It has been 20 years since I began writing the 1994 “Computer Chaos” report for Senator William Cohen. Today, I am struck by how easy it would be to draft a similar Congressional report describing the systemic problems with federal information technology management and acquisition. Thus, it is tempting to conclude that the legislative result of that study—the Clinger-Cohen Act (CCA) of 1996—was a failure. This, as with many public policy reform efforts, is a glass half-full/half-empty exercise. Parts of the Act worked for a time but need to be either re-invigorated or updated. Other aspects never really got off the ground, but still stand on sound principles.
Many positive things came from the CCA, such as expanding commercial acquisition practices, removing the centralized General Services Administration (GSA) bureaucracy that oversaw computer purchases, and improving the bid protest system. These reforms resulted in the government being able to rapidly incorporate commercial information technology (IT) into government operations and military systems.
Shift in the last five years
Unfortunately, many of these positive reforms have been undermined by a risk-averse culture that has crept back into the federal acquisition process over the last five years. Federal contracting is now dominated by the need to comply with federal procedures and regulations and not with performance and results. One only has to look at the Obamacare website debacle and the absence of true commercial contractors with experience developing successful online marketplaces, to see that the acquisition system is again failing.
Making the federal market attractive to Silicon Valley companies that currently refuse or are unable to do business with the federal government because of non-commercial acquisition barriers and requirements should be the goal of successful acquisition reform. It is vital for these dynamic, commercial companies to participate in the federal marketplace so the government can access the best private sector technology and practices. This was a primary goal of the CCA and its predecessor, the Federal Acquisition Streamlining Act of 1994. The commercial acquisition provisions of these two acts need to be strengthened and barriers removed to their effective implementation.
Compared to acquisition reform efforts, the CCA information management provisions embedded in the Paperwork Reduction Act and the Computer Security Act were even less successful. This body of statute and implementing policies should be reviewed and updated where necessary, but execution is probably a bigger problem than the wording of the law. The basic principles that underlined the recommendations of the Computer Chaos report and the CCA can continue to serve as a foundation of future information management reforms.
A look back, a look forward
These principles were at the heart of the original intentions of the Act and may not be obvious to someone reading the legislation 20 years on. They include adopting best practices from the commercial market, changing and eliminating unnecessary government business practices before automating, focusing on incremental improvements and proven commercial solutions over large “grand design” projects, and developing a strong tech-savvy federal workforce and leadership.
The foundational premise of Clinger-Cohen was the need to leverage the commercial market. This required the continuous assessment and adoption of commercial best practices. In the early 1990s, federal information management practices were stuck in a 1960s mainframe-era mentality, where government purchases and technology research dominated the computing industry. The federal government needed to take advantage of advances in commercial technology and embrace the benefits of the private sector’s greater purchasing power, lessons learned and good ideas.
Although GAO established an information management best practices baseline, this included requiring the government to continually do market research and learn from other user’s successes and failures. This continuous review and update cycle unfortunately didn’t happened. Twenty years later, the now globalized $3.7 trillion commercial information technology market has moved even further beyond the government’s ability to reinvent the wheel with its unique practices and limited purchasing power. New best practices, technologies, and approaches await the government if it can change its ways.
The second goal of the CCA was more radical than just making sure the government’s computer systems worked. It was designed to truly revolutionize the way government was performed. Prior to embarking on an IT acquisition, an agency was first required to review and change its underlying business practices. IT acquisition was to be a forcing mechanism to overhaul government transactions with the public. This has proven to be nearly impossible and the government has continued to spend billions of dollars on upgraded hardware and software to essentially “pave the cowpaths” and do the same things it has always done before.
A third goal of the legislation was to address the fact that large-scale, government-unique information projects patterned after large defense systems acquisitions never really worked in practice. The Act tried to put up roadblocks to these large projects by forcing them through a time-intensive capital planning and investment review process. The thought was that most rational agencies would rather not go through this process and instead break projects into smaller pieces, make incremental changes to government unique systems until such time as a commercial product could replace them, and use cheaper commercial products and software that had a higher return on their relative investment.
This did not work. Instead, agencies newly freed from GSA’s overly centralized process, embraced their own inner bureaucracy and continued with grand design, large-scale computer projects while turning the capital planning and review process into a procedural quagmire for most IT investments.
Finally, the most important focus of Clinger-Cohen was on the federal workforce. Chief Information Officers (CIOs) were created to be the focal points for critical skills, information systems development, management, and security. CIOs were meant to be advocates for process change in the government that would be supported by the heads of agencies and most importantly from OMB and the White House. This senior level support did not come to pass and an opportunity to establish an ongoing mechanism for continuous governmental reform was lost. Absent top-level support, the new but relatively powerless CIOs hunkered down and became mired in bureaucracy and increasingly irrelevant.
Still, there is room for optimism. The principles and goals underlying Clinger-Cohen were perhaps a bit ahead of their time and too dependent upon a shared IT vision and culture change that had yet to happen. Today’s mangers and federal workforce, having grown up with the PC and the Internet, are more tech-savvy than their counterparts of twenty years ago and offer hope that they can apply the tools of the information age more effectively than their predecessors.
The politics of agency transformation will always be difficult. However, with declining budgets, the HealthCare.gov failure, and new leaders that have experienced the effects of information revolution in their daily lives, comes an opportunity to re-initiate reform. While order may not always come out of chaos, perhaps a more effective government can be achieved if the government begins to focus on how to better use information technology and carry out the vision of the Cohen reform efforts.
William Greenwalt is a visiting fellow in the Marilyn Ware Center for Security Studies.