Print this article | Return to Article | Return to CFO.com
Two decades of failed Pentagon financial reforms put more than just dollars at risk.
Kris Frieswick, CFO Magazine
October 1, 2004
Like a second marriage, each new Department of Defense financial-management reform initiative represents the triumph of hope over experience. The DoD, with more than $1 trillion in assets, 3.3 million employees, and a budget of $417 billion for fiscal year 2005, is the largest single business organization on the planet. That it can't accurately close its books is not surprising. That it continues to try is not surprising. The surprise is that after three elaborate attempts to transform its financial and business-management processes since 1986 — together costing at least $35 billion by some estimates — the department seems not much better off than it was two decades ago.
Officials think the fourth time may be the charm. "We will get a clean audit in 2007," says Linda Furiga, comptroller of the Defense Logistics Agency (DLA), which is responsible for much of the department's procurement, supplies, and distribution. She calls the agency's experience with the newest reform effort, the Business Management Modernization Program (BMMP), "the most exciting time we've had in many years. It's like they've shone a light on accounting."
But can the world's largest business organization finally approach the transparency and integration of its private-sector brethren? True, the BMMP, which is overseen by DoD comptroller Tina Westby Jonas, has high-level support, starting with Defense Secretary Donald Rumsfeld. And advances in Internet and enterprise-resource-planning (ERP) systems make real integration and interoperability among services more feasible. Still, the dismal history of past reform efforts argues for doubts.
In 1985, in light of the infamous $600 ashtray scandal, President Reagan's Blue Ribbon Commission on Defense Management was charged with streamlining financial and acquisition policies. Two years later, the executive leading the reforms resigned, calling the Pentagon "not prepared to change the status quo."
Then, in 1989, Deputy Defense Secretary Donald Atwood launched the Corporate Information Management Initiative with great fanfare. Its target was $36 billion in savings by 1995, but a Clinton Administration study later threw cold water on its work, citing significant resistance to the initiative's proposed changes. Most were essentially abandoned, after soaking up $20 billion in spending in eight years.
The cost-focused Defense Reform Initiative followed in 1997, spearheaded by President Clinton's Defense Secretary, William Cohen. Because the individual services and Pentagon agencies were expected to foot the bill, with each one implementing its suggestions, no accurate accounting exists of what was accomplished or how much was spent on the initiative. According to a report issued several years ago by the General Accounting Office (GAO, now called the Government Accountability Office), it failed because its leadership, drawn from the DoD's various segments, couldn't put their individual interests aside to focus on departmentwide reform.
Four years later, in 2001, Secretary Rumsfeld ordered the latest study of department financial management. Its findings resulted in the BMMP.
Complicating today's BMMP financial reforms, of course, are other challenges the DoD must meet, starting with the Iraq War and encompassing questions about its Halliburton relationship, Air Force-Boeing tanker-plane contracts, and revolving-door employment agreements involving government personnel, to name just three.
But while the Pentagon's size and complexity may excuse some of the financial-reform failures, the core problem is more fundamental. "The department has never come to grips with what it is," says Dov Zakheim, DoD comptroller from 2001 to 2004 and now a partner and vice president at Booz Allen Hamilton. "You've got this fundamental dichotomy that the department has never really been able to resolve for itself." It was Kennedy Administration Defense Secretary Robert McNamara, a former Ford Motor Co. president, who first envisioned the Office of the Secretary of Defense as a corporation, with the military services and agencies as its profit centers, notes Zakheim. The services, meanwhile, tend to see themselves as independent corporations, with the OSD as their holding company.
The chief goal of all the financial-management initiatives has been not just a clean audit, but some transparency at this huge bureaucracy. Prior reforms all died on the vine because they couldn't dent the impregnable walls the military services and agencies have built up around their processes, procedures, and chains of command.
Congress is also at fault. In the main, senators and representatives want to overhaul DoD systems for the national good only so long as the change doesn't interfere with their ability to steer contract-spending to constituents. Real reform would profoundly disrupt that time-honored tradition. Financial-management reforms have also failed because, despite the big talk at their inception, turnover in senior positions has meant inconsistent leadership.
Although some of the previous efforts probably "have resulted in some marginal improvements," says Greg Kutz, the GAO's director of financial management and assurance, "they certainly haven't transformed the way the department does business." Their value, he adds, has amounted to far less than the billions of dollars that have been spent on them.
Instead, each investment in reform ended up establishing ever larger, more expensive financial and management systems, most created at the service and agency level for their exclusive use. What's left is a largely ineffective congregation of siloed financial and business-management systems.
The total number of Pentagon systems is conservatively estimated at more than 4,000, including at least 542 accounting and finance systems, 143 acquisition systems, 210 planning and budgeting systems, 665 human-resources systems, 565 logistics systems, and, ironically, at least 3 systems for inventorying all these business systems. In fiscal 2004, which ended September 30, the DoD spent $19 billion just to maintain and upgrade all the parts of these systems — $5 billion on modernization alone. Only a handful of these are interoperable, and many of them still rely on manual reentry of data between systems rather than electronic integration.
A Heavy Price
In a corporation, lack of integration and transparency would hinder efficiency, productivity, and profitability. At the DoD, it could kill someone.
The DoD still hasn't been able to track down and recall, for instance, more than 250,000 chemical and biological protective suits purchased from New York-based Istratex Inc. in 1989 and 1992 — even though in 1995, Pentagon investigators alerted the DLA that the suits were potentially defective. Five years later, after completing numerous tests that showed wearers still would be exposed to the deadly threats around them, the DLA recalled the suits. (Istratex executives were indicted in 1999 for fraud, and pleaded guilty to lesser charges.)
After the recall, which eventually resulted in the return of about two-thirds of the suits, thousands of those reclaimed were inadvertently resold to public, local, and state law-enforcement agencies by a division of the DLA. When the DLA recalled those suits, they were, incredibly, again resold. The practice continued until at least June 2003, when the GAO bought some defective suits on the Internet in an undercover operation. The GAO fears that many suits are still with active military units or local law- enforcement officers.
The problem with tracking the suits? The DLA, which works for all the military services and agencies, considers the Army, the Air Force, and the Navy/ Marines organization to be separate customers. Furiga says that once an item is "sold," it leaves DLA's inventory system and can be tracked only via its customers' systems.
If a private company bought an item from another company, says Furiga, that item would become an asset on the buyer's financial statements and would be recorded on its books.
Such an approach makes the DLA's inventory-management job easier, but it makes accurately tracking defective material nearly impossible, says the GAO's Kutz, especially since the military services have widespread asset-tracking problems of their own. Adds Kutz: "That's like Wal-Mart distribution telling you they can't find all of a batch of poisoned Tylenol because they sent it from their central distribution depot to a store in Des Moines — and now it's off their books."
What upsets Kutz even more is that the DLA is $1 billion into a new business- system modernization implementation with SAP in an effort to streamline its procurement and material-management function. Oddly, the new system has no inventory-tracking module. Instead, it interfaces with the DLA's existing inventory system, the Distribution Standard System, which tracks inventory at its 23 depots worldwide. Data is fed back and forth between the two systems through electronic data interchange technology, but the new system still offers no access to inventory data beyond the DLA's own walls. "They had their own money; they had their own set of solutions; and they bought a system to deal with their problem," says Kutz. "They didn't think of interfaces or integration or a department-wide solution."
In its defense, the DLA's modernization project was conceived and approved before the beginning of the departmentwide BMMP. The DLA is also working on a new project-management system, the Integrated Data Environment, which would allow users to pull down information on assets logged within multiple inventory-management systems throughout the Pentagon. But that project is still in its infancy, and depends on the rest of the DoD having reliable systems in place. In the interim, the status quo prevails.
And the fact remains that four years after recalling the chemical and biological protective suits, the DLA uses the same recall process it used in 2000: data calls. According to Allan Banghart, director of enterprise transformation at the DLA, an agency recall today would involve notifying — by phone, E-mail, or memo — any unit that received a defective batch, and asking for its return. For suits that left the DoD, the only alternative is to issue a general call to law-enforcement agencies throughout the country.
Could the BMMP Fail?
The Pentagon initiated the BMMP in response to such operational problems. Among its mandates: to create a business-enterprise architecture mapping all the business systems across the DoD's branches to see which ones need to interact. The BMMP is developing a set of interoperability standards that all system purchases must meet.
To ensure adherence to those standards, the BMMP created six separate "domains" based on function: accounting and finance, acquisition and procurement, human-resources management, logistics, strategic planning and budgeting, and installations and environment. Throughout the DoD, anyone seeking to purchase a new business system must now get a review and approval from the domain leader responsible for that particular type of system.
Over time, the domains will also certify existing systems to determine whether they fit with the business architecture. The BMMP sets performance metrics, establishes a compliance-assessment process, and tests integration efforts to make sure they work.
More than $203 million has already been spent on the BMMP. This year's budget request is $226 million, with spending of $494 million expected by 2009. So far, however, the three-year effort has proven as glacial as the reforms that went before, according to the GAO. While praising the attempt, the office complains of its slow speed and lack of oversight, and warns that it could fail as the others did.
The writing may already be on the wall. The House and Senate versions of the Defense Appropriation Act of 2005 cut BMMP funding by between $40 million and $97 million. The program has failed to produce an architecture that addresses all of the GAO's concerns, nor has it made significant progress in creating DoD-wide standards.
Blame the Budgeting
Some explanation can be found in the history of the U.S. military. Until the end of World War II, each service was run as a separate entity. Not until 1947 were the services joined under the OSD's civilian authority. Ever since, Presidents have attempted to unify the operational leadership of the services, with only moderate success. (Not until the 1986 passage of the Goldwater-Nichols Act, in fact, did primary responsibility for strategic wartime planning become the purview of the chairman of the Joint Chiefs of Staff.)
That parochial mind-set has left its imprint on each service's self-designed systems, policies, and procedures. The passage of the CFO Act of 1990, requiring audited financials from most government agencies, exposed much of the disconnectedness by requiring the DoD to generate coherent data. DoD insiders say the department's ability to wage war and maintain bases is a testament to the creativity of the men and women of the services, who often develop elaborate workarounds to get the job done.
The Pentagon budgeting process has abetted the stovepiping of systems. Under Title 10 of the U.S. Code, the government's rule book, each branch of the military services is funded through congressional appropriations that go directly into each service's coffers. Each service has had complete control over how its IT money was spent, without a DoD-wide financial spending control mechanism. Once money has been appropriated, very few controls govern how it is spent. In the past, this made centralized control over spending on nonintegrated business and financial systems very difficult, if not impossible, and led to a scuttling of attempts to corral outlays.
Until very recently, the BMMP fared little better. As part of the program, Congress passed legislation starting in fiscal year 2003, mandating that any business or financial-system purchase over $1 million be certified by the DoD's comptroller, to ensure that the system is developed and managed in accordance with BMMP standards.
The services, according to a GAO study, largely ignored the requirement, and bought $863 million worth of systems without certification between December 2, 2002, and September 30, 2003.
The GAO cited poor communication about the new requirement at the program-executive level, where most of the systems were being acquired. Further, those violating the certification requirement faced no consequences.
Weaknesses in the traditional DoD financial-reporting system also impede reform. Because the focus of the DoD's financial efforts has traditionally been on the budgeting and appropriations process, little attention has been paid to preparation of accurate statutory financial reports. The reports, of course, are intended to show how the department spends its money.
But a lack of integrated financial and business systems, coupled with little financial-reporting experience, leads some to find the DoD's year-end financials almost laughable. For instance, the DoD's financials report $1 trillion in assets, which the GAO claims is probably grossly understated. "No one is really sure what the real number is," says Kutz.
Keep It Opaque, Please
If the BMMP goal of a clean fiscal 2007 audit is met, it would be the department's first ever. But clean audits mean transparency, and operational transparency remains public enemy number one for many at the DoD, who, like managers at many government agencies, are generally promoted for their ability to create, get funding for, and run programs. The size of the program matters for promotional purposes, even if it contributes little to the department.
The information needed to run cost-benefit analyses on those programs, meanwhile, is hopelessly trapped in an inefficient swamp of financial systems.
"What is optimal for the program manager is suboptimal for the DoD departmentwide," says Zakheim diplomatically. Translation: little incentive exists for program executives to help implement systems that could hold them more accountable.
This may go a long way toward explaining why the 2003 certification requirements were summarily ignored, although another factor was the BMMP team's slow creation of compliance standards. In the certification requirements' early days, says David Smith, finance and accounting domain director in the DoD comptroller's office, standards for his domain were developed by codifying "all the requirements that were already out there" — from the CFO Act, the Federal Financial Management Improvement Act, the Office of Management and Budget, and other existing legislation — and identifying material weaknesses from prior audits.
But a concrete overall plan must wait until the business-enterprise architecture plan is complete. The GAO has criticized previous versions for failing to address entire functional areas. (A recently proposed version of the architecture plan "is significantly more mature," says Elizabeth McGrath, deputy director of the BMMP. "It takes onboard a lot of [the GAO's] overarching recommendations.")
Transparency is often unwelcome on Capitol Hill, too. Representatives and senators have actively demanded hearings on such troubling events as the missing chem-bio suits and persistent Army payroll problems. But protecting constituents often comes first — and that sometimes means helping companies in their districts sell products and services directly to the military branches and agencies, even if they don't meet interoperability standards.
Additional transparency also would mean the death of more inefficient programs that may involve constituent companies. Noting previous legislative initiatives designed to reform the DoD systems, Bill Phillips, global defense practice leader for IBM Business Services in Washington, D.C., says: "If we truly did everything that those acts wanted us to do, and did it right, think of the challenges that would present to Congress."
Were the reforms to take hold and make operations transparent, says Phillips — who is also a member of the Defense Business Board formed by Secretary Rumsfeld to help steer transformation efforts — it might expose that "some depot, in some state somewhere, is totally ineffective...we don't need it." And that could cause distress in the representative's district. "What I'm suggesting," he says, "is that Congress doesn't want to be that transparent." (IBM Business Services has a long history of consulting on DoD systems reform.)
No Command at the Top
Another impediment keeping financial and business-system reform initiatives from taking hold is the lack of consistent, long-term senior financial leadership throughout the DoD.
DoD comptrollers have had an average tenure of 3.5 years — just short of a Presidential term — and the workload is intense. In Zakheim's 3 years, he had to help fund the Afghanistan and Iraq war theaters, prepare four annual budget requests, and run the financial management of the DoD. He found that the BMMP alone could have consumed him full-time. (Comptrollers of the military services, where much of the hands-on work of financial-reform projects takes place, serve an average of 3 years, and all are political appointees. Many lack real financial experience.)
Michael Montelongo, comptroller of the Air Force, sees high leadership turnover as a major deterrent to reform efforts. "I have a challenge in terms of tenure because I'm a political appointee," he says. "These things take 7 to 10 years to get traction."
According to a 2001 report by the Rand Corp. think tank, Senate confirmation now takes 8.5 months on average, and one in five appointed positions are vacant or filled by an "acting" official at any given time.
Can They Do It?
With these odds stacked against it — and in the midst of war — why would there be any optimism at the DoD about the prospects for the BMMP?
One difference from previous reform efforts is that the creation of the comprehensive business-enterprise architecture, including the domain structure, will be codified into law in the Defense Authorization Act of 2005. That means it cannot be tossed out with a new Administration. The act makes it a legal violation to spend money on a business system without certification. And one Senate Armed Services Committee staffer believes the language of the act will make it more difficult for the services to ignore the rule. In prior legislation, the DoD claimed the language was unclear. "Today," he says, "I think we've nailed the door shut on that."
The GAO and the Defense Business Board both strongly believe that the only way to control proliferation of noncompliant systems is to shift the money for such systems from the individual services and agencies to the control of the domain leaders. Still, it is highly unlikely that Congress will vote to shift funding away from the services. "The services want as much control over that money as they can get," says Zakheim, and their opposition to change would have powerful support in Congress. "The way the budget works allows [the services] to move monies around as necessary. If they don't have access to the money, they can't move it around." Obviously that's not the official objection the services raise when the GAO presses them to shift control to the domains, says Zakheim, but it's still the reason.
The good news is that even without control of funding at the domain level, the domain system is starting to have an impact — by throwing a monkey wrench into the congressional practice of greasing the DoD's wheels for constituents.
Last summer, Digital Systems Group, a Warminster, Pennsylvania, business-systems vendor, was eliminated as a bidder on a financial-systems contract within the DoD because it didn't meet the systems requirements. After being eliminated, the company enlisted the support of two politicians, Rep. Jim Gerlach (R-Pa.) and Rep. James C. Greenwood (R-Pa.), who, along with a DSG vice president, met with David Smith, the finance and accounting domain leader. They demanded that he explain the decision. They also pressured Smith to allow DSG to bid on a contract for the Missile Defense Agency, which is seeking to upgrade into an integrated, cross-agency financial system. The DSG system being offered, however, was yet another small, stand-alone, siloed financial solution that didn't meet the domain's integration standards. Smith turned down the request, but reminded the vendor that it could bid on plenty of other contracts. Smith shrugs off the incident, which he calls "isolated," and adds: "I don't have a problem saying no."
Fewer Turf Wars?
At the level of the individual services, the reform mantra may also be starting to have an impact, occasionally overbalancing the "program protection" mind-set that has been a factor in scuttling previous reform efforts.
Defending program turf is beginning to lose its luster as a path to career advancement in the Air Force, says Montelongo, and job advancement more commonly reflects a focus on actions that improve overall Air Force capabilities. "I'm seeing folks who are preaching the core-capabilities model moving ahead," he says. "Those who aren't — well, the reverse is happening to them."
There is also some hope for efforts to fill the persistent leadership gap. At the end of his tenure, Zakheim, supported by the GAO, was among those recommending the creation of a DoD chief management officer or chief operating officer position, a post dedicated to overseeing the department's business and financial-management transformation. Such an office, just below deputy secretary level, would involve an appointment for more than four years. "It isn't necessarily a guarantee of success, but it guarantees focused, sustained leadership that survives an Administration change," says the GAO's Kutz.
Of course, serious reform in so enormous an organization will still take a Herculean effort. "People forget that this is extraordinarily difficult," says the Senate Armed Services Committee staffer. "It's an unbelievably complex and out-of-control system."
Yet hope is expressed even by some who have led failed efforts at DoD reform. "Achieving transparency, streamlining the processes, and [installing] interoperability is doable," says Paul Strassmann, a longtime CIO who helped run the Corporate Information Management Initiative in the early 1990s and who now works on the Department of Homeland Security's systems-integration project. But it will take "at least seven years of very persistent effort, and perhaps more," he says. "What you're dealing with is a swamp full of muck, and you want to make a swimming pool."
Kris Frieswick is a senior writer at CFO.
|If the DoD Were a Country...
...its financial and business-system spending would top the GDP of many nations.
|67 Syrian Arab Republic||$21.5|
|70 Serbia and Montenegro||19.2|
|>> Budget for DoD spending on finance and business-systems maintenance and modernization, FY 2005.||$19.0|
|73 Sri Lanka||18.5|
|Source: World Development Indicators database, World Bank|