cites instances in which the program "may have contributed" to an intelligence success. It's hard to be reassured that this legally controversial program was the best use of the available resources — especially if it was generating so many false hits.

Intelligence agencies may be discovering the "fatal conceit" that Hayek ascribed to advocates of economic planning: the belief that sufficiently brilliant experts can effectively aggregate and understand the information flowing through a modern economy. Our high-tech spies now aspire not simply keep tabs on specific suspected terrorists but to harness blazingly fast computers to automatically detect their traces in the bitstream of 21st-century financial and communications networks.

The result is a community choking on information it cannot process. Every day, according to the Post's report, NSA's collection systems "intercept and store 1.7 billion e-mails, phone calls and other types of communications," a tiny fraction of which are processed and stored in some 70 databases. A 2005 inspector general's report found that the FBI had collected, just in the previous year, a backlog of untranslated intelligence intercepts amounting to 87 years' worth of audio.

The information problem faced by analysts repeats itself at the management level. One of the handful of "Super Users" interviewed by the Post, an intelligence official meant to have full access to the Defense Department's classified intelligence activities, conceded, "I'm not going to live long enough to be briefed on everything." A similarly resigned take was offered by Pres. Barack Obama's nominee to serve as director of national intelligence, Lt. Gen. James Clapper: "There's only one entity in the entire universe that has visibility on all [Special Access Programs] — that's God."

The problem is then compounded by the intersection of perverse political incentives with the compartmentalization, the complexity, and, above all, the secrecy in which intelligence work is shrouded. While the House and Senate intelligence committees have primary jurisdiction in principle, their authority overlaps with that of the judiciary and appropriations committees — with the latter often having more practical say over intelligence expenditures, despite a paucity of the cleared staff that would be necessary to do serious scrutiny of the classified intelligence budget. And the rewards are slim for members of Congress wondering whether to invest precious time and political capital in trying to guarantee the efficiency of vital intelligence programs. Legislators seeking to face down entrenched bureaucracies — and corporate behemoths eager to protect their $50 billion share of a $75 billion intelligence budget — can't easily go on cable news to rally the public against ineffective or wasteful programs, or to trumpet their achievements after the fact if they succeed. Instead, oversight tends to follow what intelligence scholars have dubbed a "fire alarm" model: periods of intense scrutiny in the wake of a prominent scandal or failure, followed by long stretches of apathy.

Even if our burgeoning surveillance state posed no long-term structural threat to the privacy and civil liberties of ordinary Americans, it would be mysterious that many conservatives are reluctant to apply to the intelligence community the same standards and the same skepticism with which they greet any other well-intentioned government program. Why should we believe that throwing more money at a problem through government will produce better results when subject to less outside scrutiny?

One possibility is that conservative principles have, in the intelligence arena, become a casualty of the culture wars. During the debates over the warrantless-wiretap program, Sen. Orrin Hatch (R., Utah) bristled that concerns about abuse of this broad new spying authority constituted a "slap in the face to the people who protect our nation."

It was a familiar motif in a broader narrative often deployed by conservatives: Leftists attack our troops and intelligence officials, while conservatives support them. But patriotism is no vaccine against the pathologies of bloated government — nor should it be a soporific to conservatives who, in any other sector, would be wary of a bureaucrat with an ambitious plan and a request for a blank check.