Been There, Done That

ud Young liked what he heard from the new administrator of the Infrastructure Modernization Agency. Ed Alvarez sounded sharp, focused, motivated and down to earth. He seemed to have more than a superficial familiarity with half a dozen issues confronting the agency. Indeed, once again, IMA had been blessed with a presidential appointee who had some technical background relevant to the agency's mission.

It was the last issue the new administrator brought up at a get-acquainted meeting that gave Bud Young, IMA's veteran career deputy, his only trace of heartburn.

"The President told me I have a free hand on organizing the agency, but stressed that he has found decentralization to be most helpful in making operations more effective and economical," Alvarez told his deputy. The Vice President had reinforced the subtle guidance with glowing success stories about agency decentralizations, and the White House chief of staff had practically convinced Alvarez he would be graded on doing the same at IMA.

"What kind of potential for decentralization do you think we have here, Bud?" Alvarez said. Young almost blurted out, "About the same as we had in 1990, 1981 and 1973, when we tried it to our ultimate grief," but the question was a sincere one from a responsible official who had, for all intents and purposes, been given a mandate.

"You can always do a little more to assure that your capable people in the field have the authority they need to get their job done, Ed," the deputy said. "But we've spent a lot of time over the 26 years I've been here trying to achieve the ideal balance between field operations and Washington." The deputy said he would get Alvarez more information on past organizational changes before they started down that road again.

"And there's no doubt we'll be going down that road again," Young thought to himself as their meeting broke up. Decentralization is the wave of the present. People are more motivated when they feel largely responsible for what they produce, and work gets done faster and cheaper with less staff when it doesn't get hung up in the chain of command. But Congress and the White House, not to mention the press and public, hold the agency leaders in Washington accountable for progress and problems. And the operating level, no matter how competent, benefits from information and guidance from interagency, national and international connections. You can't get that with nothing more than a few note-takers in Washington.

Young hadn't always felt this way. As a young engineer, even as a mid-level manager in the field, he had chafed under headquarters guidelines and waited impatiently for the central office to make decisions. But once in Washington, he understood the need for central coordination of a large organization whose elements had limited knowledge of each other's requirements and problems.

Nevertheless, Young still favored delegation of authority to the operating level. As deputy administrator, he had been a leading advocate of the 1990 streamlining which had, among other innovations, eliminated routine headquarters review of certain projects. Tragically, one of those projects ended barely a year later when three employees were killed during a pilot test. A congressional investigation and an independent panel both cited "inadequate headquarters oversight" as a contributing factor and recommended corrective action. The 1992 reorganization addressed the apparent management weakness with more central controls.

Armed with 20 years worth of organization charts and documentation, Young gave Alvarez a briefing the next day. When he finished, the administrator leaned back and said, "We both know there are pluses and minuses with any organizational mode, Bud. But loosening central controls is the name of the game now, and I won't have any credibility with the White House or the Hill if I don't move in that direction." He told Young to use his institutional smarts to identify which powers could be safely delegated and report back in 10 days with a plan.

"You're trusted by the workforce to do what's right for the agency," Alvarez said. "I'll give you the people and resources you need to get it done right."

If there was a way to get it done right, Young reflected, the agency would have done it in 1990 or 1981 or whenever. But he had a new opportunity and he started to think.

DONALD KETTL:
Don't Look Back
Professor Donald F. Kettl is director of the Robert M. LaFollette Institute of Public Affairs at the University of Wisconsin-Madison. He is a Nonresident Senior Fellow in the Brookings Institution's Center for Public Management in Washington, and author of and contributor to works including Civil Service Reform: Building a Government that Works; Inside the Reinvention Machine: Appraising the National Performance Review; Sharing Power: Public Governance and Private Markets; and Deficit Politics.

f course there is no way to do it right. There is never an ideal balance between centralization and decentralization, and no balance that is stable for long.

Even though IMA has found a comfort level after its recent bouts of reorganization, there is no reason to presume this level is the best balance for the agency. Still, the three previous decentralization battles, and the loss of three lives, have cast a long shadow over both IMA's structure and Young's thinking. Problems, like the three tragic deaths and the other problems surrounding the earlier "reforms," teach that the only good answer to new pressures is to snug in the reins tightly enough to prevent bad things from recurring and to insulate the agency from new threats and instability.

These hard lessons will surely tempt Young to play out just enough decentralization to satisfy outside forces without doing anything real. If the agency's workforce trusts him, it is because workers count on him not to put them at risk. IMA has already suffered the twin risks of reorganization-induced instability and political attack. Young knows enough to produce the right paper trail to protect his colleagues from new risks while satisfying Ed Alvarez and other reform-minded political forces-at least long enough to allow the press of new business to make them forget their reform enthusiasm.

That, however, would be doubly irresponsible. Young has a clear and inescapable mandate to carry out the policies that Alvarez frames. Moreover, there is an even bigger risk in digging in to protect the routine: able performance of suboptimal work. The biggest threat in American bureaucracy today is not substandard performance, bureaucratic laziness or outright intransigence. It is, rather, merely competent work through routines designed to protect government workers from outside threats and pressures. Too often, the work gets done, but not well enough, quickly enough or cheaply enough to satisfy either political realities or the new challenges of society's complexity. The best way to ensure a new and greater disaster is for Young to set his course by looking over his shoulder at the last three reorganizations instead of into the future at problems IMA will have to solve.

So what should Young do? The key to satisfying Alvarez, defending his colleagues, serving the IMA mission and keeping the agency on the cutting edge is to avoid jumping one more time onto the centralization/decentralization pendulum. Young should launch a careful study that looks at how best to do what IMA needs to do. Any such study will reveal some tasks (say, policy planning and financial management) that can be best done at headquarters. The performance of other tasks might be improved by delegating more responsibility to field offices where employees know more about program management than headquarters staff.

Young, furthermore, ought to ask anyone pressing the case for centralized tasks to explain why the jobs could not be better done in the field. Even more important, as decisions are pushed into the field, Young ought to insist on a balance of accountability: more discretion in exchange for far better reporting of results. Improved performance, not reorganization, ought to drive the effort; improved reporting of outputs and outcomes, not shifting the boxes, ought to be the mechanism of reform.

Neither centralization nor decentralization ought to be ends in themselves. The key is to match the approach to the job, and then to ensure that the job is done well. Such an approach might come hard to a manager like Young whose 26 years of experience argues that no new reform will ever solve the problem but will surely create many new ones. To approach Alvarez' request this way, though, risks plunging his agency into mediocrity. Charting his way past this trap, around the political risks facing his agency and through the unpredictable complexities, is the challenge Young and thousands of top managers like him must learn anew to conquer.

ALAN DEAN and
DWIGHT INK:
Bottom-Up Review
Alan Dean was the Transportation Department's first assistant secretary for administration. He has also been associate administrator of the Federal Aviation Administration and vice president for administration of the U.S. Railway Association. As a deputy assistant director of the Office of Management and Budget, he was coordinator of President Nixon's Departmental Reorganization Program.

Dwight Ink served in federal policy positions under seven presidents. While in the Executive Office of the President, he was responsible for governmentwide management and organization reform, including the plans that established the Environmental Protection Agency and the Office of Management and Budget, and that made the U.S. Postal Service a government corporation. Ink designed and coordinated the massive decentralization of the domestic agencies in the 1970s, called the New Federalism.

mproved service to the public, reduced administrative costs, smaller staff requirements and faster action often result from delegating many of an agency's activities to field offices. Delegation also frees headquarters from the burden of day-to-day operations and enables it to concentrate on necessary headquarters functions such as policy formulation and oversight of operations. However, it is important that decentralization be designed and managed to produce the maximum benefit for the people and institutions served by the agency. Otherwise, the effort will fail, and may produce the "ultimate grief" experienced by the IMA.

Before proceeding with decentralization, Young must determine which activities can be advantageously decentralized, and which cannot. Policy determination is clearly a headquarters function. So are resource allocation, White House and congressional negotiations, monitoring and evaluation of field operations, and the operations most effectively carried out in a single unit (e.g. payroll preparation, assembly of agencywide statistical data).

In the Federal Aviation Administration's highly successful decentralization, every activity was examined and the authority to administer policies was delegated to the regional directors, except when convincing evidence showed a function could best be performed at headquarters.

Delegation of authority must be clear-cut, with operational accountability spelled out. Authority should not be delegated to the field, however, until there are policies and standards to guide those actions. In fact, a new statute may require centralized administration for a short time until the agency has developed and communicated implementation policies, standards and regulations. Failure to provide such guidance can result in serious missteps by field officials and inequitable treatment of an agency's clients.

Successful decentralization depends on qualified staff and facilities. There must be enough people in the field, and they must have sufficient training and high enough grade levels to exercise the new responsibilities. Field officials must have the rank and status needed to hold their own with headquarters officials and to be effective in working with agency clients such as state and local officials and business leaders. Because decentralization generally cuts red tape, it often permits a staffing reduction.

A senior official should lead the design and implementation of the decentralization. Career leaders must be enlisted for key roles. The hard work of career personnel years ago in closing down the government's anti-poverty agency, the Office of Economic Opportunity-an action they all opposed-shows most career personnel can be counted on to carry out the policies of the president and an agency head they respect, regardless of their personal views.

The change leader must have a support staff-an existing organization, an ad hoc group established only for the project, or a combination of the two. Task forces with members from both headquarters and the field should include employees from parts of the agency that would be most affected by the changes, and should not be limited to the highest grades. A management analysis unit can support the task forces.

The task force or staff report should include pros and cons. It should also indicate likely costs and organizational and staff changes. Schedules for implementation should be included. The report should be internally reviewed before being presented to top managers.

Useful as these temporary mechanisms are, they are not a substitute for a permanent office which provides ongoing leadership in initiating and sustaining reforms. This role used to be carried out in most departments by the assistant secretary for administration and by similar officials in most large agencies. With the weakening of these offices in recent years and the fragmentation of agency management responsibilities, it is difficult to find a suitable champion of reforms other than the deputy head of an agency.

Whatever the mechanism, the process produces internal and external repercussions. Employees, unions, interest groups, the Office of Management and Budget and congressional committees need to be informed and consulted to maximize support and minimize opposition.

The agency head remains accountable for field performance. Therefore, he or she must assure that any deficiencies are promptly identified through field visits, systematic audits, installation of information systems designed to disclose emerging problems, and regular evaluations of field operations. Headquarters must be able to quickly correct problems before they grow into public issues or scandals.

No reform is successful without the intensive involvement of agency leaders and sustained staff support. An examination of the earlier unsuccessful IMA attempts would almost certainly reveal that one or more of the above requirements for decentralization had been overlooked or badly handled.

FRANK L. DAVIS:
Consult Employees
Frank Davis is director of the HUD Field Reorganization Task Force. He is a career senior executive with 29 years of experience at HUD and GSA in a broad range of professional and managerial positions, including personnel management, labor relations, training, community planning and development, and field/regional management.

he trend to reverse the growth and authority of the federal government has been evident since the election of President Nixon and has continued through the Clinton Administration. The difference in the current movement to "right-size" the federal government is the extraordinary degree of career civil service and customer involvement in the reinvention process. The result has been a request for more and better customer service, including decision-making at the lowest possible and practicable level in the agency.

There are two conundrums which figure prominently in the organizational change process, particularly with respect to decentralized authority. First, there is an almost irresistible urge at the middle and top headquarters management levels to aggregate and maintain authority at that level. Headquarters managers feel that to decentralize strips away some aspect of a personal power preserve. Second, when forced to decentralize, these headquarters managers have an eerie sense of being programmed for the failure of the recipients of delegated authority. This "sense" manifests itself either in a paucity of training and guidance on how to exercise the new authority or in the "you had better not do anything until you clear it with me" attitude which gets conveyed with the empowerment.

Finally, sometimes one gets the feeling that organizational change initiatives have a half-life of 24 months, which is about the average tenure of political appointees. Subconsciously, career staff sometimes govern their energies and enthusiasm to last just about that long. Thereafter, the project or effort falls into disrepute and is disregarded.

The flip side of this cynicism occurs when people are convinced there is a legitimate and compelling business need to change; that noncareer and career executives are equally committed; and that there will be sincere and real opportunities for employees to provide input and to be informed and involved in both the planning and implementation processes.

Senior career managers, such as Bud Young, can and should be responsible for planning and managing the decentralization process. But all employees should be informed of the decentralization study and encouraged to offer suggestions and recommendations. A core management group, consisting of representatives from headquarters and the field, as well as union officials, should develop a charter for the organizational change study and plan. The charter should be a document which can be openly discussed internally and externally. A clearly defined communications strategy must be established, with people designated to address the inevitable inquiries.

Further, the charter should specify the goals, objectives, tasks and milestones of the study. There should be a clear process for participants to review and comment on the study, as well as the decision-making and implementation.

Technology is hardly mentioned in Bud Young's dilemma, and our experience with transformation indicates that the location of our work-particularly delivery of programs and services-is becoming less important. Streamlining, downsizing and consolidating may actually result in both decentralizing authority from Washington as well as centralizing authority in several major field office locations, which will become centers of customer service.