Deep inside James Q. Wilson’s 1989 classic, Bureaucracy, is a nugget with lasting insight. Why, too often, do bureaucracies fail? Wilson explains that different administrators, at different levels of the bureaucracy, have different perspectives, roles, and incentives. Too often, these roles fail to mesh, problems fall through the cracks—and, as a result, programs can fail, sometimes catastrophically.
He identifies three different levels of bureaucrats. “Operators” work the front lines, performing the most basic of an agency’s work. They define the agency’s culture by what they do, how they do it, and how they pass along the accepted modes of business. “Managers” are the critical shock absorbers between the front lines of operation and the often-turbulent political environment. Finally, “executives” guard the organization’s turf, seek to preserve its autonomy, build political support, and struggle to secure the resources needed to do the agency’s work.
At each level, the agency’s officials do different jobs. They need different skills. That produces different perspectives, both on the agency and on its mission. When they mesh well, and especially when the in-between managers build strong bridges between the three cultures, the agency performs well. When those bridges crack or crumble, trouble usually follows.
Consider the space shuttle Challenger accident in 1986, when the spacecraft disintegrated, taking the lives of its seven crew members and setting back the nation’s space program for years. The investigation by a presidential commission revealed a horrible fact: the night before the accident, worried operators in one of NASA’s contractors, Morton Thiokol, frantically warned that launching in the cold weather predicted for the next morning could cause the seals in the vehicle’s solid-rocket motors to fail, with disastrous results. A frustrated NASA launch manager rejected the warnings, saying, “My God, Thiokol, when do you want me to launch, next April?”
shuttle program was already troubled by delays, and the agency’s managers worried that more delays would weaken their claims that the shuttle was becoming a reliable space truck. NASA’s top executives never heard the warnings about the cold-weather risk. They gave the order to launch without knowing that front-line operators had furiously argued against it. Just 73 seconds into the flight, rubber o-rings in the solid-rocket motor failed and escaping hot gases ignited the shuttle’s hydrogen tank, disintegrating the launch vehicle.
A careful reader of Bureaucracy would instantly understand the problem, framed by the book’s subtitle: “What Government Agencies Do and Why They Do It.” After all, no sensible NASA official would willingly send seven astronauts to their death or undermine their signature program. Why did they do it? To put the crisis into Wilson’s framework, the operators worked within their culture to make sure that the equipment they built worked flawlessly. Managers were torn between the worries of operators, which were endless in a system as complex as the shuttle, and the demands of executives and outside stakeholders, whose support was crucial to maintaining the program. The executives were working through difficult, recurring public relations problems, including tough questions on the evening news. At each level, officials worked within their culture. Breakdowns between the cultures led to disaster.
As is so often the case, Wilson’s approach not only is analytical, in understanding what agencies do and why they do it. It’s also prescriptive, in charting what risks organizational culture brings to the complex workings of government agencies, what administrators need to be alert to, and how they can ensure that a narrow vision doesn’t undermine their mission. It’s an enormous insight on an important—and eternal—puzzle for government performance through complex bureaucracies, and just a tiny piece of Wilson’s enduring legacy for governance.
Donald F. Kettl is Dean of the University of Maryland School of Public Policy and a nonresident senior fellow at the Brookings Institution.