When the Government Accountability Office's top IT management investigator, David Powner, was called to testify before a House subcommittee on July 25 about the progress of federal IT reforms, he had plenty to say. If only he had taken the opportunity to say it.
Powner has submitted no fewer than a dozen reports to Congress since January detailing where and how the government needs to improve the way it acquires and manages more than $80 billion annually in IT spending. But when it came time to speak during the 80-minute hearing, Powner was given little opportunity to provide details on what government agencies can do to tackle what everyone agrees is too much duplication in government IT spending.
Instead, he gave Representatives John Mica (R-Fla.), Mark Meadows (R-NC) and Gerry Connolly (D-Va.) -- and the others on the House Oversight and Government Reform Subcommittee on Government Operations -- the kind of generalities and sound bites the committee was presumably looking for.
The question of the hour was how the Office of Management and Budget and the General Services Administration could have been so far off the mark in estimating the number of government data centers. For the past two years, OMB has maintained that number was just over 3,100. In recent weeks, OMB has acknowledged that the total number of data centers now stands at more than 7,000.
[ Would a cabinet-level Department of Technology help eliminate inefficiencies? Read Do We Need A U.S. Department Of Technology? ]
When Subcommittee Chairman Mica asked Powner during the hearing whether he considered the latest tally of government data centers to be the final total, Powner responded, "I wouldn't bet it on it."
Sadly, what could have been a constructive dialogue between Congress and federal IT officials about how to actually address wasteful IT spending amounted to another show for Congressional cameras. Even more disappointing is what the cameras revealed: Subcommittee Chairman Mica's fumbling grasp of the technical facts and governance issues that stand in the way of the reforms he and his subcommittee keep pledging to implement.
Federal CIO Steven VanRoekel tried to articulate during the hearing how he had revised the definition of what constitutes a federal data center, to more accurately reflect the government's assets. Contributing to the difficulty of getting a clear count, added GSA associate administrator Dave McClure, is the fact that CIO's aren't always forthcoming about where their data center and IT dollars are actually going. McClure heads the office that supports OMB's data center initiative.
Unfortunately for VanRoekel, his collegial style, honed from his days working for Microsoft's Bill Gates, seemed to score him few points in the lawmakers' lair and the inquiry soon turned to Powner.
The GAO director confirmed the lawmakers' suspicions that uncertainty remained about how many government data centers exist as OMB now defines them, especially within the Defense Department, or what kind of savings the government might eventually hope to see by consolidating them.
Powner may be right. But the subcommittee and the American public would have been better served if congressional questioners keyed instead on how Powner would go about improving federal IT and reducing its redundancy. His knowledge of federal IT management issues is extensive. That's reflected in Powner's latest report, which offers a useful digest of GAO's more recent findings. Among them:
-- The federal government is still failing to keep up with private enterprise. Information technology should enable government to better serve the American people. However, despite spending more than $600 billion on IT over the past decade, the federal government has achieved few of the productivity improvements that private industry has realized. That's because so much of that spending goes toward maintaining legacy systems, while the private sector is able to invest in newer, much more cost-effective systems that improve productivity and promote innovation.
-- Agencies have spent billions on failed and poorly performing investments. As of July 2013, 154 of the federal government's approximately 700 major IT investments -- totaling about $10.4 billion -- were still at risk and needed management attention, Powner found. The findings, from OMB's IT Dashboard, suggest that program leaders, not just IT leaders, need to be held more accountable.
-- Federal IT spending continues to be inefficient and duplicative. Powner reiterated the proliferation of data centers -- from 434 in 1998 to 6,836, according to GAO's latest audit last month -- as being symptomatic of how government's vast operations lead to duplicative investments.
Powner credits OMB with having taken some important steps to improving the government's management of IT spending. Here are a few initiatives it has instituted:
The IT Dashboard: A public website that provides financial and performance data on 700 major IT investments at 27 agencies.
TechStat review sessions: Face-to-face meetings with agency leaders to terminate or turnaround IT investments that are failing to produce results.
PortfolioStat: Designed to get agencies to fully assess and determine the value of their IT investments from a business or mission perspective.
FDCCI: The Federal Data Center Consolidative Initiative now aims to rationalize core and non-core data center assets and shut down duplicative operations.
But in the end, the Congressmen demonstrated little interest in getting at the root causes of why government overspends on data centers and IT in general.
If lawmakers are truly intent on helping agencies buy and manage IT more effectively, they would do better by hearing from department and agency CIOs directly. That might make it clearer why some CIOs are better than others at closing data centers. It might also shed needed light on the fact that even when CIOs try to consolidate IT assets, internal political and budget forces can stall those efforts in their tracks.
Just ask Richard Spires, the former CIO at the Department of Homeland Security. Widely regarded as one of government's most accomplished CIOs, Spires resigned in May after it became evident that even the most rational IT business cases and governance groundwork don't win the day. Spires' departure should send an important signal to legislators that the roots of duplicative government IT spending runs deeper than the CIO's office.