The Tyranny of Anecdotes
t the end of this month, agencies will deliver their first-ever annual performance reports, as required by
the 1993 Government Performance and Results Act (GPRA). The Clinton administration and members of Congress will manufacture spin and sound bites to make sure the reports get their 15 minutes in the media glare. The resulting stories will zero in predictably on agencies that have fallen short of their goals.
Despite the fact that GPRA was hatched by Congress, legislators undoubtedly will loose a cornucopia of colorful quotes about how government has yet to get any results Americans care about and will charge that agencies aren't measuring the right things. The administration will no doubt weigh in with its own observations. They'll be something like: "Look at all these marvelous improvements agencies have made," and "We've come a long way; we have a long way to go, and by the way, we've cut more than 350,000 federal jobs and reduced agency budgets in the process."
After the one- or two-day wonder is over, the lights and microphones are packed up and shipped to the next road show and the scribes are back to handicapping the presidential candidates, the essential GPRA question will remain: So what? Agencies, the administration and Congress have spent years crafting and perfecting, grading and criticizing, defending and improving their goals, plans and outcomes. But did all this Sturm und Drang amount to anything? The answer, as with so many things in government, is a resounding "yes and no."
GPRA itself, with its attendant rules and requirements and studies and grades-and the inevitable flourishing of a new industry of consultants, classes and conferences-has brought much acrimony and relatively little in the way of meaningful action. If, as its drafters and their disciples have contended, the point of GPRA was to tie agency funding to the achievement of program results, then it was doomed from the start. The concept ignores the raison d'etre of the federal budget and appropriations process: the geographic redistribution of tax receipts. Administrations and appropriators have shown themselves to be utterly unwilling to relinquish the power of the purse for any reason, least of all a GPRA-induced, quasi-scientific linkage between programs enacted primarily to bring home political bacon and whatever happens after they are enshrined in law and ensconced in agencies.
But if, by giving visibility and encouragement to a rising undercurrent of results-based management by forward-thinking federal managers, GPRA is helping turn the tide toward planning and accounting for program results, then the law has legs and at least some potential for success. As much in spite of GPRA as because of it, a number of agencies truly have changed how they operate. They really do care about measuring, managing for, achieving and even improving results. Oddly enough, however, taking results seriously hasn't won them plaudits from an approving Congress or honors from a proud administration. Instead, in many cases, careful planning and shifting of resources from unsuccessful or unneeded efforts toward results-producing ones have brought managers only bad-mouthing, bad press and bad news on the funding front. And in one high-profile case, an agency chief essentially was ousted for managing to achieve remarkable results.
Kenneth Kizer, a former emergency-room physician, once headed California's Department of Health Services. In 1994, he took over the Veterans Health Administration as undersecretary of the Veterans Affairs Department for health. VA hospitals went on "Kizer time" almost immediately. "I used to set very short deadlines. What was surprising was how many were met," says Kizer. Indeed, within a year of becoming VHA chief, Kizer began the largest reorganization of VA hospitals since Congress created the system in 1946.
Mirroring the needs of an aging veteran population less likely to suffer from war wounds than the effects of poverty-homelessness, substance abuse and poor nourishment-VHA organized its 173 hospitals and more than 500 other health care operations into 22 Veterans Integrated Service Networks (VISNs) focused on providing health care, rather than justifying the existence of their facilities. In fact, VHA shifted its focus entirely from costly inpatient care to outpatient services offered in more than 200 clinics. Between September 1994 and September 1998, 52 percent of VHA hospital beds were eliminated and bed days of care per 1,000 patients fell 62 percent. By fiscal 1998, ambulatory care visits were up 43 percent from fiscal 1994.
VHA also reflected another change in veteran demographics: the shift in population from North and East to South and West. Kizer's response was a resource redistribution plan called the Veteran's Equitable Resource Allocation (VERA) system. The system distributes operating funds among the agency's networks based on the number of veterans served. In its first year, VERA caused nine VISNs to lose some 1998 funding, while 13 gained. For example, VISN 3, in New York, lost $124 million, while VISN 18, in Phoenix, gained nearly $60 million.
VHA's shifts in focus and resources brought results. Staffing fell 11 percent between December 1994 and September 1998, while the number of patients treated per year rose 18 percent. Surgeries performed on an outpatient basis-increasing productivity and reducing deaths-rose from 35 percent of all surgeries to 75 percent from September 1995 to March 1998.
Kizer put in place a number of quality of care measurements used in private sector health care. "It was one of my premises that, if VA was to survive, it had to demonstrate its value compared with other health care players, so it had to use the same measures those players use," he says. "The essential issue in my mind is if you get the same quality of care in VHA as the University of Pennsylvania medical center, and the same level of customer satisfaction, let's compare whether medical care is a better deal at VHA or [the university]. Look at technical quality, service satisfaction, price, functional status and decide whether to buy it in the private sector or provide it with VA assets."
On a wide variety of indices, VHA is measuring up. For example, agency compliance with prevention and early-detection recommendations for cancer, smoking and alcohol abuse doubled between fiscal 1996 and 1997, besting private-sector performance on all indicators for which there was comparable data. The agency also tracks the one-year survival rate for patients with nine conditions-such as chronic renal failure, congestive heart failure, pneumonia, depression and schizophrenia-which account for nearly 12 percent of its patients. Since fiscal 1992, survival rates for several conditions significantly improved and remained high for the others. VHA mortality rates for abdominal aortic aneurysm, total hip replacement and several other surgical procedures are the lowest or equal to the lowest in the United States.
What's more, veterans are happier with their health care. On a 100-point scale for customer service, VHA outpatient clinics scored a 79 with veterans who had received care during two weeks last May. The average score for private-sector hospitals was 70. Veterans were surveyed for the American Customer Satisfaction Index, a rating produced quarterly for private-sector firms by the University of Michigan Business School, the American Society of Quality and consulting firm Arthur Andersen. The survey, released in December, found that the quality of service at VHA clinics far exceeded veterans' expectations.
Government by Anecdote
So Kizer is a hero, and VHA has been rewarded, right? Not in Washington.
In fact, after failing to win renomination in September 1998, at the end of his four-year term, and again in June 1999, at the end of a nine-month extension granted by Congress, Kizer withdrew from contention. By June, Kizer's VHA reforms had so angered legislators that they were lining up to threaten to place holds on his nomination if it ever escaped from committee.
Many legislators apparently couldn't stomach the process Kizer used to transform VHA into a patient-centered, effective and efficient health care system. Sen. Ben Nighthorse Campbell, R-Colo., repeatedly took Kizer to task for considering closure of the nursing home and outpatient clinic at Fort Lyon, Colo., a location so remote that the VA facilities there must run their own sewer system and fire department.
VHA, which already had closed the Fort Lyon hospital, had considered shifting services provided at the clinic to another location. "It is markedly more expensive to provide care [there], it is inaccessible to patients' families, it has turn-of-the-century vintage buildings," says Kizer. "Any way we looked at it, we needed to move that facility."
The Rocky Mountain VISN, which includes facilities in Colorado, Utah, Montana and Wyoming, was among networks with the lowest patient growth and is slated, under VERA, for budget reductions. Wyoming's congressional delegation also complained to Kizer last summer. Meanwhile, Sen. John Kerry, D-Mass., threatened to hold up Kizer's nomination because of hospital cutbacks in his state, which has seen reductions in veterans, and consequently in VA hospital beds and funding.
The reformed VHA system was "predicated on a concept that is very difficult for elected officials to understand," Kizer says. "[VISN] boundaries do not follow political boundaries. They were based on overall patient referral patterns. The patient is the center of the universe. It was counterintuitive to the political view of the world." When VA officials were able to make the case for closures, consolidations and resource shifts at the local level, veterans, communities and employees usually came around. But, "when elected officials throw up a roadblock, it changes the whole dynamic and it becomes more difficult," Kizer says.
In 1998, for example, James Farsetta, director of VISN 3 in New York and New Jersey, was lambasted by Congress members for returning $20 million to VHA for redistribution. Anticipating fiscal 1998 budget cuts, the network was able to cut staff in fiscal 1997 and wound up with extra 1998 funds. Already in fiscal 1996 and 1997, VERA had shifted $140 million out of the network, reflecting a declining veteran population. In 1997, VHA found that only 6 percent of New Jersey's 760,000 veterans used the VHA system. Nevertheless, when Farsetta sent back the $20 million, Reps. Gerald Solomon, R-N.Y., and Sue Kelly R-N.Y., erupted, charging Farsetta with having engineered the staffing reductions to win a performance bonus despite findings of substandard care in VISN 3 hospitals. "Farsetta took incredible grief over good management," Kizer says. "You can be assured no director will ever do that again. The next year they put $20 million back into the network. That's the first step toward compromising the system."
Particularly troublesome to Kizer are the many occasions on which legislators closed their ears to VHA's impressive overall results in favor of relating anecdotes about such problems as constituents who had to wait too long for clinic appointments. "You would think Congress would say, 'Wow,' but what you hear is, 'He didn't get into the clinic for five months.' '' Kizer recalls. "That shouldn't have happened, if it did, and we need to fix it, no question. But that's what you're judged on." In Washington, anecdotes almost always trump analysis and measurable improvements. That's why managing government agencies for results is such a thankless task. "People aren't used to judging performance in terms of quantifiable outcomes," Kizer says. "If you don't do it that way, then you do it based on anecdote.
"You need to look at the system," he adds. "If you govern [by] anecdote, you're chasing your tail because you're continually chasing that one person. You can't judge the process if you're looking at a series of one."
Shiny vs. Mundane
VHA isn't alone in having its efforts to manage for results stymied by anecdotes, political parochialism and efforts to curry favor with vocal constituents. Creating a new national park, for example, is a terrific way to make political hay. But legislators, who are adding new units to the national park system at a rate of five a year, are much slower to provide the funding needed to maintain new parks once they're open. National Park Service estimates of the maintenance backlog have gone as high as $5 billion, but Congress is only willing to invest $125 million a year in construction and maintenance.
And when, like VHA, the Coast Guard crafted plans to close facilities in order to shift resources, the Clinton administration happily reduced the agency's budget requests by the anticipated savings from those closures. But it did so knowing full well that Congress would approve the budget requests while preventing the closures, leaving the Coast Guard shorter on funds than before. For fiscal 2000, for example, the Coast Guard's budget was reduced to reflect the closure of two air facilities. Congress approved the requested amount, and not only prohibited the closures, but ordered the Coast Guard to build an additional air facility near one of those it had planned to close because it was no longer needed.
Legislators' attraction to what's shiny and new coupled with their aversion to mundane maintenance afflicts the Coast Guard. So far, the service has been unable to get appropriators to cough up money to staff and maintain new ships that they enthusiastically funded via a special fiscal 1999 supplemental measure to beef up the war on drugs. The Army Corps of Engineers, laboring under a backlog of deferred maintenance worth $329 million, has taken to shooting pictures of the backlog's effects to convince Congress to pay up.
You might expect that watching legislators and administrations cynically disregard their efforts at performance-based management would leave federal managers discouraged. But most have grown accustomed to Washington realities and take solace in the fact that at least GPRA and related reforms favor their efforts. "My personal read is that those who have made a commitment to productive, outcome-related government will bear up well under scrutiny in the near-term future and those who haven't made that commitment won't," says Coast Guard Commandant James M. Loy. "My guess is those who do will get rewarded in the budget process and those who haven't won't." Of course, Loy quickly acknowledges that he may be overly optimistic. "To some degree, that goes against the grain of what we've watched in this city over time," he admits. "Great failures have often been rewarded."
Likewise, great successes often are punished, as Kizer will attest. In his case, not only did legislators line up to take shots at him, but the administration that brought him to Washington damned him with faint support. Kizer might have overcome the objections of a changing cast of disgruntled Congress members. But not after he also ran afoul of the Office of Management and Budget.
Testifying at an unofficial confirmation hearing before the Senate Veterans Affairs Committee on Sept. 22, 1998, Kizer answered honestly a question about whether he could live with the budget VHA received under the caps created by the 1997 Balanced Budget Act. Toting up the added costs of a new Hepatitis C treatment, extra emergency care for veterans under President Clinton's proposed Patients Bill of Rights, and a federal pay raise, Kizer said that $1.4 billion to $1.7 billion in new expenditures weren't covered and that paying for them would require new funding or curtailed services. OMB reportedly read this as an unauthorized direct plea to Congress for money not requested by President Clinton and lobbied the White House against Kizer's renomination.
Facing the expiration of his temporary reappointment, the clamoring of Congress members for his head, and the enmity of OMB, Kizer simply gave up. "After being on the job for five years, effecting change that no one thought was possible-with results that showed the system was working much better-why should I have to go through such a hassle to keep the job?" Kizer asked himself. "The administration, my boss, didn't seem to care and OMB was actively lobbying against me." Thus, on June 29, he asked the administration to pull back his name. In September, he became president and chief executive officer of the National Quality Forum, a nonprofit organization dedicated to measuring and reporting the quality of national health care delivery.
Not surprisingly, Kizer now advises agency heads hoping to manage for results to fight lobbying with lobbying. "One thing I would do differently-though I do not know how because there are only so many hours in the day-would be spending more time with members of Congress on a personal level and holding the hands of some people in the veterans service organizations.
"Most problems are personal problems," he says. "[They might not have happened] if you had spent another hour explaining."
Managing for Results
|Army Corps of Engineers||B|
|National Park Service||C|
Build strategic plans that have meaning for line managers. Though it participated fully in Transportation Department GPRA planning, the Coast Guard didn't produce its own strategic plan in 1999, after canvassing line officers to set performance goals and to determine the equipment and human resources mix needed to achieve them.
Be prepared for setbacks when you try to re-apply resources to better achieve results. Build budget requests, forecasts and operations plans to account for the possibility that the administration or Congress will refuse to allow closures or redistribution of resources.
Make as strong a case for funds to maintain existing assets as you do for money to purchase new ones. Costly and growing maintenance backlogs that go unfunded for years can prevent you from achieving program performance goals.
Work constantly to deepen and expand the base of support for your agency's programs. Spend as much time as possible showing legislators how and why you make decisions about changing facilities and programs to improve results.
Make visible the costs of budget cuts, failures to provide requested funding and refusal to permit necessary reallocation of resources. True performance budgeting-showing what level of performance can be "bought" for different levels of appropriations-is vital if administrations and legislators are to make intelligent choices among competing priorities.
NEXT STORY: Government Performance Project