Point of No Return

As agencies seek newer technologies, the question is how much IT spending is too much.

Congress must choose between limiting the federal information technology budget to reduce the deficit and boosting investments to yield greater returns in the future. President Obama has urged replacing 20-year-old agency IT systems with newer technologies as well as outsourcing operations to Web-based providers. Both approaches could pay off in better, less-expensive citizen services in the long term, but they are costly in the short term. And Obama has proposed a three-year governmentwide freeze on a substantial amount of nonsecurity, discretionary IT funding in an effort to skim off waste.

The current IT budget is about $75 billion-and that's not counting the likely billions spent on classified programs. Nearly 10 years ago, when agencies were spending roughly $30 billion less on technology, Mark Forman, the first person to head the federal Office of E-Government and Information Technology, questioned whether even that figure was too high. Today, his outlook is similar. "Eventually the IT budget should be headed down, and the value of that money should be headed up," he says. Increasing that return on investment is the real challenge congressional appropriators face. That will require stepping away from numbers and zooming in on long-term agency performance goals, say other former IT decision-makers.

Lawmakers should demand a cost-benefit analysis that ensures some kind of yield in the future, says Tim Young, former deputy administrator for e-government and information technology. The Obama administration has more information than ever before to support such analysis, adds Forman, now a partner at federal advisory services firm KPMG. The Office of Management and Budget has instituted the IT Dashboard, a Web site for tracking tech spending, where agencies must report monthly on the cost and schedule of major IT investments. In justifying IT expenditures, "I hope that with the things like the IT Dashboard we start to look at 'is this the best way to close that performance gap,' " Forman says.

In addition, agencies and appropriators need to understand that technology isn't always the panacea for the nation's ailments, observers say. The debate about how to curb illegal immigration is one situation in which employee eligibility verification systems might not be the only answer, Young notes. "Certainly technology can enable mitigating illegal immigration across borders," says Young, now a senior manager with Deloitte Consulting's federal government services group. "But I don't view technology as being able to solely solve the immigration problem; it takes policy and regulation, and other kinds of carrots and sticks."

Other former federal officials say Congress should take a cue from the private sector and calculate an IT budget that reflects an organization's deficiencies and overall strategy, rather than a percentage of revenue. "Corporations make annual investments in the known and future technology needs of the organization because this is central to staying competitive," notes Kevin Novak, former director of Web services at the Library of Congress. "Similarly, government should invest in upgrading technology infrastructure. . . . In addition, there are efficiencies to be gained by investing in the latest technology," says Novak, now vice president of integrated Web strategy and technology for the American Institute of Architects. The executive branch already has embraced this philosophy. Obama, acknowledging that the federal government is not as technologically astute as the corporate world, wants agencies to ramp up productivity by looking to the tools that have worked for the private sector.

Young sees room to emulate the private sector in pay practices and procurement, areas noted for wasteful federal IT spending. The government should tie compensation and bonuses to program performance, a common corporate sector practice, he says. Shared services, or outsourcing IT work to an office that serves multiple agencies, should be a mandatory consideration in agency budget justifications, he adds. The government has invested hundreds of millions of dollars in shared services, and yet there is still a reluctance to embrace their economies of scale.

Studies seem to back the goal-oriented method of picking budget numbers over the deficit-reduction approach. Andrew Bartels, a research analyst at Forrester who specializes in IT spending and budgeting trends, says there are three issues Congress should consider when determining budgets. One is opportunities to advance an agency's mission by using technology. For example, regulatory agencies, such as the Federal Trade Commission and Federal Aviation Administration, could benefit from money for sophisticated collaboration tools that let them better interact with citizens. The other two factors to assess: Which IT purchases will help stimulate the tech sector? And what investments will improve government efficiency in the long term so when the time comes to cut funding the investments already are yielding dividends?

But Bartels says the first priority should be resuscitating the economy, which means boosting IT spending. "Right now, it's critical the government be spending-to grow the economy," he says. Bartels notes the federal government accounts for 40 percent of state, local and national government IT spending. One valuable investment, he adds, would be spending to retain younger IT employees. "The biggest crisis facing the federal government right now is going to be one of skills . . . in the face of impending waves of retirement of existing IT staff," he says.

IT expenditures are called investments for a reason, stresses Young. "That is because they are expected to yield, at some point in the future, a return," he says. According to Young, when it comes to capital planning, the executive branch and Congress should ask: What does this agency need to accomplish? What do we need to accomplish these outcomes-policies, laws, executive oversight, technology or a combination? Do we need to build that technology? Or can we borrow it or buy it from somewhere else and how much does that cost? "Dollar figures are important, but they shouldn't be the first step," Young says.