Performance Monitoring Systems for Public Managers

Author(s): Stan Altman

Reviewed work(s):
Source: Public Administration Review, Vol. 39, No. 1 (Jan. - Feb., 1979), pp. 31-35
Published by: Blackwell Publishing on behalf of the American Society for Public Administration
Stable URL: http://www.jstor.org/stable/3110375 .
Accessed: 24/05/2012 15:03

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and American Society for Public Administration are collaborating with JSTOR to
digitize, preserve and extend access to Public Administration Review.

http://www.jstor.org
 31

PERFORMANCE MONITORING SYSTEMS FOR

PUBLIC MANAGERS

Stan Altman, State University of New York, Stony Brook

The Expansion of Government's Role as a Provider sciousnessof the need for more fiscal control, cost effi-
of Services ciency, and improved effectiveness in government services
provided.
Failureto realizethe promiseof the 1960scan be traced
A majorshift in the role of governmentas a providerof to limitationson government'sability to develop policies
servicesoccurredduringthe 1960s. The JohnsonAdmin- that solve social problems. Transportation policies
istration's recognition of social problems in American directed at building highways and urban and suburban
societyand the need to take action to solve themled to the housingpolicies,for example,creatednewproblems,with-
developmentof the social goals of the "Great Society." out solving the problemsthey were originallydesignedto
Broad social programswere undertakento improve the address.Limitationsalso exist in government'sability to
plight of the poor, improve educational opportunity, implementpolicies,evenwell designedones. Pressmanand
increase health and medical service, and expand public
Wildavsky'sbook2,Implementation,is a classiccase study
legal services. As these programs focused funds and of just sucha situation.
personnelfrom federal, state, and local governmentson
social ills, Americanswere encouragedto expectand ulti-
mately demand more and better governmentservices. It
soon becameclear,however,that a commitmentof dollars Students and practitioners in government
and personnelwere not sufficientto ensurethat the social have now come to realize that policy mak-
goals and performanceobjectivesof theseprogramswould ing and implementation are two separate
be realized. One of the many lessons learned from this
period of social experimentationwas that "you cannot processes, each with their own set of actors
solve a problemby throwingmoneyat it." and each requiring different sets of skills.

Problems Confronting State and Local In retrospect, it has become clear that the managerial
Governments as Providers of Services technologies employed by state and local government
agencieswere inadequateto meet the demandsplaced on
them since the 1960s. The administrationof government
Gaps betweenpublic expectationsfor more and better
government services and the government's ability to programswas often executedwith a limitedunderstanding
of the dynamicsof bureaucracies,inadequatefeedbackon
providethem have producedpressuresat all levels of gov-
ernment.Thesepressureshavebeen particularlyevidentat programoutputs,and poor managementcontrolsover op-
the state and local levels. State and local governments,as erations. Lack of adequatemanagementtechnologiesin
direct providersof governmentservices, have borne the governmentagenciesmadeimplementationof new policies
bruntboth of public dissatisfactionwith governmentand and proceduresa risky and often unsuccessfulbusiness.
of the increasedcost of financinggovernmentservices. Studentsand practitionersin governmenthave now come
In the early 1970s,EdwardHamilton,thenbudgetdirec- to realizethat policy makingand implementationare two
tor of New York City, describedthe cause of the financial separateprocesses,each with their own set of actors and
eachrequiringdifferentsets of skills.
pressuresconfrontingthe City by saying that "revenues
are increasingat 5 per cent per yearwhileexpendituresare Repeatedexperiencesof risingcosts, public dissatisfac-
increasingat 15 per cent per year."' By 1975 it was clear tion, and unsuccessfulimplementationefforts have had
the effect of shifting state and local government attention
that the problem of growth in expendituresoutpacing
growthin revenueshad becomewidespread,touchingstate away from broadpolicy makingto improvedimplementa-
and local government units throughout the country. tion throughbetter public management.To improve the
Attemptsto addressthis problemby increasingtax revenue
only intensified the dissatisfaction that already existed Stan Altmanis associateprofessor,W. AverillHarrimanCollege
within the public over the cost of financinggovernment of UrbanandPolicyScience,StateUniversityof New York,Stony
activities. Brook. His major interests are in managementinformation
The concernfor controllingthe burgeoningcost of gov- systems,developingstrategiesfor improvingthe operationof gov-
ernmentservicesand the perceiveddeclinein the qualityof ernmentagenciesand frameworksfor analyzingandimplementing
servicesprovidedhas becomea matterof great interestto publicpolicy. He is the directorof the Instituteof PublicServices
governmentofficials and has resultedin an increasedcon- Performance,Inc., andeditorof the Journalof UrbanAnalysis.

JANUARY/FEBRUARY1979
32 PUBLIC ADMINISTRATION REVIEW

managementof government services requires improve- A performancemonitoring system is purposeful. Its

mentsin the informationsystemsand controlmechanisms purposeis to providemanagementwith beacon lights for
currentlyavailablein governmentagencies. steeringthe serviceorganization'scourse.To steera course
impliesa destinationor a senseof direction,and in organi-
zationalterms,destinationsand directionsaredescribedby
If managers are to become the backbone of goals and objectives.
government agencies, then management in-
formation systems must become the nervous Relationships Between Activities
system of government agencies. That Goals and Objectives Direct

Thereare differencesin the use of goals and objectives

If managersare to becomethe backboneof government to define what is to be achieved.Goals are
generalstate-
agencies,then managementinformationsystemsmust be- ments of desiredsocial or organizationaloutcome. Goals
come the nervoussystemof governmentagencies.Efforts are
generallynot time limited.Theygenerallydeal with ab-
to develop managementinformationsystemsas part of a stractionsand
qualitativelydescribedesired values, atti-
largerprogramof publicmanagementreformsin stateand tudes,appreciations,andunderstandings.
local government agencies have provided the initial Social goals and organizationalgoals differ in termsof
impetus for the developmentof performancemonitoring the universewith which they are concerned.Social goals
systems. are broaderand are statementsof the social ends to which
a government,at any level, aspires. Organizationalgoals
Performance Monitoring Systems are statementsabout the ends to which an organization
aspires.As such, it is only concernedwith the universeover
A performancemonitoringsystemis a systemof infor- which it has, or perceivesit has, control. The organiza-
mation and action. Its function is to provide a dynamic tionalgoals of governmentagenciesare definedin support
view of a serviceorganization'soperationthrougha series of the social goals espoused by the governmentwithin
of static snapshots.It providesmanagementwith a tool to whichit functions.
detect organizational breakdowns. These systems are Eliminatingpoverty, improvingeducationalopportun-
based on establishingmeasurableobjectives,definingthe ity, increasinghealthand medicalservices,and expanding
operationsor work processesnecessaryto achieve these publiclegalservicesareexamplesof socialgoals. Reducing
objectives, and then monitoringthe performanceof the the administrativecost of publicwelfare,reducingthe ab-
work processesand the attainmentof the objectives.This senteeism in public schools, reducing waiting times in
enablesthe managementto determine: healthservicesclinics, and increasingthe numberof cases
1. Whetherthe organizationis achievingits objectives; handledby legal servicesare examplesof organizational
2. Problemsin workprocessesthat may be impedingor goals. Both types of goals are generally the concern of
facilitatingattainmentof organizationalgoals. policymakers.
Goals do not tell what specificallyis to be achieved.To
All performancemonitoringsystems are composed of direct the activities of an organization, more specific
threemajorcomponents:a data component;an analytical statementsof desiredends are
required.Performanceob-
component; and an action component.3 The data jectives are "quantifiable, time-limited statements of
componentprovidesthe frameworkfor measuringandcol- plannedresults."4Theyembodythe theoryof whatorgani-
lectingdatacontaininginformationaboutthe performance zational performancehas to be achieved to realize the
of that which is being monitored. The analytical com-
goals of the organization.Thus, at the very least, a per-
ponent provides the frameworkfor processingthe data formanceobjectivemustinclude:
from the data componentto extractthroughanalysisthat
informationwhichis to be disseminatedto thosewho need 1. A PerformanceStatement:Whatis expected(in mea-
to know. The action componentprovidesthe framework surableterms)to be done;and
for actingon the informationdisseminatedby the informa- 2. A CriterionStatement:How well the plannedactivity
tion component by discriminatingfrom among all the must be performedto be consideredacceptable.
actions possible those for
appropriate dealing with prob- Importantconditionsunderwhichthe performanceis to
lems resulting from detrimentalorganizationalpolicies, be achievedshould also be includedin the statementof
inadequatejob training,inadequatejob engineering,and objectives.5An example of an acceptable performance
unacceptableworkerperformance. objectiveis:
The continuous nature of the information provided by a "Withinthe next 24 months,to reducethe fatal accidentrateper
performancemonitoring system suggests that the ways 100 million passengermiles driven from 215 fatalities to 2.0
data are collectedand processedshouldnot interruptbasic fatalities."
organizationalroutines.The potentialuse of information This performanceobjectivealso tells us what performance
provided by a performance monitoring system to guide indicator is to be measured, namely: Fatal accident
correctiveaction suggeststhat informationshouldbe pro- rate/100millionpassengermiles.
duced in sufficient time to permit organizationaladjust- Achievinga performanceobjectivedoes not ensurethat
mentsto be made. an organizationalgoal is achieved. For example, assume

JANUARY/FEBRUARY 1979
POLICY ANALYSIS IN STATE AND LOCAL GOVERNMENT 33

an organizationalgoal of a statedepartmentof transporta- "Providelife sustaining

financialaid for thosepersonswithno
tion is to increasethe safety of drivingon the state roads. meansof support."6
Then, achieving the performanceobjectives of reducing As currentlyworded, this statementis a goal and not a
fatal accidentsdoes not ensurethe organizationalgoal is
performanceobjective.It is a desiredconditionthe Depart-
achieved, because more non-fatal accidents may be mentwouldlike to achieve.In fact, however,the provision
occurringat the sametime fatalitiesaredeclining. of financial aid (public assistance) was more narrowly
Within an organizationthere are severallevels of per-
definedin the New York State Social ServiceRegulations
formanceobjectives.Thereare performanceobjectivesfor
(Title 18, New York Code of Rules and Regulations, 18
programsand more specific objectiveswhich describethe N.Y.C.R.R.) which covers public assistance. The New
work levels that must be achievedto ensurea program's YorkStateRegulationscontainthe followingprovisions:
performanceobjectivesare met. Performanceobjectives
tiedto specificworktasksarecalledtask objectives. 1. To receivepublic assistancean individualor family
Performancemonitoringsystemsrequirethat these dis- mustapplyfor it;
tinctions be made. Unless managersare able to identify 2. Eligibility for financial assistance is determinedto
explicitly the theories and assumptions by which they exist when there is a deficit between the amount
designwork, procedures,and other organizationalinstru- requiredto supportan individualor family as speci-
mentsfor directingactivities,thereis no guaranteethat the fied in the regulationsestablishingthe Standardsof
organization'sactivitieshave any relationshipto the orga- Assistanceand the resourcesavailableto the individ-
nization'sgoals or to largersocialgoals. ual or family.
Relationshipsbetweenperformanceobjectivesand task 3. A decision concerningeligibility must be made by
objectives must also be explicitly stated if the linkages local social serviceoffice withinthirtydays of apply-
between the information provided by a performance ing. In the case of an emergency,needs(as definedin
monitoringsystem and the correctiveactions to be taken the regulations)areto be met immediately.
wheneverbreakdownsare identifiedare to be effective.By
monitoringtask activity, it is possible to determinehow Using these restrictionsfrom the New York State Social
work patternsaffect the performanceobjectivesthey were ServiceRegulation,the originalperformanceobjectivecan
designed to meet. The different levels of performance be restatedas:
objectivesand task objectiveshelp assign responsibilityto Toprovidewithinthirtydaysof applying, orimmediately inemer-
managers at different levels of an organization. This gencysituations, financial
assistance
to 100percentof theindivid-
providesa frameworkfor establishinga systemof account- ualsor familieswhoseavailableresources(?35.2)are less than
ability. Table I summarizesthe hierarchicalrelationship Standards for Assistance
definedin Section352.1of NewYork
betweengoals and objectivesand governmentunits they StateSocialServiceRegulations.
aredesignedto direct.
In comparingthis performanceobjective to the goal
Thereis an hierarchicalrelationshipbetweenthe goals
and performance objectives of a service organization. originallystated, it should be clear that it is possible to
Goals define what the service organizationdesiresto ac- satisfy the objectivewithout satisfyingthe goal. The goal
impliesthat anyonein need of assistanceshould get it but
complish, and performanceobjectives define what the the performanceobjectivepoints out that New York State
organizationhas to accomplish. Goals are global and Social Serviceregulationslimit assistanceto those in need
missionorientedand performanceobjectivesare local and
who apply for it. Unless there is another performance
taskoriented.
objective defined for an outreachprogram, eligible seg-
mentsmightnot knowto applyfor assistance.
TABLE I The restated performance objective would allow a
numberof task objectivesto be identified. For example,
HIERARCHICAL RELATIONSHIP workersresponsiblefor processingapplicationsmust com-
BETWEEN GOALS & OBJECTIVES
plete their work on an applicationwithin 30 days. If the
& GOVERNMENT UNITS workerswho verify the correctnessof documentationpro-
vided by applicantsare different than the workers that
Level Goal or Objective
Government process the application, then they must complete their
SocialGoal workin sufficienttime to allow the processof applications
GovernmentAgency OrganizationalGoals to be completedwithin thirty days. To carry out these
Program PerformanceObjectives
tasks, the coordinationof two separatebut dependentsets
Operational TaskObjectives of activitiesis required.
The restatedperformanceobjective also defines other
tasks. For example, processed applications must be
auditedto determineif they are being processedcorrectly
Most public service organizations,particularlygovern- and to determineif the impliedzero errorrate specifiedin
mentagencies,confusegoals and objectives.For example, the performanceobjectiveis beingachieved.By addressing
New YorkCity'sDepartmentof IncomeMaintenancePro- this task, the question of achievinga zero error rate in
gramsdefinedone of its performanceobjectivesfor fiscal processing applications becomes explicit and must be
year 1976-77as: directlyaddressed.

JANUARY/FEBRUARY1979
34 PUBLICADMINISTRATIONREVIEW

The dilemma and challenge which confronts policy- Performance Data

makersand managersis bridgingthe gap betweenpublicly
stated goals and organizationallyachievableperformance Data are to a performancemonitoringsystemas founda-
tions are to buildings.Data providethe factualfoundation
objectives.If performanceobjectivesprovidea framework
for monitoringperformance,then performancemeasures upon which efforts to directthe behaviorof an organiza-
tion are based. Data are used to describedifferentcharac-
provide the data required by performancemonitoring
teristicsof an organization.Everyoperationis susceptible
systems.
to data collectionwithin existingtechnologyand resource
limitations.Data often exist only partially,particularlyin
Measuring Performance governmentand public administration;what is not meas-
ured or measurableis an importantpart of organization
and operation!Therefore,thereis always some degreeof
Types of Performance Measures
uncertaintyassociatedwith conclusionsdrawnfrom data.
To describean organization'sperformance,three gen- It is true that a full descriptionof the organizationcan be
eral classes of measuresare used: effectivenessmeasures; obtainedby increasingthe numberof organizationalchar-
efficiency measures, and workload measures. Effective- acteristicsfor which data are collected.However,such an
ness measuresprovideinformationon the organization's approachbringsalong with it increasedcomplexityin col-
progress toward achieving its program objectives. Effi- lectingandinterpretingthe data.7
ciencymeasuresand workloadmeasuresprovideinforma- In developinga system of performancemeasures, the
tion on the organization'sprogresstoward achievingits questionwhichmost often arisesis "to what extentdo the
task objectives.Brieflydefined, performancemeasuresrespondto the need for informa-
tion?" This questionhas elicitedstrongdebatefrom vari-
Effectivenessmeasuresdescribethe extent to which
the servicesprovidedmeet definedperformanceob- ous research groups advocating either "crude" or
"precise"measures.However,the decidingfactor in such
jectives.
a debatemust be to ensurethat the data collectedreflecta
Effectivenessas definedhere is broaderthan the notion serviceorganization'sperformance,as stated in its objec-
of publicsatisfactionwith the servicesprovided.Effective- tives and, secondarily, that the data reflect significant
nessis definedin organizationalterms;i.e., has the organi- changesin a serviceorganization'sperformance.As used
zation achieved the performanceobjectives established? here, significantchange is defined in organizational,not
Thedefinitionof whatis effectivemaybe shapedby public statisticalterms.Therefore,the questionmustbe answered
concerns,by legal requirements,or by decisionsinternalto on a caseby casebasis.8
the organization.Effectivenessgenerallyis not concerned To decide how many and how precisethe performance
withthe cost aspectsof performance. measuresusedshouldbe requirescarefulthoughtand anal-
Efficiencymeasuresdescribethe magnitudeof serv- ysis. This is not a simpletask to perform.Yet, if not per-
icesprovidedat anygiven level of humanand capital formed, the data collectedmay prove useless. New York
resources. City's productivitymeasurementprogramis an example
wheredata werecollectedusing hundredsof measuresbut
Efficiencyas defined here is used interchangeablywith littleuseful informationresultedbecauseit was neverclear
the notion of productivity.Efficiencycan be measuredin whatthe data meant.However,the principleof parsimony
absolutedollar terms. However,it has becomea common shouldbe kept in mindin developinga systemof perform-
practiceto expressefficiencyas the ratioof dollarworthof ance measures.Also, it should be rememberedthat in de-
outputsto dollarworthof the inputresourcesused to pro- ciding what action is to be taken to correctan organiza-
duce the output. Conceptualproblemsin defining dollar tionaldeficiency,morepreciseinformationwill be required
worth of outputs have led to alternativedefinitions.The than a performancemonitoringsystem will provide. The
most commonalternativedefinitionis: Numberof output job of the performancemonitoringsystem is to indicate
units of servicesproduced/person-hour.This definitionis the extent to which an objectiveis reached.Furtheranal-
particularlycommonin publicserviceorganizationswhere ysis and evaluationof the actualsitutationis calledfor be-
laborcosts arethe largestcomponentof inputcosts. fore makinga decisionto takeaction.
Workloadmeasuresdescribethe type of servicesre-
questedand the amountof workrequiredto deliver What is the Future of Performance
them.
Monitoring Systems?
Workloadas defined here can differ from output be-
causeorganizationscan developbacklogswhichwill not be Performancemonitoringsystems will have an increas-
reflectedin its reports.Anotherreasonfor includingwork- ingly importantrole to play in the managementof govern-
load is that parts of it may be generatedby the organiza- ment service agencies. Students of governmenthave fo-
tion itself and as a resultmay be controllableby improving cusedon policy makingsince the existenceof government.
the organization'soperation. Policy makers focus on decision making, not execution.
Eachof the threebroadcategoriesof performancemeas- However,like a generalwithout an army, policy makers
ures identifiedabove have subcategorieswhichdefine dif- cannot run a governmentwithout management.This is
ferentaspectsof the measure. particularlytrueof stateand local governmentsbecauseof

JANUARY/FEBRUARY1979
POLICY ANALYSIS IN STATE AND LOCAL GOVERNMENT 35

their direct responsibility for providing public services ra-

ther than simply regulating service delivery mechanisms.
What factors suggest a bright future for performance
monitoring? For one, the recent rebirth of concern for Help your students
public accountability of public management. This concern
for accountability suggests that support for public pro- better understand
grams will be determined in the future by the ability of
public managers to achieve their program objectives. In their role
addition, public managers have come to realize that they
need more timely information than is currently available to in determining
them. With the use of advanced data technology, one can
expect to see major growth in performance monitoring sys-
tems at all levels of government and for all programs.
public policy...
Moreover, one can expect that performance monitoring
will be included as an integral component of public pro-
grams.
Here's how:
There are other factors, of secondary importance, which
will play a role in the future of performance monitoring A New Book.
systems. Performance monitoring systems' objectives are BUREAUCRACY,POLICY
focused and limited. The corrective actions it requires are AND THE PUBLIC
incremental rather than global. In addition, because its
corrective actions are oriented toward improving the use of By StevenThomasSeitz.April,1978. 226 pages,
129 illustrations.
Price,$9.95.
existing resources and existing organizational arrange-
ments, it is less likely to unsettle labor/management re- Contemporary in everyway,this newbook offersstudentsa
lationships-in contrast with the controversy that often introduction
well-written to publicadministration.Theauthor
drawsfroma varietyof alliedfieldsas he examinesindetailthe
surrounds productivity programs. Finally performance functionof bureaucraciesin modernsociety, bureaucratic
monitoring systems can provide managers with feedback decision-making,and the relationshipof bureaucracyto
on the status of efforts to implement new programs, proce- democracy.Ratherthan treatadministrationas an endinitself,
dures, and policies. he emphasizestherecenttrendtowardrelatingadministration
andmanagementdirectlyto publicpolicy.Hisbook:
* Includesfar more solid researchand empiricaldata
thanthe usual"descriptive" booksofferedinthisarea;
* Discussesthe historicaldevelopmentof bureaucracy
andlinksthisto currentconcerns;
Notes * Relatesto the roleand problemsof publicpolicyin a
practicalandusablefashion,andfeaturesa uniqueemphasis
on the publicsector;
* Considerssuch contemporaryissues as the public
1. EdwardK. Hamilton, "Productivity:The New York City
administrator'srole in coping with the energy crisis,
Approach,"PublicAdministrationReview,VolumeXXXII, intemationalaspects,andecology;
Number6, November/December,1972,p. 784 (Washington, * Providesan excellent,timelychapteronadministrative
D.C.) laws;
2. Jeffrey Pressmanand Aaron Wildavsky,Implementation: * Includesa teacher'sguideandtest manual.
How GreatExpectationsin WashingtonAre Dashedin Oak-
land;Or, WhyIt's AmazingthatFederalProgramsWorkAt
TABLEOF CONTENTS (Abridged)MACRODYNAMICS OF
BUREAUCRACY * Birthof bureaucracy* Foundationsof
All, (Berkeley:Universityof CaliforniaPress, 1973). *
bureaucracyBureaucracyand *
productionBureaucracyand
3. Stan Altman, "Approachesto MonitoringProductivityand social control * Bureaucracyand intemationalrelations*
Performance,"Institute of Public Services Performance, MICRODYNAMICSOF BUREAUCRACY* Multiple
NewYorkCity:1976.
bureaucracies* Bureaucratic leadership* Bureaucracyand
cybernetics * Bureaucraticmanagement * Bureaucratic
4. JohnC.W. Cadoo, Jr., "Measuringa Government'sProduc- decision making * APPLICATIONS * Bureaucracy and
tivity In the State of Washington,"PublicProductivityRe- democracy * Glossary
view, Volume 1, No. 3, Centerfor ProductivePublicMan- Formoreinformation on thisbookorotherMosbytexts(orto
agement, John Jay College, City Universityof New York, havea sales representative contactyou), simplycalltollfree
1976,New YorkCity.p. 36. (800) 325-4177 andask forextension30duringourregular
5. Robert F. Mager, "PreparingInstructionalObjectives," business hours. A90097
(Belmont:FearonPublishers,Inc., 1975.) Priceeffectivein U.S.A.only.
6. ManagementPlan of the City of New York, Fiscal Year
1977,HumanResourcesAdministration:Departmentof In-
comeMaintenancePrograms.
MOSBY
7. JohnTukey,"TheFutureof DataAnalysis,"Proceedingsof TIMES MIRROR
the 10th Conferenceon the Design of Experimentsin Army THE C. V. MOSBY COMPANY
11 830 WESTLINE INDUSTRIAL DRIVE
Research, Development and Testing, pp. 691-729, (U.S. ST. LOUIS, MISSOURI 63141
ArmyResearchOffice, Durham:1965.)
8. John Tufte, Data Analysis in Public Policy and Politics,
(EnglewoodCliffs:PrenticeHall, Inc., 1973),p. 71.

JANUARY/FEBRUARY 1979