ED 134 114

AUTHORTITLE

INSTITUTION

SPONS AGENCY

PUB DATEGRANTNOTEAVAILABLE FROM

J *

° 15-t4104.4DOCUBBI RESUME

H4 008 596

Purls, Ralph A.; Glenny, Lyman A.Stat Budgeting for Higher Education: InformationSystms and Technical Analyses.California Univ., Berkeley. Center for Research andDevelopment in Higher Education. .

Depaitment of Health, Education, and Welfare,WashIngton, D.C.; Ford Foundation, New York, N.Y.76730-01552; NE-G-00-3-0210246p.'s--Centei for Research and Development in igherEducailon, University of CalifOrnia, Be4kêiey,CalifOrnia

EDRS PRICE MF-$0.83 HC-$12.71 Plus Postage.DESCRIPTORS *Budgeiing; *Data Analysis; *Educational Finance;

EducatiLonal Planning; Higher Education; InformationSystemp; Program Planning; *State Agenci s;*Stateyide Planning

IDENTIFIERS California; Colorado; Connecticut; Flori a; Hawaii;IllinOis; Kansas; Michigan; Mississippi; Nebraska;New YOrk; Pennsylvania; Tennessee; Texas Virginia;Washirigton; Wisconsin

ABSTRACTIThe eitent to which state agencies are implementing

information systems and analytical methods for budget re

i

iew areexamined. Focus is on '17 states: Califormia.-Colorador-Cn necticut-e-Florida, Hawaii, Illinois, Kansas, Michigan, Mississippi Nebraska,New York, Pennsylvania', Tennessee, Texas, Virginia, Wash ngton, andWisconsin. Trends in baget information to state agencie and thebases for current concerns are reviewed. Higher educatio budgetinformation categories are defined and a typology of inf rmation usesis presented. An overview is given of the trends in info mation andanalysis activities typifying each of the state budget a encies, andthe principal styles of budget review used in the states are alsodescribed. The informational content of state budget doc ments is,delineated, followed by a discussion of the technical an politicalconsequences of using information and analysis systems. he reportconcludes with a discussion of several major considerati ns to beweighed in setting up state-level information and analysis systems.(1BH)

***********************************************************************Documents acquired by ERIC include many informal unpublished* materials not available from other sources. ERIC makes every effort ** to obtain the best copy available. Nevertheless, items of marginal ** reproducibility are often encountered and this affects the quality ** of the microfiche and hardcopy reproductions ERIC makes available ** via the ERIC Document Reproduction Service (EDRS). EDRS is not* responsible for the quality of the original document. Reproductions ** supplied by EDRS are the best that can be made from the original. *

***********************************************************************

RALPH A. PURVES

LYMAN A. GLENNY.

CENTER FOR RESEARCH A\ N'_-) L)EVELOPMENT N HIGHER EDUCATIONUNIVERSITY OF CALIFORNIA BERKELEY

c'ENTEn Fon n"sEncli ANn DI.:V1-1.onN11.1\"1

UNIVElts1T1' 01. CALIF(MNIA.

e Center for lksearch and De \ elopment in I ligher Edo-Th

catok, i,.,...,,,., in research designed to assist.individuals and organi-zations responsible for Anwrican higher education to improve thequality. efficiency, and availability of education beyond the high school.In the pursuit .of these objectives. the (enter conducts studies which:I ) use the theories and methodologies of the behavioral sciences; 2)seek to discover and to disseininate new perspectives on educationalissues and new solutions to cthicational problems::: seek to add sub-

...,,4stantially to the descriptive and analytical literatT on colleges anduniversities; :il contribute to the systeinatic knowledge of several of thehehavioral s(inces, notably psychology, sociolog4 economics, and

,political science; and 5) provide models of research 'and developmentactivities for colleges and universities planning tuid pursiiing their own

..-priTrams in institutional research.

el

suprord . . L. 47E-f.;-00-3-0210lie21.7,1:, Eda3ation, an4 ;,17,fare, and Urant'

fr7m under:42ng suchjcocruncnt eponorz,kip aro .:nc 3ura:x.-.1. ac'press freely

judg:rr:,?nt Zt :.he Points of14, ropresent th

0.1 .izotit:Ato ,:-,f,Elitc,a-tion ortl2e

State Bucketing for Higher Education:

lacmination Systems andTechnical Analyses

RALPH A. PURVES

LYMAN A. GLENNY

CENTER FOR RESEARCH AND DEVELOPMENVIN HIGHER EDUCATIONUNIVERSITY Of CALIFORNIA, BERKELEY, 1976

.I.rnder_tlw general. title State Budgeting for Higher Education theCenter is issuing nine publications, each with its own subtitle'andauthors. The volumes report three separate but interrelated proj-ects carried on from July 1973 to August 1976, funded as follows:one on state fiscal stringency by the Fund for the Improvementof Postsecondary Education (FIPSE), another on state generalrevenue trends by the Lilly Endowment and the American Coun-cil on Education, and the third on selected aspects of state budg-etary theorY and practice by a joint grant from the National In-stitute for Education and the Ford Foundation. The principalinvestigator for all the projects was.Lyman A. Glenny; the princi-pal author or authors of each volume carried the inajor respon-sibility for it. To varying degrees, all members of the researchteam contributed to most of the volumes, and their contributionsare mentioned in the acknowledgments. This report is the fifth tobe issued in the series.

LIBRARY oF CONGRESS CATALOGING IN PUBLICATION DATA

Purves, Ralph A 1937-

Information systems and technical analyses.

(State budgeting for higher education)Bibliography: p.Includes index.1. Education--United States--Finance.

'2. System analysis. I. Glenny, Lyman A., jo'.nt

.author. I/. Title. III. Series.

L82825.1387 379'.1214'0973 76-51452

Acknowledgments

We are especially grateful for the financial supportprovided jointly by the National Institute of Educationand The Ford Foundation. This cooperative venture ofpublic and private funding agencies has provided a veryhospitable environment for research.

An exceptionally high degree of cooperation andassistance was provided by state and instit'utional officials,often at considerable time and expense and, in many in-stances, during particularly busy periods. Needless tosay, the study was entirely dependent on such help and weare especially grateful for their generous contributions.Answering questionnaires, providing documents, arranginginterviews, and responding to letters and telephone in-quiries are not the most rewarding.activities for some ofthe busiest officials in state government. Yet all weremost helpful.

The representatives of the national organizationswho comprised our National Advisory Panel and the indi-viduals who comprised the Technical Advisory Panel madesignificant contributions to this report through theircritical comments and useful suggestions in response toproject plans and report drafts.. . Their assistance broad-ened our perspectives and provided many useful insights.We are most appreciative. The members of both panels

-dre-listed below.

NATIONAL ADVISORY PANEL

American Association of Community and Junior CollegesDr. Edmund J. Gleazer, Jr., Executive Director

American Association of State Colleges and UniversitiesDr. Allan W. Ostar, Executive Director

American Council on EducationDr. Lyle .H. Lanier, Director, Office of Administrative

Affairs

Education Commission of the StatesDr. Richard'M. Millard, Director, Higher Education

Services

National Association of State Budget OfficersMr. Edwin W. Beach, Past President

National Association of State Universiti,i-and Land-Grant Colleges

Dr. Ralph K. Huitt, Executive Director

National Center for Higher Education ManagementSystems

Dr. Paul Wing, Senior Staff Associate

State Higher Education Executive Officers AssociationDr. Warren G. Hill, Past President

TECHNICAL ADVrSORY-PANEL

Dr. Frederick E. BalderstonChairman, Center for Research in Management ScienceUniversity of California, Berkeley

Mr. Edwin W. BeachChief, Budget DivisionCalifornia State Department of Finance

Mr. Harry E. BrakebillExecutive Vide .ChancellorCalifornia State University and Colleges

iv

Dr. Allan M. CartterGraduate School of EducationUniversity of California, Los Angeles

Mr. Loren M. FurtadoAssistant Vice President--Budgeting and Planning,

and Director of 'be BudgetUniversity of California

Dr. Eugene C. LeeDirector, Institute of Governmental StudiesUniversity of California, Berkeley

Mr. John C. Mundt, Executive DirectorWashington State Board for Community College 'Education

Dr. Morgan Odell, Executive SecretaryAssociation of Independent California Colleges

and Universities

Dr. Aaron B. WildavskyDean, Graduate School of Public PolicyUniversity of California, Berkeley

Dr. Paul E. ZinnerProfessor of Political ScienceUniversity of California, Davis

In addition, we gratefully acknowledge the helpfulcriticism and comment of Frank Bowen, Anthony Morgan,Mel Orwig, and Frank Schmidtlein, who reviewed drafts

may not necessarily agree with the final form theircomments have taken, but they have immeasurably assistedus An_avoiding errors of fact and interpretation. _Weespecially thank Janet Ruyle for guiding the reportthrough production, Harriet Renaud fOr editing the finaldraft, and Mildred Bowman and Margaret Elliott for pre-paring several versions of the manuscript, includingthe final one. .

Preface

From.July 1973'-'to August 1976 three studies of statebudgeting and financing of higher education were conductedby the Center for:Research and Development in Higher Educa-tion at the University of California, Berkeley.

The present study began in July 1973 when the Centerundertook a three-year, 50-state study of the processesused bY state agencies to formulate the budgets of collegesand universities. Seventeen states were studied in-

tensively.*

Financial support was furnished jointly by the eNational Institute of Education (60%) and The Ford Founda

:ftion (40%). The study was endorsed by the following organ-.

American Association of Conmunity and Junior

CollegesAmerican Association of State Colleges and

Universities

* The 17 states were: California, 'Colorado,

"Connecticut, Florida, Hawaii, Illinois, Kansas, Michigan,iMississippi, Nebraska, New York, Pennsylvania, Tennessee,Texas, Virginia, Washington, and Wisconsin.

American Council on EducationEducation Commission of the StatesNational Association of State Budget OfficersNational Association of State Universities

and Land-Grant CollegesNational Center for Higher Educatim Management

SystemsState Higher Education Executive Officers

Its twofold.purpose is to advance budgetary theory and togive state and institutional budget.professionals a broaderunderstanding of: 1) the interrelationships, roles,functions, and objectives of the several state agenciesin the budgetary process; 2) the congruence or incongruenceof such objectives among the several agencies; and 3) thepractices and procedures that build confidence in thefairness of the budgetary process.

Reports based on the study describe and analyze theorganizational st7uctures and staffing of state-levelagencies and the progress of institutional budget requeststhrough these agencies from the time that prebudget sub-mission instructions are first issued by a state agencyuntil appropriations are enacted. The primary emphasis ison the budget review and analysis process and the proceduresused by the state agencies; the study concentrates on theadministrative interfaces among the several state agenciesthat review and analyze budgets and between these agenciesand the institutions, or systems of institutions, ofhigher education.-

Latensive-interviews-r-doeument-review-i-and-questionnairesjn the 17 states selected formed the basis for anarrative and tabular description and comparison issued

1975.. Less detailed data were collected from 50 states-by questionnaire only; these are examined and presentedin a second.descriptive report.

The other volumes resulting from the three-yearstudy are analytic in nature. This volume focuses on thedevelopment and use of information systems and analytictechniques; Others concentrate on the creation and useof budgetary formulas, the cooperation, redundancy, and

vii

duplication of effort among the beveral state agenciesthat review budgets, and the dilemmas involve& in thedesign of budget processes, along with a step-by-step&nalysis of budget progress througb the labyrinth ofstate agencies and processes.

The second study, sponsored by the Fund for theImprovement of Postsecondary Education (PIPSE), examineshow state colleges and universities respond when statesmake substantial reductions in their appropriations.This one-year study encompasses experience with fincalstringency in about a dozen states, primarily in tne fivestates presented in the case studies. The latter havebeen brought up-to-date as of late spring 1976.

The third study, sponsored by the Lilly Endowmentand the American Council on Education, analyzes the trendsin state general revenue appropriations for higher educa-tion from 1968 to 1975. Refining earlier work at theCenter, the study compares trends among the states forthe several types of institutions in both appropriatedand constant dollars, comparing dollar increases withenrollment trends in each case and also comparing dollarsappropriated for higher eddcation with those for elementaryand secondary education.

Each volume resulting from the three studies drawson significant findings of the other studies yet standsalone as a complete book. However, awareness of the fullpanoply of social, political, and economic variables thatwe found in state budgeting for higher education can begained by-review-of-wkl-the volumes. We-earnestW-hope_the readers learn as much from our resealch as we did inconducting it. 4 complete list of the volumes is foundon the back cover of this book.

viii

Acknowledgments

Preface

1 Introduction

Contents

2 Budgetary and Financial Information:Taxonomies, Uses, and Systems/

Page

vi

1

3 Information and Analysis Developmentin State Agencies: An Overview 30

4 Styles of Higher EducatiOn Budget Review 44

/5 The Context of Formal/Budget Documents 83

e

6 Payoffs from Anaiiisis_and Information Bystems 127

/7 Znformation and/Analysis Systems:

Considerations fcr Design 162. .

Raerences

General/Bibliography

Appendix 1-1. NACUBO Current Funds ExpendituresAccounts, 1968

198

207

209

Page

Appendix A-2. NACUBO Current Funds ExpendituresAccounts, 1974 212

Appendix B. HEGIS Taxonomy of InstructionalPrograms in Higher Education 216

Appendix C-1. Number of Four-Year Public Insti-tutions with Budgets Reviewedby Techniques Discussed 228

Appendix C-2. Number of Two-Year Public Insti-tutions with Budget ReviewsSimilar to Four-Year Institu-tions 229

Appendix D. An Interpretation of a Non-ZeroSum Game: The Prisoner's Dilemma 230

TABLES

1 State Higher Education Agency (SHEA)Responsibilities, 1974-1975

2 Number of Four-Year Campuses without MedicalPrograms Reviewed, by Inclusion in Multi-campus Systems, Focus of Review, and Level ofInforamtional Detail Provided

3 Budget Data Submitted to Executive andLegislative Agencies

4 State Agencies that Prepare Formal BudgetDocuments

EXHIBITS

Page

42

89

92

109

1 Organization of the Program ClassificationStructure 18

2 State of Connecticut Objects of ExpenditureClassification

3 Three Formula Methods for Developing BudgetEstimates of Instructional Costs

4 U.S. Bureau of the Budget PPB Timetable

5 Formal Education Program Structure inthe State of Hawaii

6 Measures of Effectiveness and Units of Measurein the State of Hawaii

7 Summary Decision (SD) Itc, , University ofWisconsin System

8 Examples of Unit Decision Items, Universityof Wisconsin, Madison Campus

9 Proiram Measures Required in PennsylvaniaBudget Submissions, Fiscal Year 1974-1975

10 Overlap of Budgeting Phases

11 Contradictory Elements in Organization andEvaluation

xi

14

50

56

63

66

67

76

78

102

137

184

11.

Introduction

In all, public colleges and universities in theUnited States currently receive approximately 40 percentof their operating income from state tax appropriations.This figure rises to 60 percent when income from noneduca-tional activities and services such as auxiliary enter-prises is subtracted (Carnegie Commission, 1973). Withthe advent of comprehensive executive budgets in moststates, virtually all institutions now request these fundsby means of a formal request procedure routed through theexecutive budget office. Requests from higher education'institutions are considered along with requests from otherstate services and are ultimately the basis for a legis-lative appropriation, signed by the governor. There isconsiderable variation among states, but on the averageabout one-seventh of all expenditures from state generalfunds is appropriated for the operating expenditures ofpublic higher education (Glenny & Kidder, 1973).

The entire scope of the state higher education budget'process from request to appropriation is considered in an-other study in this series (Schmidtlein & Glenny, in pre-paration), which will report on specific details andvariations in the procedures in use in 17 states. Basedon the same 17-state data, the present report is focusedprimarily on the informational and analytical aspects ofbudget requests to the state and the technical proceduresused by state budget agencies to review these submissions.Particular attention is given to the application of methodswhich objectify and rationalize the budget process, suchas program budget submissions, new information reporting

structures and systems, and various microeconomicanalytical techniques that have been developed forbudget preparation and review.

A considerable volume of literature prescribingthe implementation of budgetary reform has followedthe experience with Programming-Planning-Budgeting(PPB) in the federal government. As yet, relativelylittle empirically based research has been reportedon the implementation of reforms to alter and improvethe technical aspects of the state higher educationbudgeting process. For many years budgetary formulassubsumed a great deal of the information and analysisused in higher education. Miller (1964) reported onthe use of "formulas and cost analyses" in state budg-eting for higher education, and Gross (1973) recentlysurveyed the states to determine the extent to whichformulas are used. Identification of the techniquesof budget development and state-level review withformulas has emphasized the inflexibility of formulas

,

whose application has often tended both to lock infunding inequities as well as preserve levels ofsupport. Under current conditions, budget requestsgenerated by formulas cannot always be funded byavailable revenues; furthermore, institutions havedifficulty adjusting to lower enrollments when funding !

is mechanically tied to enrollments by formulas. Thus,

the greater flexibility and disCretion needed in theuse of resources when funds are limited require amore sophisticated technical review of operationsand issues. Consequently, the term formula is usedin this report in a fairly restricted sense, althoughthe subject matter does coincide generally with whatMiller (1964) has treated in his discussion of thetechnical features of formulas. The report in thisseries on formulas (Meisinger, 1976) treats theorganizational and political uses of budgetaryformulas, but does not describe the technical useof quantitative information and its use in supportof budgets, aspects with which we shall be concerned.

16

2

FOCUS OF THE STUDY

A major portion of this report is descriptive, inorder to provide an up-to-date review of the extent towhich state agencies are implementing information systemsand analytical methods for budget review. This is impor-tant for at least two reasons: A descriptive overviewcan lead to more realistic expectations of what stateagencies can achieve in implementation, and it can iden-tify more clearly the options in budget informationgathering and review that state agencies have open tothem. In aadition to describing what each of the 17states does, we have developed analytical categories ofinformation and its use to show both commonalities anddifferences from state to state.

We have.also attempted to evaluate budgetary infor-mation and review systems by considering some of theimplications and consequences 'cd their _use. That analysiscan improve public decisionmaking is taken as an articleof faith, but current use of analysis in the process mustbe understood before changes in its use can have a bene-ficial effect. One university president interviewedexpressed the belief that the outcome of the budgetprocess in his state had relatively little to do witthtechnical budget review procedures. The procedures, hefelt, were only a way of'rationalizing a decision madeon the basis of political realities. At the same time,it was clear to us that the capacity of interest-grouppolitics for adequately resolving all budgetary alloca-tion decisions is overtaxed by the number and complexityof the decisions that must be made.

The methodology for this study has not involvedthe testing of hypotheses nor the application of rigor-ous analytical techniques. Our objective has been todetermine actual state practice rather than to developbudgetary theory. Budget documents and materials fromeach of the 17 states were reviewed to see the kind ofinformation that was submitted and the manner in whichit was displayed. In addition, site visits were madeto each of the states, and to determine the methods ofreview, all state-level officials in executive, legis-

3

17

lative, or higher education agencies involved in theformal examination of higher education operating budgets

were interviewed. Unless one can actually work in anagency for an extended period, "use" of a system or atechnique is difficult to ascertain. On the other hand,

by_being able to talk with budget process participantsat all levels of the state bureaucracy, the outsider maygain an awareness of features of the process that areobscured from the participants. The description in thisreport is intended to describe actual use, not planned orhypothetical use, although plans are important in suggest-

ing future development. If anything, this report may erron the side of implying too great a level of implementationof standardized information collection and rational quan-titative budget analysis, but the trend in this direction

is clear. Current dissatisfaction with the way publicagencies or institutions are serving public needs canrarely be attributed to the use of too much analysis in

the decision process.

PLAN OF THE REPORT

The remainder of this chapter reviews trends inbudget information to state agencies and the bases forcurrent concerns. In Chapter 2, higher education budgetinformation categories are defined and a typology of inT-formation uses is presented, and the sense in whichmanagement information systems are to be understood in

this report is explained. Chapter 3 presents an over-

view of the trends in information and analysis activitiestypifying each of the state budget agencies. Chapter 4

begins the major description of technical budget reviewby describing the principal styles of budget review usedin the states. Chapter 5 delineates the informationalcontent of state budget documents, and is followdd inChapter 6 by a discussion of the technical and politicalconsequences of using information and analysis systems.The report concludes with a discussion of several major

considerations to be weighed in setting up state-levelinformation and analysis systems.

19:

4

TRENDS IN INFORMATION SUBMISSION

The-submission of information to state budget reviewagencies in support of operating budgets prior to legis-lative review is certainly not a new practice, althoughit is one that has increased dramatically over the lastfew years. Discussions of state budgeting for highereducation, even.for the postwar period, are sparse, butsources such as the Council of State Governments (1952),Glenny (1959), Miller (1964), and Moos and Rourke (1959)have suggested that recent requirements for additionalbudget information were initiated long ago, although per-haps only in more isolated instances. For example, theCouncil of State Governments reported for 1948 that outof 162 governing boards reporting, 140 boards (governing327 institutions in 43 states) submitted their budgetrequests to a central budget authority for revision beforethese requests were consolidated and submitted to thelegislature. Only three boards submitted requests direct-ly to the legislature.

What actually may have changed, more than the sub-stance of what is being reported and the way it is re-viewed, is the changed attitude of the general publicand its elected officials, which now see higher educationas an integral part of the state governmental structure.Public institutions have always faced the problem of howto present budgetary demands to state agencies. Currentcriteria for an effective submission or a justified re-quest may relate more to changes in the environmentalcontext in which higher education operates than to itsintrinsic operations.

Although higher education budget submissions tostate agencies have a long history, changes in variousfeatures of these submissions point up some significanttrends:

Quantity. There has been an increase in theamount of data required of budget submissions asstate agencies have sought information on insti-tutional operations in greater detail.

195

Comparability. Attempts at making data morecomparable through the application of uniformdefinitions have peen intensified.

New categories. Information has been requestedin new types of aggregations and categories--quitefrequently in multiple data structures.

New items. Information has been requested onnew items such as educational outputs, targetgroups, and educational unit costs.

Systems. Attempts to build systems for collect-ing and reporting information, particularly throughthe application of computers, have become common.

THE BASES FOR INFORMATION CONCERNS

Seasoned budget practitioners tend to be unimpressedby the promises made, or the expectations held out for manysuggestions for budgetary reform, including changes ininformation submission. ,Thus, one sees. few states in-which these reforms are championed with the same enthusi-asm that they are in some management consultant circlesor graduate professional schools. Yet, state and insti-tutional officials continue to seek new methods for makingbudgetary choices (with new information requirements) inall public agencies, with perhaps greatest intensity inhigher education. Among the many reasons for thisconcern are seven general factors.

1. There is the general concern over establishingmanagement control in a public agency to foster efficiencyand constraint in the use of public funds. Niskanen

(1972) answered the question "Why new methods?" with thegeneral observation that "Governments do not serve us well."He argued more specifically that the budget process isincoherent, that decisions on the parts are inconsistentwith decisions on the total. "Our political processessuggest an increasing demand for individual programs,but opinion polls indicate a substantial and increasingpopular concern about total federal spending" (p. 156).

6

20..

Further, as important as any temporal factor, accord-ing to Niskanen, is the unique factor inherent in

governmental budgeting which is absent in budgeting forcorporate firms: Governments do not face the conditionsthat make the economic calculus and the balance sheetrelevant to a business firm. Governments are nonprofitmonopolies "whose objectives do not provide consistentlystrong incentives to use the economic calculus" (p. 156),and as a consequence, various budgetary procedures arecommonly seen as a substitute for incentives that areformed in a competitive market.

2. The notion'that a minimal level of statewidecoordination is necessary has taken root in virtually allstates, although the individual states differ with respectto the kinds of higher education decisions they considerto be properly within the domain of state responsibility.In budgetary matters, public institutions have clearlyhad to involve state government officials in the processthrough which they receive appropriations from state taxfunds. A well-established method for achieving coordi-nation, which usually stops short of according actualdecisionmaking authority, is creation of a state-levelagency with the responsibility for knowing what thestate's colleges and universities are doing so that theactivities of one can be related to the activities ofthe others. An important resource in fulfilling thisfunction is a formal information gathering and report-ing system.

3. The belief exists i- oome quarters that admin-istration and management in higher education institutionsare somehow not on a par with, say, the management ofcorporate enterprises of similar size. There is there-fore an interest in strengthening formal procedures,especially in the areas of budgeting and fiscal control,which are the core of management responsibility in theprivate sector. A state legislator reflected this viewwhen he noted that many of his legislative colleagueswho had managed private businesses were astonished bycollege administrators' inability to provide informationthat the legislators themselves had considered essentialin their own businesses. This fueled the move in his

7

2 :I

state, as it.had in others, to require that certain in-stitutional information be provided to state budgetaryagencies as.a way of ensuring that the institutions them-selves had the information.

Even when state officials are not greatly concernedabout the capabilities of institutional administrators

. . .

and their systems of management, they may still expressg wish for more detailed information on institutionsbecause they have substantially less control over collegesand universities than they do over other state serviceagencies. One state director of administration indicated,"I would feel more comfortable if we knew more about theuniversity in detail, even though decisions wouldn't bebased on this detail."

4. In the higher education budget process, at leastthree state-level agencies may be striving to carve outa legitimate role. Introduction of reforms into thebudget process is often either the initial step in shift-ing influence or a direct response to it. Because thestate higher education agency is a new participant inthe budget process in some states, it has a relationshipto establish with both the legislature and the governor'sbudget office. At the same time, many legislatures areseeking an expanded role in state policy along the linessuggested by the Citizens Conference on State Legislatures(1971). Policy involvement requires access to informa-tion, and therefore the attempt to take on new or ex-panded functions by these state agencies quite oftenbegins with a proposal for more systematized informationdistribution. Reforms or innovations of the budgetprocess are rarely neutral with respect to power andinfluence relationships that exist between state agencies,and reforms are rarely accepted wholeheartedly by allagencies involved.

5. In some, but by no means all states, recentlyhired budget staff or their ditectors have been.instilled,through academic training or experience in other organi-zations, with the zeal for activism and reform thatcharacterizes the core of ahy established profession.Professions are built around the practice and applica-

228

ly

tion of a codified bOdy of knowledge, and in budgeting,this body of knowledge or discipline stems largely fromeconomics--accounting, public finance, or operationsresearch, and political science. The major tenet of thisprofessional core has been that scientific thinking should

.

be applied to the solution of social problems, and it there-fore tends to emphasize systematic, rational, and quanti-tative approaches to budgeting. An interest in programbudgeting is likely to come from this quarter.

Berdahl (1971), in his treatment of state governmentand higher education relations, has linked informationsystem development in the states with attempts to intro-duce program budgeting for higher education. While theimplementation of program budgeting depends on the avail-ability of systematized information in special categories,so does virtually every other kind of budget reform. Wefound strong and continued interest in developing state-wide information systems even though attempts to implement .

formal program budgeting systems have waned considerably.

6. Although much of the discussion surroundingefforts to improve the management of higher education isexpressed in terms of improving efficiency, there are anubber of indications that the real concern is about thelevel and kind of higher education being provided. Publicpreferences for higher education appear to.be changing,as has been indicated in the recent decrease in age-specific participation rates. The kinds of educationbeing provided are also changing in response to changinglabor market conditions. But because the public doesnot pay for public higher education services in individualunits, the pressure of this changing demand is felt onthe budgetary process. Demand for information is atactic, or ploy, in the bargaining that determines the"price" state government will pay for an entire bundleof higher education services over the course of a yearor a biennium. Having to provide information on somerelatively simple indicators of funding level or perform-ance can serve to weaken the budgetary arguments ofthose institutions which are funded richly according tothese indicators, without making it necessary to increasefunding for institutions which are funded poorly. Those

2 39

who are getting less have the weight of history goingagainst them,,for their existence is evidence that theymanaged to survive the previous year.

Even where the demand for higher education does notappear to be falling in a particular state, the budgetConatraints-of recent years-may make it essential thatstate budgets be pared, and higher education is both amajor expenditure item and vulnerable on political grounds.It is also a highly discretionary item, being funded byalmost all states from the general fund.rather than fromvarious earmarked funds. Except for the community collegesin some states, public higher education is rarely funded'through statutory formulas, as elementary and secondaryeducation or public assistance often are.

Even without any strong connection with the demandfor higher education, new budget techniques are oftentried selectively in higher education. (Apparently scmestates have singled o'jt higher education as an exceptionfrom program budgeti'lg, but x:one of the states examinedin detail for this Litudy had done so.) Howard (1973)has suggested that new budgat techniques be attemptedfirst in gx agency under one head which has limitedduplicaticA and .overlap with other agencies. Highereducation may present special problems when program budg-eting is attempted because of the complexity of its out-puts, but for organizational reasons, higher educationmay be an ad7antageous agency in which to begin.

Recently, the concept of accountability hasbeen invoked repeatedly to justify the increasing penetra-tion of state agencies into the management of higher edu-cation institutions. Elected officials are, of course,ultimately accountable to the electorate and may be votedout of office. But other state executives--departmentdirectors, for example--are not elected and thereforemust be accountable to the electorate and its represent-atives in some other sense.

Accountability has been used in a number of sensesas a characteristic of the web of relationships whosesum total is the governance process of higher dducation

10

1,4

institutions. Accountability appears to have at leasttwo ii4terpretations, both of whiCh lave far-reachingimplications for centralized inforMation reporting andbudgetary review procedures. The first is that account-ability is an extension of the fiduciary responsibilitythat administrators have to assure that budgeted fundsare not spent illegally or improperly. This means thathigher education administrators are accountable to stateofficials for the specific performance of their institu-tions. Because an institution supported by state resourcesis presumed to have a schedule of services to perform,institutional administrators are accountable in the sensethat a postaudit would be possible, at least in theory,which would check performance against specific objectives.This notion of accountability has obvious implicationsfor budget submission and the provision of informationbecause these educational outputs (services) must beidentified, quantified, and linked to budgetary require-ments

The second interpretation focuses attention on deci-sions and choices made by state officials rather than onperformance. Accountability in this sense literally callson administrators to account for their decisions or, inparticular, their budgetary choices, by providing anappropriate supporting rationale that demonstrates theymade considared choices on the basis of adequate andappropriate information and analysis. This is analogousto the management audit function in administration whichtries to ensure that accounting systems are adequate toprevent fraud and mismanagement. To provide decisionaccountability, methods of decision are scrutinized fortheir adequacy in supporting "good" decisions. Budgetary !

requests, then, must be documented with information toshow that decisions have not been reached arbitrarily.

From these seven general factors which seem to liebehind efforts to improve the substance of budget requeststo state agencies, one should not infer that all theinitiative lies at the state level and that institutionsare merely passive respondents to the information demandsof state agencies. More than ten years ago Miller (1964)commented that two problems that act as catalysts for the .

11

2 5

development of budgetary formulas and cost analyses arethe problem of.securing an equitable distribution of fundsamong institutions, and the proclem of providing sufficientlyobjective justifications to satisfy state budget officesand legislatures. He noted some expressions of surpriseat the slowness with which objective devices have beendeveloped and adopted for higher education budgeting, andsuggested.that this was indicative of substantial legis-lative trust and good will toward colleges and universities.At present, *t appears that the reserve of good will towhich he referred is becoming exhausted (more so in somestates than in others, of course), and that the incentivefor providing sufficiently objective justifications tosatisfy state-level agencies is operative once again.The institutions, like the state agencies, have a clearinterest in developing budget information formats andsubstance which will win the approval of their statereviewers.

2 3

12

2.Budgetary and Financial Information:

Taxonomies, Uses, and Systems

Although the states differ markedly in the detailsof their budgetary procedures, the broader issues involvedin the submission of higher education data to supportrequests for state appropriations are remarkably similar.These issues commonly pertain to:

41, The kind of data to be provided

The format or organizing structure withinwhich the data are collected or displayed

The uses of the data in the budget process tosatisfy various budgetary functions

The development of systematic procedures forgathering and reporting data

Each of these topics is the subject of a lengthier dis-cussion in subsequent chapters. However, in order todefine the substance of these issues and the terms usedto describe them in this report, this chapter is devotedto a brief discussion of the concepts and theory whichwill be applied in the subsequent description and analysis.

KINDS OF DATA AND SYSTEMS FOR ORGANIZING THEM

Data are simply recorded facts or observations whichbecome information when they are related to a decisionin a particular context. Budgetary data are principally

2'7

expenditures of some sort, but they may also includemeasures of institutional activity related to students.,facilities, instruction, research, administration, andeducational outcomes expressed in various dimensions. In

addition to the distinguishing subject matter to whichdata pertain, there are three important characteristicsthat distinguish the data used in the budget process fromdata used in other organizational and management activi-ties: the extent to which data are combined or aggregated;the origin of the data as either observation, estimation,or projection; and the perspective used in combining thedata.

Data which describe a single event or transaction,rather than a sum of many similar events over an extendedperiod of time, are not very useful during the early stagesof the budgeting process because of the emphasis on theplanning function during this stage. Attention is directedto aggregates of expenditures and activity levels becauseit is usually unnecessary, if nut impossible, to specifytransactional details in advance. Because the focus ison a future period, the data in budget requests are largelyestimates and projections rather than what might be called"hard" data, that is, actual observations from the histor-

-i.Cal past.

Systems for organizing data primarily arise from twoperspectives: an operational perspective which relatesdata to the day-to-day operations and activities of in-stitutions and a programmatic perspective which relatesdata to the objectives and goals of an institution. Inpractice, operational data systems are structured in termsof the items or objects for which budgetary expendituresare made, or the functional and support activities whichinstitutions are engaged in. Programmatic data structures,in theory, differentiate the purposes of governmentalactivity into distinguishable programs which may or maynot parallel organizational lines, depending on thestructure of bureaus, agencies, and institutions. In

practice, however, a programmatic structure is difficult.. to distinguish Irom functional or activity classifica-tions, because an agency's goals and missions are mosteasily stated in terms of its activities, which also

214

serve to distinguish the agency along organizational lines.Data structures for each of these expenditure perspectivescould be unique to a state's procedures, but in the interestof achieving comparability they are usually applied accord-ing to standardized conventions, with various modifications.

Object of expenditure categories refer, of course, tothe specific items which a budget will purchase. Followingis the object of expenditure classification used in federalbudgeting (U.S. Bureau of the Budget, 1960).

PERSONAL SERVICES AND BENEFITSPersonnel compensationPersonnel benefits

.Benefits for former personnel

CONTRACTUAL SERVICES AND SUPPLIES,Travel and transportation of personsTransportation of thingsRent, communications, and utilitiesPrinting and reproductionOther servicesSupplies and materials

ACQUISITION OF CAPITAL ASSETSEquipmentLands and structuresInvestments and loans

GRANTS AND FIXED CHARGESGrants, subsidies, and contributionsInsurance claims and indemnitiesInterest and dividendsRefunds

Most states employing such a classification use a morecondensed version which only shows Personal Services,Contractual Services, Supplies and Expenses, and CapitalExpenditures. The principal merit of this kind of class-ification, at least in its more detailed version, is thatit merges directly with the budget execution phase of cen-tral budget office activity. Because budgets are spent on

15

2 9

objects in individual transactions, budgetary planning inthis same mode can be linked directly with the allotmentand control of funds during budget execution. The set ofcategories in use is clearly applicable to all kinds ofagencies.

The standard classification of institutional functionsand activities is that devised by the National Associationof College and University Business Officers (NACUBO), thebasis of the accounting structure that virtually all in-stitutions use for their financial reporting. The detailsfor the current funds expenditure accounts are shown inAppendix A-1. Its major expenditure categories, whichare the aggregates used in functional budgeting are:

EDUCATIONAL AND GENERAL EXPENSEInstruction and Departmental ResearchOrganized Activities Related to Education

DepartmentsSponsored ResearchOther Separately Budgeted ResearchOther Sponsored ProgramsExtension and Public ServiceLibrariesStudent ServicesOperation and Maintenance of Physical

PlantGeneral AdministrationStaff BenefitsGeneral Institutional Expense

STUDENT AID

AUXILIARY ENTERPRISES

The source for this structure of accounts is the NationalAssactation, of College and University Business Officers(1968), 2nd edition. It should be noted that the 3rdedition of this volume (1974) is now available and pro-vides a new set of functional classifications which aresomewhat revised from those shown above. At the time ofour field investigations, this new set of accounts was

rInOXY

16 ,

not yet in use at the state level, and therefore "func-tional" in this report should be associated with theaccoun1s from the 1968 edition. Institutions will grad-ually shift to the new structure, and it may be assumedthat state-level classifications will change as a result.Although some changes in assignment of activities tovarious functions have been made in the new structure,the major changes involve new aggregations of major func-tions and slightly altered names. These are changes whichshould be accomplished at the institutional level withvery little delay. Appendix A-2 contains a descriptionof the most recent NACUBO current funds expenditureaccounts.

The standard programmatic structure is that developedby the National Center for Higher Education ManagementSystems (NCHEMS) in the Program Classification Structure(Gulko, 1972). Exhibit 1 displays the version that appearedin the 1st edition of the Program Classification Structure.This structure was intended as the basis for developing aformal program planning and budgeting system for highereducation. However, inspection of the individual programs-reveals that they are not very different from the func-tional categories of the NACUBO classification. As notedearlier, a longstanding technical difficulty in the appli-cation of program budgeting has been the specification ofa program structure that can be linked unambiguously withan existing system of bureaus or institutions. In March1974, a Joint Accounting Group, consisting of represent-atives of NACUBO, the American Institute of CertifiedPublic Accountants, and NCHEMS designed a set of expend-iture subcategories and "cross-over" conventions whichmakes it possible to relate expenditures reported accord-ing to the NACUBO chart of accounts developed in the1974 edition of College and University Business Adminis-tration to Gulko's (1972) Program Classification Struc%ure.A comparison of these two account structures can be foundin Collier (1975).

For all intents and purposes, any conceptual dis-tinction between the two systems is removed when data tobe formatted according to the Program ClassificationStructure is derived from an operational rathern than a

3117

Exhibit 1

ORGANIZATION OF THE PROGRAM CLASSIFICATION STRUCTURE

CAMPUS

1.0

INSTRUCTION

2.0

ORGANIZEDRESEARCH

3.0

PUBLICSERVICE

4.0

ACADEMICSUPpORT

5.0

STUDENTSERVICE

INSTITUTIONALSUPPORT

7.0

INDEPENDENTOPERATIONS

1.1 GeneralAcademic

2.1 Institutes& Research

3.1 CommunityEducation

4.1 Libraries4.2 Museums &

5.1 Social &Cultural

6.1 ExecutiveManagement

7.1 InstitutionalOperations

Instruction Centers 3.2 Community Galleries Development 6.2 Fiscal 7.2 Outside1.2 Occupational 2.2 Individual Service 4.3 Audio- 5.2 Supplementary Operations Agencies

& Vocational Of Project 3.3 Cooperative Visual Educational 6.3 GeneralInstruction Research Extension Services Services Administrative

1.3 Special . Service 4.4 Computing 5.3 Counseling ServicesSession Support & Career 6.4 LogisticalInstruction 4.5 Ancillary Guidance Services

1.4 Extension Support 5.4 Financial 6.5 Physical PlantInstruction 4.6 Academic Aid Operations(for credit) Administration 5.5 Student 6.6 Faculty &

& Personnel Support Staff ServicesDevelopment 6.7 Community

4.7 Course & RelationsCurriculumDevelopment

Reprinted from Gulko, W. W. Program Classification Structure (Technical Report 27). Boulder, Colo:

National Center for Higher Education Management Systems at WICHE, 1972, p. 19.

programmatic set of accounts. The major practical conse-quence of using functions rather than programs is that

.

use of the functional set of accounts results.in a largernumber of budget categories and slightly different aggre-gations of the major subcategories. For the budget year1974-75, both sets of accounts were being used by variousstates. In 17 states we found no use of a programmaticaccount structure based on higher education goals andobjectives. Although conceptually one can envision thatsuch an account structure might be designed (see Dyer,1970),. in practice it is expedient, if not essential, tobase "program" expenditures on operational data. Inhigher education, as with many social services, a program-._

matic structure for budgeting is difficult to establishbecause the recorded operations of the institution, itsbasic operational transactions (enrolling a student,hiring a faculty.member), do not explicitly involvemeasuring attainment of its stated goals and objectives(student development, expanding knowledge). Nonetheless,in many instances, the use of programmatic classificationsin budgeting is indicative of a subtle shift towarda moresubstantive review of agency activities and intentions.

Contained within the Program Classification Structureis another standard structure, that of the various disci-plines issued in the National Center for EducationalStatistics (NCES) Taxomony of Instructional Programs(Huff & Chandler, 1970). This structure, like the Pro-gram Classification Structure, does not usually coincideexactly with the organizational arrangement of instruc-tional departments in a college or university. However,the extens.we use of this structure in the data gatheringactivities of NCES makes it a logical choice as an in-

r. program structure for use in budgeting. Itis known as the Higher Education General InformationSurvey (HEGIS) Taxonomy and is shown in Appendix B. TheHEGIS taxonomy and codes are used widely in budgetingat the two-digit and four-digit levels.

The broad uses to which budgetary data may be putin planning, decisionmaking, or controlling are obviouslynot equally served by the various means of data organiza-tion. Consequently, budget submissions often include

19

3 3

data in more than one format. It may be particularly use-ful in some instances to mix display formats by further

disaggregating data already classified by program orfunction into object of expenditure categories. It is

also possible to preserve the underlying higher education

organizational structure by displaying organizationalunit data in terms of objects, functions, and.programs,or by some mix of these, and then recombining them at the

state level.

USES OF DATA IN THE BUDGETARY PROCESS

Use implies purposes and objectives, and therefore

use of data in the budgetary process must be interpreted

with reference to the functions or the purposes of budg-

eting. The budgetary process moves through a cycle of

phases which may be described as "preparation and sub-

mission," "approval," "exedution," and "audit" (Lee &

Johnson, 1973). Each of these phases serves particular

functions. In the context of state budgeting for highereducation, the first two of these phases were described as"institutional budget preparation and submission," "agency

, review and analysis," and "legislative review and appropria-

tion" (Glenny, Bowen, Meisinger, Morgan, Purves, & Schmidt-lein, 1975, pp. 27-31). This study of information andanalysis systems, like the broader study reported on in theother volumes'in this series, is concerned primarily with

agency review and analysis and legislative review and

appropriation. Legislative approval of the budget and the

use of the governor's veto are procedural features of the

process which bave interesting variations among the states,

but do not directly involve the use of information andanalysis and thus are outside the scope of this report.

Analysis and the use of quantitative justificationshave their primary.application during budget preparation,

but information submission continues during execution of

the budget and is also a primary concern in connection

with the process of postaudit. We examined the process

of institutional budget preparation and submissionselectively because it precedes the process of state

agency review, but execution and postaudit were not

20

generally considered. The portioh of the budget processon which we concentrated attention--budget preparation andsubmission--is the one most suited for relating plannedexpenditures to the pel.acy objectives of state officials.and higher education,administrators. In the everyday workof budget agencies, the entire range of budget cycle ac-tivities is often carried on simultaneously because pro-cesses overlap for past, current, and future budget cycles.As a consequence, there are many opportunities for inter-action between budgetary functions and the use of dataduring different phases of the process. Thus, use ofdata during the agency review and analysis phase cannotbe completely decoupled from the control function ofbudgeting, which is more directly associated with budgetexecution.

Current discussions of budgetary functions areheavily influenced by the functions derived by Schick(1966) from a functional classification Of managementplanning and control systems described by Anthony (1965).From this perspective the budgetary process is simply aparticular planning and control process. Anthony definedthree categories of managerial process as: strategicplanning--the process of deciding on objectives of theorganization, changes in those objectives, the resourcesused to attain those objectives, and the policies thatare to govern the acquisition, use, and disposition ofthose resources; management controlthe process by whichmanagers assure that resources are obtained and usedeffectil,ely and efficiently in the accomplishment of theorganization's objectives; and operational control--theprocess of assuring that specific tasks are carried outeffectively and efficiently. Management control isdistinguished from operational control by requiring some-what greater judgment and by being concerned with thedirection of people rather than the accomplishment ofspecific tasks. Although Schick's (1966) own discussiontended to support treatment of these processes as separateprocesses, he emphasized that:

Operationally, these processes often areindivisible, but for analytic purposes theyare distinguished here. . . . Planning is

21

3 5

linked most closely to budget preparation,but it would be a mistake to disregard themanagement and control elements in budgetpreparation or the possibilities for planningduring other phases of the budget year. The

management process is spread over the entirebudget cycle; ideally it is the link betweengoals made and activities undertaken. . . .

Control is predominant during the executionand audit stages, although the form of budgetestimates and appropriations often is deter-mined by control considerations. (p. 244)

-Schick's treatment has also stimulated a conceptualiza-

tion of the budgetary process as the analog of a zero-sum

game in which increased emphasis on planning and manage-

ment lead to decreased emphasis on control and vice versa.

However, as the discussion of public budgeting has been

elaborated by further studies of governmental budgeting,

most o6servers have come to regard shifts in emphasis

on the various phases of the budget process as responses

to changing budgetary environments and policy contexts

rather than phased shifts from control to management and

planning. Thus, with the reservation noted above concern-

ing the interaction of functions among phases, the pre-

paration and review phase is predominaritly one that

emphasizes planning, although the degree of emphasis on

planning varies among states and within a state itself

as revenue conditions and policy ooncerns change.

Our impressions are that state managerial control

systems function largely within higher education systems

or institutions. Operattonal control, where.it exists,

is clearly at a lower level within the institution. How-

ever, state-level managerial and operational control

systems do exist in the form of position control, re-

strictions on the transfer of funds between budget

classifications, and other procedures designed to achieve

compliance with central policy at the operating level.

Although these systems are outside the budget preparation

and review process, they are often important in setting

the basis for subsequent controls.

Schick's classification neither explicitly treatedthe overtly political or organizational functions of thebudgetary process nor recognized the distinctions betweeninternal budgeting within an agency and budgeting involvingthe overhead control organizations which are participantsin the state budgeting process for higher education(Simon, Smithburg, & Thompson, 1970). As overhead controlagencies, state budget agencies are involved in managementmost frequently through monitoring the activities of lineorganizations. Often excluded from direct responsibilityfor higher education management decisions by subtle andnot so subtle differences in authority stemming fromconstitutional status or tradition, budget agenciesnevertheless influence management decisions throughtheir ability to impose budget sanctions.

During state agency review, the explicit use of in-fbrmation appears to fall into four activities,.describedbelow.

1. Checking. Involves the detection of error inbudget requests, and ranges from detecting and removingarithMetic errors to the more substantive checking in-volved in eliminating unintended consequences of proposedexpenditures--such as a legal conflict, or a conflictwith federal or local expenditure programs.

2. Costing. Not merely calculation, costing isthe attempt to estimate the justifiable fiscal needs ofa proposed program or organizational unit in light ofhistorical accounting costs. Usually, the basis for thiscalculation is the preceding year's budgetary requirements,but inflation, changes:In.adtivity levels, the existenceof joint products, or the fact that there has been noprevious expenditure for a comparable item may necessitatethe application of judgmental estimates and cost-alloca-tion conventions.

3. Evaluation. Ultimately requires, as the wordimplies, setting a value on the program in question inrelation to other expenditure programs. It usually beginswith procedures for examining programs in comparison toother programs in terms of their costs, or with respect

to the technology of the program, as in a considerationof the use of inputs for the program. Evaluation mayalso include an examination of earlier program impactand performance. .Formal evaluation procedures are fre-quently grounded in economic theory and cost-benefitanalysis. Values are measured by market prices wherethese are available, or are estimated by means of avariety of techniques (see Chase, 1967). The cost concept'employed is that cost is the value of resources when em-ployed in their mos.t valued alternative use. Clearly

this may be very different than historical cost. For

example, under the concept of historical cost, facultycosts would be estimated from what faculty salaries have_

been historically. In a cost-benefit analysis, however,faculty costs might be estimated from the faculty salariesof the highest paying comparable institutions, or thesalaries of comparable professionals in industry orgovernment.

Final valuation of programs tends to be implicitrather than explicit, and often comes as a result ofbargaining--as in the determination of a price for atransaction in the market place. The price an individualis willing to pay for a service depends on his personalvalues, although if the provider is to stay in business,the price agreed on must cover the cost of providing theservice. In state-budgeting, it is a widely establishedprinciple that the price paid by the state,for the pro-

. vision of public;higher education services be determinedby the accounting cost of that service. This, however,

tends to suppress what are really questions of societalvalues, especially when the provision of a new socialservice or the suspension of a continuing service isbeing considered. In the interest of consensus, bindingbudgetary procedures, such as formulas that combine cost-ing and pricing in a single operation, are often used toblur the distinction between costing and pricing (evalu-ation).

4. Bargaining. The final adjustment of recommend-ations and appropriations at the margin in which commonground in the preferences of participants is sought, after

Which the result is adjusted to reflect the realities of

33-24

political influence and authority. In l'argaining, infor-mation is used as a resource in trying to persuade, orto gain a competitive advantage over other budget process,participants.

From a broader and more behavioral perspective,Burkhead (1956) described governmental budgetary functionsas "expertise," "communications," and "responsibility"functions. The purpose of the expertise function is toassess economic., sociological,'administrative;-and tech-nological considerations that can be infOrmed by facts.It consists of measuring and comparing a broad range ofcharacteristics of current governmental efforts., the-consequences of such effort's, and their alternatives.The communications function involves gathering the viewsand preferences of interested and affected groups, andthe responsibility function employs means and procedures I

for appraising the political acceptability of alternativeproposals and establishing patterns of responsibility for.,making budgetary choices. In effect, the budget processmust establish ground rules as to who will have access tobudgetary decisions and who will take the responsibility.BurkheAd argued that a legislative budget system is nottruly a budget,system at all because responsibilitypatterns are so often ill-defined. Budgetary functions,however, cannot be compartmentalized; all functions aremerged in the final budgetary decisions.

Burkhead's conceptualization of the functions ofgovernmental budgeting encompasses far more than the useof information and analysis systems. Indeed, because itrecognizes the political and the legitimating functionsof the budgetary process, it covers more than the pro-cesses of planning, management, and control. This reportwill shed more light on the technical use of information,that is, on the use of information which satisfies the"expertise" function in budgeting. These technical uses,however, must be considered within the context of thewider functions of the budget process because, as sug-gested in Chapter 1, new methods of budgetary choicesare not being considered solely or even principallybecause of technical inadequacies in the present process.Deficiencies in responsibility and Communications

3 9

functions are currently considered to be a problem in thearea of accountability.

INFORMATION AND ANALYSIS SYSTEMS

Administrative procedures are increasingly referredto as management systems, for in many instances thesepractices not only involve comprehensive and centralizeddirection and management, but sometimes also automated.data processing equipment and computer systems. Thisreport deals with the information reporting and analyticalprocedures used in state agency review's of higher educa-,tion-,budgetsand-thus,-by-extension-,-,with:-the-information_,,___,_

and analysis systems that state agencies are developingfor meeting their specific budgetary responsibilities.

During the course of this study, we sensed that budgetprocess participants had a fundamentally ambiguous attitudetoward information and analysis systems. On the one hand,virtually every state agency and higher education institu-tion spoke of the need for developing comparable sourcesof information across a bl,..ad spectrum of higher educationnd socioeconomic phenomena; on the other-hand, very fewagencies or institutions have developed these sources intoa widely recognizable, acceptable management informationsystem (MIS). A large part of this uncertainty over theuse of management information systems in the state highereducation budget process results from an inability of those'who would implement these systems (as well as those whowould save us from them) to precisely define what theymean.by a higher education management information systemat the state level.. Some consider MISs to be anotherchapter in the ma,a.-.7-!.Lent philosophy which sired PPB(Planning-Programming-Budgeting), MBO (Management byObjectives), and PERT (Program Evaluation and ReviewTechnique). Many who use the term emphasize the use ofautomated data processing equipment and computers. Othersuse it to.mean little more than an extension of what "good"managers, planners, and analysts "have always done" intrying to gather more reliable and more comparable datato aid managerial decisionmaking. From a somewhatnarrower perspective, according to Weiss (1970), manage-

ment information systems may be patterned after fourbroad models which are, however, not mutually exclusive:

A combination of functional internal and ex-ternal information gathering networks which serve theinfcrmation needs of the organization.

A part of the management process which assistsin fulfilling the decisionmaking function of management,primarily in the areas of planning and control.

A managerial reporting system in which informa-tion is routinely reported on a preset schedule for cen-tral storage, from which it can be withdrawn by varioususers.

An information processing system consisting ofhardware, software, and human beings.

From the description of information and analysisprocedures discussed in the following chapters, aisess-ments can be made of how extensively MISs have been imple-mented. It is important, however, in making these esti...mates, to have in mind what seem to us to be some definitefeatures of MISs, the primary one being that they are in-tended for management and should be regarded as part ofthe management system. The elements of information systems(other than MISs) can be specified by listing informationtAsks comprised in a general automated system.

Data collection: Observing and recordingevents and transactions.

Data conversion: Changing data containedin an original document into a more suitableform for processing or storage.

Data transmission: The process of movingdata from one location to another byphysical or electrical means.

Data,represontations Representing data ina machine-readable form so as to facilitateprocessing.

27

41

Data organization: Arranging data in theform of files, records, so that they can beprocessed and retrieved efficiently andeconomically.

Data storage: Storage of data on appropriatemedia so that they will be easily accessibleto the central processor.

Data manipulation: Selecting, sorting,merging, and/or editing data so as tofacilitate further processing.

Data calculation: Performing various arith-metic and logical operations on data so,asto produce desired information outputs.

Information retrieval and display: Retrievinginformation, or processed data, from infor-mation storage, and making it available inthe proper format and medium at the righttime to interested users. (Dippel & House,1969, pp. 5-6).

In planning and budgeting information systems which givegreat attention to specific Aggregations of data, however,the functions of organization, manipulation, and calcula-tion must include such extensive filtering of data thatnot all data are passed on. These systems may also in-clude various aggregation routines involving crosswalkprocedures between data structures. But because thisprocess merely defines, collects, and transmits data, inour view it is not an MIS. Authentic management infor-mation systems are directed toward the accomplishment ofspecific tasks or the support of specific decisions orkinds of decisions. These are systems that are obviouslyconstrained, therefore; by the responsibilities andfunctions which are assigned to officials and agencies.

Another'important consideration is that MISs arenot comprehensive, nor are they the total integration ofall organizational subsystems. In any complex organiza-tion, whether public or private, the range of organize-

4 228

tional functions is simply too great for any single MISdesigner to have command of the information needs ofadministrators or managers for the full range of decisionsto be made. Resorting to the use of a sufficiently largestaff of diverse functional experts presents severe coor-dination problems, and as Dearden (1972) pointed out,"If any of.the MIS people are competent to tell the func-tional experts what [they need], they should be in the

.

functional area" (p. 96). In short, comprehensive manage-ment information systems are a mirage because no complexorganization has or would tolerate an all-powerful, all-knowing central management. Real-world management infor-mation systems are limited in scope, and provide particulardecisionmaking points within an organization with limitedsets of homogeneous information.

Further, an MIS is not merely computer-based activity.In the state budgeting process, an MIS may not even re-

. quire the application of computers and automated dataprocessing systems. As larger amounts of data are reported,some use of automated data processing may facilitate datahandling and analysis. The development and review ofhigher education budgets, however, do not generally re-quire quick retrieval of data, and thus on-line systemsdo not have a high priority. Some states have implementedautomated budget data systems which tabulate state agencyactions and recommendations on budget totals, but thesesystems are used as record-keeping devices and not fordata submission.

Systematic efforts to gather and report informationconstitute a notable development for the state-level admin-istration of higher education. But the most importantachievements in automated management information systemsin higher education are most likely to be at the institu-tional and operating levels--in admissions, purchasing,libraries, and other business operations. With respect tobudgeting, the most significant developments regardinginformation will lie, with few exceptions, not in theimplementation of automated data-processing technology,but in attempts to refine and use the information pro-vided by improved information gathering and reportingprocedures.

4 329

3.Information and Analysis Development-

in State Agencies: An Overview

Most state budget review agencies examined in thecourse of this study indicated that they had certain in-formation deficiencies. Most identified the difficultyof getting adequate information for conducting a satis-factory review of institutional budget submissions as asignificant problem, but few considered this the mostcritical issue they faced in recent budget cycles. Otherconcerns, primarily the highly constrained resources ofmost state governments, have extremely high saliency atthis particular time, and in addition, most state agencieshave limited capacities to devote to information gather-ing and analysis.

The information gathered by state agencies,except perhaps for state higher education agencies,

appears to be specifically addressed to the activitiesand functions they perform. Because agencies usuallycannot afford the luxury of maintaining a comprehensiveinformation system and the associated automated dataprocessing resources, such activities as budget develop-ment and review rely on a virtually independent informa7tion gathering and distribution network. The end resultis that budget requests and the entire process of revieware largely self-contained. Budget request documentsand supporting information form a single package. Somenew information may be injected from time to time duringthe course of hearings and negotiations, but for themost part the information base for analysis and reviewis set in the call for budgets, sdbsequent instructions,and the submitted requests. Information for use during

30

4 4

budget execution may require separate submissions of in-formation, but it remains essentially isolated from thebudget development process. The use of automated systemsfor submission of information is extremely limited,although it is becoming quite common for the reviewprocess-within individual state agencies to involvemachine-readable records and automated data processingin recording the changes and recommendations over thecourse of that agency's budget review.

-Significant exceptions to these generalizations-will--be pointed out during the course of the report. Thegeneralizations made above must also be qualified becausethey were formulated from evidence gathered at one par-ticular point in timethe year 1974 and the review ofthe fiscal 1974-1975 budget then in progress. At thepresent time, these same agencies are working on their1976-1977 budgets, ind in most instances two additionalbudget cycles have passed since our review. There also areindications that many states recently have become sub-stantially more involved in statewide inforniation collec-tion and quantitative analysis of institutional budgetrequests. _

Generalizations about administrative activities inthe 50 states will almost always be wrong because of thesubtle distinctions that slip through anything but a de-tailed analysis and description, and yet they may also beright because at .least one state will be doing the exactopposite of another. Furthermore, in approaching overalltrends in information use in terms of'the practices ofspecific types of agencies, there is always the possi-bility that any description will fall short of completeaccuracy because the roles of executive, legislative,and higher education agencies vary so widely from stateto state. It is also impossible to divorce informationand analysis of budgets entirely from agency activitiesin other aspects of budgetmaking. The use of specialstudy commissions and task forces, and the developmentof new categories for appropriations in the budget bill,are clearly substitutes for reforms in the handling ofinformation because often these procedures produce thesame results that can be achieved through the routine

,

31

4

gathering of more information and the application of newforis of analysis. However, some consideration of theoverall norms of activity in regard to information arein order as a basis for considering more specificactivities.

Accompanying the vast differences between statehigher education governance structures and proceduresare very different general attitudes toward informationcollection and disclosure at the state level, which pro--duce-very different information-environments. In somestates, for example, salaries of public officials, in-cluding those of university presidents, are fairly wide-spread knowledge and may be the subject of regular publicattention in relation to either general or specific fiscalissues; in other states, such disclosures are not generallymade. In one state, salaries of public university ;presidents

and legislative staff directors were neither generalpublic knowledge nor made available to researchers forthis study.

Personnel systems are the source of other variationsin the kind of information available at the state level.In a few states (Texas and Virginia, for example),personnel records were maintained at the state level,and detailed information on individual faculty memberswere made available to state agencies, although thedata supplied seemed to play very little role, if any,in the budget formulation pro-ess. In other states,the mere suggestion of such a file at the state levelwould have been unthinkable. We mention these differencesnot to suggest approval or disapproval of one practiceover the other, but rather to illustrate the wide vari-ationstin what different states consider the properprovince of state agencies in maintaining and providingaccess to information files.

EXECUTIVE BUDGET AGENCIES

Although all state agencies may approve the formand substance of budget instructions, the executivebudget office is usually the issuing agency, and it

432

therefore has a key role in determining information sub-missions to the state. Information from institutionsobviously can be requested outside the more formal processof budget instructions and subsequent budget submissions,but the formal instructions lay most of the groundworkfor the information on which technical reviews will takeplace. During the course of review, questions or issuesmay arise which require other information. The executivebudget office, in issuing its instructions, is perhapsin a better position than the legislative staff to satisfyits information-demands because it very often specifiesmost of the substance of the budget instructions. Legis-lative staffs also may have the opportunity to specifyany data they feel will be required, but in practice theyusually depend very heavily on the executive budget office.

Currently, executive budget offices are quite widelyrequesting both more information on institutional opera-tions and also more detailed information, that is, in-formation at a lower level of aggregation than they havetraditionally sought. This is especially true for theinformation asked of doctoral-granting and researchuniversities, which generally have more complex opera-tions and have not traditionally submitted data to state;agencies on the complete spectrum of their activitiesand sources of revenue. Information which calls eitherfor new classifications, such as program categories, orfor information that has never before been gathered oneducational outputs and the target groups for educationalservices, is in some instances being required from allinstitutions. Data schedules are usually devised inwhich to display these data, and instructions are givenfor their inclusion in the institution's budget submission.A request for new data often signals a new approach tobudget review in the form of a shift to formula or pro-gram budgeting,.or in the case of more incremental changes,a request may merely be a tactic to keep the institutionsoff balance.

Many of the requirements for changes in categoriesand types of information are associated with the imple-mentation of program budgeting. Several execu.tive budgetoffices in Atates which were leaders in program budget 1

33

4 '7

development have now pulled back substantially fromattempts to budget higher education on a program basisin anything more than a superficial way, such as simplycontinuing the use of the program categories. California,

New York, and Pennsylvania are states which mounted sub-

stantial efforts to do program budgeting (see Schick,1971) that have now been largely abandoned. Vestiges of

the program budgeting effort remain in the form of infor-mation categories used either in the appropriation billor in budget requests. These program categories are in

--use-at-highly-aggregated-levels, In-New York, for-example,, -

their use seemed to relate more to an interest in split-ting up the traditional lump-sum appropriation to highereducation than to program budgeting.

At the same time, in several other states in the17-state sample, executive budget offices were takingsteps to move exclusively to program budgeting or weredeyeloping program budgets in tandem with the more tradi-tional budget format. This results in a substantiallyhigher information load during the process of budgetdevelopment for the agencies reviewing'budgets. In a

few states, two governors' budgets are produced--one inprogram format and one in a format more related to the

traditional practice of review, recommendation, andlegislative appropriation. In at least one state,Connecticut, the governor's budget in program formatwas developed from parallel institutional program budgetsubmissiong which accompanied the traditional object of

expenditure budgets. There was no indication, however,that the program budget played any role in decisionsduring the formal review process. The budget was infor-mational; it displayed a state expenditure plan by pro-gram, and showed what might be termed a "reconstructedlogic" for, "he budget as opposed to a "logic in use."

Because of their limited staffs, executive budgetagencies have little capacity for engaging in specialanalytical ,,*Idies of a nonroutine and ad hoc nature.In Wisconi.n, the governor and the legislature mandated

the Vnivo, Lty of Wisconsin Central Administration staff

to complete certain "special studies" and include them

with their budget submission for the following biennium.

34

The University of Wisconsin Central Administration, havinga much larger staff than the state budget agendy, andhaving access directly to relevant information, was in abetter position to carry out these studies. The execu-tive-legislative mandated studies (ELMS) resulted inreports that were more than just the submission of data;they involved in some instances the evaluation of pro-posals and proposed program.activities designed by execu-tive and legislative budget review agencies. In thisparticular case, the request for the studies had been anoutcome of a prior executive staff analysis.

The staffs of governors' budget'offices have grownover the last few years, however (Schmidtlein & Glenny,in preparation), and some of these staffs have signifi-cantly increased their analytical capacity-and resources.NeVertheless, in most of the,17 states examined, thenumber of persons reviewing budgets fOr th4 governor haskept pace only with the growth of statigeryices; thesestaffs remain small and do not usually include individualswith highly analytical or research backgrounds or experi-ence in statistical analysis and quantitative techniques.The size of these staffs does not generally allow, there-fore, for the review function to be specialized beyondthat of assigning individual staff members:as budgetreviewers to doctoral-granting universities, statecolleges, or community colleges.

Research and analyticii backgrounds are more likelyto be found on performance audit staffs that have beenestablished in the executive branches of some states.Such staffs, however, even when they are in the proximityof budget review staffs, have difficulty relating theirefforts to budgetary matters because their studies oftentend to lack timeliness and relevance.

LEGISLATIVE BUDGET AGENCIES

In all but a few states, the governor is nominallyresponsible for the preparation of a comprehensive statebudget. The executive branch, therefore, is most oftenthe source of changes in information requirements, as

4 935

well as of other reforms in the budget process. But

legislatures may also exercise a direct role in informa-tion and budget review reforms by legislating specificbudgetary procedures (as in Hawaii) or by assigning theresponsibility for certain budgetary functions to a

,coordinating agency (as in Illinois, Texas, and Virginia).By establishing their own budget review staffs to assistthem in taking a more active policy-setting role, legis-latures also add their own information requests to thoseinformation demands already placed on institutions.

Legislative action en the budget in general is oftencharacterized by heavy dependence on the executive branch,both for policy direction and information. Legislative

budget committees have traditionally depended on budgethearings and informal sources of information for indica-tions of institutional competence and credibility and onspot checks for budget control. As a court of last appeal

on budgetary matters, the legislature is likely to be

concerned with very specific and concrete issues, notgeneral questions of policy. The hiring of staff analystshas increased the legislature's capacity to go beyondthese spot checks and to develop a policy context withinwhich responses to specific issues can be made. Conse-

quently, we found legislative budget staffs growing muchmore rapidly than the corresponding executive staffs.

Nonetheless, with the exception of those state legis-latures that actually prepare comprehensive state budgetson their own, or are virtually coequal with the executive

in developing such a budget, legislative budget reviewis not comprehensive, nor does it often propose alter-

natives. In the main, it continues to consist primarilyof checking the implications of executive proposals.Of course, where the legislature is a coequal in develop-ing the state budget, as opposed to merely adopting it,legislative information demands may be directed towardcomprehensive programmatic information.

In several states, we found examples of legislativestaffs actively involved in requesting additional infor-mation as a part of the budgetary process. Vor example,

in Colorado, the Joint Budget Committee worked throughthe Commission on Higher Education to gather ard analyze

36

5 't)

information on departmental faculty workloads. InMichigan, the legislative fiscal agency staffs requestedsubstantial amounts of additional by-lget and planningdata from inStitutions in an organizational format thatwould support their review, conducted differently fromthat of the executive budget office, which relied onprogrammatic data. As mentioned above, the Hawaii legis-lature, through legislation, has established a compre-hensive Planning-Programming-Budgeting System (PPBS)for state program and financial management. This system,-which requires the development of program information and

_program analysis.resources,,was first.used_in_the1975 fiscal biennium. In Washington, the state legislaturis developing the Legislative Evaluation and AccountabilitProgram (LEAP), a computerized information system whichprovides budgetary planning and expenditure informationto the legislature as an aid in preparing the budget andmonitoring public expenditures.

Excepting those state legislatures which tradition-ally have been the dominant branch of state government,there is currently a rather broad reaction againstexecutive dominance and a move toward strengtheninglegislatures. The studies and proposals of the CitizensConference on State Legislatures (1971) support thistrend. Expanding staff and improving systematic infor-mation gathering are always major components of theseproposals for strengthening legislative capabilities.There is, of course, much for these legislative budgetstaffs to do if they merely respond to the initiativesof the other budget participants. In addition to theproposals of the governor's budget, which usually receivemost of their attention, they must also react to thedetails of institutional requests (or of a statewidegoverning board) and possible formal or informal recom-mendations of a coordinating agency. Even when legis-lative staff do not officially receive the initialinstitutional budget requests submitted to the executivebudget office (as in the federal process, where agencyrequests do not go to the Congress), these requests areusually available informally and are in virtually allcases much more informative about institutional plansthan the governor's buaget.

37

51

As guardians of the state treasury, legislativeinterest in controlling spending continues to influencethe style of legislative budget reforms very heavily.Changes in the format of information submitted with

budgets frequently are tied directly to the format of

the appropriation bill. Often, their purpose is to split

up a lump-sum appropriation rather than provide theopportunity for programmatic review by the legislature.In New York, agreement was reached to relax restrictive

personnel and allotment controls on the State Universitywhen the university's appropriations were made in the

-seven-categories-of NCHEMS Program_Classification_Structure. Legislatures also have become interested inexpanding their.fiscal postaudit function to include orencompass performance audits of agency operations. The

legislative auditor of the Hawaii legislature has con-ducted performance audits in higher education that have

resulted in formal reports and recommendations (see the

discussion of the Hawaii budget process in Chapter 4).

The Joint Legislative Audit and Review Commission in

Virginia has also completed a performance evaluation of

the state's community colleges. The number of performance

audits that have actually been carried out on higher

education activities is fairly small, at least in those

states we observed, and certainly the coverage of these

audits has been limited. However, there appears to be

growing interest in using these studies to focus legis-

lative interest on substantive policy matters.

Performance audit staffs are likely to be furtherseparated from budget review in the legislature than are

the performance audit staffs in the executive branch.

It appears that legislative performance audit staffs

are more likely to have done studies on higher education

than are executive performance audit staffs (Glenny,

et al., 1975).

HIGHER EDUCATION AGENCIES IN THE STATES

With the exception of three states (Delaware,Nebraska, and Vermont), all states have a state-level

coordinating or governing board for public higher

5 238

education which could play a role in budget review. Eventhough they may not have formal budget review authority,these boards and their staffs may still have a role instate-level information collection or in establishinganalytical bases for budgetary review. Although thebudget-related activities of these agencies range acrossa broad spectrum, it is essential to draw at least theminimal distinction between state higher educationagencies: some are statewide governing boards and somecoordinating boards. Because of their inherent respons-ibilities, governing boards always have statutory budgetauthority. They*also are able to take fundaMehtellydifferent actions in developing information bases..,

STATEWIDE GOVERNING BOARDS

Many statewide governing boards actually developbudget requests for the governor and legislature insteadof serving in the capacity of first level of state review,as do coordinating boards. In most instances, institu-tions under a statewide governing board do not submitseparate formal budget requests to be subsequently re-viewed by executive and legislative budget agencies.Statewide boards usually consolidate or aggregate=theseinstitutional requests, hence campuses may or may notbe identified in the statewide governing board's budgetrequest. These boards and their staffs tend to play amore active role in designing the format of the requestand its substance, but in this respect there are greatvariations among the states.00In Florida, for example,the institutions do not subm#,g formal written requestto the statewide board for revlew, and in Kansas theinstitutional requests, although reviewed and changedby the statewide board, remain separate individualdocuments throughout the entire course of state-levelreview.

Statewide governing boards are likely to have much ,

more operating and budgetary information than coordi-nating boards; in fact, they may be the producers andaccumulators of this information for all the institutions.Consequently, in contrast to the information reforms

5 339

undertaken by a coordinating agency, which are likely tobe consistent with its role as a "watchdog" or overhead

-control agency, the information reforms that governingboards undertake resemble the actions of an institutionreforming its own internal administrative systems. In .

general, statewide governing boards have somewhat largerstaffs than coordinating boards, and their staffs alsohave direct access to operating information because theyor their supervisors have line-management responsibilities.It is also likely that the budget unit can draw supportfrom an analytical studies unit. Staffs of three of the-five statewide governing boards in our-sample-worked veryclosely with the budget preparation unit. Although theremay be little need for the legislature to assign respons-ibility for statewide gathering of information where thereis a statewide governing board, there is some indicationthat the growth of budget and analytic studies units insystem administrations resulted frcm executive and legis-lative information requirements.

Because governing board administrations have aroutine need for operating data, they may be more likelyto use an automated information system of some type.Three states with governing boards, Florida, Mississippi,and Wisconsin, had information reporting systems withautomated featuxes of various types that made thesereporting systems considerably more sophisticated thanthose of other states examined.

COORDINATING BOARDS

Virtually all of the coordinating boards studiedhave staff with responsibility for statewide data col-lection and distribution. In a few states, where datacollection has been a longstanding effort of the coordi-nating board, a routine reporting system, usually manual, -is in operation, in addition to the data reporting thattakes place in the submission of budgets.

In carrying out their data collection responsibili-ties, these staffs are frequently involved in definingdata inventories that include the budgetary support

40

5 ';

information requested by other state agencies. Coordi-nating agency staff also are likely to have the respons-ibility for preparing data reports or comparisons of datafor executive and legislative budget staff. One of theirprimary responsibilities, at this early stage of state-wide data systems development, is implementing a commonreporting system to provide comparability of data fromassorted institutions. Some attempts at this were under-way in fiscal 1974-1975, but the efforts have been expandedsignificantly with the adaptation of the informationexchange program for the state level and the State-LevelInformation Base Project at NCHEMS. Increasingly, many

.more states have been availing themselves of the infor-mation exchange procedures and the accompanying workshopsfor institutions and coordinators. In Colorado, coordi-nating agency staff have served as consultants to insti-tutions in their development of data systems, such asResource Requirements Prediction Model (RUM) and Compre-hensive Analytical Methods for Planning University Systems(CAMPUS), but these systems were not as yet supportingbudgetary requests.

On a routine basis, some coordinating agency staffsprepare average faculty compensation reports and unit-cost studies. These staffs may also be responsible forthe development and revision of funding formulas. Ofthe states in our sample,.a majority appeared to bereviewing and developing budgets on a nonformula basisor abandoning formulas altogether. Illinois, a statewith a number of years of experience with a cost-basedfunding formula, is perhaps the most notable example ofa state with a tradition of formula use which has nowabandoned it. Five states, three of them with coordi-nating agencies, are making use of formulas (Table 1).In each of these, the coordinating agency has or willhave a significant role in formula application. InTexas, the coordinating board actually certifies thatinstitutional requests have been submitted in accordancewith the formula devised through a statewide formulaadvisory committee. In Tennessee, the higher educationcommission collects the data for the determination ofunit costs on which the formula is based. In Virginia,the Council Of Higher Education will play a role in

41

5 3

Table 1

STATE HIGHER EDUCATION AGENCY (SHEA) RESPONSIBILITIES, 1974-1975

State

Kinds of

SHEAs holding

statutory

budget review

authorit

70 percent

of budget

or more

reviewed by

formula

Prepares Prepares Number of

Develops formal faculty professional

budget or unit-cost salary staff assigned

budget esti- com- to higher educe-

reviews mates ar:sons tion bud et

California

Colorado

Connecticut

Florida

Hawaii

Illinois

Kansas

Michigan

Mississippi

Nebraska

None*

Coordinating

Coordinating

Governing

Governing

Coordinitting

Governing

Coordinating

Governing

New York None

Pennsylvania Coordinating

Tennessee Coordinating

Texas***

Virginia****

Washington None

Wisconsin Governing

Not app. General review Yes Yes 0

No Detailed review No Yes 2

No Detailed review No No 1

Yes Develop budget Yes No 11

No Develop budget Yes No 6

No Detailed review Yes No 5N

No Detailed review No No 1

No General review No No 1

Yes**

no coordination

No No 2

Not app. Not app. No No 0

No Detailed review Yes No 4

Yes Detailed review Yes Yes 7

Yes Not app. No Yes 3

Yes General review No Yes 2.5

Not app. General review Yes Yes 1.5

No Develop bud et Yes No 31

Source: Glenny et al., 1975

* The California Postsecondary Education Commission, which superseded the ,:oordinating Council on

Higher Education in 1974, is expected to review budgets in order to advise state officials on policy

issues and to recommend general levels of support.

** The Mississippi Board of Trustees review of budget requests is !general, but review of institu-

tional operating budgets, which takes place before the beginning of the fiscal year, is detailed.

*** The Texas Coordinating Board does not have budget review authority, but has statutory authority

to develop and certify the use of budgetary formulas.,

**** The State Council of Higher Education for Virginia will have statutory budget review authority

for the fiscal biennium beginning in 1976.

future revision of the instructions for formula use,thereby assuming some of the role once played by theexecutive budget office.

A questionnaire survey of the development of state-wide infoimation systems by Barak (1975) indicated thatonly two states (Ohio and Tennessee) met certain criteriafor having fully developed state-level management infor-mation systems; several other states--Colorado, Georgia,Maine, New Mexico, North Carolina, Oklahoma, Oregon,and Washington--were reported to be well along in theirdevelopment; and nine states reported they had no state-level management information system at all. California,Florida, Hawaii, Nevada, Rhode Island, Utah, and WestVirginia were listed as nonrespondents. Our impressionsfrom visits to 17.states do not contradict the generalimpression given by this survey--that centralized manage-ment information systems are far from fully developedin the states. However, the criteria used in the Baraksurvey for determining whether a state has an MIS repre-sent such a high order of achievement that developmentefforts being made by state higher education agenciesmay be overlooked. The criteria usea for the Baraksurvey were as follows:

A formally planned, integrated informationsystem that utilizes data generated by opera-tional level programs to develop informationthat can be related meaningfully to theobjectives or problems of the top leveldecisionmakers.

A system consisting of analytical toolsand/or-programs encompassing the areas offinance, facilities, students, faculty,nonacademic staff, and academic programs.

Ali the analytical tools or programs areat the fifth stage of development, that is(as defined in the survey instrument), in-formation from these systems is being usedfor decisionmaking at the statewide level.

5 P"

43

4.Styles of Higher Education Budget Review

The budget process examines and explicitly or im-plicitly answers two essentially different questionsduring the course of review It first addresses thequestion of whether the activities proposed by the insti-tutions are proPerly costed out (what it costs to operatea university program) and secondly, it determines whetherfunds should be appzopriated to condubt these activities(whether a uniVersity, or some segment of it, ought tobe funded). The technical process of budget review isof much greater assistance in answering the first ques-tion than the second; some would argue that it cannotbeneficially address the latter at all. Traditionalbudget procedures have been directed chiefly at the"costing" of agency activities, but almost without excep7tion recent reforms in budget review have been directedat developing formal procedures for explicitly dealingwith choice within the confines of the budget process.

STYLES AND DIMENSIONS OF BUDGET REVIEW

To generalize about state-level budget processesfor higher education, it is useful to describe severalstyles of budgeting that were observed in the 17 states.Practice in some states may coincide precisely with oneof these types, but it is far more common to find theelements of several types combined in a state's pro-cess. Thus, the styles described below are attempts atdefining relatively pure forms of budgeting which embody

44

5 '3

a particular-set of budgeting principles and concepts.The styles are in two groups which coincide with the twodimensions of budget review: scope of review anl'reviewtechnique. Scope of review is described by the,two cat-egories, zero-base review and structured-incremdtal review,which distinguish between budget review approaches Pro-viding for review of items in the budget base ainOpproacheswhich structure review alound increments to the base, there-by virtually accepting a continuation of expenditures forbase activities. Review techniques are identified with theapplication of various analytical techniques, whether ap-plied to the base or only to increments to the base. Theapproaches in this group are the traditional object ofexpenditure budgeting, performance budgeting, formula budg-eting, programming-planning-budgeting, and tactiCal budgetplanning.

The styles discussed here are intended to be usefulanalytically, and should not be taken to exaggerate thedifferences between states. They have also been selectedto relate to differences in budgetary information require-ments rather than to budget process outcomes. Althoughthere are many differences between the various officesin their procedures and level of participation, thebudgetary process in any state must still solve the samefundamental issues. The styles which have been definedare also intended to represent existing state practicesrather than models for reform. In terms of the broadspectrum of practices described by Schick (1971), thevast majority of state budget procedures are heavilyoriented toward control. Consequently, we are describinga set of arrangements which include devices for providingwhat is considered to be a sufficient degree of controlover budget execution. In concentrating on the budgetformulation process, the present study gives relativelylittle attention to means for exercising budgetary con-trol. It should be obvious, however, that some reviewstyles are more appropriate than others for exercisingbudgetary control and that, therefore, actual stateprocedures are often a blend of the review styles weshall describe. The emphasis on the budget developmentprocesses in the states may also change from year toyear in response to changes either in revenue expecta-

45

5 9

tions or in the governance of higher education. Some of

these changes in procedure may be relatively superficial,but the submission and review phases of the budgetaryprocess are unaergoing regular scrutiny for revision andimprovement.

One characteristic feature of higher education budg-eting in all 17 states is that the higher educationrequest for general funds isia request for partial fundingof its total budget. Unlike most other state agencies,public higher education has various sources of operatingincome. This is particularly true of the large doctoral-granting research universities, but it is also true tjdome extent for any institution which receives income inthe form of student tuition and fees. Thus, in the highereducation budget process in most states, institutionsrequest a state appropriation which is the differencebetween their total budgetary needs for the coming periodand an estimate of their other sources of revenue. Techni-

cally, this places the state in the position of fundinga residual, with little or no control over-nonstate funds.However, by virtue of various control devices in use atthe state level, the institutions may realize relativelylittle discretion in expenditures from their nonstatesupport. Depending on arrangements for control, informa-tion demands concerning these nonstate funds are commonlyan issue in determining an appropriate format for informa-tion submission.

SCOPE OF REVIEW

Structured Incremental Review. Virtually all govern-

mental budget review involves comparisons of budget esti-mates with expenditures in the previous year, and in thissense review is incremental. Some budget submissions,however, focus attention on the increment over last year'sappropriation by establishing cross-cutting review.cate-gories, such as the adjusted base from the previous year,costs to continue, workload incieases, and program improve-

ments. Review within each of these categories may proceedaccording to any of several methods. Review of selected

46

expenditures within the budget base is not necessarilyexcluded, but in this format emphasis in review is nec-essarily directed to the additional increments. In someinstances, supporting information and documentation arealmost entirely limited to justifying these increments.

Definitions of the expenditure items assigned toeach incremental category differ among states, but intheory, assignment is straightforward. "Adjustments tothe base" are negotiated, but would typically includecosts of inflation and annualized costs of programsapproved in-the previous year less nonrecurring costs."Costs to continue" consist of cost increases that willbe incurred in continuing the activities included in theprior year's base. "Workload-increases" are those costincreases necessitated by higher activity levels, suchas enrollment increases, increases in library usage, andincreases in information reporting. "Program improvements"and new programs are expenditures that arise from increas-ing the quality of service or adding new services.

Clearly these increments to the base may be justifiedmore or less appropriately 1:4 different kinds of supportinformation. Herein lies one of the reasons that budget-ing methods'are likely to involve a mix of approaches.Costs to continue may be more clearly represented in termsof objects of expenditure. Workload increases are mosteasily justified through formula factors or performanceindicators. New programs and program improvement involvequestions of new outputs and may tend to rely more Onbudget methods that consider program outputs, alternatives,and the relationship of the proposed program to institu-tional missions, goals, and plans.

Zero-Base Review. In contrast to review whichexplicitly assigns a lower priority to scrutiny of itemsfor which appropriations were made in the prior year,some methods of review either make no distinction betweenfunds for continuing activities and funds for new activi-ties, or they make special provision for reviewing certainitems within the base budget. In keeping with the times,more and more state budget review agencies are developing

4 7

procedures to periodically examine selected items in thebudget base, but we do not mean to limit zero-base reviewto any one approach. In essence, central budget agenciesmerely have to establish the principle that ongoing acti-vities are not self-justifying, and that some of theseactivities will be reviewed in each budget cycle as thoughthey were new expenditures. Even though review may besomewhat more comprehensive, budgetary outcomes are stillonly incrementally different from previous budgetary out-comes. However, see Williamson (1967) and Bailey andO'Connor (1975) for a discussion and critique of incre-mentalism as an analytical concept describing the budg-etary process.

Pyhrr (1973) describes a general method and approachfor structuring a zero-base review, which he calls zero-base budgeting. This budget methodology has been appliedin corporate planning and budgeting and was used to developthe entire executive budget recommendation for the Stateof Georgia for fiscal year 1973. The method's distinctivefeatures are the presentation of an organization's entirerange of activities in terms of "decision packages," andthe ranking of these packages through cost-benefit analysisor subjective evaluation. Decision packages are identifiedeither as different ways of performing the same function,or as the performance of a particular function az a dif-ferent level of effort. A minimum level of effort package,for example, might be 50-70 percent of current operations,and additional levels of effort above this would be re-ferred to in terms .of a decision .package.

BUDGETING TECHNIQUES

Traditional Object of Expenditure Budgeting. The

most straightforward method of presenting a budgetaryplan is according to the items for which funds will beexpended. This ensures that all expenditures will beidentified, since they are expressed in terms of theplanned purchases of various concrete services and items.The federal object of expenditure classification has been'noted as a typical framework for summarizing and aggre-gating the expenditures of any governmental agency(see Chapter 2, page 15).

48

6

States reviewing budgets in terms of object classes usevery similar categories. However, for a budget to bereviewed in terms of objects of expenditure, it is neces-sary to provide data in substantially more detail withineach of these categories. As an example, see Exhibit 2for a listing of the object categories and the detailused by the Department of Finance and Control of theState of Connecticut. In addition, typically includedare supporting schedules for personnel, listing eachposition by institution and department with the presentand proposed salary level, and position reclassifications.Because a very large portion of a college or universitybudget is determined by personnel costs, this is thecategory for which detail is most important..

Review of higher education budget estimates in thisformat does not require a sophisticated understandingof educational methods or programs. Government agenciesmay differ in their use of various inputs, but theprinciples of review will be the same for any agency orbureau. A comparison with the previous year's expendi-tures in the same format is perhaps more important withthis approach than with others. Personnel costs arereviewed in terms of the number of positions listed andadditions; merit and anniversary salary increases; pro-motions; and the historical salary lapse rates resultingfrom personnel turnover. In general, budgetary issuesare related.less to programmatic considerations than tothe personnel regulations and procedures specific to theparticular state. Guidelines may also be promulgatedwithin the reviewing agency for evaluating the reasonable-ness of institutional requests. Experienced budgetexaminers use their judgment and intuition in determin-ing how-large an increase an institution or an agencycan reasonably handle, and the appropriate relationshipsamong expenditures for various categories. Virtuallythe .entire budget is reviewed against the estimatedexpenditures for the prior year. Rules of thumb andstandard operating procedures that apply to all agenciescan be used as well in reviewing highe- education budgets.:

49

6 3

STATE OF CONNECT= OEJECTS OF EXPENDITURECLASSIFICATION

CURRENT EXPENSES

Personal ServicesPermanent full-time positionsOther positionsFeesOvertime

Contractual ServicesAdvertisingPrinting & bindingSubscriptionsFeesLicensesTravel in stateTravel out of stateTransportation of personsFreight, cartage, & expressUtility servicesTelephone & telegraphLaundry, dry-cleaning, & towel serviceRents & storageGeneral repairsMotor vehicle.repairsInsuranceBoard & care in other institutionsSundry operating expensesFees for outside professional servicesMotor vehicle rentalsData processing rentalsData processing servicesPostage

CommoditiesAgricultural, horticultural, & dairyFoodClothingPersonal suppliesMaintenance supplies t 1,16,

50

Laundry, cleaning, & disinfectingDrugs, medicines, & serumsMedical & laboratory suppliesFuelMotor vehicle suppliesOffice supplies

Educational, religious, & recreationalData processing suppliesMiscellaneousRepair materials

Commodities purchased for resale

Sundry

Source: State of Connecticut. Department of Financeand Control, 1974.

6 5

51

Examples of such rules of thumb are that:

An agency usually cannot make responsible useof a budget.increase of more than 15 percent overwhat it got last year.

Agencies always overstate their budget estimates,and therefore trimming back i:== always required.

Because of their well-known control features forapplication during budget-execution, budoet data inobject of expenditure classifications often supplementinformation provided in some other format in budget sub-.missions. Summaries limited to the primary e-tegories(personal services, contractual services, and suppliesand equipment) may suffice for these purposes, and usuallyappear in the appropriations bill. The result is called a"line-item" budget, but it should not be concluded thatthe review process was restricted to the review ct suchcategories. It is possible that it was, but a budgetrequest can be reviewed functionally or programmatically,and in order to set up corresponding allotment accounts,it can retain the object of expenditure implications foreventual inclusion in the appropriations act.

Performance Budgeting. The budgetary question--what it will cost an agency or institution to operatefor another budget period--is answered in an object ofexpenditure review by considering the kinds of expendi-tures an agency plans to make during the budget period.This question may also be answeredby estimating thecost of activities planned for a future budget period onthe basis of currently incurred activity costs or work-loads. This approach is widely referred to as performancebudgeting because it involves explicit consideration ofthe input-output on production relationships in an agency,whereas object of expenditurz zeview is restricted to aconsideration of agency inputs.

The outputs of most public agencies, not only ofhigher education, are diverse and difficult to specify;relating budget dollars to these outputs has not proveda satisfactory basis for either institutional formulationof budget reqUests or state agency review and revision of

52

6 3

these requests. However, a budget estimation can be madeby relating dollars expended to units of the process activ-ity rather than to a specific output. Thus, while it maybe difficult to relate expenditures to educational outputssuch as increases in student knowledge or level of skill,it is possible to relate expenditures to units of educa-tional activity--such as a full-time student enrolled forone year, a student earning a credit-hour, or a professorspending a contact-hour with a student. The common unitsof educational activity used in higher education budgetingare the full-time equivalent (FTE) student, that is, afull-time student enrolled for one year, or the equivalentin terms of part-time students, and the student credit-unit, that is, a completed semester or quarter credit-hour by one student. Both student-years and credit-units,as activities, can be specified more precisely by level,discipline, or instructional program. Completion of adegree is frequently suggested as a unit of activity oroutput, but it was not in use in any of the states in oursample.

During budget review by state agencies, as opposedto budget formulation or development by institutions andsystems, performance budgeting is characterized by theapplication of activity and workload measures as indica-tors of the reasonableness and justifiability of aninstitution's request. Frequently, these measures, orindicators, are applied ad hoc during review, and theymay not be specified in budget instructions as-the basison which institutions should determine their budgetaryneeds. When they are so specified, they become thefactors of budgetary formulas. The budget process thenbecomes sufficiently different that we have treated itas a distinct approach to budgeting. Budget instructionsmay request that specific activity measures, usually inthe form of activity-related workloads or costs, becalculated as a part of the budget submissionor thereviewing agency itself may calculate them. The use ofthese measures as checks or indicators, rather than asnorms or standards, is the significant feature of per-formance budgeting. Because of the selective characterof budget review, among other reasons, it may be impor-tant for state-level budget agencies to keep budget

53

6 `i

submitters somewhat in doubt as to the exact criteriaused in technical budget review.

Comnonly, performance measures reiate levels ofinstructional or other activity levels either to posi-tions (faculty-tstudent ratios, for example), in whichcase they are referred to as workload measures, or toexpenditures (dollars per student credit-unit, forexample), in which case they are called unit costs.Although these measures are usually calculated from datain budget submissions, they may not be a true repres9ta-tion of conditions in the institutions, both because ofunfilled but authorized academic or administrative posi-tions, and the myriad of cost-accounting conventions thatmust be specified in the absence of elaborate cost-recharge systems. Nevertheless, these indicators doallow comparisons of institutions, and review agenciesmay attempt to promote efficiency.by basing budget rec-ommendations on the budget levels of the least-costinstitutions. This is.not always possible because offixed costs or costs associated with initial program orinstitutional development, and the review agency maysimply base its budget recommendations on projections of

an-institution's own performance indicators rather thanon the lowest in the state or on the state average.

Formula Budgeting. A long-established method ofdemonstrating budgetary need is by means of budgetaryformulas in which agreed-upon cost standards--input-output, or input-input ratios--are used to developbudget estifttes. Whereas all budget development pro-cesses use such standards or guidelines-, certain features

set some budget processes apart such as the widespread .legitimacy attached by all parties to the standards used inthe process, and the fact that such relationships are usedto develop virtually the entire operating budget request.Thus, a formula budget process involves more than a refer-ence to some standard student-faculty ratios or student

credit-hour unit costs. The budgets for instruction, as

well as for other functions or programs, are developed with

these standards, and then once developed these standards

654

are accepted, in principle, by the state-level budgetreview agencies. There will, of course, always beinstances when the full formula-generated amount maY note funded because of resource limitations. But in thisevent, the formula amount is nevertheless regarded aslegitimate, and there is general recognition that fundinghas been deficient.

The formula concept is best illustrated by givingsome examples of basic formulas. By far the most commonare,those that estimate budgetary requirements for somelevel of activity by multiplying estimates of futureactivity levels by a unit workload or cost factor. Threeexamples of formulas of this type used in budgetinginstructional costs are shown in Exhibit 3,(see Meisinger,1976, for other formula examples). Budgets for supportareas, such as general administration, may use formulasin which numbers of administrative staff at variouslevels per FTE student are used as workload factors.Operation and maintenance of plant budgets are frequentlydetermined by a formula in which the activity level isestimated in square feet of floor space multiplied by astandard cost factor in dollars per square foot, possiblyspecified for floor space of various grades.

Although there are numerous variations among formulas,the formula concept is a simple one; the conceptual andpractical difficulties are introduced by the conventionsand procedures for determining workload and cost factors.Each state, and perhaps each type of institution, maydevelop its own unique units of measurement for anactivity in addition to its own conventions for mappingthese activities into budgetary expenditure classifica-tions. To use formula budgeting, each educationalactivity and support activity to be used as an elementin the formula must be assigned to a budgetary expendi-ture classification. It is clear that unless a verycomplex formula is devised, the number of activitie's(elements) to be specifically funded by the formula willbe quite small. In the instructional area, for example,the only activity generally recognized is instruction;the formula gives no explicit recognition to departmentalresearch as an activity. Furthermore, how activities are

GO55

Exhibit 3

THREE FORMULA METHODS FOR DEVELOPINGBUDGET ESTIMATES OF INSTRUCTIONAL

COSTS

Inputvariables

Formula, parameters(norms, institutional,segment, or stateestimates)

Formulaoutput

I Projectedenrollments

II Projectedstudentcredit-hourproduction

III Projectedstudentcredit-hourproduction

Student-facultyratios, averagefaculty salaries

Average studentcredit-hour load,student-facultyratios, averagefaculty salaries

Direct instructionalcost per studentcredit-hour

Totalfacultysalaries

Totalfacultysalaries

Directinstruc-tionalcosts

70

56

specified can lead to a great many formula versions.Formulas f.)r the instructional area might distinguishbetween total student credit hours produced at variouslevels--lower division, upper division, and graduate--based either on student level or course level. Theselevels might further be broken down into discipline orprogram areas, according to the HEGIS taxonomy, anddifferent workload factors or costs could be applied toeach of these activities (such as, producing a studentcredit-hour in lower division history, or producing astudent credit-hour in graduate anthropology.) Thus, avery simple formula structure is easily transformed intoa calculation problem that most budget agencies would nottackle without the aid of computers.

The only other formula methodology in use in the 17states was one generally known as the "base formula," inwhich the estimated expenditures for the various supportfunctions are determined as a standard percentage of theinstructional and departmental research budget, whichitself is determined by a workload-factor formula. Useof the base formula is by now relatively limited; it isnot, of course, a method for budgeting instruction, which

.

forms the largest portion of any institution's operatingbudget.

In our estimation, a complete formula process must*include these fundamental characteristics: a stablemethod for determining the workload or cost factors tobe used in the process; the use of formulas to developbudgets; and the widespread acceptance of the formula-generated expenditure estimates by all or most of theparticipants in the process, including the legislatureswhich make the budgetary appropriations.

A recent survey by Gross (1973) reported that 25states were employing formulas in budgeting for highereducation. Judging from our observations during thecourse of this study, many of the states designated byGross, and perhaps by themselves, as using formulas,are relying on budgetary methods which are closer toperformance budgeting methods or another budgetingmethodology. Commonly, the factors used to estimate

5 7

budgetary requests have an ad hoc development; they areused to test the reasonableness of requests rather thanto develop them, and the resulting budget appropriationsbear little consistent relation to any formula-calculatedrequest. A unique application of formulas is found inthe State of Washington, where institutions use budgetaryformulas to develop an instructIonal budget and a com-plement of faculty positions comparable to that in a top-quality private institution. Since the inception of thispractice, however, no institution in the state has everbeen budgeted at these levels, which are simply used asa standard against which actual funding.levels can becompared. Most Washington institutions were funded atabout 70 percent of formula for fiscal year 1975. Anincrease in the level of funding it relation to theqtandard is interpreted as an improvement in quality,and a decrease in the level of funding is regarded as adeterioration of quality.

Once a formula structure has been devised, itsapplication in the budget process has just begun, forthere remains the crucial task of determining the work-load or cost factors to be used. These determinationscan take place in a number of ways, but they almostinvariably involve both a statewide advisory group ofrepresentatives from institutions to be affected.by

*the formula, and the state agencies that review theinstitutional budget requests. In essence, there aretwo bases for determining formula factors once thestructure of the formula has been specified--norms ordesiderata which are determined either from practicein other states or from some general sense as to whatconstitutes good educational practice, and measures of

. existing conditions in the state. Either of k ieseapproaches may involve extensive study directed towarddeveloping either a aet of fixed factors or the processby which such factors can be determined for each budgetperiod. Although some states continue to use formulafactors that are really norms rather than historicalrelationships, there appears to be a trend toward thedetermination of historical unit costs, and continualimprovement in the data systems that make it possibleto measure these costs.

7

58

Because of the considerable time spent in developinga framework for a formula and appropriate factors, budgetreview under a formula process is likely to be simplified.State-level budget activity for some participants may onlyinvolve certification that approved fr.niulao have beenused in developing budget requests. OfF cottrse, in nocase is the entire budget of an institution or system ofinstitutions based entirely on a formula; requests forcertain functional areas and requests for new programsmay be outside the formula. Adjustments for inflationand other technical factors, such as fringe benefits,also require review. Even though negotiation' and ulti-mate reduction of the formula-based request is requiredunder this approach to budget development, the formula

structure still provides the basis for discussion and theframework within which adjustments to budgetary totalsare made.

Program Budgeting. The budget review styles describedthus far do not tend to emphasize either the actual out-puts of institutional operations or the specific objectivesthat institutions may attempt to achieve through expendi-ture f-,f funds. A consideration of outputs or objectivesiS more likely when budget review is formally directed atanswering the question: Should funds be expended to under-take the activity et all? This is a question of choice,not of costs alone, and it is intimately connected withconstraints on the budget process which define the avail-able options.

The review of new instructional programs or proposalsfor new activities may involve some consideration ofobjectives, but the major portion of the budget request,deals with activities and programs already underway, andtraditional methods of budget formulation give relativelylittle explicit consideration to whether base activitiesshould be continued. The emphasis in the formal budgetprocess is on costing out institutional plans and deter-mining the budgetary requirements for what is largely acontinuation of the current year's activities. Theissue of choice, which confronts every budget process,

7359

is handled less explicitly, and certainly less formally,outside the technical process, through negotiation andbargaining.

Interest in developing budget techniques which explic-itly address the issue of choice in the budget process hasbeen created, however, by normative proposals for budgetreform extending over the past two decades, and recentdeclines in available state revenues. Because reformsare often not fully incorporated in the budgetary process,it is easy to fail to distinguish betweell proposals forreform and actual practice. Therefore, the descriptionsthat follow of Planning-Programming-Budgeting (PPB) andwhat we refer to as tactical budget planning draw heavilyon the budget processes in Hawaii and Wisconsin, unlikethe descriptions of budget styles discussed earlier,which did not reflect the practices of any particularstate.

In considering public service outputs, both of thestyles of budget review described below have introducedplanning techniques into the budget process in one formor another. The distinguishing feature is that Planning-Programming-Budgeting systems modeled after federalcivilian department systems rely heavily on the concept oflong-range strategic planning that includes the specifica-tion of long-term goals and objectives, and on the indenti-fication of alternative means for achieving these goals andobjectives. The analytical base for this kind of planningis derived from normative, rather th'n descriptive, economictheories of utility and welfare maximization.

There are, however, other planning concepts whichcan be connected with the budget process, such as con-tingency planning, advocacy planning, coordination,innovative-futuristic planning, and tactical planningfrom a short-range perspective. In addition, rather

than combining planning and budgeting, most statebudget processes have explicitly sought some separa-tion between budgetmaking and long-term strategy.The analytical base available for their planning alter-natives is the much larger one of descriptive theorydrawn eclectically from various disciplines. Although

60

7 4

this may appear to be merely a modification in the application of federal PPB or a reflection'of various phases inPPB's development, differences in the substance of techni-cal budget procedures between PPB and tactical budgetplannihg are substantial enough to distinguish them asalternative styles of budgeting. While recognizing thatthe PPB spirit may be more important than the letter ofthe law, it is essential to deal with actual practicesto the extent that they can be determined and described,rather than with the general principles on which reformsor styles of review are based.

Planning-Programming-Budgeting (PPB). The litera-ture of analysis and commentary on program budgeting isso extensive, and the debate that has resulted over whatPPB is and 'how it can be implemented has been so intensethat it rivals religion and politics as subjects to beavoided in polite conversation. Without debating theconclusion of Howard (1973) that the federal style PPBis dead in the states, it is important to treat programbudgeting as a formal alternative style in state highereducation budgeting because many of the informationaland analytical reforms in use or under consideration arederived from the federal style of PPB. Therefore, afterreviewing what are generally understood to be the princi-pal features and elements of program budgeting, and theattempts, at implementing;them in the states, the formalfeatures of the budget process of the state of HawaiiU. be.used to illustrate the extent to which the prin-

cipal,PPB elements have been implemented in pcificsituation.

Program budgeting involves five activitLes richtend to distinguish it from styles of bt)14r3ting already

, described.

1. Development of information and ,Aata sttm.uresbuilt around programs (groups of activita contl.:%outingto a common objective) which specify meaSum.s n( programobjectives, outputs, activities, and cost:.

61

2. Multiyear planning which results in five- orsix-yearprojections of'expenditures based on the impli-cations of current decisions.

3. Formal consideration of alternative means ofprogram output, production, and delivery.

4. Formal analytic studies, especiallf in the formof cost-benefit and cost-effectiveness stuiies, whichsupport preparation of the multiyear plan and the con-sideration of alternatives.

5. Program postaudits which.evalulte agency achieve-ment of program objectives..

Various conceivable arrangements and tiico. schedules havebeen devised to.incl4 Ode some or all of tNise activities.The timetable designed for use in federal Bureau of theBudget review of civilian agency programs illustrates'the cycle of activities and is shown in Exhibit 4.

The technical side of the PPD process involves prep-aration and review of various formal documents by request-ing and reviewing agencies: the comprehensive andmultiyear Program and Financial Plan (PPP), i,rogramMemoranda (PM), Program Change Requests (PCK), and SpecialAnalytical Studies (SAS). The special nature of each of.these is readily apparent from its title, except for theProgram Memoranda, which are summaries of major agencyrecommendations and decisions presented within a frame-work of long-run agency strategy, accomplirthment ofagency objectives, and available alto:matives. (For adescription of these documents see Bureau of the BudgetBulletin, 1967; reprinted in Lyden and Miller, 1968.)

Several states participating in this study hadimplemented various fragments of this package of formaldocuments, or were continuing to use various fragments,after having attempted implementation of program budget-ing on a much more comprehensive scale. In the main,states have turned increasingly to the Program Classifi-cation Structure (Gulko, 1972) and the REGIS Taxonomy

7 '3

62

Exhibit 4

U.S. BUREAU OF THE BUDGET PPB TIMETABLE

MONTH

September

October-December

January

January

January

February

February-July

April-August

July-September

Year around

BUDG2T PREPARATION ACTIVITY

Agency submits Program Memoranda (PMS)--in final.Eam, multiyear Program and Financial Plans (PFPs),t.h.o annual budget, and the annual legislative

:grhm to the Bureau of the Budget (BoB).

BoB reviews and recommends to the President;presidential decisions made and communicatedto.the agency.

Executive budget is presented to the Congress,major elements in the legislative program areindicated in the State of the Union message,the economic report, or in other communicationsto Congress.

Agency reviews special study program and submitsproposed list for the calendar year to BoB.

Agency updates the PFP to conform to the execu-tive budget.

BoB indicates to agency its request for SpecialStudies and for issues to be covered in PMsduring the upcoming cycle.

Agency brings Special Studies to completion andprepares drafts of PMs.

BoB responds on Special Studies and draft PMs.

Agency head makes final decisions on his programrecommendations; agency revises draft PHs; agencyupdates PFP, adding one year and making it con-form to agency head recommendations.

Special Studies are begun, carried on and com-pleted, as appropriate.

Sources U.S. Bureau of the Budget Bulletin No. 68-2, 'July 18,1967. Reprinted in Lyden & Miller, 1968, pp. 440-441.

7 763

of Instructional Programs, the principal data structures

to which states have turned in programmatically structuring

higher education activities and outputs. In a few states,

multiyear budget plans are submitted along with budget

requests with virtually the same complement of estimates

that are included for annual or biennial review. These

projections of estimated budgetary requirements are cora-monly based oilsimple extrapolations of past trends, and

are therefore given very little weight or attention in

the process of budget review. Several states were using

one or two isolated program budget documents, such as

Program Policy Guidelines issued by a governor prior to

submission of budget requests, or Program Revision

(or Change) Requests included as a part of an insti-

tution's or system's budget submission. Where implementa-

tion is fragmentary, the overall budget process remains

at most a hybrid process, and there is little to sub-

stantiate the view that these fragments have materially

altered the decision process.

HAWAII

To avoid fragmenting their implementation of PPB,

the Hawaii state legislature has written a full PPB

procedure, closely patterned after the federal model,

into the state's administrative law. Eventual imple-

mentation for all state departments began in the

Vepartment of Education in 1966 and continues for all

state agencies at the present time. No other state

visited during the course of this study (and we suspect

this is also true of states not visited) has so fully

implemented the formal procedures for doing program

budgeting.

The availability of-the program information described

earlier in this chapter is a necessary, but not a suffi-

.cient, condition for a program budgeting system. Thus

the development of a statewide program structure and its

elaboration within higher education are the first order

of business in irplementing PPB. The necessary program

structure is fairly well established in Hawaii and is

based on the Program Classification Structure. The

64

7 0

Program Classification Structure has required very litt...*!Oditional refinement or expansion in order to be applieL.instead it is the specification of program measures andactivity indicators that have been much further advanced.It is noteworthy that the educational program structureadopted for state-level budgeting extends only to theseven campus-level programs indichted in Exhibit 5 and notto the level of instructional programs of, say, the HEGIStaxonomy. This structure gives approximately the samedepth of program specification that has been developed forthe other state departments or programs.

The primary budget submission for state-level reviewis prepared by the University of Hawaii Central Administra-tion and consists of the first two years of a six-yearProgram and Financial Plan submitted as a single document.This is a highly structured document consisting of briefnarratives covering objectives, descriptions, and the sizeand scope of Level II and Level III programs, and detailedquantitative descriptions of Level IV programs. In aprescribed format, the elements of these quantitativedescriptions cover the program's objectives, its targetgroup, client group, measures of program effectiveness,,required physical resources, financial costs, programreviews, and a narrative explanation of the programanalysis performed in establishing budgetary requirements.Every quantitative element, whether a budget estimate, anactivity measure, or a measure of effectiveness, is pro-jected for each year of the six-year period. The measuresof effectiveness and their measurement units are shown inExhibit 6. The budgetary plan prepared for 1974-1980 didnot actually include the majority of these measures;those that were available are designated by a cross (+).It is evident that many of these measures are still in astate of development. Target and activity indicatorsshown in Exhibit 6 were much more widely available. Thosethat could not be provided are indicated by Na.

The formal analytical activities appear to be some-what more limited, however, than the massive program planwould indicate. Although executive, legislative, andUniversity of Hawaii headquarters staff are all involvedin the development of the program structure and measures

65

7 9

Exhibit 5

FORMAL EDUCATION PROGRAM STRUCTUREIN THE STATE OF HAWAII

Level I. Formal Education

Level II. Lower or Higher Education

Level III. University of Hawaii, Manoa Campus

Level IV. Instruction

Organized Research

Public Service

Academic SUpport

Student Support

Institutional Support

Independent Operations

Levels III and IV are represented on each of the

University of Hawaii campuses

Source: State of Hawaii. The multi-year prograM andfinancial plan for the period 1974-1980. Submitted to

the Seventh State Legislature, December 1973.

66

Exhibit 6

MEASURES OF EFFECTIVENESS AND UNITS OFMEASURE IN THE STATE OF HAWAII

Level I - Formal EducationNo measures indicated

Level II - Higher EducationAverage CEEB raw score of graduating classTotal cumulative income earnedAverage alumni gift to higher education

.

Number of people entering institutions of highereducation

Number of graduates who began not more than5 years ago

Alumni evaluation of worth of college

Level III - UOH-ManoaAverage CEEB raw score of graduating classTotal cumulative income earned during 5-10 yearsAverage alumni gift to higher educationNumber of people entering as a percent of the

eligible populationNumber of graduates who began not more than

5 years agoAlumni evaluation of worth of college

Level IV - Instructional ProgramAverage CEEB ranking of entering freshmenAverage CEEB raw score of graduating classAverage GRE ranking as percent of average

CEEB rankingAverage national GRE rankingMerit scholars entering as freshmenNumber of graduates that continue to graduate

schoolNumber of graduates that continue to good

graduate schoolsNumber of students graduating after 4 yearsNumber of students graduating after 5 years

8 .167

-

Number of students graduating anywhere after5 years

Level IV - Organized ResearchAmount of research funded from extramural sources

. Average number of journal articles furnished pertenured faculty member

Aerage number of citations of faculty members'work

Average number of books published per tenuredfull-time faculty

Estimated economic value of discoveries, etc.

Level IV - Public ServicesIncome produced where fees chargedIncome where fees charged as percent of cost of

public offeringsTotal attendance or average class size in other trainingPublic evaluation of quality of public offeringsNumber of faculty hours donated to public offeringsAmount of fees earned for consultant projectsEstimated economic value of public.services

Level IV - Academic SupportDegree to which overall program objectives are met

Level IV - Student ServicesAttendance at intercollegiate activitiesIncome produced from intercollegiate activities,Participants in intercolleg:Late activitiesParticipants in intramural activitiesStudents receiving financial aidAverage amount of financial aid received by

type of aidAverage amount of financial aid received as

a percent of costStudent evaluation of quality of student serviceIncome produced where fees chargedNumber of graduates who obtain jobs through

placement service

Level IV - Institutional SupportDegree to which overall program objectives are met

68

82

Level IV - Independent OperationsDegree to which this program's objectives are metDegree to which this program makes a net contribu-tion

Level IV - Program Target Group and Activity IndicatorsINSTRUCTIONTarget: Population--local countyActivity: Enrollment

Student credit hoursWeekly student contact hoursNutber of coursesNumber of classesSemester hours

ORGANIZED RESEARCHTarget: Total state populationActivity: Graduate student enrollment

Nutber of projectsScientist man yearsNumber of publications

PUBLIC SERVICETarget: Total state populationActivity: Courses and curriculum--number of

participants/studentsSpecial and professional programs

(nutber of projects)

_Oruanized,teachinoAnumberpersons--,as'sitted)

Direct consultations (number ofpersons contacted)

Lib-industrial relations (number ofpersons)

Seminars (number of each)ACADEMIC SUPPORT

Target: EnrollmentActivity: Enrollment

NuMber of facultyNumber of titles publishedMaterials added to the library

(volumes)Circulation (library)

Reference service (number of contacts)Nutber of requests for instructionalmedia

69

STUDENT SERVICESTarget: EnrollmentActivity: Enrollment

Number of financial aid applicationsprucessed

Number of students counseledNumber of applications for admissionNumber of graduatesNumber of bed spacesNumber of admission and records

inquiriesNumber of student-sponsored activities

INSTITUTIONAL SUPPORTTarget: EnrollmentActivity: Total acreage maintained

Total buildings (thousands of sq.ft.)Na Total dollar sales (bookstore, food, etc.)

Number of parking permits issuedNa Number of government vehicles maintained

Pieces of mail handledRequests for duplication serviceNumber of pages reproducedNumber of faculty and staffTotal general fund

INDEPENDENT OPERATIONSTarget: Total state population

Source: State of Hawaii. The multi-year program andfinancial,plan for the period 1974-1980. Submitted tothe Seventh State Legislature, December 1973.

+ = Measures of effectiveness and their measurement unitsthat were available in the budgetary plan preparedfor 1974-80.

= Target and activity indicators that could not beprovided in the budgetary plan preipared for 1974-80.

70

of performance, the university staff is the onYy stafflarge enough to provide these analytical resources.For purposes of internal allocation and management, theheadquarters staff prepares cost analyses and other stud-ies. The first executive Program Memorandum on FormalEducation, submitted to the legislature in January 1974,referred4po Issue Papers that had been prepared on highereducation access, residency status, enrollment in highereducation institutions, and the limiting of higher educa-tion growth. A special analytic study was recommended todeveloP policies for controlling the growth of enrollmentin the University of Hawaii system. The substance ofthis 60-page Program Memorandum is suggested by its majorsLction headings:

I. Overview of Formal Education

II. Costs and Effectiveness of the RecommendedPrograms. (Some of the supporting dataunavailable for the Program and FinancialPlan is included here.)

III. Program ChaRge Recommendations. (Aggregatedbudgetary implications of program changes aresummarized. No details.)

_

IV. Emerging Conditions, Trends, and Issues.(Some alternatives such as the open uni-versity, special education, and computer-assisted instruction are discugsed briefly.)

V. Selected Problems for Possible Study.

The performance audit function is well establishedandsis the responsibility of the legislative auditor.Although the auditor has no budgetary responsibilities,his staff appeared to be at the cutting edge of changesin both "le higher education program structure and themeasures c.f effectiveness., and thereby directly affectedthe formal budgetary procedures. The legislative auditstaff, a permanent staff, is also large enough to con-stitute an analytic, resource to the legislature on higher

8 571

education--possibly more so than legislative appropria-tion committee staff, many of whom were temporary. Per-

formance audits had been conducted lpy the legislativeauditor's staff on the educational television stationand on faculty workload standards (see LegislativeAuditors of the State of Hawaii, Audit of the HawaiiEducational Television System, Audit Report 71-4, April1971, and Audit of the University of Hawaii's FacultyWorkload, Audit Report 73-2, February 1973). The budg-etary implications of these audits were not apparent,although it was clear that both of the audits, especiallythat on faculty workloads, had been highly critical ofthe administration of the university, and thus had con-tributed to creating a less sympathetic audience forhigher education budget requests.

Some qualification of the principal formal elementsin use in the Hawaii system of PPB is necessary in orderto properly understand the use of these documents andteaniques in the budget process, even though there isno question that they exist. This is considerably moredifficult in the case of PPB than in the case of per-formance budgeting because PPB is an attempt to radicallyreform the nature of the budget decision process. Even

_performance budgeting is quite frequently hybridizedwith tina.ditional object of expenditUre re;.riew, castingdoubt on whether new procedures have altered the processof budgetary decision.

Schultze (1968) has said of PPB that no one shouldconceive of it as a system that makes decisions, and thesame should be said of any of the technical styles ofbudget review that we have discussed. As various pro-ponents of PPB have indicated, its purpose is to assistbudgetmakers in making decisions by providing them withinformation and analysis ana a context for making budg-etary and policy choices.

Because our investigation cf budgetary proceduresdid not extend to examining their impact on budgetary oreducational outcomes, we make no evaluation of the budgetprocess in any state. Nonetheless, our impression ofthe various budgetary participants in Hawaii is thata great portion of the information that was included with

72

. R

the Program and Financial Plan had very limited utilitywithin the prevailing decision framework. Although someof this data may have been used in the sense that it waslooked at by examiners and officials, the link betweenfairly immediate budgetary consequences and the indicatorEof activity and effectiveness is so tenuous that its use-fulness in budgetmaking is limited. Whatevei else thebudget review process accomplishes, it must Produce abudget--an expenditure plan--and all plans ake notnecessarily budgets. To what extent the perspectives ofthose making budcjetdry decisions haw. shifted to the'moresubstantive programmatic considerations is imposSible tomeasure. The fact remains that enough cost and'expendi-ture data is included in the program information providedto permit a fairly traditional object of expenditure re-view to undermine the programmatic perspective. The costsderived are based on accounting conventions and, therefore..clo not allow for analysis of total social costs in thesense implied by cost-benefit or cost-effectiveness usuallassociated with PPB.

The staff activities necessary to maintain the PPBprocedures in Hawaii are extremely time-consuming,. andbeing tied to the budget timetable they must be completedon a fairly rigid schedule. Within this structure, the

___planning-that-takes-place-has-the-appearamof-teillgrelatively ihflexible--dictated by the budtletary pro-cedures and their timetable. Planning, as reflected informal docurentso.is almost .entirely of a rational-objective nature, and does not encompass a sense orspirit of advocacy.

Tactical Budget,Planning. Unlike PPB, there is nosystem of procedures to serve as.aoprescription and pointof reference for doing what we have called tactical budg-eting. Wisconsin, whose higher eduCation budget processhas been selected as illustrative of this approach, wasone of the first states to experiment with program budg-eting. Admittedly, one interpretation might be thatWisconsin's budget procedures are simply operationalprogram budgeting. However, without prescribing these

73

- 8 7

procedures as a recipe for carrying out program budgeting,

we shall provide an informative contrast by pointing out

a number of characteristics which distinguish tactical

budget planning from PPB.

WISCONSIN

Wisconsin is an especially important st,'_e to study

with respect to analytical and informational budgetsupport activities because so many of these activitiesare documented in its budget process with staff reports

and written communications. It should be noted thaother states may be following similar procedures insomewhat less formal manner. These formalities may a:times distort or becloud for an outside observer the trut,nature of.the choices being madr- and the parameters sur-

rounding these choices, but they nevertheless the

unchanging matrix within which bmyetary issues are shape-

and refined.

As in PPB, information display clic!. analysis arecrucial elements of tactical budget ?laing. The range

of options which these data and analysis inform, however,

is considerably more limited, and opions are expres------sed,--as-fairly---ediateoperational- 434,...cislons_ratherthan,

long-range goals. State-level budget review agencies .

work primarily with institutional (caMPUs) data struc-

tured by a slightly modified Program ClassificationStructure'combined with the HEGIS .Taxonomy of Instruc-tional Programs.- The specific arrangement of these data,

however, is dictated more by a budget device known asaDecision ItemiNarrative (DIN) than it is by any program

structure.

Decision Items are identified througii negotiationbetween the University of Wisconsin Central Administra-

tion and the Department of Administration, but represent

essentially t...te incremzIntal programmatic changes for

which the unaversity requests funding for the comingbiennium. Me Decision Item Narrative is the star.aardformat and sp.ification of supporting information used

to describe these changes. Although Wiscohsin has

8374

foilowed a structured-incremental approach in highereducation budgeting, Decision Items may refer to itemsin the base budget as a consequence df imposed pro-ductivity cdts. (The productivity cut in Wisconsinrefers to the prior specification by the governor of a7.5 percent base cut for the 1973-1975 biennium forwhich the university could request a reallocAtion throughthe DIN procedure.) Decision-Items are ana_ogous toProgram Change Requests, but at a.considerably lower levelof detail. The biennial request for 1973-1975 included42 summary Decision Items aggregated at the system levelwith an appendix in which thesa were disaggregated atthe unit level in well over 200 unit Decision Items.Each unit DIN identifies the program, unit (campus),type of funding and budget change category, and giyes asummary of-budgetary requirements in terms of objects ofexpenditure. The narrative includes a justificationbuilt around the objective and an analysis of need, ananalysis of alternative solutions, a proposal and intendedaccomplishments, and identification of potential perfor-mance indicators. A typical unit DIN is several pageslong (4-5 on the average), and deals with expendituresf from several thousand to several million dollars.

Because 'the sumnary Decision /tem Narratives cutacross campuses, they are aggregated in terms of salary_

inceaSes Of various kinds (enroll-ment, support functions, etc.), new academic programs,programs for minorities and the disadvantaged, andvarious specific program improvements (library, instruc-tional programs). (See Exhibit 7 for a iis.ing of theSummary DINs.) Consequently, they not s_ggest theprogrammatic character of the underlying unit DINS.Exhibit 8 shows the unit DINs of an illustrative campuswithin the system. Note,that unit DINs are cross-referenced to the summary DINs.

More extensive examination of bu&,:atary isues inWisconsin are frequently the subject of policy or issuepapers prepared by the Department of Administration,the Legislative Budget Bureau staff, or the Universityof Wisconsin Central Administration staft. For example,for the 1973-1975 biennium, the Central Admiri:Aration

8 )75

Exhibit 7

SUMMARY DECISION (SD) ITEMSUniversity of Wisconsin System

SD# Description Requested by/for units

100 Academic Staff Compensation $ 41,205,200

101 Classified Merit Steps 4,083,900

102 General:Pr"ce Increase Allowance 1,318,500

103 Financial Aid Matching Funds 519,400

104 Potential Minimum Wage Increase (Informational)

150 Productivity Program (Cut) -21,558,700

(Reinvestment) 15,676,900

151 Specific Price Increases 2,836,100

152 Dormitory Conversion 1,231,000

153 Equipment Replacement 644,900

154 Continue Special Programs 1,045,800

155 Vet/Public Patients (Hospital) -8,000

200 Enrollment Funding 5,996,500

201 Phy.Plant: Workload 5,658,400

202 Utilities and Heating Coss 4,198,800

203 Financial Aids-Incr.Workload 165,600

204 Debt Service on Acad._Bldgs. 12,506,500

205 AuXiiiery Enterprises (§-e-11:-ti.ifiij)-

250 Additional Minor Remodelling 357,000

251 Staff, Equip.+ Oper.New Bldgs. 1,545,000

252 Fed.Indirect Cost Support 7,600,000

253 Computing Workload 823,000

254 Student Service Workload 314,300

.255 Extension Contin.Educ:Enrollment 1,558,500

256 General Opns.+ Svcs.Workload 853,600

257 Instructional Svcs.Workload 2,487,600

258 Other Workload Increases 317,800

259 Univ.Hospitals Revenue Budget -5,460,700

260 Fee Income Offset (771fo. only) 13,690,000)

261 Federal Funding Growth 14,000,000

300 Sea Grant Match/Marine Studies 417,300

° 302 Minority/Disadvantaged Programs 3,147,000'

350 New Acad.Program Proposals 3,778,400

76

0 0-

SD# Description Requested by/for units

351 Health Science Curricula $ 3,690,200352 Instructional Improvements 3,367,500353 Wisconsin Idea, Ext.+ Publ.Svc. 8,951,300354 Teaching Hospital Support 7,488,500355 Field Stations and Farms 186,8120356 Affirmative Action for Employment 554,900357 ETV Network Capability (p.ar ECB) 453,800358 Computing + Library Improvement:: 829,100359 Improved Mgt.+ Support Programs 172,500360 Student Services Improvements 612,700361 Research Programs + Grad.Assts. 1,811,600362 Other Program Improvements 570,000

So ce: University of Wisconsin System 1973-1975 BienniaBudget Documenation, Book V, U.W. System BiennialBudget Request and Decision Item Narrative, August 1972

77

Exhibit 8

EXAMPLES OF UNIT DECISION ITEMSUniversity of Wisconsin, Madison Campus

Description (+Summary DIN X-Ref.)* Unit request

Required Base Cut for Productivity (SD 150) -7,647,700Restoration to Cover Fed.Match.Requirements

(SD 150) 1,014,100Accelerated Studies (SD 150) (3 yr.degree opt.) 778,700Library Improvement/Statewide Resource .(SD 150) 807,400Restore Balance of Base Cut (SD 150). 5,737,500

Conversion of Res.Hall to Academic Use (SD 152) 144,200Obsolete Equipment Replacement (SD 153) 436,200

Utilities (SD 202) 2,054,700

Small Scale Waste Systems (SD 154) 243,700

Specific Major Price Increases (SD 151) 270,000

Additional Funds for Minor Remodeling (SD 250) 125,000

Dept.Support for Fed.Funded Programs (SD 252) 4,000,000(PR)Maint.St Custodial Costs,of New Bldgs. (SD 251) 399,100

--eperatiJmy-o-f-Animal-&-yolatiie-Waste:-Incinerator (SD 251) 122,900

Staff, Equip.& Operate-New Bldgs. (SD 251) 442,900

Teaching Univ.Hosp.-Full Funding of Educ. .

Costs (SD 354) 1,488,500Environmental Curriculum & Course Develop-

ment (SD 350) 83,400Improvement in Legal Education Progs. (SD 352) 565,000Allied Health & Physician Asst.Progs. (SD 351) 563,900Improve Undergrad. Health Science Instruction

(SD 351) 645,600 .

Improve Bagic Med.School Programs (SD 351) 1,004,700New Advanced Med.School Programs (SD _351) 504,000Campuswide Animal Resource Center (SD 361) 139,600

Improvements in Postgrad. Med.Education (SD 351) 216,200Improvement of Instructional Media (SD 150) 258,200 ,

78

Description (+SumMary DIN X-Ref.)* Unit request

Increased Opportunities for Minority& Disadvantaged Students (SD 302) 791,600

Improved Research & Instruction in Col.ofAgr. & Life Sci. (SD 361) 743,000

Improvement in Instructional Use ofComputing (SD 358) 367,600

Establishment of Statewide Public Servicefor Mental Retardation (SD 353) 578,000

SUBTOTAL (Certain items)

NET TOTAL -(All items)

22,129,800

$23,078,000

Summary DecisionlItem Number cross-referenced.11

Source: Universi.ty of Wisconsin.System 1973-1975Biennial Budget:.Documentation, Appendix, Book V,Unit Decision Item Narrative, August 1972.

79

. 9 3

of the university was mandated by the governor and thelegislature to prepare 10 studies on varimis budget-related policy issues and to submit them with the biennialbudget request. (In some instances, the policy issue hadalready been the subject of an issue Paper or study-by theexecutive or legislative staffs.) The result was a for-mally.documented policy dialogue on the designated policyissues. The character of these studies can best beillustrated by indicating the subject of each study andsummarizing briefly the questions to which the universitywas asked to respond:

User,Fees. Study ways in which user Eees couldbe assessed to defray general revenue costs ofnoninstructional activities and use of educationalfacilities for noninstructional purpose's. (The

university had previously been dirc7ted by the gov-'ernor to study the potential for raisin4.additionalrevenue in intercollegiate athletics, intramurals,and other noncredit use Of physical education facil-ities.)

FaCilities Utilization. Examine current andprojected campus utilization, how utilizationcould-be---improvedj-arid-vhis-A-easmus-should-be-closed. Develop criteria for maximum and minimumcampus size and acceptable ranges'of cost perstudent for various disciplines and instructionalcategories.

Physician's Assistant Program. Develop am, in-ventory of information on physician's aSsistantprograms in other states and a model programstatement.

Academic Program Evaluation. Develop specificcriteria for evaluating academic degree programs.Identify the range of academic programs which dreto be offered at each institution. Establishcriteria for minimum and maximum class size.Determine the nuMber of comparable programs justi-

,

9 `It

0

fied among various campuses within the Universityof Wisconsin system. Develop a plan for theequitable dismissal of faculty. Indicate the pro-gress to date in elimination of unnecessary programs.

Plans for Wisconsin Idea Funds. Indicate theefforts to date in addressing the broad questionsof the extension (outreach) program'sdorganiza-tion, coordination, and planning.

Student Employment Budgeting.. Develop a methodfor revising the budget and accounting system sothat student employment becomes a separate DecisionItem.

Center System (Community Colleges) Programmdng.Study methods for improving high cost and/or under-utilization of small classes at certain Centercampuses. Set specific performance targets. De-velop coordination plan with the vocational-technical campuses. Develop merger plan for lowenrollment centers. 2,

Educational TeChnology-Academic Computing.Rejustify and.specify budget request to take pro-

--ductivity-savings-and-Teinveat-themAn-these-to --areas.

Physical Plant Study. Reevaluate present methodused to develop management objectives and workloadprojections on newly constructed facilities withoutside consultants. Consider outside-contractingand establishment of systemwide standards.

Excess Dormitory Capacity. Respond,to previous, communications on methods for reducing dormitorycapacity.

There is clearlv much more detail in the tacticalbudgeting approach blan in program budgeting. In programbudgeting, the objectives of the.instructional programof an entire campus are considered, whereas tactical budg-eting is directed to the more discrete objectives of,

j81

say, attempting to improve the instructional uses ofcomputers or clarifying the goals of a new teaching hos-pital. Issue papers are related to budget options thatare under consideration within the decision horizon orin'the very near future. The papers are'related to whatappear to be the tactical issues of operating the univer-sity, rather than,to the more global concerns of who shouldhave access to higher education and how it should be paidfor: The methods used in the various policy papers areeclectic; because they are identified with specificagencies and represent agency positions on various issues,they poSsess a clear sense of advocacy. A number ofthese issue papers, by the nature of the questions that

:they pose, require the university to consider variouscontingencies. In this way especially, they stimulatethe institutions or the university system to respondtactically to a condition that may arise instead ofmerely providing an open planning framework within whichinstitutions are more likely to respond strategicallywith more abstract organizational objectives.

As we noted of Hawaii'L PPB system, the informa-.tional demands.of the Wisconsin procedures also aregreat. .Similarly, we have no way of really knowing theextent to which various kinds of information play a

ificant-rol-e-4-n--b:idget---choicee:-----It-sh(5uld-be---noted

that budgetary information in WisCOnsin can be relatedto both programs and objects of expenditure, and thatdetail for the latter is sufficient to support a tradi-.tional object of expenditure revieW as well. Within theuniversity's budget planning section in the CentralAdministration there are two groups which prepare thebudget submission. The budget preparation group handlesthe mechanics of data processing, including the rationaliza-tion of objects of expenditure with institutional programs.The planning and analysis group prepares the issue paperanalysis and.establishes budgetary procedures, criteria,and budget planning parameters. Although the manpower re-quirements are considerablyygreater, this arrangement pro-vides for a definite, although minimal, separation betweenthe routine of budgetmaking and the flexibility that plan-ning activities require.

82

9 t)

5.The Context of Formal Budget Documeni.s

The te4Inical approaches to budget review discussedin the previous chapter are implemented primarily throughrequests for inclusion of data in the documents sUbmittedby the institutions for state-agency review. These docu-ments, because of time constraints and the limited sizeof review staffs, become the.principal vehicle, albeitnot the only one, of course, for channeling informationto those who must determine the appropriation for publichigher education. This chapter isA.n two sections: Thefirst deals with requests from institutions and statewidegoverning'boards; the second with the exchange of docu-ments at the state level and the relationship of thesedocuments to the original requests.

A discussion of the content of budget documents tendsto emphasize the formal as opposed to the practical aspects

--of-budget-review,-One-uftiversity-president-interviewedexpressed the belief that the outcome of the budget pro-cess in his state, the legislative appropriation, hadrelatively little to do with the technical budget reviewprocedures. He felt the procedures were only a way ofdeveloping a rationalization for decisions made on thebasis of political realities. Be that as it may, budgetdocuments do give some indication of how state agenciesreview budgets. They, also represent ii)art of an agency'sor institution's attempt to be accountable on budgetallocation decisions, for these documents are, in.onerespect, a public statement explaining why certain budgetchoices were made. Although reformed budgetary procedures

83

9 7

do not decide budgetary issues, we have found it diffi-

cult to square the view that "New budgetary methods can't

be used" with the apparent changes in state budgetarypractice. A discussion of the consequences of thesechanges*is reserved for Chapter 6.

INSTITUTIONAL AND SYSTEM BUDGET REQUESTS

In describing the content of formal higher educationbudget requests, we make a distinction between requestssubmitted to a governing board of a multicaApds systemand those submitted to a coordinating board. If insti-

tutional requests to a governing board are not subsequentlyreviewed by executive and legislative fiscal agencies,they were not considered for this report. Therefore,

when a state has a statewide governing board, our discus-sion refers to the budget submission from the board staffto executive and legislative agencies. This is also true

of multicampus systems, such as the University ofCalifornia, where individual campus eubmissions to themulticampus system board are not reviewed subsequently

by state-level agencies. In reviewing the budget requestsof states with coordinating boards, the requests to thecoordinating board are discussed, rather than the sub-sequent coordinating board recommendations. Where neither

a coordinating nor governing board exists'at the state

level, we discuss the budget submissions that are reviewed

by executive and legislative agencies, that is, the insti-

tutional or multicampus system requests.

Following the convention used in Glenny et al. (1975),requests of campuses with a single governing board thatare grouped together during the process of review are

referred to asither aggregated or consolidated. Aggre-

gated implies that all campus detail is provided in such

a way that the equivalent of an individUal campus review

is possible. Consolidated implies that all campus detailshave been lumped together in a multicampus total, andthat therefore review of individual campus programs is

generally not possible.

84

NONBUDGETARY CHANNELS OF DATA SUBMISSION

The principal budget data submission channel for allstates participating in this study, and we suspect forall states, is the budget request document itself. Inmany instances this document is accompanied by machine-readable supplements of all data included in the budgetdocument. Because of the short tine span during whichbudgets are reviewed at the state level, it is desirableto expedite the process by including all relevant data ina single document.

In one state, Florida, budget submissions are supple-mented by extensive data submitted on magnetic tape.This submission is related to use of a formula requiringdata at relatively low levels of aggregation (instruc-tional/department level) and the need to make large numbersof calculations in arriving at formula-based requestsdetermined on a system or statewide average.

Channels other than the budget request document it-self are used when formal submissions are relativelyshallow in providing data, when issues arise in thebudgetary process for which relevant data have not beenprovided, or when more timely data become available laterin the course of review. Information sought on a non-routine basis is frequently gathered over the telephone,and separate submissions often are made once the processof review has begun at the state level and issues areraised which require more data. Hearings before theegislative-commi-ttees-or-the-gavernor-,-and-more-informai

meetings, usually between staff, also provide an oppor-tunity for the exchange of information.

These information-gathering processes directly re-lated to review of budgets should be distinguished frommore general collection of information for other state-level planning or review activities, or for externalreporting purposes such as the Higher Education General-Information Survey (HEGIS). Many state-level highereducation agencies have a routine reporting schedule forthe submission of various reports which are entirelyseparate from the budget process. Such reports may

85

9 9

include more detailed or more accurate historical enroll-ment figures; descriptions of facilities; certain studentcharacteristics, such as counties of residence, to clarifyenrollment patterns; and other historical information.Primarily because of timeliness, these reports have littlebearing on the budget process. They may actually containdata that could have a bearing on the review process, butbecause of the importance of having all relevant budgetdata in one document, most budget submissions are v.irtuallyself-contained.

In several states, most often in connection with theactivities of the executive budget office, an informal,but very important information "system" is_the routinereporting of required information that must a6companybudget administration or execution. Where they are used,various administrative forms used for such matters,among others, as personnel changes and requests for changesin allotments constitute a valuable information-gatheringnetwork for monitoring institutional operations. In onestate this highly informal but effective system was re-ferred to as "those little slips of paper which trickleinto the budget office."

BUDGET REQUESTS

The physical formOf institutional and system budgetrequests.ranges from glossy documents with relativelysuperficial presentations of budgetary plans to extremelydetailed documents running to over 600 pages of single-spaced narrative and tables. In essence, they all describe.the fiscaLneed.for state resources for the coming budgetperiod, either a biennium or a year. Usually, the re-sources required for operating the entire institution orsystem of. institutions are described and matched againstestimates of revenues available from nonstate soUrces.The state is requested to provide funding for the balance.

In a-few instances,- where other budget materials andinformation are submitted, the formal budget document maybe primarily a public document rather than an administra-tive form. This is the case in California and New York,

f

86

states in which, to varying degrees, the systemwide budget'.equests of the state university and college system, theuniversity, and the state university do not contain thesupporting detail on programs and expenditure items thatmost budget requests include. Budget requests maY alsobe simplified where the use of budgetary formulas cover-ing the major expenditure areas is well established.Such is the case in Mississippi, although institutionsthere also submit an object class budget request likethat of all other state agencies which have been backedout of their formula request.

For each expenditure category (whether objects ofexpenditure, functions, or programs), data are uSuallyincluded on expenditures of the most recently completedyear for which totals are available, estimates of expendi-tures for the current budget year, and estimated expendi-tures for the forthcoming budget year or years underreview. As discussed below, some states require budgetestimates for five or six years beyond the current year.We emphasize again that a large amount of the data includedin a budget request is not historical data, but estimatedor projected. Because of the increasingly long leadtimes required for the budget process, budget requestssubmitted for a subsequent year, say budget year t+1,will certainly not include actual data for a year laterthan t-1, the past year, and possibly t-2, becauseaccounts will not have been closed yet. Budgetary plan-ning at the institutional level must usually begin farenough in advance that the appropriation process for theforthcoming year will not have been concluded.

The quantity of information in budget requests isrelated to the number of institutions or systems whichSubmit requests to state-level agencies. Channelingrequests through multicampus system administrations orthrough'various coordinating or governing boards, sothat the resulting requests are consolidated or agc7re-gated, does reduce the amount of budgetary infor,--it ionhandled by executive and legislative budget agc.c1.1.The type of institutions involved--whether underyJuate,undergraduate through master's degree, or doctora:.granting, and with or without independent departmPntal

medical or health science schools--also affects the com-plexity of the information flowing to the state level.There is, of course, no absolutely clear relationshipbetween type of institution and amount of budgetaryinformation submitted because some of the more complexdoctoral-granting research institutions have constitu-tional autonomy, and this may reduce somewhat the amountof information they provde to state agencies.

For the 17 states included in this study, Table 2.shows the number of nonmedical four-year public campuses

'in each state, the number of campuSes which submit budgetsto be reviewed as-part of a multicampus system, and thelevel_of data aggregation at which review, informationcollection,and information use take place. Appendix C-1

shows a tabulation of the institutions that are excluded,- that is, that are not necessarily reviewed according tothe procedures described in this report. A table suchas this'can only grossly represent differences betweenthe states, but it does serve to introduce sone of thevariations which exist in state practice.

Medical programs and the other institutions notincluded in the table are reviewed at the state level,but the review procedures discussed here and the commen-tary which follows do not generally apply to them.Obviously, the budgets for medical schools are one ofthe major items undergoing review at the state level,representing one of the most costly activities in thehigher education budget. Also not thoroughly examinedwas state-level review of community college budgets.

In a few states indidated in the tate, budget-reviewfor community colleges, technical or vocational colleges,and university branch campuses is similar to the reviewof four-year institution budgets, or is included in the

same process (see Appendix C-2 and Schmidtlein & Glenny,

in preparation). Budget review for these colleges, .

therefore, depends on the same parameters as does budg-eting for four-year institutions, and not on statutory

---,formulas, which leave much less discretion to reviewing_

agencies.

Table 2 indicates that states with large numbers of

institutions have some sort of consolidating or aggre-

88

10 '7)eoo

,

4.1,1

Table 2

NUMBER OF FOUR-YEAR CAMPUSES WIMOUT MEDICAL PROGRAMS REVIEWED, BY INCLUSION IN MULTICAMPUS SYSTEMS,FOCUS OF REVIEW, AND LEVEL OF INFORMATION DETAIL PROVIDED

States and numberof four-yeArinstitutions re-

viewed

California

Colorado

Connecticut

Florida

HawaiiIllinois

KansasMichiganMississippiNebraska

New York

Pennsylvania

Tennessee

Texas

VirginiaWashingtonWisconsin

Number of four-year campuses with- Focusout medical programs included in ofmulticampus systems which are SHEAaggregated (A) or consolidated (C) review

Focus ofexecutiveand legis-lativereview

Lowest level of detailreceived or used byexecutive office orlegislative branch

28

12

5

University of CaliforniaCalifornia State University

and collegesNoneUniversity of Connecticut

(C)

(A)

(C)

8

20

0

1

Connecticut State Colleges (C) 49 State University Systems

of Florida. I

(C) 9

2 University of.Hawaiii. (A) 2

12 University of Illinois (A) 2

Board of Governors (A) 4

Board of Regents (A) 4

Southern Illinois University (A) 2

6 None 015 None 0

a None 07 University of Nebraska (A) 3

Nebraska State College (A) 4

22 State University of New York (A). 22

University of Pittsburg (C) 2

23 Temple University (C) 2

Pennsylvania State University(C) 5

10 None 0University of Texas (A) 6

30 University of Houston (C) 2

Texas A&M University (A) 3

15 None 06 None .013 University of Wisconsin (A) 13

Not app. System.

Campus Campus

System System

Campus Statewide

Received Used

SystemSystem

and campus*

Below campus Below campusSystem . Systemand campus* and campus*

Statewide Statewide

Campus Statewide Campus CampusCampus Statewide Below campus Campus

Campus CampusCampus . CampusCampus StatewideNot app. System

Not app. CampusSystem System

and andcampus* 'campus*

Campus Statewide

Not app. Campus'

CampusCampusCampus

Campus CampusBelow campus CampusCampus CampusBelow campus Below campus

Below campus CampusSystem Systemand andcampus* campus*

Campus Campus

Below campus Statewide

Campus Campus CampusCampus Below campus CampusStatewide Below campus Cluster

* /n California, Connecticut, and Pennsylvanip, state colleges are reviewed or-p.rovide information atthe campus level, while the universities are reviewed or provide information at the system level.

103

gating process for combining institutional budget re-quests and thereby reducing the number of budget units

which must be reviewed. In some instances, a coordinatingagency or a statewide governing board reviews individualinstitutions at the state level, making possible a moreaggregated review by-executive or legislative agencies.As discussed more fully below, submission of requests'consolidated at the system or statewide level does notentirely preclude review and analysis of institutional orsubcampus data. A request consolidated at the systemlevel may still contain indivi,-ual campus details onpertinent budgetary issues.

The entries in the table for information collectionand use are also subject to some qualification. Thetable indicates the aggregation level at which data iscollected more or less comprehensively. Similarly, thecolumn indicating use specifies the aggregation levelfor which data has some clear-cut use in the-total rangeof budget-review activities. Use of information isdifficult to determine when all that.is known is thatdata are available. Use of data in a formula or a com-parison is more easily determined. In the review of newprograms, supporting data is necessarily at the subcampuslevel, even though the main body of the budget requestis n terms of multicampus systems or institutions.Actually, therefore, in all states where new programs con-stitute a separate budget category, some data is reviewedat the subcampus level.

FORMATS OF INSTITUTIONAL AND SYSTEM REQUESTS

The types of budgetary information categories orformats used in budgeting Were discussed in Chapter 2..

To-descr-ibe-each-individual-institutional_or_systembudget request requires more than a taxonomy of formats:

-- because these documents often use multiple formats and'a variety.of ad hoc representations of data. The formatsof appropriation bills categorized in Glenny et al. (1975)

do suggest formats that must be a part of the requests,but do not clearly indicate the breadth of informationsupport that adcompanies the original requests. Ordinarily;

just as the-appropriation bill is only suggestive of thebudgetary control that is possible, the budget bill givesvery little indication as to the budget review methodologyemployed. The examples drawn from Connecticut, Hawaii,and Wisconsin in Chapter 4 illustrate the difficulty inlabeling these according to a simple taxonomy. All three

--Involve-some-use-of the Program-Classification-Structure:-yet all three also employ an object of expenditure classi-fication. Format alone is hardly sufficient to describethe data requirements of a budget process.

INGREDIENTS OF BUDGET SUBMISSIONS

The informational content of budget requests variesalong so many dimensions that it is difficult to general-ize about them. Table 3 describes institutional or systembudget submissions in the 17 states in terms of whetheror not the submissiOns include several features which aresignificant in determining their information content:

SubmisSion-of wage and salary detail for bothacademic and nonacademic positions disaggregatedby individual positions.

Submission of budgetary information on enroll-ment by level of student or course for explicit

........... ..... or4mplicit consideration in determining funding.

Submission of budgetary information on the fullrange of instructional programs for explicit orimpl-icA-use-In-fund-ing-oh--the-bas-is-of-d-i-ffer--ential costs.

Submission of multiyear budget prOjectionrbeyond the year for which funds are requested.

Biennial budget submissions.

Use of budgetary formulas to generate the majorportion Of the budget request.

Submission of educational output measures.

91

1 5

Table 3

BUDGET DATA SUBMITTED TO EXECUTIVE AND LEGISLATIVE AGENCIES

Budget data Data sources

Multi- Out-Wage & Enroll- Instruc- year Bi- putsalary ment tional projec- ennial For- mea- Institutional and

States detail levels programs tions budget mula sures 9 system requests

California Yes Yes No No No No1

No University of CaliforniaCalif. State U. & Colleges

Colorado No Yes Yes No No No No No institutional requests

n.)

iuJ

info. info. examined-in-Coloradc------------- -- --

Connecticut Yes2

Yes No No No No Yes3 Univ:,rsity of Connecticut

Florida Yes4

-. Yes Yes 6 yrs.5

No Yes No State Tl. System of Florida

Hawaii No No No 4 yrs. Yes No Yes6

University of Hawaii

Illinois Yes7

Yes No 5 yrs. No No Yes8

University of IllinoisNorthern Illinois University

Kansas Yes No ,,a No No No No University of Kansas

Michigan Yes9

Yes ies No No No Yes1° University of Michigan11 i

NoMississippi No Yes No No. Yes No Board cf Trustees Request

Nebraska Yes12

Yes Yes13

1 yr. Nc. No Yes14

Unive:lity of Nebraska--

Nebrasta State College

New York No Yes Yes No No No No State U. of NY at Albany

Pennsylvania No Yes No 4 yrs. No No Yes University of PittsburghIndiana U. of PennsylvaniaUniversity of Pennsylvania

Tennessee No Yes Yes No r) Yes No Middle Tennessee State

Texas No Yes Yes No Yes Yes No No institutional requestsinfo, info, examined in Texas

Virginia Yes Yes No No Yes Yes15

No College of William and Mary

Washington Yes16

Yes YPs No Yes No Yes University of Washington

Wisconsin No Yes Yes No Yes17

No Yes University of Wisconsin

1

The California State Universities and Colleges used a formula to gennrate faculty position requests for

internal purposes, but state agencies did not,use forollas. The University of California did not use

a formula in detormining ingructional budgets.

2

Wage and salary detail provided only by the University of Connecticut.

3

Only student contact-hours.

4Wage and salary detail for currea and new positions highly aggregated at the universitywide program

level.

5

Multiyear projections were abandoned in the FY 1975-76 budget cycle.

6See Exhibit 6 for a listing of the output measures used by the University of Hawaii.

Wage and salary detail highly aggregated, showing only the number of priiiiions'EFliVii-brOrbgtapT---------

For FY 1974-75, the University of Illinoh; included a special study of nonacademic salaries, showing

detail on positions and levell'..With its budget request. All institutions provided detail on faculty

positions by level and tenure stai'us.

B

Earned degrees.

(.4 '

9

Wage and salary detail is highly aggregated.

10

Student credit-hours, student years completed, and earned degrees by level by program.

11

In FY 1975-76, the Board of Trustees began use of budgetary formulas based on instructional program

costs.

12University of Nebraska submitted wage and salary detail only for academic staff by level and school.

Nebraska State Colleges submitted wage and salary detail only for administrative staff.

13Only the University of Nebraska included instructional program detail in its request.

14

Enrollments and student credit-hours by level and school.

15

Formulas in virginia were used only as guidelines by state agencies.

16Wage and salary detail highly aggregated, showing positions by level aggregated by program.

17The biennial budget process included a formal process for annual review.

. ,

107 \

r. L

To sumMarize just how informative these documents are,we would have liked to identify the "key" dimensionsof budget submissions. It has been impossible to dothat satisfactorily; some dimensions clearly transmitmore information than others. However, the impressiongiven about information content from observing the numberof columns shown in Table 3 is not entirely inaccurate.The University of Washington budget request was undoubtedlyone of the most extensive we examined, and the budget re-quests of the California State University and CollegesSystem and the Mississippi Board of Trustees were undoubt-edly the least extensive. In all cases, the intent of the

chart-is--to--treat-those-request s-which_receive_prattention by legislative or executive agencies. As weshall note later, legislative agencies sometime:-; focustheir review on the governor's budget rather than on theinstitutional or system requests.

FORMULAS

The use of formulas according to the strict defini-tion given in Chapter 4 makes the use of budgetary informa-tion in relation to budgetary outcomes fairly explicit.Their use at the state level, the only use we are con-sidering, provides a structure for the information inbudget requests and generally makes that information moreintelligible. Note that formulas may be used for reallo-catioh within higher education systems, as in theCalifornia State University and Colleges System, but notin cieveloping budget requests. Formulas are likely toreduce the information load in budget submissions be-.cause their use obviates a need and an interest for themore sophisticated types of educational program and out-put measures. The use of formulas usually goes hand-in-hand with a funding methodology which takes into

--account-differences-in-instructional-and-support-costsassociated with course or student level and the disci-pline of the instructional program. Because. formulas

_provide a direct linkloetween data and budgetary outcomeswhen used in the strict sense, their ut3e in'this manneralwayi represents considerable st eaf'. wQrk and negotiationbefore their application. This deviolopment or measufe-

94

10

./

ment process provides a vehicle for exchanging infola-tion of all kinds which then can be left out of the

Iformal budget process. "Formula" use without these pre-agreements requires information exchange and adjustment

.after the fact.i

iAlthough the use of formulas involves a greatdealof information on higher education activities in termsof apparent scope and substance, it should be keptfinmind that much of the data involved is fictitious 'withrespect to actnal institutional operations, or is at sucha level of aggregation that it indicates very littleabout institutional practice. All the budgetary formulas

. encountered in this study employed budgetary parameterswhich were statewide averages. This being the case, aformula budget indicates only a planned level of enroll-ment or student credit-hour production-for the institu-tion or system in question, and not institution-specificdata on actual incurred costs, student-faculty ratios,or other workload parameters. The Texas formula is anexception in that enrollment levels are actual; its costfactors, however, are statewide averages.

Formulas rarely generate the entire operating budgetand the formula-generated amounts rarely survive theentire review process uncut. Therefore, even formulabudgets require additional support information, whichsometimes may call for considerable detail.

DATA TO SUPPORT COST DIFFERENTIALS FOR LEVEL OF STUDENTOR COURSE AND INSTRUCTIONAL DISCIPLINE

Where formulas are not used in a strict sense, theexplicit use of data disaggregated by level and instruc-tional program is less obvious. When provided, thesedata tend to appear in the budget submission as some-thing of an appendage, and frequently are part of ajustification for workload increases (and higher plannedexpenditures) brought on by enrollment shifts to moreexpensive instructional programs. In states wherebudgetary review is more judgmental or discretionary,

95

109

this kind of data may be subcampus data pertaining to a

particular institution rather than to a statewide average

of the kind used in formulas.

The present study discovered no workload measures or

parameters that would not be familiar to practitioners.

However, some of the distinguishing ft?atures of these

budget parameters should be noted. Enrollment data, for

example, is quite widely reported in budget documents and

is widely reported to be the prime element "driving".

budgetary outcomes. In most instances, the enrollment

figures reported are not determined by actual enrollment

counts, but rather by dividing the number of student

credit-hours earned at various course levels and programs

by a standard student credit-hour load, usually 15. In

rare instances, periodic surveys may be made to determine

a more accurate credit-hour load factor. Thus, because

enrollments are really secondary data and student credit-

units primary, the use of student credit-units as a

budgetary datum is much more common. It was found to be

the principal workload measure (either as a unit cost or

a productivity index) in 12 of the 17 states. Seven

states employed student-faculty ratios. Many states, of

course, calculated more than one such measure.

Actual enrollment data is the basis for generating

budgets under the formula structure employed in Texas.

Audited histoiical enrollments as of the 12th class day

of the fall quarter is the only workload element for

which the legislature will appropriate funds for instruc-

tional activities. Because Texat 'has a biennial budget,

there is a critical necessity to update the enrollments

used in the formula in order to base tppropriations on'

the most recent enrollment level possible. The updating

of these enrollments is the responsibility of the Texas

Coordinating Board.

As students increasingly take courses at higher levels

than their student status, the distinction made between

workloads by course or student level has become an impor-

tant one for budgetary purposes. Most states that make

this distinction use course level to determine workloads.

The number of levels to be distinguished also ham grown.

.1.1 0

96

Four levels--lower division, upper division, graduate(master's), and graduate (doctoral)--now seem to be re-quired in order to adequately reflect differences inresource use among programs. Some states are now con-sidering using the method of instruction (lecture,laboratory, or tutorial) as a factor reflecting resourcerequirements. Ultimately, essentially the same proceduresare used by states wishing to take into account levels bycourse, student, or variations in instructional programs,there being only three or four different parameters in-volved.

The desire, primarily held by legislatures, forsimple indices that encompass the whole range of activi-ties, is reflected in Connecticut's use of the SCHLDE(Student Credit Hour Lower Division Equivalent), an out-put measure relating student credit units earned atvarious course levels to student credit units at thelower division level. The number of SCHLDEs per FTEfaculty computed for various institutions or departmentsprovides a single index measure of workload or facultyoutput.

UNIT COSTS

In spite of the attention given to unit costs indiscussions of budgeting, management, and planning forhigher education, the actual use of unit costs in statebudget analysis in any exacting sense is relatively in-frequent. A unit cost can be calculated quite simplywith budgetary data by dividing any of the many expendi-ture figures available by some workload (production)measure; for example, enrollment or student credit-hourproduction. For the unit cost calculated in this fashionto have any usable budgetary implications, it must be

. determined for some cluster of organizational activitieswhich can be distinguished either in terms of activityinputs or outputs.

Unless one is concerned only with direct costs, anyorganization has trouble accounting for its use of inputsin a production process if it does not have a well de-

97

11 1

veloped cost accounting and recharge system. Such systems

are often lacking in higher education because the ill-

defined nature of educational outputs makes it a partic-ularly_difficult task to_compute costP_With,respect_to

outputs. Thus, only two of the 17 states, Florida andTennessee, make explicit use of unit costs (direct costsattributable to the production of student credit-hours)in their budget processes. Several other states performroutine unit cost studies, but do not make explicit use

of them in the budget process. A larger number of statesrequire that unit costs of one sort or another be reported,

-but make limited use of them in budget analysis. The-

inclination to use this kind of,data does seem to beincreasing in state agencies, however, and we can expectinterest will continue in developing institutional andsystem information systems to provide accurate data of

this kind.

Unit cost data in all instances we observed wereaverage unit costs and, therefore, made no distinction

between fixed and variable costs. These data are based

on the explicit or implicit assumption that unit costs

are invariant with the scale of operations, that is,that costs increase or decrease in direct proportion toincreases or decreases in enrollment, student credit-hour production, or other measures of educational activity.

Cost data which is a direct input to a budget formula

(as in Tennessee, for example), are measures of the direct

use of nonrestricted funds for instruction. The computa-

tion of estimated total instructional budgetary require-

ments follows directly from projections of student

credit-hour production. Nonsate revenues are subtractedfrom to.tal budgetary requirements to arrive at the level

of funding requested from state sources.

In Tennessee, direct costs of instruction were de-

termined biennially for 30 disciplines at five different

instructional levels. Similar cost data for five differ-

ent disciplines and four course levels were used in

Wisconsin to adjust funding for instruction for marginal

changes in enrollment. Florida generated and used unit

cost data very much like Tennessee, but has developed

common accounting categories and procedures and a computer-

1

based system to a fuller'extent. The Texas formula isone of the oldest, even though it undergoes continual re-vision and expansion to cover more functional areas.Strictly speaking, it is not a cost-based formula, but

---rather-one that relies on average-faCulty -salary andstudent-faculty ratio parameters in the instructional area.

It should be made clear that the unit cost data inuse in the states is highly.aggregated and more accuratelydepicts relative cost patterns in general than relativecosts between specific organizational units. It is there-fore of very little value as an analytic device foridentifying efficient or inefficient educational practiceor properly costing out educational activities in anaccounting sense. Its use is highly pragmatic insofaras it is an acceptable methodology to all budget processparticipants for estimatihg future expenditures in theinstructional area.

In other states, the development of unit cost data,even when it hasbecome routine, takes on the appearanceof a special analytic study rather than that of part ofa budget process. In California and Washington, thecoordinating agencies have performed this function; forsome years, the staff of a council of university presi-dents conducted a cost study in Michigan; and in Hawaii,an analytic study of instructional unit costs was con-ducted in 1973 by the University of Hawaii's centraladministration staff.

These cost studies tend to be somewhat morerigorously conceived than cost studies which feeddirectly into a budget process. They also are morelikely to attempt to allocate and determine total costs(both direct and indirect) by making assumptions thatmight not be acceptable in a process used to developbudget estimates. To estimate such total costs, it isnecessary to make assumptions about the allocation ofsupport costs to instruction and about the allocation ofinstructional effort and resource use for various levelsof instruction. The more accurately such cost studiesportray institutional operation, the greater is thetemptation for central budgetmakers to try to control

11 ,;)

.99

operations. However, because most of these cost alloca-tion rules may not in fact be consistent with actual linesof organizational responsibility and authority, they aretherefore unsuitable for establishing the accountabilityof responsible administrators other than those at thehighest organizational levels.

WAGE AND SALARY DETAIL

Higher education operating budgets, like most of thebudgets reviewed for public sector activities; are devotedpredominately to expenditures for labor inputs, referred'to as personal services in object of expenditure terms.Therefore, a significant portion of the data provided withbudget requests may be related to what is virtually a line-item budget for every budgeted position in the organizationrequesting funds. The budget submission may include indi-vidual positions and position numbers, lacking only thename of the persons holding the positions. Provision ofthis kind of information, especially. if state governmentmaintains control of positions, is an indication of classicline-item review. Where position control does not exist,and where wage and salary detail is shown only for largeaggregates, the significance of added increments of infor-mation is appreciably reduced. Even in those cases(clearly the majority) where the institution or systemis not bound to the descriptibn and complement of posi-tions in the request, a clearer depiction of institutionaloperations.emerges from this description of the personnelcomplement.than from budget requests comprised of expendi-ture estimates based on formulas or programmatic considera-tions.

OUTPUT MEASURES

Even though formal program budgeting has not beensuccessfully implemented in its entirety in any statebudget process, various fragments of the methodology'have been adopted, and the associated information report-ing requirements have been incorporated. This is espe-cially so with respect to information on educational

100

114

outputs. Many of the output measures that have been de-vised are not very sophisticated and merely representpreviously reported data under new names. Data on enroll-menLJ, 2tudent credit-hour production, and degrees awardedfall anto this category of data that are justifiably quanti-::tative, yet suspect as true measures of educational produc-tion. Such data have been collected for some time as apart of HEGIS (Higher Education General Information Survey),and they constitute nd real additional reporting burden,except that reporting of information to be used in thebudget process must be more timely. Student contact hours,defined as.the clock hours per week actually spent inscheduled teaching, thesis supervision, and meeting withstudents in connectiqn with course content, have beenadopted in a few states as a new measure, although it isan only slightly modified version of an old instructionaloutput measure, the student credit hour. Collection ofstudent contact hour data requires additional informationreporting systems and procedures.

Hawaii, Pennsylvania, Washington, and Wisconsin haveattempted much more sophisticated approaches to measuringeducational outputs in conjunction with budgeting. Theoutput measures for Hawaii and Wisconsin were describedin Chapter 4. Output measures required in Pennsylvaniaare listed in Exhibit 9. As was the case in the Hawaiiinformation requiremants, great variation existed amongthe Pennsylvania institutions in their ability to providethese measures for FY 1974-1975. The budget request ofthe University of Washington for the same year containedthe most sophisticated discussion of education performanceand output measures that we observed in any of the 17states. In this Washington request, measures of extensive-ness, efficiency, effectiveness, and program benefits wereincluded along with a narrative discussion of a qualitythat might well have qualified it for acceptance in ascholarly professional journal. Whether this presentationwas effective with the legislature is another matter, andone we could not determine. It is noteworthy, in ourestimation, that this kind of detail and this level ofdiscussion is considered necessary to justify publicexpenditures.

101,

115

Exhibit 9

PROGRAM MEASURES REQUIRED IN PENNSYLVANIA.BUDGET, spEims.sIoNs, FISCAL YEAR 1974-1975

'Full-time equivalent studentsNumber of associate degrees conferredNumber of bachelor degrees conferredNumber of masters degrees conferredNumber of fir§t professional degreesNumber of'doctoral degrees_ conferredNumber of degree recipients working in their major field

one year after graduationNumber of medical school graduates interning in,

PennsylvaniaNumber of education graduates not seeking certification

from the Department of EducationNumber of registrations in adult/continuing education

,activities for degree credit

Number of registrations in adult/continuing educationactivities for noncredit

Number of registrations in adult/continuing educationactivities for conferences, institutes, andworkshops

Number of students receiving financial.assistande%Number of research projects currently supported by%

state fundsNumber of graduate students supported by state funds,

Number of full-time professional staff supportedby state funds

Number of new projects initiated with state fundsNegotiated overhead rate on federal contracts and

grantsPercent of state unds devoted to institutional

support

Source: Commonw alth of Pennsylvania, Budget Instructions

for Colleges a d Universities, FY 1974-1975, Appendix B.

102

BIENNIAL BUDGETS AND MULTIPLE-:YEAR PROJECTIONS

Five of the 17 states studied prepared biennialbudgets. At least one of these, Wisconsin, had a formalannual review process after one year, involving formaldocument.submission for updating or-revising.budgetaryrequirements for the second year of the biennium. Thebulk of data required for a biennial process is:approxi-mately twice as great as for an annual process, but thereprobably is considerably less work involved in producingthe information load. There are clearly returns toscale in data collection and submission. A biennial pro-cess with annual review requires not only biennial sub-mission of data, but in addition, only a slightly smallerdata submission for the second year. Although theinformation workload will not be significantly greaterthan for two annual submissions and will probably beless, if the opportunities for incorporating some mid-range planning decisions in the budget process aretaken advantage of, there Will, of dourse, be a con-siderable increase in the analyses required. Checkingand costing difficulties are probably not increased two-fold, but evaluation and bargaining issues are undoubtedlymultiplied within the longer budget time frame.

A biennial process does introduce desirable plan-.

ning considerations into budget preparation, preciselybecause of its longer time frame, which may not bepresent in an annual process. Budgetary estimates forthe second year will be considerably r. .a uncertain,however, and this fact will underm07. their eventualaccuracy. As a budgeting practice, biemial budgetingraises many other considerations than information load--principally, the uncertainty of expenditure plans in atime of transition and price inflation (see Schmidtlein& Glenny, in preparation, for a discussion of theseconsiderations).

The submission of multiyear irojections beyond thebudget period in question is a device associated withPPB, which was employed for higher education in five ofthe 17 states for FY 1974-1975. That year was the firstin which the procedure was used in Illinois, and the

1 1103

last in Florida. Typically, although multiyear pro-jectiong add considerable bulk to budget requests andappreciably increase the number of figures that must beestimated, neither the providers of the projections nor

.reviewers give them much credence. In the main, the pro-..jectiOns are arrived at by extrapolation or guesswork.The University of Nebraska, for example, provided itsadditional year of budget estimates by merely duplica-ting the figure for the preceding year. In general, no

,budget participant wants to be committed to figuresarrived at in this way, and thus any serious effort byinstitutions or state agencies to determine the direc-

-:....tion or magnitude of trends in expenditures is not likely

to be forthcoming because of the fear of locking theseexpectations into the budget process prematurely.

OTHER DIMENSIONS OF BUDGET DOCUMENT CONTENT

The seven characteristic information items includedin budgets, described above, cannot adequately describethe full spectrum of data found in budget submissionsbecause so much of this information relates to idiosyn-cratic features of each state's review procedures or its

..higher education governance structure. One such feature,which we have been unable to treat in any systematicway, is the submission of information on nonstate revenuesources. As noted earlier, most states request the,balance between their expected nonstate revenues andtheir expected total expenditures from the state. Tosupport estimates of this balance, institutions mustestimate the magnitude of nonstate support. A few statesrequire that revenue source and expenditure data be pro-vided by program. Hawaii provides a good example of aprocess in which institutions provide revenue sourcedata by expenditure program for the prior year, thecurrent year (estimated), and the budiet period.

The principal issue involved in submission of dataon nonstate sources of revenue is not entirely provisionof information, however. If the institutions retain con-trol over nonstate funds, provision of these estimates,which are frequently underestimations, will do little

104

118

toward promoting state control, although it may invite it.Some institutions have balked at providing this informa-tion to state agencies on grounds of constitutionalautonomy, for reporting information may be only a preludeto its use. On the other hand, there does appear to havebeen a trend, at least in the states involved in this study,to reduce the operating flexibility that institutionsderive from their multiple sources of funding--legislativeappropriation of more and more of their total funds, andexecutive budget office allotment procedures which off-set surpluses in nonstate funds or otherwise reco4p unex-pended balances. A substantial amount of information onnonstate revenue sources is contained in most budget sub-missions, but its quality and usefulness to budget exam-

.

iners are difficult to evaluate except in the contextof a particular buAget process.

Budget submissions also contain varying amounts ofnarrative, ranging from virtually none at all to extendeddiscussions of institutional or system role and scope,supporting arguments for specific items in the request,and assorted budgetary issues such as performance evalua-tion and educational effectiveness. In many instances,these narratives are essentially freeform, adjustingcontent and format to the subject at hand; elsewhere,they conform to elements prescribed in budget instruc-tions, and cover program objectives, target groups, andmeasures. Narratives also may attempt to explain specialfeatures of the context in which the request is beingmade by referring to trends or patterns that the quanti-tative portions of the submdssion do not show.

A subjective feature of budget submissions which isprobably very important for its effect on elected andtop administrative officials, who are not likely to bemuch concerned with detail, is how much one could learnabout a particular institution from its budget request.A budget examiner's concern with determining just whatan institution really needs and what is additional paddingwill eventually lead him to develop certain sensitivitiesthat help him review budgets. His focus will more likelybe on the most objective elements of the request. Butothers who have a say about what funds an institution

105

119

gets will want to 'have information that explains how the

alma mater is organized, what its real tasks are, and

how the athletic program is doing. Many will get this

information from personal contacts with institutionalofficials, communications from concerned individuals,

and the media. We did not explore these alternative

avenues of transmitting information'outside the formal

process. They are without doubt as important as therequest document itself, certainly even more important

for some Process participants. Yet in considering theinstitution's formal submissions alone, one is struckby the differences, from state to state, in the effortthat is put into making the budget request a persuasiveargument for support of higher education programs.

Schick (1971) has commented that a considerableamount of the narrative material he observed in statebudget documents was added to bolster agency buiget

claims and to supplement public information appearingin statistical tables. Ir. few lAstances, he said, were

.the narratives helpful tr, budget ;Irocess "insiiiders," nor

did they deal directly ct rigorously with questions of

agency.performance. Our impressims were largely thesame, except t!-7.At we often found the bulk of these nar-

-ratives apprec,bly-yreater than Schick suggested when

he described narratives as citing the agency's legalstatus or discussing somewhat what the agency is doing.

Many states are calling for elaborate narrativedescriptions el role and scope, mission and performance,but it apparently takes from two to three budget cycles

for the institutions to fully realize the opportunities

in these narratives for "selling" themselves. Under

formal PPB, these narratives were envisioned as a dis-

tillation of issues and implications that had been ex-plored in special studies and program memoranda carriedout considerably in advance of the submission of budgets.

Plans for carrying out these preplanning activities and

linking them with the budget process do not seem to have

materialized. As a consequence, calling for the inclusion

of these "items" in the budget documents results in staff

and others substituting such knowledge as they have and

relying on a sense of what will help "sell" a budget.

12 7)106

TEE DOCUMENTS 0? STATE BUDGET AGENCIES

In his study of the federal budget process, Wildavsky(1964) has suggested one perspective from which to viewthe budgetary process.

Throughout this volume we shall be concernedwith budgets as political things. Taken asa whole the tederal budget is a representa-tion in monetary terms of governmentalactivity. If politics is regarded in partas conflict over whose preferences shallpreVail in the determination of nationalpolicy; then the budget records the out-comes of this struggle. If one asks, "Whogets what the government has to give?' thenthe answers for ornament in tine are recordedin the budget. (p. 4)

--If we take this 1terally, then each request, recommenda-tion, or alternative is a proposal for budgetary outcomesand a tentative answer to the question, "Wbo gets whatthe government has to give?" The appropriation bill isthe final score. But there is an alternative perspective,which really is the impetus to many of the informational

_and-analytical reforms suggested.for-state-budgeting.This alternative view is that budgeting is much more thandetermining who' (in the narrow sense of which specificagencies, government employees, consultants, and con-tractors) gets what (namely, an authorization to encumberpublic accounts); budgeting is, to name a few examples, aprogram or a plan for reducing poverty, hunger, or disease,or for implanting technological and professional skills:A budget in this sense is a plan for doing something.

'As should be clear from the previous discussion of institu-tional and system requests, these documents now are muchmore specific about who will get what than they areabout concrete plans for action, although informationreforms have forced some attention to the latter.

Institutional and system requests provide informa-tion on institutional operations, contextual influences,and other supporting data. Subsequent review by state

107

121

agencies generates further information on priorities, con-

straints on revenue, and political expectations. State-.

level agencies communicate in many ways besides theexchange of formal documents, but this exchange is in many.instances the principal means through which staff analystsobtain techn4.cal information. Table 4 shows the formalbudget doCuments prepared by the various types of agencies

in the 17 states studied. These do not include staff issue

or 'policy papers intermediate to a final recommendation or...

approoriation. Of course, budget processes in all statesconclude with the passage of an appropriation act which is

the definitive formal budget document.

All of these documents are public information in the

broad sense, but there are considerable differences amongthem: Some are merely administrative forms tabulating acomprehensive set of expenditures with no supporting argu-ment; others describe the substance of agency programs and

give some sense of alternatives being considered and choices

being made. In the introduction of the first of The Brook-ings Institution's series of yearly volumes on setting

national priorities in budgetmaking, Schultze (1970) wrotethat the executive budget presented to Congress is designed

to persuade, not to infcirm. This describes many state

budget documents as well. Although as "outsiders" welacked sufficient timc, Incight, and information to judge_

the adequacy of these documents, many state staff person-nel indicated that their review also was hindered by a

lack gf .Lnformation.

With very few exceptions, higher education budgetsare actually reviewed thoroughly by only one state agency

or in depth jointly by two agencies. The task falls to

the other agency staff or staffs to check the resultsboth technically and for policy implications by testing

and probing under severe constraints of time, expertise,

and information. The "norm" of state.agency interaction,therefore, is for institutions to submit requests to acoordinating agency (or a statewide governing board) wherethey are examined for internal consistencies, adherenceto program approval procedures, and general congruence

with a state plan. Coordinating agency recommendations,along with institutional reqUests, then go to the execu-

12 Th

108

Table 4

STATE AGENCIES THAT PREPARE FORMAL BUDGET DOCUMENTS

States

Statewidegoverningboard

Coordi-natingagency

Execu-tivebudgetoffice Legislative budget staff

California Na1

No Yes Yes, Legislative Analyst'sanalysis of the budgetbill contains recommend-ations

Colorado Na Yes Yes No

Connecticut Na Yes Yes No

Florida Yes No Yes No

Hawaii' Yes No Yes No

Illinois Na Yes Yes No

Kansas No Na Yes (Yes), no staff for FY 1974-75.

Michigan Na No Yes

Mississippi Yes Na No

Nebraska Na No YesNeu York Na No YesPennsylvania Na

2Yes Yes

Na Yes Yes.TennesseeTexas Na No Yes

Virginia Na No YesWashington Na No YesWisconsin Yes Na Yes

Yes,

Yes,

YesNoNoNoYes,

No

No

Yes

Legislative researchdepartment has since pre-pared staff reports, butnot recommendations.

House & Senate fiscala4e0Cies jointly preparesumriary of net stateappropriations.

Commission of Budgeting& Accounting reCommendsin annual budget

Legislative budgetboard's budget estimatescontain rdbommendations.

1Na Not applicable

2Department of Education makes recommendations to the governor throughan internal.memorandum.

123109

tive budget office where they are reviewed in technicaldetail and adjusted to meet the priorities of the gover-nor's program. As a part of the governor's budget,executive recommendations are checked by the legislature,modified to meet legislative program priorities, andpassed in an appropriation bill which the governor even-tually signs. If the budgetary role of multicampus'systems in California and New York is considered equiva-lent to this one, nine of the 17 states displayed such apattern of interagency relationships (California,Connecticut, Florida, Hawaii, Nebraska, New York, Virginia,Washington, and Wisconsin). By allowing only for a shiftin technical budget review from the executive budgetoffice to the coordinating agency, three more states can'..)e regarded as exhibiting this budgetary pattern (Illinois,Kansas, and Tennessee) for a total of 12 out of 17. Theremaining states, Colorado, Michigan, Mississippi,Pennsylvania, and Texas, all differ from the 12 in somerespect, principally in an expanded legislative role asa coequal of the executive in constructing a comprehen-sive state budget. (For a discussion of the implicationsof the latter arrangements', and a more complete discus-sion of the "norm," see Schmidtlein and Glenny, in pre-paration'.)

It would be presumptuous to recommend the "norm" asa process that all states ought to adopt, but this processdoes illustrate one way in which the budget review rolesof various interests can be coordinated. What amountsin estence to joint executive-legislative budget develop-ment and review in Texas and Mississippi appears also tobe a stable pattern of shared responsibility. Informationneeds are obviously not the sole source of conflict, butthey can become the focal point of tension and interagencydisagreement because information requirements are greaterfor an agency which develops its.own recommendations thanfor an agency which simply modifies those of another agency.

When there is fundamental disagreement over the pre-mises on which review is conducted, all agencies may thenfind it necessary to develop their own information basesand ignore the technical review of other agencies. The

110

12 1

review process can evolve far beyond a state: of beneficialredundancy to a state of "mass confusion." Such funda-mental disagreement was evidenced most strongly in Michiganand Pennsylvania, two states in which program budget reformshave been initiated by the executive. In Pennsylvania, wefound little evidence of a technical or analytic review bylegislative staff, and therefore the intonation needs of

. the legislature were met satisfactorily with crosswalksfrom a program classification to an organizational struc-ture. In Michigan, however, duplicative informationrequests were made of the institutions to support adetailed review by the Department of Management and Budgetin terms of program categories and a similarly detailedreview by the legislative fiscal staffs in terms oforganizational units. The gravity of the conflicting --review procedures in Michigan is indicated by excerptsfrom a September 18, 1974 memorandum by the state budgetdirector to legislative leaders:

It would be useful to iterate more pre-cisely the legal framework for the budgetprocess and the specific disruptions whichhave been caused by legislative demands uponthe agencies for information. . . .

Taken in the aggregate, the statutes estab-lish in rather clear terms the budget systemas a whole--not an executive budget systembut the entire system. All powers, duties,and functions related to planning and pre-paration of the proposed budget are vestedin the State Budget Director. This includesall information he requires, as-well as thatdeemed wise and of material value to theLegislature in its consideration of thefinancial needs of the state. This broad,comprehensive authority is backed up bysubpoena power. By being so clear, itprecludes legislative initiatives whichinterfere or contradict or complement thesystem. In short, there is no specific,stated legal basis for a legislative budgetsystem. In fact, there can be only one

111

125

state budget system, according to statute,and that is determined by the State BudgetDirector, an appointee of the Governor.

Requests for information to prepare budgetpositions which are not consistent witheither those established by the StateBudget Director as required to prepare andpresent the budget or contrary to estab-lished reporting systems on fiscal manage-ment result in a time disruption thatdeters agencies from meeting executivetimetables.

In short, the spontaneous requests forinformation which periodically emerge fromthe fiscal agencies confuse the authorityvested in the State Budget Director by theConstitution and state statutes.

Other manifestations of a less serious lack of coor-dination in the use of budget information were presentin Colorado, where the Commission on Higher Educationmade budget recommendations in terms of student-facultyratios which included both filled and unfilled facultypositions, whereas the staff of the Joint LegislativeBudget Committee wanted to make its recommendations onthe basis of information about more specific operatingconditions. In Connecticut, differences in requiredinformation bases between the-Department of Finance andControl, which reviewed budgets in terms of detailedobjects of expenditure, and the Commission for HigherEducation, which reviewed budgets in terms of aggregatedfunctions and programs, were met by a dual institutionalbudget submission.

The remainder of this chapter is devoted to a sur-vey of the documents of state higher education agencies,executive budget offices, and legislative budget staffs.The contents of the documents will be discussed withrespect to the amount and kinds Of analysis and support-ing arguments found in them. The approaches to review.

taken by the various agencies are described more completelyin Glenny et al. (1975). In this report we are trying togive the flavor of these documents as information sourcesand as indicators of the substance and style of review.In some instances it is necessary to qualify the impressionof comprehensive rationality given by the documents becausebudgetary totals must usually be reached through recourse.to political bargaining and attention to revenue limita-tions. Subaggregates and details may then be "backed out"after the'fact. Although it may be that these "rational-izations" have little direct impact on the substance of

_budgetary_outcomes,.the _fact.that_they_are_recorded_at_all,,,even as "reconstructed logic," suggests some appeal for

--support-on the basis of reason and analysis. These docu-ments both provide an agenda for discussion in_subsequentreview, to some degree, and also focus reviews on a smallerand a more manageable number of issues. .

COORDINATING AGENCY RECOMMENDATIONS

State higher education agencies, other than statewidegoverning board staffs which have statutory budget author-ity to make recommendations to the executive branch or thelegislature, usually do so in a formal document. Con-solidated governing boards frequently develop rather thanreview the higher education budget request submitted toexecutive and legislative budget agencies, and a discus-sion of these formal requests was included in the firstpart of this chapter. The Kansas Board of Regents andMississippi Board of Trustees are exceptions to this role'of consolidated governing boards among the five consoli-dated governing board states included in the 17-statesample. The Kansas Board of Regents' staff thoroughlyreviews campus requests to the governor and legislature,but the results of this review are reflected in revisedinstitutional requests to the legislature and not indocuments of the Regents' staff. The revised campusbudgets subsequently are published by the Regents in asingle document which becomes the Regents' Budget Request.The Mississippi Board of Trustees' staff checks insti-tutional requests submitted in object of,expenditure.format, but'prepares the formula-based request.

113

12'7

In this section we shall describe the formal recom-mendations made by five coordinating agendies, in Colorado,Connecticut, Illinois, Pennsylvania, and Tennessee. State-

wide governing board submissions'were included in the firstsection of this chapter, which dealt with institutionaland system requests. It should 'not be inferred that theseformal recommendations constitute the only way in which astate higher education agency can influence the budget pro-

cess.

Coordinating agency recommendations, in general,condense institutional requests to one summary volume,but they do not necessarily supplant the institutionalrequests which are also available for subsequent state

agency review. The recommendations eliminate considerabledetail and tend to use formats not heavily influencedeither by PPB-procedures or hidebound object of expendi-ture details. With the exception of Illinois, where theResource Allocation and Management Program (RAMP) pro-cedures were being implemented, coordinating agencies werenot emphasizing a programmatic format, nor was there ex-tensive use of narrative or clarifying argument. TheProgram Classification Structure was used in some recommend-ations, but the principal focus was on functional categoriesof expenditure. Coordinating agency recommendations wereaggregated at the same level as organizational requestsWere reviewed in all states but Connecticut; there therecommendation on the state colleges was consolidated,although individual campus requests were reviewed. In

Colorado,;requests for certain support functions (e.g.,student financial aid and automated data processing) wereconsolidated and an appropriation made to the Commissionon Higher Education for allocation among the institutions.

The responsibility to formally recommend requiresthat the coordinating agency recommendation be compre-hensive and sensitive to political realities. Conse-

, quently, these formal recomnendations do not reflectthe substantial amount of analysis that often goes into

them. They are, by and large, influential for the"answers" they give, not for the supporting justifica-tion and argument they provide. In Tennessee especially,

but also in Illinois, these recommendations are derived

114

from the most thorough analytical review conducted by anybudget agency in the state. Historically, the recommenda-tions of these agencies have been quite influential. Ingeneral, lack of analytical substance may not necessarilydetract from recommendations in the view of executive andlegislative budget agencies, but without such analysis,there is no factual leavening for the.political processand little help in makin4 the recoMmendations credible.Only one of these formal recommendations was cast explicitlyin the context of longstanding policy and planning considera-tions. .As we have indicated, these formal recommendationsare frequently important in framing the agenda for subse-quent review; none of them, however, were structured tofocus this review on sharply defined or specific issues.

FIVE COORDINATING AGENCIES

In the five cases where the coordinating agency Makesa formal recommendation, the agency either plays the domi-nant role in specifying the format and information supportto be included in institutional or system budget requests,or it serves as a conduit in expressing the informationneeds of other state fiscal agencies. (The process isonly just beginning in Pennsylvania, but it seems likelyto reach fruition for the state-owned institutions.) Eachof-the five coOrdinating agencies to be discussed had de-signed a review methodology'6f varying degrees of sophisti-cation and complexity together with the supportinginformation requiremdnts. In Colorado, this was exemplifiedby the workload and productivity data desired by the staff__of the Joint Budsjet Committee; in Connecticut, data was pro-vided by program and by object class according to "A," "B,"and "C" priorities; in Illinois, the Resource Allocationand Management Program (RAMP) procedures, which are a PPBstyle information format, dictated the Torm of requests;in Pennsylvania, calculation of the Department of Education'sformula factors were required; and in Tennessee the instruc-tional unit cost formula structured the request.

Review by these coordinating agencies, and subsequentreview by executive and legislative budget staffs was not

115:

129

limited to these devices or techniques, except in thecase of the Tennessee formula, but subsequent review wasinfluenced, or at least structured, by their presence.Only in Pennsylvania, where the use of technical.budgetprocedures at the state level were relatively new, is itlikely that these devices had a very limited impact, andthis only with reference to the state-related and state-aided institutions. The review process for the state-

s,

oWned institutions was structured by the Department ofEducation's workload indibators. Until recently, region-alism and partisanship in the legislature virtuallyswamped any analytical review of higher education budgets.

- _ The short summaries which follow describe the formalrecommendations of these coordinating agencies. Pennsyl-

vania is omitted because the recommendation process wasnot fully established at the time of our site visits, andconsequently we had no recommendation document to examine.

Colorado. The formal recommendations of the ColoradoCommission on Higher Education showed actual, estimated,requested, and recommended budget-year expenditures forall campuses by functional classifications. The narrativecontained-general supporting analysis and justification,with little eXplicit justification for item-by-item recom-mendations among campuses. For example, in the instructionalarea the basis for recommendations was summarized as followsin the 1974-1975 Operating Budget Recommendations of theColorado Commission on Higher Education:

For 1974-75, recommendations for numbers offaculty and related academic administrativestaff were determined from a combination of(1) the institutions' requests includingbudget narrative and budget hearing pre-sentations; (2) consideration of the insti-tutions' role, mission, and program withinthe Colorado postsecondary education system;(3) analysis of student/faculty ratios basedupon guidelines for each of the three groupsof institutions described above (large,small, medium-sized); and (4) calculation

13

of need based upon actual historical stu7,dent/faculty ratios,_particularly that for i

1973-74, related t6 full-time equivalent \

students expectedj4 1974775. (pp. 32-33)

Connecticut. The Commission on Hi§her"EdUC-ation (CBE)makes budgetary recommendations for the constituent units(regional community colleges, state technical colleges,state colleges! University of Connecticut, and the Uni-versity Health Center) of the state's postsecondaryeducational institutions. jormally, the recommendationsof the commission are supposed to be reviewed as the con-stituent units request. In fact, however, they were not,because the informational detail of the CHE recommenaa-tions did not meet the needs of the budget review approachof the Department of,Finance and Control. Nonetheless,CHE recommendations did provide a rationale for budgetaryrequirement estimates at a fairly aggregated level. In .

addition, they provided a priority indication for fundsrequested above.the level of expenditures for the previousyear. For the state colleges, they included a recom-mendation on'the allocation among institutions of thebudgetary augmentation recommended by the commission.The CHE recommendations contained a moderate amoiunt ofnarrative explanation and justification.

Illinois. The Board of Higher Education, throughthe Director's Report on the Operating Budgets for insti-tutions and activities, makes annual budgetary recommenda-tions to the Governor and the General Assembly. Theserecommendations combined the level of disaggregationfound in the _Colorado commission's recommendations witha slightly more extensive level of narrative and explana-tion than appeared in the Connecticut commission'srecommendations. The Board of Higher Education's rec-ommendations were completely rationalized within theDirector's Report: One can start with the prior year'sbase, follow the recommended adjustments, add on salaryand price incraases at common rates, and determinedeficiency adjustments and new program and/or enrollmentsupport. All items were summarized, but at the campuslevel one has a picture of expected expenditures interms of a total base and adjustments to that base. It

131117

is noteworthy that very little of the RAMP documentationwas used.in the Director's Report in 1974, and that func-tional or programmatic categories were eschewed in therecommendations. Issues in the sense of new academicprograms and price and salary increases were discussedrather broadly, but there was no discussion of otherexpenditure categories, even in the increment over theprior year's base.

Tennessee. The Higher Education Commission (THEC)submits an aggregated budget request based on the insti-tutional requests to the Division of the Budget. Becausemost of the requeit was generated using formulas deter-_mined from the biennial cost study administered by THEC,little analysis and justification was required with theirformal recommendation.

GOVERNORS' BUDGETS AND EXECUTIVE-LEGISLATIVEINFORMATION-SHARING

The governor's budget is the formal communicationof a comprehensive state expenditure plan by the governorto the legislature. Many observers have suggested thatanalytical budget reforms and accompanying proceduralchange (what Howard, 1973, appropriately called "ratio-nalistic budgeting") offer an opportunity for the governorto take greater initiative in policy leadership. One

_might expect, therefore, that executive documents wouldmore commonly reflect this approach, and highlight thesubstance of programs rather than operational detail.

There are two ways in which governors and executiver-budget-o-f-flces-have tried to increase policy leadershipthrough involvement.in earlier stages of the-budget pro-cess. The first is exemplified in Michigan and Pennsyl-vania by the governor's issuance of Program PolicyGuidelines prior to budget submission. These guidelinesidentify the governor's policy and program prioritiesand are intended to provide a vehicle for coordinating'executive and agency planning in the final stages of

budget preparation. In practice, they do not go fartoward achieving that end because of their broad non-

118

...

specific nature. The second way, found in virtually allother states, is through the specification of new informa-tion categories and budget structures which were discussedin the first section of this chapter. These changes aremuch more pervasive, but as we shall discuss in Chapter 6,the focus of review has not yet shifted substantially.

The executive budgets themselves, with exceptionsnoted below, are generally much less informative thancoordinating agency recommendations, both-in-their-de-scription of institutional activities and programs andin their narrative justification. Further, they condense

sexpenditure estimate detail, at the same time mainaining_. .

the consolidated or aggregated distinctionS present inthe requests or recommendations they reCeive. Though itis true that the state budget is the most importantpolicy document in state government (Dye, 1973), andtherefore that the governor's budget is the most importantstatement of executive policy intentions, most guberna-torial budgets state policy only implicitly, and givevery little indication that they are anything but, inWildavsky's words (1964, p. v), "the province of stodgyclerks and dull statisticians."

In Connecticut, Michigan, and Pennsylvania attemptswere underway to make the executive budgets more explicitstatements of policy through the use of programmaticinformation, although at a highlyAggregated level.Pennsylvania had the most elaborate example of a dualbudget presentation in which crosswalk data from organiza-tional,units to 15 programmatic categories were used tocut across instructional programs and institutionalsuppori. activities. These attempts at changing the sub-stance of review did not appear to have penetrated verydeeply into the underlying budget processes. In Connecticuthis process was still dominated by a traditional objectof expenditure review, and in Pennsylvania and Michiganthe,dominance of political forces in higher educationresource acquisition considerably lessened the significanceof technical or analytical review.

Executive budgets in California and New York, whichwere not-referred to. by budget analysts or state officials

119

133

as "program budgets," offered extensive narrative andsupport justification. In New York, legislative staffdescribed this material as providing useful information-fOr their budget review and policy planning. Both execu-tive budgets, especially in New York, focused attentionon particular budgetary issues, whereas the more tradi-tional governor's budget, by comprehensively outliningall expenditure categories, gives little indicationthat some recommendations are more important than others.In some respects, the much criticized routines of programbudgeting 'do a better job than traditional formats ofcondensing the enervating array of budgetary figures.But the condensation itself constitutes, of course, onesource of executive-legislative conflict--over informationand the format of the budget. The failure or inabilityto implement new procedures is hardly due to ignorancein the executive branch or to lack of leadership. Strategicand political constraints clearly limit the changes thatare possible.

Perhaps the most unique governor's budget is that inWisconsin, which maintains the integrity of the detailedbudget categories, the Decision Items Narratives (DINs),of the original request. Major Decision Items groupedas costs to continue, workload increases, and new andchanged services were itemized with justifications and thegovernor's recommendation on each item. The explicitnessof these executive recommendations is virtually uniquein the budget processes we examined. Other budget pro-cesses maintain the identity of original budget categoriesin executive-legislative review, but rarely for expendi-ture categories other than large aggregates. The onlyother comparable public statement of budget recommenda-tions of which we are aware is that of the CaliforniaLegislative Analyst discussed below.

Executive-legislative information-sharing takes placeinformally as well as through formal documents, but itshould not be assumed thet there is always a free flow ofcommunication between staff analysts. Executive budget ,

staff may provide data, but not the analyses and the budg-etary implications derived from them. In other situations,executive and legislative staff may hP shared, or they may

I Q A

JL120

work together very closely, as they do' in Kansas andVirginia. Where the governor does not have the dominantrole in preparing the state budget, as in Mississippiand Texas, information-sharing often ceases to be an issue.

The California'Department of Finance and the Legisla-tive Analyst are a notable example of executive-legislativeinformation-sharing. In order for the analyst's Analysisof the Budget Bill to be timely (it follows submission ofthe governor's budget to the legislature by less than amonth), higher education requests, their analysis, andthe governor's recommendations must go to the LegislativeAnalyst's staff substantially in advance of printing.

LEGISLATIVE BUDGET STAFF RECOMMENDATIONS

State executive-legiSlative relationships are suf-ficiently different from presidential-congressionalinteractions that few parallels can be drawn, especiallyin consideration of the number of variations in executiveand legislative styles that exist in the states. None-theless, there are trends in the states similar to thoseapparent at the-federal level toward obtaining more budg-etary information from executive sources, formalizinglegislative budget proposals, and developing more staffand analytical resources. The importance of the Governor'sBudget as a unified statement of executive priorities,and its domination over alternatives considered in thebudget process, cannot be overemphasized. However, thepreparation of formal legislative documents indicative oflegislative priorities (and containing or reflectinganalysis by legislative staff) is becoming increasinglysignificant in broadening the range of alternatives con-sidered.

Differences in executive and legislative perspectiveson state higher educational policy, and differences inpolitical influence are discussed in Glenny (1976) andSchmidtkAn and Glenny (in preparation). As is apparentfrom these discussions, there is increasing debate overthe necessity for legislatures to increase the size oftheir staff bureaucracies and growing concern-over the

121

135

implications of the increases. The polarity that can

develop between staff Over information because of diverse.executive and legislative policy perspectives was nowheremore apparent than in Michigan. Yet other states in which

.legislative staff pursue equally aggressive analyticalroles appear to have worked'out acceptable solutions toinformation reporting and the division of responsibilityin budget review.

To summarize: We have restricted our attention to

formal documents. Their implications as to the level of

.legislative staff activity must be qualified by notingthat legislative staff may make budget recommendations tobudget-committee members-internally without-the-use,of- v.

formal documents. In.addition to those staffs describedbelow, such recommendations are made by legislative budgetstaff in Colorado (Joint Budget Committee), Florida (HouseWays and Means Committee), and New York (Senate and HouseAppropriations Committees).

In Mississippi and Texas, the agencies we have labeledas legislative agencies are staff to a commission or board

which includes either the gow-nor (Mississippi) or thelieutenant governor (Texas). Consequently, such agencystaffs are not Strictly comparable to staffs of joint orseparate legislative budget committees; the formal docu-ments these agencies prepare are comparable to many execu-tive budgets in comprehnsiveness and format. The commis-

sion budget was the only coutprehensive state budget inMississippi and there was, in effect, a legislative budget

in Texas. The higher education portion of the BudgetReport of the Mississippi Commission of Budget andAccounting was consolidated; even though the requests itreceived from .Ehe stz ,c1, le governing board were aggregated,-but this consolidated tormat feeds directly into the format

used in the appropriation bill. The Texas LegislativeBudget Board's. Budget Estimates were aggregated and com-paratively detailed at the institutional level (approxi-mately 15-20 lines per institution). Similarly, thesebudgetary items furnished the categories of the appro-

priation act. Neither of these documents includednarrative expianation or other justification.

13 3122

Although the Nebraska budget process itself is un-like that in Mississippi or Texas, the Nebraska Legisla-tive Fiscal Analyst and staff submit a similar formalrecommendation on postsecondary programs to the legisla-ture. These recommendations were in tabular form andincluded two years of historical data on appropriations,current year estimates, the agency requests, and theanalyst's recommendations. A separate volume (which wedid not review) provided narrative and an explanation ofthe underlying assumptions and workloads on which theailalyst's recommendation was based.

What were probably the best examples of legislativeanalytical staff documents were found in California,Kansas, Michigan, and Wisconsin".

In California, the analysis of the budget bill isprepared annually by the Legislative Analyst's staff,which serves the Joint Budget Committee but providesthis report to all members of the legislature. Theanalysis ran in excess of a thousand pages, selectivelycovering the entire range of the Governor's Budget(itself a five-volume document) with recommendations andexplanation. Postsecondary education programs were treatedas one of the total range of state services. The unitsfor recommendation were the multicampus systems (segments),but recommendations applied to specific elements and activ-ities of the systems and campuses. The analysis wasreleased immediately upon submission of the Governor'sBudget to the legislature, which made clear that theDepartment of Finance had cooperated in making availabletheir information and analysis and the thrust of execu-tive recommendations. The Legislative Analyst's reportis used as the agenda for legislative budget hearings.

The Kansas legislature had no staff on higher educa-tion for FY 1974-1975, but subsequently the LegislativeResearch Department provided staff reports to the Houseand Senate Ways and Means Committees. They are noteworthyas an illustration of what new staff can provide. Thestaff report on higher education consisted of a briefdiscussion of approximately a dozen general issues per-taining to all campuses, and specific issues pertaining

123

137

to the budgets for each of the six institutions. Theformat.of the reports focused attention on the budgetaryimplications of these issues by providing background in-formation and institUtional comparisons but made no recom-mendation. The governor's recommendation was provided asa reference point.

Until 1971, the Michigan legislature was staffed forbudget review by the Legislative Fiscal Agency. In 1971,a separate House Fiscal Agency was established and thename of the Legislative Fiscal Agency_waachanged to the__Senate Fiscal Agency. These legislative budget staffsprepare specific recommendations for fiscal committeemembers, but these are not public documents. The onlyformal public document prepared by the staffs is theSummary of Net State Appropriations to State SupportedInstitutions of Higher Education. This is a so-called"tracking surnamwel of state funds to colleges and depart-ments within each institution (Arts and Sciences, Fineand Applied Arts, Education, Nursing, Business, etc.)and to seven other support and public service functions(activities) which the Michigan public institutions havein common. The tracking summary is used to describe thepurposes for which legislative Appropriations are made,and it therefore supports the appropriation bill whichthe legislature passes. Its use in_ a budget process wasuniquein the 17 states we exam4ned, and it evidentlyhad resulted from the juxtaposition of information de-ficiencies attributable to a high degree of institutionalautonomy coupled with executive and legislative disagree-ment over the use of program rather than organizationalunit data. The executive budget office has introduceda number of program budget reforms into the.process andprogram categories were used in the governor's budget.However, the legislature has continued to want data onspecific organizational units, and to make its own re-'quests for this data from the institutions. It is notentirely clear whether the data actually being providedto state agencies in Michigan was appreciably more de-tailed than in other states, but it was certainly morevoluminous and difficult to reconcile because of thelack of coordination between the executive arid the legisla-tive budget staffs.

The staff of Wisconsin's Legislative Fiscal Bureauprepared issues papers for the members of the JointCommittee. on Finance, but theli did-not appear to be pub-lic documents in the sense of most legislative staffdocuments. In addition, a summary comparing major budg-etary Provisions of the Governor's Assembly and theConference Committee Recommendations was prepared at theclose of the session. Language from the Budget Bills wasincluded, but this was not an analytical document.

Formal legislative budget documents neither suggestthe full measure of legislative budget staff activity norits influence on the budget, but they give an indicationof the extent to which legislative review is assisted byinformation reporting and analysis. Where legislativebudget staff are involved in.preparing a co ehensivestate budgetary plan, their analytical activl,ies willbe somewhat broader, but in most other inst es legis-lative staff can be very effective in filter g theavailable mass of information and focusing l islativeattention on the few key items for which dub antivechoices can be made during each budgetary cy I e. Whereno one else performs this function, elected o ficials,especially legislators, are likely to be ove helmedwith technicalities and therefore have no alt rnativebut to make budgetary choices entirely on the basis of,"Who gets what" rather than, "What is this b eau orinstitution doing?"

These formal documents and the staff act itifs thatsupport them have another consequence which i equentlyoverlooked. When asked how current fiscal con Monsin his state had effected the budget process f r hisinstitution, one institutional vice-president: ommentedthat the effect on the process had been slight and wenton to emphasize that the form of the appropria ion billhad not changed. Obviously, the culmination Of thebudget process in a budget bill is the act that poststhe "final score." But this overlooks the manifer inwhich staff documents, particularly those whicg havea proximity to the appropriation bill, can be ind arebeing used to control or influence execution ort-rig-BUdget.

125

139

f.17 f79'177 F711.1

The "stump" approach to budgeting, whereby the legis-lature puts an appropriation on a stump and the institutions

-takelit and run, has been largely abandoned, if not throughlanguage of legislative intent inserted in the appropriationbill, then by the fact that .institutions must come back tothe legislature in subsequent years.to be confronted withthe formal documentation and analysis of-budget processesfrom prior years.

126

6.Payoffs from Analysis and

Information Systems

Evaluating the consequences of improved informationand analysis in the budgetary process is especially dif-ficult because there has as yet been no comprehensive .

implementation of information systems and sophisticatedforms of analysis. A questionnaire survey by Barak (1975)indicated that, at most, a handful of states have de-veloped management.information systems which meet a setof formal criteria for total development and implementa-tion. Our impressions from site visits in the 17 statesagree quite generally with the survey responses reportedby these same states for the Barak study. Furthermore,for implementation to take place at the state level,considerable development work must already have been doneat the institutions. Yet, in responses to the survey byBogard (1972), only 24 percent of the institutions re-ported having a full-time office of institutional research;13 percent had computerized management information systems;and 31 percent used some form of program budgeting. Only2.8 percent had all three.

These surveys of implementation,admittedly reflectonly superficial evidence of implementation and use ofsystems because of the difficulties in defining exactlywhat these systems are. In comparing the characteristicsof information and management analysis systems in Floridaand Wisconsin, two states widely regarded as being rela-tively advanced, one finds considerably different systems.

.Furthermore, in emphasizing the formal procedural featuresof these systems, one loses touch with qualitative andcontextual factors which are ultimately more important

127

141

than techniques. A state with all the formal routinesand documents of PPB may actually be reviewing budgetswith very little attention to the PPB presentation.Another state may conduct traditional line-item reviewwith periodic attention to agency performance studies.In which of these states can one expect payoffs fromnew information and analysis?

Paradoxically, in spite of the inability of budgetreformers to fully transform the budget process, a great

"many-states, if not all, have altered higher educationbudget formats and information reporting procedures inthe last five years. Fenno (1966), Merewitz and Sosnick(1971), Wildavsky (1964, 1966, 1969, 1972), and numerousother critics of PPB have explained.why the PPB processis unworkable and cannot be implemented comprehensively.But these observers have given relatively little attention\to the significance of the many incremental changes thathave been occurring in the budgetary process. Like Marxistexplanations of fatal flaws in capitalism, these critics

-overestimated their capacity to deal PPB a death blow bytheir criticisms, and instead have brought about itsgradual reform.

In his discussion of the payoffs from urban informa-tion systems, Downs (1967) argued that the provision ofurban services is finally the true measure of benefitsfrom these systems. Many protagonists of the use ofinformation and analysis in manemept have emphasizedthat improved efficiency will be the outcome. Collegeand university budgeters maintain that unless the useof new budgetary procedures results in significantlydifferent levels of state appropriations, they are notworth the.effort. Although we tend to agree with Downs,the impossibility of evaluating payoffs of informationand analysis, in terms of their influence on highereducation services, was implied by Corbally (1974), thepresident of a major public university:

Regardless of the data that our machinesproduce, regardless of the reports and'information that underlie our administra-tive decisions, the teaching-learning pro-cess remains basically constant. (1). 3)

142

128

He further observed that educational effectiveness mayremain at its peak until an institution's final day,even though ineffective management has led to bankruptcyand closure. The separation between teaching, the primarytask of colleges and universities, and administration isso great that any relationship between educational outputsand administrative procedures is very tenuous. Thus,we see no way to evaluate the contribution of informationand analysis by the final educational product. Is theperformance of a university president, for example,measured by_the. knowledge-of-graduates-of-his-institution4-or by the starting salaries of graduates on their firstjob?

Given that public budgeting in the 1920s was con-ceived as a device for reducidg governmental expenditures,for clarifying governmental responsibilities, and forsecuring significant governmental reforms, we should looknow to similar concerns as the underlying sources of inter-est in budgetary reform in higher education. Higher educa-tion services and objectives, as well as its efficiency,are involved, but in the fundamental way described byBailey (1973) and not simply as a problem of resourcemanagement:

Only the woefully naive contend that thereal problem is efficiency--that govern-ment bureaus and universities will receivevotes of confidence when they can masterPPH (program planning budgeting) and relatedcost-benefit techniques and thereby be heldaccountable. The basic issue is politicaland psychological--a growing belief thatwhat government bureaus and universities dois not worth the cost: that governmentsreduce freedom too much, and that universi-ties foster too much license. The absenceof sophisticated systems of accountabilitysimply adds to the already substantialfrustration of politicians and publics.(p. 133)

129

4 43

If analysis and information systems are to producechanges in higher education performance, institutionaladministrators and faculty must sense a performance gap.between what institutional outputs are and what they mightbe, rather than maintaining with Corbally (1974) thatthe teaching-learAing process is constant and independentof administrative factors. Although state executives andlegislatures show dissatisfaction with institutional per-formance by mandating faculty teaching loads and by increas-ing amounts of control and expressions of intent in appro-

-priation-bills, they-do not-know-exactly what they-wantfrom higher education, and much less exactly what they canget. Our interpretation is that, as a substitute for con-trols on activities and operations of institutions, legisla-..tures and executives are turning to budgetary procedureswhich show that responsible college and University adminis-trators are at least procedurally accountable for doingthe "right" thing. Centralized information reportingsystems attempt toAncrease the floW of information at allorganizational levels so that administrative decisions canbe checked against relevant data or checked procedurallyto ensure that supporting data was available to thosemaking decisions.

Schick (1971) treated the general consequences andproblems of implementing information and analysis systemsin budgeting in his discussion of PPB reforms in thestates. Silverman and Gatti (1975) reported on someteChnical problems associated with the Pennsylvania PPBexperience. The innovations assoCiated with PPB havebeen thoroughly critiqued by Wildavsky in a number ofpublications, as well as by others noted previously. In

general, critics of information and analysis reforms haveemphasized the procedural defects, and advocates havestressed the technical advantages. Both the proceduralbenefits and tech dcal problems of increasing the avail-ability of information and analysis in the budget processappear to be the most often overlooked consequences. Inthe discussion whi.h follows, we therefore restrict ourattention to thE, procedural benefits and associatedtechnical probl Is. A discussion of the power and in-fluence imp; :4..aLlons of information and analysis trendsdescribed in earlier chapters concludes the chapter.

1.130

PROCEDURAL BENEFITS OF CENTRALIZEDINFORMATION AND ANALYSIS

It seems almost axiomatic that better informationand analysis lead to better decisions, and hence, toimproved management and planning. Unfortunately forpurposes of proving or disproving the point, we are notdealing with alternative processes of production, onewith good information and one with poor information, forwhich different decisions yield correspondingly differentoutcomes. Moreover, in public sector management thecriterion ultimately applied to define good or poor out-comes is the political one: Do constituents and interestgroups approve?

Cheit (1975) indicated his tentative belief thatinstitutional planning and management have been improvedas a consequence of implementing management systems, butnoted that this is only an impression and,that littleevidence could be mustered to prove it. No evaluationstudies in the interim have provided that evidence, norare they likely to resolve that question to everyone'ssatisfaction.

The formal budget process reported on here, emphasiz-ing staff roles, is an adjunct to the political process;hence,'the ultimate standard against which payoffs fromchanges in the budget process must be evaluated arepolitical in nature. As noted earlier, much of thecriticism of recent budgetary reform has centered on thefailure of these reforms to satisfactorily meet politicalcriteria. Therefore, we begin with a discussion of theprocess benefits we believe are associated with the useof centralized information and analysis systems.

EQUITY OF BUDGETARY OUTCOMES WITHIN HIGHER EDUCATION

Free and open exchange of information and analysisfosters equity in the determination of budgetary outcomes.Although fair and impartial budgeting procedures can go along way toward satisfying budget participants that theallocations they eventually receive are equitable, analysis

131

145

of those appropriations in relation to objective factorsis an important ingredient in influencing perceptions ofequity."

Rivlin (1971) explained the weakness of analyticalmethods in determining interprogram.allocations as relatedto their inability to compare diverse program benefits,but notes the prospect of demonstrating through analysiswhat implicit weights are attached to alternative programsby the political process. A convincing case has been made.by Steiner (1974) for a sequential decision model in which_decisions about the total level of expenditure and relative-. --size of major federal programs are made sequentially ratherthan simultaneously. Accordingly, the total budget isfixed, and then major program segments are backed out.Interprogram choices, therefore, are constrained by priordecisions, although they may be more flexible. WhileSteiner's sequential decision model was intended to de-scribe the federal budget process, it agrees completelywith our impressions of state-level budgetary practices.Thus ex ante analyses of budgetary outcomes most likelydo not effect the distribution of appropriations acrossprograms (say, highway construction), but they can be afactor in intraprogram allocations (for example, withinhigher education) because there decisions are more likelyto 'be simultaneous rather.than sequential and becauseprogram outputs, workload levels, and other indicatorsof budgetary need are more likely to be comparable.

A budgetary dialogue that includes information andanalytical input covering all of higher education, aswell as information on how budgetary decisions arereached, is conducive to a more objective process, onein which special treatment and privilege are made moredifficult to dispense: The role of information andanalysis in higher education budgeting is least well-developed in states where the:allocation among iolsti-tutions is overtly political. Equity in these *:tates

is achieved principally through "logrolling," mviprobably at a somewhat greater cost to the statc':.;taxpayers.

14 3132

INSTITUTIONAL CREDIBILITY AT THE STATE LEVEL

Institutional credibility at the stata level dependsgreatly on interpersonal and interagency relationships,and it can be undermined as well as supported throughexploitation of the opportunities for clouding or clarify-ing issues through information reporting. Credibility(and trust and confidence) do not follow alone fromprocedure--from doing PPB, as it were--what count arethe content and legitimacy accorded information andanalysis. .

Cheit (1975) concluded that the credibility of highereducation iristitutions is not much improved by new infor-mation and analysis procedures, but this conclusion wasbased primarily on information and analysis systems asinstruments for improving institutional, efficiency. It'is true that protagonists for these systems have emphasizedtheir ability to improve technological efficiency, but inpractice improved efficiency in operating a public serviceagency means primarily lowering expenditures or loweringthe rate of expenditure growth, which is to say, in thecurrent vernacular, retrenching. If that alone is thegoal of state agencies, then it can be achieved much moreeasily within the confines of a less analytical budgetingprocedure. State agencies can request that institutionssubmit budget requests within 105 percent of last year'sexpenditures, for example. Howard (1973) and Schick (1971)both commented that budgetary formats which include sup-porting program and fiscal detail do not seem to befavored by political administrations seeking retrenchment.Providing the information to support cutbacks program-matically and in terms of service outputs also opens theopportunity for requesting increases on the same basis.

Expenditure cutbacks have become a reality in a fewstates, but most public institutions continue to receivebudget increases when state revenues permit them (Ruyle& Glenny, in preparation). Although it will not beuniversally true because of the presence of other factors,some institutions and systems have found it distinctlyto their advantage to take an "open system" approachin making information available to state agencies because

133

147

this provides an opportunity for them to credibly sub-stantiate budgetary needs. In several instances, intro-duction of information reforms has been tied to reductionsin state personnel system or other controls, or to morefavorable budgetary recommendations on appropriations.

While information and analysis may contribute to amore receptive state-level review, whether the resulteventually will be a higher level of higher educationappropriation is not really an appropriate question. Allgovernmental agencies are underfunded, if one were to

-judge-by-the level of funds they-request versus the levelof funds they receive. It is certainly the rare agencywhich does not receive a budget appropriation below theamount requested. If state revenues are available,higher education may receive marginally greater appro-priations, based on relatively gross political considera-tions. Institutions may get greater appropriations basedon technical factors as well, although with the bias instate-level budget review necessarily toward reduction,good technical support usually only prevents an insti-tution from losing funds. In one institution examined,however, the use of special analytical studies to demon-strate deficiencies in their funding did result in greaterhigher education recommendations for that institution andsubsequent increased funding. The adoption of certainbudgetary classifications also may be encouraged by thepromise of greater appropriations, although it is notclear that such presentations are all that is needed toconvince state officials. Certainly by using programclassifications or various workload measures, it iseasier for an institution, on the basis of its documentsalone, to justify additional budget requests. Someinstitutions bolster their budget justifications con-siderably by using the rubric of program budgets, specialstudies, and program and mission statements.

Miller (1964) noted that the development of formulascould serve as a vehicle for improved communicationsbetween institutions and state agencies by requiring the

.productive involvement of state agency staffs earlier inthe process, and of institutions in the latter stages ofthe process. He identified the cost to institutions as ,

1 1 3134

a willingness to agree to compromises on their requests,and the cost to state agencies as a willingness to re-frain from arbitrary decisions during joint phases ofthe process. "The overall result . . . is that eachparticipant is better informed and had a longer periodof effective participation and influence" (p. 83). ThesaMe can be said for reasonable exchanges of informationand analysis in the budgetary process.which, in essence,are only less highly stylized and structured substi-tutes for the "guidelines" and "aids to calculation"provided by formulas (Meisinger, .1976).

SUBSTANTIVE BUDGET DIALOGUE

Introduction of information and analysis into thebudgetary process can shift the focus of disagreementsfrom what the facts are to the implications of thosefadts; that is, what programs should receive state sup-port? As a consequence, the budget dialogue becomesmore relevant to the ultimate purpose of the budgetaryprocess, the determination of policy. Systems forinformation and analysis have freed some state highereducation agency staff to devote More time to policyissues, and they can make it possible to respond quicklywith information and analysis when issues are raisedduring the legislative process. The inability to re-spond at such times with reasoned support can lead tothe imposition of relatively arbitrary budget cuts.

An analytically based budget process tends to raisethe level of sophistication of the budget dialogue byadding another dimension to communications. Instead ofpreferences alone, objections or support can be raisedon matters of substance. In a bargaining situation,information is power, but so is lack of information.Bargaining depends on getting someone to believe some-thing, and the availability of information, or the lackof it, may be crucial. We do not know that substantiveinformation and analysis lead to better decisions, butmost administrators believe that access to adequateand relevant information is. a characteristic of a gooddecision process.

135

14 9

Critics of the substance in the dialoguedecision argue that consensus may be fatilitated ,hy de-

bate over more easily resolved matters of fact. But

decisions tied to relatively'inconsequential matters donot establish policy, and as a result may establish a_tradition of arbitrary and capricious decisions., Onecan gain as well as lose from seemingly arbitrary choices,

.but on the whole analysis raises the tough questionsrequired to stimulate adequate evaluation of newand ongoing programs in relation to policy. And it may

eventually supplant the questions raised solely for-p0iitical-purposes--about-the-presidentis-salary-or the----reasons why an interested person's son was not admitted

to medical school.

From a societal rather than an individual point of

view, choices made from the broadest possible set of'alternatives are more likely to meet the requirements ofchanging social preferences than choices made fromalternatives constrainad by lack of information. Infor-

mation and analysis can force attention to problems thatwould otherwise receive little attention. Current fiscal

difficulties in cities and universities offer goodexamples of instances in which available options warereduced by the postponement of attention to criticalproblems that might have been raised by better informa-tion and analysis. Consensus, however, is not so easily

achieved.

By and large, few would argue against the advantagesof a more substantive budget process, but there is doubtas to how systems and procedures can be used to accomplish

such a process. Bowen and Glenny (1976), in their studyof institutional response to fiscal stringency, found thatretrenchment and midyear budget cuts have added urgency tothe review of existing programs and accelerated interestin and use of allocation procedures based on more sub-stantive program measures. Further, Exhibit 10 shows

.1that in any year, the activities of a budget agency mayinvolve preparatory planning, budget preparation forthe coming fiscal year, execution of the current budget,

and postaudit. When revenue expectations force a cut in

expenditures for the current year, the question of the

150136

Exhibit 10

OVERLAP OF BUDGETING PHASES

Year(t)

t t+1 t+2 t+3

Budget forlast Post-year (t-1) audit

Budget forcurrent Execu- Post-year (t) tion audit

Budget for Develop- Execu- Post-year t+1 ment tion audit

Budget for Pre- Develop- Execu- Post-year t+2 paratory ment tion audit

planning

15 .1.

137

consequences of the cut for postaudit evaluation anddevelopment and planning for subsequent budget years isimmediately raised, and more substantive planning and

--allikatioh-idrOdeddres may result.

STIMULUS FOR ORGANIZATIONAL CHANGE

Although organizational change is not always for thebetter, frequently a far greater problem than slowing therate of change is how to achieve any organizational innova-tion at all. Examples of the ways in which the develop-ment and impleMentation of information systems haveb,-4imulated change in the management of higher education,in addition to changing other practices in higher education,could produce a very large report. One cannot often claimthat such changes have necessarily been stimulated byinformation reporting and analysis; it is somewhat morelikely that the resources to develop and implement aninformation system and the interest in it provide an in-fusion of energy that is then rechanneled in variousways.

Information task forces may serve some of the samepurposes as formula development committees in providinga focus and a forum for discussion of management-relatedissues. Analytical studies and performance evaluationaudits, which are sometimes misdirected or ill-conceivedin themselves, do tend to have the effect of releasingpent-up energy within the system, and this in turn resultsin administrative revisions. Analytical studies andevaluations-also may bring problems to light which can beremedied, even when the original issue is insoluble or

misconceived. As such, these studies may indeed repre-sent solutions in search of problems, but that may beall that is feasible. (See Schmidtlein's dissertation,1976, for a discussion of the outcomes of Departmentof Finance audits in the University of California librarysystem.)

1-'-mv ;)

138

"ORGANIZATIONAL MEMORY"

When procedures for information submission and analysisfocus and condense the issues relevant to a given annualbudget review, as well as produce a historical record ofthem, the budget process may avoid repeated "rediscovery"of these issues and their solutions. Discussion andanalysis can facilitate the achievement of agreed-upon

. goals by reminding participants of previously establishedfacts, premises, and conclusions. When the discussion ofthese issues is not formalized with good information anddocumentation, the issue is often lost as a referencepoint, and institutions and state agencies may return tothe iscue in circular fashion in succeeding periods. Be-cause of the relatively high turnover of personnel instate agency budget staffs, particularly junior personnel,these agencies, compared with other agencies, tend tohave relatively limited organizational memory.

TECHNICAL PROBLEMS WITH INFORMATIONAND ANALYSIS SYSTEMS

The problems commonly associated with informationand analysis systems concern their failure to meet cer-tain political process requirements. For a discussionof these issues, which helps to explain the inabilityof system designers to implement their products, seeGlenny (1976), Schmidtlein and Glenny-(in preparation),and the "critique" literature referred to earlier. Thissection considers the technical problems confronted inimplementing and maintaining budgeting information andanalysis systems. While efforts to reform informationgathering and analysis have fallen far short of thosecalled for in full-blown PPB systems, state agenciesand institutions are making changes in establishedbudgetary procedures. The assertion made by detractorsof PPB, that "no one knows how to do program budgeting"(with its implication that no one need worry about thenitty-gritty details of implementation), is not muchhelp to an agency trying to design improved, workable

. information and analysis procedures.

139

15"

NONCOMPARABILITY OF DATA

Perhaps the most frequent information deficiencymentioned by budgeters is the lack of comparabilityof data submitted b3 institutions within the state.Although institutions use common accounting structures,this in itself is no assurance that common practices arebeing used in allocating costs or applying definitions.Institutions in many states continue to use personneland pUrchasing systems of their own design, with variousresults. Responsibility for salary levels and for fillingpositions may reside at the departmental dean level, forexample, and consequently state-level involvement inapplying standards and classifications may, at least foracademic positions, be virtually nil.

Lack of comparability also stems from the variedcharacteristics of institutions, especially those engagedin research and gradUate instruction. An inspection ofcollege course catalogues comparing even those insti-tutions which have similar missions in graduate educationand research usually reveals that institutions go togreat lengths to differentiate their course offerings.Thus, any comparability of educational outputs is forced,unless individual courses are made more similar or courseand discipline area designations are made more precise.Two economics departments, for example, one of which isheavily oriented toward mathematical economics, econo-metrics, and managerial economics, and another that isprincipally an economic history department, will havequite different cost and workload measures.

Differences in departmental organization among .

campuses require the use of common course taxonomies,such as the HEGIS structure, which break any connectionwith program responsibility. Two specific examplesfound in the University of California system, but un-doubtedly common in other states, are the existence ofinstructional programs in operations research andsystems analysis, both in schools of engineering andbusiness administration, and programs in marketing inschools or departments of agricultural economics, busi-ness administration, and administration (which includes

15.1140

public sector management). It is instruction, of course,that makes the largest single use of funds within thebudget, and consequently, it is the function for which itis most critical that some minimal level of comparabilitybe achieved. Short of the common course-naming and number-ing systems being developed in Florida, comparisons ofcosts and workloads in the instructional area are severelycompromised.

Noncomparability is also attributable to differentinstitutional capacities for providing data. Apparentlybecause of differences in administrative style and interest,some institutions have been developing their own manage-ment information systems for several years as part ofsome of the NCHENS development project, or in connectionwith locally initiated efforts. These institutions arenow much more capable than others in the state of pro-viding information on costs or other performance measureswhen state agencies request it, although problems inmaking comparisons with other institutions are stilllikely to be present.

The prospect of achieving comparability of informa-tion on instructional support activities, which are notunlike activities of noneducational organizations, seemsconsiderably better because there is greater similarityin practice and because common measures of activity levelare often available to serve as common denominators. Theimpression should not be left, however, that becauseproviding comparable instructional data presents dif- .

ficulties, data on costs, activity levels, instructionaloutputs, and other measures are not being widely reported.However, the dangers in forcing comparability throughchanges in instructional practice, for which diversityis highly desirable, are mentioned almost universally. Al-though this study focused on the state level, it did not pro-vide an opportunity to observe how accurately differencesin institutional practice are perceived at the state level.Because institutional diversities obviously continue,comparisons among institutions in the course of budgetreview are, to some extent, always artificial. Budgetarydecisions very rarely follow directly from such compari-sons. What effect comparisons would have on budgetary

141

153

decisions, if made on the basis of comparabledata, re-mains an hypothetical question. Institutions fear thatsome comparison may become the crucial factor in reducingtheir requests, and state agencies are noncoMMittal as tohow they would use more comparable information,

DATA NOT USED

Use of data is difficult to judge, but there arenumerous indications that a great portion of the datasubmitted to state agencies is not used in any explicitway. The admission of nonuse.by budget analySts andother participants in the budget process seeMS unequivo-cal, and the small size of staffs of most executive andlegislative agencies virtually precludes the regularanalysis of large quantities of information. One mustallow for the fact that state agencies legitinlately re-quest data they will use only selectively, aS well asdata for which there is a targeted or planned use. None-theless, nonuse of data appears to exceed such allowanceswhen institutions are burdened with providing substantialquantities of information which can yield feW, if any,positive effects because so much of it is scarcely lookedat. In positing that lack of use is a probleM, we igmorethe suggestion of various critics that actual use ofmore of the data provided might constitute a far greaterproblem.

Some of the data reported are not used because theyare somewhat artificial and do not provide a good indi-cation of institutional plans or practices. Further,unusable and useless information swamps data that couldbe informative. A common problem of information systems,in general, is that once reporting is initiated it con-tinues indefinitely, although this is still rare instate higher-education reporting systems becaUse theyare so new. In-:two states we found data being reportedat the state level that had been required under pre-vious reporting sYstems, but were no longer Used in thecurrent budget or planning process. There is also verylittle tendency for new information to replace-informa-tion previously collected. New information is almost

1 5142

always added to that already being cpliected, and not onlybecause organizations are slow to change. Once data cate-gories and their definitions are determined, revisionsupset the historical continuity for trend analysis andcomparison. There is, consequently, alwayS considerablepressure to maintain old structures and add new elementsrather than consolidate and revise.

Routine analytical studies freqUently cannot be usedin the budget process because the issues raised duringthe course of review change from year-to-year. If studiesare used routinely, they are likely to he highly struCturedcost studies that have virtually become a formula. Whenanalytical studies do not have a well-defined use in thebudget, however, they are easily shUhted aSide or ignored.If they do have an impact on the process, it is usuallyin highly simplified form or as amMUhition in the bargain-ing process.

The requirement that certain data be included inbudget requests often stems from a belief that the pro-cess of decision, from the department to the state level,can be influenced by forcing data into new classifica-tions or requiring new kinds of data. Although we foundmany examples of new data classifications being used, theimpact on budgetary decisions appears to be limited. InHawaii, a programmatic presentation of the University'sbudget resulted in certain programs, such as the Educa-tional Television Station, being treated separately.Previously, the TV station had been folded into theuniversity's budget. In some instances, inforvation onprogram detail, which is not related to organizationalcategories or the functional accounting categories thatinstitutions use, is included with the subnlission foreventual connection with appropriation categories, butstill receives little attention in the process of review.Generally, Schick 41971) described the new kinds ofhigher education information quite well when he assertedthat participants are more likely to seek and use datawhich suit their preferences than to alter their behavioras a consequence of information availability.

157143

Jernberg (1969), in examining congressional commit-tee behavior, found that committee budgetary inquirystyles, determined by questions posed during hearings,did not change when the budgetary format used by federalagencies was changed. On the contrary, regardless ofthe manner in which budgetary requirements were presented,the committees on which he reported tended to followparticular lines of questioning dominated by concernsover agency integrity, agency capability and reliability,or agency performance. Agency integrity was examined withquestions focusing on congressional oversight; capabilityand reliability through an object of expenditure emphasis;.and performance evaluation through examination of programgoals and achievement. A question Jernberg did not pur-sue, which arises naturally from the data he presents,is whether the style of inquiry which dominated each com-mittee was related to the character of agency activitiesand functions. An agency, if we may call it that, likehigher education, may be more effectively examined withreference to capability and reliability, to use Jernberg'stypology, because oversight (in the state context, super-vision) lacks legitimacy, and performance expectationsare so difficult to establish. It is also likely thatin a period of fiscal stringency, when there is a dis-position toward reducing higher education budgets, thereis also a marked tendency for state agencies to ignoremuch of the new information based on growth assumptions.

UNSOPHISTICATED COSTING TECHNIQUES

The emphasis on providing unit cost Jata of variouskinds has been so great that we have singled it out forattention. Attempts to develcip unit cost data alsosharply illustrate the difficulties that confront attemptsto apply sophisticated theoretical rational-analytic con-cepts within the routines and politics of the budgetprocess. Because of gross variations in accounting sys-tems, it has been no small task to change accountingconventions to the point that acceptably comparabledirect average costs per student credit hour could bereported. This is the practice in several states con-sidered in this study, and it is frequently the basis

144

1_5(3'

for requesting workload increases for changes in enroll-ments. However, comparing institutional average costsas a technical solution to the problems of identifyingefficient educational practice or ascertaining a directrelationship between funding support and average costsis on balance a nonproductive exercise. The comparisonsare based on assumptions that do not hold in actualpractice, and invariably adjustment and compromise arerequired when enrollments fall because of the realitiesof institutional operation.

The root of the problem is that use of average costdata assumes that increases in instructional costs aredirectly proportional to the number of enrolled studentsover the entire range of enrollment levels. Another wayof saying this is that such cost data recognizes no dif-ferences between fixed and variable costs; all costs areassumed to be variable. It is, of course, a complicatedmatter to determine which costs are variable and whichare fixed; in actual practice those that are fixed withrespect to one decision May be variable with respect toanother. Nonetheless, in times of growing enrollments,institutions have found it convenient to ignore thesecomplexities and to request workload increases on thebasis of average costs because their marginal costswere undoubtedly less than their average costs. Unfor-tunately, when funds are cut back under the same formulathey discover, to their discomfort, the fallacy in thelogic.

The computation and analysis load could well beunmanageable if regular attempts were made to estimateand use marginal costs, although this has been approxi-mated during periods of growth through the use ofgraduated cost or workload standards based on increasingreturns to scale. Reliance on average cost standards(ratios, costs, or other parameters) during a period ofgrowth may allow some institutional flexibility th=ughover-budgeting, but it is generally believed that theuse of average-cost formula factors.during 'he 1960skept budgetary demands below what they might have beenbecause realized enrollments exceeded forecasts. On

145

1.59.

the downside, formula fact.rs end z!tant,71r:Is giNe way,

and their demise necessitate:.; Fadden to ad hoc

methods of funding which often la:7k any consistent cost

basis.

Application of costing techniques also is hamperedby the jointness of educational outputs when the inst:uc-tional activities within the institution are treated as

production processes. The use of physical plant, automateddata processing equipment, libraries, and faculty and stu-dent inputs in activities that "produce" instruction of,various kinds (regular degree programs at varying levelsand extension), in addition to research and auxiliaryenterprise outcomes, makes cost assignment a ratherarbitrary process. Cost allocation conventions are usedto assign costs in the absence of recharge systems, butthis sets up a potential source of conflict between thosewho are attempting to cost outputs and those who mustmanage inputs. For management purposes, the dictum appliedin industry--that cost reporting should be consistent withthe managerial responsibilities of the various organiza-tional units--conflicts with the attempt to assign coststo the variety of educational outputs which may have somevalue in the planning process. State budgetary'planningis dominated by a philosophy of control and input costingwhich is much more consistent with the assignment of coststo .points of management responsibility (cost centers)than with.assignment on a programmatic basis, where respon-sibility almost always cuts across organizational bound-aries.

The lack of comparability in expenditure classifica-tions and the.failure to recognize fixed and variablecosts restrict the eventual application of program costsas analytical devices. Nevertheless, if the budget pro-

cess can permit the flexible application of program costs,these costs may find wide appeal as cost indicators orin formulas. The attractivenebs of formulas for insti-tutions depends on the acceptance, on principle, ofthe formula as a device for requesting funds, but not forspending funds. Program costs.based on the ProgramClassification Structure do observe such a principle,although it is not by design, if actual institutional

iuJ4.

146:

cost centers and decision points are sufficiently dif-ferent from those implicit in the Program ClassificationStructure. Such a program cost structure has the effectof establishing an information veil which prevents state-level involvement in operating-level decisions, while atthe same time providing information in aggregates thathave potential uses in planning and coordination. However,limiting access to input and responsibility control dataat the state level runs counter to current demands forsatisfying administrative accountability.

DISTORTED INCENTIVES THROUGH WORKLOAD FACTORS

Problems caused by the use of average rather thanmarginal standards have led some states to apply thehistorically incurred average costs at the institution ,as workload factors for the forthcoming year. Although'these budget estimates may be altered during subsequentnegotiations in the budget process, this methodology hasa tendency to build in historical inequities or dif-ferences. Another technique to blunt the sharp edge ofenrollment changes is to use a three-year moving averageof enrollments (or other workload element) to determineworkload changes. This approach may reduce the severityof shifts caused by ignoring fixed costs, but it isunlikely to be responsive to factors that are the actualcause of the shifts.

Most states employing formula workload factorsdetermine an average factor for all institutions in thestate and apply this to.the appropriate workload elementto generate the budget estimate. If an institution isincurring higher than average costs, a formula of thissort will reduce its costs if the institution can cut.expenditures fast enough to avoid a deficit.

a

Current accounting procedures in higher educationlead to cost estimates that give very little insightfor management purposes because they bear little re-lationship to the manner in which instructional costsactually are incurred. Such a methodology, coupledwith the relatively inflexible use of faculty inputs,

147

184

will invariably show increasing costs whenever enrollmentor student credit hour production declines. Rarely willan institution be able to reduce faculty or other factorsfast enough to avoid a rise in unit costs. Thus, thelevel of unit costs is as much a function of the currentgrowth trend of an institution as it is of any insti-tutional or programmatic characteristic. To avoid in-creases in costs, institutions and depar-ments areinduced to stabilize or increase enrollments, even at therisk of reducing quality.

ADEQUACY OF OUTPUT INFORMATION

Ideally, the program review should be separate frOmthe budget process rather than a part of it, but theremust be a bridge between the two processes which ensuresthat program planning decisions are reflected in the budget.Currently, much of the programmatic information only addsbulk to budget submissions; and the program review processis completely isolated from the budget process. For validprogram evaluation, information on outputs is needed thatgoes considerably beyond the "assemblyline" measures thatare included in most budget submissions, namely, degreesgranted and student credit hours produced. Educationalvalue-added, labor-marked supply and demand conditions,student preferences, program quality, and the like arefar too complex to be introduced into the routine calcu-lations that characterize most of the budget process.One state requiring this exacting information on programoutputs, Hawaii, was not able to use it in any analyticalmanner in the budget review process. The.university wasunder some pressure to reduce the number of output mea-sures to a more manageable number, but these measuresare probably among those that would be most useful inprogram review. (See output measures for the Universityof Hawaii in Chapter 4.)

DATA QUALITY AND CREDIBILITY

We have emphasized repeatedly that budget data con-sists, in large measure, of estimates of expenditures

or148

for future plans. Where these data do not reflect actualinstitutional practice or intentions, credibility sufferseven though institutional autonomy may be maintained. Aslack of confidence in the data provided by institutionscarries over into the reSults of studies of various kinds,as for example, faculty activity analyses, which uniformlyhave very low credibility at the state level.

Higher education institutions are in a unique positionvis-a-vis other state agencies because they have a varietyof funding sources. Generally, data on the magnitude ofthese alternative sources are available to state agencies,but often (where the institution feels it can reserve accessto actual data) these data reflect only gross estimates.Usually no link is made between sources of funding andexpenditures, and this effectively precludes state involve-ment. State agencies often request this information, evenwhen the institution claims complete responsibility forits allocation and the credibility of the data is low.

It is difficult not to be impressed by the apparentaccuracy and precision of quantitative results, partic-ularly in the course of a process which itself mustdetermine a quantitative dollar appropriation. Thisapparent precision frequently misleads politicians. Thequality of budget data cannot be assessed by casual in-spection, nor should it be expected that all inaccUraciescan be removed. Expenditure estimates and projections oftrends are necessarily speculative. Unfortunately, thequality of data is often assumed to have improved simplybecause it appears on punch cards or computer print-out.

MISUSE OF DATA.

From our discussion of misuse of data, referred toabove in the consideration of costs, it is apparent thata widely applicable criterion for data misuse cannot bespecified because of the multitude of rationales thatare applied in the budget process. In the course of thebudget process, participants introduce data and conclu-sions derived from technological, economic, legal,political, and social principles.

149

16 3

What is reasonable from-an economic standpoint (in-creasing the portion of educational costs paid for bystudents, for example) may be unthinkable in politicaland social terms. (See Diesing, 1962, for a discussionof this hierarchy of applied principles in a generalframeworK and Fincher, 1975, for a discussion of currentproblems faced by higher education in these same terms.)Misuse in our discussion refers to misuse in a techno-logical, economic, or legal sense rather than in apolitical or social sense because the former areas havemore clearly established rules for determining error.Political and social rationales are also used to supportand influence sociopolitical transactions and, of course,ultimately to dominate the determination of public choices.

As explained earlier, the use of average cost-formula factors has little support in technological oreconomic principles, but other redeeming qualities maymake it an acceptable budgeting device. When used toreduce operating ftnds as a consequence of workloaddecreases, average costs are a very weak basis on whichto proceed because marginal operating costs are surelylower than average costs. The use of direct costsof instruction rather than full costs moderates thisdeficiency somewhat, but the full annual per-studentcosts recommended by the National Commission on FinancingPostsecondary Education as a budgeting device have nological application under conditions of decreasingworkloads.

Misapplications of other data also axe related tolack of comparability. The practice of aZ.tempting todetermine a standard for appropriation levels by com-paring states on the basis of expenditures on highereducation Per capita, or by other measures of highereducation support, is particularly unsuitable becauseof gross variations among states in the funds that areincluded in expenditure, totals and in the costs ofservice. We found several instances of this argumentused to increase higher education appropriations, andit is also used occasionally as an argument to limitfunding. For a discussion and an example of the kindof analysis that is a necessary prelude to interpreting

1 4'6 11ti

150 d

e.

these state expenditure data, see McCoy, Cherin, Makowski,and Weldon (1976). Although cross-state comparisonsshould not be uaed to discourage socially valuable dif-ferences between the states of a federal system, it isimportant to mention here that cross-state comparisonsdo have a useful application--but more in determiningminimal standards for equality of opportunity than inidentifying technically efficient practice.

A similar practice is the use of average facultysalaries at "comparable" institutions as a factor informula calculations to support instructional budgets.We do not refer here to the common practice of includingaverage faculty salary data compiled either by theAmerican Association of University Professors or theinstitutions themselves as support for faculty salaryincreases. The practice of costing faculty salariesat levels found in comparable institutions appears tobe an application of the notion of opportunity costs.Under this cost concept, the cost of any factor is thehighest price that would be paid for the factor inalternative use. Although advocates of allocating fundsto public agencies on the basis of benefit-cost analysismight approve, the advancing of opportunity costs tosupport a request for higher academic salaries at thesame time that other factors are costed on a historicalbasis cannot be justified except on grounds of expediency.Funds generally are not allocated on an opportunity costbasis if they are appropriated.

THE DATA PROVIDED BY INFORMATION SYSTEMS

In any in depth review of a particular budgetaryissue, the data required for adequate review are likelyto be details and specifics rather than uniformly com-prehensive program or functional activity data at afairly high level of aggregation. Budgetary informationsystems are not very good at specifying much of thisdata before the budget is submitted because institu-tions may actually define issueS by the nature of theirreqUest. Program budget classifications, or for thatmatter all functional classifications, usually aggregate

16 5151

all expenditures and do not separate out the details forspecific new or changed programs so that these canexamined. The depth of information is frequently uniformacross the budget when what is needed is greater depth onthe few issues that can be effectively reviewed in any onebudget cycle.

Because issues change from year to year (even thoughprograms and functions may not), budget agencies reviewingissues may require information of varying kinds. Thisfrequently leads to requests for information that is notcollected routinely because systematic collection of allthis information would require an information system ofunmanageable size and cost. The assumption behind pro-vision of program data is that the data serves as a basisfor evaluative dedisions on the effectiveness and effi-ciency of higher education activities. This is not,however, the function that staff analysts at the statelevel usually fulfill. Staff analysts must review thetechnical elements of the agency's request for propercosting and accuracy. Thus a comprehensive programmaticfocus may prevent analysts from receiving the data theyneed, as well as fail to provide the issue orientationneeded for policy decisions by upper-echelon agencyofficials.

Consolidations of data that have developed histor-ically, whether from patterns of institutional or systemgovernance or tradition, frequently do not combine wellwith the submission of new budget-documents,.such as pro-gram revision requests or program change proposals. Forexample, although budget data for the continuation ofexisting programs may be consolidated at the systemlevel, information regarding requests for new prognams,necessarily being more specific, must be at the campuslevel. But state agencies will have problems evaluatinga proposal which is an increment to an undescribed base.

OTHER TECHNICAL PROBLEMS

A common fault reported at the instituLional levelis that most special information gathering efforts

152

16 3

I ,

stimulated by requests from systems administrations orstate agencies rarely result in compilation's or studiesbased on the data that are returned to the linstitution.While the studies may relate to 'systemand'state agencyresponsibilities, institutions still wani"to,see howthey stand in comparison with others,'and'MaY actuallybe able to use the data to greater effect in their ownplanning processes. When there is adequate feedback,errors in the data can be corrected and interpretationof the data facilitated.

The requirements of the U. S. Office of EducationHEGIS reporting, and its associated heavy informationload, have Lequired that institutions devote considerabletime and effort and the assignment of substantial ADPresources to preparing and submitting these data oncethis reporting capacity has been developed.! Requestsfor data by state agencies can be filled quite readilyif the HEGIS reporting categories are appropiiate.Quite frequently state requests for information addressspecific organizational issues neither encompassed nordefined by the HEGIS categories. InstitutiOns are thenunable to respond appropriately because available report-ing resources are already committed.

POWER AND INFLUENCE IMPLICATIONS OFINFORMATION AND ANALYSIS

The implications for power and influence of theincreasing emphasis on technical budget review are dis-cussed in the context of the full range of interagencyrelationships in Glenny (1976). For completeness of thepresent discussion, we sketch below the most significantthemes associated with changes in information reportinganalysis at the state level, along with our impressionsof trends in the states.

Information and analysis are both cause and effectin that they may produce shifts in the locus of powerand influence when they are introduced, while,being, atthe same time, an indication of existing power and in-fluence relationships. Furthermore, there are two

153

16'7

distinctly different kinds of power-and-influence shifts

in question. One is a tendency toward centralization--that is, a power shift toward the upper levels of thestate higher education budget hierarchy; the other, ashift of influence toward administrators and staffs whocontrol and understand the techniques and knowledgethat are being applied in information and analysis systems.The present study can offer relatively little new concreteevidence that would contradict the widespread impressionthat both of these power shifts are indeed the dominant

trends. This view also seems clearly supported by_reportson the growth of staff in state budget agencies, the re-

placement of lump-sum appropriations with line-item appro-priations, and the increasing number of decisions inhigher education that must be reviewed or approved atthe state level (Glenny, 1959, 1976; Schmidtlein & Glenny,in preparation). What is not so clear is whether thesetrends are actually having the dire consequences that

have been predicted.

CENTRALIZATION

The underlying strategy in current attempts to in-

crease the rationality or accountability of the state

budget process for higher education has been to increasethe information available at the state level. Althoughthe availability of this information at the state levelhas led to instances of greater state agency involvementin institutional operations, much of this involvement

has taken the form of second-guessing administrativedecisions rather than attempting actually to make them.Direct involvement in educational decisions is L rictlylimited by the separation of educational and administra-tive functions within the institutions. However, state-

level involvement in administrative decisions at theinstitutions in the form of centralized approval andreview procedures is greatly facilitated by management

information systems. The importance of systems as

mechanisms for institutional control seems somewhatexaggerated when'one considers that states with a long

tradition of central control over state agencies

1L6154

and functions (Connecticut and New York, for example)have been able to exert such control without the useof systems.

Additional doubt is cast on the role of f-Ilese cen-tralized information and anlysis systems in transferringinstructional decision authority and respunFilbility upwardwhen one considers the experience in one conbolidated-system of public higher education in this st isinstance, central university administration had an infor-mation reporting system in operation which they and otherstate agencies felt served the budgeting function at thestate level adequately. Central administration officialsnoted, however, that they could not get administrators atthe campus level to use the information in managing theirown affairs. Consequently, a pilot project is underwayto encourage use of this information in campus management.In addition to this kind of difficulty in bridging thegap between campus and system or state-level administra-tors, there is an even greater difficulty in involvingfaculty in administrative decisions. The obstacles tomaking institutional management systematic though academicare implied in the quote below of a renowned teacherspeaking to his colleagues.

We can also be sidetracked into administra-tive functions. I remember wasting threeentire weeks on.preparing a budgetary pro-jection to cover the next five years--aplan which was rendered completely meaning-less by unforeseen and unforeseeable events.(Highet, 1976, P. 40)

Professor Highet would probably not be contradicted byadministrative colleagues on the usefulness of most five-year budgetary projections as they are currently prepared,but faculty also ignore information on predictable eventsand actual choices.

Gross (1969) commented that centralizing informationand analysis leads to centralized decisionmaking only iftop-level officials want to use the information to makedecisions. The acceptance of responsibility for decisions

155

169

at the institutional level has perhaps been abdicated asmuch as it has been usurped. There also are indicationsthat state-level officials do not desire to Make the re-allocation decisions that accompany the steady state inhigher education, but that they believe programmaticinformation and substantive analysis will assist insti-tutional and department officials to,make these decisions.This may be true, but institutions frequently deny thatthe information state agencies request is useful forthese purposes. On this point see Bowen rnd Glenny (1976).

GROWING STAFF INFLUENCE

. As the public sector has increased in size and com-plexity, the accountability of public officials hasbecome more vague, and similarly, the specification ofpolicy by elected officials has become less sharply de-fined. The differentiation of state governMent servicesthat has followed on the increase in size has been, in,itself, a major factor in shifting discretioh from electedofficials to administrative staff. To some extent, in-formed analysis has been used to complement staff dis-cretion because of the lack of clarity in Policy directionifrom elected officials. Considerable attention has been ,

given in the recent literature on public organizations(Ellul, 1965; Wolin, 1960) to the loss of poWer byelected officials, especially legislators, arld to theacquisition of this power by agency staffs Of the execu-tive bureaucracy. The development of information andanalysis systems which serve this bureaucracy are fre-quently noted as factors in this shift of PoWer andinfluence. Development of supporting staff for legisla-tive communities is often part of an atte ingt to regaincontrol of a complex executive bureaucracy, but legisla-tive staff growth is also likely to be folloWed by aneed for more information at the state level.

Increasing the free flow of information and analySisin an attempt to rationalize and objectify the budgetingprocess, and increasing the number of interested staffraises the specter of limiting political Options. Staffin a number of the states visited commented that the

J.56

higher education portion of the state budget was one ofthe most political. Thus, there iS considerable opposi-tion to implementing these info rmation systems in thehigher education segment from many quarters and, at thesame time, an increased interest in using them in anattempt to limit some of the politics. One staff direc-tor noted that the impacts of analYsis and programevaluation are likely to be greater where the leadershipfunction of either the executive or the legislature isweaker. But the introduction of these review activitiesis not likely in itself to be a caUeal factor in weaken-ing the leadership function.

Considering only budget staff/ there clearly axedifferences in the staff-line relationships of the execu-tive and legislative branches of state government, attrib-utable especially to executive resPcnsibilities forpreparing a comprehensive budget and executing it. As aconsequence, executive staff have clear information re-quirements and review responsibilities. In those statesin which the governor is not the elected official re-sponsible for preparing the state blIdget, joint executive-legislative budget responsibilitieS require the legislativebudget staff also to have rather clear budget informationneeds and review responsibilities. Where the legislatureconcentrates on review of the governor's budget, staffwork may be substantially less technical and well-definedthan staff work done in the executive budget office. Theeffect of this is that executive staff do have Substantialinfluence, especially with respect to budgetary details,through their technical review; but those iecativebudget staff not involved in putting togethec a 0,.6.eye-hensive legislative budget have considerablY lesn in-fluence. This being the simplest and most Iirect 'Aely

that staff influence is felt in budgetarli notconlef.:, inthose instances in which the norms Of exec,,itive aPclegislative.roles apply, staff inflUence 3-4 greatex inthe executive than in the legislative bra.:?c.h. Legx%la-tive staff are much closer to the elected offIclot;s theyserve and, because of this may exert a dir4t '.;,Ipact onspecific igsueS.-

1.7 1.

157

Kingdon (1973), in a study of congressional voting de-cisions and the influence of various information andanalysis sources, found that of various informationsources,-such as congressmen, lobbyists, publications,and the media, congressional staff were the least impor-

tant. We have no data other than our impressic)ns toverify or refute this conclusion as it applies to state-legislatures, but our impression is that in a ranking onbudgetary matters legislative staff would fare considerablybetter in terms of their influence, and that their rankin relation to other factors would probably Le higher.

Of course, there are great differences among the states,and it is certainly not clear that being low in terms of

a ranking is in any way comparable to an abLolute rating

from very influential to not influential. Balutis (1975),

:in a survey of legislators and other close legislative

observers in New York, reported that most respondentsrated budget committee staff as either influential or

very influential. Butler (1975) concluded that staff of

the Texas legislative budget board were highly influential,

although this would be expected because, in effect, theboard prepares a legislative budget.

Other legislative staff are not so influentinl, as

Kent (1975) suggested about Illinois. There are, of

course, other staff analysis functions belides tchnical

budget review which provide a means for influonr:ing

budgetary outcomes, such as_the preparation of policystudies and program evaluatiOns. With respect to dilution

of the influence of elected officials, one generalization

seems clear: Influence possessed by elected officialsand staff is not fixed in amount, so that growth in staff

influence necessarily decreases elacted official influence.

To the contrary, most state executive or legislative

bodies which are influential are sc, in large part, be-

cause of the capabilities of their staff.

STRATEGIC USES OF INFORMATION

Frequently, information and analysis-prattites are

seen more clearly as the short-run consequences of

established power and influence than as.factors in the

158

17 ")two

gradual shift of authority. Institutions sometimes claimthat state-level requests for information are made simplyto plade the institution on the defensive or embarrass itbecause it cannot provide the data. On the other hand,state-level agencies accuse institutions of using datasubmissions as "vindicators" of'their programs rather thanas_indicators of program performance.

In most states, public colleges and universities aretrying to increase tntair credibility with state budgetreview agencies by provi,2'...ng more information, more com-parable information, and information of better quality.Institutions rarely volunteer information, however, therebeing considerable concern that some item of data pro-vided will become the crucial piece of information that

.

-will-turn a budget decision against them.- Although trying to remain c.00perative, institutions and multicampussystems are clearly not fully committed to centralizedinformation and analysis systems because they do notwant to legitimize additional planning, analysis, anddecisionmaking responsibilities in the other agencies.Internal institutional "score keeping" that might atsome point be used against the institution externallyis often not done.

Greater use of information and analysis is oftenlinked to more complex methodologies for justifyingbudgets. The more autonomous the institutions, the lesslikely that complex information will depict actualinstitutional plans and operations; even with relativelycomplex and detailed information, little control isachieved under theae circumstances. Where possible,technical factors will always be used to institutionaladvantage in making budgetary requests. The end resultis often a justification which is more difficult tocomprehend and review, with no additional real under-standing of institutional operations for the purposes ofplanning or state-level coordination. Meisinger (1976)has,noted both the susceptibility of compli-cated budgetarymethodologies to manipulation, and the eventual loss ofcredibility in some circumstances when requests havebeen formulated with complex and highly structuredmethods. Where there is less institutional autonomy, as

159

17 3

in state colleges, the prospects may be better that moreexacting and complicated budget methodologies will pro-vide mutual benefits for both state agencies and insti-tutions.

Topics of informational concern vary a great deal_from_state to state, but the problem of providing infor-mation in a variety of formats and aggregations is onethat is generally shared. Executive budget offices areusually interested in aggregates bailt around state-levelprogrammatic considerations. Legislatures and theirstaffs have a predictably much greater interest in theorganizational units to which appropriations are actuallymade. Where these norms of executive and legislative re-sponsibility apply, there is a clear conflict between-executive and legislative interests. Executive agenciesmust also be comprehensive in their treatment of thebudget, particularly in states with strong governors.Where the legislative role consists more of checking thaninitiating, the staff can restrict its view of the budgetto specific details and may not require comprehensivedetail.

INFORMATION POWER

One state-level budgetary function which virtuallyall executive and legislative agencies are willing toassign to state higher education agencies is that ofresponsibility for state-level information gathering.However, by itself statutory authority to collect infor-mation is no guarantee that the information gatheredwill be both valid and comparable enough to be used inplanning and budgeting. An agency must "spend" con-siderable influence and political goodwill with theinstitutions to assure their cooperation. Gettinginformation is usually a manifestation of already exist-ing power relationships; top officials are more likelyto get data they request than staff analysts. Thedevelopment of information and analysis is given a fairlyhigh priority by state-level agencies in most states,but we found numerous examples of.system developmentdelayed by lack of funds from the state or by a conflict

1.7.1160

with priorities at the campuses. Not surprisingly, somestate higher education agencies have been unable to ful-fill this information gathering function to the satis-faction of the other state-level agencies.

Information is power, but the tendency to treat infor-mation as a free resource overlooks the fact that in thestate_budgeting process information is not always dis-seminated or accepted when it might remove comparativeadvantages in negotiating budgetary allocations. -Not onlyis information not a "free good" in ,the economic sense,because it costs something in both physical and politicalresources to acquire it, but its acquisition and use arerestricted by constitutions, laws, and customs. Thus,"perfect information" as the standard of informationavailability for the exchange of private goods in a com-petitive-market-cannot be eXpected to apply to the ex-change of budgetary information. .

17.5161

Information and Analysis Systems:Considerations for Design

The previous chapters of this report have described

the actual development of information and analysis systemswhich support higher education budgeting in 17 states.These operational alternatives in themselves suggestconsiderations for design, but it is now time to take asomewhat more prescriptive stance and discuss the direc-

tion which changes in the design of information and anal-

ysis systems should take if they are to be effective.

We shall do this by posing a series of considerations

which, depending on the characteristics of a given state's

process, may have vastly different implications for state

practice. We believe that awareness of limitations and

alternatives, and of the underlying nature of data analy-

sis itself will result in better procedures for analysis

and data collection. Although several of our recommend-

ations do appear to be broa.dly applicable, they aresecondary to the broader understanding of the limitations

and alternatives.

THE NATURE OF INTELLIGENCE FAILUREIN THE CURRENT SYSTEM

An interest in changing the present level of avail-

able information and analysis in the budget process can

be taken to signify, at least implicitly, that present

arrangements are inadequate in some respect. Criticism

is frequently expressed, most often in general terms,

that senior state-level and institutional administrators

do not have the information they need, and that existing

162

17:3

information systems either compress and distort informa-tion, reducing its quality, or fail to filter out ir-relevancies, producing information overload. If theseinformation deficiencies exist, there must be consequences:Either the choices stemming from educational and adminis-trative procedures based on this deficient informationare inefficieht in an economic sense, or the choices ofeducational outcomes that result from budgetary decisions,again based on deficient information, are somehow politi-cally or'socially inappropriate. It is presumed thateither efficiency or educational outputs, or both, wouldbe improved by changes in information availability andthe extent of analysis. But the path to improvement isnot wholly apparent because the exact nature of the in-telligence failure-is not generally clearly specified.

On the basis of our investigations in a variety ofstates, we would not argue that intelligence failures donot exist. Currently, there is some general dissatis-faction with the outputs and activities of higher educa-tion institutions with respect to the inadequate'vota---tional preparedness of college graduates, the emphasis putby some faculty on research activitieg rather than in-struction, and to some extent with the attention givento the highly qualified full-time student to the possibledetriment of the less-qualified part-time student who mayrequire new avenues for participation in higher education.There is also some indication that fluctuations in enroll-ment have not been anticipated by institutions of highereducation. Certainly there is implicit evidence thatchanges in labor market demands for particular kinds ofprofessional training have not been accuiately anticipatedeither by institutions or students. However, if currentinformation and analysis procedures are to be changed,it is essential to determine specifically how the currentsystem is failing, and if it can be corrected by theproposed changes.

The aspect of the intelligence failure, if it canbe called that, that most concerns those who would alterinformation and analysis systems in higher education,is the extent to which budgetary choices can be accountedfor with explicit rational and supporting information.

163

177

To some degree, the increasing intercst in requiring aminimum level of decision accountability as part of the

---b-u-dge-tarTalid--O-Er1-6-i----adininistrative processes is sympto-

matic of a loss of trust in highei.-.6ducation faculty andadministration. State agencies generally have consider-ably less post hoc control over the activities (e.g.,budget e-.ecution) of higher education institutions than-they possess over the activities of other state agencies.Consequently, changes in information reporting to thestates are almost uniformly a substitution of ex antecontrol of higher education appropriations through thebudgetary planning process for the post hoc controlavailable for other state bureaus. Increasingly, somemechanism for insuring this same level of formal decisionaccountability is becoming a requirement for an accept-able and legitimate budgetary process. The logic behindsuch a requirement is increasingly difficult to faultunder a financing mechanism in which public higher educa-tion raises a majority of its operating funds through thecentralized taxing power of the state. Ashby (1971) hasexplained the logic of funding policies and institutionalaccountability in terms of alternative patterns of univer-sity responsibility to students, employers of graduates,the public-at-large, or the universities' own guildtraditions. Clearly, the current interest in account-aoility arises from a divergence of views as to wherethis dominant responsibility lies.

The designed flexibility of funding formulas andother funding methodologies which have served to controlthe level of higher education expenditures in the past,and provided institutions with operating latitudesufficient to adjust to changing demands, is indicativeof a tradition in which emphasis has been placed on theuniversity's unique capacity to judge its own performance.An analogy has been Jrawn between managing an operacompany and managing a university which illustrates therelationship between administration and faculty and theformer's restricted ability to manage and be responsiblefor instrUctional performance. As with decisions con-cerning magical interpretation and performance in anopera company, decisions concerning the educationalprocess are highly decentralized and in the hands of

173164

a relatively autonomous faculty. The measure of manage-ment, therefore, is likely to be its ability to provideand manage resources and to foster work conditions condu-cive to high morale and creativity. Although this wasnot an issue on which data were collected extensively inthis study, it appears that the most obvious example ofintelligence failure in public colleges and universitiesconcerning planning and budgetary matters is the limitedflow of information from state and institutional adminis-trators to the faculty. Before overhauling funding arrangr?-ments and supporting information systems, it is essentialto identify current educational and information deficienciesand determine that changes will tend to foster improvementsrather than exacerbate present shortComings. If it is anadaptable and innovating system of higher education thatis desired, budget models and information systems shouldbe sought which allow for adjustment and diversity rather .than those which foster bureaucratic routine and theentrenchment of procedures.

MANAGEMENT INFORMATION SYSTEMS AS PARTOF THE TOTAL MANAGEMENT SYSTEM

The design and implementation of information andal4..14ysis -systems-sometimes-fail-to TevogTrize 'that-current--information ahd analysis procedures are.part of an existingmanagement system which determines how decisions are madeand who makes them within the organization. Furthermore,

. centralized information systems carry with them a model--usually implicit--of how managerial decisions are made'or how they ought to be made. With due allowance for thefact that procedurally defined information flows anddecision responsibilities are likely to be simpler thanactual flows and responsibilities, a management informa-tion system is not likely to be implemented as intendedif the informat.on system and the management system arefundamentally inconsistent. In practice, either the in-formation system is altered to neutralize those character-istics which oppose the existing management system, or tothe extent that an information system can be imposed, themanagement system is altered to reflect the influence ofinformation flows and analytical resources. Changes in

-

165

179

information flow aryl access are not neutral with respectto organizational power and influence.

Formal information sysiems are designed with a parti-cular concept of managerial functions and activities inmind. Two divergent theoretical conceptions of whatmanagers do illustrate how conceptions of management re-late to the kinds of management information sy.,,tetw, tuatwill appear useful. One of the first models formulatedis exemplified by a model of public management whichassigns planning, organizing, staffing, directing, coordi-nating, reporting, and budgeting (POSDCORB) functions toadministration. ,The distinguishing feature of this modelis.its representation of managerial activity in terms of,a set of formalized normative principles and procedures,a feature it shares with other models of management whichemphasize the unity of decision responsibility. Modelsof management which are based on the manager as rationalutility maximizer have been used successfully in develop-ing testable hypotheses about business firms and othexorgahizations with no normative implications. However,

as behavioral refinements have been introduced to makethese models conform somewhat more closely with whatmanagers are actually observed to be doing, a tendencyhas developed to make the underlying notion of managerschoosing goals and acting purposiVely to db-hieve thesegoals a prescription for as well as a description ofmanagement. Management information systems designed withthis conception of management will be designed for managerswho define problems, seek alternative solutions to solvethe problems, and make decisions on courses of actionwhich will implement the solutions.

In searching for a more realistic description, .critics of the POSDCORB model argue that managers cannotfunction and make decisions in the way that the modelimplies because organizational goals are often vague,ambiguous, and contradictory. In addition, a singularlybest method for achieving any given objective very,rarelyexists; knowledge, responsibility, and authority cannotbe consolidated sufficiently in large and complicatedorganizations to make this possible. Contrasting viewsof what top management actually does have been offered

13)166

by Braybrooke and Lindblom (1963), among others, whichemphasize the remedial and fragmentary nature of managerialaction. The popular literature on management also hasproduced many examples which stress the unstructured,nonquantitative, and time-constrained features of managerial

'work,'all of which limit the application of the rational-analytical problem solving ideal.

Cohen, March, and Olsen (1972) went far toward bring-ing together theoretical models of management and populardescription in their "garbage can model of organizationalchoice." They hypothesized an organization characterizedby the lack of preference structures which satisfy thestandard consistency requirements of the conventionaltheory of choice, an unclear technological relationshipbetween organizational inputs and outcomes, and irregularand fluid participation. In such an "organized anarchy,"triey claimed, at least a partial uncoupling of problemsand choices takes place.

Such organizations can be viewed for somepurposes as collections of choices lookingfor problems, issues and feelings lookingfor decision situations in which they mightbe aired, solutions looking for issues towhich they might be an answex,_and_clecisionr__makers looking for work. (p. 1)

Although all organizations from time to time mayexhibit such characteristics, Cohen et al. asserted theyare particularly conspicuous in public, educational, andillegal organizations. They did not prescribe this kindof organization as one especially suited for resolvingproblems, for that it certainly is not. On the otherhand, there may be certain organizations in which thepreconditions for a garbage.can process cannot be elimi-nated, and in fact should not be eliminated. Their modelemphasizes the ambiguity in group decision stimuli moregraphically han most and implies the need for a differentmanagement information system. It also suggests why astate higher.education agency, coordinating as it doesa set of institutions which are well described, to someextent, as organized anarchies--Ashby (1971) has called .

167

4 81

them benevolent and useful anarchies--encounters difficul-ties in gathering information that one might assume wouldbe available to it. Part of the explanation may lie inthe fact that the orgariizations involved do not have ahierarchical and functionally articulated organization.

Actual organizations and management systems are muchmore complex than these two models, neither of which treatsthe implications.of multiple layers of management encoun-tered in-public higher education nor the division of laborbetween political officials and agency staff in the publicsector. However, these models do suggest a contrastbetween the information needs of top officials, which aremore likely to correspond to the current intelligence needsof the "garbage can model," and of agency staff whose workis less time-constrained, although more structured, andwho therefore may make use of comprehensive, historicaldata.

Most, if not all, information systems are designedby staff working from a set of assumptions very much likethose of the rational problem solving.ideal. Consequently,the comprehensive and partially automated informationsystems designed primarily to meet the needs of staffperforming routine budget review activities of checking

-and-costing-are-not-like ly-to-provide-the-inforpatio.nneeds of top management, which are more likely to berelated to the evaluative functions of budgeting. Al-though an increasing number of attempts are being madeto provide evaluative information by squeezing it intocategories and'formats that can accompany the formal budg-etary requests, this information tends to be unsuited tothe needs of top officials. The evallative and strategicplanning decisions which these officials make are morelikely to require information that deals with programsexternal to the one in question, that is, the relation-ship of the program under consideration to the total arrayof state programs and to choices and opportunities aboutwhich there is as yet little hard or historical data.

168

DATA REQUIRED FOR PLANNING VERSUS CONTROL

Even though the data required for planning and thatfor control are.fundamentally different, attempts to changethe informational content of the budget request processhave been directed simultaneously at improving both ofthese management functions. Program budgeting reformshav- attempted to instill the budget process with planningactivities by reviring information on institutional goals,program missions and performance, and alternative programactivities. . At the same Lime, changes in budgetary infor-mation have attempted to tighten the points of control byeither increasing the number of appropriation categoriesor shifting their focus from agency inputs to activities.

Planning, in the sense of strategic planning /annoychoice, requires tangible detail about where the organiza-tion "is" currently and, at the same time, informationabout trends and selected "trigger" items that infcrmofficials about expected developments and problems.Aggregates may have some limited use, but are needed lessthan details.which sharpen understanding of future trendsand their implications for current operations. Historicaldata,. except those which help to identify future trendsand indicate current status, are of evem'more limited

than budgeting and tends to emphasize the merits of pro-grams to the,axclusion of costs. It requires data on thesubstantive features of programs across the full rangeof state activities more than it requires cost data abouta single program. Evaluative data of this kind are likelyto be highly speculative, and may be more easily communi-cated orally than in writing. Buigetary planning placesmuch more explicit emphasis on costing out programs, andthus can rely on data pertaining to single programs.

The data important for budgetary control need notnecessarily be more detailed in the sense of being disaggre-gated, however; what is critical is that control of in- .

formation be concrete rather than conceptually ambiguous.The data must relate to actual lines of authority andresponsibility, which means they must pertain to existingunits of organization rather than, as they often are in

19

Fi

program classifications, to hypothetical ones. Further-.more, control data must describe or pertain to actualoperating transactions which the agency or'institutionuses in its own administration.' Thesacontrol data arevery likely to be concrete items of expenditure or aggre-gates of these that the institutions make in the processof routine operation. Control may also_be maintained byrelating institutional activities to a specific set ofeducational outputs agaiinst which institutional performancecan be measured, in the same sense that control in a profit-making enterprise is maintained in terms of cost centersand profit-generating activities. The transactional datasystems of profitmaking organizations produce informat .1nwhich is directly relevant to Organizational objectiveLbecause the units of measurement for goals and activitie:-are either the same or can be suitably converted. There-

fore, control exercised through trrinsctional data alsoserves.to control organizational ci. ttainment, even

though, for reasons given above, tactional data willnot be very effective in strategic pla7 'ing. The goals

of higher education institutions, howe:-: are not onlyconsiderably more ambiguous because the; cr.not be con-verted to common units of measurement, but :.keir attain-ment-is also farther removed from the C:iy-to-day trans-actions within the institutions--especLa:'l y. from its

administ'rati-ve-,--purcha'sing.,--and-personnel- transaa-ions-.--This effectively decouples the transactional data system,for which there is relatively extensive data, from in-formation on educational outcomes as well as from otherdata that would be relevant to strategic planning.

Attempts to determine the emphasis on control versusplanning in state budgeting, although imprecise, usuallyfind a heavy emphasis on control (Friedman, 1975; Schick,1971). The costs, in the broad sense of political costs,of being deficient in cont:olling the mismanagement offunds are yery great; they re far greater, evidently,than the benefits that accrue from providing a precisely"correct" menu cervices as a result of strategic'

planning and su..:ssful meeting of contingericies. The

iisk, again in ,he political sense, is greeter in allyi-ing flexibility fr the futlire than it is in maintainingcontrol for the present. The economic costs of control

18170

are also lower than the costs of strategic planning.Whatever the reasons, state agencies with responsibilityfor planning and oversight do emphasize the latter. How-ever, institutions are likely to realize more flexibilityunder a system which emphasizes traditional object ofexpenditure review and allotment controls than they areunder one in which state agencies do more planr:_ng.Aggregated object of expenditure allotments leave morediscretion to institutions in terms of the purpose ofexpenditures than do functional, program, or output class-ifications which specify expenditure objectives moreprecisely.

Thus, in the budget request process, there is notonly a significant difference in the characterfstics ofinformation required for the two different functions ofcontrol and planning, but also a tendency for stateagencies to concentrate on one function to the exclusionof the other. At the same time, those who support moreplanning attempt to impress it on the ioutines of thebudget process by including data that relate to educa-tional outcomes and other program and policy considerations.Data for the planning process and the analysis that mightprove useful tend not to be routine, however, and thesetherefore come into direct conflict with the rigidity,

---tight-time-seheduler-inforffisitim- formats-, -- detail of the budget process.

To separate planning from the budget process entirelyis ill-advised. Indeed, we urge their ihtegration, forinfluence in the control process is the princjal meansthrough which influence over implemention is gain-N.-L.Nonetheless, the routines of the budget process effectivelydrive out the opportunity for the more speculative insightand information needed for the planning activity to havean effect. Because planning appears to be dc facto re-moved from the budget process, which is to say, itS-in7fluence and impact on budgeting cannot bc--or a. ieasthas not as yet been--formalized or routinized, seriousconsideration should be given to conscious and deliberate,although less formal and routine, ways for rTanning tohave an impact on budgetary decisions. In many instances,the planning detail included in budget requests. ap theyare now reviewed, is virtually of no consequence.

171

18

Effective techniques for bridging this gap betweencontrol and planning will be very dependent on the statecontext and on the quality and sophistication of theagency staffs. There obviously will be great variatonfrom state to state in the extent to which it is deemedpolitically advisable for state agencies to improve theirplanning capability. However, coordinating agencies mayfind it to their advantage not to have responsibilityfor detailed budgetary review so that they can concentrateon the strategic planning process, with its very differentinformational demanas. Likewise, executive and legisla-tive budget agencies might find it a better diVision oflabor if they concentrate on resource and fiduciary con-trol and reduce their involvement in comprehensive pro-gram and policy review.

DIVISION OF LABOR IN BUDGET REVIEW

Other reports in this series have treated the questionof effective differentiation of budgetary review and re-sponsibility (Glenny, 1976; Schmidtlein & Glenny, in pre-paration). However, the present discussion of factors tobe considered in changing information and analysis pro-cedures cannot be complete without a reference to the-implicAtions of--agency '&1p-1i-cation -of-effortand-lack -of

coordination. Clearly, a sharp delineation of responsi-bilities, both vertidally between state agencies andinstitutions and horizontally among state agencies, wouldsimplify 4nd'rationalize the budgetary process,-.reduceduplication and overlap of responsibility, and raise thelevel of coordination. The difficUlty lies in the lackof criteria for making such a division. Various generalnormative principles distinguish, in theory, among orbetween the functions of elected officials and the bureaus(policy and administration) and between the executi.ve andthe legislature (the executive proposes; the legislaturedisposes), but these prove to be far too coarse to dis-criminate among the various state agency and institutionalstaff functions in higher education budgeting. In abargaining context, unless one agency is widely presumedto be a source of objective and independent analysis andinformation, each agency will rely on its own own analyticand informational resources to some degree.

172,

1 E '3

In studying the overlap of state agency reviewactivities, we found, generally, that at most, one stateagency does a really thorough review of higher educationbudget requests, that is, a review that is both compre-hensive and goes into considerable.detail. Where thestate higher education agency develops the budget request,either the executive or legislative budget agency conductsa thorough review.. Depending on a s:ate's statutorydelegation of responsibility and the histor:.:-al role ofthe.legislature, this agency may be the state highereducation agency; the executive budget offine, or thelegislative budget agency. In most cases, the otheragencies take a reactive stance and either do little orno technical review .or merely check the recommendationsof the reviewing.agency. most states have neither theagency sources nor the time to concern themselves withoriginal budget submissions in detail in addition to con-ducting a further review of subsequent-recomMendations.One method for reducing the complekity of reiewing anentire submission is to focus on specific issues. Thisalternative review approach, patterned atter that of theCalifornia Legislative Analyst, can go considerably beyondmerely "checking" an earlier, more or less comprehensivereview, and is the approach newly established staffsoften attempt.. Such multiple reviews, to the extent they .

-do --teke--pl-ac e-, -usual ly- additional in-f ormatdemands on the institutions. To the extent that thisduplication of effN,F,t at the state level becomes a polit-ical problem, that is, an intolerable manifestation ofagencies jockeying for competitive advantage, it must besettled through political means. To the extent that itis a technological problem, it may be moderated by theprudent use of suitably designed automated data systems.Rarely does it appear that the information reporting load,even though it might be more suitably coordinated, isthe true source of interagency friction. The plea of an"intolerable load" may be just, but it is also almostalways part of the arsenal with which this conflict isconducted.: The source of the conflict iz more likelyto be quite basic disagreement over fundamental respcnsi-bilities in state government or the governanc t. ofhigher education.

187173

Even in the absence of basic disagreement, a probableproblem is a lack of communication between agencies whichperform a detailed review and those that subsequentlycheck or test these recommendations. These latter agenciesare sometimes thwarted in their desire to understand thebasis of recommendations. Once an agency performs a de---tailed review and reaches specific recommendations, it ismore likely to place its efforts in justificatdon of therecommendations than in opening them for. Analysis.

The vertical division of labor is the heart f theinstitutional autonomy question, and the legitimatedresponsibilities in budget review have been determined,like the division of responsibility between executive andlegislative bodies, through constitutional action, thepassage.of legislation, and evolutionary development.There is clearly no single optimum cc figuration, butrather numerous alternatives which evolved from theparticular set of conditions existing in each state.Many of these have been discussed in Glenny (1972) andSchMidtlein and Glenny (ih preparation). And as we pointedout in Chapter 5, the extent to which institutions orsystems provide Institutional detail to state agenciesvaries considerably. What is acceptable by all partiesconcerned in one state would be anathema to the partici-

-pAnts in-another. --

Two criteria for providing access to information,which have been offered elsewhere, have shortcomings asprecise guides to the provision of information, but theymay serve some purpo-se as guidelines by establishing a .

point of reference. Unfortunately, they may run cotinterto the spirit of "sunshine" laws and freedom of informa-

.

tion legislation. The first of these is the analog,

of

the !'need to know" criterion, which,is applied in accessto classified materials. Access to information is limitedto information to be used in making decisions. Onlyinformation with a specified use, therefore, would beprovided to those with legitimate responsibility forusing it. Because the decision process, as we have seen,is so complex, this is not an easy principle to apply.The formalist model of problems, solutions, constraints,and information impinging on individual choicemakers is

174

1.3E3

clearly an oversimplificaticn of group decisi.nnmaking withmultiple choicemakers in a bargaining context. Nonethe-less, the current assumption--that the public is best servedby having all information regarding agency choices madeavailable to anyone who is interested--must be challenged.A distinction should be maintained between informationcommunicated formally in public documents and public hear-ings, and that exchanged in private communications. Forexamp)e, an institutional chief executive officer statinginstitutional priorities for budgetary alternatives ina public document will undoubtedly express them somewhatdifferently when he makes the same priorities knownprivately tO a governor or legislative committee chairman.We all know that some of the best evaluative informationcomes from the candid and confidential impressions ofknowledgeable individuals. Attempts to make this kind ofevaluative information public can destroy its value orbring undue attention to those concerned by it. Inattempting to do away With "politics" in public choice,such reforms merely ignore it.

The second criterion is that data for budgetaryplanning should Only pertain'to linkages between resourcesand inputs for the educational.process, and should avoiaroutinized consideration of.the hypothetical outputs of.

. __the _p_rac.ess . aormactions-b.etween-r-esourc-e-use-,---inputs-,

and educational outcomes are certainly appropriate plan-ning concerns over the long run, but they are, accordingto this criterion; unsuitable annual or biennial budgetaryissues. The purpose served by this criterion is theavoidance of locking-in production relationships to whatare suiAle and unquantifiable educational outcomes% Thisprinciple is rather generally applied in budgetary practice;although budgeters may sometimes ne convinced taat theyreally are measuring educational outputs in student credithours. Student credit hours are truly only surrogatemeasures of educational outcomes, albeit some would feel"that even their use in budgeting determines and confinesthe educational process too severely. It is certainlytrue that innovative nontraditional programs can only befunded if this principle is applied because such programsdo not measure student progress by traditional measures.

175

1S3

Formula budgeting has proven to be a workable solu-tion for funding according to the principle of input andresource linkage. But it has run into difficulties whenthe inherent flexibility in the procedure has been abusedby institutions or state agencies which have tended toview formula allocations as detailed work plans. Con-strained rev_nues and falling enrollments have also proveddifficult to handle within the formula framework (Meisinger,1976). As information reporting becomes more detailed anddisaggregated, state agencies and institutions logicallyseek the mutual accommodation and flexibility providedby a formula: Unfortunately, in the context of levelingenrollments, the application of simple formulas is'un-------satisfactory, a state of affairs complicated by the factthat the level of interagency trust is often below thatrequired for the use of complicated formulas.

COMPUTERS IN MANAGEMENT INFORMATION SYSTEMS

'New institutional information requirements largely .arise from increased demands made by state and federalagencies; they are not merely a response to the ease withwhich new technology makes it possible to secure andprocess information. Still, the identification of infor-

---mation-systerns---igi--th-the- 'autmated or computer.--j---- --systems does suggest that a great deal of the interest inkmplementing the=e systems in higher education is stimu-lated by the prospect of a successful transfer of tech-nology from industry to the domain of social policy.This orientation is so strong, and the technology of in-formation systems so much more advanced lhan our under-standing of public management itself, that initial effortsto meet new information demands are often focused on hard-ware and software systems rather than on information.This has obvious advantages as an implementation strategybecause the agreement on information system technologyexceeds that on the informational content of informationsystems. But this approach is also undoubtedly respons-ible for a large-fraction of the mismatch between expect-ations from information systems and what is actuallyachieved. A key issue in altering budgetary information,therefore, is the role of automation in these systems and

176

the extent to which it contributes to the success of thesystem. We have already discussed automating the budgetreview process itself. Here we refer to the more compre-hensive activity of automating the entire budgetary infor-mation support system.

In many instances, the large quantity of data thatis called on to support the budgetary process makes auto-mated systems attractive, if not essential. Unfortunately,these systems of themselves do little to improve dataquality; when systems merely report information that hasbeen untried in an analysis, the principle that unuseddata is usually "dirty!' data operates with full force.Only when informationas used in an analysis is there thelikelihood that its deficiencies will be brought to light.A good example of this is the process of improving theHEGIS data, which is now underway as a result of use madeof these'data by the staff of the National Commission onFinancing Postsecondary Education and at the NationalCenter for Higher Education Management Systems (McCoyet al., 1976). Use of the HEGIS data has depended on theapplication of computerized analysis, but to a large extenttheir use in a policy context awaits further improvementpl timeliness and quality of the data.

Automated systems and manual systems also havedifferent capacities to cope with the so-called "80-20rule," which dictates that large portions of a data basewill receive very limited use. (Conversely, relativelyfew items in a data base 4tp very heavy use.) Automatedsystems obviously handle much larger quantities of datawith ease, but they are less able to filter and winnowthat data for eVentual presentation in a policymakingcontext.

Experience with the application of automated dataprocessing in the corporate world suggests the adminis-trative activities in the public sector most likely.tobenefit from technolonr transfer. For this purpose, itis useful to think in terms of levels of increasingsophistication in the application of computerizedsystems. .

191177

Leve/ 1. Basic accounting, sales analysis, andautomation of routine clerical tasks.

Level 2. Partial integration of separAte subsystemsand data bases with few automatic decisionrules.

Level 3. Inventory control and production scheduling,formal'optimization techniques, and sequen-tial linear programming solutions.

Level 4. Simulations of corporate strategy, studiesof-budgeting alternatives, and plantlocation. -

Analogous applications are readily founiT in higher educa-tion management, although there are fewer piocesses whichcorrespond to Level 3 unless one equates the more hypo-thetical flows of the instructional prOcess. Classroomscheduling at the institutional level is one adtivity inwhich computerized systems have been usefully applied.The extent of implementation of computerized systems atthe various levels described is undoubtedly changing, butonly at Levels 1 and 2 have there been very extensiveimplementation in industry. Level 3 is much more rarely __-a-dbleVed, and- Lével'4 is only at an experimental stage.

, ,

Using Anthony's (1965) classification of kinds ofplanning and control systems (strategic management,

managerial control, and operational control)- it is fairlyclear that, because .of computers' capabilities for per-forming a high volume of well-structured operations ina context of few interdependencies, computers will beused most in operational management and control, and leastin strategic planning. Their greatest successes asdecision-aiding systems have come in-operational controlsituatiops involving logistics, schedules of projectcompletion, and budget execution. Computerization isless valuable in managerial control or strategic manage-ment because of the nonroutine and less highly structurednature of the activities that must be carried cut inperforming .0ese functions..

178

91. 7)

In one instance of higher education budget management,the use of an automated information system at the institu-tional operating level provided up-to-date information onlevels of expenditure, and threby helped that institutionspend virtually all of its state appropriation that hadbeen subject to complicated allotment transfer and salarylapse controls. This graphically illustrates that it isat th', operating level that management needs quick infor-mation retrieval. The higher the organizational manao.,2-ment level, the less the need for instantaneous informationretrieval and, consequently, the less the need for auto-mated systems. Trend data for planning based on dumo-graphic, economic, and social characteristics can I:0.,

maintained to advantage in a data base automated for 1%-,--

trieval by management.

A final consideration in the use of automated datasystems is cost. Currently it is the costs of software--the developing of computer programs for data base manage-ment and analysis--that are critical, rather than thecosts of computer hardware. Computer capacity is quitelikely in oversupply, but the costs of designing softwarepackages to handle unique data bases can extend beyondthe resources that most state budget agencies can bringto bear. We found no cases where these costs had beencarefully documented.

MODES OF INTRODUCING INQUIRY AND ANALYSISINTO THE BUDGET PROCESS

Greater analytic capability in the budget processis sought principally through the introduction of infor-mation reforms, new procedures, and the addition of staff.Viable alternatives in information reform and new pro-cedures can be determined by considering the range ofpractices described in Chapters 4 and 5. In initiatingor augmenting analytic staff capacity, the principalconcerns and considerations are the character of thestaff as either "neutral professionals" or policy-advocates,the skills and expertise of staff analysts, the use ofspecial studies as a substitute for enlarging centralizeddata systems, the use of program evaluation units, andthe use of outside consultants.

179

1. 9 3

Information and analysis systems can be thought ofas attempts to apply so-called scientific methods to theproceSs of organizational choice. An examination ofalternative theories of knowledge may therefore provideadditional insights to the organization of the "inquiry"function of the budget process. For an introduction tothis approach, see the short summary on alternative formsof policy analysis by Mitroff and Pondy (1974). Any of

the work of C. West Churchman, the originator of thisapproach to operations research, might also be consulted.

Staff usually can be expected to adopt a mind-setthat corresponds to the role of their agency in the budgetprocess. Thus, one does not expect the executive budgetoffice staff to take a stance as neutral professionalsin developing and formulating analytical support for thegovernor's budget policies. Nor does one expect thisneutral stance in the staff of a statewide governingboard. Legislative staff must often adhere slavishly tothe sentiments and inclinations of the committee membersthey serve, and coordinating agency staff, particularlythose of agencies with formal budget authority, are oftenin the no-man's-land between institutional interests andthe interests of the executive or legislative branches.Although they can ill afford to be more objective, of allthe state-level agencies the staff of the coordinatingagency may have to deal with the broadest range ofpolicy alternatives.

Although no state agency staff has resources foranalysis that equal those of the U. S. Congress, theCongressional Reference Service provides a good exampleof a staff serving primarily as an analytical resourcefor policymakers. The service is to provide "factffalinformation, analysis of issues, alternatives to pro-posals, and evaluation of alternatives without eitheradvocacy or partisan bias" (Beckman, 1975, p. 403). In

addition to its policy analysis, the CongressionalReference Service is formally charged to provide foreach House, Senate, and Joint Committee at the openingof a new Congress a list of subjects and policy areaswhich the committee might profitably analyze in depth;it submits to the appropriate committees of Congress

19180

lists of programs and activities which are scheduled toterminate during that session of Congress; and it main-.tains continuous liaison with all congressional committees.At the state level, this function is usually carried outby a legislative reference service or counsel, but exceptfor those states in which the legislative budget committeeshave no permanent staff, these counsel staffs have arelatively minor role in budgetary analysis.

The newly established Congressional Budget Officewill attempt to provide neutral analyses of budgetaryissues as the Congressional Reference Service does onpolicy matters. According to staff director Alice Rivlin,the Congressional Budget Office will not be taking posi-tions or making recommendations; its function is to giveCongress better information about budget choices. Thisit must do within the framework set by the requirementthat on each April 1 the director of the CongressionalBudget Office report to the Budget Committees on fiscalpolicy for the coming year with:

a discussion of national budget priorities,including alternative ways of allocatingbudget authority and budget outlays foreach fiscal year among major programs orfunctional categories, takinc into accounthow such alternative allocations will meetmajor national needs and affect balancedgrowth and development of the UnitedStates. (Congressional Budget and Impound-ment Act of 1974, 88 Stat. 297)

We found legislative budget staffs in the states tc,

be taking somewhat more assertive roles in making recom-mendations than these formal statements would indicateof the staffs serving the U. S. Congress. Of course,the congressional staffs which are comparable to thelegislative budget staffs in the states are those attachedto the appropriation committees of the Congress. Thesestaff are more numerous than the new staffs mentionedabove, and they are, in addition, extremely aggressivein proposing budgetary recommendations. Even thoughmany legislative staffers in the states thought of them-

181

1 9 5

selves as ihdependent professionals, they are expectednot only to avoid interjecting their own biases in orderto retain the confidence of the policy officials theyserve, but also to serve and justify the biases of thelegislators for whom they work.

Without abandoning plans for eventual developmentof centralized data systems, some state budget partici-pants have found that improved programmatic review isfacilitated more by ad hoc and nonperiodic special ana-lytic studies than by large-scale reorganization orcomprehensive collection of data. Some of these specialstudies, such as cost studies or surveys of faculty work-load, do become routine. The special analytic study cantake advantage of bursts of energy and interest in aparticular issue to mobilize a research effort, whileattempts to design comprehensive information systems cantake many years and still not include the data that maybe needed for researching an issue of importance forbudgetary planning. Frequently the very nature of thesestudies--for example, the necessity for analyzing apractice that is not being carried out currently--pre-cludes the use of routinely available comprehensive data.

Program evaluation units in both the executive andlegislative branclkts are also sources of planning infor-

mation and expertise. Their evaluations, like the specialanalytic studies, depend to a great extent for theirimpact on timing. To influence the outcomes of the budgetprocess, studies must be abstracted or condensed andchanneled to those who review current budget submissions-arid make decisions about them. A few states havefound------it possible to sensitize budget examiners to the lessroutine and less atructured conclusions of these evalua-tions and studies by encouraging mobility between budgetreview and program evaluation units. It is much more

common, however, for the program evaluation unit to be

somewhat isolated, and thus less able to function effect-ively as an adjunct to the budget process.

Two factors are paramount in reducing the effective-ness of these evaluation units, and they should be con-

sidered in establishing a role for them. The first of

182

19 3

these is the more easily overcome; for the latter, theremay be no effective solution. Staff for these evaluationunits predominantly have research backgrounds and as a rule,limited practical experience. In itself, this,is not a

--S1ertebitiii4,-fer-there MUSt-be-SOMe'diViSiOn Of-labbi inthe budget process. However, the staffs' heavy researchorientation tends to result in analyses and documentationthat are more appropriate for academic reseazch. Notinfrequently these staffers, like faculty at universities,are looking to their own academic discipline or professionas the arbiters of taste and style rather than to the stateofficials their analysis is intended to serve. Policyanalysis and evaluation that can have an impact tend tobe a lot more like investigative journalism than likeacademic research. They are likely to be synoptic andeclectic rather than highly focused and original. Theymust also be sensitive to the requirement that buildingsupport for policies is a crucial part of designing orimproving them.

Secondly, the essence of evaluation is preference,and preferences differ both among individuals and amonggroups who have an interest in policy outcomes. Wildavsky(1972) captured the intractability of the evaluationproblem when he concluded that evaluation and organiza-tion may be contradictory terms. Exhibit 11 below suggestssome of the dimensions along which this contradictionexists. This pattern of contradiction makes it clear whyorganizational tension is a normal accompaniment of aproperly conducted evaluation. One presumes that evenWildaYskhowever,_evaluates his graduate studentswell as the work of assistant professors seeking tenurein his department. Evaluation may well be contradictoryto organization, but evaluation is also an essentialingredient of choice.

Evaluation efforts can make the implication:.past,or pending choices more explicit by definim:problems to which pUblic higher education addreYsr.L... it-self and noting how these problems are distribu. ,

Evaluation also can identify who is helped by spl.cificprograms and how much, examine which programs do themost good, and probe how different kinds of social

183

1 9

Exhibit 11

CONTRADICTORY ELEMENTS IN ORGANIZATIONAND EVALUATION

Dimensions Organization Evaluation

Basis for action Consensus Reason,analysis

Areas affected by Programs, Goals,

organizational action clienteles objectives

Implications for Stability Change

structure

Source of values Participants Philosophy,

Desired staff values

ideology

Commitment Skepticism

19 3

184

services can be produced most effectively (Rivlin, 1971).These analyses go well beyond the time and resources avail-able for annual and biennial budgetary analysis, but itseems clear that formal evaluation should be one of theelements of the total process that determines which socialprograms are attempted, which are abandoned, and whichare changed. However, no one should expect that evalua-tion alone will determine the choices which are made.The extent to which state agencies can induce institutionsand systems toward self-evaluation may determine theeventual effectiveness of these evaluations.

Formal evaluation efforts by state agencies haveonly recently begun on any sizeable scale. Some have beenill-conceived and most have had relatively little impact.In general, however, one cannot say that this overalleffort has been expensive in terms of the total effortexpended. Although evaluation efforts may not have beena precision tool for correcting organizational deficiencies,they have served to bring to light administrative andprocedural deficiencies. They have probably been sub-stantially more efficient in their use of resources thanhave attempts to design comprehensive, centralized infor-mation systems.

The present study of state budgeting procedures foundfew instances in which outside management or economic con-sultants were used to provide expertise and analysis forthe budget process, either directly or indirectly. Ofcourse, technical budget review as practiced in moststates neither requires nor lends itself to the_use_of__outside personnel. Outside,consultants, however, wereused to design budget procedures in Michigan and Pennsyl-vania, to design the information and analysis proceduresused at the state level in Illinois, to perform a manage-ment analysis of higher education in Virginia, and todesign various information systems, such as an informationsystem for the legislature in Washington and a communitycollege information system in Illinois. There were un-doubtedly other examples in the states we visited. Theuse of these specialized consultants may be effective inbringing greater analytical and conceptual resources tobear on a design problem because the total number of

185

199

state-level budget personnel is limited in all but thelargest states. and the qualifications and special skillsof agency personnel do not usually match those of the

consultants. Aside from the quality of the work done,

which is an obvious.factor in implementation, use'of out-

side consultants appears to have its greatest drawback

in securing eventual implementat.Lon. All too often aproduct is "delivered" which, even if it meets the most

critical objections, will not be implemented because the

individuals who actually must use it were not sufficiently

involved in its development. Consultants also are in-

clined to ignore organizational constraints on implementa-

tion, and they may spend considerable time learning facts

already known by staffs.

Generally, the use of outside expertise for inter-

jecting more analytical substance and expertise in the

budget process at the state level is not as prevalent as

it might be. We refer to the use of all kinds of outside

sources, not just independent consultants. A major out-

side source of procedural assistance is the National

Center for Higher Education Management Systems (NCHEMS)

which has developed many of the tools and techniques onwhich state agencies depend. Through these "products"

and through seminars and assistance with implementation,

institutions and state agencies have been able to conservetheir own resources and also achieve some of the common-ality that has been desired by state and federal agencies.

Perhaps tne largest source of specially qualified

__persons is the state agencies themselves, but aside from

NCHEMS there seems to be no effective clearinghouse for

identification of these sources. The National Association

of College and Business Officers (NACUBO) and College and

University Systems Exchange (CAUSE) provide that assist-

ance to institutions and do not usually serve state

agencies. To some degree, NCHEMS has also served such

a role for the institutions, and is beginning to do the

same for the state higher education agencies. But execu-

tive budget office staff and especially legislative budget

staff do not have a formal means for effectively commun-

icating about their professional interests, except at

the very highest levels through the National Association

J186

of State Budget Officers and the Council of State Govern-ments. The Education Commission of the States (ECS) isbeginning tr) bring these officials together and couldserve an In reds inqly valuable role in the future. Onema7lor ;:ontlIbution it could make would be to establish a:learinghouse for exchange of state higher educationstudies and descriptions of analytic budgeting techniques.Legislative staffs are gradually developing linkages forexchange of these kinds of information through such organ-izations as the Legislative Program Evaluation Section ofthe Governmental Research Association.

LIMITATIONS IN THE APPLICATION OFANALYSIS TN POLICY CHOICE

It may overstate the intent of research and analysiswhich supports higher education budgetary review and state-level planning to consider it as social science, yet thespirit motivating attempts to improve and increase theseanalytical efforts14.the same as that behind attempts toincrease the use of social sCience knowledge and methodsin public policy administration. Therefore, we haveexamined the limitations on the use of centralized in-formation and analysis systems in higher education budget-ing by considering the utilization of social sciencemethods more generally. In using the term analysis below,we mean the application of social science knowledge andtechniques in the determination of public policy.

A number of factors may be responsible for thecurrent distrust of science and its application in public

--pottry-formulation; one certainly is the- concern that-"hard" science and technology are generating effectsbeyond our control, and this concern has undoubtedlycarried over to social science. If anything, this ts acountertrend to the more widely spread and overly opti-mistic expectations of both those who make policy andthose who administer and analyze it. Millikan (1963)has cogently summarized a number of misconceptions fromboth the action-oriented side (which we shall call thepolicymaker or policy administrator) and the researchside (which we shall call the analyst). A 1973-1974

187

AJi.

survey made by the Institute for Social Research of over200 top-level executive branch political appointees andcivil servants on their use Of social sdienoe-knowledge-----4suggests that these misconceptions are still widely held(Caplan, Morrison, & Stambaugh, 1975). Quoting Millikan

(1963):

[The policymaker] frequently has an exag-gerated notion of the degree to which thesolution of his problems can be effected bythe collection of additional factual infor-mation. . . . He tends to expect predictionin situations in which this is clearly be-yond the capubilities of present day socialscience . . . he too easily ascames that theconclusions of a research project willassist him when the important factor isactually the process of analysis underlyingthose conclusions. (p. 170)

The analyst likewise tehis to have an exaggerated faithin the division of labor between action's and analysf.s..He also underestimates the intellectual or substantivecontent of the policymaking process because those whoadminister policy usually do Aot articulate the inter-pretation of their judgments and ccurses of action inthe terminology of social science. Consequently, theanalyst assumes that decisionmaking is overly intuitive.The ability of most policymakers to define the conceptualframework in which they state their 7.,perational problemsalso is overestimated.

Pmtagonists--for-analysis-have tended- to -ignore- the---

fundamental conflict in political problem solving whichLindblom (1968) identified as being between the ideasthat stress, on the one hand, man's fallibility and the-...onsequent need for liberal democratic political institu-tions, and on the other, man's competence, his potentialfor theoretical formulations ..-ufficien to guide socialreconstruction, and the consequent acceptability ofauthoritarian leadership in the hands of the competent.Primed with the derivations of theoretical welfareeconomics, which Boulding has called one of the most

0ANC .4

188

barren excursions in the history of economic theory,analysts completely disretjarded the separations beteenscience and moral and political philosophy. The inabilityof science to cope with the hierarchy of social andpolitical ends was thought to have been overcome, and asa consequence, the advocates of managing our public affairswith mathematical models and scientific Qnalysis promisedtoo much. (See Moynihan, 1973, for a description of oneman's conversion and subsequent recantation.)

At present, the prospect of raising the analyticalcontent of the public budgeting process is greatly re-stricted by the failures associated with PPB and otherbudgetary procedures which have attempted similar reforms.After these excesses, however, the healthy skepticisnwhich undoubtedly exists may provide an environment formore effective reforms.

The introduction of analytical procedures into,thebudget process has been critiqued so thoroughly that areview here is both unnecessary and impossible. (See

especially Merewitz and Soznick, 1971, and the publicationsreferred to at the beginning of the previous chapter.)However, to emphasize considerations revealed by ourstudy we shall mention five important problems.

1. Because good analysis cannot be reduced toprocedural routines, it therefore will'always encounterfrictions with_the budget process which must precede itarid be completed according to a tight time schedule.Thus, a crucial problem for management is devisingarrangements for freeing analysis from the routine ofbudget_review,___and at the_same time_providing_channels _

for the consequences of staff analysis and evaluationto have an effect on budgetar ald planning outcomes.The difficulty this introduces is illustrated by thedistinction between algorithmic and heuristic problemsolving. Algorithmic problem solvin_ involves the solutionof problems with a fixed structure, an identifiable setof variables and parameters, and a completely specifiedset of decision rules. Budgetary planning and reviewtak(,1 place in a context of such time constraints andresources that most participants would undoubtedly fine

189

it desirable if this process could be reduced to anAlgorithm. In practice, us t. of budgetary formulas in arigid sense comes as close as any budgetary procedure tosuch an algorithm. In fact, budgetary procedures makevarious attempts at simplification by reducing certainactions to algorithms in order to accomplish a difficultand complex task in a limited time span. Introdrctionof analysis, which cannot be predefined or specified,makes the budget process more heuristic in that it servesonly to specify guidelines and provisional methods forcoping with a dynamically shifting problem structure.Thus, analysis may make budgetary problems more intract-able in the limited time available to solve them.

2. It has been suggested that introducing analyticalprocedures which make assumptions, procedures, and conse-quences more explicit raises the likelihood of conflict.Because the budget process must result in unambiguouschoices on levels of funding, it is argued that it mustbe from beginning to end a process which builds consensus.The lack of fuzziness in a final dollar appropriationimay be moderated by a fuzziness in the specification ofassumptions, procedures, and consequences.

3. Analysis may unwittingly introduce certain biasesin its applicatior of explicitly logical methods whichire predominantly quantitative. Emphasizing the quanti-tative may ultimately result in more significant andimportant qualitative factors being ignored. Further-

more, one of the primary strengths of analysis, thesubstitution of reasoning for experience through simula-tion and theoretical explanation,.implies that even ifdecisions are not,made,_they_can at least be consideredmore remotely from operating responsibility. Thus,

analysis may foster centralization and control. By in-

sisting that the world is too complicated far commonsense alone tp resolve its problem:,. analysis sets itself

in conflict with democratic de,..7icic,0 rules which assume

that no one can know enough, an.L. it cons:msus ormajority decision are the only acceptable mechanismsfor choice.

4. The bargaining context of the budget processalso raises doubts about the efficacy of information

190

2

and analysis which would not arise in a simpler -ingledecisionmaker context. The theory of games treats a sub-class of games known as nonzero-sum, noncooperative gamesof which the game of the prisoner's dilemma is alfamousexample. Dror (1968) offers this game as an example ofa bargaining situation in which a logical or rationalproblem solving approach leads to results which would beinferior to intuitive or "extra-rational" methods. Theinterpretation of this game situation given by Luce andRaiffa (1957) is provided in Appendix D. We have notbeen able to draw a valid budgeting application of thisgame, but the recent experience of the State UniversitySystem of Florida with the exploitation of budgetaryformulas by two of its campuses comes to mind (seeVan Dyne, 1974). Clearly, there are cooperative (freeexchange of information) game situations in which thedisclosure of information (on preferences, activities,etc.) would be irrational, that is, would not serve theinterests of the participants in a bargaining situation.

5. As noted above, difficulties may arise in theuse of analysis because of the overriding attention givento the results by both the policymaker and the analyst.But what often happens is that results are ignored wherethey are counterintuitive, and are accepted uncriticallywhere they are consistent with intuition. If resultsseem reasonable in the light of commonsense, they areoften felt to be intuitively obvious by a policymaker,and analysis is considered a waste of time. The surveyundertaken by the Institute for Social Research (Caplanet al., 1975), and mentioned earlier in this chapter,concerned the use of results from program evaluation,survey research, demographic research, social statistics,and field experimentation, which may go beyond much ofthe,analysis that comes to bear on problems of budgetaryanalysis. Nonetheless, the conclusion of that study,indicating a preoccupation with the results of analysis,would undoubtedly be confirmed in a similar study ofstate budget officials.

With all of these problems in introducing analysisinto the decision process and its alleged detrimentaleffects, one wonders why the budgetary process continues

191

205

to involve any analysis. Despite the warnings, however,

there is hardly any indication that present-day budgetingand planning processes involve too much analysis, althoughthere are some indications that the current burden of in-formation for both federal and state agencies that must becarried by staff and administrative channels is excessive.What appears to be developing is a skl,e of analytical dueprocess, analogous to administrative due process, thatthere is a minimal level of logical and explicit analysisthat should be a factor in legitimately reached publicsector decisions. While this presents a potential report-ing problem if it is the upward flow of post hoc infor-mation that is emphasized, the notion that improveddecisionmaking would be served by a minimal level of in-formation and analysis does not seem unreasonable. Filter-ing and biasing of the upward flow of information orsimply swamping those at higher levels with more infor-mation than they can handle limits the effectiveness ofupward reporting systemc.

If many of the expectations for analysis are exag-gerated by its proponents, what then are some more reason-able expectations? First of all, analysis is likely toachieve more tangible results in terms of the process ofpublic policymaking than in relation to specific results.On identifying organizational objectives, Hitch (1961)commented:

So what does the analyst do? If he cannotfind anyone to give him acceptable objectives,where does ha obtain them? The only answerI have is that learning about objectives isone of the chief objexts of this kind ofanalysis. . . . We have never undertaken amajor systems study at RAND in which we wereable to define satisfactory objectives atthe beginning of the study. (p. 49)

In this respect, analysis does not give answers to policyquestions, but it may contribute to their formulation.

Millikan (1963) has suggested that the task of socialscience is to provide the substance of argument through

rj460

192

extendiny the policymakers' capacity for judgment. Agencyofficials may lack the time for analysis themselves, butthrough their use of staff a greater range of analyticalsubstance can be brought to bear on issues to raise thelevel of the poli.y dialogue. Further, policy choicesare often surrounded by clusters of intuitively obvious,but partially conflicting statements. Sound analysiscan determine which of the contradictory, but intuitivelyobvious conclusions about a situation is in fact true andunder what circumstances it may be expected to hold.This application-of analysts may appdar somewhat restrictedin comparison with the role given it by Utopian socialengineers and planners. But it is a far more activistrole than that allowed for by those who would restrictanalysis (and social science) to the mere interpretationof events because they believe that social developmentis unfolding according to various immutable patterns andtrends. Not forgetting that our primary concern is anappropriate role for budgeting information and analysissystems, we should note that our discussion is influencedby the philosophical criticisms of Popper (1962), whoargued:

Only a minority of social institutions areconsciously designed, while the vast majorityhave just "grown," as the undesigned resultsof human actions. . . . The main task ofthe social sciences . . . is the task ofanalyzing the unintended social repercus-sions of intentional human actions.(pp. 93, 95)

Various observers have pointed out that organizationsdo not learn well from experience. This is attributable,in part, to the weakness of the intelligence function,the fact that organizational memory is often weak. Wehave noted that the political process often thrives onthis dull organizational memory because of the necessityto produce consensus, and may oppose changes that wouldsharpen memory. In the special case of state budgetagencies, where the turnover of analytical staff isparticularly high, this memory function may be improvedby a greater capacity to store and analyze information.

193

20

The pluralist or checks and balances model, which

describes most accurately the process employed at all

governmental policy levels in the United States, emphasizes

that the process for decision is paramount. If that

process is legitimate, then the outcomes of the process

are acceptable. But as Schick (1969) has pointed out,

there are imbalances in power in political markets, just

as there are in economic markets for goods and services.

Analysis, hard logic, and the scrutiny of implicit assump-

tions, processes, and consequences are needed as a "check"

on the political process. As in economic markets, political

participants try to gain comparative advantage over the

allocation of political goods and services. If facts are

ignored or interested parties are excluded, mere agreement

(consensus) is not in itself an adequate criterion for

evaluating the outcome. This is not to suggest that reason

and analysis become the final arbiter of public choice,

but that reason and analysis have a legitimate supporting

role to play in determining these choices.

As for the specifics of budgeting, it is in the

costing and evaluating functions that analysis has the

greatest role to play. Rivlin (1971) posed four questions

that.anyone making decisions on social action programs

should answer:

1. How do we define social problems, and how

are they distributed?

2. Who would be helped by specific social action

programs, how much, and at what cost?

3. How can particular kinds of social services

be produced more effectively?

4. How do the benefits of different kinds of

programs compare?

The budget process does not explicitly address all

of these questions for all state programs every cycle,

but implicitly and in varying levels of detail it does

answer these questions either with analysis or through

the political process. In view of our earlier conclusions

194

203

about the importance of methods versus results, we areinterested in analysis as a method for addressing these.questions in the context of higher education. It must benoted that only with respect to the first two questionsabove has much progress been made. Although the thirdquestion appears tractable, Rivlin observed that fewattempts have been made to organize social services insuch a way as to address this question. No analyticalmethod is available for addressing the fourth question.With respect to public higher education, analysis hasprovided a fairly sound basis for looking at educationalproblems and their distribution; it has provided numerousapproaches to costing, but only a very limited indicationof benefits, either private or public; and as is true formost other social services, it has contributed very littletoward evaluating alternative means of providing highereducation.

Although we shall not offer an agenda for analysisbecause that depends very much on the particular issuesand concerns in a state, two issues relating to budgetinformation could be illuminated by analytical effort.The first of these is the development of alternative modelsfor budgetary and programmatic review. Currently,'thefield is almost totally dominated by the "production"model in which, for budgetary purposes, virtually everysocial service agency is treated as a production processconsisting of a set of inputs, a technology of production,and a set of outputs. Many of the difficulties in develop-ing adequate and useful information systems are relatedto difficulties in applying this model to organizationalprocesses which are not truly production processes, sincetheir outputs cannot be practically specified. This modelis, of course, very attractive because it meshes well withthe costing function of budgeting, and because it providesa connection with the status quo, so that incrementalchanges are encompassed. Because it can be used to answercrucial costing questions, it is unlikely that it willbe completely supplanted in the budget process. In ano-growth environment, however, simple versions of theproduction model are inadequate. It should be the taskof analysis to explore the nonlinear features of univer-sity financing as they relate to the critical mass

195

2U0

characteristics of departments, size of student body andfaculty, and breadth of course offerings. This analysiswould be related to an investigation of the simple ratios,such as student/faculty ratios and average-cost ratios,which are used in budgeting without any valid informationabout the operational implications of funding determinedin this way. This requires analysis of the kind carriedout by Carlson (1972), which developed more suitableindices of the cost and production behavior of institu-tions. While this kind of analysis cannot, like formulas,become a part of the budget process, its results and con-clusions should at least inform the budget process.Because of its ability to examine both intended and un-intended consequences or proposed changes, analysis canexplore alternatives in funding arrangements, such asperformance contracts and student vouchers.

Further, analysis can be effective in developing atheoretical base which supports data structures and theireducational or social indicators. The success of thenational economic data collection for policy and researchuse has been partially due to the existence of theoriesof the macro-economy which suggested, first of all, thestructure and relationships between variables, andsecondly, determined which of the many variables wouldbe vseful economic indicators.

Ultimately, we are faced with the question posed byMiller (1964) in his discussion of formula methods ofbudgeting which, before the advent of PPB, must haverepresented a pinnacle of achievement in incorporatinganalysis into the budget process. Is the recognitionand acceptance of the value content in the final budgetchoices incompatible with the advocacy of a rationalprocedure for considering them?

If by rational we mean a procedure governed by theprinciples of science, which gives us criteria for ascer-taining facts or knowledge, then the answer must be yes,for determining the priority of social values and theproper course for social action is not just a problem offinding facts. However, if rational refers to distinctsets of technolgoical, economic, legal, and sociopolitical

210196

principles, as Diesing (1962) has argued, then this in-compatibility is removed, albeit by definition. Each ofthese sets of principles has its own concepts, and mostimpy.Lcantly, its own problem-defining and problem solvingmethods which may interact with each other from one extremeto the otner as values influence facts and facts influencevalues. Although it is far better to recongize the mutualdependence of fact and value than to separate them, thetask for those who try to apply these sets of principlesis to recognize when each set should be dominant. Forthis reason, the best educational policy or budget analystis likely to be a teacher-economist-social philosopher.

CI441.1

197

References

Anthony, R. N. Planning and control systems: A frame-work for analysis. Cambridge: Graduate School ofBusiness Administration, Harvard University Press, 1965.

Ashby, E. Any person, any study. New York: McGraw-Hill,1971.

Bailey, J. J., & O'Connor, R. J. Operationalizing incre-mentalism: Measuring the muddles. Public Administra-tion Review, January/February 1975, 35(1), 60-66.

Bailey, S. K. A comparison of the university with agovernment bureau. In J. A. Perkins (Ed.), The univer-sity as an organization. New York: McGraw-Hill, 1973.

Balutis, A. P., & Butler, D. K. (Eds.). Political purse-strings: The role of the legislature in the budgetaryprocess. New York: Halsted Press, 1975.

Balutis, A. P. The budgetary process in New York State:The role of the legislative staff. In A. P. Balutis,& D. K. Butler (Eds.), Political pursestrings: The

role of the legislature in the budgetary process.New York: Halsted Press, 1975.

Barak, R. J. State-level management information systemsfor higher education: A survey summary. Unpublishedreport, Iowa State Board of Regents, April 1975.

198

2 .1 2

Beckman, N. Use of a staff agency by the Congress: Thecongressional reference service. Bureaucrat, January1975, 3(4), 401-428.

Berdahl, R. 0. Statewide coordination of higher education.Washington, D.C.: American Council on Education, 1971.

Bogard, L. Management in institutions of higher educa-tion. In A. A. Mood, C. Bell, L. Bogard, H. Brownlee,& J. McCloskey, Papers on efficiency in the managementof higher education. Berkeley: Carnegie Commissionon Higher Education, 1972.

Boulding, K. E. The management of decline. Change,June 1975, 7(5), 8-9.

Bowen, F. M., & Glenny, L. A. State budgeting for highereducation: State fiscal stringency and public highereducation. Berkeley: Center for Research and Develop-ment in Higher Education, University of California, 1976.

Bowen, F. M., Ruyle, J., & Glenny, L. A. State Budget-ing for higher education: Budget processes in the 50states. Berkeley: Center for Research and Developmentin Higher Education, University of California, in pre-paration.

Braybrooke, D., & Lindblom, C. E. A strategy of decision.New York: The Free Press, 1963.

Butler, D. K. The legislative budget in Texas. InA. p. Balutis, & D. K. Butler (Fe-.), P,litical purse-strings: The role of the legislature in the budgetaryprocess. New York: Halsted Press, 1975.

Bureau of the Budget. Bulletin No. 68-2, July 18, 1967.

Burkhead, J. Government budgeting. New York: JohnWiley, 1956.

Caplan, N., Morrison, A., & Stambaugh, R. J. The useof social science knowledge in policy decisions at the

199

21,3

national level. Ann Arbor: Center for Research onUtilization of Scientific Knowledge, Institute forSocial Research, 1975.

Carlson, D. E. The production and cost behavior of highereducation institutions (Ford Foundation Program forResearch in University Administration, Paper P-36).Berkeley: University of California, December 1972.

Carnegie Commission on Higher Education. Higher education:Who pays? Who benefits? Who should pay? New York:McGraw-Hill, 1973.

Chase, S. B., Jr. Problems in public expenditure analysis.Washington, D.C.: The Brookings Institution, 1967.

Cheit, E. F. The management systems challenge: How tobe academic though systematic. In J. F. Hughes (Ed.),

Education and the state. Washington, D.C.: AmericanCouncil on Education, 1975.

Citizens Conference on State Legislatures. The sometime

governments. Kansas City, Mo.: Author, 1971.

Cohen, M. D., March, J. G., & Olsen, J. P. A garbage can

model of organizational choice. Administrative ScienceWarterly, March 1972, 17(1), 1-25.

Collier, D. J. Higher education finance manual (Tech-nical Report 69). Boulder, Co.: National Center forHigher Education Management Systems at WICHE, 1975.

Commonwealth of Pennsylvania, Budget Instructions forColleges and Universities,.FY 1974-1975, Appendix B.

Corbally, J. E., Jr. The impact of MIS on and within amajor institution and institutiona: system. In Informa-tionoimpact-collision with tradition (Report No. 49).Denver: Education Commission of the States, May 1974.

Council of State Governments. Higher education in the

forty-eight states. Chicago: Author, 1952.

21 A

200

Crecine, J. P. Governmental problem-solving: A computersimulation of municipal budgeting. Chicago: RandMcNally, 1969.

Davis, O., Dempster, M. A. H., & Wildavsky, A. On theprocess of budgeting: Am empirical study of congres-sional appropriation. Papers on Non-Market DecisionMaking, 1966, 63-132.

Deardeni'J. MIS is,a mirage. Harvard Business Review,. January/February 102, 50(1), 90-99.

Diesing,.P. Reason in society: Five types of decisionsand their social c nditions. Urbana: University ofIllinois Press, 1962.

Dippel, G., & House( W. C. Information systems. Glen-view, Ill.: Scott, Foresman, 1973.

Downs, A. A realist,ic look at the final payoffs fromurban data systems. Public Administration Review,September/October 1967, 27(5), 204-210.

Dror, Y. Public policy reexamined. San Francisco:Chandler, 1968.

Dye, T. R. Politics in states and communities (2nd ed.).Englewood Cliffs, N.J.: Prentice-Hall, 1973.

Dyer, J. S. The use of PPBS in a public system of highereducation: Is it cost-effective? Academy of Manage-ment Journal, September 1970, 13(3), 285-299.

Ellul, J. The technological society. New York: Knopf,1965.

Fenno, R. F., Jr. The power of the purse. Boston:Little, Brown, 1966.

Fincher, C. On the rational solution of dominant issuesin higher education. Journal of Higher Education,November/December 1975, 35(6), 625-628.

201

24 5

Gerwin, D. Budgeting public funds: The decision process

in an urban school district. Madison: University ot

Wisconsin Press, 1969.

Glenny, L. A. Autonomy of public colleges. New York:

McGraw-Hill, 1959.

Glenny, L. A. State budgeting for higher education:

Interagency conflict and consensus. Berkeley: Center

for Research and Development in Higher Education,

University of California, 1976.

Glenny, L. A., Bowen, F., Meisinger, R., Morgan, A.,Purves, R., & Schmidtlein, F. State budgeting for higher

education: Data digest. Berkeley: Center for Research

and Development in Higher Education, University of

California, 1975.

Grosz:, B. M. The new systems budgeting. Public Adminis-

tration Review, March/April 1969, 29(2), 113-137.

Gross, F. M. A comparative ana1y.t.4s of the existing budget

formulas used for justifying budget requests or alloca-

ting funds for the operating expenses afstate-supported,colleges and 4niversities. Unpublished doctoral disser-

tation, Unive.-sity of Tennessee, 1073.

Gulko, W. W. Program classification structure (Technical

Report 27). Boulder, Co.: National Center for Higher

Education Manasement Systems at WICHE, 1972.

Highet, G. The need to "make it new." Chronicle of

Higher Education, June 21, 1976, 40.

Hitch, C. J. On the choice of objectives in systems

studies. In D. P. Eckman (Ed.), Systems: Research and

design (Proceedings of the First Systems Symposium atthe Case /nstitute of Technology). New York: John

Wiley, 1961.

Howard, S. K. Changing state budgeting. Lexington, Ky.:

Council of State Governments, 1973.

21 3202

I 3,

Huff, R. A., & Chandler, M. O. A taxonomy of instruc-tional programs in higher educqtion (National Centerfor Educational Statistics). Washington, D.C.:U.S. Government Printing Office, 1970.

Jernberg, J. E. Information change and congressionalbehavior: A caveat for PPB reformers. Journal ofPolitics, August 1969, 33(3), 72.-740.

Kent, J. D. Legislative fiscal staffing in Illinois.In A. P. Balutis, & D. K. Butler (Eds.), Politicalpursestrings: The role of the legislature in the budg-etary process. New York: Halsted Press, 1975.

Kingdon, J. W. Congressmen's voting decisions. New York:Harper & Row, 1973.

Lee, R. D., Jr., & Johnson, R. W. Public budgetingsystems. Baltimore: University Park Press, 1973.

Lindblom, C. E. ne policy-making procLes. EnglewoodCliffs, N.J.: Prentice-Hall, 1968.

Luce, R. D., & Raiffa, H. Games and decisions. New York:John Wiley, 1957.

Lyden, F. J. The budget cycle as a basis for decision-making in higher education. Planning for Higher Educa-tion, October 1975, 4(5).

Lyden, F. J., & Milier, E. G. Planning programming budg-eting: A systems approach to management. Chicago:Markham, 1968.

McCoy, M., Cherin, E., Makowski, D., & Weldon, K. Stateand local financial support of higher education: Aframework for interstate comparisons 19 7-1974.Boulder, Co.: National Center for Higher EducationManagement Systems at WICHE, 1976.

Meisinger, R. J., Jr. State budgeting for higher educa-tion: The uses of formulas. Berkeley: Center forResearch and Development in Higher Education, Univer-sity of California, 1976.

203

217

Merewitz, L., & Sosnick, S. H. The budget's new clothes:A critique of PPB and benefit-cost analysis. Chicago:Markham, 1971.

Miller, J. L., Jr. State budgeting for higher education:The use of formulas and cost analyses. Ann Arbor:Institute of Public Administration, University ofMichigan, 1964.

Milliken, M. F. Inquiry and policy: The relation ofknowledge to action. In D Lerner (Ed.), The humanmeaning of the social sciences. New York: Meridian,1963.

Mitroff, I. I., & Pondy, L. R. On the organization ofinquiry: A comparison of some radically differentapproaches to policy analysis. Public AdministrationReview, September/October 1974, 34(5), 471-479.

Moos, M., & Rourke, F. The campus and the state.Baltimore: Johns Hopkins Press, 1959.

Moynihan, D. P. Coping: On the practice of government.New York: Random Houbc. 1973.

National Association of College and University BusinessOfficers. College and uniVersity business administra-tion (::nd ed.). Washington, D.C.: Author, 1968.

National Association of College and University BusinessOfficers. College and university business administra.-tion (3rd ed.). Washington, D.C.: Author, 1974.

National Commission on the Financing of PostsecondaryEducation. Financing postsecondary educatic.; in theUnited States. Washington, D.C.: U. S. GovetnmentPrinting Office, December 1973.

Niskanen, W. A. Why new methods of budgetary choices?Administrative aspects. Public Administration Review,March/April 1972, 155-161.

213204

Popper, K. The open society and its enemies (Vol. 2).Priaceton: Princeton University Press, 1962.

Pyhrri P. A. Zero-base budgeting. New York: John Wiley,1973.

Rivlin, A. M. Systematic thinking for social action.Washington, D.C.: The Brookings Institution, 1971.

Ruyle, J., & Glenny, L. A. State budgeting for highereducation: Trends in state revenue appropriationsfrom 1968-1975. Berkeley: Center for Research andDevelopment in Higher Education, University of Calif-ornia, in preparation.

Schick, A. Budget innovation in the states. Washington,D.C.: The Brookings Institution, 1971.

Schick, A. The road to PPB: The stages of budget reform.Public Administration Review, December 1966, 26(4),243-258.

Schick, A. Systems politics and systems budgeting.Public Administration Review, March/April 1969, 29(2),137-151.

Schmidtlein, F. A. Decison processes and their structuralimplications: A case study of library planning at theUniversity of California. Unpublished doctoral disser-tation, University of California, Berkeley.

Schmidtlein, F. A., & Gienny, L. A. State budgeting forhigher education: The political economy of the process.Berkeley: Center for Research and Development in HigherEducation, University of California, 1976.

Schultze, C. L. The politics and economics of publicspending. Washington, D.C.: The Brookings Institution,1968.

Silverman, E. B., & Gatti, F. C., Jr. PPB on the statelevel: The case of Pennsylvania. The Bureaucrat,July 1975, 4(2), 117-146.

205

219

Simon, H., Smithburg, D., & Thompson, V. Public admin-

istration. New York: Knopf, 1970.

State of Connecticut. Department of Finance and Control.

1974.

State of Hawaii. The multi-year program and financialplan for the period 1974-1980. Submitted to the 7th

State Legislature, December 1973.

Steiner, P. 0. Public expenditure budgeting. In The

economics of public finance. Washington, D.C.: The

Brookings Institution, 1974.

U.S. Bureau of the Budget. Bulletin No. 68-2, July 18,

1967. Reprinted in Lyden & Miller (1968),pp. 440-441.

U.S. Bureau of the Budget. Circular No. A-12, July 221

1960.

University of Wisconsin System. 1973-75 biennial budgetdocumentation, Book V, U.W. System Biennial BudgetRequest and Decision Item Narrative, August 1972.

Van Dyne, L. Tallahassee hassle: A scholar's explosive

memo. The Chronicle of Higher Education, July 1974,8(37), 10-11.

Weiss, S. D.Cuadra & A.science andBritannica,

Management information systems. In C. A.

W. Like (Eds.), Annual review of informationtechnology (Vol. 5). Chicago: Encyclopedia

1970.

Wildavsky, A. The politics of the budgetary process.

Boston: Little, Brown, 1964.

Wildavsky, A. The political economy of efficiency:Cost-benefit analysis, systems analysis, and program

budgeting. Public Administration Review, December 1966,

26(4), 292-310.

Wildavsky, A. Rescuing policy analysis from PPBS. Public

Administration Review, March/April 1969, 29(2), 189-202.

n4.0

206

Wildavsky, A. The self-evaluating organization. PublicAdministration Review, September/October 1972, 32(5),509-520.

Williamson, 0. E. A raticaal theory of the federal budg-eting process. Papers on Non-Market Decision-Making,1967, 2, 71-89.

Wolin, S. S. Politics and vision. Boston: Little, Brown,1960.

General Bibliography

American Institute of Certified Public Accountants.Audits of colleges and universities. New York: Author,1973.

Anthony, R. N., & Herzlinger, R. E. Management controlin nonprofit organizations. Homewood, Ill.: RichardD. 'Irwin, 1975.

Blumenthal, S. C. Management information systems: Aframework for planning and development. EnglewoodCliffs, N.J.: Prentice-Hall, 1969.

Brewer, G. D. Politicians, bureaucrats, and consultants.New York: Basic Books, 1973.

Chartrand, R. L., Janda, K., & Hugo, M. (Eds.). Infor-mation support, program budgeting, and the Congress.New York: Spartan Books, 1968.

Davis, G. B. Management information systems: Conceptualfoundations, structure, and development. New York:McGraw-Hill, 1974.

Ebert, R. J., & Mitchell, T. R. Organizational decisionprocesses. New York: Crafie, Russak, 1975.

Galbraith, J. Designing complex organizations. Reading,Mass.: Addison-Wesley, 1973.

207

2 2

Hill, M. J. The sociology of public administration.New York: Crane, Russak, 1972.

Hussain, K. M. Development of information systems foreducation. Englewood Cliffs, N.J.: Prentice-Hall, 1973.

Johnson, C. B., & Katzenmeyer, W. G. (Eds.) Managementinformation systems in higher education: The state ofthe art. Durham, N.C.: Duke University Press, 1969.

Lyden, F. J., & Miller, E. G. (Eds.). Planning program-ming budgeting: A systems approach to management (2nded.). Chicago: Markham, 1972.

National Research Council, Assembly of Behavioral andSocial Sciences, Committee on Department of LaborManpower Research and Development. Knowledge and policyin manpower: A study of the manpower research anddevelopment program in the department of labor. Wash-ington, D.C.: National Academy of Sciences, 1975.

Niskanen, W. A. Structural reform of the federal budgetprocess. (Domestic Affairs Study No. 12). Washington,D.C.: American Enterprise Institute for Public PolicyResearch, April 1973.

Rosenthal, A. Legislative performance in the states.New York: The Free Press, 1974.

Toombs, W. Productivity: Burden.of success (ERIC/HigherEducation Research Report No. 2). Washington, D.C.:American Association for Higher Education, 1973.

Weidenbaum, M. L., Larkins, D., & Marcus, P. N. Matchingneeds and resources. Washington, D.C.: American Enter-prise Institute for Public Policy Research, March 1973.

Wildavsky, A. Review of Brewer's politicians, bureaucrats,and the consultant. Science, 1973, 28, 1335-1338.

Wilensky, H. Organizational intelligence. New York:

Basic Books, 1967.

2"oqp,

208

APPENDIX A-1

NACUBO Current Funds Expenditures Accounts,1968

EDUCATIONAL AND GENERAL

INSTRUCTION AND DEPARTMENTAL RESEARCHAccounts by divisions, schools, colleges, and depart-ments of instruction, following the administrativeorganization of the institution.

ORGANIZED ACTIVITIES RELATED TO EDUCATIONAL DEPARTMENTSAccounts for each such activity for which revenueaccounts have been established.

SPONSORED RESEARCH

Accounts by individual projects, classified by.organizational units.

OTHER SEPARATELY BUDGETED RESEARCHAccounts by individual projects, classified byorganizational units.

OTHER SPONSORED PROGRAMSAccounts by individual projects, classified byorganizational units.

EXTENSION AND PUBLIC SERVICEAccounts for each organizationa/ unit in thiscategory, such as:Agricultural Extension ActivitiesContinuing EducationDepartmental Research BureausPublic Lectures and Concerts

Source: National Association of College and Univer-...

sity Business Officers (2nd ed.), 1968.

209

223

LIBRARIESAccounts for all libraries, both central as well as

departmental.

STUDENT SERVICESAccounts for all organizational units, such as:RegistrarDean of StudentsDirector of AdmissionsFinancial Aid OfficerHealth or Infirmary Services--un/ess classified as

an Auxiliary Enterprise

OPERATION AND MAINTENANCE OF PHYSICAL PLANTAccounts for all organizational units and functions,

such as:AdministrationCustodial ServicesMaintenance of BuildingsMaintenance of GroundsUtilitiesPolice and WatchmenTrucking ServicesFire ProtectionMotor Pool and Transportation Services--un/ess

classified as a Service DepartmentProperty Insurance

GENERAL ADMINISTRATIONdetailed as needed;for example:Governing BoardPresidentChief Academic OfficerChief.Business OfficerInvestment OfficerLegal Counsel

STAFF BENEFITS--detailed as needed; for example:Retirement Contract PremiumsGroup Life Insurance PremiumsHealth and Medical Insurance PremiumsWorkmen's Compensation Insurance

210

GENERAL INSTITUTIONAL EXPENSE--detailed as needed;for example:Alumni OfficeAuditingCatalogues and BulletinsCommencementsConvocationsDevelopment OfficeGeneral Insurance--other than Property InsuranceInterest on Current Funds LoansLegal FeesMembershipsPublicationsPublic RelationsTelephone and. Telegraph--unless charged todepartmental budgets

STUDENT AID

Accounts as needed and desired for undergraduate andgraduate scholarships, fellowships, grants-in-aid, prizes,and awards.

Tuition remissions--unless classified as scholar-ships or fellowships

Accounts may be set up for instructional divisionsand departments, such as:

School of MedicineDepartment of Physics

AUXILIARY ENTERPRISES

Accounts as needed and desired for the same enterprisesincluded in the Current Funds Revenues accounts.

SERVICE DEPARTMUNTS

Nominal, or Interim, accounts for all organizationalunits classified in this category. These accountsshould be closed out at the end of each fiscal year.

211

APPENDIX A-2

NACUBO Current Funds Expenditures Accounts,1974

EDUCATIONAL AND GENERAL

INSTRUCTIONAccounts by divisions, schools, colleges, and depart-ments of instruction following the administrativeorganization of the institution. The four functional

subcategories are:General academic instructionOccupational and vocational instructionSpecial session instructionCommunity education

RESEARCHAccounts by individual projects, classified by organ-izational units. The two functional subcategoriesare:

Institutes and research centersIndividual or project research

PUBLIC SERVICEAccounts by activities, classified by type of activity,

such as:Community ServiceConferences and InstitutesCooperative Extension ServicePublic LecturesRadioTelevision

From College and University Business Administration,Third Edition (Washington, D. C., 1974), by permission ofthe National Association of College and University BusinessOfficers, pp. 212-214.

212

22 3

ACADEMIC SUPPORTAccounts by activities, classified by type ofactivity, such as:

Academic Administration and Personnel DevelopmentAudiovisual ServicesComputing support (excluding administrative dataprocessing), unless distributed to usingactivities

Course and Curriculum DevelopmentDemonstration SchoolsLibrariesMuseums and Galleries

STUDENT SERVICESAccounts by activities, classified by type ofactivity, such as:

Admissions OfficeCounseling and Career GuidanceCultural EventsDean of StudentsFinancial Aid AdministrationHealth and Infirmary Services if not an integralpart of a hospital nor operated as an essentiallyself-supporting operation

Intramural AthleticsIntercollegiate Athletics if operated as an inte-gral part of department of physical educationand not essentially self-supporting

RegistrarStudent OrganizationsRemedial Instruction

INSTITUTIONAL SUPPORTdetailed as needed, for example:

Governing BoardChief Executive OfficeChief Academic OfficeChief Business OfficeInvestment OfficeLegal CounselAdministrative Data ProcessingAlumni OfficeAuditing, internal and externalSafetySecurity

213

227

Catalogues and BulletinsCommencementsConvocationsDevelopment OfficeEmployee Personnel and RecordsFund RaisingGeneral Insurance other than Property InsuranceInterest on Current Funds LoansLegal FeesMembershipsPrintingProvision for Doubtful Accounts and NotesPublicationsPublic RelationsPurchasingService Departments

There should be interim accounts for all organi-zational units classified in this category; theseaccounts should be closed out at the end of eachfiscal year.

Space ManagementTelephone and Telegraph unless charged to depart-mental budgets

Transportation including motor pool, unlessoperated as a service department

OPERATION AND MAINTENANCE OF PLANTAccounts for all organizational units and functions,such as:

AdministrationCustodial ServicesMaintenance of BuildingsMaintenance of GroundsUtilitiesTrucking ServicesFire ProtectionProperty Insurance

SCHOLARSHIPS AND FELLOWSHIPSAccounts as needed and desired for scholarshipe,fellowshipe, grants-in-aid, trainee stipends,prizes, and awards.Tuition and Fee Remissions unless properly classifiedas staff benefit expenditures Accounts may be set up

for instructional divisions and departments, such as:School of MedicineDepartment of Physics

AUXILIARY ENTERPRISES, HOSPITALS, AND INDEPENDENTOPERATIONS

AUXILIARY ENTERPRISESAccounts as needed and desired for such enterprisesas included in the Current Funds Revenues accounts.Provision should be made for identification ofmandatory and nonmandatory transfers--to and from--by significant subcategories.

HOSPITALSAccounts as needed and desired. Frovision should bemade for identification of mandatory and nonmandatorytransfers--to and from--by significant subcategories.

INDEPENDENT OPERATIONSAccounts as needed and desired fo- organizationalunits.

Provision should be made for identification ofmandatory and nonmandatory transfers--to and from--by significant subcategories.

229215

APPENDIX B

HEGIS Taxonomy of Instructional Programsin Higher Education

0100 AGRICULTURE AND NATURAL RESOURCES

0101 Agriculture, General0102 Agronomy (Field Crops, and Crop Management)0103 Soils Science (Management and Conservation)0104 Animal Science (Husbandry)0105 Dairy Science (Husbandry)0106 Poultry Science0107 Fish, Game, and Wildlife Management0108 Horticulture (Fruit and Vegetable Production)0109 Ornamental Horticulture (Floriculture, Nursery

Science)0110 Agricultural and Farm Management0111 Agricultural Economics0112 Agricultural Business0113 Food Science and Technology0114 Forestry0115 Natural Resources Management0116 Agriculture and Forestry Technoldgies0117 Range Mangagement0199 Other, Specify

0200 ARCHITECTURE AND ENVIRONMENTAL DESIGN

0201 Environmental Design, General0202 Architecture0203 Interior Design0204 Landscape Architecture

Source: Huff, R. A., & Chan(41,r, M. O., 1970.

216

230

0205 Urban Architecture0206 City, Community, and Regional Planning0299 Other, Specify

0300 AREA STUDIES

0301 Asian Studies, Gen,ral0302 East Asian Studies0303 South Asian (India, etc.) Studies0304 Southeast Asian Studies0305 African Studies0306 Islamic Studies0307 Russian and Slavic Studies0308 Latin American Studies0309 Middle Eastern Studies0310 European Studies, General0311 Eastern European Studies0312 West European Studies1313 American Studies1314 Pacific Area Studies0399 Other, Specify

0400 BIOLOGICAL SCIENCES

0401 Biology, General0402 Botany, General0403 Bacteriology0404 Plant Pathology0405 Plant Pharmacology9406 Plant Physiology9407 Zoology, General0408 Pathology, Human and Animal0409 Pharmacology, Human and Animal0410 Physiology, Human and Animal9411 Microbiology0412 Anatomy0413 Histology0414 Biochemistry0415 Biophysics9416 Molecular Biology.9417 Cell Biology (Cytology, Cell Physiology)9418 Marine Biology0419 Biometrics and Biostatistics0420 Ecology0421 Entomology

217

2 3 1

0422 Genetics0423 Radiobiology0424 Nutrition, Scientific (exclude Nutrition in Home

Economics and Dietetics)

0425 Neurosciences0426 Toxicology0427 Embryology0499 Other, Specify

1500 BUSINESS AND MANAGEMENT

0501 Business and Commerce, General0502 Accounting0503 Business Statistics0504 Banking and Finance0505 Investments and Securities0506 Business Management and Administration0507 Operations Research0508 Hotel and Restaurant Management0509 Marketing and Purchasing0510 Transportation and Public Utilities0511 Real Estate0512 Insurance0513 International Business0514 Secretarial Studies0515 Personnel Management0516 Labor and Industrial Relations0517 Business Economics0599 Other, Specify

6Goo COMMUNICATIONS

0601 Communications, General0602 Journalism (Printed Media)0603 Radio/TV0604 Advertising0605 Communication (use of videotape, film, etc.,

oriented specifically toward radio/TV)0699 Other, Specify

0700 COMPUTER AND INFORMATION SCIENCES

0701 Computer and Information Sciences, General0702 Information Sciences and Systems0703 Data Processing0704 Computer programming

2 3218

0705 Systems Analysis0799 Other, Specify

0800 EDUCATION

0801 Education, General0802 Elementary Education, General0803 Secondary Education, General0804 Junior High School Education0805 Higher Education, General0806 Junior and Community College Education0807 Adult and Continuing Education0808 Special Education, General0809 Administration of Special Education0810 Education of the Mentally Retarded0811 Education of the Gifted0812 Education of the Deaf0813 Education of the Culturally Disadvantaged0814 Education of the Visually Handicapped0815 Speech Correction0816 Education of the Emotionally Disturbed0817 Remedial Education0818 Special Learning Disabilities0819 Education of the Physically Handicapped0820 Education of the Multiple Handicapped0821 Social Foundations (History and Philosophy of

Education)0822 Educational Psychology (include Learning Theory)0823 Pre-Elementary Education (Kindergarten)0824 Educational Statistics and Research0825 Educational Testing, Evaluation, and Measurement0826 Student Personnel (Counseling and Guidance)0827 Educational Administration0828 Educational Supervision0829 Curriculum and Instruction0830 Reading Education (Methodology and Theory)0831 Art Education (Methodology and Theory)0832 Music Education (Methodology and Theory)0833 Mathematics Educetion (Methodology and Theory)0834 Science Education (Methodology and Theory)0835 Physical Education0836 Driver and Safety Education0837 Health Education (include Family Life Education)0838 Business, Commerce, and Distributive Education

219

23 3

0839 Industrial Arts, vocational and TechnicalEducation

0899 Other, Specify

09C0 ENGINEERING

0901 Engineering, General0902 Aerospace, Aeronautical and Astronautical

Engineering0903 Agricultural Engineering0904 Architectural Engineering0905 Bioengineering and Biomedical Engineering0906 Chemical Engineering (include Petroleum Refining)0907 Petroleum Engineering (exclude Petroleum Refining)0908 Civil, Construction, and Transportation Engineering0909 Electrical, Electronics, and Communications

Engineering0910 Mechanical Engineering0911 Geological Engineering0912 Geophysical Engineering0913 Industrial and Management Engineering0914'Metallurgical Engineering0915 Materials Engineering0916 Ceramic Engineering0917 Textile Engineering0918 Mining and Mineral Engineering0919 Engineering Physics0920 Nuclear Engineering0921 Engineering Mechanics0922 Environmental and Sanitary Engineering0923 Naval Architecture and Marine Engineering0924 Ocean Engineering0925 Engineering Technologies0999 Other, Specify

1000 FINE AND APPLIED ARTS

1001 Fine Arts, General1002 Art (Painting, Drawing, Sculpture)1003 Art Hisppry and Appreciation1004 Music (Performing, Composition, Theory)1005 Music (Liberal Arts Program)1006 Music History and Appreciation (Musicology)1007 Dramatic Arts1008 Dance

23;220

1009 Applied Design (Ceramics, Weaving, Textile Design,Fashion Design, Tewelry, Metalsmithing, InteriorDecoration, Commercial'Art)

1010 Cinematography1011 Photography1099 Other, Specify

. 1100 FOREIGN LANGUAGES

1101 Foreign Languages, General1102 French1103 German1104 Italian1105 Spanish1106 Russian1107 Chinese1108 Japanese1109 Latin1110 Greek, classical1111 Hebrew1112 Arabic1113 Indian (Asiatic)1114 Scandinavian Languages1115 Slavic Languages (other than Russian)1116 African Languages (non-Semitic)1199 Other, Specify

1200 HEALTH PROFESSIONS

1201 Health Professions, General1202 Hospital and Health Care Administration1203 Nursing1204 Dentistry1205 Dental Specialties1206 Medicine1207 Medical Specialties1208 Occupational Therapy1209 Optometry1210 Osteopathic Medicine1211 Pharmacy1212 Physical Therapy1213 Dental Hygiene1214 Public Health1215 Medical Record Librarianship1216 Podiatry or Podiatric Medicine

221

233

1217 Biomedical Communication1218 Veterinary Medicine1219 Veterinary Medicine Specialties1220 Speech Pathology and Audiology1221 Chiropractic1222 Clinical Social Work1223 Medical Laboratory Technologies1224 Dental Technologies1225 Radiologic Technologies1299 Other, Specify

1300 HOME ECONOMICS

1301 Home Economics, General1302 Home Decoration and Home Equipment1303 Clothing and Textiles1304 Consumer Economics and Home Management1305 Family Relations and Child Development1306 Foods and Nutrition (include Dietetics)1307 Institutional Management and Cafeteria Management1399 Other, Specify

1400 LAW

1401 Law, General1499 Other, Specify

1500 LETTERS

1501 English, General1502 Literature, English1503 Comparative Literature1504 Classics1505 Linguistics (include Phonetics, Semantics, and

Philology)1506 Speech, Debate, and Forensic Science (Rhetoric

and Public Address)1507 Creative Writing1508 Teaching of English as a Foreign Language1509 Philosophy1510 Religious Studies (exclude Theological Professions)1599 Other, Specify

1600 LIBRARY SCIENCE

1601 Library Science, General1699 Other, Specify

4)3222

1700 MATHEMATICS

1701 Mathematics, General1702 Statistics, Mathematical and Theoretical1703 Applied Mathematics1799 Other, Specify

1800 MILITARY SCIENCE

1801 Military Science (ArmY)1802 Naval Science (Navy, Marines)1803 Aerospace Science (Air Force)1899 Other, Specify

1900 PHYSICAL SCIENCES

1901 Physical Sciences, General1902 Physics, General (exclude Biophysics)1903 Molecular Physics1904 Nuclear Physics1905 Chemistry, General (exclude Biochemistry)1906 Inorganic chemistry1907 Organic chemistry1908 Physical chemistry1909 Analytical chemistry1910 Pharmaceutical chemistry1911 Astronomy1912 Astrophysics1913 Atmospheric Sciences and Meteorology1914 Geology1915 Geochemistry1916 Geophysics and Seismology1917 Earth Sciences, General1918 Paleontology1919 Oceanography1920 Metallurgy1999 Other, Specify

2000 PSYCHOLOGY2001 Psychology, General2002 Experimental Psychology (animal and human)2003 Clinical Psychology2004 Psychology for Counseling2005 Social Psychology2006 Psychometrics2007 Statistics in Psychology

223

Qr7

2008 Industrial Psychology' 2009 Developmental Psychology

2010 Physiological Psychology2099 Other, Specify

2100 PUBLIC AFFAIRS AND SERVICES

2101 Community Services, General2102 Public Administration2103 Parks and Recreation Management2104 Social Work and Helping Services (other than

Clinical Social Work)2105 Law Enforcement and Corrections2106 International Public Service (other than Diplomatic

Service)2199 Other, Specify

2200 SOCIAL SCIENCES

2201 Social Sciences, General2202 Anthropology2203 Archeology2204 Economics2205 History2206 Geography2207 Political Science and Government2208 Sociology2209 Criminology2210 International Relations2211 Afro-American (Black Culture) Studies2212 American Indian Cultural Studies2213 Mexican-American Cultural 3tudies2214 Urban Studies2215 Demography2299 Other, Specify

2300 THEOLOGY

2301 Theological Professions, General2302 Religious Music2303 Biblical Languages2304 Religious Education2399 Other, Specify

233224

4900 INTERDISCIPLINARY STUDIES

4901 General Liberal Arts and Sciences4902 Biological and Physical Sciences4903 Humanities and Social Sciences4904 Engineering and Other Disciplines4999 Other, Specify

5000 BUSINESS.AND COMMERCE TECHNOLOGIES

5001 Business and Commerce Technologies, General5002 Accounting Technologies5003 Banking and Finance Technologies5004 Marketing, Distribution, Purchasing, Business, and

Industrial Management Technologies5005 Secretarial Technologies (include Office Machines

Training)

5006 Personal Service Technologies (Stewardess,Cosmetologist, etc.)

5007 Photography Technologies5008 Communications and Broadcasting Technologies

(Radio/TV, Newspapers)5009 Printing and Lithography Technologies5010 Hotel and Restaurant Management Technologies5011 Transportation and Public Utility Technologies5012 Applied Arts, Graphic Arts, and Fine Arts Tech-

nologies (include advertising design)5099 Other, Specify

5100 DATA PROCESSING TECHNOLOGIES

5101 Data Processing Technologies, General5102 Key Punch Operator and Other Input Preparation

Technologies5103 Computer Programmer Technologies5104 Computer Operator and Peripheral Equipment

Operation Technologies5105 Data Processing Equipment Maintenance Technologies5199 Other, Specify

5200 HEALTH SERVICES AND PARAMEDICAL TECHNOLOGIES

5201 Health Services Assistant Technologies, General5202 Dental Assistant Technologies5203 Dental Hygiene Technologies5204 Dental Laboratory Technologies

225

5205 Medical or Biological Laboratory AssistantTechnologies

5206 Animal Laboratory Assistant Technologies5207 Radiologic Technologies (X-Ray, etc.)5208 Nursing, RN (less than 4-year program)5209 Nursing, Practical (LPN or LVN--less than 4-year

program)5210 Occupational Therapy Technologies5211 Surgical Technologies5212 Optical Technologies (include Ocular Care,

Opthalmic, Optometric Technologies)5213 Medical Record Technologies5214 Medical Assistant and Medical Office Assistant

Technologies5215 Inhalation Therapy Technologies5216 Psyciatric Technologies (include Mental Health

Aide Programs)5217 Electro Diagnostic Technologies (include EKG,

EEG, etc.)5218 Institutional Management Technologies (Rest Home,

etc.)

5219 Physical Therapy Technologies5299 Other, Specify

5300 MECHANICAL AND ENGINEERING TECHNOLOGIES

5301 Mechanical and Engineering Technologies, General5302 Aeronautical and Aviation Technologies5303 Engineering Graphics (Tool and Machine Drafting

and Design)5304 Archite..tural Drafting Technologies5305 Chemical Technologies (include Plastics)5306 Automotive Technologies5307 Diesel Technologies5308 Welding Technologies5309 Civil Technologies (Surveying, Photogrammetry,

etc.)5310 Electroni,.ls and Machine Technologies (TV, Appliance,

Office Machine Repair, etc.)5311 Electromecl.anical Technologies531.2 Industri,. Technologies531s Textil ,,. l'achnologies

b3l4 In ..tation Technologies

5315 Mf.lchanical Technologies

5316 Nuclear Technologies5317 Construction and Building Technologies (Carpentry,

Electrical Work, Plumbing, Sheet Metal, AirConditioning, Heating, etc.)

5399 Other, Specify

5400 NATURAL SCIENCE TECHNOLOGIES

5401 Natural Science Technologies, General5402 Agriculture Technologies (include Horticulture)5403 Forestry and Wildlife Technologies (include

Fisheries)5404 Food Services Technologies5405 Home Economics Technologies5406 Marine and Oceanographic Technologies5407 Laboratory Technologies, General5408 Sanitation and Public Health Inspection Technologies

(Environmental Health Technologies)5499 Other, Specify

5500 PUBLIC SERVICE RELATED TECHNOLOGIES

5501 Public Service Technologies, General5502 Bible Study or Religion-Related Occupations5503 Education Technologies (Teacher Aide and 2-year

Teacher Training Programs)5504 Library Assistant Technologies5505 Police, Law Enforcement, Corrections Technologies5506 Recreation and Social Work Related Technologies5507 Fire Control Technology5508 Public Adminsitration and Management Technologies5599 Other, Specify

211227

APPENDIX C-1

Number of Four-Year Public Insfitutions withBudgets Reviewed by Techniques Discussed

States

Number offour-yearpublicinstitu-tions

Number ofcampusesincludedin thisreport*

Number of campusesexcluded fromthis reportSeparatemedical

Otherspecial

California 31 28 1 2

Colorado 13 12 1 0

Connecticut 6 5 1 0

Florida 9 9 0 0

Hawaii 2 2 0 0

Illinois 13 12 1 0

Kansas 7 6 0 1

Michigan 15 15 0 0

Mississippi 9 8 1 0

Nebraska 7 6 1 0

New York 40 22 5 13

Pennsylvania 24 23 1 0

Tennessee 11 10 1 0

Texas 34 30 4 0

Virginia 15 15 0 0

Washington 6 6 0 0

Wisconsin 13 13 0 0

* Medical programs on these campuses not incluf:;Ad.

228

212

APPENDIX C-2

Number of Two-Year Public Institutions with BudgetReviews SFmilar to Four-Year Institutions

States Communitycolleges

Vocational,technicalinstitutions

Universitybranches

California 0 0 0

Colorado 7 0 0Connecticut 12 4 5

Florida 0 0 0

Hawaii 7 0 0

Illinois 0 0 0

Kansas 0 0 0

Michigan 29 0 0

Mississippi 0 0 0

Nebraska 0 0 0

New York 0 10 0

Pennsylvania 0 0 25Tennessee 10 0 0

Texas 0 0 0

Virginia 23 0 1

Washington 0 0 0

Wisconsin 0 0 14

229

2 4 3

APPENDIX D

An Interpretation of a Non-Zero Sum Game:The Prisoner's Dilemma

Two suspects are taken into custody andseparated. The district attorney is certainthat they are guilty of a specific crime,but he does not have adequate evidence toconvict them at a trial. He points out toeach prisoner that each has two alternatives:to confess to the crime the police are surethey have done, or not to confess. If theyboth do not confess, then the districtattorney states he will book them on somevery minor trumped-up charge such as pettylarceny and illegal possession of a weapon,and they will both receive minor punishment;if they both confess they will be prosecuted,but he will recommend less than the mostsevere sentence; but if one confesses andthe other does not, then the confessor willreceive lenient treatment for turning state'sevidence whereas the latter will get "the book"slapped at him. In terms of years in apenitentiary, the strategic problem mightreduce to:

Prisoner 2

Primner 1: Not Confeu Confess

Not Confem 1 year eacil 10 years for 1 and

3 months fur 2

Confess 3 munths for 1 and 8 years each

10 yean for 2

230

24 4

. . The problem for each prisoner is todecide whether to confess or not. The gamethe district attorney presents to theprisoners is of the non-cooperative variety.(Luce & Raiffa, 1957, p. 95)

If.both suspects make their decisions to confess ornot on the basis of pure rationality, both will confess,and as Dror (1968) comments, they will spend the nexteight years thinking about the limits of pure rationality.This is the rational choice for prisoner 1 because ifprisoner 2 does not confess, then prisoner 1 gets onlythree months instead of one year if he confesses. Ifprisoner 2 does confess, then prisoner 1 must also con-fess to avoid serving ten years. The same reasoning holdsfor prisoner 2. However, this is clearly a worse outcomethan if both were not to confess (on the basis of a hunchor extrasensory perception) and then serve only one year.

215

231

Publications in the Series ofState Budgeting for Higher Education

Clenny, L. A., 0- al. Yoh' innigetinv, for higher (duration: Data digestBowen, F. M.. Buvle, J. I.. & (jenny, I,. A. Stair.. budgeting for hIuher

education: Midget proce.vses in the .50 states11,10e, J. TI., & (jenny. I_ A. Statc budLi,cting for higher education.

7rmd. 1t :te rercraw appropriations from MRS-1915Schniidtlein, F. & Clentiv. I. A. :;tate budgeting for hiulwr Mara-

tion: The political ei'Onomr; of the process-Purve, H. A.. & Clerniv, I.. A. Stale budgeting for hi,oher rducathm: In-

formation visionsv mul technical (author'sC.k.miV, I.. A. Jtalf, for higlicr education: Interagency conflict

and Ci)/ISCl/SUS

Nkisinger. B. J., Jr. ,tole budgeting for higher education: The uses offormulas

Bowen, F. I., & C:lennv, L. A. State budgeting for higher educati(w.State fiveal stritownen and public hIgher education

246