Networked Government
It's not about outsourcing vs. bureaucracy. It's about managing diverse webs of relationships to deliver value.


or a century, the predominant organizational model used to deliver public services and fulfill public policy goals has been hierarchical government bureaucracy. A complex society is forcing this model to change. Although the traditional model isn't dead yet, it's steadily giving way to a fundamentally different approach in which government executives redefine their core responsibilities from managing people to coordinating resources for producing public value.

To be sure, the delineation of authority between different levels of government and among the public, private and nonprofit sectors never has been totally clear, but this century's challenges and the means of addressing them have become more numerous and complex than ever before. Nearly all the major public policy issues of the day-reviving urban communities, providing a better education for children, protecting the nation from terrorists, responding to disasters-require activating, nurturing and managing networks of federal, state and local governments, and private businesses, contractors and nonprofit institutions. In these activities, government's role is not necessarily reduced, but rather transformed. Government agencies, bureaus, divisions, units and offices become less important as direct service providers and more important as levers of public value inside the web of multiorganizational, multigovernmental and multisectoral relationships that now constitute modern government.

This "government by network" has become a fixture at every level of government. In Iraq,the U.S. military relied on thousands of contractors to do everything from maintaining computer systems to setting up base camps. In Kansas, a network of nonprofit and for-profit providers delivers all foster care and adoption services. Private contractors, not public employees, now run all the welfare-to-work programs in Milwaukee and soon will operate the entire information technology and communications infrastructure for the Navy. In New Zealand, the country's entire highway infrastructure is designed, built and repaired by private firms.

Third-party government is nothing new. What has changed is the breadth and scale of the trend. Service contracting at the federal level over the past 10 years, for example, has grown by 33 percent at civilian agencies and 14 percent at the Defense Department-even taking into account the huge Defense cutbacks after the Cold War ended.

There's one big problem: the federal government's organizational, management and personnel systems are designed for a hierarchical, not a networked, model of government. Many of the highly publicized federal contracting failures of years past are a direct consequence of trying to use traditional hierarchical controls to manage a more horizontal government. But instead of talking about how to improve government's capacity to manage networked and outsourcing relationships, we spend all our time endlessly debating whether agencies should be doing more contracting or less. Although important in a political sense, this essentially is a side issue to the more fundamental question of what kinds of systems, organizational structures and skill sets are needed to operate a government that increasingly focuses-rather than owns-resources, and purchases-rather than provides-services.


Sometimes networked government fails not because of how a particular venture is managed, but because of what was delegated to the private sector in the first place. All too often, precious little thought is given to what policy goals an agency is trying to accomplish and how they relate to what is contracted out. Instead, agency officials pick up their organizational chart, look for something they're not doing very well, and then get the private sector to do it for them.

But before federal executives think about how they should do something, they need to figure out what they're trying to do in the first place. "The biggest challenge in contracting out in government is determining what it is we want to accomplish," says Deidre Lee, the Defense Department's acquisition chief. "What is our mission? What do we need to accomplish the mission? Oftentimes, government is not clear about all this when we go out to bid."

The government executive, hamstrung by precedent and reinforced by well-intentioned bureaucratic practices, often will find it difficult to step into the larger, more important and more exciting role of conceptualizing new models and solutions. The critical point of departure is the problem definition stage. A government agency shouldn't let its historical processes, organizational chart or existing capabilities-or even the private sector's capabilities, for that matter-dictate what it should contract out. Traditional outsourcing models-inside a narrow procurement box and based mostly on what's not working well-just shift your problems to someone else, rather than using outsourcing as a lever to create a new solution and transform existing operations.

"In the private sector there's the notion of 'design space,' which refers to the space needed to design to an outcome, rather than a preconceived notion of what it should look like," explains Stan Soloway, President of the Professional Services Council. "When I was at DoD, we spent a day at Federal Express. The key point hammered into us was that the key to success is recognizing that before you outsource, you need to completely reassess everything you're doing today, do a process map, and then get rid of anything you're now doing that doesn't plug into your new model."

This was the approach the Coast Guard used when it went out to bid several years ago for the modernization of its fleet of aircraft and ships that patrol at least 50 miles from shore. The fleet, including 90 ships and 200 aircraft, was old, falling apart and unsuited to present conditions. The standard way of replacing the $10 billion fleet would have been to purchase each plane, boat and piece of technology separately as they wore out and only later figure out how to put them all together. The inevitable result would have been higher prices than would have been paid through bulk purchasing, and multiple solutions and platforms that weren't integrated.

Determined to avoid these problems, the Coast Guard tried a much different model: In a project called Deepwater, it contracted to replace its entire inventory as an integrated package over a multiyear time frame. The agency challenged bidders to help increase mission effectiveness through better and newer technologies and new ways of operating. The Coast Guard assigned internal teams made up of specialists from various areas to each of the three final bidders to help them understand exactly what the agency wanted to achieve. The request for proposals spelled out the agency's desired outcomes and the capabilities it needed-search and rescue, identifying someone adrift in the ocean, providing surge capacity to meet national security and disaster response requirements. Then the Coast Guard left it up to the vendors to design systems of boats, ships, aircraft, satellites, information technology and unmanned aerial vehicles that met the criteria. The ultimate goal: to revolutionize the way every man and woman in the Coast Guard does his or her job.

The Coast Guard spent several years coming up with this approach. The lesson is clear: The success or failure of a networked government model can often be traced back to how it was originally designed.


The Coast Guard Deepwater project, along with the Navy-Marine Corps Intranet (NMCI) and the National Security Agency's Groundbreaker-two huge IT outsourcing projects-all reflect the recognition that sometimes the reason for going out to the market is that the government concludes that the private sector can integrate and manage a set of service delivery and infrastructure components better than an agency can itself.

Particularly in cases where the government is trying to move away from a narrow, stovepiped model of service delivery, there is often an inescapable logic-if not an imperative-to integrate services under large contracts. Proper integration may even involve meshing the activities of a large number of entities-some of them other agencies and other levels of government. In these and other cases, breaking up the pieces into smaller contracts could cause serious-if not disastrous-operational problems down the road.

The State of Kansas found this out when it privatized its child welfare system in 1996. The Kansas Department of Social and Rehabilitative Services (DSRS) divided the state into five regions and then put out bids for family preservation and foster care separately in each region (the adoption process was bid out in one statewide contract). "We divided it the way we did by region and service because we wanted homegrown Kansas providers, rather than the 'big, bad managed care providers from the East' to provide the services," explains Teresa Markowitz, the former DSRS commissioner who spearheaded the privatization effort. This political goal was achieved, but the flip side was that a single child might be shuffled from one provider network to another several times, depending on where her case was classified at any one time along the child welfare continuum. Continuity of care was wholly dependent on tight coordination among different providers working across contracts and across service areas. Not surprisingly, such collaboration was more the exception than the norm. The end result: The state's effort to "unbundle" the contracts inadvertently undermined one of its main policy goals: integrated service delivery.

All of this is not to say that purchasing an integrated solution is always the right way to go, or that small businesses' objections to bundling service contracts should be ignored. However, it's important to understand that today's complex problems often require carefully integrated solutions. In certain instances government can act as its own general contactor, but that role requires the federal executive to think creatively across product lines and agencies, build an intergovernmental network before the procurement process starts, and find internal management talent that can creatively configure the best possible solution. When the capacity to do this is absent, executives must recognize that the ability of the private sector to properly integrate the parties into a solution might, in fact, be the most important asset to be procured.


Executives who craft and manage such contracts are pinned between conflicting challenges. On the one hand, they want to prevent a contractor from bidding low only to modify the contract terms later to increase profits. At the same time, the most valuable relationships are dynamic, learning relationships. With a good vendor and government manager, the goals and outcomes of the contract will stay sharply in focus, but the inputs and processes will change as required. Strict adherence to the contract terms is a sure way to leave value on the table. Managing in a dynamic way while protecting the taxpayers-and staying out of the crosshairs of the inspector general, GAO and Congress-might be the ultimate challenge of networked government.

The dilemma is to retain as much flexibility as possible in the relationship, but to do so in a way that keeps the vendor honest with consistent standards and outcomes. Balancing this tension between accountability and flexibility requires "both parties to understand at the conceptual level the goal that has to be reached," says Rear Adm. Charles Munns, the director of the $8 billion NMCI program, the largest government outsourcing project in the world. "Once you have that shared understanding, you can recognize some of the weaknesses and ambiguities in the contract, put each one on the table and talk through it."

Munns and his team have come up with a novel way to get NMCI's prime contractor, EDS, to increase performance and make substantial modifications in service without changing a single line of the contract. Clauses in the contract provide EDS with a generous performance bonus for achieving a high level of customer satisfaction, giving the Navy and the Marine Corps the leverage they need to suggest changes they believe will help boost customer satisfaction numbers. EDS, in turn, has a strong financial incentive to fix the problems. "We understand the Navy, they understand service delivery and the network infrastructure," says Munns. "Together we try to figure out how to boost the customer satisfaction numbers without ever having to get the lawyers involved."

The Coast Guard Deepwater project was the brainchild of its former leader, Adm. James Loy, and is now led by Patrick Stillman, a two-star admiral. Such senior-level leadership of a contract relationship has become standard in the private sector, where responsibility for managing and nurturing alliance partnerships, joint ventures and outsourcing relationships often rests at the top. However, it's rare in the federal government, where executive attention typically is more focused on political issues, public affairs and putting out fires, leaving little time for supervising and fostering partnership arrangements.

"In government, the people doing the work managing the contractors are not the people that really have the skin in the game," says Jack Brock, managing director of sourcing and acquisition issues at GAO. "It's been pushed down the organization and doesn't get the attention it deserves." Even at agencies such as Energy and NASA, both of which have become de facto contract management agencies because they outsource so much of their work, there's a disconnect between senior agency leaders and the contract administrators, says Brock.

For years, the Internal Revenue Service was no different, with one failed multimillion-dollar contract after another. But then in 1997, the IRS got its wake-up call in the form of highly publicized congressional hearings taking the agency to task for a host of problems, including dreadful customer service and archaic information systems. Meeting many of the new goals set by Congress was dependent on the success of its multibillion-dollar modernization project, managed by a consortium with Computer Services Corp. (CSC) at the helm. Failure would mean another round of nasty hearings and a total loss of confidence in the agency. This wasn't a relationship that could be relegated to the procurement shop.

"Driving a multiyear, business transformation outsourcing relationship across a complex organization requires senior-level commitment," says IRS Deputy Commissioner Dave Mader. The agency set up a management structure in which IRS executives are lined up with CSC executives in joint integrated process teams for all of the agency's modernization projects. This approach reaches up to the highest levels of the agency. Fred Forman, the deputy commissioner who leads the modernization effort, has a full-time counterpart at CSC whose office is adjacent to his own.


Leadership at the highest levels allows agencies to solve legal and functional issues more quickly. Also, it serves to bring together the many lower-level program officers who possess important skills and information, but who individually can't easily collaborate across impermeable boundaries. For these middle managers, the day-to-day business of working in networks is infinitely more complex and more difficult than managing a traditional bureaucracy. It requires a whole different set of skills. In addition to knowing about planning, budgeting, staffing and other traditional government duties, networked management requires proficiency in a host of new tasks, such as business process reengineering, negotiation, mediation and network design.

Unfortunately, such skills aren't exactly plentiful in the federal government, nor are they typically recognized or rewarded. The way to get ahead in the federal government has been to be an adviser on policy issues or demonstrate a solid ability to manage government employees, not to show proficiency in negotiating deals and managing third-party service providers. As a result, some agencies don't even have effective contract management capabilities, much less the capacity to handle the vastly more sophisticated requirements of network management. "When you look at sourcing issues, government's need for access to services has grown faster than its management capacity to control that they're getting what they want," explains GAO's Brock.

Building such a capacity requires not only far-reaching training and recruitment strategies, but also a full-blown cultural transformation. What is required is nothing less than changing the definition of what it means to be a public employee. Contracting and relationship skills can no longer be just the province of acquisition employees. People with these skills-skills that currently are not highly valued in government-need to be recruited, rewarded and promoted.

One model is in Great Britain, where the central government created an organization called Partnerships UK as part of a broader effort to improve its public-private partnerships capacity. Partnerships UK helps agencies become smarter purchasers of services by standardizing contracts, providing help desk support, highlighting best practices, and rotating employees in and out of agencies for up to six months at a time. Employees of Partnerships UK come from a variety of backgrounds, including investment banking, law, management consulting and engineering. They all have commercial experience in managing projects. This fills a major void in the central government. "The civil service career strategy doesn't lend itself to developing commercial deal capabilities," says Helen Dell of Partnerships UK.

The workforce also needs to update its skills. Traditionally, the most important thing was to know the rules and then to follow them."The old days of the stereotypical . . . acquisition worker are at an end," says Joseph Johnson, director for administration and services at the Defense Acquisition University. "The worker of the future cannot be just rule-bound. Acquisition is no longer about managing supplies; it's about managing suppliers."

As defense acquisition experts are learning, the government landscape has changed. Many of the traditional assumptions about outsourcing and public-private partnerships no longer hold sway. Unfortunately, you wouldn't know this from observing the contemporary debate inside the Beltway. The reflexive opposition on the political left to all things outsourced, and the failure of those on the right to acknowledge that far too many contracting endeavors fail to measure up to expectations, are symptomatic of a stale debate that's still stuck in a 1980s ideological box. To succeed in an age of networked government, we not only need to update our approach to government, but also our thinking.

William D. Eggers is the global director for Deloitte Research-Public Sector, the research arm of Deloitte Consulting, and a senior fellow at the Manhattan Institute for Policy Research. Stephen Goldsmith is a professor at the John F. Kennedy School of Government, where he chairs the Innovations in American Government program, and a senior vice president of Affiliated Computer Services Inc. Goldsmith and Eggers advised George W. Bush on federal management issues during his presidential campaign.