Statements on Introduced Bills and Joint Resolutions

Floor Speech

STATEMENTS ON INTRODUCED BILLS AND JOINT RESOLUTIONS -- (Senate - April 28, 2009)

BREAK IN TRANSCRIPT

Ms. COLLINS. Mr. President. I am pleased to join Senator Carper in reintroducing a bill that will improve agency performance and Congressional oversight of major federal information-technology, IT, projects. We introduced this bill last Congress and offer it for consideration again because it will strengthen oversight of technology investments to help prevent the waste and misuse of taxpayer dollars.

The well-publicized cost and performance problems with the Census Bureau's handheld computers for the 2010 Census--with its troubling implications for the next House reapportionment and for the allocation of Federal funds--represent only the most recent and conspicuous failure in a long trail of troubles that also includes critical IT projects like the FBI's Virtual Case File initiative.

The 2010 Census is notable among projects that have drawn our attention, not only because of its great scope and expense, but because of its history of unheeded cautions. For years, warnings of potential dangers came from experts sought out by the Census Bureau itself and from the Commerce Department's own Inspector General.

The implications of this lack of proper planning and oversight are evident in the burgeoning estimate for the life-cycle cost of the 2010 Census. The Bureau initially estimated that the 2010 Census would cost the taxpayers about $11.3 billion dollars; today, the estimated cost is more than $14 billion.

Another example is the Department of Homeland Security's, DHS, efforts since 2004 to integrate its financial management systems. DHS spent approximately $52 million on one failed attempt before abandoning the project nearly two years later. According to GAO, this attempt likely failed because DHS had not developed an overall financial management transformation strategy that included financial management policies and procedures, standard business processes, a human capital strategy, and effective internal controls. DHS spent approximately $52 million and now has little, if anything, to show for it.

The Department of Homeland Security is now attempting another consolidation of its financial information technology systems. It is essential that, this time, the Department sufficiently plan and monitor its cost, schedule, and performance targets.

During the 108th Congress, the Committee on Governmental Affairs investigated the botched automated record-keeping project for the federal employees' Thrift Savings Plan, TSP. This project was terminated in 2001 after a four-year contract produced $36 million in waste that was charged to the accounts of TSP participants and beneficiaries. A second vendor needed an additional $33 million to bring the system online, years overdue and costing more than double its original estimate.

In a 2004 letter from the Federal Retirement Thrift Investment Board to the Governmental Affairs Committee, the Board characterized the project as ``ill-fated,'' and acknowledged the importance of careful planning, task definition, communication, proper personnel, and risk management--all of which were lacking on that project.

Large IT project failures have cost US taxpayers literally billions of dollars in wasted expenditures. Perhaps even more troubling is the fact that when Federal IT projects fail, they can undermine the government's ability to defend the nation, enforce its laws, or deliver critical services to citizens. Again and again, we have seen IT project failures grounded in poor planning, ill-defined and shifting requirements, undisclosed difficulties, poor risk management, and lax monitoring of performance.

Unfortunately, as the Government Accountability Office, GAO, continues to report, Federal IT projects still fall short in their use of effective oversight techniques to monitor development and to spot signs of possible trouble.

The GAO reported that the Federal Government spent over $71 billion in fiscal year 2009 on IT projects. Most of that spending was concentrated in two dozen agencies that have approximately 800 major projects underway.

When the GAO reviewed a random sampling of these major Federal IT projects, they found that 85--nearly half the sample--had been ``rebaselined.'' Eighteen of those projects have been rebaselined three or more times. For example, the Department of Defense Advanced Field Artillery Tactical Data System has been rebaselined four times; a Veterans Affairs Health Administration Center project has been rebaselined 6 times.

Rebaselining can reflect funding changes, revisions in project scope or goals, and other perfectly reasonable project modifications. But as the GAO notes, ``[rebaselining] can also be used to mask cost overruns and schedule delays.'' All major federal agencies have rebaselining policies, but the GAO concludes that they are not comprehensive and that ``none of the policies are fully consistent with best practices.''

The bill that Senator Carper and I are introducing will go far toward addressing the weaknesses identified by the GAO and will reduce the risks that important Federal IT projects will drag on far beyond deadlines, fail to deliver intended capabilities, or waste taxpayers' money.

Our bill will improve both agency and Congressional oversight of large Federal IT projects. For all major investments, the bill requires agencies to track the Earned Value Management index, a key cost and performance measure, and to alert Congress should that measure fall below a defined threshold.

The bill requires additional reports to Congress as well as specific corrective actions should those same indicators continue to worsen. Further, because the bill's performance thresholds are based on original cost baselines, rebaselining can no longer serve as a tactic to hide troubled projects. Where severe shortfalls remain uncorrected, agencies are prohibited from committing additional funds to the project until the required corrective actions are taken.

Our bill would not make Congress a micro-manager of Federal projects--especially in so complex a field as information technology. But it will ensure that, for these important investments, agencies will be required to track key performance metrics, inform Congress of shortfalls in those metrics, and provide Congress with follow-up reports, independent cost estimates, and analyses of project alternatives when the original projects have run off course.

The bill also provides that each covered agency identify to Congress their top mission-critical projects. Those ``core investments'' would be subject to additional upfront planning, reporting, and performance monitoring requirements. This will help ensure that agencies apply extra vigilance to these projects at the planning stage, and not just when execution begins.

In addition to tracking cost and schedule slippage, agencies making core IT investments must provide a complete ``business case'' that outlines the need for the project and its associated costs and schedules; produce a rigorous, independent, third-party estimate of the project's full, life-cycle costs; have the agency CIO certify the project's functional requirements; track these functional requirements; and report to Congress any changes in functional requirements, including whether those changes concealed a major cost increase.

To help agencies deliver IT projects on time and on budget, the bill also provides two new support mechanisms.

First, agency heads would be required to establish an internal IT-management program, subject to OMB guidelines, to improve project planning, requirements development, and management of earned value and risk.

Second, the Director of OMB and its E-Gov Administrator would be required to establish an IT Tiger Team of experts and independent consultants that can be assigned to help agencies reform troubled projects. In addition, the E-Gov Administrator can recommend that agency heads mentor or replace an IT project manager, reinforce the management team, terminate the project, or hire an independent contractor to report on the project.

These and other provisions will help improve project planning, avoid problems in project execution, provide early alerts when problems arise, and promote prompt corrective action.

In projects where difficulties persist, our bill provides strong remedies. For projects that exhibit a performance shortfall of 20 percent or more, the agency head involved must not only alert Congress but also provide a summary of a concrete plan of action to correct the problem. If the shortfall exceeds 40 percent, agencies have six months to take required remedial steps or else suspend further project spending until those steps are completed.

If the provisions of this bill had been in force during the past decade, early indicators of trouble and prompt warnings to Congress might have helped prevent much of the added cost, decreased functionality, and increased anxiety we now see surrounding the handheld computers that were intended to streamline the 2010 Census. The additional scrutiny of plans and costs required by this bill might have saved some of the billions wasted on other IT projects that ultimately landed on high-risk lists.

I urge every Senator to support this much-needed and bipartisan bill.

BREAK IN TRANSCRIPT

Source