Respond to this Article November 1989

Congress's Watch Dog: The General Accounting Office

Mostly it still goes for the capillaries.

By John Heilemann

It was barely two weeks after election day when the General Accounting Office (GAO) sent George Bush a 26-volume "transition series" report, telling him the government he was about to inherit was falling apart. "Neglect and mismanagement" had become the government rule. At the Environmental Protection Agency, the Internal Revenue Service, and the Pentagon. At the Social Security Administration. At Housing and Urban Development. The cost of fixing the problems would be "staggering," the GAO said, including at least $ 130 billion to modernize the plants that produce the nation's nuclear weapons, $ 20 billion to fix dilapidated public housing, and $ 50 billion--off by at least $ 100 billion, it now turns out--to bail out the S&Ls. Even the national parks needed $ 1.9 billion to halt their deterioration. Though they rarely mentioned him, the reports added up to an indictment of ronald Reagan.

While unflattering appraisals of the outgoing government were to be expected, no one thought the most systematic attack would come from Comptroller General Charles Bowsher--whom Reagan himself had handed a 15-year appointment, an $ 89,500 salary, a $ 300 million-plus budget, and a staff of more than 5,000. "Gratuitous advice," harrumphed Edmund Haislmaier of the Heritage Foundation. "Mr. Bowsher's latest political exercise," griped The Wall Street Journal. Less partisan voices sounded puzzled. A story in The New York Times marveled over "the habitually cautious" GAO's use of adjectives like "staggering." These reactions say a good deal about Washington, not least of which is that the city expects government agencies to write in euphemisms. But the reactions say a good deal more about the GAO. Here was Congress's watchdog doing its duty--and the sound was so unfamiliar that no one knew what to make of it. Washington's confusion shows both how far the GAO has come in recent years and how much farther it needs to go.

As Congress's investigative agency, GAO ought to be the government's great ombudsman--tracking programs not only for evidence of fraud and abuse but to discern which work, which don't, and why. Its size alone makes its potential powers formidable, with 80 audit sites within the federal agencies and 15 field offices spread across the country. It has the authority to dig through the files of almost every government program and to grill its personnel. But, historically, the agency has suffered a form of institutional myopia, asking too many small questions and asking them too politely to fulfill its mission.

Under Bowsher, the GAO has shown increasing signs of outgrowing its instinct for the capillary, with the transition reports being just one piece of evidence. But the agency still spends too much time on isolated examples rather than trends. It gives too much attention to questions of minor fraud and not enough to questions of broader government performance. And it's much better at dissecting disasters after they happen then it is at detecting and disarming them as they arise. GAO: You've come a long way, baby, but you're still not the ombudsman we need.

If the call for greater GAO oversight seems to repeat the call for greater press oversight, greater congressional oversight, and greater Office of Management and Budget oversight--it does. Maybe someday the duplication of effort won't be needed. Maybe there'll be so many reporters and government inspectors worried about whether the mine safety laws are being enforced that they'll be bumping into each other in the Harlan County mines. But for now it's safe to say that the odds of such mine-shaft collisions are slight; would-be watchdogs needn't be deterred.

GAO: DOA?

When Congress created the GAO in 1921, it vested the agency with a sweeping mission: to investigate "all matters" involving public funds. And Congress went beyond mere rhetoric. To help protect the GAO from intimidation, it created a 15-year term for the agency's head; though appointed by the president, he could be ousted only by a joint congressional resolution approved by the White House--a formula designed to leave him politically invulnerable.

But this grand potential was squandered under the direction of the first comptroller general, J. Raymond McCarl, whose tenure stretched from Warren Harding to FDR. He was a man who couldn't see the trees for the bark. While Harding's pals were cutting sweetheart land deals over Teapot Dome, McCarl marshaled his 1,700-member staff to sift through piles of canceled government checks. McCarl did make a big fuss late in his career when he caught the Tennessee Valley Authority in the unauthorized purchase of a single mule. It was not an auspicious start.

The GAO muddled along in green-eyeshade fashion without significant change until the arrival of Joseph Campbell as the fourth comptroller general in 1955. Campbell left behind a reputation at the agency as a man who "loved to chase the crooks down the halls"--hyperbolic, no doubt, but then look what the GAO had grown accustomed to.

Campbell's biggest bang came from his investigations of defense contracts in the newly named "military-industrial complex." GAO audited hundreds of defense department contracts and found just the kind of thing Eisenhower had in mind when he coined his cautionary phrase. It found huge cost overruns. It found all kinds of cozy relationships between contractor and contractee. And it named names: Hewlett-Packard, Boeing, Lockheed, General Electric, Westinghouse. Even in his heyday, Campbell defined GAO's work more narrowly than he should have--he was looking at the propriety of the weapons' contracts, not whether the weapons would work. But it was a start.

It was also a finish. Listening to Eisenhower carp was one thing, but relinquishing power to a bunch of auditors was something else indeed, and the contractors knew how to fight back, mobilizing allies in the Pentagon and Congress. In 1965, Rep. Chet Holifield held hearings on the GAO's defense work, to explore what he called "the difficult and sometimes awkward situations created by the GAO audit reports." No shortage of Pentagon and industry witnesses stepped forth to describe the difficulties and awkwardnesses. Though the hearings found virtually no errors of fact in the GAO's work, the critics argued that words like "overcharge," "unnecessary costs," and "failure to protect the government interest" had no place in the ombudsman's reports. They called the straight talk "inflammatory, colored, and sensational." Ridiculous as it was, this contention carried the day. Campbell quit.

And the GAO's golden age ended before it began. Following the Holifield hearings, Lyndon Johnson appointed a comptroller general who came to embody bureaucratic caution, the aptly named Elmer Staats. A man who'd rather switch than fight, Staats cut a deal: he promised Holifield that the agency would ease off on the Pentagon, in exchange for a committee report that contained only a mild rebuke of the GAO. This set the stage for a 15-year tenure marked by the scrupulous evasion of controversy. An unsound accounting practice detailed here. A weakened financial control system cited there. Thousands of pale-blue reports issued, but none that would reach a bold conclusion like "Hey, this program is great; expand it" or "This agency has become irrelevant; kill it." In 1970, with Staats's support, Congress even passed a law giving the GAO specific authority to "review and analyze the results of government programs." (Our emphasis.) It was a potentially revolutionary idea, but the accent, then and now, has remained on potentially.

Bowsher's beginning

No doubt about it, Staats's successor, Charles Bowsher, has helped build a better GAO. Judged only by his credentials--Reagan appointee, certified public accountant, partner at Arthur Andersen--Bowsher didn't seem the likeliest candidate to shake things up. But he's made the agency an increasingly interesting place to work, and an increasingly relevant one. Last year The Washington Post cited a GAO report at a rate of just slightly under one a day. Bowsher has put an end to the longstanding practice of issuing anonymous reports; these days, they carry the evaluators' names. He's also instituted a pay scale that pegs advancement to performance, a change the rest of the government should follow. "They've become more courageous in what they go after," says Frederick Mosher, a historian who has authored two books on the GAO. "Bowsher seems to be using his position more in the manner it was originally intended--as the government's ombudsman."

For a case in point, witness the GAO's renewed involvement in reviewing defense work, a sign that the ghost of Holifield may finally be fading. What's especially significant about the agency's recent defense reviews is that they haven't focused just on potential fraud but have gone further to ask "Is this weapon needed?" and "Does it work?" In the case of the Bigeye Bomb, a chemical weapon with an unfortunate habit of exploding before it's dropped, the GAO answered, "No, it doesn't." Likewise with the Sgt. York, an anti-aircraft gun known for mistaking latrines for airborne targets. The GAO's critical reports helped the weapon meet its deserved demise. And the agency hasn't been simply a bearer of bad news; its favorable review of the Head Start program helped stave off the budget axe.

Efforst like these have sent GAO's stock soaring on the Hill, where members used to gripe that the agency was slow and unresponsive. Requests from individual representatives or committees now account for about 80 percent of the nearly 900 reports GAO churns out annually; 15 years ago, that figure was only 20 percent, and surveys showed that a majority of congressmen had never even read a GAO report. Now, hardly a day goes by without one member or another referring to GAO as "indispensable."

And that's one of GAO's current problems: its reliance on others for direction. For one, this keeps it from doing enough to head off disasters in advance. More often, GAO arrives on the scene in the aftermath, like some super State Farm agent, clucking over the mess and adding up the bill. It took the nation's nuclear weapons production sites 30 years to deteriorate, but GAO didn't begin looking into the situation until Congress specifically asked it to--first, with some very narrow audits, after the accidental release of uranium dust from an Ohio plant in 1984, and later with a somewhat broader look at safety demanded by Congress in 1986, in the wake of Chernobyl (not the kind of clue you want your investigators to be waiting for). "It just wasn't something we'd thought to check into--there had been no need to," says Carl Bannerman, GAO's assistant director for energy issues.

In 1988, an FBI sting called "Ill Wind" implicated scores of defense contractors and consultants in bribery and influence peddling, and GAO is now working on a follow-up report about the industry. But why didn't the agency catch on while the problem was occurring? "Well, that's a good question," said Jim Johnson, a GAO director in the defense division. "And quite frankly, I don't know why. We hadn't been asked to do so, and nowadays most of our work is in response to congressional requests. With hindsight, I think it's obvious we should've been looking. Every time one of these scandals breaks, I say, 'Jesus, where was GAO on this?'"

The habit of waiting for orders from Congress not only makes GAO late but promotes a ridiculous narrowness. "We'd rather look at broader programmatic areas, ones that have national importance," said one GAO issue area director. "But often these congressmen say that it'll take us too long, and they don't have time to wait. Either that, or they're unbelievably blunt. They say, in essence, 'I don't want you to go beyond my specific, individual request because I don't care about the broader issues. I only care about my problem.'"

Thus are born countless GAO reports like "Corps' Permitting of Louisville Floating Restaurant," in which the agency double-checked a decision by the Army Corps of Engineers to let the "Islands" restaurant bob in the Ohio River. (Thumbs up, said GAO.) Or "Labor-Management Relations and Customer Services at the Simi Valley, California, Post Office" (GAO: They were bad, but then they got better). While the FBI was breaking the "Ill Wind" investigation that involved hundreds of millions of dollars in military contracts, GAO was authoring its tour de force, "Award of a Contract at Whidbey Island Naval Air Station."

GAO shouldn't wait for Congress to cut the apron strings. One way to stop saying "Jesus, where was GAO on this?" is to institute a regular schedule for reviewing major agencies and programs. While it's true that GAO has shown more initiative than it once did, the distance it still needs to travel is suggested in the agency's nomenclature--staff members still call themselves "auditors" rather than "evaluators." It's a revealing term. Auditors, as one senior staff member explained, aren't interested in "coming in and looking for what's wrong. We assume that everything is OK until we find out differently. . . . Investigating just isn't part of our job."

Isn't part of our job?

Writer's block

Another set of problems arises not from the way GAO conducts its studies but from the way it reports them. If you want the world to take notice of your findings, you have to report them loud. And clear. And in terms that convey their importance. One government agency that excels at this is GAO's congressional cousin, the Office of Technology Assessment, which also reviews programs and policies (see Scott Shuger, "How to Revolutionize Washington with 140 People," June). Here is the way OTA helps convey the meaning behind its report on infertility treatments: "Lately you cry at the drop of a hat--when you see a diaper commercial on television, see a pregnant woman at a grocery store, or get an invitation to a baby shower. The whole world seems to be having babies."

Now here is the GAO, commenting on the federal government's failure to assess how well the American education system is working: "The mission statements of the three units we examined are general, providing considerable latitude for the development of a portfolio of information. To insure that these functions are being properly carried out, different review methods have been tried out, varying the authority given to the reviewers (policymaking versus advisory) and the regularity of the review (periodic versus ad hoc)."

The first thing the GAO should do to solve the problem is start hiring people who can write. Like journalists--the best of whom also know how to root out the truth about what's happening in government. The best journalists have mastered the skills that good program analysis requires: an ability to sift through the numbers, yes, but also, and more important, a talent for firsthand observation. For asking good questions. For thinking through the meaning of the answers and drawing incisive conclusions. But instead of hiring talented writers, GAO sends its new evaluators--a mix of public policy school graduates, accountants, lawyers, business administrators, and economists-- through a day-and-a-half writing course. A day-and-a-half! As though someone who's plodded through high school, college, and graduate school without learning to write will suddenly blossom in the hands of the bureaucracy's Berlitz. The fact that the course director explains it's designed to teach the staff to write "like journalists" only punctuates the point: Why doesn't the agency hire journalists in the first place?

The second thing the GAO should do is ease off on the writing review process. Studies must pass through as many as eight layers of editorial and content review, often with line-by-line editing at each stage. As one GAO staff member says, "We write for the next level of review"--turning the agency's reports into chain letters for bureaucrats.

Duds at HUD

If ever the new, improved GAO had a chance to show its stuff, it was at the Reagan administration's Department of Housing and Urban Development. On paper, it was a match made in heaven: He was a multi-billion-dollar agency, woefully mismanaged. She was a government watchdog with 40 auditors on the department's site. But while the two seemed made for one another, GAO brought only its worst qualities to the relationship.

In early 1984, HUd was subject to the first of what would become an ongoing series of GAO "general management reviews." The report promised to evaluate HUD's success both in improving its program management and strengthening its internal controls. Thus, it had a chance to uncover programs that weren't working, or had perverted their mission. (Of which there were plenty.) It also had a chance to ferret out old-fashioned abuses, like sweetheart deals. (Of which there also were plenty.)

The report was more than a year in the making. And here's how it began: "The Department of Housing and Urban Development has played a major role in upgrading the nation's housing and providing community development aid to cities and counties." It was downhill from there--212 pages of pap. Dig through it long enough (as embarrassed GAO officials recently have), and you'll find a few remarks that seem vaguely on the mark. "HUD's top management changes frequently. . . . As a result, institutional memory at a high level does not exist to understand the causes of HUD's longstanding management problems and the implications of alternative solutions."

Revisionist stories in the press have seized on lines like those to cobble together an argument that GAO was ringing a bell while HUD was burning down. More to the point, however, were lines like these: "HUD has made significant strides toward resolving many of its management problems. For example, HUD has increased its efforts in monitoring to prevent fraud, waste, and mismanagement and is improving its financial management activities." (Emphasis added.) Now if these words were superfluous bromides, they didn't belong in the report. And if they were the candid assessments of the evaluators, then they didn't belong in the report.

GAO followed up on this incisive look with a few of its exercises in congressional errand-running--reports like "Conditions at the Laurel Park Project in St. Petersburg" and "HUD Oversight of the Villa San Pedro Project." Some of these reviews did find evidence of the mismanagement that characterized HUD programs nationwide, but no one at GAO thought to piece the fragments together. GAO also conducted at least two reviews of the Section 8/Moderate Rehabilitation program that captured last summer's headlines. Neither mentioned that the program had become a cash cow for powerful Republicans.

"I guess we got a little too complacent," said John Ols, who took over GAO's HUD oversight a year ago. "HUD was not a big priority in the administration and Congress didn't care enough to give it decent oversight. Nobody cared about HUD--and we just went the same way as everyone else."

"You get very settled," he said. "You start to see the agency like the agency sees itself."

It's an old problem, for the nation's one-eyed watchdog.

John Heilemann is a contributing editor of The Washington Monthly.


Mission   Masthead   Features Archive    Writers Guidelines   
Feedback   Customer Service    Subscribe Online    Make A Donation

This site and all contents within are Copyright © 2006
The Washington Monthly 1319 F Street N.W. #710
Washington DC. 20004.
Comments or questions ... please email editors by clicking here