performance management: a “start where you are, use what ... · this report by chris wye,...

Ma

na

gi

ng

f

or

Re

su

lt

s

Se

ri

es

Chris WyeDirectorCenter for Improving Government PerformanceNational Academy of Public Administration

Performance Management: A “Start Where You Are, Use What You Have” Guide

O C T O B E R 2 0 0 2

1

Performance Management: A “Start Where You Are, Use What You Have” Guide

M A N A G I N G F O R R E S U L T S S E R I E S

Chris WyeDirectorCenter for Improving Government PerformanceNational Academy of Public Administration

October 2002

2

T A B L E O F C O N T E N T S

Foreword ........................................................................3

Author’s Note..................................................................4

Introduction ....................................................................6From Theory to Practice ..............................................6The Issues on the Ground............................................7From Practice to Theory ..............................................7Organization of the Report..........................................8

Making the Case for Performance Management ............9“The Public Sector Is Different” ................................10“This, Too, Shall Pass” ..............................................12“GPRA? Never Heard of It” ......................................14“It’s Not My Responsibility” ......................................15“We’ve Got It Covered” ............................................16“We Just Give Away Money” ....................................17“Our Program Is a Drop in the Bucket” ....................18“Our Work Is Very Far Removed from the

Front Line” ............................................................19“We’re the Third Largest Government in

the World” ............................................................20“My Boss Isn’t Interested”..........................................21

Designing Performance Indicators ................................22“Our Program Is Different” ........................................23“We Don’t Have the Data”........................................24“We Can’t Find a Valid Measure of Performance” ....26“We Have No Control Over Outcomes” ..................27“No One Wants to Do Crosscutting Measures” ........28“Our Outcomes Take Years to Achieve” ....................30“Our Data Are a Mess” ............................................31“Our Data Are in Narrative Form” ............................32

Aligning Performance Processes....................................33“Well, We’re Waiting on ______” ..............................34“Someone Else Does That” ........................................35“Our Budget Account Structure Is a Mess” ................36“We Have No Control Over That” ............................38“We Don’t Have Unit-Cost Data”..............................39“Everyone Wants the Existing Account Structure” ......40“Our GPRA Plan Is Not Our Operating Plan” ..........41“We Don’t Have Control Over Grantees”..................42“There Is No Way to Include Overhead Expenses” ....44

Using Performance Information ....................................45“What’s the Use? Decisions Are Political” ................46“No One Uses Indicators Anyway”: Part I..................47“No One Uses Indicators Anyway”: Part II ................48“No One Uses Indicators Anyway”: Part III................49“Our Data Are on the Wrong Cycle” ........................50“Our Organizational Structure Impedes Use”............52

Communicating Performance Information ....................53“No One Here Gets It”..............................................54“We Don’t Have Money for New Reports”................55“Whom Would We Communicate To?” ....................56“No One Will Understand” ......................................58 “But We Didn’t Meet Our Goal” ..............................59“What Would We Say?” ............................................60

Publications of the Center for Improving Government Performance ............................................61

About the Author ..........................................................64

Key Contact Information ..............................................65

3

PERFORMANCE MANAGEMENT

F O R E W O R DF O R E W O R D

October 2002

On behalf of the IBM Endowment for The Business of Government, we are pleased to present this report by Chris Wye, “Performance Management: A ‘Start Where You Are, Use What YouHave’ Guide.”

This report continues the Endowment’s long-standing interest in performance management andthe challenge of managing for results. In 2001, the Endowment published Managing for Results2002, edited by Mark A. Abramson and John M. Kamensky. That volume contains six Endowmentreports all related to aspects of performance management, including strategic planning, the use ofprogram evaluation to support performance management, managing for outcomes, developingcross-agency measures, using the Government Performance and Results Act as a tool for manag-ing third-party government, and the use of performance data to enhance organizational account-ability. This report by Chris Wye adds substantially to our understanding of how governmentmanagers can overcome common problems in the design, alignment, use, and communication of performance measures and information.

This guide builds on both Mr. Wye’s distinguished 20-year career as a leader in government per-formance management and the recent work of the National Academy of Public Administration’sCenter for Improving Government Performance. We trust that this report will provide valuableadvice to government managers seeking to enhance their organization’s use of performance man-agement by developing effective responses to common obstacles that they might confront whenimplementing performance measurement systems.

Paul Lawrence Ian LittmanCo-Chair, IBM Endowment for Co-Chair, IBM Endowment forThe Business of Government The Business of Government

4


In the following pages, strong emphasis isgiven to the relationship between performanceand public service.

From the perspective of this report, this con-nection—performance and public service—is the single most important lesson to haveemerged over the last decade of experience in the implementation of the GovernmentPerformance and Results Act and related performance-based initiatives.

Why is this so important?

Because it identifies one of the most—if notthe most—important motivation for public servants.

To talk about improving government perfor-mance alone, without connecting it to ahigher vision of public service, is, more oftenthan not, to focus on all the things that needto be fixed—to see the glass as half empty.

When successive new political administrationstalk about the need to make the governmentwork better, is there not a strong implicationthat things are not what they should be andsomeone is falling down on the job?

Even for the best civil servants, this approachcan sometimes seem to be—at the least—lack-ing in political art.

But to talk about the high calling of public ser-vice and the relationship between good publicservice and good performance (as opposed totalking just about the need to improve perfor-mance) is to go deep into that private place inthe heart of many public servants where lifechoices matter and career goals have realmeaning—to see the glass as half full.

The fact is that most civil servants want to dowhat they are doing. They have specificallychosen a career in government. They want tobe public servants because they want to makethings better in their communities and country.

Author’s Note

5


They want the best and highest performancefor themselves and their fellow citizens.

Over the last 10 years I have had the privilegeof talking with many groups of civil servantsabout their work and, in particular, their effortsto improve government performance. Overtime it became clear that an effective way tobegin the discussion was to ask one or both ofthe following questions.

Do you want to be a civil servant? The answers, recorded on three-by-five cardsto preserve anonymity, almost without excep-tion are “Yes,” and they are frequently accom-panied by short, incisive affirmations, such as:“No question.” “This is it.” “I don’t want to doanything else.” “I do not want to work in theprivate sector.” When the audience is invitedto speak, one after another the hands go up.

How did you make the decision to become acivil servant?Someone speaks. Someone else speaks.Several people speak at once. The audiencehas to be asked to slow down. The vignettesare real. “I always wanted to help make thingsbetter.” “My father was a public servant andhe instilled a desire to make my life count.” “I wanted to be on the cutting edge of myfield—but in service to my country.” Truly it isa moving experience. I invite anyone to try it.

The point is that public servants themselvessee public service as the best, highest, andmost important motivation for improving gov-ernment performance. It is their motivation.

Not that of the Congress. Not that of theincumbent administration.

Theirs.

We could not wish for more.

6


From Theory to PracticeI had just returned to my office after listeningto an expert presentation on designing perfor-mance indicators. The presenter was veryknowledgeable and the slide presentation was quite impressive.

The phone rang, and I found myself talking toan agency staff member who had a question.As the conversation ended, he said, “Youknow, the problem is we’re different. Our outcomes really can’t be measured.”

I was startled. I had just come from a work-shop where a leading expert talked aboutdesigning indicators for difficult-to-measureprograms. Yet this caller said, basically, that hecould not find performance indicators for hisagency’s outcomes.

Why was I startled? For one, both individualswere extremely knowledgeable. After amoment’s reflection, it occurred to me thatthese different perspectives illustrated the dis-tance between theory and practice. The expert

was telling others how to do it; the caller actually had to get it done.

In 2002, nearly a decade after the Govern-ment Performance and Results Act (GPRA) wasenacted, many still struggle with fundamentalissues. This is true both for those who havesubstantial knowledge about GPRA and out-come-oriented measurement, and for thosewho have just begun to learn about GPRA andbarely know what the word “outcome” means.

The U.S. Office of Personnel Management’sEastern and Western Development Centers—which provide training to thousands of govern-ment employees annually—report thatattendees have only a rudimentary knowledgeabout these topics. Indeed, many have neverheard of GPRA.

For experts and beginners alike, many of themost challenging issues remain far below thelevel of theoretical discussion and public dis-course. They are much more basic.

Introduction

7


The Issues on the GroundThe purpose of this report is to respond tosome of the mostly frequently raised issues onthe ground. The intent is to provide simple,practical, timely, low-cost strategies to helptranslate theory into practice.

As the director for the Center for ImprovingGovernment Performance at the NationalAcademy of Public Administration, I have met,observed, and talked with many of the nation’sleading experts on performance-based man-agement over the last 10 years.

As director of the Center’s PerformanceConsortium, an organization made up of 30agencies that have come together to fund anannual program of peer-to-peer dialogue onemerging practices in performance-basedmanagement, I have come to know manyagency representatives who are implementingGPRA and related initiatives.

Sitting between the experts and implementers,I sometimes have found myself wonderingwho is who. Many expert presentations areframed far beyond the horizon of day-to-dayreality. They stimulate thought and promptenergetic discussion, but agency attendees can nonetheless return to their offices feelingunsure about how to deal with issues that arevery ordinary yet very important.

Even worse, the agency person can feel veryalone. The world of theory seems a long wayoff and the problems at hand can seem notworthy of conversation. There sometimes is asense that raising them will make one appearnot entirely confident, successful, or compe-

tent in the circle of office colleagues or thecommunity of professional peers.

This report is intended to help these people. It gathers a group of issues raised far more frequently than generally realized and offerspractical suggestions. The issues are framed as they often are raised in daily conversation:“Our outcomes are hard to measure.” “Wedon’t have the data.” “What’s the use? Decisionsare political anyway.”

How many times have we heard these state-ments? How many times have we respondedwithout taking the trouble to learn the context?Or, how many times have we responded withtheory rather than practice?

The issues have been culled from meetings,reports, workshops, and conferences held atthe Academy over the last decade. Many morecould have been added. Judgments had to bemade in selecting the final list; no two lists likelywould be the same. In some cases, a specificissue is one of several that could have beenframed in a general set of issues. The assump-tion is that the reader will make that connec-tion. Not every issue could be taken up directly.

The responses also are taken from this richdialogue. Much of this conversation has takenplace after, rather than during, an event. Then,candor can be more freely expressed and theorycan more easily bear the weight of experience.

From Practice to Theory Beginning from the standpoint of practice hasseveral implications, as the reader will soon dis-cover. The first point is that some of the answers

given to these issues may seem repetitive. Thesecond point is that some of the answers givento these issues may seem repetitive.

This is not a typographical error.

To begin with issues on the ground is to beginat the outer edges of a circle whose circum-ference represents practice and whose centerrepresents theory. A single theory will betested in many different contexts of implemen-tation. Since this document starts with individ-ual contexts, many answers return to the samefundamental points. That can make the docu-ment seem repetitive. Nonetheless, the intentis to provide a list of frequently raised issuesthat readers can access when a particular issuearises in their daily work. It is hoped that read-ers will find everything needed to address thatissue in one easily accessible place.

Although repetition is evident in the report,this can serve as a valuable reminder that theunderlying concepts really are quite simple.We may “complexify” issues as the interplaybetween theory and practice adds newdimensions to our thinking. Yet the underly-ing concepts related to performance-basedmanagement are neither complicated nornew. Basically, we are talking about:

• Assuming responsibility as individual public servants for the high trust inherentin our calling

• Searching continuously for the highestquality public service at the lowest cost

• Using creatively whatever information canbe found to improve programs

8


• Doing something (to improve perfor-mance) in the face of all obstacles, asopposed to doing nothing

• Placing boundaries on discouragementand moving constantly toward the highand noble goal of public service

• Remembering that the money supportingpublic endeavors is not ours but the pub-lic’s, and that we are their trustees

Performance-based management itself is notcomplicated. Outcome indicators can be onlyas good as the human beings who design themand the resources available to implement them.If we had infinite time and resources, it mightbe possible to think about perfection. How-ever, no appropriation was made for designingand implementing indicators. Also, imple-menters are squeezing their attention to per-formance into already crowded schedules. Thebest that can be done is always related to thetime and resources available. The worst is notto do something with what we have. We needto start where we are and do what we can.

If the Office of Management and Budget(OMB) and the Congress do not like theresults, they can provide more guidance and more resources.

Organization of the ReportThis report is organized into five major sec-tions, presented in the order in which theissues discussed would arise in everyday prac-tice. The first section, “Making the Case forPerformance-Based Management,” addressessome of the major lines of resistance to perfor-

mance management and measurement. Thefollowing four sections consider importantstages in the performance-based managementcycle: “Designing Performance Indicators,”“Aligning Performance Processes,” “Using Performance Information,” and “CommunicatingPerformance Information.” Each section dis-cusses several issues, each of which is presentedand discussed on a single page. The top of thepage presents the issue in everyday languageand in bold type. Following discussion, thepage concludes with recommended responsesfor that issue.

The intent of this guide is to provide a conve-nient way for practitioners to find practical,low-cost help when they need it in their specific situations on the ground.


9

Nearly a decade after the enactment of theGovernment Performance and Results Act of 1993, it may seem that there is no need to “make the case” for performance-basedmanagement.

The world is different. Attitudes have changed.There is a broader awareness of the ratchetingup of performance standards around theworld. Information and communication tech-nology add dramatic increments of speed,breadth, and depth to the world economy.

Few people snicker, as they once did, aboutstrategic planning and performance indicators.They do not complain, as they once did, aboutthe variety of reform strategies: reinvention,reengineering, quality management, and per-formance measurement.

At least not openly.

There is a growing understanding that thereare central concepts underlying them all: per-formance, results, and the bottom line.

Yet the U.S. government is an enormous enter-prise. It can take years for new concepts togain general acceptance, one by one. As indi-viduals come to grips with new concepts,fairly predictable reactions can be expressed.Some are discussed in this section.

Making the Case for PerformanceManagement

10


The short answer to this line of thinkingis: “True enough, but not that different.”Performance indicators are relevant in anymanagement system, public or private, andthey should be a sine qua non in the publicsector.

Keeping track of performance in government is challenging and complicated by factors not present in the private sector. There, theprofit/loss motive adds impetus to efficiencyand effectiveness: There is less need to trackoutcomes. Programs designed by Congress,based on social goals and political issues, areless amenable to these management issues.Executive branch agencies often must retrofitthe elements of good management to programsnot designed to fully accommodate them.

This may make performance-based manage-ment more challenging, but not less necessary.At one level, government is a public trust, andits purpose is to faithfully pursue the publicinterest. How can it not hold to the higheststandards of accountability and performance,regardless of any challenges that may arise?Performance-based management should bethe basis for all public management.

Nor does the challenging endeavor of publicservice make the use of performance indica-tors less possible. Remembering that all effortsto track performance must be accomplishedwithin the bounds of existing resources andanalytical constraints, the standard is not theperfect but the possible. If Congress or the

citizens want more, they can provide moreguidance and more resources.

The absolutely unacceptable response to thecall for performance indicators is to do noth-ing or do something with weak intent.

Recommended Responses 1. As a public servant, view your role as

holding and managing in trust for thecommon good the resources provided toyou by your fellow citizens. The key wordhere is “trust.” Think about what the word“trust” implies. Does it mean to be trust-worthy? How would you put that intopractice in your particular activity? Howwould you be a trustee of the commongood? How would you explain the valueof what you do as a trustee to a neighbor?

2. Develop performance indicators that youwould find reasonable if you were a citi-zen wishing to understand what a givenprogram is doing with your tax dollars. Tryto see yourself as a citizen responsible toyour next-door neighbor, to your friends inchurch, to your colleagues in communityand civic action associations, for theeffective use of tax dollars. Ask yourselfwhat people want to know about theresults of their tax dollars.

“The Public Sector Is Different”

11


ExcerptApplying the strategic planning concept in the public sector is complicated. Goals often are multiple andunclear, sometimes conflicting, and often much larger than can be achieved. Leadership is divided amongpolitical parties and is relatively short term (in the executive branch). Congress and others reflect a multi-tude of pressures and counterpressures, complex intergovernmental and public-private partnerships, andinterventions via the ballot box and other citizen action. Planning, data, and program evaluation functionsoften have not been valued highly enough in the public sector to keep them a robust part of the decision-support systems for management.

These complications make strategic planning even more necessary and valuable in the public sector thanin the private sector. Strategic planning offers a process to identify diverse views, debate, clarify, andresolve them, and guide actions that need to be taken.

The benefits of strategic planning and strategic management can be great if they become integral parts ofhow the agency does business. Planning is a continuous process, regularly available to federal agencies toopen dialogues with the administration, Congress, and their constituencies about the changes taking placethat demand responses from them.

It helps federal agencies to:

• Develop agreement with Congress about missions and goals.

• Show the rationale for their programs and support successful programs.

• Design immediate implementation actions to deliver planned results.

• Focus on outcomes important to the public and demonstrate performance to the administration,Congress, targeted clientele, and the American people.

• Ensure that the organization will not be blind-sided by unexpected external forces.

• Broaden stakeholder involvement to clarify issues, broaden agreement about missions and goals, andstrengthen support for effective programs.

• Take a long-range perspective to help ensure that the bigger, longer-range objectives are not sacrificedto less important short-term gains.

• Facilitate tracking of performance, accountability, reevaluation, and renewal of programs, so they donot become stale.

Center for Improving Government Performance, Helpful Practices in Improving Government Performance (Washington,D.C.: National Academy of Public Administration, 1998), pp. A-4, A-6-7.

12


One of the earliest and most persistent strainsof thinking about performance-based manage-ment is that it is just the latest managementimprovement fad and that it will pass away likeothers before it. Previous efforts to establishgovernment-wide management improvementsystems, such as PBBS (Performance-BasedBudgeting Systems), MBO (Management ByObjectives), and ZBB (Zero Based Budgeting),are frequently cited as examples.

In fact, performance-based management is aglobal trend, cutting across the public and pri-vate sectors. Also it is rooted in fundamentalchanges taking place in the worldwide eco-nomic system. These changes are beingimpelled by dramatic advances in informationand communication technology. Together, theyallow more people to communicate moreinformation faster than ever before. In turn,these forces are drawing the world’s economicmachinery together, increasing its efficiencyand strengthening the systems of accountabil-ity through which resources are transformedinto products and services.

There hardly is a major government that is notaware of and taking some steps to support aperformance-oriented approach to manage-ment. Australia, New Zealand, Canada, andGreat Britain have been engaged for 10 yearsor more. Japan, China, and the nations ofSouth American are actively mining theirexperience.

In short, the public and private sectors arebecoming more efficient, as they must for thelong term. Performance indicators, as requiredby GPRA and related initiatives, are essentialto this trend. In a sense, our progress towardimplementing performance indicators can tell ushow well we are adapting to these new trends.

As these new trends take hold, comparisonsalmost inevitably will be made within andacross service sectors, continually ratchetingupward the standards for performance.

Recommended Responses1. In today’s world, with an increasingly

global economic and political system, professional public servants should bebroadly aware of management trends inother nations. The world is shrinking. Ourneighbors, competitors, friends, and ene-mies are closer to us than ever before. Weneed to understand the policy, political,and economic issues and challenges fac-ing the world community.

2. Public servants should have a specificworking knowledge of emerging perfor-mance-based management trends, includ-ing performance indicators, in their areasof management responsibility. If the worldis becoming smaller, then emerging newways of doing business are coming nearer.We need to monitor the emergence ofnew ideas and better practices whereverthey occur.

3. There is a great need to develop a man-agement culture that cultivates new ideasand better practices. If we simply monitorthe practices of others, we can be no bet-ter than good imitators. But if we create aclimate in which new and better ways ofdoing things is valued and encouraged,we can generate a culture of improvementwhere the pursuit of quality can becomean overriding mission.

“This, Too, Shall Pass”

13


ExcerptThe Results Act is based on organizational performance driven by strategic missions and out-come goals linked to those missions. It has a basis in statute, and it affects how Congress works.Essential factors for the future success of Results Act implementation include the following:

• Receiving support from top leadership in both the legislative and executive branches

• Driving the plans and goals down into organizations to change the management cultureand incentives to focus on results rather than process

• Allowing time to develop, redesign, or reengineer internal processes

• Integrating performance plans with budgets, performance measures across the activities ororganizations, program outputs with the accountability for the outcomes they are toachieve, and accounting with managerial needs …

• Balancing goals and measures among program results, customer satisfaction, andemployee commitment

• Changing information technology to make paperwork and reporting less burdensome …

Other countries have been wrestling with the same issues, particularly how to improve perfor-mance and accountability while at the same time becoming more efficient and customer-oriented. Cross-functional or horizontal coordination issues among government programs alsoare receiving attention. Answers have come in the form of separation of policy and operatingfunctions …; use of term performance contracts with agency heads and key managers; severedownsizing of headquarters staffs so that coordination became much easier; and delegation tojunior ministers portfolios focused on missions or outcomes that cut across agencies.

Center for Improving Government Performance, Questions and Answers for Improving GovernmentPerformance: Operationalizing Performance Management, (Washington, D.C.: National Academy of Public Administration), pp. 9, 10.

14


For those who have labored to support GPRAimplementation, it is hard to imagine thatmany government employees have never heardof it or do not have a good understanding ofwhat it is. Yet this appears to be the case.

Ever since the law’s enactment in 1993,instructors at the Office of PersonnelManagement’s Eastern and Western TrainingCenters have reported that very few of thepeople walking through their doors know very much about GPRA.

This does not mean that no one is aware ofGPRA or that awareness is not growing insome areas. On the one hand, the Congres-sional Research Service has done several stud-ies showing that references to GPRA have significantly increased in congressional dia-logue and activities. On the other hand, theU.S. General Accounting Office (GAO) hasdone several government-wide surveys show-ing modest levels of awareness.

Everything about GPRA is so rightly intended,so vital to the times, and so essential to goodpublic service that those who are aware of itmust actively spread its message. Should allother motivation fail, GPRA is the law. In fact,one way to view GPRA is as a legal constitu-tion for good management. From this perspec-tive, GPRA can be seen as support for buildingimportant management capacities that mightotherwise not be pursued. For example, strate-gic planning—which can be highly desirablebut is often not pursued—can be justified with

the force of law and supported with appropri-ate resources.

Following a decade of budget cuts and down-sizings, often implemented on an across-the-board, pro-rata basis, GPRA makes clear thatperformance counts. This can mean that awell-performing program can be treated differ-ently from a bad, poorly performing program.This distinction constitutes the fundamentalbottom line of good management.

Recommended Responses1. Every public servant who has taken the

oath of public service—in other words, all public servants—and is familiar withGPRA has an obligation to understandhow the law applies to his agency. Thestrategic plans, annual plans, and annualreports required by the law are, if nothingelse, the major vehicles for communicat-ing an agency’s mission and activities toall major stakeholders, including citizens.

2. Every public servant should understandthe specific ways that GPRA affects herown area of responsibility and be able toarticulate how her particular work fits intothe larger departmental picture. If all pub-lic servants could articulate simply,clearly, and persuasively how what theydo contributes to the mission of theirdepartment, and ultimately how that con-tributes value to taxpayers, much wouldbe achieved.

“GPRA? Never Heard of It”

15


There is a tendency to see GPRA and relatedperformance-based initiatives as “not in myarea of responsibility.”

This perception has many roots. In GPRA’searly days, agencies tended to appoint “pointpersons” to handle its requirements. Large, com-plex organizations—like cabinet agencies—have many layers that can become isolatedfrom each other; there is a natural tendency to think “someone else is handling that.”

Such factors suggest that the reaction of someto recent performance-based management ini-tiatives has been “business as usual”: “Here isone more thing to do. Thank goodness it’s notmy responsibility.”

This misses the fundamental point of GPRAand related initiatives—an enlarged sense ofresponsibility, accountability, and performancemakes it unacceptable for one part of an orga-nization to say to another, “It’s your faultbecause I did my part.”

The successful manager of the future mustunderstand and accept some level of respon-sibility for all parts of the service deliveryprocess, extending from inputs to outputs andoutcomes. This sometimes is called the logicmodel for service delivery. As standards ofperformance rise and the techniques for track-ing performance improve, parts of the servicedelivery process needing improvement willinevitably be identified. Other parts willdemand that action be taken, because each is held to a performance standard.

It is in everyone’s interest to ensure that every-one else is meeting established performancestandards.

Recommended Responses1. Managers should construct a “logic

model” for the service delivery system ofwhich their operations are a part. A logicmodel is essentially a graphical display orflow chart—usually arrows and boxes—depicting the sequence and relationship ofindividual parts of a process as they mustbe carried out to accomplish an objective.It may be helpful to have more than onesuch logic model for a given process: forexample, one for the major steps in thedepartment-wide process and one or morefor the more detailed aspects of particularsteps in the process. The utility of thesecharts will be exponentially increased bythe addition of dates and resource require-ments at major points.

2. Coordination, including sequencing, timing, and quality standards for servicedelivery processes, should be included inthe performance appraisal process for allmanagement levels. Although it may seemthat this may expose an individual to therisk of missing goals because other partsof a process are not performing, in fact itcan establish a reasonable context forassessing individual performance—if theoverall logic model sufficiently explainsthe interrelationships and dependenciesbetween parts of the overall process.

“It’s Not My Responsibility”

Excerpt… It is the responsibility of public servants,political and career, in both the legislatureand executive branches of government toseek improvement of government programs.

One of the essential responsibilities of publicservice is to use public resources effectivelyand efficiently—and to look for ways toimprove service delivery. The professionalpublic administrator charged with theresponsibility to manage public programsmust accept some level of responsibility forthe effective management of programs asthey relate to each other.

If there was no responsibility for the relation-ship among programs, then there would beno cabinet agencies, no large organizationsto manage groups, and no inter- and intra-agency coordinating mechanisms. While our system of government was clearly notdesigned to produce perfect programs orperfect coordination among programs,neither was it designed to be blind to goodprogram administration.

Center for Improving Government Performance,Helpful Practices in Improving GovernmentPerformance: Improving Performance AcrossPrograms: Thinking About the Issue—Taking theFirst Steps (Washington, D.C.: National Academyof Public Administration), pp. 5, 7.

16


A short story, which is true and representativeof others, illustrates this point.

Some time ago a Washington-based person,knowledgeable about management trends inthe public sector, gave a speech on the WestCoast about the importance of GPRA and performance-based management. The next dayhe got a call from a high-ranking military offi-cial who said the speech had been very effec-tive and was exactly what he had been tryingto tell “Mr. X” back in Washington. Would hebe willing to make the same presentation toMr. X and his staff? The meeting was arranged,and the presentation was given. At its conclu-sion, Mr. X looked around the table and said,“We’ve got this covered, don’t we?” Everyonesaid yes, and the meeting was over. No ques-tions. No dialogue. No interest.

To some extent, this response may be explain-able. The Department of Defense has a longhistory of attention to performance-basedmanagement, dating back to former SecretaryRobert McNamara’s tenure. The departmentmay have more data on its operations thanmany other federal departments. At the sametime, there is a persistent sense by the publicthat the department’s approach to manage-ment, cost effectiveness, and accountabilitydoes not take performance issues into accountto a sufficient degree.

People in other agencies of the governmentcould be cited, as well. They feel that they“have it covered.” Maybe they do, from their

perspective. Yet the trend toward performance-based management is more than a new way ofdoing business; it is a ratcheting up of thestandards for conducting business. Even if wedo “have it covered,” we ought never to feelthat we have arrived.

Recommended Responses1. In these and future times, it will not be

sufficient for managers to be content withexisting management systems, no matterhow good they may be at any one point intime. They will need to cultivate an openmind and maintain a proactive sense ofpursuit in relation to identifying, sortingthrough, and ultimately using thoseaspects of cutting-edge theory and prac-tice that can improve their operations.With so much thinking and practice avail-able to so many so easily via the Internetand computers, today’s manager worksunder the scrutiny of a thousand eyes. Theimage of a TV camera focused on thethinking and practice of an individualmanager may convey a sense of the extentto which each will ultimately be judgedbased on a comparison with all othersworking in the same general area.

2. Leaders, especially, must promote changein pursuit of performance. For many of thesame reasons, the awareness of change(new thinking and practice) will need tobe accompanied by a willingness to act.Speed—the speed with which new

approaches are adopted—will be an issue.Ideas slowly implemented will lose out toideas more quickly implemented. Skill atfacilitating the implementation of newapproaches, articulating direction, buildingconsensus, solving problems, and meetingchallenges will be more and more essen-tial for the managers of the future.

“We’ve Got It Covered”

17


Fortunately, this sentiment is less pervasivethan it once was. Yet it remains in some areasas one of the reactions to the broadened andstrengthened sense of accountability thatunderlies the trend toward performance-basedmanagement in the public sector.

Examples are programs with broad discretionfor the use of federal funds at other levels ofgovernment or administering entities, such asrevenue sharing, block grants, and other grantprograms. The feeling can be that the task is togive out money and the job is over. It some-times is pointed out that the authorizing legis-lation has no provision for tracking programoutputs or outcomes; hence, there is noauthority for data collection requirements.

It clearly is a challenge to track the effects offunds administered through secondary and ter-tiary agencies where there is broad discretionfor its use and no specific legal basis for suchactivity. However, it is increasingly appropriateand possible in many areas.

A broad shift in opinion seems to have takenplace over the past decade. Many now feel thatpublic administrators must assume some levelof responsibility for tracking program results,even where there is no specific legal require-ment or empowerment to do so. This reasoningholds that public administrators should dis-pense public funds within a context of publictrust. This trust requires that they make someeffort to be aware of program consequences.

Given that this sense of heightened responsi-bility is still emerging and there are few gen-eral guidelines or requirements, the approachapplied in each situation will have to beworked out directly with the parties and entities involved.

Recommended Responses1. Programs without apparent responsibility

for tracking program activities beyond theactual dispensing of dollars should reviewtheir procedures to add some level ofmonitoring to program operations.Whatever is done must be limited inscope and inexpensive. One option isinformal telephone or mail surveys basedon small samples. In a zero budget situa-tion, it might simply be done by regularrecording, organizing, and reviewinginformation gathered through routinephone calls to field offices. Professionalassistance may be needed to design anaffordable system because the task ofretrofitting a credible evaluative methodol-ogy to an information base not originallydesigned to sustain it can be challenging.But it need not be expensive.

2. These programs should reach out to sub-grantees to see where joint activity orcooperation in tracking program operationscould benefit overall service delivery andreduce costs. As noted, since there are fewguidelines in this area, this will require ini-

tiative and creativity to see what kind ofactivity may be helpful. Over time theseexperiences, even if individually less thanperfect, can aggregate into useful lessonslearned. Among these, one of the mostfundamental—and potentially useful—maybe improved understanding and communi-cation among and between administeringentities and levels of government.

“We Just Give Away Money”

18


“In the larger scheme of things, we’re not veryimportant,” some may say. In a government aslarge as ours, this feeling is both more preva-lent and less recognized than it should be. It isa silent, invisible current working against thehighest level of performance.

People connected to smaller programs some-times feel that their activities are not conse-quential enough to enter the fast runningcurrents of major management initiatives. Thisfeeling can be rooted in a true sense of smallworth; for others, it can be a conscious orsemi-conscious strategy.

In either case, it is misguided. Spending publicdollars is a public trust. We betray our sense ofpublic service when we do not treat each taxdollar with the same respect.

Ideas leading to performance improvementscan emerge from even the humblest activities.These ideas can impact far beyond the originalcontexts. The larger view also might suggestimproved ways of connecting smaller pro-grams and activities with larger ones. This“connectedness” is itself a fundamental goal of performance-based management.

A side effect is the potential career progress itmay offer energetic and innovative managerswho are seeking the management lessons oflarger programs and whose own work withsmaller programs can become known to alarger audience.

Recommended Responses1. Those managing smaller programs should

recognize that their perceptions of man-agement must be as robust as those heldby managers of larger programs. Thinkingof oneself, one’s job, or one’s programactivity as insignificant is a breedingground for all kinds of unhelpful thoughtsand inclinations. It is not humility buthumiliation. All parts of the managementprocess are worthy. It is true that no one is likely to come up to you and say, “Hey,you’re important,” or “Hey, you’re part of the team.” But that’s the way humanbeings are. How could one ever think toprogress from managing smaller activitiesto managing larger ones without demon-strating the necessary capacities.

2. Smaller programs should understand theirrelationships to larger ones, and focus man-agement attention on these relationships.The nexus of understanding has benefitsfor both program management and careeradvancement. Understanding the largercontext will help you to improve thesmaller one, and vice versa. Understandingboth will help you to be seen as someonewho understands the big picture alongwith the little one, and thus will identifyyou as a potential candidate for largerresponsibilities. Almost every manager of a large program started out managing a smaller one.

“Our Program Is a Drop in the Bucket”

19


This reaction is representative of a broaderclass of reactions: “Our job is to make thewheels go round.” “We don’t deliver the fin-ished product: We deliver the parts.” “Nobodyknows what we do.” “We’re too far removedfrom the action.”

Typically, this reaction is found in administra-tive, logistical, and support functions that aresomewhat distant from the delivery of out-comes. Not all. Some are so large (the DefenseLogistics Agency) or so important (the SecretService) that they do not fit these characteriza-tions. Yet the feeling is persistent in manyplaces. Years of working behind the scenes—away from the front lines of service delivery or on the fringes of policy and program dia-logue—have encouraged detachment and isolation.

As noted elsewhere, there can be a sense ofremoval and distance from the accountabilityor performance chain that leads to the finaloutcome. This sense of distance is the problem.

No matter the temporal, geographic, sequen-tial, or logistical distance, each part of the ser-vice delivery process must share a sense ofimmediacy with regard to its role in the deliv-ery of outcomes. This requires a broad recogni-tion of the entire delivery process and a preciseunderstanding of the specific component.

It also requires active anticipation in bothdirections: the process leading up to the spe-cific component and the one leading out from

it. Good communication must be cultivated inboth directions. Also, personnel incentive andappraisal systems must support this anticipa-tion and communication.

Recommended Responses1. All, especially those feeling distant from the

achievement of outcomes, must activelyanticipate processes leading up to and out from their own function. If you do notunderstand the relationship of your activityto the larger activities of which you are apart, who will? If you do not articulate therelationships of your activity to the largeractivities of which you are a part, whowill? The point is that all need to under-stand and be able to explain how theiractivity fits into departmental operations.

2. Personnel incentive and appraisal systemsshould support this kind of anticipation.There ought to be a specific aspect of theappraisal process that allows employees toreceive a rating in relation to their under-standing of their own specific activity andother activities leading up to the depart-ment’s mission. Continuous pursuit of this understanding, hopefully leading tosuccessively finer-grained knowledge ofdepartmental operations over a period ofyears, would make all employees morevaluable to the organization and wouldhelp prepare each one for career progress.

“Our Work Is Very Far Removed from the Front Line”

20


Among all of the reactions to GPRA andrelated performance-based initiatives, this isone of the most insidious and disappointing.The comment is almost always made as a pref-ace to explaining why a given agency or func-tion will not do anything different from, or inaddition to, what it already does. The impliedjustification is that the agency or function is soimportant, large, and complex that its leadersare staggering forward, under the weight ofenormous challenges, to preserve the republic.

There is truth in this response. But there also isignorance and sometimes arrogance. Whenasked what his function was doing in relation toperformance-oriented management, the follow-ing response was given by a high-ranking politi-cal appointee just prior to President Bush’sJanuary 2002 State of the Union Address:

Well, you know, OMB has a job todo. They represent what the Presidentsays, put together the budget, andfocus on the President’s priorities. Wehear what they’re saying. But a lot ofpeople don’t realize this agency is the(xth) largest government in the world.And we have some real and urgentproblems over here. We have somebig tickets from GAO. We have someproblems that are core problemsaffecting our core management. Mystrategy will be to focus on those, andget this place to where it runs. When Ido that, we should have a better shot

at getting things done, including thePresident’s priorities.

A compelling aspect of this vignette is that the official was a senior manager with greatresponsibility who was at the end of a longand distinguished career in the private sector.He had won many awards for his leadershipin performance-based management, as evi-denced by the many plaques displayed on his office wall.

Recommended Responses1. The President must effectively make and

keep performance-based management atop priority. The sentiment expressed in thestatement “We’re the third largest govern-ment in the world” is an arrogant way ofsaying, “Tell the President we’re plentybusy.” The only cure is for the President to say very clearly what he wants done,and then hold people accountable. ThePresident alone is responsible for the competence and fealty of his cabinet. Anunusual but possibly very effective additionto the President’s leadership strategy wouldbe for him to make unannounced calls onagency appointees at various levels. Thatcould be electric.

2. Top-level political appointees should—ontheir own and without any guidance orrequirement—be aware of the global trendtoward performance-based management.They should lead the charge to make the

United States the number one country andeconomy at the table of nations. In thisday and age, with performance-basedmanagement sweeping the globe, sup-ported by a worldwide ratcheting upwardof performance standards, each person,and certainly each executive, needs toclosely monitor emerging managementtrends. Those who don’t will be leftbehind.

“We’re the Third Largest Government in the World”

21


On its face, this issue may seem odd. After all,why would a staff person even think of pursu-ing something not on his or her supervisor’sagenda? Why would anyone offer guidance tosuch a person?

In fact, some of the earliest, purest, and mostactive support for GPRA and related perfor-mance-based initiatives came from early tomid career professionals: young people whoseenthusiasm and vision for public service arealive and well, whether they work on CapitolHill or in an executive branch agency.

This does not discount the support and majorcontributions from others, including cabinetsecretaries, deputy secretaries, assistant secre-taries, deputy assistant secretaries, chief finan-cial officers, senators, representatives, andothers.

All public servants who understand the impor-tance of performance-based public servicehave a responsibility to try to engage thosearound them, including their boss. But eachmust decide for himself what that engagementshould be. Everyone wants to be seen as doingwell, seeks a good year-end “performanceappraisal,” and wants to receive merited promotions.

The task is to identify some way of engagingone’s superiors in a useful way. Some ideas:

• Show how performance-based manage-ment and budgeting can help your unit.

• Build an information file (hard copy orelectronic) on related current events.

• Call on colleagues in other agencies andon the Hill to invite your supervisor tomeetings and events.

• Draw up a simple proposal showing howyour function fits into the organization’sperformance strategy.

• Ask your supervisor if you can serve ondepartmental committees working on thisissue as a career development activity andregularly report back to your office.

Recommended Responses1. Do something. Doing nothing despite

awareness of the increasing importance ofperformance-based management is todiminish your role as a public servant. Inother words, do something yourself. Dowhat you can do.

2. Try to find an acceptable way to showhow a performance-based approachwould improve the operations of whichyou are a part. Or, do what you can do tointerest and motivate those above you.

Note: In both of these recommendedresponses, the suggestion is to do what “youcan do.” Each person has to decide what thatis. Many factors will go into the decision. Noone will want to jeopardize his career oremployment. But somewhere between doingnothing and doing everything is something.

What that is will be different for each person.Doing nothing at all is inconsistent with thepurpose of public service and the role of indi-vidual public servants.

“My Boss Isn’t Interested”

22


Performance indicators seem like such a goodidea. The words “indicators” and “perfor-mance” can seem a little academic, and cer-tainly a notch more formal than “managementinformation systems,” but they intuitively areeasy to grasp.

Developing indicators is another story altogether.

“You mean for my program?”

Fear can take over, often unnecessarily inhibit-ing or complicating effective action.

A sense emerges that indicators must be devel-oped with indisputable scientific precision.Hours, days, weeks, and months can be spentin search of the perfect indicators. Emotionscan be high. Non-technical people begin totalk like professional methodologists. Notinfrequently, the dialogue turns to the issue ofcausality: “How can we be certain that thereis a direct, verifiable relationship between anoutcome and an indicator?”

Indicators proliferate as the quest for precisionintensifies. If the purpose is to make sure theelephant is walking in the right direction, weinstead end up defining every feature of theanimal itself and its environment.

All of this is not necessary. The task at hand ismore simple, less technical. Are we on or offcourse? If we can answer that question well,we will have accomplished much.

Designing Performance Indicators

23


Many, if not most, programs have a sense ofuniqueness. Part of this notion is related to anauthentic sense of value: “We do somethingimportant, and we do it well. You can’t fullyunderstand what we do because you do notknow much about it.” This probably is more afeeling of worth than uniqueness, and it ismore a good thing than bad, as long as rea-sonable effort is made to develop performanceindicators.

Yet some feel very strongly that their activitiesdo not lend themselves to being tracked withsuch indicators. The most often cited exampleis basic research, especially related to thehealth or physical sciences. A typical expres-sion of this view might be:

We’re trained scientists with years ofeducation and experience exploringthe frontiers of human knowledge. Weneed to keep exploring those frontiersfor the betterment of humanity and ournation. We have no idea how to focusour research to be performance based.How can we tell how well the trip isgoing when we don’t know where wewill end up? But we know that ourwork over time will lead to scientificbreakthroughs with great potential forimproving human experience.

This case can be powerful and apparentlycompelling. Nonetheless, it is rooted in thefundamental error of assuming that the task isto measure, not indicate. At some point in the

basic research process, some measurementmay be useful. This probably would require aformal evaluation involving significant timeand resources. Yet the task at hand is simpler.It is to indicate—to communicate to managersand other stakeholders—what is happeningwith public resources.

Basic measures of time, dollars, and units orstages of work can almost always be supplied.Indicators pointing toward outcomes can beprovided through regular narrative commen-tary. Semiannual or annual reports can gatherthis information to inform stakeholders and thepublic. Much research, even basic research,begins with a general approach that can berecorded and reported against. Approachesleading to dead ends can be valuable ininforming future work.

Recommended Responses1. All activities funded with public dollars

should provide indications (this term isdeliberately used to capture the basicintent) of how funds are spent on a regularbasis and in a systematic format. For everypublic dollar that is spent, we shouldknow two things: what has been accom-plished and what have we learned alongthe way about accomplishing that kind ofthing. This is our minimum level ofresponsibility—the basic starting point—for the management of public resources.To the extent that we cannot articulatethese things, we have not fulfilled ourmost fundamental mission.

2. Research and similar activities shoulddevelop a specific strategy to communi-cate to citizens what they do. Publictrusteeship and accountability do not endwhen the going gets tough; just becausean activity is difficult to measure, assess,or track does not mean we should give upcompletely. A better approach would be totry harder to communicate. The notionthat some topics are so complicated thatonly experts can understand them is atvariance with the most basic tenets of ourdemocratic political system. True enough,all citizens may not be trained scientists,but that is exactly the point.

“Our Program Is Different”

24


Perhaps there is no reaction related to perfor-mance-based management that is more preva-lent—or more frequently erroneous—than thisone. This reaction is frequently rooted in thesense that performance must be measured,and that measurement must be precise, thor-ough, reliable, and methodologically sound.

The first response to this claim is to rememberthat the requirement is for indicators, not mea-sures. Indicators can be less precise, less reli-able, and less costly than measures. This is notalways true, but they can be.

The second response is to note that Congressdid not appropriate funding for GPRA orrelated performance-based initiatives. Whateverit is that an agency does concerning perfor-mance indicators must be done within theexisting budget.

The agency must do the best it can with the resources at its disposal. This will meanreviewing existing data systems to see whichelements can be used, which can be modified,and which can be added—all within theresources available.

However, even the best efforts to use an exist-ing data set may not yield sufficient informa-tion. Here, the effort should be supplementedwith mail, phone, and in-person data gather-ing based on samples. If properly designed,very small samples can yield valid and usefulinformation. If there is insufficient funding foreven this modest effort, agencies should use

existing communication channels, such as routine phone conversations between centraloffices and the field, and procedures for mak-ing written notations in a standardized format.

When agencies report these methods toCongress, they have the opportunity to pointout how they used existing resources and whatadditional benefits could be gained from alarger appropriation.

Recommended Responses1. An agency should rarely be comfortable

saying that it cannot proceed with devel-oping performance indicators because itdoes not have the data. This is an unusualthing to say. And it is intended. The pointis that when an agency begins to thinkthat it does not have the data, the thoughtitself gathers momentum and official sanc-tion as it is repeated. In turn, this some-times stalls progress as everyone throwsup their hands and no one steps forwardto take responsibility for next steps. Notinfrequently this sentiment begins with aremark made by someone who does notreally have the expertise to make it, butbecause that person has some relatedtechnical responsibility in relation to datait is taken as fact.

2. Agencies should mine existing data sys-tems and supplement them with low-cost,sample-based data collection procedures.This is a creative exercise that probably

will need to be led, or supported withexpertise, outside the normal departmentalchannels of database operations. It mayrequire using such things as surrogateindicators—pieces of information that areintended specifically to track one thing butwhich also indicate related things. It mayrequire combining data elements or datasystems. Just going to the data systems andlisting the data elements—which is themost frequent approach to the search forusable information—is a very narrow andunimaginative approach.

“We Don’t Have the Data”

25


ExcerptManagers are sometimes reluctant to use performance data because of the fear that it will be usedagainst them and their programs. One of the biggest pitfalls in the use of performance data is using thisdata prematurely in the performance appraisal system. Positive reinforcement is the best approach to getting a buy-in by employees to new approaches.

Another major pitfall is to set expectations too high, too soon, or both.

Other pitfalls and ways to avoid them … include the following:

• To avoid setting expectations too high, too soon, develop a longer-term, staged approach.

• To avoid the piecemeal approach, agencies should think systematically and implement the newapproach gradually.

• To avoid premature and incorrect use of data on results, establish valid performance goals andensure that the data collected is accurate and properly analyzed.

• To employ new business processes that are untried, embed them in the framework that alreadyexists.

• To avoid a headquarters orientation, ensure that results data are included at the point of servicedelivery.

• To avoid a last-minute implementation frenzy, change the whole orientation of management.

One of the keys to overcoming pitfalls of developing and using performance data and incorporating it inmanagement by results is to involve customers and employees in the whole process. Unless “knowledgeworkers” and stakeholders are involved, neither will be committed to the whole process of goal setting,gathering performance data, and using this data to improve results.

Center for Improving Government Performance, Questions and Answers for Improving Government Performance:Using Performance Data to Improve Program Effectiveness (Washington, D.C.: National Academy of PublicAdministration), pp. 6-7.

26


Discussions about performance measuresoften get bogged down in issues concerningempirical or statistical validity. Are particularmeasures valid pieces of information or validstatistical measures? Sometimes these discus-sions are complicated by methodology expertswho themselves may hold different views.

This debate is unnecessary when one recallsthat GPRA’s basic purpose is to help managersto manage, not researchers to research or eval-uators to evaluate. The fundamental test iswhether the indicator selected is useful to theprogram manager.

Fundamentally, the task is to find performanceindicators rather than performance measures.The former should be few in number. Theyresemble channel markers for boaters in thatthey help to establish key points along thechannel path, rather than mark every point ina continuous, uninterrupted line. Therefore,the test of validity is whether information tellsmanagers whether they are on or off course.

It is possible, and sometimes desirable, that aparticular piece of information is an “indica-tor” as well as a valid “measure.” It is not nec-essary for it to be so, however.

Let us imagine a federal government agencytrying to monitor customer satisfactionbetween its state-based field offices and citi-zens. A valid measure could be derived from asurvey based on a statistically valid methodol-ogy. A useful indicator, however, could be

based on a small number of informal phonecalls to field offices.

Recommended Responses1. Never lose sight of the fact that GPRA’s

primary goal is to improve managementand that the law envisions program man-agers in a central role. Its intent is to helpprogram managers improve program per-formance by giving them a primary role indeciding what pieces of information—indicators of performance—they need tomanage their programs. It is helpful toremember that GPRA was enacted follow-ing a decade-long period of budget cuttingduring which management capacities forsuch things as management informationsystems, policy and program analysis, andprogram evaluation were substantiallyreduced. GPRA says that information onperformance is legitimate and important,and gives program managers a key role indetermining what information is needed.

2. Remember that GPRA requires indicators,not measures. If the primary purpose of thelaw were to measure performance, muchmore attention would have been given inthe legislation to criteria for measurement.There would be discussions of such thingsas sample sizes, confidence levels, andmeasures of association. None of these isin the law. Also, it is very likely that therewould have been a specific appropriationof funds to support training in these areas.

“We Can’t Find a Valid Measure of Performance”

27


This issue arises in many guises: “We have nocontrol over our outcomes.” “We have noauthority to track outcomes.” “Our outcomesare not easily measured.” “The time frame forour outcomes is too long.” “Our outcomes aretoo abstract.”

At the core of these statements is the notionthat the outcome is beyond the manager’sreach and responsibility: “I can’t (measure)reach it, so I can’t track it.” Or, more bluntly:“I can’t (measure) reach it, so it’s not myresponsibility.” Even more bluntly: “I can’t(measure) reach it, so I can’t be held account-able for it.”

This is not a measurement issue, but it is anaccountability issue. Public servants who holdor dispense public resources do so in a con-text of trust. In this regard, they are trustees onbehalf of their fellow citizens.

If managers fear accountability and do notmake a good faith effort to track programresults, they betray the public trust and fail tomeet its most fundamental requirement. At aminimum, the guideline for this trusteeshipshould be what can be done to responsiblyaccount for the spending of this money, notwhat cannot.

The issue of precision is subordinate to theissue of accountability. The degree of precisionthat can be achieved is a matter of methodsand resources. Methods, in turn, can be pro-vided with expert assistance. Resources are apolicy matter.

In more basic terms, the public servant whodispenses public resources must develop anddeliberately pursue a heightened sense ofawareness about the effects that occur relatedto those resources.

Recommended Responses1. Those managing public resources need to

make a good faith effort to be aware of theresults of public expenditures. This is afundamental aspect of the public trustwith which they are vested. Having littleor no control over outcomes or having lit-tle or no legal or regulatory authority fortracking outcomes is an insufficient reasonfor 100 percent ignorance about whathappens with resources once they arepassed on to another entity. The languagehere is deliberately colorful. The issue is:what can be done, what can be known,and how it can be used to inform theprocess.

2. Managers who do not know how to trackoutcomes should make it known that theyneed help. This should be done on therecord. It may be that a program managerwill need expert assistance. But that can-not even be considered until someoneraises the issue. Tracking outcomes can bea challenge requiring the creative—andsometimes even unorthodox—applicationof available methodologies. If the decisionis made not to seek expert help, then atleast the program manager has taken theright steps.

3. For programs with difficult-to-measure endoutcomes, managers should use the tech-nique of measuring intermediate outcomes.Intermediate outcomes show reasonableprogress toward end outcomes. While perhaps not every end outcome can bemeasured, the point is that what is possi-ble can and should be done. The conceptof intermediate outcomes should substan-tially remove the “our outcomes cannotbe measured” response, because it is hard to imagine a program that does nothave an intermediate outcome that can be tracked.

“We Have No Control Over Outcomes”

28


The terms “crosscutting measures” and “cross-cutting programs” have been used loosely tomean similar measures for similar programs.The intent is to gather together similar pro-grams so that the total resources and effectscan be presented to the public for review andscrutiny. This is a fundamental, irreducible stepfor providing true accounting to citizens fortheir tax dollars.

At the same time, there is another effect ofbringing together similar programs in a com-mon measurement frame. The comparisoncould reveal duplication, overlap, lack ofcoordination, or a recognition that a program’sfunding should be reduced or its life ended.From the standpoint of managers responsiblefor similar programs, there is another adverseeffect of gathering together similar programs:being misunderstood, mischaracterized, mis-represented, and just plain not appreciated.

There only is one way to handle these feelings:Come and sit at the table of similar programsand make the case, whatever it is. This is theright thing to do as a public servant. In today’sincreasingly performance-oriented world, italso offers the best chance for survival.

Congress enacts programs that citizens want.Yet these programs are rarely enacted in acoordinated fashion so that they all fit per-fectly together. No one should expect perfec-tion. Nonetheless, enacted programs andresponsible public service require sensibleimplementation, taking into account othersimilar programs.

Recommended Responses1. Every program manager should take the

initiative to identify similar programs andto develop an understanding of how theyare related. This seems to be only commonsense. If you are working in a given area, itwould seem natural to know of other pro-gram areas doing similar things. And,many are aware of other similar programs.Unfortunately, this awareness is often ofthe competitive or bureaucratic variety. Inthe coming world of performance-basedmanagement, managers will need to moni-tor similar and related programs to stayabreast of lessons learned, emerging prac-tice, helpful innovations, and comparativeperformance, or they will risk fallingbehind and not knowing they are behindor, worse, falling behind and not knowingthey are behind while others do.

2. Every program manager should be pro-active in staying current with emergingpractices in related program areas. Thiswould include seeking and reading thelegislation and reports on related pro-grams; staying abreast of current budget,congressional, and press dialogue; know-ing the structures, processes, outcomes,and indicators of similar activities; goingto conferences and workshops; and takingthe initiative in building relationships withthe managers of related programs. It islikely that building relationships will beappealing to a relatively small number ofindividuals, but it is also likely that those

relationships may provide opportunitiesfor mutual benefit that are not easily avail-able to those who do not have them.

“No One Wants to Do Crosscutting Measures”

29


ExcerptThe examination of recent experience … suggests that careful thought should be given to how andwhere to begin activity to improve performance among related programs. Since the Results Act does notspecifically require the kind of performance measurement in related areas that is the focus of current dis-cussion, it does not provide explicit guidelines about where and how such activity should begin.

A good place to start is within an individual agency.…

However, once a beginning has been made within an individual agency, attention should turn to pro-grams in related areas outside the agency. In fact, taking initial steps within an agency may suggeststeps that should be taken to address related programs outside the agency.…

Progress will relate to the strategy chosen. Each agency will need to decide the best course of action,taking into account likely costs and potential benefits.

Listing and describing related programs is a useful way to begin gaining a sense of what may beinvolved in attempting to develop measures for related programs.

Programs can be characterized in many ways: by goals, strategies, administrative structures, and targetpopulations; by whether they are large or small; by the level of government and type of agencythrough which their services are delivered; and by their policy, political, and practical contexts.Agencies can assign weights to these (and other) relevant factors.

A focus on goals is a useful and practical place to begin. Both OMB and elements of Congress haveindicated that initial attention to the relationship among programs should begin at the broadest andhighest level….

At this relatively general level of aggregation, goals provide an obvious first step for looking at programsin relation to each other. If every federal program could explain its relationship to other federal pro-grams that are grouped under the same goals in the President’s budget and could identify and demon-strate its contribution to that goal through specific performance measures—much would beaccomplished.

Such activity should be initiated first in regard to related programs within an agency and then proceedto related programs outside the agency. When addressing related programs outside the agency, anattempt should be made to coordinate with all agencies involved; and, if possible, work should pro-ceed on a joint basis and lead to a joint agreement on the measures to be used. If it is not possible toconduct a joint effort or to reach agreement, then agencies should proceed on their own to propose aninitial set of measures for related programs.

Center for Improving Government Performance, Helpful Practices in Improving Government Performance(Washington, D.C.: National Academy of Public Administration, 1998), pp. B-10, B-11.

30


There is concern that most reporting againstintended objectives takes place annually inconnection with the annual federal budgetcycle and GPRA’s required progress report.However, some program outcomes may takeyears, even decades, to accomplish.

The answer seems obvious. Indicate the over-all expected time frame across multiple yearsand tell what progress has been made in aspecific year. Indeed, this provides a perfectopportunity for program managers to useGPRA’s strategic planning component to pre-pare interested stakeholders for the roadahead, especially if it is a long one. GPRArequires a five-year strategic plan, which canbe updated every three years. A longer lookahead can provide the context within whichstakeholders can be educated about thelonger-range horizon of multi-year outcomes,possible contingencies, intervening variables,and possible scenarios.

At the same time, tracking progress can bebroken into annual increments. These may bedimensions of time, quantity, or quality. Theyalso may include narrative explanatory com-ments. Further, nothing prohibits a managerfrom recapping the longer-term strategic planin GPRA’s two other required documents: theannual plan and the annual report.

In regard to multi-year outcomes, the generalrule should be to give a progress report to citi-zens on what has been accomplished annuallyand also a progress report against the multi-year target.

Recommended Responses1. Provide an overall estimate of the time

needed to affect the outcome. Update thisestimate as appropriate. Use this as anopportunity to communicate a rich senseof program substance and context to stake-holders. Think in terms of explaining andmaking clear what it takes to deliver theend outcome. Draw into the narrative thework of related and supporting activities asa way of building improved relationshipswith colleagues. Consider explaining therole of Congress, OMB, and changingadministrations. In other words, take aproactive position. If you were a stake-holder interested in this program, and theprogram took a number of years to accom-plish, wouldn’t you be interested in know-ing how your tax dollars are being used tocome to grips with these issues?

2. Provide an annual progress report againstthe multi-year plan, and use performanceindicators that provide significant mile-stones along the way. Ditto all of the com-ments made above. An annual progressreport can do all of these things, and at the same time provide the considerableadvantage of allowing you to report inmanageable bites. Not the least of theopportunities provided by annual reportingagainst a longer-term plan is to prepare theway for needed revisions in the longer-term plan. These may be much more palat-able if the context for the changes iscommunicated to stakeholders in advance.

“Our Outcomes Take Years to Achieve”

31


This statement is a variation on the “We don’thave the data” issue. It often is made with greatfeeling, as if the problem were so enormous,complicated, expensive, frustrating, or intractablethat all hope for improvement is gone.

The point is not to make fun of such expres-sions, but to suggest that an errant data systemcan generate a culture of hopelessness thatreally is not necessary. In fact, a less than per-fect data system may have some advantages,chief among them that a system, or some sem-blance of one, exists. Money has been allo-cated, some collection system has been put inplace, and some data are being generated.Even if the system or the data are fatallyflawed, they still may offer a starting point thatis one step ahead of zero.

The answer is to design a sample of the uni-verse in question. Do it without regard to theexisting data system so you will know that thesample design is correct. Once the sample hasbeen designed, look for potential overlap withthe existing system, allowing at least part of thedata to be collected from within it. The remain-ing sample data must be collected de novo,but there are inexpensive ways to do this, suchas mail, phone, and in-person surveys.

Once data are collected from both sources,the sample data from the bad system can bechecked for accuracy. This may involve furtherdata collection and some corrections to exist-ing data. The point here is that when there are“no data” or “bad data,” the cheapest andquickest approach may be to design a sampleand extrapolate findings to the universe.

Recommended Responses1. Recognize that an existing data collection

system may offer a partial route, at a mini-mum, for collecting of some of the neededdata, possibly at reduced cost. Look at theparts of the existing data system to seewhat can be useful in obtaining the datayou need. Here are some questions thatmay help: How is the data collected? Howis it recorded? Is it subjected to any trans-formation process along the way, such asa coding procedure to transform narrativedata into countable categories? How is thedata manipulated (retrieved, analyzed,and arrayed)?

Each of these steps may offer an opportu-nity or suggest ideas for using parts of anexisting data set by selecting, adding, ormodifying data elements. The existingcapacity for managing the data system,whether it is in house or under contract,can provide expertise in making theseadaptations.

2. Design a sample—based on an existingdata collection system, new collectionprocedures, or a combination of the two—and extrapolate to the universe. It isalmost always cheaper and easier to use asample than it is to collect data from anentire universe. In fact, it would be a goodprocedure, if circumstances allow, to thinkin terms of starting out with a sample first,in order to see if it will do the job, beforegoing to the more expensive option ofbuilding a complete database. Using a

sample also offers the advantage of beingeasier and less costly to change. If a uni-form sample is needed over time, multiplesamples can be used. For example, onesample could be maintained pretty muchwithout change to provide a good basisfor multi-year comparisons or tracking,while another sample could be changedfrom year to year to capture emergingissues. Two 10 percent samples willalmost always be cheaper than one uni-verse data collection.

“Our Data Are a Mess”

ExcerptDon’t overdo precision requirements. Usesmaller samples. Many evaluation and surveyprofessionals press for high levels of precisionsuch as 95-percent (or higher) confidencelevels (which means that the sample findingsthat would occur in 95 out of 100 samplesdrawn from the population would be withinthe number of percentage points indicated bythe statistics). The higher this confidence levelis set (such as 95 percent), the larger thesample size needed and the more expensiveis the data collection effort. When outcomemeasurement resources are tight, programsshould consider lower confidence levels....For many, if not most, outcome indicators, a90-percent confidence level will be fully ade-quate (and likely to be more accurate thanmany of the other pieces of information thatmanagers use to make decisions).

Harry Hatry and Shelli Rossman. Obtaining Timely, Cost-Effective, and Useful Outcome Data(Washington, D.C.: National Academy of PublicAdministration, June 2000), p. 4.

32


For a variety of programs, all or some of the dataare in narrative form. This kind of informationis seen as being unstructured and not usable.

For example, the Community DevelopmentBlock Grant Program in the U.S. Departmentof Housing and Urban Development, whichdistributes money by formula to cities andstates for such things as housing rehabilitationand street improvements, requires a very shortreport on how the money is used. Much of theinformation is in narrative form. This minimallevel of reporting is a part of the “block grant”approach, intended to streamline granteeadministration.

The reaction to this type of narrative informa-tion base is that it is not “objective” or “system-atic,” and that it prohibits the aggregation anddisplay of countable units of accomplishment.

It is possible to overcome these issues byemploying a methodology to transform the nar-rative data into specific categories of accom-plishment and count the units in each category.

The Department of Housing and UrbanDevelopment did this with a coding system fortransforming narrative data into countableunits and objective categories. The system wasdetailed in a manual consisting of sequentialdecision rules for placing narrative statementsinto objective categories. This was, of course,a highly subjective process, yet it allowed aninitial array of program accomplishments.

Once this stage was completed, results werediscussed in detail with each grantee, esti-mates were fine-tuned, and analysis resultswere weighted by the sample design andextrapolated to the universe of grantees. Theend result was an annual report that presentedactual accomplishments in countable form forthe entire program at a surprisingly small cost.

Recommended Responses1. Start with a sample. If all, most, or a sig-

nificant part of the data are in narrativeform, one of the first things that should beconsidered is constructing a representativesample of the data universe. The benefit ofthis approach is that it reduces the size ofthe data universe that has to be dealt withand, therefore, reduces the cost of datacollection and transformation. Constructingthe sample will very likely require engag-ing an independent professional with theappropriate expertise. And even if it doesnot seem as if a professional is needed, itcan be a good idea to engage one, as thisprovides the added credibility of an out-side professional opinion. The sampledesign will require an upfront cost, but itcan be amortized over time.

2. Develop a process for transforming narra-tive or subjective information into cate-gories, and extrapolate to the universe ofgrantees. Here, too, it is likely that the ser-vices of an independent professional will

prove helpful. The process of transformingnarrative or subjective data into standardor objective categories will be new tomany people, and there will undoubtedlybe a need for assistance in conceptualiz-ing the transformation process as well as alearning curve for in-house staff. Basicallythis approach leads to the development ofa collection of decision rules, collected ina coding manual, that guides the transfor-mation of unstructured data into struc-tured form. The end result is units ofcountable product.

“Our Data Are in Narrative Form”

33


For much of the past decade, the concept ofalignment has been intermittently and vari-ously used without broad common under-standing or application.

First there was much discussion about theneed for “logic models”—“logic diagrams”may be a more descriptive term—to mark the pathway from inputs to internal processesand ultimately to outputs and outcomes. Later,alignment sometimes was used to describe the relationship among required elements ofGPRA: strategic plans, annual plans, andannual reports. Later still, attention shifted toaligning accounting structures with perfor-mance plans. Occasionally, someone wouldmake the point that whatever planning wasdone needed to be aligned with an agency’soperating plan.

In retrospect, these early formulations can be seen as individual parts of a process not yet fully articulated. Placing them together in a performance-based accountability chaincomes close to understanding the meaning of the term today.

Alignment should mean that every processand resource is properly sequenced undereach major organization goal to achieve thedesired result. Further, there should be anorganization-wide sense of responsibility formaking the whole greater than the sum of theparts.

Aligning Performance Processes

34


Fill in the blank. How many times has anassistant secretary or deputy secretary heardthis response from a senior manager assignedresponsibility to carry out a task? There arevariations: “They’re having a problem in OGC[Office of General Counsel].” “CongressionalRelations said they needed to work on somewrinkles.” “Admin hasn’t processed the paperwork.”

Whatever.

The point is that a senior manager assignedresponsibility for a task is passing the buck. Ofcourse, there are legitimate situations wherethe official can be caught by surprise by someuntoward circumstance. Yet there are other sit-uations where the accountable manager cananticipate the problem and take proactivesteps to keep an operation on schedule.

In either case, the prerequisite for taking aproactive approach is accepting responsibilityfor the goal, and monitoring preceding or par-allel steps that must be accomplished.

The bottom line of the modern trend towardperformance-based management is that it is nolonger acceptable in today’s environment for amanager to point to a prior or related manage-ment step and simply report that the process is“off track.” Everyone in the line of account-ability bears full responsibility for a specificpart of the process and for the entire processleading up to the accomplishment of a goal.

Recommended Responses1. Managers must understand pre, parallel,

and post steps in the process of whichthey are a part. One way to approach thistask is to envision oneself as the managerin charge of the entire process. The firstthing this will require is an understandingof the various parts of the process andtheir contexts. Other activities—perhapsvery far outside one’s own area of experi-ence and expertise—will have to be stud-ied. Think of it as a learning venture thatwill take time. Study similar operations indifferent departments and levels of govern-ment. Set a goal of becoming an expert inthe delivery of the service activity ofwhich you are a part. Read. Go to confer-ences and workshops. Pick up the phoneand call experts in the field. You may besurprised to find how willing many ofthese people are to answer the questionsof those who are earnestly seeking toimprove program delivery or management.

2. Each manager who is a part of a processmust accept some level of proactiveresponsibility for the process as a whole,as well as the specific part. Once youhave mastered the overall process or activ-ity of which you are a part, look for waysto improve overall delivery. Cultivate newviews of old processes. Look for ideas thatcan be transferred from one place toanother. Be the person who transports

them. Offer to help other parts of theprocess. Approach the managers of relatedprocesses to see if they are interested inmeeting on a regular basis to exchangeinformation and ideas. Not everyone willrespond. You will have to be tactful andnon-threatening. But by taking these steps,you identify yourself as a leader, regard-less of your rank or seniority.

“Well, We’re Waiting on ______”

35


Managers making this statement reveal a senseof isolation—of being out of touch with thelarger awareness and responsibility that man-agers increasingly are expected to assume intoday’s world.

An experience reported by the UkrainianFinance Ministry is a useful, if extreme, illus-tration of the kind of isolation that can beexperienced in a large public bureaucracy.

An American expert in performance-basedmanagement was hired as a consultant toimprove the performance orientation of theUkrainian Finance Ministry. The individual wasamazed to find that each civil servant workedin almost complete isolation. Each day, the civilservant came into a very small office, sat at adesk, and went through the inbox, one item ata time in the order in which they occurred,made a notation on each, and deposited themin the outbox. At the end of the day, a clerk col-lected and distributed the items. One year toanother, there were never any meetings, andthere was very little sense of accountability.

This is an extreme example. However, if thiscase had not been introduced as an examplefrom the Ukrainian government, could it nothave passed for an example of behavior forsome parts of the U.S. government?

This anecdote provides an image of the indi-vidual manager as an “island.” It is an effectivebackdrop against which to consider theenlarged sense of responsibility emerging inrelation to performance-based management.Such islands exist in the U.S. government. We

may be better, but we are not perfect. More tothe point, we are not as good as we could be.

Today’s smart managers will master the entireprocess of which they are a part, and developa reputation among their colleagues foradvancing the success of the entire enterprise.

Recommended Responses1. This kind of attitude will get you into trou-

ble in the coming world of performance-based management. It will immediatelyidentify you as an anachronism, if not anobstruction. Yesterday, this kind of state-ment could have been taken as a fact.Tomorrow, it will be seen as an attitude.The solution: Change your attitude.Broaden your viewpoint. Enlarge yoursense of responsibility. Become moreproactive. And develop an understandingof the entire management process ofwhich you are a part.

2. Develop a specific strategy for supportingthe success of the entire enterprise, as wellas your own specific part. This will be yourown individual strategic plan—your ownroad map for a successful career as a fullcontributing member of your department.Use the strategic planning process to iden-tify where you are in relation to the depart-mental mission, where you would like tobe at the end of your career, and what arethe steps that must be taken to get fromone place to another, and then updateyour overall plan periodically.

“Someone Else Does That”

ExcerptUnless all key players participate in setting thestrategic agenda from the outset, it is unlikelythat the organization will harness all of its capa-bilities to achieve strategic results. Moreover,the process of cascading these priorities andplanned actions down and across the organiza-tion will be handicapped by the lack of owner-ship by key components and individualemployees....

Cascading and communicating strategic require-ments may be done function-by-function oragencywide. But “whether the organizationchooses to cascade functionally or cross-functionally, the objective is the same: Commu-nicate the strategic requirements, but not howthe gaps must be closed. Employee teamsdesign their own plans.” In this process it isimportant to communicate not only what eachprogram component has selected as a strategicemphasis—so that support organizations canalign their plans accordingly—but the strategiesof related support functions as well. This stepwill allow each support function to link to relat-ed support efforts aimed at the same missiongoal. It will also lead to increased employeeownership of the plan.

Center for Improving Government Performance,Helpful Practices in Improving GovernmentPerformance: Linking Administrative Support toStrategic Planning (Washington, D.C.: NationalAcademy of Public Administration), pp. 5, 8.

36


There was a time, not too long ago, whenthere was a long dialogue among GPRAimplementers concerning what might becalled the “messy,” “complicated,” and some-times “undecipherable” nature of some agencybudget accounts.

The question was: “How can we do perfor-mance-based budgeting when our accountstructure is a mess?” The word “alignment”was sometimes used to indicate the desiredstate—a budget aligned with agency perfor-mance goals, where units of cost and resultscould be linked.

Beyond the untidy nature of agency budgetaccounts, there seemed to be a general reluc-tance on the part of everyone—agencies,OMB, and congressional appropriations committees—to depart from past accountingpractices. It was occasionally said and oftenthought, “If the system is bad, at least we knowwhere we can find the money we care about.”

Today, there seems to be a growing under-standing that the first and most important stepin performance-based budgeting is understand-ing and clarifying the budget process and therelationship between costs and desired results.It is not changing the account structure.

Most fundamentally, this implies informed dis-cussion at every level of the agency budgetdevelopment process. In other words, perfor-mance-based budgeting does not depend onaligned accounts but aligned analysis and dis-cussion. To the extent that this discussion actu-

ally is presented in the budget, it can providea summary of the discussion and reference thedata and analysis on which it was based.

A reasonably aligned account structure is adesirable, but longer-term, goal.

Recommended Responses1. Do something to align costs with activities

now, right away, for this year’s budget andthe next. It’s the right thing to do. Weshould do everything possible to relatecosts and activities. We owe it to citizens.And OMB has made it a top priority. Whatthat something is will likely vary with theagency or program context. But it is verylikely to center primarily on the word“information.” If the issue is the relation-ship between costs and activities, thenwhat do we know about both, and howcan that be factored into the budgetprocess? Likely this will mean simplybringing to the table all available informa-tion at each stage of the budget process,leading up to some kind of short narrativesummary to accompany each budgetrequest. In essence, gather available infor-mation, talk, think, and explain.

2. Start the longer-term process of revisingthe accounting structure so that costs canbe aligned with activities. This will involvea substantial investment of time andresources, but in many cases will be foundnecessary. In many agencies the account-

ing structure may be likened to the growthof barnacles—an accretion of individualencrustations resulting from legislationand earmarks emerging from a collectionof periods, contexts, issues, political par-ties, and individual personalities. What weneed is an accounting process and struc-ture that facilitates an ongoing alignmentprocess.

“Our Budget Account Structure Is a Mess”

37


ExcerptOne of the major challenges facing agencies as they implement the Government Performance andResults Act is aligning performance-based accountability structures, as envisioned by the Results Act,with existing budget account structures. The objective is to assist agencies in modifying their budgetstructures to achieve effective alignment with performance-based, results-oriented management.

… Agency budget systems established to serve one set of purposes and accounts have been adaptedmany times to serve many additional purposes and accounts. The resulting collection of accounts thatmake up existing budget structures represents a major challenge to those seeking performance-basedmanagement....

Helpful practices that have emerged from current experiences of agencies seeking to achieve betteralignment of resources and goals include the following:

• Identify the major program areas that accomplish the agency’s mission, goals, and objectives.

• Take the initiative at the career level to design and implement an aligned budget.

• Define a budget account structure that aligns the budget accounts and program activities with program goals.

• Define the manner for distributing costs, direct and indirect, to the proposed accounts.

• Manage the buy-in process within the agency, with OMB, and with congressional committees.

• Submit the aligned budget request in aligned format, and with a crosswalk to the existing format.

• Account for the expenditure of funds in the same structure used for the budget request.

• Issue a financial report presenting actual expenditures in conformity with the aligned budget.

As they take these steps, agencies will have to consider factors relating to strategy (some accounts orall), timing (now or later), potential barriers (organizational, management, cultural, political), likely costs(systems development, installation, operation), and possible benefits (better, cheaper, more accountableservice).

No matter how well designed, budget accounts cannot serve all purposes over all periods of time. Aswith other aspects of management, budget alignment must be continuously improved.

Center for Improving Government Performance, Helpful Practices in Improving Government Performance: An Overview of Helpful Practices (Washington, D.C.: National Academy of Public Administration), pp. 7-8.

38


This statement frequently is made as a defen-sive response to a question about a deadlinethat either has been missed or is likely to bemissed. The answer may not be made in a“defensive” way. In fact, it may be offered in avery reasonable manner. If some untoward cir-cumstance has occurred, or if some other indi-viduals have not done their tasks, what is oneto do?

Nonetheless, the individual gives away anobvious state of mind by responding with theproblem rather than the solution.

Can we imagine the commander of a battle-engaged military unit responding this way?

What is the difference? Is it not that the militarycommander assumes a higher level of responsi-bility or has a greater sense of urgency?

Is it unreasonable to expect that a managerengaged in a non-military operation will pro-ject some sense of urgency about transactingpublic business?

Two key questions must be answered: First,who is responsible for solving the problem?Second, what should be done? The answer tothe former is anyone who becomes aware ofthe problem or whose operations may beaffected by it. The answer to the latter is what-ever may be possible and useful. Someoneindirectly connected may only be able to alertothers; someone with more direct responsibil-ity can do more.

To cite a problem and claim “We have no con-trol over that” delays the application of possi-ble solutions, ignores the potential synergygained by going to the source and offering tohelp, adds an additional cost to the process,and possibly jeopardizes the outcome.

Recommended Responses1. Try to develop your own radar for tracking

the progress of steps in the managementprocess of which you are a part that areeither prior or subsequent to your own.Add to your daily “to do” list a moment’sreflection on the overall process andwhere it is. Track newspaper, journal, andelectronic-based news sources regardingprogram activities. Perhaps make a phonecall or two to check on the progress of keysteps.

2. Do what you can to support steps in themanagement process prior to or subse-quent to your own. If a problem has arisenor seems likely to arise in a prior or subse-quent step, go to the area, raise the issue,and offer to help. Be proactive in keepingthose involved with steps subsequent toyour own informed of emerging issues orpotential delays. If the process arrives atyour doorstep behind schedule or withoutstanding issues, try to be a part of themid-course correction that will have to bemade.

“We Have No Control Over That”

39


Performance-based budgeting raises the issueof unit cost, the cost of some unit of produc-tion—whether building housing units, stagingtraining sessions, or maintaining federal high-ways. Frequently, such data do not exist or arenot readily usable in the desired format.

It also raises the issue of overhead, as in, “Wecan get some numbers on unit cost, but wehave no way to include overhead expenses.”“Overhead” can be defined many ways, but itgenerally means some combination of retire-ment benefits, health insurance, and anagency’s administrative operations, such ascontract management, policy analysis, andlegal expertise.

Both issues (unit cost and overhead attribution)have merit. But neither should be an obstacleto projecting unit cost.

How can this be? To not answer this questionis to ignore a fundamental aspect of the publictrust vested in public servants. How can wenot know the cost attributes of services pro-vided by citizens with their tax dollars?

So, what is the answer? Estimate. Gather avail-able information and come up with an esti-mate. What are total agency outlays forretirement, and how do they translate into perperson, per hour figures? Ask the same ques-tions for health care. How much time do indi-viduals in the Office of General Counselspend on your program? What are their gradeand salary levels? Ask the same about otheroverhead functions.

These estimates may be rough, but they indi-cate a willingness to accept responsibility forcost issues. If someone objects to these esti-mates on the grounds that they are not “accu-rate,” there are four possible outcomes that arebetter than doing nothing: The estimates canstimulate dialogue, or they can be accepted,improved, or rejected. Even if they arerejected, you have raised the issue.

For help in making these estimates, seek pro-fessional advice. There are methods of estimat-ing costs similar to those used in employingsurvey research samples. In other words, youmay not need a full unit-cost system todevelop useful estimates.

Recommended Responses1. Devise an estimate of costs and include an

explanation of how it was established. Themost important part of this exercise is theexplanation of how it was done. There aremany ways to estimates costs, even wherevery little data are available. No twoapproaches to estimation are likely to bethe same. And no two reactions to themethods are likely to be the same. Themost you will be able to do is to chooseone method and explain why. Someonemay disagree with your opinion. But thatis perfectly legitimate and will only lead tofurther discussion and refinement of themethod. The absolute best way to deviseand estimate is to obtain prior agreementon the estimation methods, but in practice

post—rather than prior—agreement maybe the better approach. In other words,take your best shot and express a willing-ness to revise. This has the advantage ofhelping you to get your estimates to thetable much faster than if prior agreementwere sought in all areas.

2. Become knowledgeable about cost-estimation methods. Learn the field, as itwere, in your area. Build a portfolio ofestimation methods from similar programsand activities. How do others estimatecosts in these areas? What are the advan-tages and disadvantages of each? What arethe lessons learned? Where are the bestpractices? There is a vocabulary to belearned. But mastering it will be wellworth the effort.

“We Don’t Have Unit-Cost Data”

40


Be careful. While this may be truer now thanit ever has been, making the statement intoday’s environment can have harmful reper-cussions. It reveals a certain way of looking atthe issue of performance-based budgeting, onethat has lost currency.

In the recent past, there was a sense that truebudget alignment required revamping out-moded, complicated, and misleading agencyaccount structures. It was clear that suchrestructuring was supported weakly by almosteveryone (OMB, agencies, and the Hill). Asbad as the account structures were, everyoneknew where the money could be found. Noone wanted to lose track of “his” or “her”money.

Today’s dialogue about performance-basedbudgeting concerns the concept of unit cost,evidenced in information presented for thebudget process and recorded in the budgetdocument.

What does this mean? Attention is focusedsquarely on the relationship between costsand results, rather than on the budget structureitself. There seems to be an unspoken sensethat structural alignment, while needed, is alonger-term goal. The immediate need is toencourage dialogue using available informa-tion about unit cost.

This approach seems based on the conclusionthat budgeting’s true purpose is to allocateresources to intended goals, gather informa-tion on what happens, and use this informa-

tion to revise future allocations. In effect, thedialogue has shifted from a desired end state(account restructuring) to the more basic issueof gathering, considering, and presentingwhatever information is available in support of the budget.

OMB’s initial budget for fiscal year 2002makes it very clear that a successful budgetprocess begins with the presentation of infor-mation on costs compared to results.

Recommended Responses1. Agencies should define the questions to

be answered for each major account line,present answers from existing sources, andlist questions that cannot be answered.Note that most of what is suggested hereconcerns information. The questions couldbe rephrased this way: What informationdo we need? What information do wehave? And, how can we get what we don’thave? Note also that before these are bud-get questions, they are management ques-tions. They only become budget questionswhen resource levels are tied to levels ofperformance.

The first question can be answered simplyby thinking; no data is necessary to definethe kinds of questions that should beanswered. Everyone can take this step. Thesecond question can also be answered bygathering together what is known aboutthe program. This information may come

from a database, an evaluation, a manage-ment review, or an Inspector General’s(IG’s) report. The information may not beperfect. But it is the place to start.

2. For those questions that cannot beanswered adequately from the existinginformation, or for those questions thatcannot be answered at all, agenciesshould state what they will do in the yearahead to gather the needed information.Since it is likely that additional resourceswill be needed for this, at least two andpreferably three options with different costlevels should be projected. The advantagesand disadvantages of each should be pre-sented. One of these cost levels should beprojected at a modest level and involveinexpensive data collection through theuse of sample-based surveys.

“Everyone Wants the Existing Account Structure”

41


This refers to GPRA’s requirement for anannual plan, the purpose of which is to lay outthe goals, objectives, and performance indica-tors against which progress can be summa-rized in the annual report.

The claim made here is that the GPRA plan isonly a paper exercise. The required GPRAannual plan is not the “real annual plan” forthe year ahead, while the latter is laid out inthe agency’s operating plan. The followingvignette makes the point:

The administrator of a large, important, andvisible agency spent two days every quarterreviewing progress toward the agency’s keyobjectives, as embodied in the agency’s oper-ating plan. Present in the room were theagency’s most senior political and civil serviceexecutives. The administrator sat in the roomand followed the dialogue as the deputyadministrator thoroughly examined theprogress made toward each goal. The sessionwas notable for the deputy’s sustained vigor,intelligence, and sense of pursuit. At the endof the second day, the deputy, obviously tiredand mentally already on to the next task, said,“Oh, wait a minute, we forgot the GPRAreport. Would you please tell us where youare?” At this point the GPRA coordinatorstood, meaningfully brandishing a thick sheafof paper, and said, “Here they are. They’regoing to OMB tomorrow. You’ve all seen them.You’ve all signed off. Right?” Hearing noresponse, she sat down. The deputy thankedher for the report and closed the meeting.

It is worth noting that the administrator andthe deputy were highly regarded for their man-agement skills.

What can be done?

Recommended Responses1. An individual manager may not be able to

make the entire system whole. However,knowingly participating in an exercisewith no meaning can be described asduplicitous and a betrayal of public trust.Managers should take practical steps toinitiate and incrementally expand the fron-tiers of performance-based management inthose areas for which they bear primaryresponsibility. If the department’s GPRAplan is not consonant with its operatingplan, perhaps all or part can be put inplace in a particular operating area. Inother words, do what you can do to exer-cise your responsibility for the departmentas a whole—in your area.

2. Consider applying key GPRA elements toyour operation or program. Strategic plan-ning, annual planning, and annual report-ing using performance indicators is goodmanagement. The law does not requirethese elements for every activity and level.But there is nothing to prevent a programmanager from putting them in place. Drawup your own strategic plan. Use it toexplain to others inside and outside thedepartment what you do. Do the same for

the annual planning, reporting, and perfor-mance indicator requirements. It may helpto provide direction and meaning for yourstaff. It will establish you as someone whois committed to the best in performance-based public service. And it can provideleadership and encouragement to others.

“Our GPRA Plan Is Not Our Operating Plan”

42


Many federal programs do not do much morethan pass money through to other entities orlevels of government. The primary federal roleis getting the money out quickly. A sense ofimpatience emerges on the part of grantees,who are not shy about calling their membersof Congress if the money does not arrive ontime.

For programs that operate in this kind ofatmosphere, it is not hard to understand why federal administrators take a hands-offapproach. Frequently, nothing in the authoriz-ing legislation specifically requires the use ofperformance indicators. Thus, it is not unusualto find very minimal reporting requirements.

However, this begs the question: Are federaladministrators blind? Are they without anyresponsibility?

Such programs are specifically intended toreduce and streamline the federal role. Yet do they eliminate all management functionsexcept the distribution of money? No, espe-cially given the growing sense of public ser-vice accountability.

Recommended Response1. These program managers must take affir-

mative and creative steps to maintain anawareness of how well money is appliedat the grantee level. What and how muchis done depend on existing resource levels. Here is a short list of potentialapproaches:

• Develop model guidelines for the development and use of performanceindicators.

• Request access to grantee records.

• Provide forums for grantees to con-sider indicators.

• Develop a sample of grantees andmonitor it.

• Work with the Inspector General toadd questions to ongoing audits andstudies.

• Ask the research or evaluation func-tion for help.

• Conduct a mail or phone survey.

• Collect press and trade journal reportson program activities.

Also, consider requesting additional funds to fund a basic reporting activity. In today’s world, it is becoming less andless acceptable to say, “We don’t know”or “We’re not required to do that.”

“We Don’t Have Control Over Grantees”

43


• In developing their performance goals, each agency shouldmake clear their role in the delivery of public services.Specifically, in addition to outcome measures, agencies that givegrants to third parties should develop goals relating managementand oversight of grantees’ performance in achieving outcomes.

This report has identified the discrepancy between the demand for out-come performance measures and the actual work of many federal agen-cies. Direction in the development of these goals and their alignment withthe actual work of agencies should come from the agencies’ senior man-agement. This is the only way that agencies will be able to use their per-formance plans and reports as a tool to devise management strategies thatreflect their position, function, and capacity in the implementation of fed-eral programs.

As the number of third parties that agencies must work with to implementfederal programs increases, so too does the complexity of service delivery.As a general rule, more third parties in a given program means less lever-age for the agency charged with its implementation. This fact is not anexcuse for agency executives to shirk responsibility for results of programsfor which they have only limited control. Whenever third parties areinvolved in service delivery, agencies become players in a partnership fordelivering services. Agencies should use their GPRA strategic and perfor-mance plans to coordinate, measure, and oversee the activities of thirdparties to assure that all are working toward the goals established in GPRA strategic and performance plans.

• Agencies should use GPRA not only as a means to communicatetheir performance, but also to communicate constraints thatinhibit their performance.

GPRA provides critical information to decision makers within the agenciesand in Congress. The release of the first two performance reports and thesubsequent congressional and public response to each indicate that anagency’s performance report will be judged on its own merits and notbased on public or congressional perception of the agency. One agencythat has received much praise for the quality of its performance report has been the U.S. Agency for International Development (USAID). What isparticularly praiseworthy about USAID’s report? According to the MercatusCenter at George Mason University, USAID’s report “contains thoroughdiscussion of management challenges.” Additionally, the Mercatus analystsfound that the “agency does not hesitate to criticize its own initiatives anddiscuss failures.”

Some agencies have expressed concern that the performance informationin GPRA reports will serve as additional ammunition for members ofCongress to use during appropriations and oversight hearings. Aggressivecongressional scrutiny existed prior to GPRA and will continue regardlessof GPRA’s ultimate fate. Agency executives can strengthen their hand inthese discussions by using GPRA as a tool to systematically discuss agen-cies’ management challenges. In many instances, members of Congresswill discover or be reminded that many of the factors inhibiting perfor-mance are not under agencies’ immediate control. In addition to theextensive use of third parties in the delivery of federal services, the rulesset forth in authorizing legislation impede performance. In this way, agen-cies can frame the debate about their performance and even make rec-ommendations to Congress about what it can do to help agencies meettheir performance targets.

David G. Frederickson, The Potential of the Government Performance and Results Act as a Tool to Manage Third-Party Government (Arlington, Va.:The PricewaterhouseCoopers Endowment for The Business of Government), pp. 24-25.

Excerpt

44


Total cost per unit of service is one bottomline for a fully aligned management system. Amajor challenge in coming to that bottom lineis that many overhead costs (often calledadministrative costs) are not available by pro-gram or activity areas. Instead, much of thisinformation is maintained in large, aggregateaccounts managed separately from programs.These costs include such things as salaries and health and retirement benefits, and mayinclude such crosscutting support services ascontracting, personnel administration, andinformation technology support.

If the task is to improve performance, cost per unit of outcome is the inescapable bottomline. That the information does not exist in aconvenient or agreed upon format makes itdifficult—but not impossible—to come to gripswith unit cost.

There are two paths to the needed informa-tion. One is to charge agencies with develop-ing systems and accounting structures thatattribute costs to programs and activities. OMBhas taken steps in this direction, and it seemslikely that more will follow. Nevertheless, thedevelopment of new systems and accountingstructures will take time.

The second path is to develop estimates. Thesecan and should be developed now. Managerscan estimate the number of man-hoursdevoted to overhead tasks, obtain salary levels(including breakout estimates for benefits), anddevelop overall cost projections. These may

not be perfect. As long as the process used todevelop the estimates is documented andincluded in the budget, however, the informa-tion is properly bounded.

This estimation process can have additionalbenefits. Dialogue about costs and their attri-bution can be encouraged and sharpened,issues can be surfaced, cost saving ideas maybe generated, and cost sharing may be possi-ble. Indeed, a primary goal of performance-based budgeting is to stimulate dialogue thatcan lead to improvement.

Recommended Responses1. The systemic solution to the issue of

including overhead costs in total programcosts is to develop specific procedures fordefining and attributing costs. This willtake time and resources. The presentadministration has recognized the need tomove in this direction and has taken spe-cific steps to put the necessary systems inplace.

2. For the present—and absent specific uni-form accounting procedures—managerscan develop estimates and explain howthey were developed. There are manyways to do this. Professional help may beneeded. One approach is to look at howoverhead costs are attributed in similarkinds of activities in areas of the publicand/or private sectors that employ full costaccounting. This can, at a minimum, pro-

vide a context and range of likely costs. Avery fundamental approach would be tospread all overhead costs evenly across allprogram outlays. A more focused strategycould involve sampling actual costs inspecific areas and extrapolating from thesample. Whatever approach is taken, careshould be taken to make sure it is docu-mented in sufficient detail that it can beexplained and evaluated.

“There Is No Way to Include Overhead Expenses”

45


Mystery frequently surrounds the use of perfor-mance measures.

At times, finding an example of use seems toresemble a UFO sighting. A report begins tocirculate that a program has been using indi-cators effectively. The tale gathers momentumas it runs its course, and interest intensifies asaffirmations and confirmations multiply. Alas,a flaw is found, the context becomes morecomplex, and the example often turns out tobe less than what meets the eye.

There often is a grain of truth in the story, butthe grain is thrown out with the tale.

Why? When it comes to the issue of use, wefeel a need to seek an application so pure thatit could be written as holy script. We seek aone-to-one correlation between a piece of per-formance information and a given decision.

In no other context would we have the leasthope that the use of information would deter-mine the course of decision making. At most,we would expect it to inform the decision-making process. Where indicators are con-cerned, however, we expect a perfect worldwhere truth not only speaks to power but alsotells it what to do.

Such a world does not exist, and uses of per-formance indicators are far more subtle, vari-ous, and frequent than we recognize.

Using Performance Information

46


This statement is made with the intent of iden-tifying a major problem with “the system.”More often, the claim reveals a problem withthe individual making it.

Unpleasant as this may be for some to read,there is nothing wrong with decisions beingmade on a political basis—that is, decisionsmade by appropriately vested politicalappointees who give effect to the politicalplatform on which they are elected. Unlikecivil servants, they are held accountable to the electorate at fairly short intervals.

Career civil servants watch political appointeescome and go and they may grow weary at the mechanics of change—time lost, costsincurred, directions changed, opportunitiesmissed. Yet that is the process. To paraphraseWinston Churchill, “Democracy is the worstform of government, except for all other formsof government.”

We pay a price for the open exchange of ideasand broad access to political office. Civil ser-vants would do well to anticipate changesbrought by elections, and develop explicitstrategies to help new appointees accomplishtheir platform.

This may be distasteful for some to read. It ishard to do and, if done well, may never berecognized. It has to be redone many timesover the course of a civil service career. A newpolitical administration naturally is suspiciousof civil servants who are successful, so doing

the job well may lead to a difficult time get-ting started with the next administration.

Too often forgotten is the fact that this is thenature of public service. Most of us have cho-sen to be public servants, explicitly rejecting acareer in the private sector. Discouragement,foot dragging, and criticizing appointeesdemean the concept of public service. Theyalso reveal our own inadequacies in comingto grips with reality. It may be hot in thekitchen, but we have chosen the kitchen.

Recommended Responses1. Career civil servants should have a clear

understanding of the legitimate role ofpolitics and political appointees in theAmerican system of public administration.There is nothing wrong with Democrats.There is nothing wrong with Republicans.Even if you are of the opposite party. Thereis nothing wrong with the sometimes radi-cal change that can result when the presi-dential administration changes from oneparty to another. There should be no sur-prise that when a new administration takesoffice it is suspicious of the civil servantswho served the previous administration.This is the American democratic system in action.

2. Career civil servants should have a clearunderstanding of their role during achange of administration. They are in themiddle. They stand between two cultures,

two political ideologies, two very differentsets of emotions. They have the most chal-lenging of tasks: to close out the old andbring in the new. A transition is by defini-tion a trauma. Change is always difficult.All civil servants could help themselvesand new political appointees by develop-ing an explicit strategy for understanding,participating in, and supporting the transi-tion to a new administration. Mostly, wedon’t do this. But having a specific planfor handling our responsibilities during a transition can be a very effectiveapproach.

“What’s the Use? Decisions Are Political”

47


This claim—and those like it—is based on afundamental misconception that understatesthe use of indicators.

When the issue of use is raised, it is assumedthat it applies to making decisions. It is furtherassumed that the kind of use anticipatedinvolves a one-to-one correspondencebetween the value of a performance indicatorand a decision. For example, a low efficiencyscore would lead directly to a decision basedon that score.

Yet this is a misconception that oversimplifiesthe decision-making process, which must takeinto account many facets of program activity.On any given occasion, these may involvepolitical, policy, empirical, practical, andstrategic dimensions.

Information—pure, objective, and relevant as it may be—is only one factor for decisionmaking. Everyone knows this and readilyagrees when the statement is made. Yet an oddtransmutation occurs when the same peopleare asked to provide examples where perfor-mance indicators are used. Typically, there issolemn silence and a negative answer. Whenasked for examples, the respondent frequentlyadopts the narrowest possible definition andreports that very few decisions have beenmade based on performance indicators.

The true standard should be whether or notperformance information is taken into accountin the decision-making context, not whether itis the basis for a given decision. Over time

performance information should lead to ser-vice delivery improvement. Ultimately, its util-ity must be evidenced in actual programoperations.

Recommended Responses1. Develop a general guide on the many

dimensions of use relevant to your agencyor program. This could be as short as a tri-fold brochure. Its purpose would be toeducate stakeholders on the many dimen-sions of use. Short, illustrative examples ofeach type of use should be provided. Thisbrochure could then be handed out when-ever packages of program information areassembled for such purposes as trainingstaff, reporting to Congress, and informingthe public. The point is that an effort todevelop a better understanding of what“use” is will be time well spent. Even if, asis likely, the document is not disseminatedin larger numbers, having it available toanswer questions will be very useful.

2. Develop a specific plan of use. Once theconceptual work has been done on thetypes of uses that may be anticipated, aspecific plan of use can be developed. For example, performance informationintended to monitor progress toward pro-gram outcomes may also be used by theIG, the program evaluation function, thecongressional liaison office, the budgetofficer, and so on. By simply bringingthese together in one place, a certain logic

of use appears and it is seen that the manydimensions of use become concrete. Itmay be further seen that this use is bothbroader than anticipated and far moreroutine, or perhaps less exceptional thanis sometimes surmised.

“No One Uses Indicators Anyway”: Part I

48


The word “use” is problematic. After all, itimplies an act of will, a deliberate and specificapplication. It need not have this implication,but it often does in the contemporary context.

The problem is that indicators of performancecan be useful, have utility, or contribute toeffects or results—all without being identifiedin connection with a specific “use.” In a sense,the word has placed a stranglehold on theconcept of utility.

This does not mean that performance indicatorsshould be placed without specific regard fortheir use. When designing performance indica-tors for a particular application, a key questionalways should be: What is the purpose forwhich an indicator is sought? Once chosenand put in place, however, performance indi-cators enter several other streams where theirutility may be far broader—and perhaps moreimportant—than their intended use.

Two broad categories may help to suggest themagnitude and vectors of these streams. Onecategory might be called “secondary uses,”and would include all the “other” uses towhich indicators are put, as well as thosespecifically considered when the indicator wasdesigned. For example, an indicator may havebeen selected to help a program managerachieve an annual goal. It also may be usefulfor an in-depth program evaluation. Secondaryuses might be further divided into those thatare specifically intended and those that arerelated to specific new contexts.

Another category might be called “effects.”These would include consequences that arethe secondary and tertiary iterations stemmingfrom primary and secondary uses. Some canbe very important. For example, a heightenedorganization-wide sense of urgency in relationto performance goals or standards of service.

The intent is not to iterate vocabulary. Othershave considered these issues in much finerdetail. The point is that we must be carefulabout the word “use” and understand how we are defining it.

Recommended Response1. Develop a descriptive model of likely

effects of all or part of the performanceindicator system. Depict this model ingraphic terms. View it as a visual display of intended effects. Include such things asheightened organizational awareness ofperformance issues; improved intradepart-mental communication; greater cross-function, program, and unit collaboration;improved morale; and so on. The effects ofa performance-based management systemextend far beyond the immediate contextof a specific program, process, or decision.

“No One Uses Indicators Anyway”: Part II

49


Are you sure?

How are you defining the word “use”? Here isa list of possible uses that can be expanded inmany directions. Performance indicators maybe used to:

• Improve public service

• Improve service quality

• Improve management performance

• Support budget requests

• Make budget decisions

• Design program changes

• Provide incentives

• Assess/evaluate programs

• Conduct research

• Communicate to citizens

• Communicate to stakeholders

• Communicate to Congress

• Build accountability

• Support public dialogue about priorities

• Clarify program and policy purposes

• Reward good performance

• Identify best practices

• Seek continuous improvement

• Support economies of scale and type

• Streamline intergovernmental servicedelivery

• Avoid overlap and duplication

• Coordinate similar programs across agencies

The point here is not to make a definitive casefor all performance indicator uses, but toemphasize that performance indicators consti-tute a form of information. Information is thelifeblood of good management.

It always will be difficult to document the fullutility of performance indicators and informa-tion in general.

Recommended Response1. Support organizational learning and

greater awareness of the importance ofinformation in general and performanceindicators in particular in the managementprocess. This can be done in many ways,such as a performance knowledge center,a specific element in management trainingcourses, a special lecture series on therole of information in management, peri-odic brainstorming sessions, and informalbrown bag luncheon discussions. In thedaily press of activities, it is easy to forgetthat almost everything we do in both ourpersonal and professional lives dependson information—frequently about the per-

formance of something: a roof, car, insur-ance plan, article of clothing. Informationabout performance is an essential aspectof almost any human endeavor.

“No One Uses Indicators Anyway”: Part III

50


“… and so we can’t use it” is the end of thesentence sometimes offered in response to aquery about use.

This is a line of thinking that requires a littleuntangling.

Frequently, the implication is that the “out-come” takes more than several years toachieve. It also can mean that the outcometakes one year, but that year is not coincidentwith the GPRA reporting schedule. Theresometimes is a fear that an agency is disadvan-taged because only partial progress can bereported, even under the best circumstances,until the end of the multi-year period neces-sary for accomplishment.

There are two unhelpful conceptions here. Thefirst is that multi-year or off-cycle reporting isdisadvantageous. There is no reason for this tobe true. All that needs to be done is to providea narrative explanation of the reporting cycleand the issues raised, and to use available datato comply with the reporting schedule.

It also may be possible to adjust the data byextrapolating from one reporting schedule toanother and then adding any needed correc-tions to the subsequent year’s report. Suchadjustments may require technical help, but itis feasible. If the data show partial or limitedprogress, explain why relative to the reportingcycle’s original narrative explanation.

The second misconception is that the only use of the information presented annually is

related to the multi-year objective, as opposedto the annual segments leading up to it. This isunfortunate and untrue. GPRA and subsequentOMB guidance maintain a steady focus onoutcomes as the ultimate target. For mostagencies, much time is spent on tracking,reporting, and using information that simplyreports progress to date.

This interim information can be vital to fine-tuning annual management strategies and budgets. And this is a very important use ofperformance information.

Recommended Responses1. An important first step in situations where

outcomes cannot be achieved on anannual basis is to write a narrative state-ment explaining why this is so and whatthe likely period of performance will be.This statement will need to be reviewedand updated in subsequent years. Anymid-course corrections can be explainedat this time. The final outcome should bespecified in this initial narrative, alongwith the performance indicator that willsignal its accomplishment.

2. Develop a feedback system for trackinginterim stages of performance anticipatedto lead up to the final outcome. If an out-come cannot be fully realized in less thanfive years, then what are the significantstages of progress that can be noted alongthe way? Use these as milestones—indica-tors to show that the process is on the way

to the fulfillment of its ultimate indicator.Develop specific low-cost procedures thatdemonstrate a good faith effort to reportprogress in interim years. To say that anoutcome will take several years to achieveis one thing; to make no discernible goodfaith effort to report progress in interimyears is another.

“Our Data Are on the Wrong Cycle”

51


Excerpt…For many—probably most—programs, outcome indicators do not really require high levels of pre-cision and accuracy, although they should be as accurate and precise as feasible. Requiring exces-sive accuracy and precision can make outcome measurement expensive and discourage programs from attempting to seek some desirable outcome information.

The perspective of this report is that federal managers, and their personnel, need timely, frequentdata on outcomes—as basic information—if they are to attempt continuing improvement of theirprograms so as to make them more effective for the public. This is just good management. Thetheme of this paper is that it is likely to be considerably better to be roughly right rather than com-pletely ignorant about outcome information.

… various cost cutting/shortcuts … increase the possibility that on occasion the data obtained willbe misleading, possibly supporting poorer decisions than would have been made without the data.We believe, however, that such occurrences will be infrequent and will be greatly offset by themany more situations in which the rougher data provide information for better management deci-sions throughout the year.

Clearly if the resources for more expansive outcome measurement are available, without sacrificingother important outcome measurement opportunities, programs should go for the additional scopethat those resources make possible.

Harry Hatry and Shelli Rossman, Obtaining Timely, Cost-Effective, and Useful Outcome Data (Washington,D.C.: National Academy of Public Administration, June 2000), p. 2.

52


By its very nature, a performance-based man-agement approach may establish lines ofaccountability that cross organizationalboundaries. It also may cross accounting andbudget categories and management and policyprocesses.

For example, a logic model—constructed todepict the flow of sequential and relatedresponsibilities to achieve a goal—may involvedifferent bureaus, functions, and programs.This can be seen as impeding the use of infor-mation generated by performance indicators.

These are, after all, the “kingdoms” of bureau-cratic culture, each with its own leaders andstaff—sometimes with independent policy,budget, administrative, and even politicalfunctions.

The logic model itself delineates accountabil-ity relationships for these entities individuallyand combined. The information generated ateach stage of the process also can be usefulfor prior and subsequent stages, and the over-all process as well.

This is what is meant by the term “knowledgemanagement.” Any management process is aninformation-generating series of activities pro-viding an opportunity for organizational learn-ing. In the management world of today andtomorrow, those who do not use informationeffectively will be identified. It will not be pos-sible for part of a process to hide from view,or to escape the judgment of related parts orthe organization as a whole.

Recommended Responses1. Once a logic model has been specified for

a given management process, each part ofthe process should construct a plan of usefor related performance information. Thisplan should stretch the concept of use fromdirect uses—such as achieving manage-ment milestones—to indirect uses, such asevaluation and research and policy analy-sis.

2. A similar plan of use should be developedbased on an overview of all performanceindicators for an entire process. Such anapproach likely will identify combinationsof data elements from different parts of theprocess that can yield useful information.

“Our Organizational Structure Impedes Use”

ExcerptIn some types of programs it is more difficult totrack performance than in others. For example,there are particular problems in measuring per-formance in programs that are under limited federal control and crosscutting programs admin-istered by more than one agency. Among thetechniques for overcoming problems in usingperformance are the following:

• Developing intermediate measures to showprogress which contribute to end results

• Disaggregating performance data for sub-groups with different expectations

• Using statistical models to reflect the influ-ence of key external factors

• Using qualitative measures

• Making use of program evaluations

Center for Improving Government Performance,Questions and Answers for Improving GovernmentPerformance: Using Performance Data to ImproveProgram Effectiveness (Washington, D.C.: NationalAcademy of Public Administration), p. 15.

53


Perhaps the single most disappointing aspect ofperformance-based management implementationover the last decade has been the widespreadlack of attention given to communicating per-formance information.

Notice the word used here is “disappointing.”

There are other areas where attention is seri-ously needed to “get the ball rolling,” such asdata availability, but nothing is so close to theheart of performance-based management ascommunication.

One may argue that there could be enormousimperfections in the design of performanceindicators, the alignment of managementprocesses, and even the use of indicators. Still,performance information can make a dramaticcontribution to improving government perfor-mance if it is effectively communicated tostakeholders, including citizens.

How can this be so? The ensuing dialoguewould rapidly broaden the understanding ofservice delivery issues, and improve the qual-ity of performance indicators and services.

Yet, no aspect of performance-based manage-ment is so routinely unappreciated and unaddressed.

Communicating PerformanceInformation

54


Communication?

How many times have you heard this wordused? Chances are, not many. Communica-tion—the act of communicating—seems to bea lost art in government.

One wonders why. As public servants, weshould listen and speak clearly and effectively.Even for the best of us, those superb publicservants who see public service as a high call-ing and strive to achieve the highest level ofquality in their work, the attention paid tocommunication is often less than optimal.

Effective communication is an activity com-pletely distinct from the act of achievingresults.

Achieving good results is a necessary, but notsufficient, requisite for good management.How can that be? Because others will notknow that good management has occurredunless it is communicated to them. Further, itis not possible to improve future managementwithout knowing the lessons from presentresults.

How can results be communicated effectively?By presenting them in ways that stakeholderscan understand. Stakeholders must be identi-fied, and a specific communication strategydesigned with them in mind.

How many people read beyond the executivesummary of a report or other document? Howmany get past the first page? Pages of wordsand numbers may be good and true, and theymay document superb performance. Yet thecore question remains: Do they communicateeffectively to the intended reader?

Recommended Responses1. Identify the intended audience or audi-

ences and develop a communication strat-egy specific to each. This is a fundamentalstep, though it is often left out. Identifyingthe intended audience for the purpose ofdevising a communication strategy is justas important as identifying the audiencefor the purpose of designing an outcomeindicator. In fact, the two are, or shouldbe, different parts of the same task. Anoutcome measure should always bedesigned to address a specific purpose in relation to a specific stakeholder.Communicating to the stakeholder is theculmination of the process. Now, theremay need to be more than one communi-cation strategy, because there is more thanone purpose and more than one stake-holder. At an absolute minimum, thereshould be a communication strategy toinform citizens, and a strategy to informoversight bodies such as Congress, OMB,and other agencies.

2. Use different communication media. Thetendency is to think in terms of a lengthy,written report full of statistics. Shorter,more readable reports can be very useful.These may include executive summarytype reports of 25 pages or less, smallpamphlets, and trifold leaflets. Reports alsocan be posted on the web, and could evenhave a section where readers could askquestions. An interactive chat room couldprovide useful feedback. Film can be usedto record meetings where key executivesreport and discuss major findings.

“No One Here Gets It”

55


When the issue of communicating governmentprogram information arises, the conversationfrequently turns to reports.

The government is awash with reports.Many—if not most—pieces of legislationrequire an “annual report” or a one-time“report back to Congress.” Communicationthus becomes a matter of “reporting” orpreparing a report. Of course, there are manyother forms of communication. Yet repetitivecongressional requirements move thinking inthat direction, as opposed to other venuessuch as conferences, informal briefings,newsletters, film and radio, TV, and electronicmedia.

One frequently heard lament is: “We don’thave any money for new reports.”

If it is determined that a report is the bestcommunication device in a particular situa-tion, an existing requirement and resourcesdevoted to it can provide just what is needed.An existing annual report could be cut back tofree up resources for a new report, or it couldbe reformatted to accomplish the previous andnew purposes in one document.

Many annual reports being prepared are notespecially useful. In the early years of a pro-gram, they are frequently used to report wherethe money is going. In later years they canbecome long compendiums of lifeless, and notespecially useful, statistics.

Sometimes, the very same database used topresent the existing annual report can be refor-matted to serve a different purpose, most likelyby breaking data into different categories.Some new data collection or analysis mayneed to be added, but this is preferable tostarting from scratch with no budget. Low-costtechniques, such as telephone and mail sur-veys, can be used to supplement existing data-bases and address current issues.

Recommended Responses1. Try to adapt existing reporting require-

ments and resources to new uses and formats. Many programs have a built-inreporting requirement and a specific bud-get. Some have a reporting requirementand do not have a specific budget, but,over time, a certain amount of money ismade available for report production andthis amount tends to stay in place. Thepoint is that there is likely some amount ofresources—staff, budget, or contract—thatis in place and that could be reshaped tounderwrite a new reporting format.Considering the large number of pagesnow produced in annual reports that arenot widely read, a fresh approach mayyield a high dividend.

2. Use inexpensive data collection proce-dures, such as mail and telephone sur-veys, to add new material and focus to an existing report. Try to reserve a small

amount of available resources for gather-ing new data. Most annual reports sum-marize information yielded by in-placedata systems. To the extent that these sys-tems do not change much from year toyear, they have a limited capacity to focuson new or emerging issues. The surest wayto get a report read is to include informa-tion on a topical issue. This kind of infor-mation can be included in the reportunder the heading “emerging issues.”Asking stakeholders in the third quarter ofa reporting year what issues they wouldlike to know more about can be a usefulway to build relationships and gaininvolvement in the reporting process.

“We Don’t Have Money for New Reports”

56


A surprising question?

Perhaps not, when you consider the context.

The government produces thousands of reportson program activity. Most are “required” bysome law, regulation, or administrativerequirement. In its early years, a report often is attuned directly to the interests of primarystakeholders, frequently a congressional com-mittee, OMB, or interest group.

In subsequent years, as the program or activitybecomes a part of an operating routine—and,in particular, as those who were most directlyconnected with early program activities moveon and are replaced by new people—thereport can become a fairly mechanical func-tion; and sometimes less focused on currentstakeholders and issues.

Even in a program’s early years, when theattention paid to stakeholders can be at itsheight, this attention often is not as wellthought out as it might be. Stakeholders withkey political positions or interests can haveprecedence over program customers and citizens.

Communication is important—in fact, essen-tial—in our form of government. It deservesmore careful thought than it is frequentlygiven, even for new programs.

The starting point for good communication isthe identification of stakeholders. Attention

should be given to their size, importance, andpossible interrelationships. A specific stake-holder communication strategy then should bedeveloped as the basis for allocating availableresources. Different groups may require differ-ent approaches, and these may require multi-ple reports or communication vehicles. Oncethe communication strategy has been imple-mented, its effectiveness should be monitoredand assessed so refinements can be made.

Recommended Responses1. Identify stakeholders—the intended audi-

ence. (See the first recommendation under“No One Gets It.”)

2. Develop a communication strategy specifi-cally for the stakeholders. The strategyshould consider how stakeholders obtainand use information. By what media: hardcopy, computer, film, live event? Eachcontext will be different, but some aspectsof the presentation will likely be commonto all successful presentations. Keep itshort; have a good executive summary.Use graphics to display quantitative data.In other words, be considerate. Ask howyou would like to receive information ifyou were the stakeholder. Monitor imple-mentation so that lessons learned can beincorporated into future events.

“Whom Would We Communicate To?”

57


ExcerptThere is a need for agencies and the federal government to restore trust and confidence to the pub-lic. One of the strategies for accomplishing this task is the Results Act. The Act has several key points:

• Performance plans must reflect measurable outcomes.

• The significant measures should be important to the business, stakeholders, and customers.

• Agencies need to establish improvement plans.

The processes that are used to improve customer satisfaction should be improved upon continuouslyas feedback from the customer is received. Looking at other businesses that have succeeded in pro-viding consummate customer service and using them as a point of reference will also help toincrease customer satisfaction. Finally, reworking the business processes to make them more efficientand goal oriented contributes to successful customer satisfaction.

There are several ways to measure the success an agency has in satisfying its customers. Establishinginternal measurement systems and comparing customer feedback on customer service to the originalbaseline are tools to increase customer satisfaction. After the comparison is made, it is possible tothen assess and modify, if needed, the agency’s plans and strategies.

The primary factors in improving customer satisfaction are communication with both the customerand within the agency, involvement of both parties, and the flexibility, on the part of the agency, toprogressively alter and improve plans and strategies.

Center for Improving Government Performance, Performance Notes: 2000—Dialogues, Practices, and Results onthe Government Performance and Results Act (Washington, D.C.: National Academy of Public Administration,2000), p. 4.

58


Everyone thinks his or her situation is “differ-ent,” “difficult to understand,” and “very com-plicated to explain.”

But the remedy is the same.

Tell your story, whatever it is.

If you do not, someone else will.

Those who will want to tell your story for youprobably are not your friends.

The point is that no performance indicator cantell a story, just as a brick doesn’t make ahouse. No performance indicator can yielduseful information until it is interpreted,explained, and set in context. The fear that noone will understand comes from a misplacedsense that performance indicators alone willbe taken as the “full report” on progresstoward a given objective.

Every value reported for a performance indica-tor should be accompanied by narrative thatprovides contextual and interpretive informa-tion. A good practice would be to place blankspace next to the reported value to remind theprogram managers to comment on what thevalue means.

Still, some situations and data are very compli-cated and difficult to understand, no matterhow much care is taken in their design, analy-sis, and interpretation. All the more reason,then, to make a focused effort to explain whatthey mean and to communicate that effectively

to key stakeholders. If necessary, use outsideassistance to add credibility to explaining thecomplications.

Recommended Responses1. Provide a space for narrative commentary

at every point in the performance indica-tor system where the value of a specificindicator is required. Imagine a formdesigned to record performance informa-tion. See it as a series of statements orquestions, each of which ends on theright-hand side of the page with a blankspace or box in which to record a numeri-cal score. Directly under or very close tothe statement or question should be aseries of blank lines with an invitation toprovide any relevant explanatory or con-textual information needed to properlyinterpret the numerical score. As a rule,no numerical value should be displayedwithout the relevant contextual orexplanatory statement.

2. Present an interpretive overview that con-siders all of the indicators together. This is,in effect, the same approach suggestedabove for individual performance indica-tor values—but now applied to all of theindicator values relevant to a particulargoal. The question is: How do all of theserelate? What do they say overall—what isgoing on?

“No One Will Understand”

59


Of all the concerns raised regarding communi-cating results, few are as frequently heard asthis one.

“And so,” the conversation continues, “wehave this problem, because if we say wehaven’t achieved the goal, a lot of bad thingscan happen, like GAO reports, congressionalhearings, OMB reviews, and IG audits.”

True enough. Danger may be near.

Good goals—goals well defined—shouldalways involve some element of risk andstretch if management seeks continuousimprovement. This should especially be truewhen programs, strategies, activities, or meth-ods are doing something that has inherent risk,such as launching a new kind of social, scien-tific, technical, or medical program.

Good public servants should live at the cuspof risk, pursuing the highest levels of publicservice—that place where a reasonable alloca-tion of resources under conditions of risk has areasonable chance of achieving breakthroughlevels of improved performance.

Fear is a poor reason for a failure to communi-cate results.

Under all circumstances, results should bereported in conjunction with a narrative state-ment setting forth the context and reasons forthe results. This statement should explain boththe element of risk in the original goal and thereason why the goal was not fully met.

A good and well-communicated report is thebest vehicle for addressing a missed goalbecause it is your own story. If you do notreport a missed goal, someone else will.Through ignorance or intent, the story that istold about your program may be wrong andpotentially harm the program.

Recommended Responses1. Write a good report concerning a missed

goal before someone else writes a badone. Try to write the first report. In otherwords, be the one who brings the missedgoal to public view. Include in the reporta clear explanation for the missed goal.Clear writing and honest reporting isimportant. Accompany the explanationwith a strategy for assuring that the goal isnot missed the next time.

2. Communicate individually with key stake-holders about a missed goal before it getsreported in the media or a formal report isissued. Few things are appreciated morewhen something important is at stake thana “heads up” call. The agency politicalleadership, senators and representatives,OMB, and key stakeholders, such as citizenand interest groups, can become allies inexplaining the issue if they are approachedearly and brought into the dialogue.

“But We Didn’t Meet Our Goal”

60


“Our performance indicators are reportedquarterly and posted on the web. What moredo we need?”

Answer: Posting on the web is fine. Reportingonly performance indicators—without contex-tual information—can be fatal.

Performance indicators, most of which arenumbers, can be the equivalent of loose can-nonballs rolling around the deck if they arenot accompanied by contextual information.

If a performance indicator were more than anindicator, it would be called something else.An indicator indicates; it does not explain. It isa piece of information designed to help a pro-gram manager know if his or her activity is onor off course. It is not designed to providedefinitive information on results or explainactivities leading up to them.

Imagine a goal set at 98 percent. At the end ofthe year, the performance achieved was 97percent. A news headline could read, “XYZAgency misses again!” Yet the real story maybe that agency XYZ came from a level of 35percent to 97 percent in one year. A very dif-ferent story, indeed.

This is an overly simple illustration. Yet everyreport becomes a story, whether or not it ispresented as a story. The only question iswhether you, as the program manager, will tellthe story, or let someone else do it. Who isbetter prepared to tell the story, and who has

the first obligation to do so? The public servantin charge, exercising his or her responsibilityto be accountable to citizens.

Recommended Responses1. Never release a report containing perfor-

mance indicators without contextual infor-mation. Never let a number or percentageappear in a prominent place withoutnearby textual elaboration. This point can-not be overemphasized. Even if the num-ber or percentage seems self explanatory,go ahead and explain it anyway.

Example:Percentage: 97% of the awards weremade.

Contextual statement: This means that34,605 out of a total of 35,675 peoplereceived program benefits.

2. Develop a story line for the report. Usingthe above example, “More than 34,000homeless people received food, clothing,and shelter.”

“What Would We Say?”

ExcerptTopics that might be particularly good candi-dates for making clear the linkage betweendaily activities and the organization’s strategicfocus might include:

• Justifications for legislative changes, man-agement reforms, new technologies, betterinformation sources, or other innovations

• Explanations of the contribution of admin-istrative support activities to the successfulpursuit of the organization’s general goalsand objectives

• Discussion of how alternative scenarioswere explored, and rationales for the chosen path—a procedure at the heart of strategic planning for businesses andfor state and local governments …, butseldom surfaced in the strategic plans offederal agencies

Center for Improving Government Performance,Helpful Practices in Improving GovernmentPerformance: Planning for Results (Washington,D.C.: National Academy of Public Administration),p. 17.

61


Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance, Washington, D.C.: NationalAcademy of Public Administration, 1998.

Center for Improving Government Performance.Designing Outcome Measures at the U.S.Patent and Trademark Office: A Dialogue withStakeholders and a Review of ExistingMeasures, vol. I, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.Designing Outcome Measures at the U.S.Patent and Trademark Office: Statements ofViews, vol. II, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.Performance Notes: 2000—Dialogues,Practices and Results on the GovernmentPerformance and Results Act, Washington,D.C.: National Academy of PublicAdministration, 2000.

National Academy of Public Administration.National Science Foundation’s Science andTechnology Centers: Building an InterdisciplinaryResearch Paradigm, Washington, D.C.: NationalAcademy of Public Administration, 1995.

National Academy of Public Administration.Information Management PerformanceMeasures: Developing Performance Measuresand Management Controls for MigrationSystems, Data Standards, and ProcessImprovement, Washington, D.C.: NationalAcademy of Public Administration, 1996.

National Academy of Public Administration.Deciding for the Future: Balancing Risks,Costs, and Benefits Fairly Across Generations,Washington, D.C.: National Academy ofPublic Administration, 1997.

National Academy of Public Administration.Improving the NOAA Budget and FinancialManagement Process, Washington, D.C.:National Academy of Public Administration,1999.

Panel on Improving Government Performance.Toward Useful Performance Measurement:Lessons Learned from Initial Pilot PerformancePlans Prepared Under the GovernmentPerformance and Results Act, Washington,D.C.: National Academy of PublicAdministration, 1994.

Panel on Improving Government Performance.Effective Implementation of the GovernmentPerformance and Results Act, Washington,D.C.: National Academy of PublicAdministration, 1998.

Publications of the Center forImproving Government Performance

62


Conference PublicationsCenter for Improving GovernmentPerformance. Performance Leaders Conference’99: Helpful Practices in ImprovingGovernment Performance—ConferenceProceedings, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.Performance Leaders Conference ’99: HelpfulPractices in Improving GovernmentPerformance—Executive-Legislative BranchSummit Proceedings, Washington, D.C.:National Academy of Public Administration,1999.

Focus PapersCenter for Improving Government Performance.Obtaining Timely, Cost-Effective, and UsefulOutcome Data, Washington, D.C.: NationalAcademy of Public Administration, 2000.

Center for Improving Government Performance.Performance Results Forum: Getting to Results:What We Have Learned About MakingStrategic Planning Effective, Washington, D.C.:National Academy of Public Administration,2001.

Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance: An Overview of HelpfulPractices, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance: Budget Alignment, Washington,D.C.: National Academy of PublicAdministration, 1999.

Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance: Improving Performance AcrossPrograms: Thinking About the Issue—Takingthe First Steps, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance: Linking Administrative Supportto Strategic Planning, Washington, D.C.:National Academy of Public Administration,1999.

Center for Improving Government Performance.Helpful Practices in Improving GovernmentPerformance: Planning for Results, Washington,D.C.: National Academy of PublicAdministration, 1999.

Center for Improving Government Performance.Questions and Answers for ImprovingGovernment Performance: Designing EffectivePerformance Measures, Washington, D.C.:National Academy of Public Administration,2000.

Center for Improving Government Performance.Questions and Answers for ImprovingGovernment Performance: OperationalizingPerformance Management, Washington, D.C.:National Academy of Public Administration,2000.

Center for Improving Government Performance.Questions and Answers for ImprovingGovernment Performance: Using PerformanceData to Improve Program Effectiveness,Washington, D.C.: National Academy ofPublic Administration, 2000.

Center for Improving Government Performance.Questions and Answers for ImprovingGovernment Performance: Using PracticalProgram Evaluations, Washington, D.C.:National Academy of Public Administration,2000.

Center for Improving Government Performance.Transition Briefing Paper Forum: Why IsPerformance Budgeting Important?,Washington, D.C.: National Academy ofPublic Administration, 2000.

Center for Improving Government Performance.Transition Briefing Paper Forum: Why IsCrosscutting Measurement Important?,Washington, D.C.: National Academy ofPublic Administration, 2000.

63


Center for Improving Government Performance.Transition Briefing Paper Forum: What HaveBeen the Results (So Far) of the Results Act?,Washington, D.C.: National Academy ofPublic Administration, 2000.

Center for Improving Government Performance.Transition Briefing Paper Forum: What Are theUses of Performance-based Information andMeasures?, Washington, D.C.: NationalAcademy of Public Administration, 2000.

Center for Improving Government Performance.Transition Briefing Paper Forum: Why MustTop Leaders Lead the Way to Performance-based Government?, Washington, D.C.:National Academy of Public Administration,2000.

Performance DialogueCenter for Improving Government Performance.On Helpful Practices in Program Evaluation,Washington, D.C.: National Academy ofPublic Administration, 1998.

Center for Improving Government Performance.On Designing Effective Performance Measures,Washington, D.C.: National Academy ofPublic Administration, 1999.

Center for Improving Government Performance.On Operationalizing PerformanceManagement, Washington, D.C.: NationalAcademy of Public Administration, 1999.

Center for Improving Government Performance.On Using Performance Data to ImproveProgram Effectiveness, Washington, D.C.:National Academy of Public Administration,1999.

64


Chris Wye is Director of the Center for Improving GovernmentPerformance at the National Academy of Public Administration,where he has been since 1994. The Center provides assistanceto government agencies on improving management processesusing the techniques of performance-based management,including strategic planning, performance measurement, pro-gram evaluation, performance-based budgeting, and strategicmanagement.

Wye also is the Director of the Performance Consortium, anorganization of 30 federal agency functions that have cometogether under the Academy’s leadership to provide a programof peer-to-peer dialogue, forums, workshops, and conferences to support the exchange of helpful practices related to perfor-mance-based management.

Prior to his work with the Academy, Wye spent 20 years in the federal government, directing policy analysis, program evaluation, and program monitoring functions. He has written and published widely on practical, low-cost techniques for improving government performance.

Wye graduated from Parsons College with a B.A. in 1966, and from Kent State University with an M.A. in 1967 and a Ph.D. in 1974.

A B O U T T H E A U T H O R

65


To contact the author:Chris WyeDirector Center for Improving Government PerformanceNational Academy of Public Administration1100 New York Avenue NWSuite 1090 East Washington, DC 20005(202) 347-3190, ext. 3208

e-mail: [email protected]

K E Y C O N T A C T I N F O R M A T I O N

66

ENDOWMENT REPORTS AVAILABLE

To download or order a copy of a grant or special report, visit the Endowment website at: www.businessofgovernment.org

GRANT REPORTS

E-Government

Managing Telecommuting in theFederal Government: An InterimReport (June 2000)

Gina VegaLouis Brennan

Using Virtual Teams to ManageComplex Projects: A Case Study ofthe Radioactive Waste ManagementProject (August 2000)

Samuel M. DeMarie

The Auction Model: How the Public Sector Can Leverage thePower of E-Commerce ThroughDynamic Pricing (October 2000)

David C. Wyld

Supercharging the EmploymentAgency: An Investigation of the Useof Information and CommunicationTechnology to Improve the Service ofState Employment Agencies(December 2000)

Anthony M. Townsend

Assessing a State’s Readiness forGlobal Electronic Commerce:Lessons from the Ohio Experience(January 2001)

J. Pari SabetySteven I. Gordon

Privacy Strategies for ElectronicGovernment (January 2001)

Janine S. HillerFrance Bélanger

Commerce Comes to Government onthe Desktop: E-CommerceApplications in the Public Sector(February 2001)

Genie N. L. Stowers

The Use of the Internet inGovernment Service Delivery(February 2001)

Steven CohenWilliam Eimicke

State Web Portals: Delivering andFinancing E-Service (January 2002)

Diana Burley GantJon P. GantCraig L. Johnson

Internet Voting: Bringing Elections tothe Desktop (February 2002)

Robert S. Done

Leveraging Technology in the Serviceof Diplomacy: Innovation in the Department of State(March 2002)

Barry Fulton

Federal Intranet Work Sites: AnInterim Assessment (June 2002)

Julianne G. MahlerPriscilla M. Regan

Public-Sector Information Security:A Call to Action for Public-SectorCIOs (July 2002)

Don Heiman

The State of Federal Websites: ThePursuit of Excellence (August 2002)

Genie N. L. Stowers

State Government E-Procurement inthe Information Age: Issues, Practices,and Trends (September 2002)

M. Jae Moon

Preparing for Wireless and MobileTechnologies in Government(October 2002)

Ai-Mei ChangP. K. Kannan

Financia l Management

Credit Scoring and Loan Scoring:Tools for Improved Management ofFederal Credit Programs (July 1999)

Thomas H. Stanton

Using Activity-Based Costing to Manage More Effectively(January 2000)

Michael H. GranofDavid E. PlattIgor Vaysman

Audited Financial Statements:Getting and Sustaining “Clean”Opinions (July 2001)

Douglas A. Brook

An Introduction to Financial RiskManagement in Government (August2001)

Richard J. Buttimer, Jr.

Human Capita l

Profiles in Excellence: Conversationswith the Best of America’s CareerExecutive Service (November 1999)

Mark W. Huddleston

Leaders Growing Leaders: Preparingthe Next Generation of Public ServiceExecutives (May 2000)

Ray Blunt

Reflections on Mobility: Case Studiesof Six Federal Executives (May 2000)

Michael D. Serlin

A Learning-Based Approach toLeading Change (December 2000)

Barry Sugarman

Labor-Management Partnerships:A New Approach to CollaborativeManagement (July 2001)

Barry RubinRichard Rubin

Winning the Best and Brightest:Increasing the Attraction of PublicService (July 2001)

Carol Chetkovich

Organizations Growing Leaders: BestPractices and Principles in the PublicService (December 2001)

Ray Blunt

A Weapon in the War for Talent:Using Special Authorities to RecruitCrucial Personnel (December 2001)

Hal G. Rainey

A Changing Workforce:Understanding Diversity Programs in the Federal Government(December 2001)

Katherine C. NaffJ. Edward Kellough

To download or order a copy of a grant or special report, visit the Endowment website at: www.businessofgovernment.org

67To download or order a copy of a grant or special report, visit the Endowment website at: www.businessofgovernment.org

Life after Civil Service Reform: The Texas, Georgia, and FloridaExperiences (October 2002)

Jonathan Walters

Managing for Resul ts

Corporate Strategic Planning in Government: Lessons from the United States Air Force(November 2000)

Colin Campbell

Using Evaluation to SupportPerformance Management:A Guide for Federal Executives(January 2001)

Kathryn NewcomerMary Ann Scheirer

Managing for Outcomes:Milestone Contracting in Oklahoma (January 2001)

Peter Frumkin

The Challenge of Developing Cross-Agency Measures: A CaseStudy of the Office of National DrugControl Policy (August 2001)

Patrick J. MurphyJohn Carnevale

The Potential of the GovernmentPerformance and Results Act as a Tool to Manage Third-PartyGovernment (August 2001)

David G. Frederickson

Using Performance Data forAccountability: The New York CityPolice Department’s CompStat Modelof Police Management (August 2001)

Paul E. O’Connell

Performance Management: A “StartWhere You Are, Use What You Have”Guide (October 2002)

Chris Wye

New Ways to Manage

Managing Workfare: The Case of the Work Experience Program in the New York City ParksDepartment (June 1999)

Steven Cohen

New Tools for ImprovingGovernment Regulation: AnAssessment of Emissions Trading and Other Market-Based RegulatoryTools (October 1999)

Gary C. Bryner

Religious Organizations, Anti-PovertyRelief, and Charitable Choice: AFeasibility Study of Faith-Based Welfare Reform inMississippi (November 1999)

John P. BartkowskiHelen A. Regis

Business Improvement Districts and Innovative Service Delivery(November 1999)

Jerry Mitchell

An Assessment of BrownfieldRedevelopment Policies: The Michigan Experience (November1999)

Richard C. Hula

Determining a Level Playing Field forPublic-Private Competition(November 1999)

Lawrence L. Martin

San Diego County’s InnovationProgram: Using Competition and aWhole Lot More to Improve PublicServices (January 2000)

William B. Eimicke

Innovation in the Administration of Public Airports (March 2000)

Scott E. Tarry

Entrepreneurial Government:Bureaucrats as Businesspeople (May 2000)

Anne Laurent

Implementing State Contracts forSocial Services: An Assessment of theKansas Experience (May 2000)

Jocelyn M. JohnstonBarbara S. Romzek

Rethinking U.S. EnvironmentalProtection Policy: ManagementChallenges for a New Administration(November 2000)

Dennis A. Rondinelli

The Challenge of Innovating inGovernment (February 2001)

Sandford Borins

Understanding Innovation:What Inspires It? What Makes ItSuccessful? (December 2001)

Jonathan Walters

A Vision of the Government as a World-Class Buyer: MajorProcurement Issues for the Coming Decade (January 2002)

Jacques S. Gansler

Contracting for the 21st Century: A Partnership Model (January 2002)

Wendell C. Lawther

Franchise Funds in the FederalGovernment: Ending the Monopolyin Service Provision (February 2002)

John J. Callahan

Managing “Big Science”: A CaseStudy of the Human Genome Project(March 2002)

W. Henry Lambright

Leveraging Networks to MeetNational Goals: FEMA and the Safe Construction Networks(March 2002)

William L. Waugh, Jr.

Government Management ofInformation Mega-Technology:Lessons from the Internal RevenueService’s Tax Systems Modernization(March 2002)

Barry Bozeman

Making Performance-BasedContracting Perform: What theFederal Government Can Learn from State and Local Governments(June 2002)

Lawrence L. Martin

68 To download or order a copy of a grant or special report, visit the Endowment website at: www.businessofgovernment.org

21st-Century Government and theChallenge of Homeland Defense(June 2002)

Elaine C. Kamarck

Moving Toward More CapableGovernment: A Guide toOrganizational Design (June 2002)

Thomas H. Stanton

TransformingOrganizat ions

The Importance of Leadership:The Role of School Principals(September 1999)

Paul TeskeMark Schneider

Leadership for Change: Case Studies in American LocalGovernment (September 1999)

Robert B. DenhardtJanet Vinzant Denhardt

Managing DecentralizedDepartments: The Case of the U.S. Department of Health andHuman Services (October 1999)

Beryl A. Radin

Transforming Government: TheRenewal and Revitalization of theFederal Emergency ManagementAgency (April 2000)

R. Steven DanielsCarolyn L. Clark-Daniels

Transforming Government: Creatingthe New Defense ProcurementSystem (April 2000)

Kimberly A. Harokopus

Trans-Atlantic Experiences in HealthReform: The United Kingdom’sNational Health Service and theUnited States Veterans HealthAdministration (May 2000)

Marilyn A. DeLuca

Transforming Government: TheRevitalization of the Veterans Health Administration (June 2000)

Gary J. Young

The Challenge of Managing Across Boundaries: The Case of the Office of the Secretary in the U.S.Department of Health and HumanServices (November 2000)

Beryl A. Radin

Creating a Culture of Innovation: 10Lessons from America’s Best Run City (January 2001)

Janet Vinzant DenhardtRobert B. Denhardt

Transforming Government:Dan Goldin and the Remaking of NASA (March 2001)

W. Henry Lambright

Managing Across Boundaries: A CaseStudy of Dr. Helene Gayleand the AIDS Epidemic (January 2002)

Norma M. Riccucci

SPECIAL REPORTS

Government in the 21st Century

David M. Walker

Results of the GovernmentLeadership Survey: A 1999 Survey of Federal Executives (June 1999)

Mark A. AbramsonSteven A. ClyburnElizabeth Mercier

Creating a Government for the 21st Century (March 2000)

Stephen Goldsmith

The President’s ManagementCouncil: An Important ManagementInnovation (December 2000)

Margaret L. Yao

Toward a 21st Century PublicService: Reports from Four Forums (January 2001)

Mark A. Abramson, Editor

Becoming an Effective PoliticalExecutive: 7 Lessons fromExperienced Appointees (January 2001)

Judith E. Michaels

The Changing Role of Government:Implications for Managing in a NewWorld (December 2001)

David Halberstam

BOOKS*

E-Government 2001(Rowman & Littlefield Publishers, Inc., 2001)Mark A. Abramson and Grady E. Means, editors

Human Capital 2002(Rowman & Littlefield Publishers, Inc., 2002)Mark A. Abramson andNicole Willenz Gardner, editors

Innovation(Rowman & Littlefield Publishers, Inc., 2002)Mark A. Abramson andIan Littman, editors

Leaders(Rowman & Littlefield Publishers, Inc., 2002)Mark A. Abramson and Kevin M. Bacon, editors

Managing for Results 2002(Rowman & Littlefield Publishers, Inc., 2001)Mark A. Abramson and John Kamensky, editors

Memos to the President:Management Advice from the Nation’s Top PublicAdministrators (Rowman & Littlefield Publishers, Inc., 2001)Mark A. Abramson, editor

Transforming Organizations(Rowman & Littlefield Publishers, Inc., 2001)Mark A. Abramson and Paul R. Lawrence, editors

* Available at bookstores, online booksellers, and from the publisher (www.rowmanlittlefield.comor 800-462-6420).

For additional information, contact:Mark A. AbramsonExecutive DirectorIBM Endowment for The Business of Government1616 North Fort Myer DriveArlington, VA 22209(703) 741-1077, fax: (703) 741-1076

e-mail: [email protected]: www.businessofgovernment.org

About the EndowmentThrough grants for research, the IBM Endowment for The Business of Government stimulates research andfacilitates discussion on new approaches to improving the effectiveness of government at the federal, state,local, and international levels.

Founded in 1998, the Endowment is one of the ways that IBM seeks to advance knowledge on how to improvepublic sector effectiveness. The IBM Endowment focuses on the future of the operation and management ofthe public sector.

PRST STDUS Postage

P A I DPermit 1112

Merrifield, VA1616 North Fort Myer DriveArlington, VA 22209-3195

performance management: a “start where you are, use what ... · this report by chris wye,...

Documents