Leaders Who Succeeded -- and Those Who Didn't

By Joe Davidson
Wednesday, July 8, 2009

There's no shortage of studies on how to make the federal government work better. Many of them provide important information, yet they often plow the same ground and can be as dull as a worn rug.

But one report scheduled for release today caught my eye because it deals not only with practices that work, but also with leaders who fail.

A Harvard researcher, along with the government consulting firm Booz Allen Hamilton, interviewed more than 250 peopleand reviewed agency budgets, congressional testimony and other documents to produce "What it Takes to Change Government" (pdf).

They looked at 11 case studies, eight of leaders who were deemed successful and three of leaders who were not. The failure group wasn't large enough to make a valid comparison with the successes, so the researchers added six others "who were not noted for having tried to make an ambitious change in strategy." I'll call them the mediocre crowd.

The study found the failures spent too much time hobnobbing with other bigwigs and not enough time with people on the shop or office floor. Successful leaders "manage within their organization, not just at the 50,000-foot level," the report says. "Leaders who didn't achieve their goals spent just one-quarter of their time internally."

The researchers would identify only three of the successful former agency leaders they interviewed: the Coast Guard's James M. Loy, the Government Accountability Office's David M. Walker and Charles O. Rossotti of the Internal Revenue Service. Loy told the researchers that he succeeded "because of the thoroughness of the process we went through to listen to everyone and understand what they wanted."

While the study focused on top leaders, its points also apply to "government leaders at all levels, including Senior Executive Service personnel, career staff, congressional staff and White House officials," according to the report.

One major obstacle to an agency's success is interference by political appointees, the study found. They certainly can play a leading role in helping a president create change you can believe in, but too many can stunt rather than energize an institution.

Steven Kelman, the Harvard University Kennedy School of Government professor who conducted the study, said it confirmed that "the lower the percentage of political appointees in your agency, the more successful the agency is."

"In six of the eight successful agencies the percentage of political appointees as a percentage of total managers was under the government average," Kelman said in an interview. "And in four of the eight, it was actually less than one-tenth of the government average."

Across government agencies, on average about 1.5 percent of civilian managers are politically appointed, Kelman said. But in three agencies that have been successful in accomplishing a significant change in strategy -- the GAO, IRS and Coast Guard -- politicos make up no more than 0.02 percent of the managers, he said.

The report listed several things that failing leaders generally skipped, but were on the did-do list of successful bosses.

-- The failures pursued too many goals. They should have concentrated on two or three, as their successful counterparts did.

-- Few leaders in the unsuccessful group measured performance, and "when they did, the measures focused on cost and production, but neglected quality and customer satisfaction," according to the report. Those who succeeded measured performance in their agencies as a way to communicate goals, delegate authority and establish accountability.

Yet, as important as performance measures are, "almost none of these leaders talked about pay for performance as a technique they used," Kelman said. That's important to note as the Obama administration tries to restructure the way federal employees are evaluated and rewarded. The administration is reviewing the Pentagon's National Security Personnel System, the government's largest pay-for-performance operation.

-- Bureaucrats are known for reorganizing their shops when they don't know what else to do. The study found that reorganizing might make things better. Just a third of the unsuccessful and mediocre managers reorganized, compared with three-quarters in the successful group. The successes felt they had to shake things up to meet their goals.

-- Only 11 percent of those who failed or were mediocre used a strategy planning process, compared with 75 percent of the successful managers. The researchers noted that not all of them produced a written plan, but they did effectively communicate their goals.

"Getting people involved at all levels is a risk-mitigation device, not just a motivational device . . ." Rossotti said in the report. "It provides a firmer basis for strategic planning by exposing you to a lot of details you may not be aware of."

You can view a summary of "What it Takes to Change Government" here (pdf).

Contact Joe Davidson at federaldiary@washpost.com.

View all comments that have been posted about this article.

© 2009 The Washington Post Company