Back in 2000, a team in the Virginia Department of Human Resource Management was nearing the end of its work on a new compensation plan for state employees--one that officials hoped would bring the state's pay practices in line with the needs of a new century. It was complicated work, and team members were pretty happy with it. But they had one more step to take before they could consider the job done: Ask Bob.
"Bob" is Robert Weaver, a veteran compensation analyst for the state. With more than two decades on the job, he has honed a talent for seeing what others miss--the unaddressed problem, the potential loophole, the unanticipated consequence. This time was no exception. He looked over the proposal and quickly spotted provisions that wouldn't work. The team revised the plan. What might have been an embarrassing or troublesome program turned into a policy success.
Once again, Bob got credit for an important save. "In just a few minutes, he can tell us whether we're on the right track." says Ruyenne White, a human resources management consultant in the department, "or show why the actual outcome would be totally the opposite of what we want to achieve."
Every successful organization has at least a few Bob Weavers-- employees who, thanks in part to their own talent and commitment but also to years of experience, know how to get things done. But Weaver's generation of public officials--the baby boomers--are now approaching retirement and hence may soon disappear from the ranks. A growing number of state and local officials are starting to grapple with the potentially unprecedented loss of institutional memory that may ensue. In figuring out what knowledge to save and how to save it, they are giving us a new understanding of what makes government bureaucracies work, both now and, they hope, in the future.
THE UNSPOKEN WORD
With the first baby boomers due to reach 60 this year, the challenge of holding on to institutional memory is taking on a new sense of urgency. The Conference Board, a business research group, estimates that by 2010, 64 million workers--40 percent of the nation's public and private workforce--will have reached retirement age. Just how soon they actually will retire is the subject of conjecture. Some experts believe a variety of factors, from the baby boomers' relatively good physical health to their less certain financial condition, will keep many in the workforce for years to come. But sooner or later, they will leave the workforce.
When they do, the public sector may be hit particularly hard. In a series of self-inflicted wounds, government has magnified the age imbalance of its workforce. In 2002, 46.3 percent of government workers were 45 or older, compared with 31.2 percent of private-sector workers, according to the Rockefeller Institute of Government. Government aggravated the age imbalance in its workforce by swinging from the expansive 1960s and '70s, when today's seniors were hired in record numbers, to the 1980s and '90s, when tight budgets and reductions-in-force clamped down on new hiring. In addition, privatization, as well as many early retirement programs that were put together to achieve short-term savings, thinned the ranks of middle management, reducing further the number of seasoned officials available to replace aging baby-boom managers. And state and local governments haven't fared particularly well in the competition to attract new talent.
Despite clear warning signs, the public sector has been slow to try to fix the problem. Many managers, themselves close to retirement, either don't believe there is a problem, see it as something they can leave to their successors or are reluctant to admit that their agencies are becoming less competent as they lose seasoned manpower. David DeLong, a research fellow at AgeLab, a Massachusetts Institute of Technology facility, notes that even leaders who are motivated to act can be stymied by practical difficulties. "Managers are faced with daily pressures of meeting performance and budget objectives," he says, "so it's hard to get resources to provide for future workforce development."
That is starting to change, though. "A few years ago, I had to persuade people that the aging workforce was an issue, but I don't have to do that anymore," says Mary Young, a researcher for the Center for Organizational Research, a Lexington, Massachusetts, consulting firm. In a study prepared for CPS Human Resources, a California-based public agency, Young argues that policies on recruitment, hiring, training and development, promotion and career mobility, knowledge management, program evaluation and more all need to change for government to get enough qualified people in the pipeline to replace departing baby boomers.
It's hard to pick priorities from that lengthy agenda. For people concerned about institutional memory, however, the top concern is knowledge management, which is defined as the systematic collection, analysis and dissemination of information and insight in an organization. While knowledge managers sometimes deal with relatively prosaic issues, such as creating manuals explaining how routine functions are performed, what keeps them awake at night is the possible loss of "tacit knowledge"--the higher-level understanding that enables institutions to go beyond routine, procedural activities to anticipating and solving problems. Tacit knowledge is to its counterpart, "explicit knowledge," what wisdom is to information. By definition, it is hard to capture in manuals or even put into words. It comes largely from experience, and it cannot be learned quickly.
Although tacit knowledge may be hard to define, most people know it when they see it. In Virginia's Department of Human Resources Management, people beat a constant path to Bob Weaver's door seeking advice and information--or just to have him check out something they have produced. Ruyenne White, one of his many admirers, says brain power is one reason for his success. Weaver has a deep understanding of employment law; he is the department's expert in Fair Labor Standards Act regulations, for instance, and he is a skilled computer programmer who regularly develops programs that his colleagues rely on to collect and analyze data. But his experience may be as valuable as his talent. He has conducted the state's periodic salary surveys since the 1980s and presented his findings annually to the state legislature. Lawmakers know him well and have assigned him more special studies than any other member of the department. His exposure to different viewpoints and his countless opportunities to see how government policies play out in the real world have given him a rare ability to perceive the kind of unintended consequences that frequently foil policy makers.
How does an institution hold on to that kind of knowledge after Weaver retires? Some try to collect "lessons learned" from people like him, cataloging them either in printed manuals or in searchable electronic databases. But it's unclear how well such systems work. Busy employees may not have time to read through a long list of past case studies. If they are seeking a solution to a brand-new problem-- the kind of issue that people look to Weaver to address--they may not find ones relevant to their concerns. Indeed, too much of a focus on previous experiences could make a government less flexible and effective. "Our environment is changing so rapidly that I'm not sure I want to pass on a lot of institutional knowledge," says Connie Phillips, senior human resources development specialist for Anaheim, California. "Institutional memory is about what we have done in the past. It's almost irrelevant."
Another favorite strategy is training. But it, too, is subject to questions about relevance. Skeptics argue that most learning goes on in real life, not in classrooms. That view has led the Georgia Merit Systems, the central personnel agency for that state, to emphasize another approach. It urges agencies to give promising employees accelerated opportunities to broaden their experience through job- shadowing, mentoring, job rotations and various special assignments. For practical reasons, however, such strategies almost always are limited in scope. "There's a great tendency to use the peanut butter theory--to spread everything equally and develop everybody," says Charles Brooks, the Georgia agency's human resources strategy manager. "We don't have the time and resources to do that. There are only so many sweet assignments you can give, only so many task forces, leadership slots and mentors."
A relatively new approach seeks to understand and influence the informal networks that sustain many organizations. It is known as "social network analysis" or "knowledge mapping" and is built on two basic observations about how organizations work. First, most tacit knowledge is transmitted informally from person to person, usually through personal contacts. Second, the most valuable people in organizations aren't necessarily the ones who know a lot themselves but rather are those who know whom to contact for information when it's needed. It's not a new insight; how often have you heard the saying, "It's not what you know, but whom you know?"
A pilot project by the Canadian federal government's environmental agency, Public Works and Government Services Canada, shows how social network analysis operates. In an effort to get ahead of an expected wave of retirements, officials e-mailed employees in the agency's Information Technology division one question each day for eight days: They asked what other people the employee relied on for help with different issues. Answers were compiled instantly and presented in diagrams that showed patterns of interaction within the organization. It quickly became evident who were the key "nodes"--people whose help was sought most often. By color-coding these individuals according to whether they were due to retire, the agency was able to pinpoint specific individuals who, when they retired, would leave the agency vulnerable to losing vital knowledge.
The results were enlightening, according to Marjorie Cooper, the staffer who ran the project. In some areas, expertise was widely dispersed. Because there was no excessive reliance on any individuals, the agency seemed relatively invulnerable to loss of vital knowledge due to retirements. In other areas, though, workers relied heavily on one or two individuals who either were experts or had access to experts whom other employees couldn't reach. Because loss of these key "go-to" people could severely harm the department's operations, Cooper, who described the project at a 2003 conference in Ottawa sponsored by Federated Press, an independent Canadian research organization, said it was clear that management would have to take steps to ensure that someone with comparable skills, personality and knowledge would be available after the key person retired. In the long run, it was hoped, it would find other ways to ensure that expertise was more widely dispersed.
While knowledge mapping is a powerful tool, it isn't without risks. In the Canadian experiment, privacy advocates saw it as an intrusion on employees' space--an issue that Cooper said could be addressed by creating safeguards. Somewhat surprisingly, the research also showed that a number of employees played essential functions that had little to do with their official duties. Some of these individuals feared that their bosses would punish them for not adhering to their formal job descriptions. Managers also seemed cool to the idea, perhaps because it showed that they weren't always in control. As Cooper put it, "information networks don't look the same as our organization charts." Whatever the reason, only two of 11 managers even attended a briefing on the Canadian project, and the agency as yet hasn't followed up on it.
UP CLOSE AND PERSONAL
In the United States, a number of private companies have used social network analysis, but few are willing to talk about it much--mainly because companies see effective use of knowledge as a trade secret, an important factor in determining a firm's competitive advantage. If the idea catches on in the public sector, credit will go to a new breed of government employee, such as the Virginia Department of Transportation's Maureen Hammer.
Hammer, whose early training was in library science, has long focused on how organizations learn. It quickly became clear to her that person-to-person contacts play a pivotal role. In one of her first jobs, she had to develop an information kiosk for emergency-room doctors at a North Carolina medical center. The facility gave users quick access to medical texts and research papers, but she learned that the feature doctors most valued was one that allowed them to do some networking: a list of experts they could contact for quick consultations whenever they encountered an unfamiliar case.
Hammer's interest in knowledge mapping grew after the transportation department hired her to run its newly created Knowledge Management Office in 2003. One of her first activities was to conduct a series of interviews to learn where various employees got information and technical assistance. She found that institutional knowledge was growing increasingly attenuated: older employees, who had worked in the agency in an earlier, less fiscally constrained era, had a wide network of contacts throughout the state, but younger staff members brought on in times of greater fiscal austerity had far fewer professional associates outside their immediate geographic region. "Because of downsizing and budget cuts, networks had become very localized," Hammer says.
One other experiment also helped convince Hammer that knowledge mapping would be a useful tool for the department. She asked traffic engineers in one of the state's nine districts whom they relied on to get their jobs done. Surprisingly, it turned out that one of the region's most essential employees was not a manager or a technical assistant but rather the district engineer's personal assistant. "She was so important that the district engineer told us that when she quits, he'll leave too because he doesn't want to work there without her," Hammer says.
Hammer suspects that social networks have become more important in recent years. Back in the 1950s and '60s, when many of the department's managers were World War II veterans and their main job was building Interstate highways, the organization had a strict, almost military-like hierarchy. Everybody had a narrowly defined function. But baby boomers have had a much looser management style, and the department's tasks have become much less standardized. What's more, the post-baby-boom generation of workers are more likely than their predecessors to look to their jobs for personal fulfillment; they tend to come and go from the department rather than stay for entire careers, in the process bringing--and taking away--a more diverse set of capabilities. Finally, the latest, Internet-inspired generation of workers is more attuned to networking and collaborative work.
To Hammer, all this suggests that the department urgently needs to understand its interpersonal knowledge network, figure out what key individuals know, and then make sure that it creates "redundancy of knowledge" wherever it finds itself too dependent on a few individuals. Her main tool for accomplishing the latter function is to convene communities of practice--groups of individuals with common interest but different backgrounds or positions in the agency who come together to address mutual concerns.
As of last summer, the agency had 10 such groups, and they already are having a significant impact. One group, for instance, helped identify recurrent problems that arise on major construction projects. When participants discovered that certain deficiencies in planning seemed to arise repeatedly on different projects, they persuaded Hammer's office to create another group in which people who carry out construction projects can give planners feedback. Another group is working to refine quality-control procedures. Still another is developing a manual on how to renovate historic bridges for which blueprints have long since been lost.
The value of communities of practice may seem obvious, but such groups conceivably could run afoul of agency politics. Consider the Right-of-Way and Utilities community of practice, which consists of officials in each of the agency's nine districts who are responsible for securing the rights to take over private property for new road projects and helping affected landowners. The group is developing a long-term strategic plan for replacing the large number of baby-boom officials who represent a disproportionate share of its workforce. But it also is trying to address a more immediate problem: workloads are tight, and some crucial employees--especially property appraisers-- already are in short supply. Its solution is to help districts share manpower.
Every quarter, the district officials get together and discuss their workloads. When one district has a labor shortage, others help out if they have manpower to spare. It's a logical solution to a difficult problem. But it theoretically could upset the politics of the agency. Some district supervisors have expressed concerns that if they lend staff to other districts too often, the department's central office will reduce their staff levels.
Hammer has several strategies for avoiding such pitfalls. First, she says, communities of practice have to yield short-term dividends. All too often, it's hard to get leaders interested in actions that will solve problems that are over the horizon. But by sharing personnel between districts, the right-of-way and utilities group believes it will help the department avoid costly construction delays in the here and now. Saving the agency a great deal of money, in turn, will help secure continued support from top management and policy makers--and thus buy time for the community of practice to put into effect longer- term efforts to address its piece of the agency's demographic problem. In general, Hammer believes the Knowledge Management Office has to start producing tangible results in this, its third year. She's confident it can meet that test.
Since knowledge management is about changing organizational behavior- -by encouraging staff to share, rather than hoard information, for instance--leaders also have to model the kind of attitudes they want to encourage, Hammer argues. That means putting the needs of the organization ahead of personal or turf interests. When one of Hammer's first projects, the creation of an online community of practice, stalled (participants didn't know or trust each other well enough to discuss problems openly), Hammer didn't hide the experience to protect her own career. Instead, she admitted the "dismal failure" and took responsibility for it in a widely distributed e-mail. This admission showed officials in different departments that she wasn't out to get them. It also demonstrated that she was confident she had support from the agency's leadership to try new approaches. That, of course, is exactly what she wanted other agency staff to do.
Finally, Hammer argues, knowledge management has to transcend turf battles. People in bureaucracies have a natural tendency to want to hold on to what they know. Knowledge is power, after all. It's much easier to persuade them to share it if you aren't seen as likely to use it against them. That's why the Knowledge Management Office works to facilitate, not direct, communities of practice. And it's why the office is based in Charlottesville, where it is less likely to be seen as part of the department's central office in Richmond, the state capital.
The hope, of course, is that by assiduously staying neutral, the Knowledge Management Office will keep the department on track as the state undergoes a generational change in leadership and staff.