V n _.;.,::.,:: 'l l - -‘ - - I— MUNICIPAL POLICY : MAKING .viDISSERTATION FOR-’"iTHELE‘ FUNDS ' PUBLIC ' ARATIVE. STUDY OF: BUDGETING A COMP 1.5mm _‘;UNlVERSl-'I'Y ; MICHIGAN 1,, w...w~.. . ———————i Mm ”iii ii”liiiiiiiiliiiiliiiiiiiiiiiiiiiiii I 4 .- i an 3‘ 3 1293 10165 4139 § g .iversit y J 94.}, a, £16m»: —\—a! This is to certify thét'flie. » -'~'-’. thesis entitled MUNICIPAL POLICY MAKING: A COMPARATIVE STUDY OF BUDGETING PUBLIC RESOURCES presented by LEWIS FRIEDMAN has been accepted towards fulfillment of the requirements for «REAL—degree mm SCIENCE /~\ 7 /. C 3/ / . I , r // L/‘C («4% L *2 ".7 .1; 1+ L Major professor , / f IDate 212/15 ”7} it 0-7639 ABSTRACT MUNICIPAL POLICY MAKING: ACOMPARATIVE STUDY OF BUDGETING PUBLIC FUNDS By Lewis B. Friedman The problem posed in this dissertation is first to demonstrate that variation occurs in the process of governmental budgeting and then to show its relationship to expenditure outputs. Data was gathered through structured interviews with department heads. executives, and members of the legislature in fourteen middle-sized Michigan cities. This achieves a comparative focus that enables systematic measurement among cities and the testing of the direct connection between decision making and the substantive Spending choices made. Both objectives of the research were achieved. Not only were fundamental differences among the cities dis- covered, but these were directly translated into concrete expenditure outputs. Department heads ask for more than they received In the previous year; the executive reduces their requests J u JL— .c... ... ... ... u... ... . n... u, .. . .u I o r O i - .pr f .08'. u -. 4 .s. .u. hilt?» F Lewis B. Friedman ' but still recommends an increase to the legislature; who similarly cut the budget they receive for review, but appropriate an increase in spending levels. The more that is initially sought, the more the executive reduces, but the more he still recommends. The more recommended, the less is reduced but the more the legislature appropriates. The concept of incremen- talism does not provide a fruitful explanation of these expenditure outputs or statistical interrelationships. Departments occupy a spending budget role, not because they pad their requests, but because they are the advocates of their programs. The executive is the econ- omizer, as he perceives departments to pad their requests and has the responsibility to balance expenditures with revenues. The legislature oversees the administrative branch of government, by changing the budget they receive, as well as by cutting it. But looking at these two actors solely in negative terms, fails to account for the in- creases they themselves Support. Formal influence is exercised on the basis of each actors legal responsibilities to participate in making budget choices and is made evident during the official review of spending proposals. The department head has the opportunity to persuade the executive, but this does not affect their requests, nor the executive's reductions. It does enable the departments to achieve their goal of J‘s— Lewis B. Friedman annual growth. The legislature is not a rubber stamp, as they exercise their authority to change, out, the executive's budget and then limit the size of the annual growth in funds. The departments' appeal to the legis- lature is more to keep what they already have than to restore executive reductions. Informal influence is exercised through anticipated reactions and behind the scenes explicit warnings. It does not actually suppress the articulation of expansionary budgets, but does decrease the amount of reductions imposed during its subsequent formal review. Items that are not approved and would be eliminated are not brought forth for public and official decision. Through limiting the amount of leeway enjoyed by the departments in formulating their initial requests, and the amount of inter-departmental competition over the distribution of funds, the executive strives to control the budget from the very start of the sequence of decisions. The more he does so, the less departments seek and the less they receive in the executive's recommendations. The distribution of influence is not concentrated in the executive but more segmented. Control over the departments is not associated with dominance over the legiSlature. The preparation of the administrative budget is divorced from the legislature's adoption Of the appro— Priation ordinance . L_—__‘— Lewis B. Friedman The difference between an incremental and a synoptic dedsimimaking process is associated with a control, mmmgemmfin and planning budgetary orientation. These arerespectively concerned with: the minimization of the yawlyincrease in the costs of purchases, the efficiency andeconomy of services, and the effectiveness of spending h>adfieve policy goals. Each is present, as the ad- lnrawe'M)one purpose does not preclude the application ofancuwr. However, only the management approach is cmnwcted with spending outputs as it serves as a mechanism tocontrol the pace of spending growth. The budgetary process is linked to its environment tMpugithe participation and influence of individuals amigrmum from the community and through the attitudes ofpmblic officials. In the first case, little impact is mfident in expenditures; but perceptions of limited resmnce capabilities serve as an inhibitor, and activist prmkrences toward the scope of governmental responsi— bihties encourage the expansion of spending levels. MUNICIPAL POLICY MAKING A COMPARATIVE STUDY OF BUDGETING PUBLIC FUNDS By Lewis B. Friedman A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Political Science 1973 Copyright by LEWIS B. FRIEDMAN 1973 ii To "N" and M. E. . s ‘.\ ACKNOWLEDGMENTS In the course of my journey through the "groves of aoademe," I have met some of the best and some of the worst examples of human beings and educational institutions. To those who stood in the way, the less said the better; but to those people who genuinely gave of themselves a special word of thanks is in order. Without the encouragement, understanding, intel- lectual stimulus, and friendship of Bryan T. Downes, who stood with me throughout, none of this would have come about, nor would it have seemed worthwhile. The debt I owe him is large, and can only begin to be repaid by the publication of this manuscript. A second warm acknowledgment belongs to Rita Schlesinger Gallin, and to the forbearance of her husband, Bernard Gallin. She too was supportive up to the end, and beyond; and without her able assistance, this study would have appreciably suffered. Her home was always warm and open. The material assistance of Philip Marcus, Director Of the Urban Survey Research Unit of Michigan State University and its staff needs also to be noted. Director Marcus enabled me to carry out the data collection iv beBlond my own expectations. An appreciative word to those colleagues who, in various ways , displayed assistance: Larry Adams, Josh Bernstein, Rufus Browning, Steven Burks, Ralph Kamisky. Fred Lane, Frank Pinner, Norman Powell, Patrick Rabet, and Eli Silverman. To Ellen Adams who so consciensciously typed the manuscript in the face of numerous obstacles I placed in her way, I say, a job well done. The financial support of the National Science Foundation. which made the entire effort possible through their program of Dissertation Research Grants, must also be noted. To that group of scholars whose earlier efforts provide the intellectual foundation of the present study: Thomas Anton, John Crecine, Heinz Eulau, Robert Eyestone, Richard Fenno, Arnold Meltsner, Kenneth Prewitt, Allen Schick, Ira Sharkansky, and Aaron ’Nildavsky, I say thank you. I also wish to thank the 169 public officials who graciously consented to be interviewed and who gave generously of their time deSpite very busy schedules. TABLE OF CONTENTS Chammr One: A Model of Municipal Budgeting...........l IntrOduCtionOOII.0.0.00.0OIOOIOIOIIICOOIIOOOIOl'l Budgeting as a Policy Making Process............2 Dependent Variables of Expenditure Outputs......6 An Explanatory Model of the Budgetary ProceSSlOIIIIOOOIOOO'OOOI.IOOIOIIIOOOOOUOOIOOI'J—Z Budget RoleSOIOOOOIOOOOOIOOIIOII000.00.00.0000013 Decision Making Structure......................15 Formal Authoritytfittclttclit-00......l000000l0017 Informal Influence.0...0.0...OIOIOOOOIOIIOIOOOI18 Executive Supervision of Departmental DeCiSiOnP/Iakingliol'lotIlltOCOIOIOIIIOOOOIOOOQ|19 The Distribution of Influence..................l9 Cognitive and Evaluative Process of Choice.....21 Linkage to the Environment.....................2 Plan of Analysis...............................27 Mmpter Two: Research Design.........................30 IntroductionOOlOOOOCOOOOOOOOOOOOOOIIOOOOOOOOC.030 Consistency of the Municipal Budget Process....30 Generalization.................................33 Representativeness.............................35 Data Collection...‘O...lI...IIOOOIOOOIOOOIIOOOOBB Interviews..................................... 1 Level ofAnalysisllOOlltDIOOIOIIOOOOOOOIOIIOOQ046 Statistical MalySiSOODI...OOIIICIOIOIIOOIOCOOQS2 Mmpter Three: Expenditure Outputs...................57 Introduction.OIOIOOQIIIOI.OOOIOOCCOOIIOIOOQOI0057 Criteria for Selecting the Dependent VariableSo-oococoon-onion...oocuocoouIc-nnooono57 ChangeOOIOCOUIOIOICO....I...OOOIOOOOCOIOOOCUIIOB8 TotalSpending...0|..ll.0OOOOOOOUOOCODIOOOOCOOOO Substantive Budget Outputs Department Budget Requests.....................64 Executive Budget Behavior Response to Departmental Requests..............68 Executive Budget Recommendations...............72 Legislative Budget Behavior Response to Executive Recommendations..........75 Lwfislative Budget Appropriations..............8O IncrementalismOOOO0......OCOIOOOOOOOOOOOOOOOOO. AbSOlute Size 0f ChangeOO0.0.00.00009000000000085 IEqmnditure Interrelationships.................88 Wiflfin.Year Change-- Executive Reductions hirmpartmental Requests.......................89 Yearly Changes Examnflye Budget Recommendations...............94 legislative Appropriations.....................96 ConCluSionOOOOO0.0000000000000IIOO0.00.00.00.0098 Chapter Four: BUdget ROleSoooo-oooooocoooooocoooooooloo Introduction..................................lOO Ikpartment Budget Role........................lOl Departmental Advocacy...0000000000000.00.90.001.03 AnalySiSOOO...OOOOOOOOOOOOOOOOIOOOOOO00.0.0000106 Department Padding..........oo.o..............llO malySiSOOCOOCOOOCOOOOOOOO...COOOOOOOOOOOOOIOOlll (kmmerison of Two Role Orientations...........113 Relationship to Department Requests...........114 * summaryOOooooooooocoo-000.0000...coo-0000.00.01.18 Ekecutive Budget Role.........................ll9 Budget Balancing..............................120 AnalYSiSOOOOO...00......OOOOOOOOOOOOOOO0.0.0.0121 Evaluation of Department Padding..............12u AnalySiSOOOOOOOOOCOOOO0.0000.0.00.00.00.000000127 Relationship to Executive Budget Behavior Review of Departmental Requests...............132 Emecutive Budget Recommendations..............134 Conclusion....................................135 legislative Budget Role.......................136 AnalySiScoocoooooooooooocoo00.0000000000000000139 Relationship to Legislative Budget Behavior Review of Executive Recommendations...........144 legislative Appropriations....................146 Conclusion....................................147 The System of Municipal Budget Roles..........150 CMMNer Five: Influence Exercise Through Formal AuthoritYOOOOOOOO00.0.0000000000000000153 IntrOduCtionOOO0.0000IOOOOOOOO00.0.00000000000153 Department-Executive Influence Relationship...l58 Relationship to Department Budget Requests....l70 Relationship to Executive Recommendations.....l72 vii Executive-Legislature hfiluence Relationship........................l73 AnalySiSIOOOI0.0.0.0....OOOOO'OOOQIO'OOIOOOOC0177 Relationship to Executive Budget Behavior.....l83 Relationship to Legislative Budget Behavior Legislative Changes in Executive Recommendations.....................l84 Relationship to Legislative Appropriations....l88 Department-Legislature Influence Relationship.O..00...OOOOOOOOI‘COOIOODOIIIO0.0189 Malysislcloi69......OOl00.0000000000000000000194 Relationship to Department Budget Requests....202 Relationship to Executive Budget Behavior.....203 Relationship to Legislative Budget Behavior Change in Executive Recommendations...........205 Relationship to Legislative Appropriations....205 Chaner Six: Informal Influence.....................208 IntroductionolitttooloOOOOICOOOICII000.00.00.0208 Mechanism of Prevention.......................212 Anticipated Reactions.........................213 EXpliCi‘t WarningS.....coo..o..................2114« Department‘EXeCutiVeooon!octonoooonoooooonuooQZl8 Executive-Legislature.........................223 Relationship to Yearly Change in Spending LevelS............................227 Relationship to Budget Change.................230 ConcluSiOnOOIIIOOCIODOOCOOOICOOIOOIIOIIIDI.000232 antm?Seven: Executive Supervision of Departmental Decision [VlakingOOOIIIOOOI'ODOI‘OOOOOIZBLJ’ IntroductionOIIDCCIOOOIOOCII...O.I.IIO"O'IIDQ23LP Departmental Leeway...........................236 AnalysiSIICOOOCIICCOOICCOIOIIOIIIOCOIOOIOUOODJZLFO Relationship to Department Budget Requests...............................244 Executive Budget Behavior Change in Departmental Requests...............246 Conclusion....................................248 Interdepartmental Competition.................250 Arlalij-SO'COICOOOU.IC...0.0.OOIOOOOOOIOOOIIOIOZSL" Department Budget RequeStS....................260 Executive Budget Behavior Change in Departmental Requests...............262 Executive Recommendations.....................262 CODClLlSiOn...ooo...o...........o.-oooc...o.o..263 viii fl” Mmptmrfifight: The Structure of Budget Influence....266 IntrOduCtionouooon00000000030000.0-cove-000000266 Fomal AuthorityIIIII'II'IIIIIIIIIIIIIIIIIIIII269 Informal InfluenceIIIIIIDIIICIIIIIIIIIIIIIIIIIZBO Executive Supervision of Departmental Decision Making..................286 COnCluSionIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII290 antaeNine: Cognitive and Evaluative Mechanisms ofChOiceIIIIIIIIIIIIIIIIIIIIIIIIIIII0293 IntroduCtiOnI I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 293 Con-troll I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 298 AnalysiSI I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 303 Relationship to Budget Behavior...............309 Management I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 31]. AnalySiS...oooonoouono.oo.cacovocooo-oocununao314 Relationship to Budget Behavior...............317 PlanningI I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I320 AnalySiSc I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 321‘" Relationship to Budget Behavior...............325 GonClUSiono I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I u 328 Mmpter Ten: External Pressure Introduction..................................331 Group Participation...........................334 Group InfluencelIIIIIIIIIIIIIIIIIIIIIIIIIIIIII343 Yearly Change in Expenditure Levels...........348 Change in Expenditure Levels from One Stage to the 1ext....................350 ConcluSiOnIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII360 Mupter Eleven: Decision Makers' Attitudes Toward the mvironmen’tIIIIIIIIIIIIIIIIIIIII362 IntroductionlIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII362 Expenditure Constraint........................366 AnalysiSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII.0368 Revenue Consciousness.........................374 AnalySiSIIIIIIIIIIIIIII .C.I'...'.'.'...'..U'..376 Relationship to the Annual Change of Expenditure Levels.........................377 Relationship to Change from One Stage to the Next....................381 Scope of Government...........................382 AnalySiSoooIIIIIIIIIIIIno000:100IIIIIIIIIIIIIIB87 Relationship to Budget Behavior...............393 Conclusion....................................395 ix Chapter Twelve: Summary and Conclusions. . . . . . . . . . . . . 396 Introduction”.o....o-.o.........o............396 Summary. I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 397 Budgeting as a Decision Making Process. . . . . . . . 05 Limitations of the Study......................407 AppendicesI I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I [+1.]. Appendix AI I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I all Appendix B I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 419 Appendix C I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 422 Bibliography. I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 426 General ReferenceSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII‘J’BB Kl L2 &3 3.1 12 $3 in 15 l6 19 110 M1 4.2 4.3 ]—__ LIST OF TABLES Responses to Interview and Questionnaire..........42 The Three Departments' Requests as a Percent of the Total Budget at the End of the First Stage 0f Decision MakingIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIqu The Three Departments' Requests as a Percentage of Total RequeStSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIILLS Department Budget Requests........................65 Direction of Change in Individual Department Budget RequeStSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII67 Executive Change in Department Budget Requests....69 Direction of Executive Change in Individual Department Budget Requests........................7l Executive Budget Recommendations..................73 Direction of Change in Executive Recommendations for Individual Departments........................74 Legislative Change in Executive Budget Recommendation.IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII76 Direction of Change in Individual Departments in Executive Recommendations......................78 legislative Appropriations........................8l Direction of Change in Individual Departments in Legislative Appropriations.....................82 Department Head Budget Role as an Advocate.......107 Department Head Budget Role as a Padder..........112 Executive Budget Role as a Balancer..............122 xi k4 m5 i1 7.2 13 Executive Budget Role as an Evaluation of Departmental Padding-coo-oloooooonoooonooouoooa00129 Legislative Budget Role as Oversight Of AdminiStrationOOOOCDOOCO.0...UOOOOOIOIOIOOtfifltluO Department-Executive Formal Influence Relationship (Department Head Replies)...........l62 Department-Executive Formal Influence Relationship (Executive Replies).................163 Executive-Legislature Formal Influence Relationship (Executive Replies).................l78 Executive-Legislative Formal Influence Relationship (Legislative Replies)...............l79 Department-Legislative Formal Influence Relationship (Department Head Replies)...........l95 Department-Legislative Formal Influence Relationship (Legislative Replies)...............l96 Department-Legislative Formal Influence Relationship (Executive Replies).................204 Department-Executive Informal Influence Relationship (Department Head Replies)...........219 Department-Executive Informal Influence Relationship (Executive Replies).................220 Executive-Legislature Informal Influence Relationship (Executive Replies).................224 Executive-Legislature Informal Influence Relationship (Legislative Replies ...............225 Departmental Leeway (Department Head Replies)........................2#1 Departmental Leeway (Executive Replies)..IOOOOOOOODIIOUUIOCOOOIIOOIOOZQB Interdepartmental Competition (Department Head Replies)........................255 7.4 $1 $2 %3 1&1 1&2 1&3 1&4 1&5 1L1 1L2 1L3 1L4 1L5 L1 L2 Interdepartmental Competition (Executive ReplieS)OOOOODCOIIOOOIOOOOOOIOICOOOOIOZ56 Incremental Budget Orientations (Impartment Head Replies)ocuoou-onoooooooonoonnoo305 Incremental Budget Orientations (Executive Replies)..............................306 Incremental Budget Orientations (LegiSlative Replies).OOOIOIIOIOOIIOOOIICO.I'0000307 Kinds of Groups that Make Contact................336 Disjuncture of Group Contact.........,...,,,,,,,,337 External Pressure (Department Head Replies)........................339 External Pressure (Executive Replies)lli.l.....OC'IOIOIOOIIIIOOIOU0314’]. External Pressure (Legislative RepliesoaoooottDOOOOOOOOlIllooQOGCIOBL‘J‘i’ Resource Capability (Department Head Replies).....OUIOIUDIIIIOQIOCOOOB69 Resource Capability (Executive ReplieS)OCGCIICO0.0000...00.00.100.000370 Resource Capability (Legislative ReplieS)....-....o..co.a..-.......o.37l Scope of Government (Executive Replies)000000000OIIIOOOIOOOOOOOOOIICOBBB Scope of Government (Legislative Replies)............................389 Expenditure Outputs—- Average of Individual DepartmentS................416 Expenditure Outputs of Three Department Heads....425 xiii 1.1 8.1 10.1 LIST OF FIGURES Model of Three Stages of Municipal Budgeting. . . . . . . . . 8 The Structure of Budgetary Influence. . . . . . . . . . . . . . . 270 External Participation and Influence. . . . . . . . . . . . . . .360 y Chapter One: A Model of Municipal Budgeting Introduction1 The study of governmental budgeting has long occupied the attention of those concerned with public administration and public policy. Some scholars have directed their efforts toward the development of a normative standard for making spending choices, others have been concerned with structural and procedural reform. Still others have provided an empirical description of the way the appropriations ordinance is prepared and adopted. The budget has come to serve multiple purposes as an arena of governmental decision making. It is a mechanism to assure the legal authorization of funds; a tool to achieve the efficiency and economy of agency activities; and an opportunity to establish the policy goals of government itself. To implement these multiple aims available to decision makers have been modified from line-items of expenditures to activity accounts of l. A previous formulation of the first and second chapters appeared in Bryan Downes and Lewis Friedman, "Local Level Decision Making and Public Policy Outcome: A Theoretical Perspective," in H. Hahn, ed., Urban Politics and Peoples (Beverly Hills: Sage Publications. 19727, vol. III, passim. 2 services and then again to categories of program effective- ness. Furthermore, the relative influence of budget participants have shifted. Starting with a diffuse structure of authority centered in the legislature, where the executive was noticeable weak, a centralized system of executive responsibility and control has emerged. The present research builds upon the foundation of previous efforts as it directs the study of governmental budgeting toward a policy-analysis framework. In particu- lar, it extends the contemporary concern with an empirical description of the budgetary process to a focus upon its specific expenditure outputs through a comparative analysis of fourteen cities, the connection between how decisions are made and what decisions are made is exam- ined. Budgeting as a Policy Making Process Beneath all of the efforts to devise the "best" methods of budgeting lies its fundamental political interpretation as the means by which resources are author- itatively allocated. The budget is a financial statement of the programs and policies society intends to pursue. Expenditures are the common denominator of all other political decisions, as without funds the activities 015‘ government would come to a halt. It stands at the center of the political system. The results of the .. l .‘I s . .. \'\. \ 3 competition between different interests and values are recorded in the appropriations ordinance. Who wins and who loses; who receives the benefits and who bears the costs can be discerned in the dollars and cents dis- tribution of the budget. As Aaron Wildavsky said of this view of budgeting: If politics is regarded in part as conflict over whose preferences shall prevail in the determination of national policy, then the budget records the outcomes of this struggle. If one asks, "Who gets what the government has to give?" then the answers for a moment in time are recorded in the budget. If one looks at politics as a process by which the government mobilizes resources to meet pres- sing problems, then the budget is a focus of these efforts.2 However, except for a few notable exceptions, the study of governmental budgeting has not been directed to the explanation of spending outputs.3 All too often, the financial content of the decisions made have been for- gotten. Instead, concentration has been placed solely on the process, the "how's" of budgeting, while actual 2. Aaron Wildavsky, The Politics of the Budgetary Process (Boston: Little, Brown and Company, 1964), p. 4. 3. These exceptions include Ira Sharkansky, "Agency RequestS, Gubernatorial Support and Budget Success in State Legislatures," American Political Science Review, Vol. LXIX, no. 4 (December, 1968), pp. 1220-1231, and Ira Sharkansky and Augustus Turnbull, "Budget Making in Georgia and Wisconsin: A Test of a Model," Midwest Journal gfhPolitical Science, vol. XIII, no. 4 (November, 1969), Pp. 631-645. u, expenditure choices have been ignored. An illuminating example of this failure is Thomas Anton's study of Three Cities in Illinois. Here, dif- ferences in such features of the decision making process as the, ". . . (formalization, professionalization, rationalization): length of time devoted to budgeting, amount of information gathered, existence or non-existence of the "budget" concept, and the presence or absence of budget review” are reported. Yet at no point is there an analysis of the spending policies of these three cities. Consequently, the effects of this variation in the bud- getary process upon expenditure outputs cannot be assessed. What difference does it make that these cities go about budgeting in various ways? This question is simply left unanswered. A second illustration of this limitation is in the movement for a Program Planning Budgeting System. h’luch has been written of the difference between the traditional and the PPBS mode of budgeting, yet little empirical evi- dence has been reported of the impact such a change would have upon governmental spending policy. This does not involve an evaluation of the "quality" of decisions made, as stated by Charles Schultze: "In my own View. PPBS and .4. Thomas J... Anton, Budgeting in Three Illinois Cities, Commission Papers of the Institute of Government and Public—— Affairs (Urbana Illinois: University of Illinois Press, 1964), p0 o ".9 o . n. c ' I a ‘1 A t - .- u t. --.,‘ v. ' , I - 5 systematic analysis can improve and indeed already have improved the quality of program and design and the al- location of budgetary resources within the civilian sector of government."5 Such a conclusion cannot be made on any other basis except the personal preferences of the author, as it rests on the implicit, but unverified, assumption that different budget choices are forthcoming from the PPB system compared to the decision making process it replaces. Before it can be assessed that "better" expenditure choices are made, it has to be first empirically determined that different budget decisions do, in fact, follow from the employment of such a new budgetary process. So far, no evidence has been presented that this is indeed the case. Some evidence must be brought to bear on the question of whether the substance of decisions will be altered on the basis of a new PPBS process of budgeting, for why else propose a change unless there is the expectation that Spending outputs will be altered. Change in budgetary decision making for itself is pointless. Allen Schick recognizes the importance of this question, although he is also without the necessary data to answer it. As he writes: In an operational sense, however, what difference does it make whether the central budget process is oriented toward planning rather than management? . 5. Charles Schultze, The Politics andBconomics of gblic Spending (Washington D. 0.: The Brookings Institute, 9 pp. ‘5. 6 Does the change merely mean a new way of making decisions, or does it mean different decisions as well? The case for PPB rests on the assumption that the form in which information is classified and used governs the actions of budget makers, and, conversely, that alterations in form will produce desired changes in behavior. Take away the assumption that behavior follows form, and the movement for PPB is reduced to a trivial manipulation of techniques—- form for form's sake without any significant bearing on the conduct of budgetary affairs.6 Cmmequmnfly, the primary objective of the research presented hue mimunicipal budgeting is to show that such a inmageexists: That variation in the activity of public offhflals who formulate the budget, and that the differ- entwayscfities prepare and adopt the appropriation ordnance has an affect upon spending outputs. Budget- my'dmflsion making, can account for identified differences ammg mndcipal spending patterns. The budget process, astheindependent variable is directly related to its dwmndent variable of expenditure policy. In the linkages hemmen the environment, the political system, and public pohcy Hm traditional study of governmental budgeting has imneased relevance as an explanation of expenditure outputs. Dependent Variables of Expenditure Outputs It is first necessary to define the measures of ex— Pmflihue that are the end-products of the budgetary process 6. Allen Schick, "The Road to PPBS: The Stages of flmgetReform," Public Administration Review, vol. XXVI, no.l+(December, 1966), p. 256. .- 1 7 and that serve as the dependent variables of analysis. These emerge from the process the municipal budget is prepared and adopted, and can be broken down into three 7 distinct, yet interconnected stages. As drawn in Figure 1.1, this is the sequence of department head, executive,8 and legislative decision making. Each stage makes its own decisions that can be represented by its own particular expenditure outputs. 7. Budgeting is a more continuous, year long process, that includes decisions before and after the three stages of preparation and adoption presented in this simplified model. Before or during the depart- ment's formulation of their requests, the executive calculates estimated revenues. After the budget is formally approved there is its actual execution. This involves the disbursement of funds through an allotment system; including accounting controls over personnel and purchasing. The legislature is further involved in this post-adoption stage by the legal requirement of their approval for purchases over a certain dollar amount, as well as transfers of money from one account category to another, or from one department to another. Furthermore, if revenue forecasts prove inaccurate and a deficit becomes apparent, all three actors may be drawn into the decision making process once again to eliminate expenditures in order to maintain the balanced budget. None of these budget decisions, however, are examined in the present study which concentrates only upon the process by which the first, official budget 18 prepared and adopted. 8. The present model combines the executive Stage into a single process, aggregating the possibility of separate decisions of the executive by himself and those of his staff. On the state and national level this has been more rigorously demarcated as there exists a Bureau Budget that possesses structural independence. But this does not exist in these size cities and there- fore little is lost in merging them together as a single executive decision stage. _ m. ‘ 8 Stage I Stage II Stage III Depart- Requests Executive Recommen- ‘ Legis- Appropri- ment dation lature ations \ , - ._,__ 7 ___7____, _ A g -H, ,, / Egviewkand Eévi ew éand Change Change FIGURE 1.1 MODEL OF THREE STAGES OF MUNICIPAL BUDGETING 9 Consequently, two measures of spending choice emerge. One isthe percentage change in total ex- penditure levels at the end of each stage compared to the level of the final appropriation ordinance of the previous year. This vertical dimension of expenditure outputs, represents policy change over time; in this case a single year. How much more or less are departmental requests, executive recommendations, and legislative appropriations in relation to expenditure amounts of the previous fiscal year. The second measure is the percent change in the level of total expenditures from one stage to the next. This horizontal dimension, the amount of change within a single year, records the reaction of each actor who has formal responsibility to review the budget proposals of others. It is measured twice,9 once between the department head requests and the ex- ecutive recommendations as the executive reviews initial spending figures, and then between executive recommendations 9. A third comparison of within-year change that can be made is between initial department requests and final legislative appropriations. This is utilized in the research by Ira Sharkansky, cited in footnote three, on state politics, but is not included in the present study for such a measure ignores the logic of the se- quencing of decisions that is the foundation of the present model of the budgetary process. Furthermore, this measure has no empirical reality. The legislature does not review departmental requests, but the executive's recommendations. Such a measure of inter-stage change in expenditure levels in budgeting is an artificial con- struct of the researcher that does not represent actual SPending decisions. 10 and final legislative appropriations and the city council makes a similar review of the budget submitted to them for authoritative approval.lo These two measures of expenditures are simple, yet capture actual choices made in the course of municipal budgeting. The five separate measures can be arranged in a real-world time sequence, that represents the choices made in each stage and which together comprise the budgetary process. The outputs of one stage are the inputs into the next. There is an interdependence of actions which defines the decision making system. The executive directly responds to the departments: the legis- lature reacts to the departments and the executive as both the executive and the legislature feedback upon depart- ment heads. The first stage of budgeting is composed of the process internal to each department as requests for the forthcoming fiscal year are prepared for submission to the executive. The decision to be made here is how much to seek in comparison to how much was received in the Previous appropriation ordinance (how much the depart- ment is currently spending). Although departments in 10. These measures were previously utilized by Ira Sharkansky in "State Legislatures," and Richard Fenno, Jr., The Power of the Purse: @propriations in Cogress (Boston: Little, Brown and Company, 1966), Passm. ,J “.1 ‘N 11 taml ahmmt uniformly request more and thus provide a pmfitive thrust to municipal spending, extensive dif- fermmes are evident. The range of requests extends frmnan actual decline to an increase of almost one half. These initial spending figures comprise the raw maumials for executive decision making. Here the mwstimiis what kind of response does the executive unkeas he goes about reviewing the departmental ex- pmwimmry budgets. Universally, the choice is made to remme expenditure. But once again variation is very mmh hievidence as executives cut requests from a nmfinalcnm percent to the substantial elimination of ahmmt one quarter of what was first sought. Yet, in thenwasure of the executive's own recommendations for mmualchange in spending levels, to the legislature, finreis still an increase above the previous year. De- patments obtain at the end of the second stage less thmiwhat they sought, but more than they already pos- sess. This too varies, from a further decline in the Mty where departments first proposed such a drop, to maincrease of fully one third above the preceding appropriation ordinance. The executive's budget recommendation, in turn, is Hm input for the third decision making stage. Now thecity council faces a similar problem of responding ~v 12 to executive proposals for annual growth. Although, on the average, the budget is reduced in some cities it is either increased, decreased, or left unchanged. The extreme variability of legislative spending choices are further evident in the change of the appropriation ordinance compared to the one adopted in the previous year. Again, on the average, spending displays an annual increase. but in four cities there is an absolute decline in total funding levels. So it is evident that extensive variation in each of these five measures of the dependent variable of municipal expenditure outputs does occur. Not all cities display the same pattern of spending choices. Differences among the fourteen cities, can be explained by its relationship to the municipal budget process. Variation in one is the explanation for variation in the other. The way the authoritative budget ordinance is prepared and adopted is explicitly and formally con— nected to the substance of the decisions made. Differences in the budgetary process among cities account for dif— ferences in expenditure outputs. An Explanatory Model of the Budgetary Process Next it is necessary to present the characteristic features of the budgetary decision making process that are the explanatory concepts. The independent variables account for variation in expenditure outputs. Looking at the u... .v \I 13 municipal budget process as a system of decision making draws attention to four different components to first explain how it operates and then the connection to its spending end products. Each of these concepts can be found in earlier studies of governmental budgeting, as the intention of the present effort is to amplify and qualify their meaning within the context of municipal budgeting. Their description in the following pages attempts to summarize the general conclusions of past research. But it must be kept in mind that variation in each of these elements, in spite of claims to the contrary, does occur. Not all cities go about budgeting in the same way. While there are regularities among governments, differences exist as well, that up to this time have not been brought out in past research. These four dimensions are: l) The roles occupied by each participant, 2) The structure of influence among these actors, 3) The cognitive and evaluative mechanism of choice employed by budget makers, and 4) The linkages of the budgetary process to its environment. Budget Roles The first component of the budgetary system involves the identification of its authoritative participants and the determination of their Specific budget roles. These are: the department, heads, the executive, and the city council who alone possess the formal responsibility to 14 make spending choices. The normative expectations of behavior and attitudes attached to each of these in- dividual positions within the organizational arrange- ments of government, following from their legal authority and actual spending choices. Thus departments are the "spenders of the budgetary system, as they initiate the process by their requests for increased funds. The executive, on the other hand compiles all the separate department budgets into a unified position of the departments by cutting their requests for increases and by recommending a budget to the legislature that is more consistent with the spending level of the previous year. Then the legislature, reviews the decisions of these two other actors, as they assume the role of "oversight of administration." By altering recommended expenditure totals in some way and by approving an appropriation ordinance that records still smaller an— nual growth in expenditures, they serve as a popularly elected watchdog of government operations and a guardian of the treasury. The question posed is not only to describe the ad- herence to such roles, but to relate the possession of such normative orientations to actual spending choices made. Does variation in the possession of these pre- scribed attitudes and values directly translate into concrete budget behavior? 15 Decision Making Structure The second feature of the municipal budget process is its structure. This concept describes the pattern of interactions within the decision making system. As Heinz Eulau and Kenneth Prewitt write: A structure can be likened to a snapshot of the group's behaVioral processes. It catches the positions and statuses occupied by group members, determining the flow of transactions and interactions in the group . . . when inter- actions among group members are so patterned that the position of the members vis-a-vis each other can be ascertained, we speak of structural properties. 11 Many dimensions of the structure of the budgetary process can be identified, but the present research singles out the distribution of interpersonal influence among the department heads, the executive, and the legislature. The concept of influence is a major theme of poli~ tical science and its inclusion to describe the budgetary process and its expenditure outputs is to be expected. As Aaron Wildavsky writes of its importance: . . . if” the present budgetary process is rightly or wrongly deemed unsatisfactory one must alter in some respect the political system of which the budget is an expression. It is impossible to make drastic changes in budgeting without also altering the political system and the distribution of influence within. 12 ll. Heinz Eulau and Kenneth Prewitt, Labyrinths of Democrac (Indianapolis: Bobbs Merrill Publishers, 1973 . pp. 3—44. 12. Aaron Wildavsky, Radical Incrementalism 16 A more direct reason for highlighting its presence within the budgetary system derives from the mutually conflicting roles and expenditure outputs of the three decision makers. Each actor stands in the way of some other participant achieving his own spending objectives. What one wants, someone else opposes. First the execu- tive stands in the way of the departments achieving their goal of expenditure growth by his own economizer role. Then the executive is in opposition with the city council, as he defends the integrity of his recommendations in the face of the legislature's own objectives to modify them and carry out their oversight-review responsibili- ties. Finally, the legislature stands in an uncertain position toward the departments. Some represent a further obstacle to departmental expansion by reducing the budget they receive for review, while others serve as an ally by adding to spending totals. In either case, legislative interactions with the department heads serve their own objective of changing executive recommendations and participating in the determination of municipal spending policy. Each actor is in conflict with each other. Who wins and who loses in large part depends upon the amount of influence that can be brought to bear upon the spending 12. (continued) (Washington, D. 0.: American Enterprise Institute, 1966), pp. 115—166. u s 1. Mn | u '» ‘ 1? choices of others. The departments have to pressure both the executive and the legislature not to reduce their budgets and to grant them an annual increase in spending, as influence is exerted on them to requests in the first place. Then the executive attempts to pressure the legislature not to change his budget recommendations, at the same time that the legislature exerts their influence over the executive to make just those modifications that are opposed. The spending preferences of the strong emerge from the municipal budget process. An assessment of the distribution of municipal budgetary influence includes the following three elements: 1) formal authority, 2) informal influence, and 3) the executive's supervision over the depart— ment's formulation of initial spending requests. Formal Authority The first component of the structure of budgetary influence is each actor's formal authority and respon- sibilities. The departments initiate, the executive modifies, then the legislature vetoes spending proposals. The exercise of formal influence is evident in the overt change in spending totals made at the time of the of- ficial budget meeting of the second and third stages. The essential question underlying this influence relationship is the extent each actor activates his legal nu. 4L.- ’u .18 right to modify the budget received for review. Is there bargaining and negotiation between the department heads and the exeCutive, or does the executive assert his hierarchical authority over his administrative subordinates? Then does the legislature assert their formal authority to alter executive recommendations, or do they defer to executive spending preferences? And finally, do the departments appeal to the legislature for a restoration of executive cuts or does the ex- ecutive prevent such an "end-run" from being made? Informal Influence Influence is exerted in other ways than through the exercise of formal authority during the official review of the budget. The presence of the other "Face of Power" must also be assessed.13 Through the mechan- isms of anticipated reactions and behind the scenes, explicit warnings, the dominant actor limits the artic- ulation of spending proposals only to those that are consistent with his own preferences. Demands for change in the status represented by an increase in expenditure levels are suppressed before they are brought forth for public discussion. So the question has to be asked not only of how much expenditure totals were altered during 13‘. Peter Bachrach and Morton Baratz, Power and Epgerty; Theory and Practice (New York: Oxford Univer- 51ty Press, l970),'passim. 4 o... . I 1.0 I 19 the official point of review, but of what was not even submitted: of what was not even transmitted for a con- crete decision because of the exercise of informal influence of the strong over theiweak. Executive Supervision of Departmental Decision Making The third element of the structure of municipal budget influence pinpoints the executive's particular control over the departments' formulation of initial spending requests. For the executive to dominate, he must not only respond to the initiatives of the de— partment heads, but shape them as well. He must exercise control from the very start of the sequence of budget stages by supervising the process by which department heads formulate their requests. The executive attempts to exercise this influence by limiting the amount of leeway department heads enjoy in preparing their requests and then by limiting the amount of competition among the departments over the distribution of funds. The Distribution of Influence Past research has generally concluded that municipal budgeting is an executive centered process. As John Crecine writes: In summary, the municipal budget is the mayor's budget in which the mayor's policies dominate the department totals and city—wide wage and tax policies. The council and department heads q. u h— . ‘I ‘ 4..1 c ‘N l . . . V c h a.. .- .c x .- ~ A.\.\ \ \ ....V uhuk ‘ u .I-..1 \t- s\\‘ ‘4‘ \\b\ u. .. L. . . u . . . u. . . . . ..\ . u .\ \ -.\ x : . . . . . x a . . . . .. :N .~ .u. n 20 have surprisingly little to say about munici- pal resource allocation on a macro leve1.l4 The executive occupies the central position in the structure of municipal budgetary influence. The de- partments are on one side and the legislature is on the other side, as the executive is at the fulcrum of the decision making process. Department heads are essen- tially passive participants to the decisions made by their requests. They are unable to affect the executive and are not allowed to go to the legislature on behalf of their own budgets. The legislature, too, is sub- ordinate to the executive. They are little more than a "rubber stamp" to executive spending preferences, as they impose only marginal changes in his recommended expen- diture figures. In terms of the distribution of informal influence, an executive centered system is where informal influence is exercised over the department heads at the same time that the executive is free of a similar exercise of in- fluence by the legislature. Furthermore, in such a structure, the executive possesses both formal and informal influence over the departments, while the legis- lature exercises neither of these forms of budget influence 14. John P. Crecine, Governmental Problem Solving: Momputer Simulation of Municipal Budgeting (Chicago? Rand McNally Inc.. 1969), p. 38. 21 over the executive. In the last element of the pattern of influence, an executive centered process is where supervision over the first stage of departmental decision making accompanies the executive's exercise of both formal and informal in- fluence over them. Cognitive and Evaluative Process of Choice The third dimension of the municipal budget system is the cognitive and evaluative process by which one particular course of action is selected from among a set of alternatives. As Aaron Wildavsky writes of this approac h : The making of decisions depends upon calculations of which alternatives to consider and to choose. Calculation involves determination of how prob- lems are identified, get broken down into manage- able dimensions, are related to one another, and how choices are made as to what is relevant . . .15 Two analytical models of making this choice have been re- peatedly offered in the literature: the synoptical and the incremental. The first emphasizes the rationality of the selection of values to be achieved and the com- Prehensiveness of the analysis undertaken to reach them. The relationship of devising specific means to reach agreed upon ends is made explicit. The incrementalists 15. Aaron Wildavsky, Radical Incrementalism, p. 120, 5. 22 contend that such a problem solving approach is undesirable because of the social conflict produced by the incom- patibility of social values, and unattainable because of the limited capabilities of decision makers. Instead it offers a simplier substitute. Values are essentially ignored and analysis is restricted to the margins of existing policy. Only the increment of difference is of concern. This dichotomy of the decision making process is reflected in'three orientations to governmental budgeting: control, management, and planning.16 Each represents a focus on different kinds of information and evaluation rules for making spending choices. The control orientation is the traditional mode of incremental budgeting. The value of existing programs and appropriation levels are accepted as legitimate, and beyond the scope of the annual budget review. Attention is concentrated only on the line-item objects of ex- penditures" the purchases of government-- and the size 01‘ yearly ‘change in spending levels-- the costs of gov- ernment. The decision rule is then to minimize deviation from the base of existing spending: to limit the rate of annual growth. The management approach occupies a middle position 16. Allen Schick, "The Road to PPBS," p. 143. 23 somewhere between these alternative models of bud- gating. Here, the cognitive orientation is upon the activity of government: its performance in terms of the services provided. The evaluative rule is the efficiency and economy of operations. It is similar to incremental budgeting as it does not challenge current spending al- locations. On the other hand, it incorporates the analyti- cal approach of synoptic decision making that is not featured in the first orientation. Finally, the planning approach represents the syn- optic model of decision making applied to budgeting. The informational base is the programs of government and the criteria of choice are the effectiveness of spending. If appropriations are not accomplishing the program goals they were set out to achieve, then further funding should be curtailed. Existing activities are not beyond review as the budget is utilized as a means to establish the goals of government itself. The specific impact each of these budget making orientations have upon spending outputs varies according to their acceptance of the base of existing spending. Each, in the order they wereidescribed, place less faith in the continuation of what was done in the past and therefore more change in spending totals should be made in a planning orientation than in a management focus, which is still more than in an incremental process. .‘. ‘n - o ' I "I. . uh. ‘ I- ‘n < . ‘\ . r» g u - I V . x. \" -'..‘ \ I I ‘.. \\ \. ‘t. a O \,. \.. v s \.‘ ‘~ M | w \‘ \ ‘I ‘\ 21+ Linkage to the Environment The final concept of the municipal budgetary process examined is its linkages to the environment. Governmental decision making does not occur in a vacuum. It is neither insulated nor isolated from the social and economic con- ditionsof the community that surrounds it. Demands for political action which arise from such external char- acteristics are certainly relevant considerations to explain the process of budgeting and its expenditure end-products. Past studies of governmental spending have examined this connection through aggregate measures of demo- graphic features of the population and through multi— variate statistical analysis.17 However, these concepts cannot be adequately represented by data taken from Census Bureau documents nor explained by statistical techniques. Numbers, in themselves, are not cause and effect rela— tionships. As Philip Coulter writes of this shortcoming of previous research: . . . the identification and measurement of a close statistical relationship between sets of variables is not explanation. Rather, it is the high level of correlation itself that is to be explained through the identification and expli— cation of the substantive political linkages 17. See Philip Coulter, "Comparative Community Politics and Public Policy," Polity. vol. III. no. 1+ (Fall, 1970), and Guenther Schaeffer and Stuart Rakoff, "Politics, Policy and Political Science," Politics and Society, vol. I, no. 4 (November, 1970). 25 between two or more related phenomena.l8 Governmental policy is not a deterministic response to the stimulus of the environment that automatically emerges from the maze of correlation coefficients and factor scores. Instead it requires the intervention of real-world decision makers. How social and economic conditions generate demands for spending outputs: how such inputs are transmitted across the boundaries of the political system; and how the structures and processes of government convert them into expenditure decisions must be explained in behavioral terms. Governmental expenditures, or any other policy area is the result of the explicit choice of authoritative decision makers. As Heinz Eulau and Kenneth Prewitt write of this View: That policies are purposive responses to chal- lenges from the physical and social environment rather than conditioned reflexes is not a trivial notion, for the conception of policy as purposive implies the intervention of human actors in the sequence of events that links the natural-social with an appropriate policy environment. The policy environment is not so simply an automatic or spontaneous effect of external causes.l9 18. Herbert Jacob and Michael Lipsky, "Output Structure and Power: An Assessment of Changes in the Study of State and Local Politics." Journal of Politics, vol. xxx, no. 2 (May, 1968), pp. 510-538. ”‘h" 19. Heinz Eulau and Kenneth Prewitt, Labyrinths of Democracy, p. 503. .0- 4 9‘ 1'} 41': 26 Two mechanisms for the linkage between demand-inputs that come from the environment and the government's policy-outputs response are currently examined. The first of these, is the overt involvement of individuals and groups from the community as they artic- ulate specific budgetary demands throughout the three decision making stages. These fourth set of actors express support for additional services in accordance with their own interests and values. To the extent that such external pressure is felt within the decision making system a positive thrust for the expansion of program and spending levels is present. The perceptions and preferences of authoritative decision makers is the second mechanism for connecting the budgetary process to its environment. Considering public policy as a purposive response to social and economic conditions then the impact these characteristics have upon the budget have to be first interpreted by public officials. What is defined as the problems generated by the environment and the challenges they pose for governmental action depends upon the perceptions and preferences of those making spending choices. They do not necessarily enter into the political system until, and unless, they are recognized as such. As Heinz Eulau and Robert Eyestone write: It has been the burden of our argument that the systematic study of public policy cannot .. I- .2". (é .. ... 27 be contented with correlating indicators of environmental challenges, or indicators of resource capability to policy outcomes. Rather it was our assumption that policy development is greatly influenced by the predilections, preferences, orientations and expectations of policy makers-- in short, by the political process itself.20 Two sets of such evaluations are especially pertinent to expenditure outputs. The first is toward the resource capabilities of city goVernment. These are perceptions of an expenditure constraint-- the amount of funds poten- tially available to government; and then a revenue con- sciousness-- the attention and concern paid to tax rates. In either case, they serve as a constraint upon the ex- pansion of municipal service and expenditure levels. The second set of attitudes are interpretations of the needs of the community and the role of government in meeting them. A future oriented, programmatic, and activist view of the scope of governments responsibilities as a community problem solver translates into expansive spending preferences and budget behavior. Plan of Analysis The next chapter describes the research design and the methods of data collection employed. It points out the limitations of past case study research in terms of ———_ 20. Robert Eyestone and Heinz Eulau, "City Councils and Policy Outbomesz A Developmental Analysis," Amer- glam Political Science Review, vol. LXVII, no. 1 (March, 1968)! p0 1&3. .. a. m u "z I 'hl ‘. I "‘ W v_. 5... '— 28 its inability to formulate wide scale generalizations be- cause of the unrepresentativeness of the cities observed and the imprecise measurement of what took place. The present adherence to an explicitly comparative focus and the employment of structured interviews attempts to over- come these shortcomings. Chapter Three fully describes the two measures of expenditure outputs that are utilized as the dependent variables of analysis. Why they were selected, their substantive pattern in the fourteen cities and the statis- tical interrelationships among them are examined. The concept of incrementalism, as a description of these patterns is discussed. The fourth chapter examines the budget roles played by each authoritative decision maker. Alternative ex- planations of the department heads' spending role are analyzed. Either they are the advocates of the program and spending expansion of their departments service- performance area: or they are the padder of their budgets in an attempt to offset the expected reductions in their requests. Two interpretations of the executive's econ- omizer role are also introduced. Either he is the balancer of the budget by bringing requested expenditures in line with estimated revenue: or he evaluates the department heads as padders and is therefore disposed to act out their expectations by actually cutting their budgets. 29 Finally. the single interpretation of the role of the legislature is examined. The next four chapters comprise the main components of the analysis of municipal budgeting and this is the structure of influence. The pattern of formal authority is examined in Chapter Five, informal influence is dis- cussed in Chapter Six and the executive's supervision over departmental decision making explored in Chapter Seven. Finally, Chapter Eight examines the exact stat— istical interrelationships among these three elements that together define variation in the distribution of executive influence. Chapter Nine examines the cognitive and evaluative mechanisms of choice adhered to by each of the three actors. The next chapters conclude the analysis of the municipal budget making system by its linkage to the environment. Chapter Ten studies the involvement and influence of individuals and groups from the community and Chapter Eleven looks at the attitudes and values of these three decision makers toward the environmental inputs of resources and needs. Chapter Two: Research Design Introduction Several methodological shortcomings mark past empirical studies of governmental budgeting that make an additional effort both worthwhile and necessary. An undo reliance has been placed upon the case study method, which lacks a comparative focus. Specifically, thexroblems of assuring representative observations amicollecting precise data have not been overcome. Thislms deleteriously affected the substantive descrip- tim1of the budgetary process, creating a false impression thatall cities go about budgeting in essentially the mum way. The acceptance of this consensus, not only disfimts the understanding of the decision making pro- cess,but serves to misdirect research efforts. Consistency of the Municipal Budget Process Sweeping claims for the universality of an executive centered, incremental process of municipal budgeting lmve been proposed, after the study of only a few cifies. John Crecine, in the most extensive research up to this time, makes grandiose assertions for the appficability of such a model of the budgetary system, 30 31 based on the observation of only three cities. He writes that: There is little reason to believe that this model would not describe the budget process in most large United States cities . . . smaller municipalities and local governments . . . state and federal government . . . [and] four major remaining categories of non- market organizations . . .1 Robert Lineberry and Ira Sharkansky, in their recent text entitled Urban Politics and Public Policy, make similar claims for the Consistency of their descrip— tion of municipal budgeting, and thereby enshrine a single pattern of decision making as conventional wis- dom. They write of the presence of uniformity not only among municipalities, but among all levels of govern- ment in the United States: Although studies of municipal budgeting have been conducted in widely different types of municipalities and in different parts of the country, the findings are sufficiently similar and show sufficient correspondence to studies of government budget making at other levels to 1. John P. Crecine, Governmental Problem Solving: A Com uter Simulation of Munici al Bud etin (Chicago: Rand McNally, 19595, pp. 221. 222. 221;. and 228. A further example of Crecine's exaggerated and unwarranted extension of his budgetary model is from the mayor form 01‘ government of his three cities to the manager form. As he writes on page four of "A Computer Simulation'.’: "The functions (and hence the submodels) are generali- zable to forms of government other than the mayor-council; however, in the city manager form, the city manager and his staff would correspond to the mayor's function in our model." In two sentences he completely dismisses an extensive body of research that addresses itself to such differences, without any empirical or logical support. 32 suggest a high degree of reliability.2 Although there is no disagreement with the empirical description offered of budgeting in the cities pre— viously studied, for there is little evidence or reason to question their accuracy, there is consider- able doubt that these conclusions are as widely appli- cable as these authors would have us believe. The broad extension of this particular model of decision making to all city governments is both unsubstantiated and un- warranted. The uniformity of the budget system is far more limited than has been portrayed. Not all cities display the same characteristics of Crecine's computer simulation model; instead there is extensive difference in funda- mental features of the way municipal governments go about preparing and adopting the appropriations ordinance. This ‘ belief, which comprises the underlying premise of the current research, is held for two reasons. The first is that contradictory evidence can be identified and con— flicting interpretations can be offered of the literature on governmental budgeting, and municipal decision making in particular, that challenge the generalizability of an These will be executive centered, incremental process. presented in the text as they arise, and need not be 2. Ira Sharkansky and Robert Lineberry, Urban Politics and Public Poligy (New York: Harper and Row Publishers, 1971), p. 236. “no 3 I aha a l ' out W. \ 33 documented here. But there are sufficient inconsistencies in essential characteristics to suggest that greater dif- ferences among cities occur. So, in spite of claims to the contrary, variation in municipal budgeting is indeed evident. But even more importantly, the assumption that various patterns of governmental budgeting exist, derives from the significant methodological limitations of past research. 0n the basis of their research design, there is little reason, in the first place, to believe that such uniformity is present. A case study approach to the investigation of political phenomena is simply unable to support the wide scale generalizations that have been made. The representativeness of the cities observed is open to question, as systematic data is lacking to identify the similarities and differences among cities. The present effort attempts to improve these shortcomings by an explicitly comparative research approach. Fourteen cities are examined and data is col- lected through a structured interview schedule. In this Way, variation among cities can be identified in order to formulate empirical generalizations on a more logically secure foundation. Generalization While case studies are often a necessary first step in the investigation of political phenomena, they need to "u rt. v \ 1 4’ a? n 31+ be followed by more extensive and more comparative re- search. Studying a handful of cities is, by itself, insufficient foundation for all embracing generaliza- tions that have been previously formulated. There are simply too few units to establish the representativeness of the specific observations. The logical foundation is missing to extend the findings beyond the immediate circumstance of a particular place to a wider universe of similar situations. There is no way to assess the similarities and differences between those cities under analysis to all others they purportedly represent. The case method fails to provide sufficient evidence that uniformity in the municipal budget process is, in fact, present. However they do serve the important heuristic purpose of identifying the significant explanatory concepts, of proposing hypotheses, and of suggesting empirical gen- eralizations. But they cannot formulate empirical propo- Sitions and invariant behavioral laws. Their findings have to be tested and verified in a larger research setting. It is in this fashion and spirit that the present study is conceived. It builds upon the work of those who revitalized the study of governmental budgeting from its traditional framework. From such a base the Present study moves toward a more extensive and more systematic explanatory model of governmental budgeting, "n. ,0. vC' UNI I o .n ,4- 35 Thus to go beyond previous efforts, the present research is comparative in focus. Fourteen cities, a number of itself that is greater than the total of pre- vious empirical descriptions of municipal budgeting are investigated at the same time, with the same data gathering technique.3 In this way, a direct search is made for variation, as well as regularities among muni— cipal budgetary practices from one city to another. If, as initially assumed, differences are present, this research strategy should be able to identify them. Representativeness However, the inclusion of a large number of cities, by itself, does not allow the formulation of universal generalizations. Fourteen cities are still a miniscule sample of all municipal governments in the United States. Consequently, it is necessary to specify what kinds of cities they are, and the wider context they are intended to represent. Past research has been especially insensitive to this methodological caveat, with the expression of all embracing statements, when perhaps a more restricted 3. The total number of cities previously subject to empirical studies by political scientists in the past few years is eleven (Crecine, 3; Anton. 3: Melts- ner and Wildavsky, l; and Caputo, 1+). Thus the present research is larger than all past efforts combined. 00' b 9. coo. 'N. | I ~ a , n u. 36 formulation of the relevancy of their budget model would be more appropriate. The three cities John Crecine studies-- Detroit, Cleveland, and Pittsburgh, are all large, metropolitan in character, and located in the older, industrial northeast. In their demo- graphic and political conditions, the. problems they face, and the size of their budgets, they are less typical of the entire, category of urban government than might first be thought. The current group of fourteen cities, are of a dif- ferent character and for this reason alone, a different budgetary process might be expected.“ They range from 25,000 to 200,000 in population and expenditures from three to twenty million dollars.5 The notion of city size, often mentioned in the study of urban politics, AF. Besides Oakland, the four cities of Caputo and the three of Anton are all middle size and thus many of the divergencies in the budgetary process described in the text, taken from these studies may derive from the different kind of city. . 5. There is not a continuous distribution of city Sizes in the state of Michigan that would have enabled a more direct test of the relationship between city size and the budget process. After Detroit, the next largest city has a population of 200,000. Unless state boun- daries were crossed, introducing extraneous variables, this important relationship can only be hypothesized to oust. Excluding Detroit, this left thirty-nine cities in the state above 25,000 population. First, the inde- pendent cores cities of the smaller metropolitan areas that dot the state were included and then the large (100,000 plus population) suburbs of Detroit were added. "- ”a n m». ‘II ‘I ‘q .. I" “g 37 represent many concepts associated with urbanization such as complexity, heterogeneity, and specialization. They have been previously differentiated from those below and above them in size. Robert Dahl, draws attention to these particular cities as the most appropriate units for a democratic political system. They are of "man-sized" proportions and optimal for rational self-government, widespread participation, and citizen influence.6 Be- yond this normative feature, Oliver Williams and Charles Adrian, in their study of Four Cities, suggest empirical differences in their social and political organizations. They write that it is: . . . within the limits of this range that urban places lose the face-to-face style of politics peculiar to small towns and yet retain some central system of communication in political and social affairs that is often lacking in the very large population centers. Beyond this range the local political processes become compounded around multinuclear centers and politics becomes increas- ingly a function of formally organized interest group activities.7 By selecting a different size city, the opportunity to identify variation in the budgetary system is maximized. So it must be made clear and kept in mind that the present group of fourteen cities are representative of only middle size cities. The range of generalization extends only to ——_ 6. Robert Dahl, "City in the Future of Democracy," mrican Political Science Review. vol. LXVI. no. 4 (December, 1967), pp. 953-970. 7. Oliver Williams and Charles Adrian, Four Cities I I u‘ \: le’ “3| l I 38 this group of municipal government. Data Collection A second shortcoming of case studies is its data collection techniques. By a narrative recreation of events, this method provides a vivid and detailed des- cription of what took place that is difficult to achieve by most other research procedures. However, the result is an imprecise and unsystematic, particularistic ex- planation that is based upon the impressions and inter- pretations of the researcher and his intimate, albeit highly personalized knowledge. Measurement is limited to classificatory statements of the presence or absence of the occurrence of what is being observed. The dichotomization of political phenomena into either/or categories, while a necessary first step in comparison, needs to be followed by more rigorous quan- tifications of what occurred.8 The simplification of the real world into mutually exclusive and jointly exhaustive categories has to be filled in with more precise measurements of the gradations from one situation to another. But case studies are unable to do more than classify events into discrete groups and are unable to 7. (continued) (Philadelphia: University of Pennsylvania Press,'l963), p. lit. 8. Arthur Kallenberg, "The Logic of Comparison: Methodological Note on Comparative Study of Political SyStzm," World Politics, vol. XIX, no. l(October, 1966), pp‘ 9‘83. .- .4 q. -. , . .n -u ‘u. 'n .... . . I . .- . -. . s \\ 'c ‘II \. I \ .. s u . ‘Q \ \‘l ( s I .. N ‘A .1 . ‘ I . M I‘D ‘ ‘ \‘ I ‘A u u \ Q ‘ \ 39 more exactly describe the similarities and differences among events. Consequently, differences among cities are neither recognized nor reported. The position of cities along a continuous distribution of low to high; of how much more or how much less one observation stands in relation to others have to be made in order to identify the variation within the municipal budget making process. To overcome these inadequacies of the data gathering techniques of past research, a structured interview schedule was utilized. It is only this procedure that permits fourteen cities to be investigated at the same time. The case method is, of course, limited to the number of observations that can be handled by a single researcher; and thus four was the most cities included within past studies of municipal budgeting. To over- come this natural limitation of being at one place at one time, post hoc interviews with official participants to the budgetary process were conducted. The descrip— tion of the decision making process derives solely from 9 the information collected in this interview. No 9. Representativeness of community interest groups Weren't interviewed to gather their views of the muni- Clpal budget process. Although this would have been worthwhile, it was precluded by limited financial re— sources. In any event, there is no reason to believe that such interviews would have provided any better PlCture of their participation and influence than those suPplied by public officials. Furthermore, regardless of what such individuals may say, it is the perceptions, assurate or otherwise, of what took by governmental de.. Cision makers that is most important, for only they make 'u- v_‘ ‘I . r n N. ' \ \~ ‘ ‘I :1 1+0 independent observations of the behavior of the department heads, the executive, or the legislature were made. None sat in on any of the meetings among them to check the accuracy of their report of what took place. No analysis of documents, such as minutes of these meetings or secondary newspaper accounts were made. Not only do such secondary sources not often exist at the local level, but more importantly, there is little reason to believe that such a research strategy would yield substantively different results. This alternative data gathering procedure would not yield a more valid and reliable description of what took place than is supplied by the personal interviews with actual budget participants. The interview instrument was explicitly designed to enhance the documentation and collection of systematic data. Instead of formulating open—ended questions that could only be considered a scale of nominal measurement, wherever pos- sible, closed—ended questions were posed to represent a rank-order measure of correlation employed to test the relationship between independent and dependent variables. In most cases, a five point classification of low to high was presented as fixed alternatives for selection. 9. (continued) authoritative spending choices. 41 Specifically, questions were phrased in terms of how often, and how much, a feature of the decision making process occurred. The choices were: none, slight, some, much, and great, corresponding to scores of: one, two, three, and four. Attitude questions were similarly scaled along the five point continuum of agreement, satisfaction, or im- portance, among others. Finally, open-ended questions, described in the text as they arise, were also coded along a single dimension to achieve a similar ordering. Interviews The interview instrument consisted of two parts. One was the personal interview schedule containing open- ended questions that required a direct exchange between 10 The second the respondent and a trained interviewer. part, was a self-administered questionnaire that was left at the completion of the interview to be returned by mail. This contained closed-ended questions that could be conveniently answered by merely checking off the appro- priate box. A total of 169 public officials were personally inter- viewed in the fourteen cities, and three quarters of them (129) returned the mail questionnaire (see table 2.1). The heads of the police, public works, and parks and 10. The interviews were conducted by.the professional Staff of the Urban Survey Research Unit of Michigan State University under the direction of Professor Philip Marcus. I I 42 Table 2 . l RESPONSES TO INTERVIEW AND QUESTIONNAIRE Department # of C' t Head Executives Members Legislature l y Inter- Mail Inter- Mail of Inter- Mail view view Council view 01 3 3 3 2 ll 8 1+ 02 3 3 3 3 7 5 5 03 3 3 2 2 7 7 7 on 3 2 3 l 7 6 3 05 3 3 3 3 9 8 5 06 3 3 4 3 8 7 4 07 3 2 2 2 7 6 5 08 3 3 LP 4 7 5 3 09 3 3 3 3 9 5 l 10 3 3 3 2 7 7 3 ll 3 2 4 3 5 4 L; 12 3 2 2 2 9 4 4 13 3 2 4 u 9 7 u in 3 3 2 2 8 8 4 Total AZ 37 42 36 112 87 55 h 1+3 recreation departments in each city were interviewed (forty-two). Eighty-eight percent (thirty-seven) of them completed the questionnaire. These three departments were selected for personal interviews because they are generally the largest service areas in city government. Together they comprise an average of 42 percent (see table 2.2) of the budget at the end of the first stage and an average of 45 percent (see table 2.3) of total requests. Next, a total of forty-two participants in the second stage of executive decision making were inter- viewed, and 85 percent (thirty-six) of them filled out the mail questionnaire. This ranged from two to four in each city and included the chief executive manager or mayor and all others who were identified as having participated in some way in the formulation of the executive's budget. These were individuals in such institutional positions as assistant to the manager/mayor, finance director, budget officer, controller, treasurer, etc. Finally 78 percent of the total number of councilmen were interviewed (eighty- seven out of 112) and 64 percent of them (fifty-six out of eighty-seven) returned the mail questionnaire. The size 01‘ the legislature extended from five to eleven members and a majority were interviewed in all but one city. Responses to the questionnaire were lower, and although in twelve Cities a majority of those who were interviewed returned the questionnaire, in only seven cities did a majority of 44 Table 2.2 THE THREE DEPARTMENT'S REQUESTS AS A PERCENT (E‘THE TOTAL BUDGET AT THE END OF THE FIRST STAGE OF DECISION MAKING . Parks . . Public (hty Police & Average Works Recre- ation 01 23.88 7.93 11.01 42.82 02 19.49 10.23 9.60 39.27 03 19.28 17. 76 8.67 45.71 on 19.19 5 38 8.07 32.64 05 18.113 28. 76 L1. 15 51.311, 06 17.63 7. no 14. 05 39.08 07 26. 08 14.18 5. 38 us. 64 08 22.19 7.31 7.26 36. 76 09 22.91 10.58 5.17 38. 66 10 20.02 13.12 9.85 #2. 99 ll 9 55 38.46 8 95 56.96 12 18.99 10.19 3 98 33.16 13 26. 41 5.83 6 34 38.58 14 18.87 17 96 12.13 48.96 Mean 20.21 13.94 8.19 L12.33 Sh Dev. 4.03 9.08 2.91 6.65 45 Table 2.3 THE THREE DEPARTMENT'S REQUESTS AS A PERCENTAGE OF TOTAL REQUESTS City Percent 01 39.29 02 21.91 03 46.16 04 39.97 05 55.99 06 27.94 07 47.39 08 71.76 09 30.69 10 39.03 11 15-33 12 42.14 13 42.31 14 38.84 Mean 45.13 St. Dev. 37057 r.- .u \ | \H . \‘ ~l ‘ 'II t 46 the total council complete it.11 Level of Analysisl2 The aim of this research is to describe the process of governmental budgeting and its expenditure outputs. It is not to describe the properties and attributes of public officials who participate in that process. In- dividuals are classified according to the specific insti- tutional position they occupy, such as department head. the K 11. Although it would have been preferable to obtain a majority of councilmen in each city, there is one city (city number twelve) where after repeated efforts, this was not possible (one is missing). It was decided to include this city anyway, instead of wasting the data already gathered. It was also im- poss1ble to obtain a majority of councilmatic replies. employing the mail questionnaire technique. But again the cities were included, except for number nine, where 01.111! a single member of the council returned the ques- tionnaire. Obviously, this one response cannot be taken to represent the entire legislature by itself. Conse- fluently. when a majority of questions that make up the Index come from the document, this one city is excluded from the analysis. 12. The "Level of Analysis" section is based upon the following works: Allen Barton, "Bringing Society Back In: Survey Research and Macro-Methodology," Amer- ican Behavioral Scientist, vol. XII. no. 2 (November- W Peter Blau, "Formal Organ- lZaFlO‘HS: Dimensions of Analysis," American Journal of Max. vol. LXIII. no. 1 (July. 1957 . pp..5 - 9: 1’9th Blau, "Structural Effects," American SoCiological M2. V01. xxv, no. 2 (April, 1960), pp. 178—193: P. L. Kendall and Paul F. Lazarsfeld, "Problems of Survey mall/Sis " in R. K. Merton and Paul F. Lazarsfeld. eds.. o .'. . . Cohtlnuities in Social Research: Studies in the 500 e .and Method of the American Soldier New York: Free Press. 1950 ' PP- l 7~l9 ; Paul F. Lazarsfeld and H. Menzel. “niche Relation Between Individual and Collectlve Prop- ertiest" in A. Etzioni, ed., Com lex Or anizations (New York: Holt, Rinehart, and Winston. 19 1 . pp- 1 7-206= - C. Selvin and W. O. Hagstrom. "The Empirical Clas- 47 executive, and the legislature, their behavior is reg- ulated by the normative expectations of the respective budget role, which draws attention to consistent pat- terns of conduct of all the distinct individuals who happen to occupy the same position. The idiosyncratic and personal differences among them are of no concern for these are most often effectively submerged under the. exigencies of adhering to the more consistent and regular prescriptions of their budget role. i The unit of analysis are these decision makers, as they together form a collective. Statements are made about each of the three sets of actors, as they exhibit similar or different patterns among the fourteen different cities, as entities apart from the 169 separate decision makers. So it is not a question of the distribution of responses among the department head, the executive, and the legislature themselves, but of the distribution of responses among the cities. It is not important to know that 20 percent of councilmen answered a question X and 30 percent answered Y: but what percentage of councilmen in each city answered it in these ways. These two modes of analysis are not the same. 12. (continued) sification of Formal Groups," American Sociolo ical Review, vol. XXVIII, no. 3 (June. 1963). pp. 399-74710'F'a‘nd A. s. Spannenbaum and J. G. Bachman, "Structural versus Ind1v1dual Effects," Mgrican Journal of Sociology,’vol. LXIX, no. 6 (May, 1964) , pp. 585-595. __‘..4 48 The reason for such an analytical approach is that expenditure policy is the output of a collective decision making process and not the behavior of individual public officials. As Heinz Eulau and Kenneth Prewitt write of this notion: ". . . most propositions of politics are not propositions about individual actors but about Col- lective actors-- on the sound assumption that collective units are, in fact, g_e_a_l_ [italics in original7 decision makers."13 The budget that is approved by the municipal legislature is the result of a group decision as the mem- bers of the council make their individual spending choices, the authoritative adoption of the budget ordinance is a decision reached by the council as a collectivity. In order to account for the variation in the pattern of budget outputs, the decision making process of the council must also be measured on this group level. Individual responses cannot logically represent the collective choice process, nor are they able to explain its policy decisions. Two levels of data and analysis exist depending uDon the part of the social system to which it refers-- the individual and the group. There is the system as a whole and its lower level parts. However, they are not 13. Heinz Eulau and Kenneth Prewitt, Labyrinths 0f Democrac (Indianapolis: Bobbs Merrill Publishers, 1973 9 pa 3 o on a 49 distinct, but interconnected, as the separate subunits come to make up the whole. The individual components fit together to form the more inclusive and higher level col- lective. Measurements can be taken at either of these two parts. These are integral properties of the group that exist apart from the characteristics of its members; as there are attributes of the individuals that exist inde- pendent of participation in the groups. The difficulty inherent in these two different modes of analysis is that data gathered by interviews with in- dividual members of the organization, are intended to describe the group as a whole. This risks the "individual- istic" and compositional fallacies. Concepts that refer to collective phenomena cannot be employed to eXplain the behavior of individuals, as individual properties cannot be used to account for the behavior of groups. Each has to be analyzed at its own level. However, this problem can be overcome, without arbi- trarily and incorrectly mixing levels, by following certain procedures that have been previously outlined. As Allen Barton writes: "Measurements of organizational properties may be derived from basic data gathered at any level of the components of the organization."lLL The behavior 14. Allen Barton, Organizational Measurement: Qgilege'Entrance Examination Board (Princeton: Princeton University Press, 1961), p. 2. no 50 observed can be different from the behavior analyzed by performing some kind of mathematical manipulation. In this way the gap between the micro-collection of data and the macro-analysis of the policy making process can be partially breached. By aggregating discrete responses, a group measure is constructed. These data transformations include: statistics of central tendency (mean) or of dispersion (standard deviation). These are acceptable statistics that enable personal interview data to measure collective properties. As a result the phenomena des~ cribed no longer belong to the individual himself, but to the group of which he is now only a single member. So, again the propositions and hypotheses of this study do not relate to individuals but to the three sets of department heads, executives, and legislatures in each individual city. To assist in the formation of such collective measures, whenever possible, questions were phrased in terms of the group itself. Instead of asking each in- dividual to describe his own activity in the budget Process, he was explicitly asked to describe his own activity in the budget process, he was explicitly asked to describe the activity of the group of which he was a member. Each councilman, for example, was not asked how much influence he personally had with the executive, but how much influence the council as an institution «1‘ ., .,v- -n ‘t .I . w 51 possessed vis-a-vis the executive. Impressionistic evidence, in a pretest and by reading the completed inter- views indicate that respondents did perceive the difference in these questions. Often individuals qualified their answer with statements to the effect that: "This is what I would say for myself, but the whole council would say this." Given the relatively small number of respondents in each institutional position within each city, this formulation of the questions also has the effect of offsetting random response error. The more independent observations are obtained, the less whatever bias is evident deleteriously affects the description of what took place. In effect, individuals are asked to be "informants" of the group process. This is important, not only for the legislature which is a natural group, but for the executive as well. In this case, there is a single individual who alone possesses the formal authority to make budget decisions. Thus it might have been acceptable to only interview the chief executive in each city and rely upon his single response to describe the second stage of the municipal budget process, but this was not done. Instead, all those administrative officials who in some way were involved in execu- tive decision making were interviewed. However, they were not asked to describe their own activities in 52 budgeting, but to the best of their ability, characterize those of the executive. In this fashion, the distortion that is evident in any interview situation is hopefully balanced out, resulting in a more complete and more ob- jective description of what took place.15 Statistical Analysis I Several features of the data analysis need to be presently discussed. The first is that each concept of the decisionmaking process is measured by several sep- arate questions. Instead of just relying upon a single question to represent the meaning of the phenomena under investigation multiple indicators are employed. The increased validity and reliability produced by this approach is demonstrated by the statistical relationship between each one of the single questions alone and the dependent variable of expenditure outputs (data not reported). Very few Of the separate questions, by themselves, indicate a stronger correlation than they do together as an index. Merging them, brings forth an interactive effect that more completely represents the several dimensions of the concept. The analysis 15. Although the collective of department heads is an analytical construct as each department could be considered as anentity into itself, with its own distinct decision making process and separate expendi- ture end-products, the first stage of departmental de- Cision making is also examined as an aggregate of the three individual department heads who were personally interviewed in each city. The analysis is on the level 01‘ the characteristic pattern of departmental decision 53 proceeds by first calculating the mean response of those interviewed in each city to each question. Then, the overall fourteen cities mean for that question is cal- mflatmizum.finally all the separate questions are merged into a grand mean that serves as the final index of the concept. In such a technique, each question has to contribute an equal portion to the numerical value of the index; for there is no reason to believe that one particular question is more or less important than the others as a description of the concept. Consequently, transformations of the data were sometimes necessary. Since the most frequent ordering of responses is along a five point scale, all other questions, no matter what their original arrangement, had to fit Within this pattern. This involved several kinds of mathematical transformations. If orig- inal responses were in the form of a percent, where the scaheextends from zero to one, scores were multiplied to achieve a distribution from zero to four. If open—ended qumfifions were coded in more or less than five categories, then the original responses were either extended by mul- tiplication, or contracted by division to arrive at the desired five point scale. One last rearrangement of the data was achieved by reversing the direction of the 15. (continued) making and the differences from one city to the next and not from one department to the next. ' I H . u .A .. .nu . 9 .- .- .. rt. - —« - u ‘ u -q s 54 numerical values of low to high. To avoid a response set, the phrasing of questions were mixed so that the high occurrence of any phenomena did not always correspond to a response of great extent and a low occurrence did not always correspond to a response of no extent. Therefore, to achieve uniformity, the absolute value of the scores were reversed, where an original score of four representing a low response becomes zero and a score of zero becomes four. In describing the relationship among budget actors, responses from both sides of the interaction were col- lected. For example, in measuring the distribution of budget influence between the executive and the legislature, both sets of decision makers were asked the same questions, and thus executive and legislative perceptions of their interaction with each other were obtained. But when it came to the connection between the decision making process to expenditure outputs, the statistical analysis proceeds only from the perspective of the actor who is actually making the budget decision. Each actor, can only account for his own behavior, and no one else's. Only legislative responses can be utilized to account for their own behavior in reviewing executive recom- mendations and in adopting the final budget ordinance. To establish the cause and effect relationship be- tween the process of budgeting and its expenditure end I ”5;!" .l I .ul0|d . . “co: 1. 9:: I. on. ...o':,0 :9 ,2...“ a. '; "I ~0- - 5|». I. "not. ‘I ‘n . I \“:Fl “ "' V.. ._. 1. . 1“” "-u .“I; v hu \\:"0:'fe 55 products is a difficult methodological and philosophical question. To disentangle the connection between inde- pendent and dependent variables in the social sciences is always a tentative affair. Statistical correlations are not "causes" as they are not conceptual explanations. The validity of the measures rests not on the strength of the statistic, but on the logic of the analysis offered to explain it. The interpretations placed on the strength of the correlation is subjective, and the following verbal correSpondence is employed: 0.00 - 0.19 -- weak 0.20 - 0.39 -- moderate 0.40 - 0.59 -- substantial 0.60 - 0.79 -- strong 0.80 - 1.00 -- very strong While the level of statistical significance is reported, the rejection of any hypothesis on this basis is always qualified. Empirical testing of relationships serve more to corroborate the existence of a systematic connection, than it can disprove such a linkage. There is no claim that the final answer has been presented. Furthermore, a single, simple "mechanistic" inter- pretation of the budgetary process and its relationship to expenditure outputs is rejected. Many different con- cepts of the policy making process are employed, each representing a different approach to the way political choices are made, and in this eclectic fashion they are employed. There is no one "cause" for differences in .._, u. I,‘ LH. 56 municipal expenditure policy that appears to be superior to any other. But all together they form an inclusive explanatory model of the municipal budget process. Conclusion Much of the uniformity reported in past studies of the municipal budget process is an artifact of adherence to a case study method. By focusing upon a narrative des- cription .of the decision making process, variation among cities, states, or agencies in the national government has not been identified. By failing to conceive of bud- geting as the independent variable to account for variation in the dependent variable of expenditure outputs, a con- sistency is reported that really does not exist in the way cities go about preparing and adopting the appropriation ordinance. Through an explicit comparative research design, the differences among cities in budgetary practices as well as its consequences for expenditure outputs can be ob- served and established. n, - Chapter Three: Expenditure Outputs Introduction The two measures of expenditure outputs as introduced in the first chapter are the percentage change in total spending between the three decision making stages and the percentage change in total spending from one year to the next. The present chapter examines these dependent variables in more complete detail. First the justification for their use is explained. Then their actual pattern in the four- teen cities is described. Finally, the statistical inter- relationship among five separate outputs is examined. The concept of "incrementalism" is discussed throughout as a description and explanation of municipal spending choices. Criteria for Selecting the Dependent Variables There are many ways to represent the spending decisions of government, both over time and within a single year. The present two measures are employed in the present study. be- cause they are the most appropriate expenditure products of the municipal budget process. Particularly, they examine the Percentage change in total fundS. whiCh are its Significant features compared to spending choices at the national level. 57 58 Previous research provides empirical evidence that it is the percent and not the absolute per capita amount of change that is the relevant consideration in budgeting.l Change The first characteristic of the dependent variable is that it is a measure of dollars and cents change. It does not represent the total absolute amount of spending at any one year or stage. The analysis is not of why a city spends more or less than another. but why there is more or less change in funding levels in one city compared to the others. It is not logically possible to account for the total amount of per capita expenditures at any point in time by the process of budgeting during a particular year. Such a measure of municipal fiscal policy is the cumulative result of a past series of budget choices as well as more general social, economic, and political con- ditions, that cannot be adequately captured in the decision making process of any single fiscal year. However, ex~ amining the budget process during one "slice" in time, ——-____ 1. As Ira Sharkansky and Augustus Turnbull write: "In both phases of the model there are weak or nonexistent relationships between the sheer size of each agency's re— quest and the recommendations of the governor or the appro_ priations of the legislature . . . a fixation of budget reviewers on the increments that are sought, rather than on the size of requests which includes both new increments and the agency's base of current appropriations," in "Budget Making in Georgia and Wisconsin: A Test of a Model," Midwest Journal of Political Science, vol. 13, no. 4 (Nov- ember. 1969) p. 633. .,, 'l , I |~l‘ r4. ‘u. ‘t 59 caireveal the mechanisms by which the yearly repetition of lmdgetary decisions results in a specific absolute level dfexpenditures. Studying the way decisions were made for the197l budget provides a partial understanding of how the direction and magnitude of annual change in spending even— tualLyumkes up the total absolute level of spending outputs. The use of this measure of expenditure outputs is en— cmnaged by the failure of past aggregate studies of spending pafimrns to successfully identify its statistical deter- mhmtes and its variation among similar governments. Compared to um difference in absolute per capita levels of spending, fim size of yearly change is much less adequately explained. As ha.Sharkansky writes of his study of these two questions m1the level of the fifty states: While previous expenditures show very close rela— tionships with the level of current expenditures, the level of previous expenditures does not show a close relationship with measures of change in expenditures. The provocations of change are not inherent in levels of previous spending.2 Thefactors most strongly associated with rates of change in- clude economic, social, political, and governmental character— isfirs. In particular. as Sharkansky writes, it is affected by: . . . economic resources and needs, political con- ditions, tax revenues, and the presence of leaders with intense motivations and the electorate and the —_.._,_ . 2. Ira Sharkansky, Spending in the American States (Chicago: Rand McNally. 1958). P- 37: ‘—-fi‘ L... 'I“! -. I‘ H u .0. H.‘ ,., 60 political resources to attain their goals . . the interests of the electorate and the strength of the majority party might be salient . . . immediate and transient phenomena, including the motivations and strengths of specific individuals and organizations.3 This list of ostensibly political determinants for the direction and size of yearly change in spending should also be extended to include the way public officials go about pre- paring and adopting the appropriations ordinance. The inability to statistically identify the independent variables of spending change points to the increased significance of the actual budget making process as an explanation of expenditure outputs. As discussed in the first chapter, the process of budgeting has direct consequences for the spending choices made. So, it must be kept in mind, that the present analysis seeks to explain only the variation among cities in the amount of change in spending levels as the most appropriate measure of the municipal budget process. Total Spending The second feature of the dependent variable of ex- penditure change is that it is a measure of total spending. EXpenditure outputs can be demarcated into two different kinds. There is the per capita level (or change) of spending as a whole: as a single figure. Then there is the distribu— tion of the total into its separate parts. The first measure “-1- and 72. Ira Sharkansky, Spending in the States, pp. 37, 73, u.. -. an .. .. .... ... .. .. .. . o . u ... .. . .. .s .. nl. . w. . . ...« a .. “a an“ ... o no. .H u... v o. .n. .o o. . .\ .5. on. \ \\\ \.- \NV. “\W V ‘W . I i .x .. .. 7 ... _ . i .. t .. f. a. T : u 77... . . ... ... . . no. \ o An n. \u\ . 1 \ . \.. .\. i \\ \I \ \..x. I, . . .. . s. . -n .. .- .\. :: ..\ .u :l 61 looks at the ratio of public to private spending as all governmental expenditures are compared to the amount of economic activity in society (the GNP). The question form- ulated in this measure is how much of societal resources are controlled by government and how much by the individual con— sumer. The second measure is the division of governmental resources among alternative uses. The question raised here is what kinds of services and how much of them are supplied by government. What is the financial priority among the different programs and policy goals of government. The political questions shift from the first to the second measure from the size of government's total spending pie to the size of the various slices. The present research examines only the percent change in the level of total spending from one stage to the next or from one year to the next for several reasons. First of all, local governments are required by law to adopt a balanced budget. The deficit spending that is the hallmark of the national government is simply prohibited at this level. As a result of this constraint the cognitive and evaluative framework of budgeting is altered from a con- cern for the relative priority among alternative spending Programs to the amount of the total itself} With this 4. An interesting analysis of the difference between {balanced and unbalanced budget requirement is found in William Buchanan, Public Finances in Democratic Processes (Chapel Hill: University of North Carolina Press, 1967)T c'fu Chapters Seven and Eight. no - : £“1 ‘- \I . ‘K H» RI '§ .‘ we |—_7 62 requirement of equal and offsetting revenues and expendi- tures there exists an explicit and definite upper limit on expenditures. This serves as a predetermined figure that cannot be surpassed. All decisions on departmental alloca- tions have to fit together to meet the amount of available resources. The total is not simply reached by adding up the individual parts, but exists prior to and independent of the separate department budgets. The first choice to be made is how much the total is going to be. After the determination of this expenditure figure, the whole is then divided up among the individual departments. Such a process does not exist when such a legal constraint is not present and for this reason the relevant measure of budgetary outputs on the local level is not how much is Spent on the departments, but how much is spent by government as a whole. The balanced budget requirement also forces a co- ordination between revenue and expenditure decisions that does not exist on the national level. At the same time that expenditures decisions are made so are revenues chosen. A Single, integrated document that contains both input and outputs are approved together as the appropriation ordinance. Financial decisions are not serial and fragmented, but are established, so are necessary changes in local taxes approved. Tax rates and hence revenue yields are manipulated to pro— duce additional resources in order to balance the budget. q: q : g. I. a], 63 The variability of the local revenue system in order to either support Spending increases, or to retard expenditure growth is a feature of the budgetary process brought about by the balanced budget requirement. Arnold Meltsner's research on Oakland California shows that in a forty-two year period the property tax rate re- mained the same from one year to the next only eleven times. Eighteen times it was lowered and thirteen times it was increased. He specifically points out how it was used as a technique to balance the budget: Currently the revenue subsystem uses a fixed tax rate tactic in which nonproperty tax revenue and ex- penditures are adjusted to the implications of the fixed rate. To balance the budget, either non- property tax revenue is raised or expenditures are cut. The preperty tax rate is held constant and is not used to balance the budget. Previously, when the tax base was more affluent and the pressures on the property tax were not as acute, the subsystem used a residual tax rate tactic. In that case, expendi- tures were cut to a "reasonable" level, nonproperty tax revenues were summed, and the property tax rate was set to balance the budget.5 Local property tax rates are not immutable to change, and often altered to satisfy spending preferences. So, because of the balanced budget requirement, local decision makers are compelled to consider the revenue implications of spending decisions. This does not exist on the national government and for this second reason, the most appropriate measure of muni- cipal budget outputs is one of the changes in total spending. 5. Mnold Meltsner, The Politics of City Revenue, (Berkeley: University of California Press, 1971), p. 117. 0.4 r I ..n: \ .a u ‘IIi l n u.. u, no. I I 3: ‘s- 64 A final difference in budgeting between these two levels is that on the national level expenditures are an instrument of fiscal policy. How much is spent in total is a tool of the management of the economy consciously man- ipulated to achieve varying rates of economic growth, unemployment and inflation. The size of the total amount of spending is taken out of the hands of the traditional participants in the budgetary system. Since there is less influence over the amount of total spending, the focus of attention shifts to the distribution of available funds into its component parts. But since this concern is not relevant for local governments, the fundamental decision for this budgetary system is how much is to be spent in total. Consequently for these several reasons, the analysis of municipal spending outputs focuses upon the percentage change in the size of total eXpenditure. Substantive Budget Outputs Department Budget Requests Beginning with the initial spending request of the departments in the fourteen cities (see table 3.1), a very clear pattern is evident. More funds are asked for than were received in the previous year. Departments seek to expand their funding level above currently allocated sums. The average of the total of all department requests is 20.06 percent above previous appropriations. The upward thrust of growth is almost uniform. In all but a single 65 TABLE 3.1 DEPARTMENT BUDGET REQUESTS . Percent Clty Change 01 40.22 02 19.92 03 22.27 04 11.62 05 37.94 06 14.67 07 43.34 08 -2.64 09 28.08 10 15.69 11 4.56 12 7.63 13 22.06 14 23.04 Mean 20.60 St. Dev. 13.03 66 city, total expenditures increase at the end of the first stage of budgeting. At the same time that expansion is evident, there is considerable variation in the absolute size of the increases asked for. The standard deviation of one half of the mean (13.03) attests to the extensive differences among the cities. The range of change in requests extends from the one city where an absolute decline of 2.65 percent is re- corded, to a 43.34 percent increase in another. While there are four cities where department requests are less than 10 percent above current spending levels, there are also three cities where total requests are more than 35 percent higher than present funds. Looking only at the direction of the change in individual department requests (see table 3.2), there is strong cor- roboration of the expansionary pattern of the outputs of the first budget stage. 0n the average, over three quarters of the separate departments in each city submit a budget of higher levels than they obtained in the previous appropri- ation ordinance. Except for the city where an absolute decline in spending occurred, at least 60 percent of the departments in sought increases: while in six of the cities more than 90 percent asked for more. Clearly. departments want increases for themselves compared to what they previously received; but, exactly how much more varies greatly among the cities. Departmental budgeting provides an unmistakable thrust for growth of 67 TABLE 3.2 IERECTION OF CHANGE IN INDIVIDUAL DEPARTMENT BUDGET REQUESTS + 0 - City Increase No Change Decrease % # of % # of % # of Depts. Depts. Depts. 01 92.83 (13) 0.00 (0) 7.14 (1) 02 92.83 (13) 0.00 (0) 7.14 (1) 03 87.50 (14) 0.00 (0) 12.50 (2) 0A 62.50 (10) 6.25 (1) 31.25 (5) 05 92.31 (12) 0.00 (0) 7.69 (l) 06 68.75 (11) 6.25 (l) 25.00 (4) 07 64.29 (9) 0.00 (0) 35-71 (5) 08 14.29 (2) 0.00 (0) 85.71 (12) 09 69.23 (9) 0.00 (0) 30.77 (4) 10 92.86 (13) 7.14 (l) 0.00 (0) 11 61.54 (8) 7.69 (1) 30.77 (4) 12 78.57 (11) 0.00 (0) 21.43 (3) 13 92.86 (13) 0.00 (0) 7.14 (1) 14 100.00 (13) 0.00 (0) Mean 76.46 1.95 21.59 St. Dev. 21.63 3.10 21.41 _—_ . on: '1 no" t ‘N.' 5.. . ..g 1. 68 municipal expenditures. The outputs of the first stage of decision making represent a strong, positive impetus to municipal spending. Executive Budget Behavior Response to Departmental Requests The executive's response to departmental requests for budget also displays a clear and consistent pattern of behavior, but in the opposite direction to the departments,6 as the executive reduces initial requests for increases by an average of 7.88 percent (see table 3.3). In every one of the cities, the executive reduces the expansionary Spending proposals of the departments. Even in the one city where the total of all requests was below the level of the previous year, departments were still cut. However, variation is just as much in evidence, as the standard deviation is quite close to the mean (6.45). The reductions imposed by the executive range from a minimal decrease of 0.18 percent to almost 20 percent of the amount that was initially sought. In nine cities, department budgets are lowered by less than 6. Previous research by Sharkansky and Fenno has labelled this measure the "success," or ability of each actor to get what he asked for. It is calculated in exactly the same way as the present research, but reported in the Opposite direction. That is, these two authors write that department heads obtain 85 percent of what they requested. Currently, this same decision is characterized as a 15 percent reduction by the executive. This latter formulation is more consistent with the three stage model of budgeting. It is the executive and not the departments who make the decision and this measure is consequently more appropriately viewed as the result of the second stage. The executive's reduction in de.. Pertmental requests than it is considered the success of the first stage of department budgeting. 69 TABLE 3.3 DIRECTION OF CHANGE IN INDIVIDUAL DEPARTMENT BUDGET REQUESTS City Percent Change 01 -18.26 02 -15.69 03 "' 5079 04 - 3.28 05 - 3.50 06 - 7.78 07 “19096 08 - 0.18 09 —10.41 10 - 1.75 11 " 0070 12 - 2.69 13 ~14.08 14 ~ 6.32 Mean - 7.78 St. Dev. 6.45 01-" ’ .1. I . pl I“ .‘. - I III OP ‘D'I'w ....,; ' C “I... 1.. \ a» I ”‘5" I,.‘. u v.. u“ x.‘: Cs , I“ ‘u 70 10 percent, while in the remaining five cities, they are decreased by up to 20 percent. The direction of the executive's change in individual department requests strongly emphasizes the cutting de- cisions of the second stage of the budget process (table 3.4). An average of 68 percent of the separate departments experienced some reduction in their spending request. In every city, except one, at least half of all departments were out. But even this figure underestimates the negative reaction of executives to proposals for more funds, as an average of only 10 percent of the departments are increased, and only 22 percent remained unaltered by executive decision making. In twelve of the cities, the number of departments decreased is greater than any other result; while in ten cities, the number of departments unchanged is greater than the number increased. The executive is quite unlikely to add to initial budget requests and give to the departments more than they originally sought. At best the departments can expect to minimize the size of the reduction the executive most certainly will make in their proposed budget. As strongly as the departments serve as an impetus for the expansion of municipal Spending, the executive functions as an opposing force who cuts initial requests in the course of preparing his own expenditure recommendations to the legislature . - ‘ ‘6. i‘i \. 71 TABLE 3.4 DIRECTION OF EXECUTIVE CHANGE IN INDIVIDUAL DEPARTMENT BUDGET REQUESTS + 0 - fity Increase No Change Decrease % # of % # of % # of Depts. Depts. Depts. 0.00 (0) 7.69 (1) 92.31 (12) 0.00 (0) 0.00 (0) 100.00 (14) 12.50 (2) 37.50 (6) 50.00 (8) 0.00 (0) 20.00 (3) 80.00 (12) 0.00 (0) 41.66 (5) 58.33 (7) 18.75 (3) 12.50 (2) 68.75 (11) 7.14 (1) 21.43 (3) 71.43 (10) 30.76 (4) 46.16 (6) 23.08 (3) 21.50 (3) 21.50 (3) 57.00 (8) 7.14 (1) 21.43 (3) 71.43 (10) 7.69 (1) 30.77 (4) 61.54 (8) 14.28 (2) 14.28 (2) 71.44 (10) 14.28 (2) 0.00 (0) 85.72 (12) 0.00 (0) 38,46 (5) 61.54 (8) Ban 9.57 22.38 68.04 . Dev. 9.27 14.35 18.42 O .0. '" .., 0:”: Q l I.. IIID. .,.‘u .- I no . u. . n 'I... .G u.‘ 'u:.. o. v. A 5:31 .~'--, - . n- “» .., “t. u “o. ‘ M ‘od 5 I "‘ ‘I V 72 Executive Budget Recommendations Although the executive cuts departmental requests for .ncreases, he does not eliminate all growth entirely. His )wn recommendations to the legislature still register an ncrease in spending above the level of the previous ap- 1ropriation ordinance. The Budget, is now 10.67 percent .bove what it was in the previous year. About half of the ncrease departments originally submitted remains at the nd of the second budget stage (see table 3.5). In the hirteen cities where increases were first proposed, de- artments still possessed them. The cutting behavior of he executive has held down the size of increases, but has ot turned around the direction of yearly change. In no it)! have increases become decreases as a result of ex— eutive Spending decisions. Differences among the cities re great, as the standard deviation is large (8.44). In elf the cities budgets grow by less than 10 percent, while he Other half expands by more than 10 percent, up to a pending increased of a full one third. This Pattern of annual budget growth in executive acommendations is corroborated by the measure of the .rection of change in individual department spending 150135113- leverage of over 70 percent of the separate departments in Lch City Obtain increases in the budget submitted by the :ecutive to the city council (see table 3.6). This is only percent less than the number of departments that first H 73 TABLE 3. 5 'IXECUTIVE BUDGET RECOMMENDATIONS City Percent Change 01 14.62 02 1.13 03 15.18 04 7.96 05 33.07 06 5.75 07 14.73 08 - 2.81 09 14.74 10 13.67 11 3.82 12 7.42 13 4.88 14 15.26 Mean 10.67 St. Dev. 8.44 "'Oncu . .— II q‘i . .b'CI..‘. 5. o ‘Ny u.“ t o .._J TABLE 3.6 OF CHANGE IN EXECUTIVE RECOMMENDATIONS FOR INDIVIDUAL DEPARTMENTS + 0 - Increase No Change ,Decrease % # of % # 0f % # 0f Depts. Depts. Depts. 85.72 (12) 0.00 (0) 14.28 (2) 71.43 (10) 0.00 (0) 28.57 (4) 81.25 (13) 0.00 (0) 18.75 (3) 62.50 (10) 6.25 (1) 31.25 (5) 92.31 (12) 0.00 (0) 7.69 (1) 56.25 (9) 0.00 (0) 43.75 (7) 57.14 (8) 0.00 (0) 42.86 (6) 21.43 (3) 0.00 (0) 78.57 (11) 71.43 (1) 0.00 (0) 28.57 (4) 85.72 (12) 7.14 (l) 7.14 (1) 61.54 (8) 7.69 (l) 30.77 (4) 71.43 (10) 7.14 (1) 21.43 (3) 71.43 (10) 0.00 (0) 28.57 (4) 100.00 (13) 0.00 (0) 0.00 (0) 70.68 2.02 27.30 18.63 0.93 25.40 “'l"" . . “a. 0" , I u... out ;--n-”"" o ,4 ' \ ”-- buuui'l . III : ,0. o‘p-O .- n u- i to» c s 'II—a co - ' I \ "'Ngnuh - F. H :‘—.t. . In“. \ “'. -.n a. . | D: l a -. ' -" ““ Vuu “Iq n".: 5 a '0. " K ‘ -eb.~ - Ill 75 an increase and it is only by this small number executive has denied all growth in expenditures to ts. 8 clear that the executive's budget cutting knife t sharp and deep that budget growth is entirely . Instead, the review of departmental requests :and limits expansion. Spending is allowed to but at a slower rate than the departments first As a result of executive action expenditure levels d of the second stage of municipal budgeting are istent with the level of the previous year than t heads would like. As Allen Schick describes the e position of the governor in state budgeting, he . the chief gatekeeper of state budgeting. In role, he directs the pace of spending increase rogram expansion by deciding which of the ams proposed by the agencies, or by his own e shall be included in the budget.7 Legislative Budget Behavior onse to Executive Recommendations e executive opposes the departments requests for the legislature similarly reacts negatively to ive's own expansionary recommendations. On the he city council reduces the budget they receive by 1.36 percent (see table 3.7). As small as llen Schick, Budget Innovation in the States n, D. 0.: The Brookings Institute, 1971), 9. 76 TABLE 3.7 LEGISLATIVE CHANGE N EXECUTIVE BUDGET RECOMMENDATION City Percent Change 01 2.30 02 “3010 03 -O.6l 04 -3092 OS 0 06 +1.03 O7 -0.0l 08 O 09 -0.64 10 0 11 -6029 12 -0.25 13 -6.62 14 1.11 Mean -l.36 St. Dev. 2.55 -. ~l“-. ' .“ nun-nu. .0. ’ . ’l ' I o . .3 o .. .. . . . 7:“ vqr‘ ""0 'l -.'v. ’ I up: :n: 6 p‘ a ' ' 0 . on .“ ~ It . a. a e..- .VKQD. ‘- "on ‘ «'3‘ ‘ h; ~"qu. ' . 77 re first appears to be, it hides just as much as ; about the third stage of the municipal budget The action of the legislature is much more varied x, and in the end more significant than the ex- ore extensive reductions in department requests. three distinct patterns of expenditure outputs, islature either increases, decreases, or leave the executive's budget. ast common decision taken, in eight cities, is to lget. This comes out to an average of 2.68 percent, >m a decrease of 0.01 percent to 6.26 percent. er hand, three city councils add to spending totals, .ge of 1.48 percent (1.03 percent, 1.11 percent, and t). Then, there are three cities where no alter— 11 are made by the legislature in the executive's ions.8 Finally, the average absolute amount of t is, disregarding its direction, is 1.85 percent. ied outputs of legislative decision making is the measure of the direction of change in individual udgets (see table 3.8). An average of 22 percent tments are increased in each city, 51 percent d, and 27 percent are reduced. In six cities, f departments increased is greater than the number absence of legislative change would also come h internal rearrangement of department expendi- t altering the amount of the total. While ssible, in the present fourteen cities this is Here the legislature makes no changes in total A) I. 78 TABLE 3.8 DION OF LEGISLATIVE CHANGE IN INDIVIDUAL ARTMENTS IN EXECUTIVE RECOMMENDATIONS + 0 - Increase No Change Decrease % # of % # of % # of Depts. Depts. Depts. 53.84 (7) 0.00 (0) 46.16 (6) 0.00 (0) 92.86 (13) 7.14 (1) 18.75 (3) 50.00 (8) 31.25 (5) 20.00 (3) 13.33 (2) 66.66 (10) 0.00 (0) 100.00 (12) 0.00 (0) 68.75 (11) 18.75 (3) 12.50 (2) 28.57 (4) 57.14 (8) 14.29 (2) 0.00 (0) 100.00 (13) 0.00 (0) 35.71 (5) 50.00 (7) 14.29 (2) 0.00 (0) 100.00 (14) 0.00 (0) 0.00 (0) 46.15 (6) 53.85 (7) 50.00 (7) 21.43 (3) 28.57 (4) 7.69 (1) 7.69 (1) 84.62 (11) 23.08 (3) 61.54 (8) 15.38 (2) 21.89 51.65 26.66 51.35 43.46 34.21 ’1' . ...‘ t... .,; 79 d4 while in five cities, the reverse pattern occurs. knee other cities no change is made. On the other nine cities the number of departments unaltered is than the number either increased or decreased. The e budget is just as likely to be increased by the mre, as it is to be decreased. But at the same time, st as likely to be finally approved exactly as it msed, as it is to be modified. qmred to the uniformity of the executive's reductions tmental requests, what constituted low of high amounts e does not so simply exist. Previously, there was no at a 5 percent cut is less than a 10 percent reduction. ;he present case, it is uncertain where a 2 percent n stands in relation to a 2 percent addition. There herent reason to suggest that an.increase represents ivity on the part of the city council than does a . And the position of no change is also uncertain; stand for the complete absence of activity, or only it between additions and subtractions to expenditure sequently the arrangement of the direction and mag- E change in the executive's budget does not lie in rical value itself, but in the concept underlying the F; (continued) expenditure levels recommended by the 2 because they make no modifications at all, ac~ .t exactly as submitted. .‘A. ' ”out In.“ .. -I. to: It I \ “WI nu. . M Q“ "In. \ .‘. . \ ‘ 80 measurement of legislative decision making. Thus three different rankings are identified. The first is where the direction of change is ignored, and only the absolute size of modification in spending totals forms the basis for the scale of low to high. Since it cannot be determined whether increases or decreases are more or less of a change, they are logically equivalent and only the absolute size is important in assessing the legislature's decisions on ex- ecutive recommendations. 0n the other hand, taking into consideration its direction, either increases or decreases can be viewed as either low or high amounts of change. 1 Actually they are the same, being only mirror images of each other; but for the sake of clarity and emphasis, they are considered as two different patterns of legislative decision making. Legislative Budget Appropriations The last measure of municipal expenditure outputs is the final appropriations ordinance and how much it changes from the previous year. The legislature, on the whole, OCCuPies a position similar to the executive. For while the budget is reduced, the expenditure end-product still increases. The average growth of the budget in the fourteen cities is 9.38 percent (see table 3.9), as 68 percent of the individual departments obtain increases (see table 3.10). The legislature allows spending levels to expand, but at a Slower rate than either the executive recommends, or the I I 81 TABLE 3.9 LEGISLATIVE APPROPRIATIONS City Percent Change 01 17.25 02 - 2.01 03 14.48 04 3. 73 05 33. 07 07 14.72 08 - 2.81 09 14.01 10 13.67 11 - 2.71 12 6.66 13 - 2.07 14 16.54 Mean 9.38 St. Dev. 9.89 82 TABLE 3.10 DIRECTION OF CHANGE IN INDIVIDUAL DEPARTMENTS IN LEGISLATIVE APPROPRIATIONS + 0 - - Increase No Change Decrease City % # of % # of % # of Depts. Depts. Depts. 01 92.86 (13) 0.00 (o) 7.14 (1) 02 64.29 .(9) 0.00 (0) 35.71 (5) 03 87.50 (14) 0.00 (0) 12.50 (2) 04 68.75 (11) 0.00 (0) 31.25 (5) 05 92.31 (12) 0.00 (o) 7.69 (1) 06 56.25 (9) 0.00 (0) 43.75 (7) 07 57.14 (8) 0.00 (0) 42.86 (6) 08 21.43 (3) 0.00 (0) 78.57 (11) 09 78.57 (11) 0.00 (0) 21.43 (3) 10 85.72 (12) 7.14 (1) 7.1“ (1) 11 46.15 (6) 7.70 (1) 46.15 (6) 12 71.43 (10) 0.00 (0) 28.57 (4) 13 28.57 (4) 0.00 (0) 71.43 (10) 14 100.00 (13) 0.00 (0) 0.00 (0) "man 67.99 1.00 31.01 8t.rev. 21.40 0.12 26-50 ...otl 83 m initially request. This pattern of decision consistent with the description Richard Fenno the expenditure outputs of the House Appro- Committee, which: . . grants a great majority of agency requests for dollar and cents increase over the previous appro— iation. But, of the increases granted, the erwhelming number are for less than the agency quested . . . The modal pattern is for an agency (request an increase and for the Committee to ant it a smaller increase than requested. The mmittee thereby cuts the budget but permits a .nservative growth.9 .wever, a closer examination of the distribution of ye action, reveals a standard deviation (9.89) which (me as the mean, which suggests that legislative making is more varied than this initial inter- lWOUld lead us to believe. There are, in fact, :ies where, as a result of legislative decisions, .budget is below the level of the previous year. particular cities final appropriations decline by e of -2.26 percent. Here, executive recommendations rases are cut so deeply that the result is an ab- '0p in appropriations. Even though the amount of sed by the legislature is not great and the magni- he decrease is not large, the impact of legislative making is significant and should not be minimized. Richard Fenno, The Power of the Purse (Boston: rown and Company, 1966). Pp. 356-357. a.“ PM! 5.- h . 10‘: 'n I. .4": . th'l Wu..- 84 In these cases they have not slowed the rate of growth, as did the executive, and as other city councils have done. Instead, they eliminated all growth entirely, completely turning around the outputs of both the first and second stages of the municipal budget process, the city council has placed to other budget actors. By disaggregating the outputs of the last stage, it ‘is evident that municipal legislatures play a far more crucial part in budgeting than is first indicated by the average response of all fourteen cities as other researchers would let us -'believe.10 Incrementalism The concept of incrementalism has an inclusive presence in the literature of public policy making. It purports to account for both the process of decision making, examined 10. The comparability of results from the present fourteen cities to previous studies on the state and national level is strong. Ira Sharkansky's study of nineteen states reports the following budget outputs: departments request an average 24 percent increase, ranging from a 15 percent to a 33 percent expansion, executive reductions average 14 percent and extend from a 4 percent to a 31 percent cut. The leg- iSlature reduces the budget by an average of 2 percent (from between a decrease of 8 percent to an increase of 19 percent). Legislators in six states cut the budget, eleven states increase it and two states leave it unchanged, as the absolute change is 5.6 percent. Final appropriations then increase by 13 percent from 1 percent to 39 percent. Richard Fenno's study provides a comparability to the descriptive measure of the direction of individual department Change. Here 81 percent of the department heads submit requests for more than they currently receive, l6.percent submit requests for less than they currently receive and 3 Percent submit requests for the same. The legislature cuts 74 percent, adds to 8 percent, and leaves unchanged 18 percent, as final appropriations increase in 69 percent of the cases, declines in 27 percent and remains the same in 4 percent. a . "d . I “(A u .-n I -.‘.l " p f" .‘ .‘ 85 in Chapter Nine, as well as its specific end products. In this latter regard it offers a description of the size of change in expenditure levels and an interpretation of the interrelationship among the outputs of the different stages.ll Each of these are now examined. Absolute Size of Change As a description of different size change in expendi- ture levels, either from one stage to the next or from one year to the next, the concept of incrementalism purportedly provides a standard to differentiate between a small and a large amount of change. As David Braybrooke and Charles Lindblom write: ". . . a small or incremental change is one that, within some short time period, such as five years, is small or incremental, regardless of the indefinite future."l‘2 11. The concept of incrementalism has a still third meaning and this is the relationship of spending levels in one year to spending levels in another. The research of Ira Sharkansky in Spending in the American States clearly demon- strates the prominence of prior expenditures in the determin- ation of current spending. By means of both correlation and regression statistics, the dominant influence of the past upon the present is established. The research of Otto Davis, M. A. Dempster, and Aaron Wildavsky, "A Theory of the Budgetary Process," American Political Science Review, vol. LXVI, no. 4 (December, 1966), pp. 529-547, also points to a similar set of relationships: The executive's recommendations are a function of the previous year's appropriations; and then Con- gressional action is a function of what is proposed to them. Since the present research examines only a single year slice of time, and is not concerned with the stability and recur- rence of expenditure patterns, but the amount of change within anyoone year, this component of the concept of incremental de01sion making does not concern us. 12. David Braybrooke and Charles Lindblom, A Strategy flecision: Policy Evaluation as a Social Process (New York; The Free Press, 1963), p. 64. an, n I C. I .‘l - ‘2 v: 86 this operationalization of the concept does not very far. To define a "small or incremental change? at is "small or incremental" does little to elaborate ng or to provide a precise quantity to distinguish ifferent kinds of change in public policy outputs. ver, Richard Fenno, in his study of Congress, attempts y this ambiguity, by establishing specific dollar the criteria of incremental change for the two of expenditures employed in the present study. In of a change in spending from one stage to the next, nt figure is posited as the benchmark: dominant pattern is not a wholesale slashing of my budgets. Most committee reductions (and in- ses) in budget estimates are marginal ones, ing between an increase of 5 percent and a de- se of 5 percent in budget estimates.13 of the difference in spending levels from one year xt, a 10 percent is then proposed as the standard: vear-to-year expansion of the agencies is kept inal by the Committee's action. A majority of the ittee's decisions (53 percent) involve no more than percent change over the previous year's appro- tion. If one takes changes of 20 percent as the ff point, three-quarters of all the cases are ided. At each 10 percent interval the number of 3 drops, until a few extreme examples of growth atardation remain at either end of the scale.l4 though these two amounts were originally formulated 1e the spending decisions of Congressional committees, Richard Fenno, Power of the Purse, p. 353. {ichard Fenno, Power of the Purse, p. 355. no" a ‘3‘ . In... 1 t u, o h. ‘ ”I -N \\' 3‘ 87 describe the various expenditure outputs of muni- geting. The mean of total department requests nt) falls on the outer boundary of incremental ange while the increase recommended by the executive nt) and the final expansion of legislative appro- (9 percent) are clearly consistent with this ization of marginal change in expenditure levels. reduction by the executive (8 percent) and the cut gislature (2 percent) also generally corroborates mental pattern of change from one stage to the next. ite the strong similarity in municipal expenditure 0 the dollar values, the concept of incremental policy still leaves much to be desired as a descrip- ifferent amounts of change in expenditure outputs. n the statistical regularity of the size of de- ade, there is no a priori reason for selecting numerical values from any others as the appro- viding line between a small and a large change. ires of 5 and 10 percent do not derive from the the concept itself, but simply from the empirical 'on of actual expenditure outputs. Furthermore, te value of this criterion is not invariant, but does the specific content of the decision. Thus difference in what constitutes an incremental m the 5 percent figure for within year change and cent figure for between year change: but why such ce is not explained and most importantly does not on: on. ~~oo v I 0.1. It.‘ . A. 88 from the concept itself. Ifinally, the attempt to define the scope of policy into mutually exclusive and jointly exhaustive cate- rxfizonly seems difficult to achieve, but draws onse, as reported by other research,17 it appears municipal legislature goes along with the direction .ng expansion established by the executive (-0.3702 The more of an increase the executive recommends, the legislature cuts and the more it adds to spending Conversely. legislature cuts the budget. The size ive budget growth sets the tone that is followed sized in the third stage of decision making. There etry and consistency in budget behavior of the ex- nd the legislature as the executive provides the subsequent decisions. If the executive submits a dget with a comparatively small percentage increase this is taken as a signal that money is not available. g the legislature to then make large cuts. 0n the Ira Sharkansky and Augustus G. Turnbull, "Budget Georgia and Wisconsin," p. 633. ...v . ..-¢vo . - ul.’ .5: ‘l 93 uh if the executive recommends an "expansive" Lth relatively high rates of growth, the legis- Lmilarly takes this as a signal that additional are plentiful and proceeds to cut the budget less still further spending increases on top of what 1tive has already proposed. possible explanation for this pattern, and the ab- an incremental routine is that the legislature is 1te to the decisions of the executive (see Chapters )ugh Eight). They do not make an independent of spending choices but instead rely upon the a, and follow his lead in their own budget review. mrkansky writes of such a system on the level of rernments: e findings of greater importance for the governor's :ommendation (rather than the agency's request) in elegislature's decisions indicates the legislature's endence on the governor's budget cues . . . State islators have a desperate need for cues that will de their budget performance and the governor's re- mendation is usually the best cue available.18 lature does not respond negatively to recommended because most of the budget cutting decisions had aen made. That is, the executive by this time has 1e expansionary requests of departments to more re— levels, so there is little left for the legislature ,— Ira Sharkansky, "Agency Requests, Gubernatorial d Budget Success in State Legislators," American Spience Review, vol. LXIX. no. 4 (December, 1968). au- u). - 94 RL Relying on the executive to do most of the the; the decisions of the legislature are dependent upon vious executive spending choices. All that remains for nis to modify, along the margins and in the same direction thecflmnge set forth in the executive's recommendation. fine is the case then it is expected that the more the nflfive reduces department requests, the less the legis- 1re reduces executive recommendations. But this is not rdMneted by the data (—0.0681). The expenditure decisions Hmelegislature take place independently of those of the nflflve. Executive reductions in initial requests are not dmtitute for the legislature's own review. The city lcilis not constrained in acting as a result of previous mdihue spending choices, responding not to the process mecutive decision making, but to its end products; the fetembmitted to them for review. The municipal legis- re compared to the executive does not appear to follow emental evaluation rules. Yearly Changes Executive Budget Recommendations The second statistical relationship suggested by the mental model of decision making is between the cuts made g the three stages of budget review and the size of y growth. Very simply, the more the budget is reduced, ess it expands from one year to the next. However, in of executive recdmmendations this relationship is not . In fact, a moderate positive correlation (0.2396) is ._'.Z a) 'ul-uq‘a a luOol-V \ . . 'II finn~g1 .n "‘ 44‘s.! v 22': u . U... ‘. \ 99?... ‘ u ‘N :'- 1.. . 334;." ‘:0,"‘Il\ ‘i‘ r'l ‘l . I...‘ "YA: ‘ "I 683‘} "U ‘J. .l“ : I ' ‘ \‘ r, w ‘E II n h h...“ I'm; ‘ u e i“. u“ ~L, . \ a: ‘nzel .,'\ , N \.~ \‘ t I '95 Presumably, the agencies that request the greatest increments will receive the largest percentage growth over their current budget, while agencies requesting little or no incre- ment will receive little or no growth in their budget.l9 Departments do not get something for nothing and to lude otherwise is a fundamental misunderstanding of budget process. Thus it is an error for department .3 to calculate, in terms of the executive, as Rufus ming writes that, ". . . if you ask for a small amount, legislature is likely to give you something. but if ask for a large amount, the legislature might cut it away . . ."20 Or as Otto Davis, M. A. Dempster, and m Wildavsky write: ". . . if they ask for amounts much :er than the appropriating bodies believe reasonable, the (cies' credibility will suffer a drastic decline. In [circumstances, the reviewing organs are likely to deeply, With the result that the agency gets much less 21 If such it might have with a more moderate request." ctations are followed, little growth will follow. But even more importantly, this relationship questions interpretation of the executive as the dominant budget 1‘. whose policies determine the department totals. It M 19. Ira Sharkansky and Augustus G. Turnbull, "Budget 1g In Georgia and Wisconsin," p. 634. 20. Rufus Browning, "Innovative and Now-Innovative :10n Process in Governmental Budgeting," Iin Ira Shar- Iy. Policy Analysis in Political Science. (Chicago: lam: 1970): P- 325. 21. Otto Davis, M. A. Dempster, and Aaron Wildavsky, “W of the Budgetary Process," p. 530. 1.1.”. a 'P: Nu. '“v- ‘ \ '00.. '9“: . ”'0' J. .n, L." \21 t u ‘1) 96 appear otherwise. It is the departments themselves, .size of their initial request for increases, who :he largest part in determining executive budget ions. As executive reductions in requests are only a ation of how much more is sought, executive recom- 1fions to the legislature for expansion are similarly a statistical determinant of initial department bud- Instead it appears that it is the departments elves who determine their own expenditures by the seas they first request. As Allen Schick writes: Ike agencies control the pressure gauge of state spending. By increasing their pressure, the agencies can compel the Governor to (open the gate wide to new programs.22 tive decisions do not alter in.any appreciable way attern of yearly Spending growth.first established e»departments. The executive does not break out of tructure of the relationship created by the departments elves. Legislative Appropriations Ekamining this same incremental relationship in terms gislativezreductions in executive recommendations and nount the final appropriation.ordinance increases re- the strong presence of such a model of budgeting. The :he legislature reduces the budget they receive for l‘ 2. Allen Schick, Budget Innovation, p. 179. A '0‘: fine" 'vu' , - , w: :2“; ca Ht..- '.¢Il i Q I ' o .I .. o In“! '7: ’ ‘ .5: 1- l...’ I ~oh , "an aver u.»- 10h- .‘i v w} Ole. val. Jail «'6.- 3,21 -- ~. ‘ .. u... ..‘.‘- ‘oih. . § :0 o-EQIQ 1“" .....v: z N - u‘ .h I I: n Mui..‘ ”.v u. h I . C.“ no. I ‘ 5‘. - . ‘4‘ fly. .. .. "3.93;, O nq‘.‘ “ ‘C :1' :“fihfi fl eunu‘.; \- l|‘.. ‘ .:u:: \ :y‘q ch ‘ V .12‘9 a W. 'h H: | 2.“; l ‘ l ‘ ‘ ‘ . “‘u " t‘ V“ .‘I' I ' \‘ . .u‘. | ‘ ‘ . \\c,:\eq 0“ \ I~ . k \‘ ‘ “ “ ‘~ \ \ “£31 < .: ‘ ‘ I‘m \ \‘ ’ I' "N! D” 2' \ J “I .h“ \' 97 view, the less appropriations increase over the pre- ous year; or conversely, the less they cut the executive's get and the more they add to it, the more spending levels rease over the preceding year (plus or minus 0.6002 + 25). Although such cuts do hold down the rate of expansion the final appropriation they do not in any appreciable er alter the pattern of growth set forth by the ex:-.: tive's budget. The more of an increase is recommended them, the more of an increase is approved (0.9571 + 0.001). 3 very strong relationship demonstrates the almost com- ate correspondence between the outputs of the second and rd stages of the budget process. Furthermore. this same y strong connection is evident between initialdepartment uests and final legislative appropriations (0.837LL + 0.001). more departments initially request, the more they wind with in the end. It seems that no matter how much is ninated by the executive and the legislature. the size ’anreases first requested is the single most important :istical determinant of final expenditure outputs. It mly by asking for more than departments obtain such *eases. ' The decisions of the second and third stages of the et process do not alter this pattern. However, this not mean that the expansionary thrust of the depart- 3 is not contained. It must not be forgotten that in "0:: 3: fkfi ,’ I"' I. .“' ‘ p I .. ‘00. I'”I’ a l r p—V'cbfi'- u .. . N C6“ \P eVID' u .. ran. it “‘1‘: :r .‘n;{ and] nu i u" I 'I ' “.ng. ~ - - ‘9. ".1 3““..va i... '. I. “up; ...,I "I ul-‘ .. n ‘ . ‘v‘;\ou ~“ I. ' IA ...I' ...h z u ,th‘ - v- "1..“ ~333‘ ‘ ‘ I .; _“‘r;»-. -" ‘Inih \ 'R . "~ ‘ r :‘ssm‘ 1 it? the 3,79 lEVe‘ n‘ u 98 ee of the cities the municipal legislature passes an propriation ordinance below the level of the previous ar. So even though a statistical relationship still ds, in these cities it is no longer increases in ex- ditures that are being written about. but decreases. it would not be justified to conclude from this analysis t'the legislature is unimportant. because of the close ection between final appropriations and previous ex- tive recommendations and initial department requests, t that the departments are more significant than others e portrayed them. Conclusion The two dependent variables of the expenditure outputs the municipal budget process are the percent change in a1 expenditures from one decision making stage to the t, and then from one year to the next. For several sons this measure is considered the most appropriate out- of the municipal budget process as compared to the tonal government. The specific substantive findings be summarized: slant ive Patterns Department heads request increases. The “executive reduces initial requests but still mmends to the legislature a budget higher than the ious year. The legislature reduces the executive recommendation also still winds up with an apprOpriation ordinance e the base of current spending. However, in three cities the legislature cuts the et they receive for review so deeply, that expen— re levels decline below the budget ordinance of the ious year . ' ' .- . n '...4 no- 5:- - *' ' . I!“ 5" I a 0". II:I\N" .- A ‘ -: w‘. . -| .N. -IOV 1 ‘ A pun‘n'l: 0“ I .:n..v...~ . b. :YAQHO‘”° I .‘i'v‘fi. V :vA-g. an ex“: 5!“.0 v. .V. ' I ' \ llvv.!—‘ OA‘ R . db ‘ ‘_ um. v .nggfl ‘l “HI“- 'I , "human. ' \- -\ u .u _ u , :‘a I luv -... DU: ‘5': ..RA -I an .y,‘ 5‘; I -1 In. . A \ F: ""‘.'l .chV I I ‘u. . '. x 2“ “a “rm-mu. 5 I ".~ - -| “t...“ “c“ a... ‘hvl4 A. ' I arr-«.0 .; H II‘I.L ..‘:~‘ 99 atistical Relationships The more department heads ask for the more they are cut the executive. The more department heads request, the more the executive commends for them to the legislature. Executive cuts do not bring about a decrease in the aunt of expansion submitted to the legislature. The ex- tive fails to contain the positive thrust of departmental nding. The more the executive recommends to the legislature, more the budget is cut; and the more is added to spending tals. The city council follows the lead of annual spending ange established by the executive and the departments. The more is either requested by the department or recom- nded by the executive the more is finally approved by the gislature. Departmental requests are the single most important itistical determinant of subsequent expenditure decisions. , .-n-. . . uh ' I "‘1I- V! m. 1.. . I. " IVA i - -' u ‘k. "I t I ' I I :‘:y§. Ill'i‘.‘ .‘ apter Four: Budget Roles Introduction The concept of a social role has long been the subject impflxy'by social scientists, and while no contribution mutently made to its theoretical status, the specific bemumive roles occupied by the department heads, the ex- Hive and the legislature are examined as the first component thermuficipal budget system. Thus, the normative expecta- nx;of behavior and attitudes that are attached to positions flfirIthe organizational arrangement of government are flflighted as the first explanation of expenditure outputs. rexistence of these roles constitute a proscriptive guide appropriate and inappropriate actions and decisions, and 13m extent they are clearly defined, agreed upon and ad- ed to, roles provide a regularity to the interactions r@;participants in the decision making system. The actual budget roles of the three sets of municipal yn:makers closely follow from the formal responsibilities each actor, as well as being suggested by the observed anditure outputs of each.stage. The departments initiate sequence of decisions by their requests and provide the 100 101 source for program and spending expansion. Thus they characterized as the "spenders." The executive is gaiudth.the responsibility of preparing a comprehensive integrated budget. He is portrayed as the "economizer" e opposes the expansionary thrust of departments and mmends to the legislature spending increases that are consistent with the level of the previous year. Finally, legislature possesses the legal right to review executive nmendations pursuant to adopting the authoritative :umdations ordinance. They modify executive spending simun by either increasing or decreasing spending haand are characterized by their traditional role of :sight of administration." The analysis of the present chapter not only attempts qflain.more fully the meaning of these roles and to me their presence on the level of municipal government; m)examine the exact relationship between adherence to normative orientations and concrete budget behavior. at extent do such roles directly translate into ex- ture choices made in each stage? Department Budget Role As departments prepare the initial set of expenditure as and submit requests for more funds than they re— 1 in the previous year, they are portrayed as the lers" of the municipal budget system. John Crecine : that the objective of department heads ". . . is 102 btain the largest possible amount of funds for his rtments and his purposes.":L Thomas Anton concludes ,'3 . . very few responsible agency administrators be likely to request _l_e_s_s_ [italics in original] money is currently available to them."2 Jesse Burkhead adds his common characterization of the budget role of de- ments that, ". . . administrators have a tendency to be rdelistic, that government officials have an inborn desire [mnd more of the taxpayers' money, to hire more people, LfiJfi more buildings."3 Similar perceptions of the part aitw'departments are held by those interviewed in the nnzstudy: "Department heads always ask for increases-- Imrsonnel and more equipment. Always growth . . . They muure builders. Each one wants as much as he can possibly :0 run his department . . . Some department heads put thing they can think of in their budgets. They 'go in moon,’ it's standard operating procedures." Two interpretations for the existence of this expan- ry role are evident. One is that the department head e advocate of the policy goals of his own particular rmance area. The second is that the department head ,__ 1. John P. Crecine. Governmental jroblemfiolving: A yer Simulation of Municipal Budgeting—(Chicago: Rand Ly. 1969) 9 p. 500 3. Thomas J. Anton, "Roles and Symbols in the Deter- .on of State EXpenditures, " Midwest Journal of Political 3: vol. XI, no. 1 (February, 1967), p. 28. 5 Jesse Burkhead, Governmental Budgeting (New York: hley and Sons, Inc., 1955). Po 249. i=- 2 7., I |< ’. '3 1'- 'v I I , l‘! p g ‘0' ‘uu' ' I II. 'I \~‘. - - O. ‘5 h ".52 103 he padder by asking for more than is needed and more than xpects to receive. Each of these is examined in turn as xplanation for the size of expenditure increases sought. Departmental Advocacy The first explanation for the department's spending stems from possession of an advocacy orientation. The ence of such a role derives from the makeup of cipal government as an organization. What is commonly In:of as a single administrative branch is actually osed of many separate sub-units that come together to the organization as a whole. Each of the departments has wn goals, its own values, and its own interests, as they 'de different and often conflicting services to the nfity. It is only natural that each department desires qmnd the scope of its activities, to extend the boundaries :s jurisdiction, and to improve its financial position 11the organization. Consequently, as chief operating er, the individual department head assumes the position fender and promoter of his own particular program and y objectives. Furthermore, department heads consider themselves and toked upon by others as "professionals," as experts in fields. As such, they are in the best position to set the of needs and wants for their department. They articulate goals and future programs as their professional norms them to high aspirations and establish a record of iShments. As Arnold Meltsner quotes a department 1014 in Oakland: "1 am a professional; no one, not even the ger, can understand this job unless he sits here: I ' I J I"; 3:) [I 1.: n 1' ‘J 148 3 us think we have influence, but it's an illusion . . . :ouncil ought to be operating at a high policy level. Iould be giving the chief administrator their priorities finding and deal only with the level of the type of res they think ought to be performed . . . Their part d.be one of sorting priorities and determining program u and not getting down to details. They have the re— Fibility as well as the right to say they want to move us or that direction." But this rigid distinction between the formulation of antive policy and overseeing administrative details fails derstand that it is often only through the supervision ministrative particulars that the legislature can have mact on policy: "We make an item by item inspection of mendations—- about two hundred pages of IBM paper. Some- this is silly and too much, but sometimes you have to at details to see what the situation is." By a line he review of purchases and costs, the legislature develops apabilities and resources to participate in shaping policy. ‘len Schick writes: "Legislative control over details, a guard against administrative transgression, now is a e for influencing the course of administration-made y . . . control over details has become a mechanism 37 ontrol over policy." 37. Allen Schick, Budget Innovation, p. 183. often this perv 5.0.3.: is forgotten 1202200: for PPB is um, "Ed SU‘DSTitUT is actually being 016 1:200:001 policy; < fizz-2:100. As Fish: .e long-stand fine Appropmat far :00 much *0 the specifics "09:11“ the i 9:0:ive View 01 nonly the 30:0 allow the .0. :1II€*‘S o: f:‘ ”-0 that 3 pr Omse 1nq0 #3303108 and "SS igatior 0f 21:59. icatior 6503331 0 L96‘ ~10 Effective] :dutne era: 1:»ervemlon 0 auministra. ' CapaCity to a 30319110 forms: QM =~ tive conti: o_f,? 149 Often this pervasive fact of legislative decision 'ng is forgotten in the reformer‘s zeal for improving the lity" of legislative participation in budgeting. The ment for PPB is such an example. By eliminating line 5, and substituting program categories, the legislature ctually being denied the kind of information they need ontrol policy; one goal of such a change in budget ation. As Richard Fenno writes in this regard: One long-standing executive branch criticism of the Appropriations Committee holds that they spend far too much time inquiring into and acting upon the specifics and details of agency activity, thereby impeding the implementation of programs. The ex— ecutive view has been that the Committee should make only the broadest kind of program judgments, and then allow the executive flexibility and discretion in matters of administrative detail. Executives have felt that a program budget might force Appropriations Committee inquiry into the realm of broad program judgments and steer them away from the detailed in- vestigation of the items highlighted by an object- classification budget. The truth of the matter. from the Committee's standpoint. is that the only way they can effectively find out what is going on in the executive branch is through detailed and specific intervention (or the threat thereof) in the process of administration. It is precisely the Committee's capacity to affect the implementation of programs in so many formal and informal ways that keeps the ex- ecutive continuously attentive to legislative wishes.38 38. Richard Fenno, "Impact of PPBS on Congressional priations Process," in R. Chartrand, K. Janda, and go, Information Support, Program Budgeting and the ess (New York: Spartan Books, 1968), pp. 183-I8ET The System is described e2 :2 particular part :eztazions. Each '1: 111110 role intera< start off the proce: 1::‘ expend expansio: \ 1 1111 I. .- iii” the pariic 150 The System of Municipal Budget Roles As described earlier, each role is looked upon as only particular part in the system of complementary ex- cations. Each is a component of the "division of labor" each role interacts with each other. The departments rt off the process and provide the impetus to program expend expansion. The executive pulls together the parate parts into a unified whole as he prepares a get that is more in line with previous spending than tial department requests. Finally, the legislature rsees the whole process, by reviewing the executive's get in the course of putting the stamp of approval upon authoritative appropriation ordinance. Each role makes se, in relation to each other, as the departments initiate, executive vetoes and the legislature modifies spending .cy decisions. As Aaron Wildavsky writes of such a em on the national level: The roles fit in with one another and set up a stable pattern of mutual expectations, which do a great deal to reduce the burden of calculations for the participants. The agencies need not consider in great detail how their requests will affect the President's over-all programs; they know that such criteria will be introduced in the Budget Bureau. The Appropriations Committees and the Budget Bureau know that the agencies are likely to put forth all the programs for which there is prospect of support and can concentrate on fitting them into the President's program or on paring them down. The Senate Committee Operates on the assumption that if important items are left out through House action the agency will carry an appeal. If the agencies suddenly reversed roles and sold themselves short, the entire pattern of mutual expectations might be upset, leaving the Ll- . o a. nor-TC Jar-131.12...» n u A.“ nan-.112 7"7 “’- v “yaw h- “J ’ z \ l u n. nnnI. A~-~fi‘0fl" i.' n \ u / , . vat-.nv-fic-I' . (_3 o« v u o -, ’l“ ~F-A H a a l 4- .. nu icn»- n v n q: «5 . t ‘A- I!!‘\.!n -ECP1 '3, u .u : u" ~01v-.-v punt-AI- c . Vinnfiaog -."A “flat: — :— "-.....ov vegv b- th.. . u. .1 '0 - . ‘:,:“r3 a. at: '\ r. "Iv ~ag~4.v V- duty-3 V'x..~‘ A‘,‘A‘-A s -‘ a,“ "( “fl A “Q ‘. Husky. "fl.- .‘L- . K . . '::]‘Ff‘v\- "v. - 'Vt-.u.-..: 1‘ :..‘:: ' . o AT“ :w‘fi- g . .. - 'II‘.‘.~ . v »._-,u‘~_. '\ V? ‘ A N« a: ‘1. ‘\-¢\\,’ 2“,; L ““M 3‘: ‘\- \I‘ \‘CL‘ “5‘ ‘b ,1 ““h :f‘ I :V Tale A . ~X€Cut H a. ’\h . A “Ye“ 1~~ V“ S~snt r0 4 t t‘ 151 participants without firm anchor in a sea of complexity.39 But does such a logical system of mutual budget roles fist empirically on the municipal level? Evidence, at least 1the data gathered in the present fourteen cities, and in m specific fashion each role was measured does not cor- morate the presence of such a system. By interrelating e measure of each actor's role such a rational set of tual expectations does not appear to exist. Beginning with the interaction between the department's ending orientation and the accompanying economizer role ‘the executive, only one of the four correlations are of fficient strength to suggest the existence of such a stem. This is between the departments own assessment of e padding of their requests and the executive's perceived ed to balance the budget (0.3940 + 0.10). Most sig— ficantly, the relationship between departmental padding iexecutive evaluation of such padding does not at all LS'l7(0.075LP). The two relationships between departmental rocacy and executive balancing (0.0504) and departmental mcacy and executive evaluation of padding (-0.2651) also not go together as expected. Similarly the connection ween the executive's economizer role and the adherence an oversight role by the legislature does not substan- te the presence of a system of mutual budget roles. The 39. Aaron Wildavsky, Budgetary Process, p. 161. .- 1 mung: of 1.16 _' “nib"- U 0 ‘. "‘- ‘i' R A cf‘o‘.‘ - li-‘J‘ll 21‘ :X'V Bd- . o "‘1‘ .33 flfl".fi'fi-—I: i--F "— I rd” h-.H ‘vv.. bu.- ‘V-U-‘h . I u l . Q . "A‘QQQPO “fifih‘n: u s , 1 . -'-05.Nuo [Hafiuacoj’ -- . p‘ "3 3 “CD-7‘02 .uv AVNV..VV v. I . “tC-o .. . x . h v.“ a n1 ," 3" pm” . ‘ va\--‘..u . nhv~ a... II ‘ ‘ . ‘ n: .. -~ ~ 1“": my: "‘-‘.4u-!a .V V (II 'uv.. , «a; : u. --x. V‘ :Yhap~\‘qn ' '--'~'v“u-\.u. \ “~3~..1... v '. .u, ' "'- “k: :“A-”CY‘ Hv “" “'V‘----- 21;. . . . 1: '2" a..- g A .fi‘. , --..t “H. __ . a ‘ “u“ k. \A H... ‘ I A Q:I n‘f \ ‘Nc\ i h A -‘ n “I ' ‘ ‘ N .‘;:;Y‘ Q‘lo ‘\ - \k'V.A\e “ \\ ! T‘::'u 152 rrelation of the legislature's role and executive balancing .0405) and executive evaluation of departmental padding .2509) are quite low. Finally, there is no connection tween the departments' spending role and legislative 'ersight (padding, -O.1083; advocacy, 0.1868). The absence of such a system of mutual roles, perhaps ovides an explanation of the failure of executive and gislative roles to account for the size of annual in- ease in expenditures are either recommended or approved. ese two actors do not stand in a contrasting position th one another and toward the departments, for as much they cut the budget they receive for review, they also vor and support the expansion of municipal spending vels. Looking at the executive and the legislature lely in terms of their negative position is only one ie to their spending choices and alone does not com— etely explain the decisions made. The firse eler L'fl‘lenee derives i p-zssesses in the se ie;11t1e:1theads S' .:;0ests. The fir its forms the has The; are the base 101:. the budget i 1.10001 on it. " E pter Five: Influence Exercise Through Formal Authority Introduction The first element of the structure of budget making luence derives from the formal authority each actor sesses in the sequence of decision making stages. The artment heads start off the process by formulating budget uests. The first set of spending figures are theirs, and 5 forms the basis for subsequent executive decisions: ey are the base of the budget making pyramid . . . From m. the budget is first shaped. They are the first in- ence on it." Even though the final determination of ir budgets is outside of their jurisdiction and in order successfully achieve their spending goals, department is have to influence decision makers who have formal nority over them, they possess considerable influence 0 final expenditure outputs. This is made strikingly er in the previous statistical analysis which shows that size of initial requests is the single most important erminant of all other budget decisions: "Preparing the mtments budget is the key step in the whole budget .ng process. You set the mold for what comes after." 'Set the framework for the decisions that follow. So in ry real sense the departments. as initiators possess 153 I . $ .._,~-H:r;".i$ 1’00rLE uvgzonvo ‘U.' -n" I ‘ . -l ‘ 03“,— 3",“ ‘Y- Lho Cl 1:“. ' nan-L" ..... 4.» ......-~ h .luhlbul -l ‘ A Av DT'.‘ T -5: Clev49-\e 9::0 ~.—: *Al‘c.’ new 'J e...- .4... -v v.5»- . v . a. ' ' '1: -: r3"? C'n'e'q ~Mvu| n» '- V ' -V 5.; ..... v... I‘ - . q C \l . n A. ‘1‘” " -'-:v..~-u-\r -U- not» . '\a ‘ . U ~‘F " “Dan‘y‘cvw or ""“ AI; _ Va 4-- V b h o . h... e a - 3...“... - ' 'II' V§A..\I;l me n. .. k..‘v U'» ‘ . --~.... : .-1 canch" - HVV.“ ‘... . fl . ....;;"~‘ :m‘ 1 ~.. -u »‘ ‘fl *1.“ i l uflfi““':: y" "Vvv. - ‘A--‘N \- :~"‘ ' _ .. n. ...__A_,- “life ans V \ sill k. .h‘ . '1‘ s ~..‘l:_.‘: TS fiaysg-YL " VV1.\ >- 4 ‘If‘u _.‘ n A“ a....;:_ O?‘ , ~ ‘ w‘ 1“ 15 , x “‘:l\'.’ ANN .‘ fi uttu. ‘- 0: s..V\,. ' u ~A ‘ u ~ ‘ ‘ .W Q a . “ LC: *9 . 3 . T. . ‘ ~ .316 31 ~ ‘ 0h - \ IV. ~_;e "A “it The “0" my C.:? «O. ‘ ‘ \ \1 .i "‘2‘ 0" ‘ N...“ ‘L ‘ 0'- I ‘ ' lS , a. 1 ‘}‘\:“Qy‘q q H. 1 '\ \~\\,“Vt TE .‘ ‘ -\ \. ~ \ . K \::1h‘ly\ ‘ u \‘;Ueq 0‘. ‘y ' 1 \ L ”c N“ ‘13“ \u I ‘l " \N “in: 78 W}. \ \ a, i A .‘ ‘! uh“ ,_I .I, 13‘ ‘4 (1'4 (‘f [4’ ’40U1{ o :1) I ‘ 154 nmiderable importance in the process of municipal bud- 3ting and in the determination of final expenditure Jtputs. The executive then prepares the budget, as a document rmt puts together expenditures and revenues in a single lace. He reviews department requests and makes recommenda- ions to the legislature: "As the chief executive he is esponsible for the whole operation . . . The charter learly requires the manager to prepare and present to the ity council the budget . . . As the chief administrator I ssemble all department requests and develop a total budget. put a unity and a particular thrust into it." According to previous reports of municipal budgeting, rm executive stage is the most important. Budget making filuence is centralized in the office of the executive, be 2a manager or a mayor. As Arnold Meltsner and Aaron leavsky describe the manager of Oakland: He is the key figure in making most of the decisions . . . the city manager reviews all the budgets and, for the most part, makes the decisions. He guides the city council in its consideration. He feels that it is his budget. And he uses it to make his influence felt throughout city government.l idescribed by those presently interviewed: "He is the prime ver. This is where the manager makes his administrative 1. Arnold J. Meltsner and Aaron Wildavsky, "Leave ty Budgeting Alone: A Survey Case Study and Recommendation d Reform," in John P. Crecine. Governmental Problem hflfl£;_A_Computer Simulation of Municipal Budgeting hicago: Rand McNally, 19697, p. 344. 6'! . u . Q- 90 ":"’"‘-= 753' 9:.-9 .".'-fi‘.- hov ,. . V ” o’r'fiauh A. or: I O; . a ‘ ' v ......s. J. "“' k I . . . I, .Ipa- rFFHfiy‘: ‘- r ' I r. ,- . ... 3..» navy-u... -- - . l . ‘ “n F v-Ar-u :c.. ”fl 9 .222 |:.' ”v! -J-v ’4 \- o : a .00.».o- an . _‘ nu.'.--vd 'c 'na'wyr lo yv \ '34.. a ‘- ‘ I. O -‘ w. Alr‘h‘ up- -fifl Y! . . .- ~. ....\-....-~ wco‘v - - v ... Q ‘ O Q 3 . "2 H 3H" «Orv Q “‘1‘..- ;V_-VI‘ ”.Vv-~ A H . ‘u A «15-. ,‘ fi . .NV‘.‘ 2" ""“"“ D-J..~ «.n‘y .. . A .. ‘1‘ .2 «9 .'..n_ ‘1‘, .. ... H H “‘ ""- "-‘s n‘“-q‘ :‘L‘I‘ A“ fifiw‘n- '.“‘—~.. ‘v .. , ‘ VV..~-\‘ ’ -..~ SIM-Wu»; v. Vvu._, ‘ .». n,‘ . ‘ . ~ . \ rs- \ H.-. .n- I‘D-“fix ‘ "' -~—--. 5 . .lI!“~‘ _ A . '.u“'.: “2 ‘a‘flnv‘a ‘us “Bk.” ‘4. ‘- . ‘.. h 5‘“ ' I ha uv V»..:-’ at II“: N Ins... V "H ‘ ah‘vx "V - ‘f'x \‘s\.\'; »: ajfi‘. . . I“ a .‘ 1 “"VI. \‘ “Y'.\.- ' \e g“ a “ Ru “‘3‘- u a n.‘ “ l \ I "M" Il‘M“Q~ 4“ ‘ 'Vb’ PO 5“» h -‘ n . n. he w~hgfi .A“‘\'~b ‘«~ .,~““‘ ‘ SKQC‘CT " ( Ca 5 . ‘ ‘le."h:'\ I '\ «“~‘\ll: » ‘ , [NV-5“ I ‘ . « ~Q‘.‘ 1.; k:"\ ‘k h\\\ .1.“ ”3"“ .‘ ”Qt/'0“ ~. . .5 Lo Q ~\.« \A. ~ ‘1 N ~y‘0re ‘~ ; thS ‘ U “I K‘s 2‘“; : "HHS, 1‘“ ‘\g\’ ‘Vfiq L 'Q\_' ‘4 . "“‘h I '. '5‘ '1‘.“.] Vg‘oh h, ‘.;h ”‘3 u l‘x.DQY-+ ' ‘Wl “Re "L h u e 4’ 11 “mm A len SC 155 lecisions most effectively and sets the stage for the >rograms of the year. Where he intervenes most directly .nvflmt happens in the rest of the year . . . The executive La821very key role in making budget decisions. His cfifivities go beyond the mere administration of the budget In cmmfijing the recommendations of others. He makes chml policy decisions in the budget. His perceptions of lmaallocations among departments, of what the tax rate gflfl:to be, of whether a particular function is worthy of qmnsion or contraction are all the mayor's." The executive is the fulcrum of the set of relation- nmm that define the structure of budgetary influence. He mmrols the department heads on one side the legislature [the other, as well as the interaction directly between e departments and the legislature. This institutionali- ifion of budget authority in the executive was, as Allen fuck writes, the intention of this reform movement: In the minds of its prime movers, budget reform meant executive budgeting. The two were in- separable. Thus Cleveland's dictum: "To be a budget it must be prepared and submitted by a responsible executive.2 Before this time, the executive was weak. Administra- Ie authority was dispersed, and the legislature in 'uunction with the departments were correspondingly strong. uldepartment head, independent of the executive, submitted .2. Allen Schick, Budget Innovation in the States shington, D. C.: The Brookings Institute, 1971),*67 17. J. itsovn separate bu ism”: authority t flament expendi‘. ::availabl° revem' :fhetoimpose s1 3:221 mi purchasi tziequacies of mi axhfityin the e: . .I Ava § .J,‘ A y w .he legis 156 own separate budget to the legislature. There was no inal authority to compile and consolidate individual ;mrtment expenditures into a whole that was related available:revenues. Nor was there any central finance flee to impose standard accounting controls over per- mel and purchasing expenses. Dissatisfaction with the dequacies of this system led to the centralization of lundty in the executive who was given the formal respon- ility to revise department requests and make recommenda- us to the legislature. The executive assumed the power integrate the separate department budgets into a com- rmnsive and coordinated whole. This system of formal mndty places the executive in the most influential posi- rlwithin the municipal budget making system. Finally, the legislature completes the sequence as Valone possess the authority to adopt the appropriation Lnance, without which no money could be legally spent. rpossess the right to act: "We have the final say . . . final voice . . . the ultimate determination and de- .on." The legislature has the opportunity to make their assessment of the rate of annual grOWth in total spending 15 and this they do by changing the budget submitted kwm by the executive. In this way. they influence the Cipal budget process and the determination of GXpendi- outputs. Consequently three pairs of formal influence tionships are identified. Between the department heads I" w I n.‘ . l . In. n ~,A»r ? n9 n “3.. we .5272“ w \— \Je aux-1 1. 33 eygv-Hr 6n... 1 u u- I I 5‘ .vnw.,.“~ -1, If '- UIOA-unopannv- - -. v . H ‘v- u. .v~; "‘M‘vuyv "‘~ A I . \ -«-.: 1.: “'“f‘mm “"EVVVv-a “a. ‘ M. _ . ""‘ov’ ‘P 27‘ "“"."Y. - M" ate” “‘ R n.._:‘ : Cu-n. ‘ - \r ‘ L I ‘ l ‘ “a- “a n.‘ . NVL u—“M‘ - . “is“; l. k . x... I. _ ‘ ‘4 3:. HI \ ‘ 5“." 2": ""“\-\. . . \ «fiféh mx", . no“ ‘_ ‘ "“11. M" . u“ 2‘ ‘.‘V!, “A UK, 3341‘ 71"" 1‘ \A as 55““1 . ’M ‘ ‘ vV..buLTeQ " 0 ‘\ u, I w ‘\:"‘ ‘5‘ . may L”: Q} \‘ZQ \u II“| I \‘I "‘.‘:C ‘3“ in bile Q18” e V'. ‘.., g t . y 1 ‘ ‘tl‘Dleme “ .i, , ‘1 \sl‘tfi‘t 157 | flthe executive, the executive and the legislature, and altmtween the department heads and the legislature. :h will be examined in turn. Emamining first the pattern of formal influence as atmdget moves from the first to the second decision gas, an executive centered process is where departments :only passive onlookers to the decisions made on their tial requests. Executive choices are "closed" to the luence attempts of the departments as unilateral budget 8 are imposed. The executive exercises hierarchical trol over departments and does not consult with them the course of reviewing their budget: "The manager ered a cut in the department, within a specific area. did not permit me to select the area where I thought ut best made . . . We had no discussions with the ager. When I first saw my budget, after completing forms, the budget had been cut in places I had not ?oved,:nor would I have chosen those areas if I had iconsulted." It is not that departments are unable Limit the size of cuts made, but that they are even ed the discretion within their own department of e to implement a certain percentage or dollar value ction. Other than the submission of their initial nditure requests, department heads have little tional effect upon executive decisions. On the other hand, a decentralized system is where utive decisions are "open" to department pressure, and tie interaction bet training. negotia iescribes such an i It is at this (budget game) departments h: ministrator a] balanced budg: suasive . , . contrary, the presentations 536 retained tract is close a; tensive. Departme :a=e' to an execut teatime figure exe:utive's own re iiaproduct of mu ecutive action alc 158 3 interaction between these two actors is marked by r‘gaining, negotiation and compromise. As Donald Borut nuibes such an influence structure: It is at this Stage that the advocate process (budget game) is presumed to operate. The departments heads plead their case and the ad- nfinistrator argues for reduction on behalf of the balanced budget. If a department head is per- suasive . . . cuts may be restored. If, on the contrary, the back up data and the departments presentations are weak . . . the reduced figures are retained . . .3 tact is close and continuous, and consultation is ex- sive. .Departments have the opportunity to "plead their e" to an executive who is susceptible to such pressure. enditure figures are tentative when they meet, as the cutive's own recommendations to the legislature emerges a.product of mutual accommodation, rather than of ex- tive action alone: ". . . Together the manager and I ermine places for a cut in my budget . . . He would -- 'If I out here, what would that do?’ I'd say, vt take it here, take it somewhere else.'" In these mtions, the second stage of executive budgeting is her isolated nor insulated from the involvement and uence of the departments. Department-Executive Influence Relationship The first question formulated to measure the formal uence relationship between the departments and the *— F— 33 Donald Borut, "Implementing PPBS: A Practitioner's paint." in John Crecine, Financing the Metropolis aiesttive was the. farthest heads an exe:stive's budget question was coded 0 1- bu) N I U\ Itemderlying con 232.: of bargaini these two actors. azzherity over the sore that corresp here to a high 1‘3 '13“: ‘ “ Qeiiaz‘tments 159 :eoutive was the, "Purpose of the meeting between the qmrtment heads and the executive in determining the :ecutive's budget recommendations." This open-ended Lestion was coded into the following ranked categories: - No meeting takes place Inform departments heads of previous decisions Support the executive's budget Exchange of information Persuasion Place where actual decisions made \fi-PMJN HO letumerlying conceptual dimension of the scale is the imnnzof bargaining and negotiation that takes place among mse two actors. The executive asserts his hierarchical rUundty over the departments. It extends from a low Hue that corresponds to an executive centered system nae to a high rank that represents a decentralized process ere departments are able to influence executive spending toices. The first two categories point to the highest degree executive dominance, where department heads have little iJity to change.executive behavior. Where department awe are not even provided with the formal opportunity to et with the executive concerning their spending requests ey certainly possess little influence. But simply meeting th.the executive does not represent an increase of in— ence, when such contact simply informs the departments 3. (continued) (Beverly Hills: Sage Publishing “196mm 1970). pp. 298-299. 2:" trevious executi ation that has bee :5 at executive cer A series of e: sessions are I submitted to ' sayor's budge EaW's decis very little m tfteetimes th department he to :ram': thei The middle ts executive control. :heieparments fc Tris provides the it influence thr' ‘3 ”emits use ‘s o MM "4L1 ng mOre detz 160 gmevious executive choices. It is exactly this sit- ion that has been described by John Crecine as a feature niexecutive centered process: A.series of executive budget hearings or review sessions are held before the mayor's budget is submitted to the council . . . By the time the nmyor's budget reaches this stage, most of the mayor's decisions have been made and there is very little money left, if any, left to allocate. Oftentimes these sessions are used to explain to department heads just why no funds are available to grant their particular request.4 The middle two categories point to a relaxation of :utive control, as he now displays a reliance upon departments for information he does not possess. 5 provides the departments with potential resources influence through judicious selection and presentation. executive uses this information for this purpose and ould also be used to support his subsequent dealings the legislature: "It gave the manager the necessary 3 and information so that he could justify certain ls of expenditure to the council. So hecould then 1' it to the legislature." The purpose of preparing own recommendations is, "to clarify a request by pre- ing more detailed and comprehensive response to a icular question.“ In either case, departments are rdinate to the needs of the executive. They do act nomously, on behalf of their own goals of increased 4. John Crecine, Financing the Metropolis, p. 37. m and display 1 gem-3e his rent The last mo < lie executive, now the departments '2'. gave the the ob I: is the opportun asters, at the tilt! infixed a..d eve h;np:‘ ‘ .. we.» tree the (11) ..... 161 is and display limited pressure upon the executive :hange his:review of their requests. The last two categories represent a decentralized :em of executive influence, where persuasion occurs. executive, now, does not impose unilateral decisions we departments who are allowed to "plead their case." gave me the chance to campaign for my requests . . . .s the opportunity to fight for their budgets." There alimmediate and direct consultation between these two ms, at the time of this meeting. The executive budget ufiixed and everything is up for grabs, as the budget rges from the give and take of the decentralized budget tem. ' The response to this question concerning the purpose the meeting between the departments and the executive Leates an open influence system. From.the perspective flm three departments interviewed it is midpoint between fourth category of an "exchange of information" and the ticategory of "persuasion" (see table 5.1). This cor- onds, along the five point scale, to the position of een some and much influence by the departments over the sions of the executive during their formal budget ing. From the executive's general description of all tments, this meeting serves the purpose of a forum n exchange of information. This points to a more tive centered process, where departments possess some E wit/rm TAU/.mrmw .50 0,6 34k 2 CC .............. - .IHV.fi/~02.u\fl/~3223 2 C t h C Q I Q h C Q L U; P . I 9 Why fil. "1n S .\In ~\v 0 .l 32 A122 0 add \u D.« a .Tv - . g c n "91.0 h Ave 056603,.00030000 0 “NIH“-.. M 1m.» 021/003/0.>039~000 <4 1 n. ”.0 a... ............. - - N“... 43.. H»- 3 M». 353.20 2 (J 3.“. «<1. «Jan». 3 1 any 9T.“ 0 “41- s.!“ v e a “U n .3... a: N t 1 l- l ‘J he. s 162 TABLE 5.1 DEPARTMENT-EXECUTIVE FORMAL INFLUENCE RELATIONSHIP (DEPARTMENT HEAD REPLIES) Purpose Resolve of Disagree— ggiigzg§e Mean Meeting ments 4.00-3.20 3.00 2.00—1.60 2.90 3025‘2o60 0.66 2000‘1060 1.62 4.16-3.33 0.33 1.66-1.33 1.66 3.66-2093 0.50 3.33-2066 2003 5000-4300 0050 1033“1006 1985 3.33-2.66 2.66 1.20-1.00 2.11 0.66-0.53 2.00 3.00-2.40 1.6a 2.50-2.00 0.50 1.00-0.80 1.10 5.00-0.00 2.00 1.66-1.33 2.44 3.33-2.66 2.03 1.75-1.00 2.03 4.20-3.34 0.00 0.66-0.53 1.29 3.00-2.#0 0.00 1.00-0.80 1.07 3.00-2.40 0.50 1.66-1.33 1.41 4.00-3.20 2.20 1.50-1.20 2.13 3.51-2.80 1.21 1.70-1.36 1.85 Dev. 1.10-0.88 1.40 0.73-0.60 0.75 =::e influence (see 1' either case, how azis of the scale, 3125315029 change , . .-. second cues inal influence i set the budget ar reference to cm 1 N “"193! he way C ."I 3 into: of each iv; '1‘” ‘ ‘ I m is the we; 163 e influence (see table 5.2). Individual city responses, anther case, however, are generally within the upper scfi'the scale, where the departments are able to nfizsome change in executive behavior. A.second question measuring the distribution of an.influence is the process by which disagreements rthe budget are resolved. Although there is little :rence to this feature of the budget process in past Hes, the way conflict is resolved is an important eator of each actor's influence. Who is the stronger, who is the weaker becomes apparent in a situation e spending preferences are in direct and overt conflict. riparticipant wins and which one loses is often deter- d by the application of influence to settle the issue in ention. By being able to change the behavior of others, 8 own interests and values are satisfied. Responses to this open ended question was similarly along the dimension of executive centralization of ence, from a hierarchical to a bargaining and nego- 'on situation. These codes are: 0 - Executive made the decision 1 ~ Exchange of information 2 - Compromise 3 - Deadlock 4 - Legislature made the decision Where the executive asserts his hierarchical authority the departments and hands down unilateral decisions certainly is the highest degree of executive influence 2. Ca at . .- , . . .\ finv 0 0V niv New fitk A4 a ,C a a .E . . . PK . . v1 . . . «4 My. MW ,0 0 n.» .H/zhc .Kafilaznv «sum; Rare 7. .HJ A); luv m..... ....V \HW. I... - - 0 - - Iv - - - - . 0 | I. .' U .4. . "v u \m. . . as) «a J .5». as) as .4 1x a fit/ cl 3. a 9; 3C 31/ «\J 2 31, 0 u H IA. .., .. ”L. 3. . lam :u 3 - I V . a 1M.“ . N v Mn #1. c n 1 q s N Al I V t A v A v u \ Q A ~ \~ \ Inn.» \ t I l I I Q 1 I \ § \ t \ V \ s . \\\ 164 TABLE 5.2 DEPARTMENT-EXECUTIVE FORMAL INFLUENCE RELATIONSHIP (EXECUTIVE REPLIES) Purpose of Resolve M Meeting Disagree- lean ments 3060-2088 1000 1094 3.00-2.40 0.50 1.45 4.00-3.20 0.75 1.48 3.50-2.80 1.60 2.20 3.60-2.88 1.25 2.06 3.50-2.80 1.63 2.22 3.33-2.66 1.00 1.83 2.66-2.13 0.33 1.23 3.33-2.66 1.60 2.13 2.25-1.80 0.60 1.20 2035‘”1088 0.71 1.29 3.66-2.93 1.33 1.13 3.71-2.97 0.00 1.49 2.50-2.00 1.20 1.60 3.22-2.57 0.97 1.66 Dev. 0.40-0.45 1.19 0.39 i-.V-a'U OM bar's" neve 0 VIA. . l . ¢ p n IQ, a can‘- I'I'g c e an“ vn'vu4~-'v new "“ 'i :01"? be“: 4' «4. n a‘vddavwvv L. n I u 131-! “'3‘ mp, nHT '- v ”v ¥4¢J-VV VJ w El) (1) 1’ (D ()' C.) In TC": Al's-hr“; ..v_\., VU..&,-\ lug. ‘ . _\4.'7‘..c":v-A n "-“"~.\n.b: 1 Y! ._ . a ‘i 5' Dre‘rebfi‘t 'L ‘ v. V ‘ K h,‘ ". ~Y‘Nh \ u. \. ., .\‘ DSIOI‘Q T \- ..‘ ~ .\ K “‘ V311,“ Q .1 a; thehd ' .. 9C :1.“ \.\~\;encc . Chg. Fec- “t) “We" J...t,:l' “he\ \ r .- . '11 ‘| ‘3’: 4958 Q V“ SaVe 4 x“ I <19 ;- 165 tments have practically little to say in ing the issue in contention and virtually no impact executive spending choices. They have little choice 0 acquiesce to executive spending preferences: "I no choice but to accept the decisions of the manager He made the decision. I had nothing to do with it He's the boss. He has the final say." Where compromise is the process by which budget reements are resolved, there is a partnership among two decision makers. The specific outcome is mutually :able to both of them: "We came to a meeting of the . . We try to reach a consensus . . . There are con- >ns on both sides." The distribution of influence e balanced, as persuasion, bargaining, and nego- m mark the pattern of interactions between the ment heads and the executive. Finally, where the legislature made the final de- s, the executive possesses the least amount of influence over the department heads. By not being 0 prevent the departments from beinging the con- Sy before the city council, the scope of conflict an extended to a third actor and the equation of ice changed: "The problem was presented to the city L. They made the final decision . . . The depart- eads gave their position to the council and they w determination.“ By not being able to contain the controversy vi the disageeuent '0 tea-is, the executi set its administr cipal budget proce The response '1‘": “ “ s «it. .he seca ‘: 1MP ‘ .:cv¢'\ver degree ( ‘61 h; DYNA ". JVESS Dy wh . . ‘ ."T 2;] l"; “. -O'h’ the de h i." 166 ontroversy within the second budget stage and ending isagreement by dealing directly with the department , the executive has lost almost complete control his administrative subordinates and over the muni- budget process. The response to this measure of the relationship en the department heads and the executive points to ater degree of executive influence over the process igeting than did the previous question. From both erspective of the department heads and the executives rocess by which budget disagreements are resolved allow the departments to present information. This esecond of the five categories and corresponds to apartments having slight influence over the ex- e when concrete budget conflict is evident between The exchange between them is fluid, during their Lg, but when the discussion becomes heated and the sticky, the executive quickly asserts his authority he departments to end the controversy, by uni- lly announcing a settlement. he third measure of this particular influence re- ship is in the "strategies pursued by the departments er to successfully obtain what was requested." The on here is the mechanism departments employ to ex- their influence. The specific ways they attempt age the behavior of the executive: the tactics and :alcilation depart higet are analyze to active and they explicitly a: actives review < iecenzralized budé behavior are foll . . Hum ..::.: have pI‘eS‘ Ii. Strategzes succes :g'so “r u. the EXEC'JII lines? there is :‘anfi ~‘y'u ; . o we; Choices. ~ . a ~uV‘ .Iective, t :v'v‘. "'llence. Reslosses ‘tc h ' a ”:51 N v‘ “ on a low exeCutive dom: 167 ulation department heads explicitly and consciously 0y to effect a positive executive review of their et are analyzed. An active and aggressive departmental strategy, where explicitly and overtly attempt to influence the ex- ive's review of their initial requests points to a ntralized budget making structure. Such modes of vior are followed, as they are effective. Department 5 have presumably learned from past years that these tegies successfully change executive behavior be- e the executive is open to such pressure. On the ? hand, passive departments, who do little have ied there is little they can do to effect executive ling choices. Any particular departmental strategy wffective, because the executive is immune from such lence. t Responses to this open-ended question were again .from a low to a high rank that indicates variation ecutive dominance. These codes are: - No strategy followed — Exchange of information - Prepare a budget acceptable to others Establish trust and confidence - Emphasize external considerations - Persuasion \n-PWNl-‘O I The first two responses point to essentially powerless tment heads who pursue no strategy to pressure the tive. The next two categories also point to passive tment heads, except that it indicates at least awareness of th ties: success. 3 EECI' ive behavior thezselves unable ‘21"‘8. leoartme ltzle to offset ' rcefforts, excej approval of the e: ever, displays th ;:in:s to a deceit 5:20: alter thei =I.icitly attem; If then by their 553 If requests, .5. l.("‘ The reSPORSt ”SWIG center: its :a “O: the exel :. successfullV :Qlained’ a Pas .~ ny slight i \ e at i :Quthels “OI e fihorltl anew; wept 15 measx 168 wareness of the importance of the executive for their at success. But the only way it is seen to effect utive behavior is to alter their own decisions in aring their requests in the first place. They perceive selves unable to change the executive's review of their ets. Department heads perceive that they can do le to offset the opposition of the executive by their fforts, except to act in a way that would meet the val of the executive. The last two categories, how- , displays the active and aggressive strategy that ts to a decentralized budget system. The departments at alter their own budget behavior but overtly and icitly attempt to alter the executive's evaluation men by their own tactical maneuvers. They make the same 3f requests, but pressure the executive to approve The response to this particular question points to an ntive centered process. Department heads (it was not [of the executive)5 describe the strategy they pursue ccessfully attain their goals as preparing a budget is acceptable to others. This represents, as just ined. a passive department strategy and corresponds 1y Slight influence with the executive. Departments -—.._____ 5. The third question of the strategy employed fluence others, was not asked from the perspective of ecutive's interactions with the department heads. es not employ a strategy as much as he asserts his ity and thus this question is not relevant and the t is measured by only two question.‘ heels are passive ':;:t'e executive c ric-nsciously at believing that suc 2;. be aggressive :fficia‘ aeeting, aver. to then: by The index of 1‘4”. ..... inert heads I: ieortnents do :Ych's‘: ‘n r: as n . "D ‘ i I Nliu\~'v Mk)“: V .. a ‘ :ntellv open to 1 :5 their requests soze opportunity executive as soon between t‘ hes: in «a serial autho 169 ads are passive participants to the decisions made the executive over their requests. They do not actively d consciously attempt to change executive spending choices, lieving that such attempts would be ineffective. They y be aggressive and influential at the time of their ficial meeting, but this is an opportunity that is ven to them by the executive himself. They do not go t to create a new situation, to exert such influence. The index of the questions reveals that both the partment heads and the executive attribute some influence departments during the second stage of budget review. ecutive budgeting is neither completely closed, nor tally open to pressure exerted by departments in support their requests for more funds. Departments possess ne opportunity to change the spending choices of the ecutive as some bargaining and negotiation takes place :ween them in the course of the executive's exercise of : formal authority to review their requests. Although the distribution of responses among the :ies extends from where departments possess from slight great influence, the pattern of infividual city scores skewed to the lower end of the scale. There is only ingle city where great influence is attributed to them. while on the average, departments in each city possess a influence over the expenditure outputs of the ex- :ive, the executive is the stronger partner. ;=.Z: ms possess :eguests increase 353. where depa' “[1211 the executiv greet tactical aincrease is re Ements create iiaset of comp: 23359 and take of 231536 is only We; 53:10“ 03" initie 33129113 TO iI '118 StitiS‘ iriluence SyStem 5?“: “538mm,, ,, e (1‘ :mve centered r. oeing made 170 Relationship to Department Budget Requests It is first expected that the more influence de- nents possess vis-a-vis the executive, the more their ests increase over the spending level of the previous . Where departments have the opportunity to bargain the executive, the size of their initial budget is an im- ant tactical maneuver in the negotiations. The more of ncrease is requested, the more slack resources the de- nents create for themselves that can be given away, set of compromise spending figures emerges from the and take of a decentralized influence system. However, a is only weak support to this linkage between the form- ion of initial budget requests and the ability of etments to influence the executive (0.1615). This statistical relationship shows that a decentralized ence system does not propel the departments to submit sionary requests. It also demonstrates that an ex- ve centered process does not inhibit such requests being made. The pattern of formal influence budget w of the second decision making stage does not extend ard to the department's formulation of their initial sts. The executive's authority to review department ts and prepare his own set of recommendations does ontrol the original expansionary thrust 0f departmental ting. Relatioz ’ '1 in: .7 ”‘ xore central :uzget eaemg syS' reviewing departs O O Q Lentils-.10 oudg .~ exrent the 535535. a cons: 5?:m1‘:p ‘ Mme L0 corn. ‘ A: ‘ a, ‘Qf‘h‘ “51- iment heg :32: \ “megs of v Shin n ‘ «MOng the c 171 Relationship to Executive Cuts in Department Requests More central to the place of influence within the et making system is the behavior of the executive in ewing departmental requests. In this decision, the gonistic budget position of these two actors directly h and their relative influence is most apparent. It xpected that departments utilize their influence to t the size of cuts the executive makes in their requests. he extent that the executive is susceptible to such sure, a constraint is placed on the budget cutting uts of the second stage. It takes influence for the utive to contain at this point the expansionary thrust apartment heads. A weak executive does not possess ability to impose such reductions as much as does a 1g executive. However, there is no empirical cor- ration for this hypothesis. In fact, a moderate positive stical association is evident (0.2275). There is no ent reason why the executive would make larger re~ ons when he is more open to departmental pressure, t to demonstrate that the executive does not e department requests any less when he does not ate them by means of his hierarchical authority. ing cuts result from the second decision making stage dless of variation in the formal influence relation- mong the departments and the executive. Departments neffective in preventing executive cuts in their requests . _‘ Iguana; *0 ,fi r- l.( _...J.:..vv U V *‘lonartj . . O :wnlor. Jv‘bd- b 2.5! 1 .. I - were 18 no; feisians. It as". are, mere eepa . ‘51.“.' 3 V A 9' . "we 3- ~11: il\- 31’?- “e“mv-n . ." ,. "WV-w “v; : 'x ' . ..- :15 7'1 7‘3 .vvlu-iy, L.l‘,‘- ‘ n ' *' :r‘“"‘vv-‘ mTc VXVVJ~‘\ c (":n . . "s: qIW‘ “1‘“ ~09- .- Nlle. bxu:\, ,\ V 'afi‘ ‘Rbk i\-\ _ ~ I! n ‘\ ‘ ""“It0u;;‘ “ In ‘ “I. “£.‘“"ih~ ~ OHM y‘;‘:‘ b v 'O N. x'WKq“: ‘ I ‘4» ‘\ H, _‘ vvbb.u‘ if} a... a . y‘ u.‘ b.‘::" . ‘5‘: , I 4“\A “§.." Q ~n R _ . ~\. ROG 0‘ ~.. ._ A ”h .‘k 5 i ‘1. S‘Nan'z; , i ‘ ‘ H‘liu‘na the \— .“' ~ ~ \‘;‘ 3'." | A H‘: :qou~h . q ' V n “s S lu; \p.‘ \::,‘|H uge‘lh \\ bend . ‘ , 0. m0 ‘:V:X 0,5 172 1se they do not possess a sufficient amount of zence to contain the negative thrust of executive rior. Departments possess some influence, but some 1ence is not enough influence to change executive sions. It will be recalled that the highest absolute a, where departments possess the most influence with executive in any of the fourteen cities, is only the 'h of the five ranked categories. Possibly, if there reports of greater department influence with the tive, then they would have enough influence to e executive spending choices in the direction of 'own budget objectives. Relationship to Executive Recommendations Although departments fail to prevent the executive reducing their requests, they are not completely un~ ssful in effecting executive decisions. They still ss sufficient influence themselves to achieve their ate goal of annual growth. The more influence de- ents have with the executive, the more of an increase ending the executive recommends, for them to the -ature (0.3703 + 0.10). Departments, apparently inough influence to prevent the executive from 'such extensive and severe cuts in their initial ts that they receive no yearly increase at all, in spending level. As Thomas Anton writes: ". . . f power means inability to persuade the governor to include a proposa- nth-gs cut 831d :ecessrily the S fiance they p053 is; decisions of iiizicns to thei their goal of ex; s:=..-:e. The diet: is executive and is: consequences fxe in: The second : the executive am :zazively adoptee itinnce of the 3:33 Council has .3521"; Slate and l Ethelllinois 16mm its he 1963 Ge 31W dete me legisle pPOpriatl 01' W prom Side . elsewr yr \ 0 ‘ I NBS l n ‘Thomas .liy ‘ lliTIOic “Hols PTQSS \ 173 me a proposal for change within his budget."6 Pre- .ng a cut and ensuring yearly expansion are not sarily the same. Departments utilize whatever in- me they possess, not negatively, to oppose the cut- decisions of the executive; but positively to support ions to their funds. In this way they ensure 'goal of expansion at the end of the second budget a The distribution of formal influence between ‘xecutive and the legislative body most definitely onsequences upon expenditure outputs. Executive-Legislature Influence Relationship The second formal influence relationship is between xecutive and the legislature as the budget is author- vely adopted as the appropriation ordinance. The ance of the executive, over the deliberations of the :ouncil has been widely reported in past studies of state and local budgeting. As Thomas Anton writes 2 Illinois State Legislature: )espite its formal authority over appropriations, :he 1963 General Assembly was virtually powerless .n the determination of state expenditures . . . :he legislature did nothing more than pass ap- ropriation bills . . . legislative behavior could nly produce a stamp of acceptance for decisions ade elsewhere. As an institution the legislature as incapable of doing more.7 . Thomas J. Anton, The Politics of State Expendi~ 3_lllinois (Champagne,,Illinois: University of .3 Press, 1966), pp. 202-203. Thomas J. Anton, Politics of Expenditure, p. 2%6. Solo. Crecine writ sizilar terms: The council, the ma'or's . . . the Ct to approving cipal legisi :‘r't'ne exe: his cone 111' series of Jongr. fie 'oudeet democratic o: the purs 174 n.0recine writes of the municipal.legislature in ilar terms: flflm council, in virtually every case, approves the mayor's budget almost exactly as submitted . . . the council's options are limited largely to approving the mayor's budget . . . the muni- cipal legislative body serves as a "rubber stamp" for the executive operating budget.8 This conclusion of legislative impotence contrasts to dies of Congress which contend that the "Power of the' se" is a crucial component of legislative authority elf. As Jesse Burkhead writes: The budget system developed as an instrument for democratic control over the executive. The power of the purse came to reside in the legislature in order to prevent the executive from imposing will- ful and arbitrary tax payments on his subjects. The budget is an expression of ultimate legisla- tive authority.9 hard Fenno similarly writes in his study of Congress, ’6: . . . when we talk about the appropriation process, we are talking about the most important source of legislative power. More particularly, the power of the purse is the key to the institutional power of the House of Representatives within the American Political System. The single most serious blow to the House as an institution would be to reduce its share of the money power.10 F‘— .— 8. John Crecine, Financing the Metropolis, pp. 35, and 207. 9: Jesse Burkhead, Governmental Budgeting (New York: Wiley and Son, 1956). p. 83- 10. Richard Fenno, "Impact of PPBS on Congressional opriation Process," in R. Chartrand, K. Janda, and M. c Information Support, Program Budgeting and the ress (New York: Spartan Books, 19687, p. 176. :eniitures reside 2722;: eve the £01 :riezior. ordinanr :inzeze decisio: fission that the step' to a domi' ‘l Inn“. 4 tendons :or ‘I :Y‘:hfl1c u y' "; .... Lure I‘ECC A": :T“ '\ ‘§Q\ll’ The leg Rims Ihe Zl‘ce . 175 One does have to travel far from the local Charter find that the legal authority over taxation and ex- ditures reside in the popularly elected legislature. y have the formal responsibility to adopt the appro- ation ordinance. This gives them the, "final say, the imate decision on the budget." Consequently, the con- sion that the local legislature is only a "rubber mp" to.a dominant executive has important normative dications for the functioning of the checks and bal- es of American political institutions. This conclusion rests solely upon the analysis of amount of change the city council makes in executive enditure recommendations. Because they impose "small" erations in spending totals at the time of their for- review of the budget, they are portrayed as unimportant ticipants in the municipal budget process, who only fill, the legal requirement of adopting the appropriation inance. The average absolute change of less than 2. :ent, and the average reduction of less than 3 percent, Ld appear to indicate that the municipal legislature in present fourteen cities also make minimal changes in :utive's recommendations, and are also uninfluential ;he budgetary process. However, this interpretation is not presently accepted. ith the concept of incrementalism, there is no exact dard to separate a small and unimportant change from a large and importer iescription of va sessed by amicip Legislative authc iasis, masks the 2:; expenditure ‘IIES the resul zenied expansion below the level these reductions :ities have :uch :he executiv ionsequem 176 ge and important one which would enable a precise cription of varying amounts of formal influence pos- sed by municipal legislatures. Furthermore, measuring islative authority in the budgetary process on this is, masks the impact these alterations have upon al expenditure outputs. As described earlier, in three ies the result of the third stage is to turn a recom- I ded eXpansion of spending into appropriations that are ow the level of the previous year. While the size of se reductions is still small, the legislature in these ies have much greater significance for the departments the executive than first indicated. Consequently, the amount of formal influence pos~ sed by the city council is measured by the actual tern of interactions with the executive revealed in course of their review of his budget recommendations the adoption of the budget ordinance. It is only er this analysis that a direct connection is made be- en the pattern of influence and the direction and litude of the change in expenditure levels imposed by City council. The means by which the legislature exercise their e1 budget making influence is by activating their 1 authority to review the executive's budget recom- ations and adopt the appropriations ordinance. Their a1 responsibilities gives them the ability to modify the budget submit it not they choos w.“ we ' ' Mice relati 'IM : ' .zportance 0 i,:‘. so. “a?!“ ‘. mice, that“ a; I i" "I: ‘;+‘.' ‘ wag“. Counc‘l 177 a budget submitted to them. The question is whether Iufi:they choose to assert themselves. Previous re- unh.has indicated that the city council simply chose :to activate their potential influence. To the ex- d;that they do so, they are no longer a "rubber stamp" a dominant executive: assuming for themselves an mutant part in the municipal budget process. Analysis The first question asked to executives (see table ) and councilmen (see table 5.4) to measure their luence relationship is the normative evaluation of importance of the budget as a mechanism of legislative luence, that: "The budget is the most effective means city council has of controlling the administrative mob of government." To a great extent these two icipants agree with this statement. Both are aware he significance budgeting holds for the distribution nfluence within the policy making process. The second question measures overall assessment of actual influence the legislature has upon the muni- 1 budget, that: "Council decisions on the budget had mal effect upon actual spending." Neither of these decision makers eXpress more than slight agreement that city council is so powerless. In other words, to eat extent, legislative participation in budgeting erceived as important in determining spending outputs. —,_‘ . q- 4. 'JV;;' TVT’V‘T. 1, HA- NI V O A. (l .1 Ali\ 0 AIM no «(.Jfilao r\.(,HU 0 O O 0 O O 0 “HM 3110 fié/O S..\JF\.JF\IJ0 - - - ' - - - - - - - - AIL 5(3/ fi/~ AJ‘J fives/t 9‘ \‘J 1‘ Al.» fl\.J \IC «(.2 ( I \ \ CWV 6 fi.‘ 178 TABLE 5.3 EXECUTIVE-LEGISLATIVE FORMAL INFLUENCE (EXECUTIVE REPLIES) Budget Resolve Carry . Minimal Decisions n §§§ijg Effect Left Disagree- Out Mean Control ments Anyway 1 3.00 0.00-4.00 0.66-2.66 4.00 4.00 3.53 2.00 1.66-2.33 0.66-2.66 1.00 2.66 2.13 2.00 3.00-1.00 0.50-2.00 2.00 4.00 2.20 3.00 0.00—4.00 0.66—2.66 2.00 4.00 3.13 2.33 1.66-2.33 0.00-0.00 2.00 3.33 2.00 3.33 0.00-4.00 0.00-0.00 3.25 2.00 2.52 3.00 0.00-4.00 0.00-0.00 4.00 2.00 2.60 2.25 2.00-2.00 0.66-2.66 1.00 1.75 1.93 3.66 1.00~3.00 0.33-1.33 0.00 3.33 2.27 1.50 1.00-3.00 1.00-4.00 3.00 3.00 2.90 2.50 0.00-4.00 1.00-4.00 3.50 3.50 3.50 3.50 0.50-3.50 0.00-0.00 3.00 3.50 2.70 2.50 1.33-2.66 0.00—0.00 2.75 4.00 2.38 3.00 0.50-3.50 1.00—n.00 4.00 3.50 3,60 2.68 0.90-3.10 0.45-1.86 2.51 3.18 2.53 0.82 0.93-0.93 0.40-1.61 1.49 0.79 1.01 .10 0: /O ”V .\ a A V nu AU .NJ q.70. . AU a/ .RJ 0 (J /n... he. - - - Q ' ‘ - - - - - 2 A/h fi/~ 9k fl/s W); fix} ull‘ Al» 0 .va : i: R Q \- . A l s h I v V Ah I \ h ‘u 1 s. u ~ ~ s n \ a V ~ - 179 TABLE 5.4 EXECUTIVE-LEGISLATURE FORMAL INFLUENCE (LEGISLATIVE REPLIES) Budget Legis- Minimal Decisions Resolve Carry E lative Effect Left Disagree- Out uean ‘ Control ments Anyway ‘ 3.50 1.00-3.00 O.25—l.00 2.87 2.50 2.57 2.80 0.40-3.60 0.40-1.60 1.00 1.75 2.15 2.00 1.14-2.86 0.84-3.36 3.29 3.00 2.70 3.33 0.33-3.66 0.66-2.66 3.20 2.66 3.11 2.00 1.20-2.80 0.25-1.00 2.26 1.20 1.85 2.50 0.75-3.25 0.84-3.36 3.71 1.55 2.87 2.50 0.66-3.33 0.50-2.00 1.66 1.50 2.20 2.66 2.00-2.00 0.00-0.00 2.25 2.66 1.92 -- 2.40-2.40 0.40-1.60 3.20 2.40 2.40 2.00 1.00-3.00 0.50-2.00 2.66 1.66 2.07 2.75 0.50-3.50 0.50-2.00 2.75 2.25 2,65 3.50 0.50-3.50 0.87-3.48 4.00 2.00 3.30 3.00 0.70-3.25 0.66-2.66 2.29 3.00 3,04 1.50 0.75-3.25 0.57-2.28 2.17 2.75 2.39 2.62 0.84-3.16 0.52-2.08 2.66 2.19 2.51 0.62 0.44-0.47 0.25-1.00 1.97 0.62 1,64 the next qu: tut-get decisions :3 the executive executive ‘ou :ior. upon the lei .:' certain issue TEE'S review, t :rzeohhical due 1‘" l l . : restrict; '._4 Ezore powerful .3 there most be "it City Couhcl 180 The next question asks if there were: "Certain et decisions that the council did not make, but left he executive." This defines the scope of legislative executive budget making responsibilities. A limita- upon the kind and number of decisions made by the slature means that the executive correspondingly hes greater latitude in making expenditure choices. ertain issues are outside the purview of the legisla- 's review, because they are looked upon as administrative echnical questions, then the city council is volun- Ly restricting their own influence and deferring to ‘e powerful executive. An executive centered system ere most budget decisions are made by the executive mt by the legislature or the departments: "Most ions are left to him. He pulls the budget together. his budget-- he has to make it work . . . In effect the thing up to the actual appropriations belong to the er. It's his responsibility to prepare it . . . ity council is less able to make the bulk of the de- ns of budget preparation and planning, which are to the responsibility of a competent manager." An ie 0f just half of the executives and councilmen riewed in each city replied that there were no de- 18. the legislature instead of exercising their own Ltives, left to the executive to make. The next question asked was the: "Process by which iisagreements an ::.oec‘- between t 521912!le cate Legislative stre goeseion provide 2'21 legislature hen these two e the: over spend II ‘‘‘‘‘ 181 egreements are resolved." This was previously ex- wd between the departments and the executive and the >ranked categories that extend from executive to slative strength, are employed. Responses to this tion provide evidence for a considerably more power- legislature than has been described in past research. these two actors are directly in conflict with each r over spending preferences, it is far more likely the legislature will prevail and make the final de- on than the executive. The legislature is now dominant, as executive acquiesces to their spending choices. As was described: "The issue Was resolved by the legis- @ power of the city council. We just told him this mat we wanted to do and we have the power to do it The manager works on a week to week contract and if Sagrees too much we can terminate his contract . . . . s resolved by instructing the city manager to change udget to the council's way of thinking, not his. It direct order by the council." The actual response th of these two decision makers both correspond to ourth of the five categories, where there is deadlock; describes a situation of balanced influence between arising from a combination of the third code of ision and the fifth category of the legislature making acision. The legislature is most certainly not sub- 1te to the executive. They are not reluctant to assert their for! receive they ne- egeriing goals. the last 0.11 tgoondeent of 5 tests how ofte disagree with tl bulge: recommend izplies that in stereos when it trio“ not disbur lature's influee iisolaved by th. feral review 11 troughout the ' ores Without 182 ct their formal authority over the executive if they eive they need to do so in order to achieve their own ling goals. The last question intends to represent the executive's indment of funds that takes place on the national level. sks how often: "The executive, even though he may gree with the changes made by the legislature in his at recommendations, he carries them out anyway," this Les that in situations of conflict over spending pref- :es when it comes to budget execution, the executive 1 not disburse the money as appropriated. The legis- re's influence over the executive and budget outputs, Layed by the changes imposed at the time of their Ll review must also be extended to the implementation [ghout the year. Making alterations in expenditure 1 es without having them subsequently executed: as i xecutive is able to ignore the decisions of the lature with impunity is certainly a sign of legis- e weakness. The executive responds that to a great t he carries out the changes made in his budget recom- tions, even when he disagrees with them, but the lature itself is less certain, believing that to extent the executive does not follow through on the t of their actions. tltogether the five questions describe the possession :ercise of between some and great legislative formal influence tmrd city council is ecutive as they independent ass tore totals. '1' possess a relat micipal budge Rel It is expe council, who ex loiifications ' hue some kind the executive h influence, at t View would cart the executive 1' Evening hi inthe size of reWests or the affected by the Wluence pose “93 not confi ionnectim is ”quests (0.02 legislature (c .— 183 uence toward the executive. By this measure the council is not a "rubber stamp" to a dominant ex— ive as they do exercise their authority to make an pendent assessment of the annual change in expendi- totals. They are far from powerless, but instead ess a relatively strong amount of influence within the cipal budget process. Relationship to Executive Budget Behavior It is expected that the presence of a strong city ail, who exercises their authority by imposing fications in the executive's recommendations, would some kind of effect upon the previous decisions of executive himself. That is, the exercise of formal uence, at the time of the third stage of budget re- would carry back to the spending choices made by executive in reviewing departmental requests and ‘eparing his own spending revommendations. Either w size of the reductions made in initial department sts or the amount of increases recommended might be ted by the executive's evaluation of the amount of ence possessed by the legislature. But the data not confirm such a relationship. No statistical ation is revealed, either to cuts in departmental sts (0.0299) or to the growth proposed to the Lature (0.1298). is evident influence over 1 authority to cha review does not ate in the pre budget to be re lint I111 subse this aspect of neither the exe iilfiking their Preferences an ihe authority boundaries of of decision main Eieht). Rela ] Ex: When it Cl ofexecutive 1‘: Nations and situation is In tiVeiteci to see Changes as pos turn I to imDOS PEDI‘ESent the 184 As evident in the previous analysis of executive fluence over the departments, the exercise of formal thority to change spending totals in their official view does not extend backward to affect the decisions de in the preceding stage. The formulation of the dget to be reviewed is not affected by evaluations of at will subsequently happen to it. As measured by is aspect of the structure of budgetary influence, ither the executive, nor the department heads defers making their own budget choices to the Spending aferences and influence of the legislature (executive). a authority of each actor does not reach beyond the nndaries of their own budget making arena. The sequence decision making stages are compartmentalized (see Chapter :ht) . Relationship to Legislative Budget Behavior-— Legislative Changes in Executive Recommendations When it comes to the legislature's official review executive recommendations, their antagonistic spending itions and objectives, that place them in a conflictual uation is most readily apparent. The executive is mo- ated to see his recommendations accepted with as few dges as possible, while the legislature desires, in n to impose just those changes that are opposed. The spending recommendations of the executive partially esent the result of his bargain with the department heads. lenar’uents are annual expensiou stage and expect Iith the city 0 final appropria the legislature the executive t end of the comp 0! such a situ to cooperate vi to establish 1 order to preve objective of i the executive 1 Aaron Wild atencies, the g This influence t00 often over: find that its . “fiat leopard: “illicit, is t3 Sites that is Most se‘ Bureau la \ he 11. here % (Bostc 185 partments are able to satisfy their own goal of an ual expansion of expenditures at the end of the second age and expect that the executive will use his influence th the city council to retain these increases in the nal appropriation ordinance. Subsequent reductions by e legislature, naturally upset that agreement, and exposed e executive to weakness for not being able to deliver his d of the compromise with the departments. The prospect such a situation could lessen the departments willingness cooperate with the executive. It could motivate them establish independent contact with the city council in er to prevent such reductions and assure their spending jective of increases; and thus undercut the position of aexecutive in the process of budgeting. Aaron Wildavsky's study of the relationship between encies, the Bureau of the Budget and Congress illustrates s influence pattern. If the Bureau of the Budget is noften overruled by Congress, he writes, the Bureau may d that its "currency has depreciated." for a "record of cat jeopardizes its effectiveness."11 What is made licit, is that it is not just any alteration by Con- ss that is important, but only reductions: A most serious handicap under which the Budget Bureau labors is not so much that Congress may 11. Aaron Wildavsky, The Politics of the Budgetary ess (Boston: Little, Brown and Company, 1964), p. E1. raise its e inportant) a cooperati it has rect on receivin vculd have it is cuts in as he utilizes it the same tion is the change iennc writes 0 They see conservat' estimates 0only by p by carryi House can requests hperienc cuts and formidablu activity. The assoc ' reductions in (I! confirmed. Th iorrelation sh them these Hade. It not directed at d, b“ that it re \ 12- liar 13. Rio 186 raise its estimates (though this is obviously important) but that it cannot guarantee that a cooperating agency will receive the amount that it has recommended . . . If agencies could depend on receiving the Budget Bureau's figures, they would have much greater incentive to cooperate.12 5 outs in his recommendations that the executive opposes, a utilizes his influence to prevent them from being made. be same time, it is exactly that this kind of change he change legislatures desire to make. As Richard 0 writes of Congress: They see their job, essentially, as the negative, conservative one of reducing executive budget estimates wherever possible. Basically. it is only by prescribing this job for themselves and by carrying it out that the committee and the House can be powerful. To rubber stamp executive requests would not be the road to influence . . Experience has shown Committee members that budget cuts and threats of budget cuts are their most formidable sources of control over executive activity.13 The association between legislative influence and tions in executive recommendations is moderately 'rmed. The association is 0.3891 + 0.10. This lation shows that the distribution of influence en these two actors does account for the decisions It not only indicates that formal influence is ted at decreasing the budget received for review, hat it requires the possession of more influence 12. Aaron Wildavsky, Budgetary Process, p. 40. 13. Richard Fenno, "Impact of PPBS," p. 179. cent the budge easier for the c isto out it, as bunch as he (1 Stale. by not b does by additio But beyond ecutive's reco £1debange repr be substantial Spending modifi tier. tweak 1 Spending choice if municipal bul he blanket acct Sign 0f a power dlhrent in the “it made by th as they accept \ 11h This lInold hi hi altsner ”ages to hi 187 1” It is t the budget than it does to add to it. r for the city council to add to the budget than it cut it, as the executive does not resist increases ch as he does decreases. The executive has more at , by not being cut up by the legislature than he y additions to it. ut beyond the imposition of reductions in the ex- e's recommended budget it is further evident that ange represents the exercise of legislature influence. bstantial association toward the absolute size of .ng modification (0.5781 + 0.005) supports this A weak legislature is one that accepts executive ng choices with few alterations. Previous studies icipal budgeting have been correct in explaining anket acceptance of the executive's budget as a f a powerless legislature. This is most strikingly at in the three cities where no modifications at all ade by the City council in budget recommendations, I accept the appropriations ordinance exactly as h This finding challenges the conclusion of Meltsner that the executive opposes legislative es to his budget for fear of upsetting the often- elicate equilibrium of a balanced budget. As he "The manager . . . does not want to be cut up council. To be "cut up" in this context means e council will be fiscally irresponsible and expensive projects or excessive salary increases. his budget so tight that, if the council wants ase an item, they will also have to find the to pay for it," in Arnold J. Meltsner, The Politics Revenue (Berkeley: University of California Press, . 183. This does not appear to be the case. the executive s rulers five, e ranks of the ex tub and three 1' replies and n ecutive respon legislature's f accurately ides isseeningly a However, t he decision no: of the structun Variation is v: legislature as it budget sub cOllllt'll are SO Spending Choi iiiediate eff L The effe dent in the f an mum of “lay utilize mosey to t mitting the Nineties (‘0 A 188 executive submitted it to them. These cities, bers five, eight, and ten, generally exhibit low ks of the exercise of formal influence (number one, and three respectively according to legislative lies and number two, one and ten according to ex- tive responses). So clearly, the measurement of the islature's formal budget making authority is able to urately identify the situations where the legislature seemingly a "rubber stamp" to a dominant executive. However, this pattern of influence between these decision makers is far from a universal description the structure of formal budget making influence. iation is very much in evidence in the extent the .slature asserts their formal authority to change budget submitted to them for review. Not all city mil are so impotent and subordinate to executive ding choices; and when they are not, there is an rdiate effect upon their budget behavior. Relationship to Legislative Appropriations The effect of legislative influence is further evi- in the final dependent variable of the direction amount of annual change in expenditure levels. Since utilize their formal authority to reduce the budget osed to them, this decision has the result of re- cting the yearly expansion of the appropriations oance (-0.2301). The strength of this correlation ecutive. The the sore succe holding the li tion from both iornal budgetau lbnicipal spent ] Conpletin ID11511 influent thee decision deilal‘tnent; hea Misfit maki portion of th departments eclithe 'S abi parthem hEad the city Colin of this pract chilly do n 189 just barely sufficient to be categorized as moderate. re is some evidence that a strong legislature, serves a force opposing annual spending growth; a position ectly in conflict with the departments and the ex- tive. The formal influence the city council possesses, more successfully they achieve their own goal of lding theline" on spending. In the face of opposi- rnfrom both these two other actors, the pattern of nal budgetary influence does have a direct impact upon icipal spending policy. Department-Legislature Influence Relationship Completing the description of the exercise of nal influence over the budget as it moves along the re decision stages is the relationship between the nrtment heads and the legislature. This feature nudget making influence is significant because a ion of the executive's dominance over both the rtments and the legislature derives from the ex- ive's ability to limit their direct contact. De- ment heads do not have the opportunity to appeal to city council for a restoration of previous cuts in r requests. As Thomas Anton writes of the absence his practice on the state level: " . . . agencies ally do not press for increases once the Governor i has turned the: of urban govern; the mayor's he for their pr office. No Ii seen to exist and department Then the the legislatur and have to re the city counc nth the de fluence thee recommended e Cities follow my] ’ this structur‘ to legally Do Powers and is 3MP. in the Mich making ”peeing c1 \ .De 19%5. Th 16. Pp, 50.515“ 190 ?ned them down."15 John Crecine similarly writes in government: ". . . departments do not go "over ror's head" to the council to request more funds air programs or to restore cuts made in the mayor's . No widespread system of stable relationships ) exist between particular members or committees l6 partments." hen the executive also exercises his control over gislature. They are denied an alternative voice re to rely upon the executive for information. 3y council is prevented from forming a coalition 1e departments as a resource and strategy to in- : the executive in their own goal of modifying ended expenditure figures, and eleven of the follow the city manager form of government. In :ructure, the chief executive officer is supposed .1ly possess all appropriate administrative and is to provide coordinated and unified leader- n the budget. as well as throughout the governmental making process. In many of these cities, there is fic charter provision that intends to limit the f contact between the legislature and the departments ,___¥ Thomas J. Anton, Politics of Expenditure, John Crecine, Financing the Metropolis, 51. i ieorder to ens eger as head of Ihisphrase, of the next state heal [ith the city nenager thereof shall the city manag tents of the hem of thei the mere prese Wide a desc Settle actual testigated in Schedule. tot all ‘ report the dep: EhedeCisions ‘ hedgets. on 't mite" O.'e ”Mined the Bur ment heads \ e 18. Dy. 50‘51. J0] 191 yr to ensure the position of the city man- head of the executive branch of government. rase, often repeated verbatim from one city to t states that: "The council and its members shall th the administrative service solely through the nager and neither the council nor any member shall give orders to any of the subordinates of ] manager.l7 Department heads are intended to be 3f the executive and not independent actors on )f their own goal of increased funds. However, a presence or absence of such a clause does not a description of what necessarily takes place. .ctual character of this interaction must be in- ed in more behavioral terms through the interview t all descriptions of governmental budgeting he department heads as passive participants to sions that are made by the legislature on their On the national level, Aaron Wildavsky writes . . everyone knows that agencies make end-runs 1e Bureau to gain support from Congress."18 De- heads do not always support the executive's Led spending totals for their departments. .—__ John P. Crecine, Financing the Metropolis, fl . - Nan“: " .- - , -. ‘ '_ often they d901- VIIU «. i I lettingtheird ‘ h Executive administr agencies it in 1e amarked too enthu will then originall There fol the form Executive hinders 0 man cites the money The presu asked for made expl the Admin however, demands 1 “Chard Fenno between depart that I Norman the exez \ 19. Aar 192 they declare their autonomy from the executive, ng their dissatisfaction known to the legislature, attempt to obtain more money than the executive nended for them. As Wildavsky goes on to write: The Bureau of the Budget, under Presidential direction, lays down the rule that members of the Executive Branch are not to challenge the Executive Budget. But everyone knows that the edministrative officials want more for their hgencies and are sometimes in-a position to get it in league with supporting Congressmen. The eesult on these occasions is a formalized game, thich any reader of appropriations hearings will 'ecognize. The agency official is asked whether er not he supports the amounts in the President's :udget and he says 'yes' in such a way that it :ounds like yes but that everyone present knows ‘hat it means 'no.‘ His manner may communicate .marked lack of enthusiasm or he may be just 00 enthusiastic to be true. A committee member ill then inquire as to how much the agency riginally requested from the Budget Bureau. here follows an apparent refusal to answer in he form of a protestation of loyalty to the Chief xecutive. Under duress, however, and admidst re- inders of Congressional perogatives, the agency an cites the figures. Could he usefully spend he money, he is asked. Of course, he could. he presumption that the agency would not have sked for the money if it did not need it is 1de explicit. Then comes another defense of he Administration's position by the agency, which, >wever. puts up feeble opposition to Congressional emands for increases.l9 h Fenno adds to this description of the relationship idepartments, the executive, and the legislature rmally, the agency fights for its desires within e executive branch, accepts the decisions made _—_ fifi . Aaron Wildavsky, Budgetary Process, pp. 88-89. there (hid it can get fend that then its h gran objec over the c trench, it pectatione [italics :‘ asked for. expectatic done . . . if more i: fared by David level, the WTi‘ Cffitial is uni Stage! he may ' ln‘ -.hers ”to seep: 193 here (ultimately by the President) as the best t can get, and goes before the Committee to de— end that dollars and cents figure and no other. hen its budget estimates badly distort its pro— ram objectives, and when it feels deeply grieved ver the decisions made within the executive ranch, it is accurate to say that its goal ex- gctations are that Congress should give it more italics in original7 money than it has formally sked for. The problem of communicating such cpectations are difficult, but it can be Jne .. .20 fmore immediate relevance, is the description of— y David Caputo of this process on the municipal He writes that: "If a departmental budgetary l is unhappy with the results of the formulation he may try, during subsequent stages, to persuade to support him . . . action has a very high prob- 21 Where of influencing subsequent behavior." ants have the opportunity to go to the legislature ‘se executive decisions, the municipal budget can no longer be considered dominated by the e and the determination of final expenditure than where such a pattern of formal influence t. Richard Fenno, The Power of the Purse: Appro— : in Congress (Boston: Little, Brown and Company, u 271—272. David Caputo, "Normative and Empirical Impli— >f Budgetary Processes," prepared for delivery ith annual meeting of the American Political ssociation, Los Angeles, California, 1970. The first 2:” the departme sihpiy their, ' their deliberate patience of eve the occurrence Inthe case of see table 5.5 3:71" did have 2 "iii-e. the leg: hertnent head he eehrohirhat thee no depar- ‘ I ‘K:\‘ ‘1 . Inn“ 4‘ ' hext, the Same QUEST. iep‘fl‘ihehts an 1521310“, the 194 Analysis he first question asked to measure the character departments relationship with the legislature was their, "appearance before the city council during ieliberations on the budget." Because of the im- :e of even a formal meeting with the legislature, :urrence of this contact was measured separately. case of the three department heads interviewed ble 5.5) an average of one half of them in each d have an official meeting with the city council; the legislature reports that they met with most ent heads (see table 5.6) in the course of adopting ropriations ordinance. This practice extends from o departments made an appearance, to where all of i. xt, the "purpose of this meeting" was examined. a question was asked of the meeting between the ants and the executive and with the addition of :gory where departments "appeal" executive budget a, the same ordering of responses are utilized.22 The scale of responses was slightly altered ibe the purpose of the department heads meeting legislature. The first code of no meeting d separately and therefore excluded from the coding. Therefore all the others were moved position. The category of "appeal to legisla— S then added as the highest indication of ntal influence. Consequently, the range of 3 still extends from zero to five. “um we. / h» 31/ \lih. O O O phi} Nil. AU 0 /.hL C 0 O Ra ..h. a xfiv 30 0U O rhhv AJeJ fihaO \slU AV 0 0 0 «b/ DC at: — QV - - - - - - - ' - - - - - - - - 3Q. hu. 1.1. 1|. 0 w». AHU My «It. .ui 1H». 0 A/~ 1,th 0 AC «Ii 0 h 5. Wu. . - . . - . . . ~ ~ ~ . ~ ~ ~ ~ Dr. [A it fit} .htU AU Wu \IV as}. A...) 0 AU (he 0 0 .HJ AM/ .Il‘ ... .. . xx» nt/ AK n...V fin,» \hhv .th «it; \wv O fAU fit.» AU FNJ 3911/.» . C. eh. 3.1 q 3....” he. hhhu hunwhufitnh i... he. «.1 e .h . .~ I V . ~.~. .. ‘4 - -.h\.4~..1 ~ ..- .- es - I ~ I . u I I 9 \ t t Q \ \ e \ V 195 TABLE 5.5 PARTMENT-LEGISLATURE FORMAL INFLUENCE (DEPARTMENT REPLIES) Appear Pugpose Dept. §§k gggiare Mean Meeting Requests Support Indepen- dence >—2.66 5.00—3.33 0.66—2.66 1.00-4.00 1.00-4.00 3.37 3-l.33 2.00—1.33 0.33-1.33 -- -- 0.80 )-0.00 -— 1.00-4.00 -- -- 0.80 )-4.00 3.00-2.00 0.33-1.33 0.50-2.00 -- 1.87 )-0.00 -— 0.33-1.33 -— -- 0.27 >-4.00 4.60-3.07 l.00-#.00 1.00-4.00 0.66-2.66 3.h0 h-l.33 5.00—3.33 0.66—2.66 1.00—4.00 -- 2.27 -l.33 3.00-2.00 0.00-0.00 0.33-1.33 -— 0.67 - .00 2.80-1.87 1.00-4.00 —- -- 2.24 -0.00 -- 1.00-4.00 —- -— 0.80 -2.66 3.00-2.00 1.00-4 OO -- -- 1,73 -4.00 3.50-2.33 1.00—4.00 -- -- 2.07 -0.00 -- 0.33-1.33 -- -- 0.27 -2.00 3.00-2.00 1.00- 00 -- -- 1,60 -l.95 2.49-1.66 0.69-2.76 0.27-0.67 0.12-0.48 1.59 -O.82 1.85-1.23 0.36-0.36 0.07-0.07 0.03-0.03 1.05 . .t- wqw Ma. «C 00 a/- «)2 AU we; 0 0 NV .50 O 5 <1, WU 00 m- h e /C 30 (J .4; r) n/. r330» {C /C K) 730 O 5 Aux Q\ .u: “U. - | - - - I - - - - - - - - - - he.“ nu. 3‘ 1 he. WV «(J 1 fie / m/s 1C «.3. fl: m/a a/~ 1.. 3 a/L 0 Duo liu _...u _ J h AI . l e I. n w V ~ H A s N \w A ‘ ~\ ‘5 \tVb \ um: 4 196 TABLE 5.6 )PARTMENT-LEGISLATURE FORMAL INFLUENCE (LEGISLATIVE REPLIES # Purpose Dept. %:k Sept. N ec are :ean Appear Meeting Requests Support Indepen- dence 3.62 2.41-1.61 0.50-2.00 1.25 1.00 1.90 1.80 3.00-2.00 0.80-3.20 0.33 2.00 1.87 0.57 3.00—2.00 0.71-2.86 0.66 1.00 1.42 3.33 3.42-2.28 0.33—1.33 1.50 3.75 2.44 1.50 3.20-2.13 0.43-1.71 0.00 0.00 1.07 3.71 3.50-2.33 1.00-4.00 3.00 2.40 3.09 2.50 3.72—2.48 0.88-3.33 1.66 2.50 2.50 2.80 2.83-1.89 0.00-0.00 1.40 1.00 1.42 3.60 3.00-2.00 1.00-4.00 1.40 1.00 2.40 2.66 2.80-1.87 0.50-2.00 2.75 1.00 2.06 2.50 3.00-2.00 O.25-1.00 0.00 0.00 1.10 2.75 2.80-1.87 0.75-3.00 1.00 2.25 2.18 1.85 2.83-1.89 0.57-2.29 2.25 1.00 1.86 3.00 2.87-1.91 1.00-4.00 1.14 1.80 2.37 2.58 3.02—2.02 0.62-2.48 1.13 1.48 1,98 0.90 0.34-0.22 0.31-0.62 0.92 1.03 0.58 his situation- enhority: ”I e additions that case for certaf . . . They were the manager ha: the legislator feet for the The respc TeSpohds to th emange of i 'n': e . t; Mn ~~wlne One' 197 situation describes a decentralization of executive rity: "I was able to ask the city council for ions that the mayor had cut . . . I can plead my for certain items left out by the city administrator They were there if they wanted a budget above what anager had recommended for them . . . They went to egislature if they were not satisfied with his t for them . . . Departments seek audiences with ity council to obtain an increase in their budgets." The response of the three department heads cor- ldS to the midpoint between the third category of inge of information" and the fourth category of hing one's case." This indicates a position sub- hte to the executive. If the four cities are led from the analysis, where no department heads - appeared before the city council, the response wes one whole category and is midway between hing one's case" and the "place where actual de- 6 are made." This indicates the weaker control executive over this interaction, where departments e actual opportunity to meet with the city council. SPonses of the legislature supports this latter ralized pattern, as they report the purpose of meeting with all departments was to provide the unity for departments to plead their case. The executive by a not being able eeeeimess ove .tsecono‘ control over i satiation of t with the execu Laure." Int 335 presented; its executive ' Practice is a Executive be», is hrnold Fiel' ;n his de “35 insi: not show wants eh. not crea- the FeQu. Fanager, end as than 4.‘ . -- been oud. ‘l‘e . :.cshlmably th “Nations do so ~ . dome t '1'“: . ‘NlnlShed . ‘\~i\\\~“ 11.33' Arm “JUdgetin 198 :ive by allowing this contact to take place, or ring able to prevent it, is most certainly revealing mess over the process of municipal budgeting. .second structural feature of the executive's 1 over these two other budget actors is the pre- ion of the, "Departments' initial requests along he executive's own recommendations to the legis- ." In this case, two sets of spending figures esented; original requests of the departments and ecutive's subsequent recommendations. Such a :e is a potential breach in the dominance of the ive by loosening his monopoly over information. eld Meltsner and Aaron Wildavsky write: 1 his desire to gain control, the city manager LS insisted that the departmental requests do ht show in his submission to the council. He hhtS the budget to appear to be his own and m create any cause for controversy by showing e requests of the departments. For the city .nager, the budget's form is almost as impor- .nt as its substance.23 th budgets are presented to the city council, bly the evaluation of the executive's recom- ns does not take place on a clean slate and ing the unity of the executive budget is ed. The frame of reference for legislative Arnold J. Meltsner and Aaron Wildavsky, "Leave geting Alone," p. 342. ‘ ' ' o I 'seifl‘.“ T1211”. I J o to ma 9.... nee-"eive deco: ‘3'qu ”A: "2‘9”-“ y. .u'. J‘VUM- Lg. u Exciter}: ‘oesi .. :N-qu‘: fifi!f‘-fi We -I» ygy‘u tun ye .. n: 5ka I‘D/“1h“ tVru b. L.‘ e : Aw." Wfififiaq m. “.1“ N‘ -' x "“v 'W‘n n :‘Mie .u] ‘C \ .:7\\‘ “gen: ’18 u. A e '1 " x‘d\ K. W A ‘- :. ‘7 .ng . \‘eh :‘ , “L r‘ the 31% imnc 1 VA: 199 sion making is changed from why should they modify utive recommendations to why did the executive re- departmental requests. The legislature is now able irectly compare the executive's budget to some other hmark besides previous appropriations and the ex- ive could be expected to be called upon to justify )wn reductions in departmental requests as well as )wn increase over previous funding levels. For the executive this is a situation to be avoided: manager does not want to be in a position of de- ng his budget from two sides . . . I wouldn't it. I can't put myself into the position of having tment heads debate with me before the city council, ey are going to respect the fact that I am manager The council isn't intended to compare department sts with the executive's recommendations. To do 11d open a door for a schism between the department as manager." Responses to this particular question indicate that ives in the present set of fourteen cities are not owerful. An average of three quarters of the de- nt heads interviewed and members of the legislature iewed report that initial budget requests are for- to the city council. .he importance of this feature of influence relation- xtends beyond the pattern of institutional arrangements. 011' g 'h‘ M V “.3 L, '...'-.'. ' nH—gr 341:. C. VLU‘VH ' ' c n ""1 runes Ind-In ..AJ ' c 3"“‘2": WC. -... ...v.. st. .4... "- "‘1 cvton-i- "' "W Humans u “NH ~w. ‘ H w... '2' “C I)” (. 3;“. EX In. ‘ -,5: 27‘; A RV: "nun. ‘. \. U‘.‘ V n ‘ I . ‘ ' ‘ '2."~'h-‘ TA “ ‘v; :xng \Ic -‘ ‘fiw- ~ I :‘fi‘.U’~‘. . ~ ‘- r. “u ‘ 4 v- " me »~ - .~ I ‘ \ “4:“th \u" 4.x. L 0 "w ““hfi‘w ‘n: “\a' , E - A .‘ ~_fly\ ‘ w.-\:; r: f‘ N xx.“ ge V \‘ ‘ .. \.H§ ‘l "a 5‘ s \ «‘\ flex“ . N K‘ "V‘N v , y ."w. he!) \ h \‘fha‘ ‘ ‘24“ . ‘ o \. Eu. ,V . a" v. ,A ‘1 ~V:‘|:‘|\ O . M'l * \‘ o1 ‘ L .‘V ~21»- u“ that" " '[~\ | ' I u t lfll‘cia :S‘E‘nq ‘. Q V‘¥\_ as a S {‘1‘ e .N the 200 'esentation of department budgets provide only wential for a comparison to be made; nor he absence of such information within the official document prevlude the legislature from making quiries. Therefore, the actual behavior of de— ts and legislatures was more directly measured extent they both use their meeting as an oppor- to "go around" the executive. in an executive centered process, the department .re only agents of the executive in the final stage slative decision making and appear in support of ve spending totals. They are not autonomous actors subordinate to the needs of the executive: "I am t the request of the legislature to answer their 18, but I am not able to bring items directly to oil without first screening them through the and getting his OK . . . to justify the requests manager and I had arrived at . . . The manager artments to sell his reasons for supporting a . the departments are there for the manager. ose was to have the departments justify requests of the manager." On the other hand, in a more ized system the legislature, whether or not they nitial budget requests, probe for such a dif- S a strategy in their own influence relationship executive. Then, departments faced with this question, use independence f :‘2r executive patent and a :e their budge So first ieparment hea recomendatior response 1 5199M or not !\ t‘x '.‘- ‘ .. ‘v l____E_ 201 ion, use it as an opportunity to declare their endence from the executive. They do not support ecutive spending recommendations for their de- ent and appeal—- influence the legislature to add ir budgets. 0 first it was asked if, "The legislature asked ment heads if they supported executive budget endations for their departments," and then if, sponse to this question did the department heads t-or not support the executive budget." However, coalition does not appear to extensively exist. ' five cities does at least one of the department .nterviewed indicate that such a legislative chal- ’o the integrity of the executive budget was made, y to an average of a slight extent does the report having formulated such a question. So appear that the municipal legislature does not establish a break between themselves and the ex- Then, when such a question is posed, departments eadily take up the gauntlet to oppose the ex- In only two of the cities where this question d do department heads report having notsupported utive's recommended budget for their department to a slight extent does the legislature reply departments declare their autonomy from the 9. mogethe extent does a legislature a ( )1) tween these t‘h rating stage. and the exe 1 | o n H .5 3M? ....-. of thei the legislatur Trity to inf] geai of effect mag Cities :' 202 Altogether these five questions indicate that to some at does a coalition between the department and the slature against the executive mark the contact be- 1 these two decision makers in the third decision 1g stage. The departments make some attempt to go 1d the executive to influence the legislature in .f of their own goal of increases. To some extent egislature themselves see this meeting as an oppor- ;y to influence the executive and achieve their own of effecting changes in the budget. The difference icities is extensive, ranging from where departments no formal contact at all with the city council to there is almost very great contact between them. Interestingly, by both accounts, the three mayor s comprise three of the four highest ranks. While are too few cities to make more rigorous comparison n these two types of governmental organization, is a suggestion that city managers are more able trol the interactions between the departments and gislature than mayors. At least by this measure, r governments display greater executive central- n in the budget making process. Relationship to Department Budget Requests here is no apparent relationship between the amount uence departments perceive themselves to possess is the legislature and the original formulation ef' expansionar expected that Legislature wc the case. is revealed in ti expenditure 1“ i review does n: Pref-res the t 203 xpansionary requests (0.0491). While it might be cted that the departments ability to influence the slature would lead them to ask for more, this is not case. As encountered earlier, the pattern of influence aled in the exercise of formal authority to change nditure figures at the time of each actor's budget ew does not extend backward to that stage that first ares the budget that is being reviewed.‘ Relationship to Executive Budget Behavior Asking the same questions to the executive (see table reports his perceptions of a slight coalition be- tthe departments and the legislature. However, there eapparent connection to his own spending choices. nfluence relationships evident at the time of the stage of budget review does not affect the executive's w of departmental requests and his own recommendations ht be expected. If such an "end-run" is a sign of ive weakness, it is not evident in smaller reductions 'tial budget requests (0.4200, a negative correlation ected), nor in larger increases in recommendations legislature (0.1866). As before, the exercise of influence within each stage, does not carry over decisions made in an earlier stage. 0 «L. w ”I. a nu AUOOOVN.IF\IU (IOOSO/nbo 7 WI; . MW new 0 .rlu “UV A U 9‘ 0 7 0 O A/K PD/HV 5 5 0 nnV\\ .T. a - - - - - - - - - ' I. - - - ' - Hm D. «(‘J “iv flu UNI nllv Au\./ A/~ OVIIV 0 An/~ Kn‘wlo 0 n ‘ u - Inn _.L. I F I a - I ~ I \ DEPARTMENT-LEGISLATURE FORMAL INFLUENCE 204 TABLE 5.7 (EXECUTIVE REPLIES) Ask Dept. # Purpose y Appear of Rgeggsts If Declare mean Meeting q Support Indepen— dence 3.00 4.00-2.67 1.00-4.00 1.33 —- 2.20 0.00 0.00-0.00 1.00—4.00 0.00 -- 0.80 0.00 0.00-0.00 1.00-4.00 0.00 —— 0.80 4.00 2.66-1.77 0.00-0.00 1.00 —- 1.35 0.00 0.00—0.00 0.00-0.00 0.00 0.00 0.00 3.25 4.50-3.00 1.00-4.00 2.75 3.00 3,20 2.00 4.3302.88 0.33-1.33 1.00 1.00 1,60 0.75 2.00-1.33 0.00-0.00 0.00 0.00 0.42 4.00 2.00-1.33 1.00-4.00 1.33 0.00 2.13 0.00 0.00-0.00 0.00-0.00 0.00 —— 0,00 2.25 2.66-1.77 0.25—1.00 0.25 —- 1,05 3.50 3.00-2.00 1.00-4.00 2.00 1.00 2.50 0.66 3.33—2.22 0.25-1.00 0.00 -- 0.78 0.50 2.50-1.66 1.00-4.00 0.00 -_ 1,h6 1.57 2.21—1.47 0.56-2.24 0.69 0.36 1.08 1.00 1.10-1.10 0.17-0.17 0.14 0.07 1,10 Le F hange Examinir iepartaents 2 the legisla‘u apartments 2 restoration e fluehce depa.‘ that are abl lative “3le H'er *7 205 Relationship to Legislative Budget Behavior-- Change in Executive Recommendations Examining the direct relationship between the artments and the city council to the changes made by legislature in reviewing the executive's budget, artments are moderately successful in appealing for a toration of previous executive cuts. The more in- ence departments possess toward the council, the more rare able to persuade them to add to executive budget ds (0.2910 from legislative replies and 0.4706 + 0.05 [the departments). The executive is correct in wanting estrict their contact for when department heads meet .the city council during the formal review of the et, the departments effectively change the behavior he legislature. Department heads are able to regain some he cuts that were imposed by the executive from the :ipal legislature. ‘ Relationship to Legislative Appropriations H0wever, there is no connection between departmental hence on the legislature and the yearly increase in ‘inal appropriations ordinance (-0.0099 from legis- e replies and 0.1824 from the departments). The t does not expand any faster as a result of the de- ents seemingly successful pressure on the city council ktore the executive's earlier reductions. The explanation for this contradictory finding is . . 0 a that the 1m lu effective in 1‘ uterine: of cl fro: the large lrzest increz :36 fade in 0 206 the influence of the departments is not actually ctive in restoring executive cuts. Althoughthe ring of changes in executive recommendations extends the largest reduction as the lowest rank, to the est increase as the highest rank, legislatures do not 1y add to the budget, as a result of departmental sure. In terms of the three departments, additions made in only five cities, and for all departments eases occur in only three cities. The average total ge for all departments is actually a reduction of 1.36 percent, and for the three departments it is a 3f (-) 0.41 percent. So departments' appeals to the slature to restore executive cuts are effective in a few of the cities. The impact of department in- me is more to defend and support what they already in the prospect of further cuts by the legislature it is to get back what was lost: "Departments ap- d before the legislature to assure that the city i1 kept in items the manager agreed to keep in . . . ere afraid that the council would "monkey" around heir budget. They were afraid we would cut their S . . . I am here to attempt to keep what the r has already budgeted for my department in the face islative cuts." n this case, departments are still successful, for gh they have not been able to influence the council :3 add to thei gets any more executive; th rate by the l! 207 add to their budgets, they have at least been able keep more of the increases they already have than if ey did not possess this influence. They have been fectively persuaded the council not to cut their bud- ts any more than it has already been reduced by the ecutive; even though there is a linkage to the changes de by the legislature, when it comes to the second asure of the size of yearly increases in spending vels, there is no statistical relationship to the cunt of department influence (see Chapter Eight). Shatter 31)“ - influenc alihg concre A to ihillaie' no: complete] :‘luence. It of Power." has to inClu‘ 11 the inter2 art and '1an of other deC outcomes mus fsiorton Beret 'uoh~decisio exists when: ...A inforci institu of the of only innocuc 1h doir purpose mter Six: Informal Influence Introduction Influence is exerted in other ways than through ing concrete spending choices. The formal authority initiate, modify, or veto spending proposals does completely describe the structure of budgetary in- ence. It represents only one side of the "Two Faces Power." An analysis of the distribution of influence to include its less apparent and less direct presence :he interactions among budget participants. The co- : and informal exercise of influence over the behavior ether decision makers and over municipal spending :omes must also be investigated. David Bachrach and on Baratz write of what has become known as the edecision making" process, interpersonal influence ts when a . . . A devotes his energies to creating or re- inforcing social and political values and institutional practices that limit the scope of the political process to public consideration of only those issues which are comparatively innocuous to A. To the extent that A succeeds in doing this, B is prevented, for all practical purposes, from beinging to the fore any issues 208 that mig detrmen he most illu of influence lahl's famous to the I 3A3 (the (through Views 01 did not CaUSfie anticipz Strenu01 0f the C lyzing , limited did not resDect did not the may. the CAC as plat): (in its POWerfu find W0” In Shor iii In t. 1' Pet Smith: The W Fave .2. Pet Qt Po. 209 that might in their resolution be seriously detrimental to A's set of preferences.l most illustrious example of this aspect of the concept influence is Bachrach and Baratz's criticism of Robert 1's famous study of Who Governs in New Haven, Connect- t: To measure relative influence solely in terms of the ability to initiate and veto proposals is to ignore the possible exercise of influence or power in limiting the scope of initiation. How, that is to say, can a judgment be made as to the relative influence of Mayor Lee and the CAC (the business group) without knowing (through prior study of the political and social views of all concerned) the roposals that Lee did 393 [italics in original make either be- cause he was warned explicitly or because he anticipated that they would . . . provoke strenuous opposition and sanctions on the part of the CAC. Dahl was not interested in ana- lyzing or appraising to what extent the CAC limited Lee's freedom of action . . . Dahl did not consider that the CAC might in this respect have exercised power. That the CAC did not initiate or veto actual proposals by the mayor was to Dahl evidence enough that the CAC was Virtually powerless: it might as plausibility be evidence that the CAC was (in itself or in what it represented) so powerful that Lee ventured nothing it would find worth quarreling with.2 In short, the exercise of influence is not visible, more covert. It is employed not to change the course overnmental policy once a proposal is brought forth 1. Peter Bachrach and Morton Baratz, Power and flafl,Theory and Practice (New York: Oxford Uni- Lty Press, 1970), p. 7. 2. Peter Bachrach and Morton Baratz, Power and 31. pp. 15-16. for decision. using of den 1: prevent th :ussion. to others and pu calculates ti froze articule 1:. this way, not submitte< cation of in is expl of formal au if their rev ifiore. The Quests does requests: TIC executive ' s ChoiCES. hoe Weep is a 3 GI mflUERCI 210 m decision, but to create barriers to the public ring of demands for action. Influence is utilized prevent the issue from entering the arena for dis— ssion. No challenge is made to the preferences of hers and public conflict is muted. The weaker actor lculates that he could not win and sees no advantage om articulating specific demands for a change in policy. this way, behavior is changed as spending items are t submitted for review if it were not for the appli- tion of informal influence. As explained in the previous chapter, the exercise formal authority to alter spending totals at the time their review does not affect the decisions that went bre. The executive's reduction of departmental re- sts does not inhibit the submission of expansionary uests, nor does the legislature's reduction in the cutive's budget affect previous executive spending ices, however, the exercise of non-decision making er is a process for such a backward looking exercise influence. The executive and the legislature exert Ormal influence to shape the decision made earlier the sequence of budget making stages by the depart- t heads and the executive respectively.3 Influence 3. A third informal influence relationship that be identified is between the department heads and legislature. However, this is not included in the Sent analysis because it violates the logic of the is directed t TIinspected f official revi all Of what review in the not even 3839 The 000” of budgetary executive 130i executive. T Strong and it: re contained Ejevever, its ecutive would the municipal influence str lature is nor less powerful legislative i the adoption 211 m directed to affect the budget before it is officially mansmitted for review. instead of waiting for their wficial review. So, the question has to be asked. not nly of what was changed at the formal points of budget eview in the second and third stages; but of What was ot even asked for in the first place. The occurrence of this dimension of the structure f budgetary influence between the departments and the xecutive points to the increased dominance of the ecutive. The weak defer to the preferences of the mong and the positive spending thrust of departments e contained before they find official expression. waver. its existence by the legislature over the ex- utive would not be consistent with the portrayal of e municipal legislature as a rubber stamp. Now. the fluence structure is more decentralized and the legis- ture is more important. It is the executive who is 53 POWerful as a result of the exercise of informal §iSlative influence. The executive no longer dominates iadoption of the final appropriations ordinance as he \ 3. (continued) model of a three stage sequence 0f get decisions. The departments and the legislature are far removed from one another. to more than lnc}dent“ y affect the decisions they each make. The legislature iews the budget of the executive. not of the departments. departmental requests are only 1mp0rf6a1jlt as they are tered through previous executive dec1s10ns. The de- tment heads are more directly concerned With the entive's review, for any set of expendlture figures t emerge from the second stage will, be defended by executive before the legislature. is no longer encoding recc reflecting lo the wielder by which inf tified. As ' I o t the mec must be Consoio the sys the gap the lik some 3; argue c 0f inf] ihurning tc Emach and us identifj Theother is budge‘cing. 212 no longer in control of the preparation of his own ending recommendations, without incorporating and Electing legislative spending preferences. Mechanism of Prevention As in the concept of interpersonal influence it- .f, there must be some concrete connection between wielder of influence and its recipiant. The means which informal influence is exercised must be iden- ied. As Frederick Frey writes: . . ..the analyst must proceed to demonstrate the mechanism of prevention. This mechanism must be some form of the exercise of power-- conscious or unconscious-- by other actors in the system. One may speak loosely of rules of the game, dominant values, biasing belief and the like; but these must be disseminated through some specifiable influence process if one is to argue convincingly for a nonissue or suppression of influence attempt. 4 urning to the original formulation of the concept by mach and Baratz, two such "preventive mechanisms" identified. One is "anticipated reactions" and other is "explicit warnings." Both of these two esses are evident within past studies of governmental sting. 4. Frederick Frey, "Comment" on Raymond Wolfinger 1e American Political Science Review. vol. LXV, no. 4 amber, 1971), pp. 1092-1093. inticip? to each decis :0 others; a: patients do future cuts, requests, CO executive. proposals in conposite es they on ex; The exi is widely re 'cccgeting. Searching 1‘ is likely t the appropr their initi T————_———’ 213 Anticipated Reactions Anticipated reactions are a calculation, internal :0 each decision maker. of what is likely to be acceptable :0 others; an "estimate of what will go." Although de- nrtments do not pad their requests in anticipation of uture cuts, they still include, in formulating their equests, considerations of the probable response of the xecutive. Department heads do not prepare their budget roposals in a vacuum, but instead. ". . . arrive at omposite estimates of what to ask for in light of what 5 hey can expect to get." 1 The existence of this feature of informal influence 5 widely reported in Aaron Wildavsky's study of national udgeting. Department heads engage in a process of aarching for and receiving cues and signals of, "what : likely to go over" with the Bureau of the Budget and m appropriation committees. Departments, in formulating mir initial budgets, make "informed judgments" and ucated guesses of the disapproval of others as a iterion of what to exclude within their requests. They not simply add up the costs of all projects they may nt and deem necessary, but instead seek increases in iht of the possible and probable funds they can expect 5. Aaron Wildavsky, The Politics of the Budgetary 2£§§§ (Boston: Little, Brown and Company, 1964), p. 24. :oobtain. expenditures approval is . previously u :othe behav see process executive wi 307.038 prepa u,'g , . u ill need,"6 Sending Chg Suture of f E 3'“ accords. In 0rd rcacti must b muflica m poliCy the ye will m of tho Some 3 214 o obtain. Calculations of non-support are made, and ernditures are not requested when objection and dis- aproval is anticipated. While, most of what has been *eviously written of this concept relates specifically >the behavior of departments and their heads, the me process also describes the relationship of the ecutive with the legislature. Both of these two tors prepare their budget proposals with considerations . ". . . what they can get, rather than what they actu- ly need,"6 as informal influence is exerted over their nding choices. Explicit Warnings The second mechanism of prevention is through plieit warnings given by one actor to another. This Lture of informal influence relationship is more ob- 'vable than the calculation of anticipated reactions, .according to Richard Merelman this feature is crucial: In order to meet the problem of anticipated reactions, the pluralist focus on decisions must be expanded to cover patterns of com- munication [italics in original7 within policy areas . . . the powerful must communicate policy preferences and threatened sanctions to the less powerful . . . political initiators will make certain that the current desires of those who count are ascertained through some sort of overt communication.7 6. Aaron Wildavsky, Budgetary Process, p. 26. 7. Richard Merelman, "On the Neo—Elitist Critique 0mmunity Power," American Political Science Review, LXII, no. 2 (June, 1958), P- 456- The ovev when the bud-e point of int: :ore frequen' pence of but lease with i: formal neeti: Behind the s ciesance is ended for r John Cr tier. does It 55 he desori imam dlI‘ Earthen req app"Popl‘ia’cic it‘ll ieedbac hi the mayc neceSsaerv tc this, on. the ‘ 215 The overt and public contact among decision makers when the budget is formally reviewed is not the only point of interaction among them. Informal and often more frequent communication occurs throughout the se- quence of budget stages and throughout the entire year. Those with influence do not necessarily wait until the ormal meeting to express their spending preferences. ehind the scenes interaction occurs, as unofficial learance is sought before an expenditure item is for- arded for review. John Crecine's model of municipal budgeting, how- ver, does not include such behind the scenes contact 5 he describes the flow of communication in a uniform ownward direction from the executive's call for de— artment requests to the legislature's adoption of the ppropriation ordinance. He writes that: ". . . the sly feedback provided through past budgetary decisions n the mayor's letter."8 But he then believes it cessary to qualify this conclusion as he writes that is, ". . . is not literally true. Informal communi- tion actually does exist between areas A, B, and C in e diagram."9 However, Crecine does not go on to either 8. John P. Crecine, Governmental Problem Solving: omputer Simulation of Municipal Budgeting (Chicago: :1 McNally, 19697, pp. 36-37. 179. John P. Crecine, Governmental Problem Solving, explain the < nor provide : overt relati raters are i However informal pat his own node includes nun The bro but may relativ rent 0: soient relati valuab the po exist cess.1 Arnold P-OJ tunicitul ‘r thuger' n. Committee < Midget. .. ll rECOmmenda. By Statute 216 :plain the character of these informal relationships, )r provide any room for them. Only the formal and @rt relationships among the authoritative decision Lkers are included within his model. However, other studies have described these formal patterns of influence. David Caputo, presents s own model of the sequence of decision making that eludes numerous feedback loops: The broken line arrows are often overlooked, but may be the most important. Informal relationships may be crucial to the develop- ment of subsequent policy; however, if political scientists refuse to consider the informal relationships, their analysis will be less valuable . . . The broken line arrows indicate the possible informal relationships that may exist during any fiscal year's budgetary pro- cess.lO Arnold Meltsner's and Aaron Wildavsky's study of {land further identifies their presence within the dcipal budget system. They write that the city ager. ". . . works quite closely with the budget sub- mittee of the council to develop support for his ll He does not simply submit his spending get . n Jmmendations to the legislature at the time required :tatute, without having some prior exchange. Instead 10. David Caputo, "Normative and Empirical Impli— ons of Budgetary Processes," prepared for delivery 19 65th annual meeting of the American Political Ice Association, Los Angeles, California, 1970. ll. Arnold J. Meltsner and Aaron Wildavsky, "Leave BUdgeting Alone: A Survey Case Study and Recommenda- and Reform," in John P. Crecine, Governmental Problem El p0 338. he. '. ' ‘ f accepi- ' ' up their in iescripiion sort, that i .omelly ti“. its presenta reet frequer departments priorities v council with1 then involve forming the formal press least two up by the time already con help With t the budget. have period assess thei duly more he interac \ u 12. p “it Budget 217 ne,'h . . feels out what the council possibly might accept . . . takes the time to get to know councilmen and their interests."l'2 This most certainly is a lescription of the non-decision making process at Kuhn that is often echoed in the current study: Tbrmally the council has no part in the budget until .ts presentation to them for adoption. Informally I met frequently with them, even before I meet with the bpartments in order to get from the council what their udorities will . . . be, I try to find out what the :ouncil wants and what they do not want. I try to get wmm involved beforehand . . . I believe in the council forming the budget along with.the city manager . . . The formal presentation of the budget was the result of at east two months of prior "advice and consent" so that y the time the budget was submitted to the council it lready contained the ideas of the council . . . They elp with the initial formulation and development of in budget. Before the presentation of the budget we ve periodic work sessions, once a month, so I can sess their priorities." Who influences whom is cer- inly more complex than it appears by only examining e interaction between actors when budget decisions are ______ 12. Arnold J. Meltsner and Aaron Wildavsky, "Leave ty Budgeting Alone," pp. 3&2 and 343. officially In Two que preventive n operates bet tive. The 1' they can ob‘ Ihis statene On this com the executi ivoertnenta The se denn artments exeCUiiVe “ leads make :it absenCe I“ onlv rs. and in mi iht avoida; eupected O] departments their Otis 218 officially made. Department-Executive Two questions were asked to describe the first preventive mechanism of anticipated reactions as it operates between the department heads and the execu- tive. The first was that: "Departments request what tflmy can obtain, rather than what they actually need." Thus statement was taken directly from previous writings on this concept. The response of departments (see table (L1) and the executive (see table 6.2) is that only to a.slight extent are initial requests formulated in anticipation of the probable negative response of the executive. Both actors report the relative absence of the executive's exercise of informal influence over departmental decision making. The second question asks even more directly if departments: "Avoided making budget requests that the executive was likely to oppose, or did the department leads make them anyway." Responses similarly reveal the absence of this dimension of budget making influence. In only five cities does any individual department head, and in only six cities does any single executive report the avoidance of expenditure requests because of the expected opposition of the executive. By either account, departments are not constrained in the preparation of their original expansionary requests by the anticipated ebtain r‘\ f\ ( {h f /0 1.0 0 0 fi)/ «316 0 0 0 0 O 0 IO 0 1.0 IO 0 #51133/0 0 O O 0 506 5‘1. 9 0 - ' - - - - - - - -- - - - 1.0?M09~fld039~021fl/~0 219 TABLE 6.1 DEPARTMENT-EXECUTIVE INFORMAL INFLUENCE RELATIONSHIP (DEPARTMENT HEAD REPLIES) What % Dept. Executive M Obtain Avoid Contact Indication lean 1.66 0.00-0.00 0.66 1.66 1.00 0.66 0.00-0.00 1.66 1.33 0.92 2.00 0.33-1.33 0.33 1.66 1.33 0.50 0.66—2.66 0.66 2.00 l.#6 2.33 0.66-2.66 1.33 1.66 2.00 0.33 0.00-0.00 0.00 0.33 0.16 0.66 0.00—0.00 1.00 1.00 0.66 3000 0033“1033 0.66 2.66 1.93 2.00 0.00-0.00 1.00 1.33 1.08 0.00 0.00-0.00 0.33 0.00 0.08 2.00 0.00-0.00 1.00 1.66 1.16 1.50 0.00-0.00 1.00 0.00 0.62 2.00 0.33-1.33 0.00 0.66 1.00 0.66 0.00-0.00 1.00 0.50 0.54 1.38 0.17-0.67 0.76 1.18 1.00 0.90 0.25—1.00 0.48 0.79 0.57 that 530 fine finJ/O O 0 O 000 .310 Oat/SORJ/OOOOOOquo 0L001Mlflwiin.rlZA/.~Ol\i‘ uet- .v/b he / . ~ wk (lulu h/I WHHV AU/ Add» is 1r~ 31/ ulv Ii \ i A.) I Al I nu 1“ I‘ve 1V iifi fliiv wit ‘I. Q I t \ V v. ‘i‘ Vi 1.29 0.69 .9331 .ty. 220 TABLE 6.2 DEPARTMENT-EXECUTIVE INFORMAL INFLUENCE RELATIONSHIP (EXECUTIVE REPLIES) What % Dept. Executive V Obtain Avoid Contact Indication hean 0.05 0.00-0.00 1.33 1.33 0.79 1033 0.00-0.00 1.66 1033 1.08 0.50 0.00-0.00 1.50 0.50 0.62 1.00 1.00-p.00 2.00 0.00 1.75 1.33 0.33-1.33 3.00 2.00 1.79 0.66 0.25-1.00 0.75 2.25 1.14 1.00 0.00—0.00 1.00 1.00 0,72 3.00 0.66-2.66 2.33 3.33 2.83 1.00 0.00-0.00 2.00 0.66 0.92 2.00 0.66-2.66 2.66 3.00 2.44 2.00 0.25-1.00 2.00 1.75 1.44 0.50 0.00-0.00 2.00 1.00 0.88 1.25 0.00—0.00 0.50 0.75 0.62 1.00 0.00-0.00 2.00 3.00 1,50 1.29 0.22—0.90 1.70 1.56 1.32 0.69 0.34—1.32 0.65 1.03 0.68 negative reac‘tion O ahead to firSt prep sals for increasesr overtly eliminate 1 review of the secor i‘ne next two C communication outs: that provide expl'l in the departments department initiat budget requests he the executive to i would not be accey initiated contact requests, how 0ft: heads indications aeeepta‘ole to him letap the "cues national level. priorities of otl u. . mind takes p12 Me Where spem this the occurr. . lease“ SBeeches to . not generally in . 221 itive reaction of the executive. Departments go id to first prepare and to then submit their propo- 3 for increases, leaving it to the executive to rtly eliminate items at the time of the formal Lew of the second decision making stage. The next two questions look into the presence of munication outside of the formal meeting points, t provide explicit warnings of what not to include the departments' original requests. First, there is artment initiated contact: "Before submitting formal get requests how often did department heads contact executive to find out if a particular budget item ld not be acceptable to him." And then of executive tiated contact: "Before submitting formal budget Jests, how often did the executive give department 8 indications of what budget items would not be ptable to him." These two questions are intended ap the "cues and signals" that are described on the anal level. Decision makers learn of the spending rities of others, before the formal budget review ing takes place. This is not the first and only e where spending proposals are discussed. And e the occurrence of press conferences, news re- es, speeches and other such institutional practices ot generally exist on the municipal level, informal cations of what will not be supported does take place. ituch of mi throughout the anti :aeet with the C16! what is not going 1 available to his d‘ degree do departme‘: either of these tw executive reports extent. in either very pronounced pa the departments ar Altogether, < presence of the e: uloo the formulat The executive. ho- some exercise of inmoe over depar lau‘iation among t the tOtal absence °C°Urrence. the v the lower end of influence is me every marked pa the rim and se 222 . Much of this informal interaction takes place ghout the entire year: "We do this all year long. t with the department heads so that they know is not going to be approved and what funds will be able to his department." But only to a slight e do department heads repert the occurrence of r of these two forms of explicit warning, as the tive reports both take place at the level of some t. In either case, explicit warnings are not a pronounced part of the influence relationship among epartments and the executive. Altogether, departments perceive only the slight ence of the executive's exercise of informal influence the formulation of their initial budget requests. xecutive. however, reports between slight and exercise of this specific dimension to his dom~ e over department heads. Although there is consider- tion among the cities, as scores range from almost otal absence of executive influence to its great rence, the overall pattern of responses is within ower end of the scale. So, although such informal ence is present on an absolute level, it is not I marked part of the structure of influence between .rst and second decision making stages. Execv Examining the the legislature 0W too questions were ueohanism of antic whether the execut validations that th or did he make the my executive deci to the spending pr lature (see table of the legislators al’Oidence practice The second qt antimparted reactz' review of the budg cause he was cart“ This partiClllar S‘ conversation with pm“ his own bu the Spending pref 223 Executive-Legislature Examining the exercise of informal influence by legislature over the decisions of the executive, [uestions were also asked to measure the preventive Lnism of anticipated reactions. The first was er the executive: "Avoided making budget recom- mions that the legislature was likely to oppose. .d he make them anyway." In only five cities did xecutive decision maker report such a deference e spending preferences of a more powerful legis- e (see table 6.3). while an average of one quarter e legislatOrs (See table 6.4) perceived such an ance practice to characterize executive decision The second question measuring the occurrence of ipated reactions was: "There was little close v of the budget submitted by the executive, be- he was carrying out the policies of the legislature." varticular statement was suggested in a preliminary sation with one of the city managers who inter- his own budget responsibilities as implementing ending preferences of the city council. It was tention to recommend a budget that recorded legis- policy. so that subsequent review and modification oe unnecessary. This particular city manager went ay that he "kept a score card of legislative VA ab a; new. - v .m. . .1. ...t. u... durum tn. .1 «luv. .u... «U erry tut .1 V i -... u N V be Policies 0300fi7400§41J0000D 0 QJJ\J0 nmputer Simulation of Municipal Budgeting (Chicago: 1 McNally, 1969), p. 52. The impact of thl ation of its "'50 rroviding an arbl “4» C quest. . . an e.. L: or decreases) . " departments, ". . to M [italics evaluation of thf if the computer s all the others a: and current spent Adherence t 238 e impact of this letter occurs through the interpre- tion of its "tone," which, ". . . has the effect of oviding an arbitrary ceiling on the department's re- est . . . an estimate of allowable percent increases r decreases)."u Enhancing its important is that the partments, ". . . are perceived as explicitly responding ggly [italics ming7 the mayor's pressure."5 The aluation of this letter is the only behavioral component the computer simulation of department decision making; 1 the others are quantitative aspects, such as past d current spending. Adherence to this guidance has direct consequences the amount of additional funds that are requested. Crecine writes: "In most instances, then, the sum of a budget requests reaching the mayor's office repre- ms a nearly (within 10 percent) balanced budget."6 10 percent is the norem of budget increases sought the departments in a situation where executive policy followed and an executive centered budget making tem exists. The second pattern of this influence relationship vhere departments are relatively autonomous from the 4. John P. Crecine, Governmental Problem Solving, 5. John P. Crecine, Governmental Problem Solving, p. 59. 6. John P. Crecine, Governmental Problem Solving. p. 68. executive in prep described in the the executive ale pride the departh hdgets, but it < formulation of re pidance letter i It does not stru. problem"7 The 3 in the averarge l A epartments. b The explana do not pay atten found in the iss because both cit Instead, what is none that accorr (’1 1 eecifically, tc tail to hold the racked up with n 239 ecutive in preparing their initial budgets, as scribed in the study of Oakland, California. Here, e executive also issues a letter of budget policy to ide the departments initial preparation of their dgets, but it does not serve as a restraint upon the rmulation of requests for more funds. Instead: "The idance letter simply initiates the budget process: does not structure or delimit the actor's decision oblem."7 The resulting effect is direCtly apparent .the average 15.67 percent increase proposed by ten partments.8 The explanation of why departments in this city, not pay attention to executive guidance. cannot be und in the issuance of this letter or its contents, cause both cities follow the same set of practices. stead, what is most important is the normative expecta- >ns that accompany it that brings about compliance. cifically, to serve as an instrument of control, the l to hold the line has to be enforced. It has been ked up with more than words, but by deeds; and viewed as a request, but as an order. Sanctions have to be osed upon non complying departments. Those who ignore P; 7. Arnold J. Meltsner and Aaron Wildavsky, "Leave rBudgeting Alone: A Survey Case Study and Recommendation Reform," in John P. Crecine, Governmental Problem in 1 p. 33.1.. 8. Arnold J. Meltsner, City Revenue, p. 168. the instructions unalties and cos ienefits have to the executive. i punishments exist is absent in Oakl spread, it is acc penalties."9 De] lave nothing to j control the firs Two questior degree or" leeway their requests. nothing to lose ' ceived in the pr of the perceived 3the the nature As the question tiered in each C I‘m there is no the ' - ”18 indeed \ 9~ Ar J. nol my Budgetihgdh Arnold 10. .Ly Budgeting A 240 the instructions do so at their own risk, and suffer some mnalties and costs for doing so: at the same time, some mnefits have to be evident to those who went along with he executive. Presumably such a system of rewards and unishments exists in Crecine's three cities while it s absent in Oakland, and: "Since noncomliance is wide- read, it is accepted as part of life and it carries no nalties."9 Department heads soon learn that, ". . . they ve nothing to lose by asking.”10 The executive fails to ntrol the first stage of the budget process. Analysis Two questions were asked to measure the varying agree of leeway enjoyed by departments in preparing sir requests. The first is that: "Departments have thing to lose by asking for more funds than they re- ived in the previous year." This is a general statement the perceived existence of sanctions, without speci— ing the nature of the costs involved for Seeking more. the question is phrased, the department heads inter- ewed in each city (see table 7.1) agree only slightly n there is nothing to lose. Or in other words, that ire is indeed much that can be lost by submitting *— 9. Arnold J. Meltsner and Aaron Wildavsky, "Leave y Budgeting Alone." P- 333- 10. Arnold J. Meltsner and Aaron Wildavsky, "Leave y Budgeting Alone," p. 330. TABLE 7.1 DEPARTMENTAL LEEWAY (DEPARTMENT HEAD REPLIES) Nothing to Chances for M Lose Increases ean 1 2.50 1.50-2.50 2.50 2 1.00 0.33-3.66 2,33 3 2.00 0.33-3.66 2.83 A 0.50 0.50—3.50 2.00 5 3.33 1.33-2.66 3,00 6 0.33 0.33-3.66 2.00 7 1.00 0.66—3.33 2.16 B 2.50 0050-3050 3000 9 1.33 0066'3033 2:33 3 0.66 0.66—3.33 2.00 L 0.50 0.33-3.66 2.08 3 2.00 1.00-3.00 2.50 3 1.50 0050‘3050 2050 L 1.33 1.66—2.33 1.83 an 1.46 0.74-3.26 2.36 . Dev. 0.90 0.Le6—0.LeLe 0.37 expansionary req ree t0 l—O.) :hey are 33:: want without However, the “9 report as strong etent is there inally worded. 5 recalcitrant de; The second the costs invol\ obtaining expenC too large an inc ceiving increase executive is let iepartments the“ those departmen' volley instruct: from either actl Egreat extent, 5031 of obtaini: Together, isPatients bel ac”ctive attribu Whey than they hanCTiOnS . The “mil independ 2A2 expansionary requests. Department heads do not believe they are freeto seek whatever additional funds they may want without experiencing some kinds of disadvantage. lkmwver, the executive himself (see table 7.2), does not report as strong an incidence of control. Only to some extent is there agreement with the statement as orig- inally worded, as only some sanctions are imposed upon recalcitrant departments. The second question specifies the exact nature of the costs involved in seeking more, in terms of actually obtaining expenditure increases: "If departments request too large an increase, they harm their chances of re- ceiving increases at all." The sanction imposed by the executive is larger cuts and smaller increases to those departments that do not follow budget guidance than to those departments that adhere to the intent of executive pmlicy instructions. But there is only slight agreement, from either actor that such.penalities are imposed. To a great extent, department heads do not threaten their goal of obtaining more by explicitly asking for it. Together, these two questions reveal that while the departments believe they possess some leeway. The ex- ecutive attributes broad leeway to them to ask for more money than they received in the past without suffering Sanctions. The distribution.of responses is in the upper end of the scale. from some to very great depart- mental independence from executive control. This 243 TABLE 7.2 DEPARTMENTAL LEEWAY (EXECUTIVE REPLIES) City Nothing to Chances for Lose Increases Mean 01 4.00 0.00-4.00 4.00 02 2.00 0.33-3.66 2.83 03 4.00 0.50-3.50 3.75 on 2.00 1.00—3.00 2.50 05 2.00 0.33-3.66 2.83 06 1.66 0.00-4.00 2.83 07 2.00 0.50—3.50 2,75 08 2.50 1.50—2.50 2.50 09 2.00 0.66-3.33 2.66 10 0.50 2.00-2.00 1,25 11 2.00 1.50-2.50 2.25 12 1050 0050-3050 3050 13 3.25 0.00—4.00 3.62 14 3.00 1.00—3.00 3.00 Mean 2.32 0.70—3.30 2.81 St. Dev. 0.96 0.62—0.60 0.66 situation would 21 percent incr toCrecine’s st close executive airing is far I“ :ities, departm on decision ma spending reques Rel Even thong siderable leewa exact relations. confirms the ex dame-+0.10) perceive they a Sitting expansi funds they will M- \ 11. City All” of the err ‘10: and if it edit. the corre 103905 for thi articular case cent 13%) yet 18 That ’ . % there 1 “en lost. It 244 situation would have have been inferred from the average 21 percent increase submitted by the departments compared to Crecine's standard of 10 percent. It appears that close executive supervision of departmental decision making is far from the norm. In the present fourteen cities, departments are relatively autonomous in their own decision making and the formulation of expansionary spending requests. Relationship to Department Budget Requests Even though on the whole departments possess con- siderable leeway in seeking expenditure growth, its exact relationship to the size of increases asked for confirms the expected positive connection between them (0.3912 + 0.10).11 The more department heads in each city perceive they are free of executive sanctions for sub- mitting expansionary budgets, the larger an increase in funds they will indeed request. If they believe there is 11. City number eight by itself contributes one half of the error term in the computation of the statis- tics and if it were removed from the analysis as a deviant case, the correlation would increase in strength (0.7012 + 0.005 for thirteen cities). An explanation for this particular case, where departments perceive high leeway (rank 13%), yet submit low requests for increases (rank 1) 18 that there is nothing to lose, because all has already been lost. It will be recalled that in this city, depart- ment expenditure requests were below the absolute level of the previous year. So, in the view of the department heads, there is nothing left, as everything has previously been taken away, and they are no longer apprehensive of executive imposed sanctions. nothing to lose have, they will growth. If, on advantage will asking for incr Rufus Bro agencies percep upon the formul :2” the process: . . . for may get th refers to requests. simple rej enough to , first place are involve ridicule . quests, and damage to t is described by amount or" politi 3'1? the requests ESSStion and de< “it-ii it to hir Given the < 2% nothing to lose by asking for more than they already imve, they will go ahead and submit requests for spending growth. If, on the other hand, they believe that a dis- advantage will result, they will be constrained in asking for increases. Rufus Browning provides an illustration of how the agencies perception of costs, operates as a constraint upon the formulation of expansionary budget. He writes of the process: . . . for eliminating or modifying requests which may get the agency into trouble . . . (which) refers to political penalties for making certain requests. For some agencies and agency heads, simple rejection of requests may be painful enough to prevent its making the request in the first place, but usually some other penalties are involved-— public criticism, insult, and ridicule . . . loss of support for other re- quests, and loss of confidence from others with damage to future requests and to career chances.12 As described by one department head: "I have a certain emmunt of political capital to use up. I carefully weigh out the requests I present to the manager that would question and decide whether or not it is worthwhile to Submit it to him." Given the opportunity, department heads will act out their own advocacy-spending role into actual dollar and 12. Rufus P. Browning, "Innovative and Non-innovative Decision Process in Governmental Budgeting," in Ira Shar- kansky. Policy Analysis in Political Science (Chicago: Nhrkham Press, 1970I. P- 309' cents requests l 1110 see]: to exp: to do so. They program and sper the ability of t the department 1 the outputs of ' process. Exeo Change The execut: departmental de. hitial budget : 0m influence 0‘ the imposes hi] Sequence of bud, indeed his own. 0f the budget h tents through attended the so 0f the formal to is now, not the department recognizes the influence of t eXel‘cise great Increases, the 246 cents requests (0.4532 + 0.10). Departments in cities who seek to expand their appropriations have the chance to do so. They can freely translate their expansionary program and spending inclinations into concrete behavior. The ability of the executive to limit the independence the department has direct and immediate consequences upon the outputs of the first stage of the municipal budget process. Executive Budget Behavior Change in Departmental Requests The executive's ability to limit the leeway of departmental decision making and constrain the size of initial budget requests is an important component of his own influence over final spending outputs. If the execu- tive imposes himself from the very beginning of the sequence of budget stages, final expenditure totals are indeed his own. He is able to affect the composition of the budget he receives as well as influencing depart- ments through his review of their requests. He has extended the scope of his influence, beyond the confines of the formal authority of his own decision making stage. He is now, not merely responding to the initiatives of the departments. but shaping them as well. Allen Schick recognizes the implications of this feature for the influence of the executive when he writes that: "To exercise greater control over the pace of expenditure increases, the governor would have to step in earlier, before the initi Otherwise, the rate of peat department reque this is indeed p leeway departmez quests are redue expansion in the partments' intei is compelled to this end by wt Mint in the se to utilize his . the expansionar: aKain writes of the executive b1 . . . did , taining ag have been and conseq alternat iv Execut However. t reassertion of m...— 13. Allen 11;. Aller. 247 before the initial agency decisions were made."13 Otherwise, the only way for the executive to limit the rate of yearly growth in spending is to reduce department requests at the time of the formal review. This is indeed what happens (0.6835 + 0.005). The more leeway departments possess, the more their budget re— quests are reduced. Failing to limit the pace of expansion in the first place, by controlling the de- partments' internal budget making process, the executive is compelled to adopt an alternative route to achieve this end by cutting departmental requests at this latter point in the sequence of the budget process. He has to utilize his formal authority to cut in order to contain the expansionary thrust of departments. As Schick once again writes of this situation and its implications for the executive budget cutting position, the executive: . . . did not consider the possibility of con— taining agency budget pressures before they have been allowed expression in the estimates, and consequently he could not conceive of an alternative to the budget-cutting role.lu Executive Budget Recommendations However. this budget cutting does not represent the reassertion of executive control, but actually the lack 13. Allen Schick, Budget Innovation. p. 179. 14. Allen Schick, Budget Innovation, p. 175. 4 4 LEGISLATURE HEADS FIGURE 8.1 THE STRUCTURE OF BUDGETARY INFLUENCE of the pyrami< to the decisie lature is a r1 spending prefv the final app: Although formal influe‘ the two arms executive is betteen the e rents and his eFiecutive cen correlation r 0V9? the firs CiSion making i‘iOWeVerI soon a Consis imminent he to the eXecut not assert hi tive SUbOI‘dip OWh Spending 271 of the pyramid. Department heads are passive onlookers to the decisions made on their requests and the legis- lature is a rubber stamp to the executive. Executive spending preferences dominate and are incorporated into the final appropriation ordinance. Although chapter five individually examined these formal influence relationships, the connection between the two arms of this triangle that centers around the executive is now analyzed; the statistical association between the executive's interactions with the depart- ments and his interactions with the legislature. An executive centered process would be revealed in a positive correlation between them. The executive is dominant over the first stage of the sequence of budgetary de- cision making as he is dominant over the last stage. However, the actual empirical data does not corroborate such a consistent structure of executive authority (0.0098).5 Department heads and the legislature are not subordinate to the executive within the same city. The executive does not assert his hierarchical authority over his administra- tive subordinates while the legislature defers to his OWn Spending judgments. Such an omnipotent central stage 5. This correlation is derived from the perspective of the department heads and the legislatures description Of their relationship with the executive. It is -0.0822 from the executive's own description of his relationship with the department heads and the legislature. does not seem aoompletell’ correlation) - checked by th oepiible to d sometrical a Instead of formal inf dent of each the executive executive's 1‘ submit to the sacrificing h is Open to th SEOpardiZing 15311 insulat one decision Pattern was p it was reveal to make Spend review does n ChOiQeS 0f th budget influe A Second lOoks at the l o“ enISIature , S I : V 272 does not seemingly exist. But it should be noted that a completely decentralized system (evident by a negative correlation) also does not exist. The executive is not checked by the municipal legislature while being sus- ceptible to departmental pressure. Influence is not as symmetrical as others have pictured. Instead it seems that the two sides of the structure of formal influence do not go together. but are indepen- dent of each other. The distribution of influence between the executive and the department does not affect the executive's relationship with the legislature. He can submit to the budget authority of the legislature without sacrificing his control over the departments; or he can be open to the persuasion of the departments without jeopardizing his superiority over the legislature. There is an insulation of the exercise of formal authority from one decision making stage to the next. Such a segmented pattern was previously indicated in Chapter Five, where it was revealed that the possession of formal authority to make spending decisions at the time of the official review does not extend backward to affect the spending Choices of the preceding stage. The scope of formal budget influence is more restricted. A second component of the structure of formal authority looks at the connection between the departments and the legislature's relationship with the executive to their on. direct 00 this dimensio of the triang only one who the legislatu prevent them his ovm budge Such an evident in a eiements. in rents vis-a-v can go to the sionary budge to withstand their initial them from app assertive leg 05 the direct Change is als relitionship or; . . u‘qo _ *- ptlns the EII:f_________________________________4 273 own direct contact and communication. In other words, this dimension examines the association between the arms of the triangle and its base. A strong executive is not only one who is dominant over the department heads and the legislature individually, but is one who is able to prevent them from forming a coalition that would diminish his own budgetary authority. Such an executive centered structure would be evident in a positive correlation between these two elements. And this is indeed the case. Strong depart- ments vis-a-vis the executive are also departments who can go to the legislature on behalf of their own expan- sionary budgets (0.h396 - 0.10).6 The executive's ability to withstand the pressure of departments as he reviews their initial budgets accompanies his ability to prevent them from appealing to the legislature. Similarly an assertive legislature, who makes an independent evaluation of the direction and magnitude of annual expenditure change is also a legislature that has an independent relationship with the department heads in the course of adopting the appropriation ordinance (0.4099 + 0.10).7 6. The association is 0.6638 + 0.001 from the report of the executive as he describes his influence over the departments and his perceptions of the department inter— actions with the legislature. 7. From the replies of the executive it is 0.3498, 0f the connection between his influence with the legislature and their contact and communication with the departments. Accepting the along with a restoration 0 The last authority com influence int over the muni than his indi as it include actions betwe directly. Th related to it cents formal irfluence wit formal influe: with the depa: This int The less the I the executive to the legisl: ‘ | ] N ‘9' slSlature at 274 Accepting the executive's spending recommendations goes along with a rejection of the department's appeal for a restoration of previous cuts. The last dimension of the structure of budgetary authority combines the two separate elements of formal influence into a single index. The executive's dominanCe over the municipal budgetary process derives from more than his individual authority over these two other actors, as it includes considerations of his control over the inter- actions between the departments and the legislature directly. The two arms of the influence triangle are now related to its base; the connection between the depart- ments formal influence with the executive plus their influence with the legislatures and the legislatures formal influence with the executive plus their contact with the departments. This interpretation is also supported by the data. The less the departments possess the ability to persuade the executive, the less they have the opportunity to go to the legislature at the same time that the less the legislature asserts their authority over the executive and the less they establish autonomous contact with the dePartment heads (0.5393 + 0.05).8 8. This is from the viewpoint of the department twads and the legislature; but it is also exactly 0.5393 from the executives' description of his influence relation- Ship with each of them, plus his perception of the interactions between them. Compared and direct for department hea it would seem pattern of but structure of i tralization. andnodified ; CO-YlOSite ind: Within the nu: Vaiiaill‘e of e the Previous idiluence is is not as C011 EPincers, Departn larger increg hfih the eXeC if Their Own departmem he result of the +the executiVo 275 Compared to the previous analysis of the separate and direct formal influence relationship between the department heads and the legislature with the executive, it would seem that these two subsequent measures of the pattern of budgetary authority point to a more consistent structure of variation in the degree of executive cen- tralization. However, this conclusion has to be withheld and modified as a result of the association between his composite index of the distribution of formal influence within the municipal budgetary system and the dependent variable of expenditure outputs. By such an analysis, the previous discription of a compartmentalization of influence is supported. The distribution of influence is not as concentrated in the hands of the executive as it appears. Department heads do not submit requests for any larger increases, on the basis of their combined influence with the executive (0.1938), than they do as a consequence of their own direct influence with him (0.1615). The department heads are not motivated to ask for more as a result of their ability to both bargain and negotiate with the executive and then go to the legislature when they are not satisfied with that compromise set of expendi- ture figures as compared to the situation where they do not perceive the possibility of appealing executive decisions. Nor does the executive reduce departmental requeStS any expected)! ‘ti actions alone executive rec' such an appea strained in C possibility C decision maki departments 8 to obtain for interactions While tl interactions influence wit of formal buc‘ ability to cl eXpected. Tl bifurcated th to appeal to add lity to p5 276 requests any less (0.4154, when a negative correlation is expected), than he does onthe basis of their inter- actions alone (0.2275).9 In fact, it seems that the executive reduces departments more, when he estimates that such an appeal will be made. He is certainly not con- strained in cutting their expansionary budgets by the possibility of such departmental activity in the third decision making stage. Finally, it even appears that departments actually lose the increases they were able to obtain for themselves (0.3703) when including their interactions with the legislature (0.1978).lO While the extent and character of the departments interactions with the legislature is a component of their influence with the executive, which describes the structure of formal budget authority, it does not enhance their ability to change actual executive spending behavior as expected. The structure of budgetary influence is more bifurcated than first indicated. The departments ability to appeal to the legislature is an extension of their ability to persuade the executive, it does not seem to 9. From the replies of the department heads, the correlation is 0.1938 between this combined measure of the pattern of formal influence and the size of executive cuts in their spending. 10. From the replies of the department head, the correlation is 0.3325 between this combined index of the structure of budgetary authority and the amount of in— creases the executive recommends for them to the legislature. be able to u" strategy to E The eXPi structure to departments 1 lature is m departments ; goals of the rnicipal lei against the . conflicting ' want an incr wants to hol alliance, th further the not to do. their abilit the executiv will appeal SW1 a threa WOil-1d more 1 ihissed by t This bi FGVealed in humid and That is! the EII:T__________________________________i 277 be able to utilize their influence as a resource and strategy to affect executive spending choices. The explanation for the failure of such an influence structure to enhance the spending objectives of the departments is readily apparent. The municipal legis- lature is not any more amenable to the entreaties of departments in behalf of their expansionary expenditure goals of the departments than is the executive. The municipal legislature is not allied with the departments against the executive because they have different and conflicting spending preferences. While the departments want an increase in expenditure levels the legislature wants to hold the line on annual growth. To form an alliance, the legislature would have to endorse and further the objectives of expansion, and this they prefer not to do. So department heads cannot really employ their ability to contact the legislature as a threat to the executive not to cut their requests, or else they will appeal to the legislature to get back what was lost. Such a threat is empty, and to actually carry it out would more likely result in a further reduction being imposed by the legislature, than other outcomes. This bifurcated pattern of influence is further revealed in the analysis of the other side of this pyramid and its impact on legislative spending choices. That is, the bifurcated influence pattern is revealed in the connecti< strength (the their intera: decisions the As in the pr: exist as a d: it comes to ; it does not a the legislatt NOW, wi to their for] tical associ; decisions ar even Complet iflflUEnCe an “mm 0.5701 3310th of re growth of th iselimihate The eXp decisiOn mak \ Th 3’ t 11. Sdpplied b s 12. 0* ”Wig: , _ 13. 01 the Well: 278 the connection of the measure of the legislature's strength (their own influence with the executive plus their interactions with the department heads) to the decisions they make in reviewing executive recommendations. As in the previous case, while such a connection does exist as a description of the process of budgeting, when it comes to its impact upon concrete expenditure outputs, it does not appear to enhance the influence position of the legislature. Now, with the addition of the departmental component to their formal authority over the executive, the statis- tical association to the three measures of legislative decisions are either greatly weakened in strength or are even completely obliterated. The connection between such influence and the absolute amount of change is lessened (from 0.5701 to 0.3599),ll as both the association to the amount of reductions imposed (0.3891 to 0.0978)l2 and the growth of the final appropriations (—o.2301 to —o.0219)13 is eliminated. The explanation for this pattern of legislative decision making is also readily apparent and similar to 11. The correlation is +0.6187 from the description SuPplied by the executive. 12. The correlation is -0.0120 from the viewpoint 0f executive replies. 13. The correlation is 0.1395 from the perspective of the executive. the one offel with the exec in appealing apouerful do powerful leg: influence as legislature. ounicipal but spending objl The dep legislative cribed earl i 1‘19 departme 279 the one offered of the department heads interactions with the executive. The department heads are successful in appealing executive decisions, as it requires both a powerful department head to make such an appeal and a powerful legislature to accept it. But the department influence actually works to the disadvantage of the legislature. It weakens their actual position within the municipal budget process, because they have opposing Spending objectives. The departments are able to offset and neutralize legislative opposition to expenditure growth, as des- cribed earlier in Chapter Five. Again it is evident that the departments are not allied with the legislature against the executive, but instead allied with the executive against the legislature. In spite of the budget cuts the executive makes, in all cities where departments asked for more, they received it at the end of the second stage. The same does not hold true for the final outputs of the legislature's review. In three cities where the executive recommended annual spending growth, the legislature completely eliminated it in the appropriations ordinance. Consequently, once a set of expenditure figures emerges from the bargaining process of executive decision making, the department heads are motivated to defend these amounts as the best they can hope to obtain in the face of further reductions by the city council legislature : recommendati it as submit tion of form. The sec budgetary in snug}. anti Help, 1“) Pronour. —>— 280 city council. They utilize their influence with the legislature not to undercut the position of the executive but to reinforce it. They support the executive's recommendations and appeal to the legislature to accept it as submitted. The segmented and bifurcated distribu— tion of formal budget authority is fully apparent. Informal Influence The second element of the structure of municipal budgetary influence is exercised through informal means. Through anticipated reactions and behind the scenes warnings, the preferences of the more powerful actor are incorporated into the spending proposals of the less powerful one. This feature of budgeting, while not a very pronounced part of the pattern of influence, never— theless does exist and does have an impact upon expenditure outputs as it is exercised by the executive over the department heads and by the legislature over the executive. In an executive centered process informal influence is exercised over the departments at the same time that he is free of a similar application of informal influence on the part of the legislature. The executive is dom- inant as he asserts control over the departments while he himself is autonomous of the city council. However, this symmetry of an executive centered process is not corroborated by the data. The presence of informal legislative influence over the executive is not translated 1: executive ' s (h3989,‘whe' Theexecutiv icing his 0 isnottotal totally weak asit was ju authority is Certainly in actions, he The bit —>— 281 translated into an absence of a similar exercise of the executive's informal influence over the department heads (0.3989, when a negative correlation is expected).14 The executive defers to the legislature without sacri- ficing his control over the departments. The executive is not totally dominant, but at the same time he is neither totally weak. The structure of informal budget influence, as it was just revealed in the analysis of formal authority is more compartmentalized than it is concen- trated in the hands of the executive. While, he is most certainly in the middle of a triangular set of inter- actions, he is not at the top of a pyramid. The bifurcation of the structure of municipal budget influence is further revealed in the analysis of the association between the "Two Faces of Power." That is, the connection between the possession of formal and informal influence over the budget process; first between the executive and the departments and between the legis- lature and the executive. In terms of the executive's control over the depart- ments, influence is concentrated in the executive. This is evident in each of the three aspects of the structure 14. This is derived from the executive's description Of his relationship to the departments and the legislature. From their respective vantage point it is 0.0875. of formal in exercises hi their budget his informal proposals th in terms of those of the able to pers fioial meeti too preventi This pa informal ini authority. control over are able to mental respc Then there :' index of the influence (. from those c his influenc 01‘ the munio informal me: the first 8' the Second . as Well as —>— 282 of formal influence. First of all the more the executive exercises his authority over the departments to review their budgets during the second stage, the more he exercises his informal influence over them to shape the budget proposals that are initially submitted to him (-O.2000 in terms of departmental replies and -0.l934 in terms of those of the executive). The less the department heads are able to persuade the executive at the time of their of- ficial meeting, the more they are also subject to the two preventive mechanisms of informal influence. This pattern is emphasized by the connection between informal influence and the second measure of formal authority. Thus the more the executive displays informal control over the departments, the less the departments are able to go to the legislature (-0.4088 from depart- mental reSponses and -O.3736 from those of the executive). Then there is a last negative relationship to the combined index of the two elements of formal authority to informal influence {-0.3509 from departmental responses and -O.4041 from those of the executive). A strong executive displays his influence over the departments throughout the sequence of the municipal budget process. Through both formal and informal means the executive dominates the departments in the first stage of the preparation of their initial requests, the second stage when he officially reviews their requests as well as in the third stage when the legislature adopts the final 8P However legislature ' executive, ’8 each other. mechanisms f influence wi they are alt the more the to modify e) their reviev fluence over is submitte< These 2 assert thei: ashieves th Preferences Where the c Submits his eXPeImi'lture fitures. T Set of Cons mendatiOnS . \ g 150 seen the i 283 the final appropriation ordinance. However, in the case of the application of the legislature‘s formal and informal influence over the executive, the "Two Faces of Power" do not accompany each other. They do not operate together as reinforcing mechanisms for a strong legislature to assert their influence within the municipal budget process. Instead they are alternative means for influencing the executive. The more the city council exerts their formal authority to modify expected budget recommendations at the time of their review, the less they exercise their informal in- fluence over executive decisions to chape the budget that is submitted to them (-o.5ooo + 0.05).15 These are two different means for the legislature to assert their budget making influence. Each by itself, achieves the same purpose of incorporating their spending preferences within the final budget ordinance. One is where the city council waits until the executive formally submits his spending recommendations to express their expenditure priorities by explicitly altering spending figures. The alternative is to act earlier by devising a set of constraints on the preparation of those recom- mendations. The application of one form of influence makes 15. The executive description of the connection be- tween the legislature's informal and formal exercise of influence is 0.0950. unnecessary of the execu recommendati only be elim has already have already is less need Nowhere ”where the is exactly as s leéiSlative highest Scor lature itsel 11': and 13). does not mes ation OI" mur t0 a dominar informal inf formal authc seen Worm This p; by the “ess-at —— 284 unnecessary the other. Changing the budget behavior of the executive not to officially forward budget recommendations that they do not approve and that would only be eliminated later, means that the legislature has already exercised their power of budget veto. They have already effected final expenditure outputs, so there is less need to act during the public review. Nowhere is this more evident than in the three cities where the legislature accepted the executive's budget exactly as submitted. According to both executive and legislative replies legislature exercise two of the highest scores (numbers 12, 8, and 14) and by the legis- lature itself, also two of the highest ranks (numbers 5, l4, and 13). The absence of change in expenditure levels does not mean that they are unimportant in the determin- ation of municipal spending outputs. They are not subordinate to a dominant executive, because by the exercise of their informal influence they do not have to activate their formal authority. Their spending preferences have already been incorporated into the executive's recommendations. This pattern of legislative influence is corroborated by the negative association between informal influence and their contact with the department heads (-O.2592).l6 16. The correlation is 0.4265 of the association between the legislature's informal influence and their contact with the department heads from the perspective of the executive. The more imp over the exe mating syste with the day is authorite for the depz decisions he :uta‘ole to . already bee; in the detei that remain 285 The more important the legislature's informal influence over the executive to their position within the budget making system, the less significant in their interactions with the departments at the time the appropriation ordinance is authoritatively adopted. By this time, it is too late for the departments to influence the city council, because decisions have already been made and are essentially im- mutable to change. Legislative spending choices have already been established and their influence already felt in the determination of the final budget ordinance. All that remains at this point is to officially, and somewhat symbolically approve the appropriation ordinance. This is further evident in the negative connection between the legislature's informal influence and the combined measure of their formal authority. The correlation is also nega- tive (4.4090 + 0.10).17 It is unmistakably evident that the exercise of formal budget authority is not a complete picture of the structure of municipal budget influence. The legislature does not perceive the need, as does the executive, to exercise both formal and informal influence. While they certainly describe a different process of budgeting, either one by itself is sufficient to achieve legislative 17. The correlation is -O.1308 from the executive's description of this index. spending 505 concrete defi end inaccurz possesses w: recommendat out that th and in othe within the decisions c' and are on: of municipe between the and varied 286 spending goals. Consequently, relying solely upon concrete decisions made at the time of the formal budget review in the published budget document is both insufficient and inaccurate of the amount of influence the city council possesses within the municipal budgetary process. The absence of activity does not mean that the municipal legislature is powerless. Accepting the executive recommendations, does not mean they are a rubber stamp, but that their influence has been utilized in other ways and in other times to incorporate their spending priorities within the budget they first receive for review. Official decisions describe only part of the spending choices made and are only a part of the influence exerted in the process 18 As the influence relationship of municipal budgeting. between the legislature and the executive is more complex and varied than first thought. Executive Supervision of Departmental Decision Making The last element of the structure of municipal budget influence pinpoints the executive's supervision of the 18. Since the impact of formal and informal influence upon actual spending Outputs operates in contradictory directions; that is, formal authority results in greater changes being made and informal influence brings forth less changes combining them into a single index of budget influence would display little effect upon spending choices. They would counteract each other and produce no apparent empirical relationship to the direction and amount of Spending changes in each of the dependent variables. first stage limiting th heads in fo the amount intends to requests, a sequence of executive v tween his 1 heads and 1 01" initial Howev‘ tw0 feature not appear Structure In the con Various me is 9V1 dent Supervisi C reQUESts C authority 1neplieg a: lack of e: mean that TGViewS i] SuperVis ‘1. 287 first stage of departmental decision making. Through limiting the amount of leeway enjoyed by the department heads in formulating their requests and through limiting the amount of interdepartmental competition, the executive intends to inhibit the formulation of expansionary spending requests, and assert his own influence throughout the sequence of budget stages. Consequently, a dominant executive would be revealed by a positive connection be— tween his formal and informal influence over the department heads and his particular supervision over the preparation of initial budget requests. However, utilizing the combined measure of these two features of this supervisory relationship, this does not appear to be the case. As described earlier. the structure of municipal budgetary influence is segmented. In the connection between executive supervision and the various measures of the exercise of formal authority it is evident that they do not go together. Executive supervision of the departments' preparation of their requests does not accompany the exercise of his hierarchical authority over them. (We have -0.0l§4 from departmental replies and —0.0740 from those of the executive). The lack of executive COntrol over the first stage does not mean that he is weak in their direct interaction as he reviews initial budget requests. Nor does the lack of Supervision imply a weakness on the part of the executive :hat is trai heads to apj sponses of ‘ theexecuti' kneen thi influence a aver depart departments The pa 288 that is translated into the ability of the department heads to appeal to the legislature (-0.2#60 from the re— sponses of the department heads and 0.0000 from those of the executive). And finally, there is no connection between this particular aspect of executive budget influence and the combined index of his formal authority over departmental decision making (-0.3140 according to departments and 0.0398 from those of the executive). The pattern of these correlations clearly indicates that a strong executive, who exercises his formal authority over the departments in the budget review of the second and third decision making stages, does not also exercise control over the departments in the initial preparation of spending requests. Departments formulate their budget free of executive interference, regardless of their ability to influence other decision makers once their requests are submitted for review. As we can see, a set of executive imposed constraints, intending to direct the size of depart- mental budget requests does not necessarily accompany the exercise of formal authority over the departments. Instead, the general consistency of the negative direction of the correlations (especially in terms of departmental responses), although weak in strength, Suggests that they are alternative mechanisms of executive control. That is, the absence of initial supervision Of the process by which the departments prepare their initial reg? ence their ' Having the epartments fluence sub iepartments :ere overt if departre NG distim 1luence . The c 289 initial requests is accompanied by passive departments once their budgets have been forwarded to the executive. Having the autonomy to formulate their requests, the departments do not further possess the ability to in- fluence subsequent budget decisions. They do not have the opportunity to persuade the executive or appeal to the legislature. They are able to ask for what they want, but not the ability to "fight" for them once they are formally proposed. The executive can allow the departments the independence without jeopardizing his more overt influence upon the budget. The supervision of departmental decision making is apparently a separate and distinct feature of the structure of budget making influence. The connection between the executive's Supervision of departmental decision making and his informal influence over them is more difficult to interpret, as there is a significant discrepancy in this correlation from the point of View of the department heads and the executive. Up to this point, the same interpretation could be made of the structure of municipal budget influence from the perspective of either participant to the interaction. But in the present case, this is not true. In terms of departmental perspective, there is a strong positive association between the executive exercise of informal influence and his supervision over the preparation of their initi vantage p013 themselves pmssess lee m prefers complete is executive f perceive ti of I H hepartme :ermal inf Z The s is more 001 Previously imminent 0‘ lature. 3: I lh° we ‘UEhCe Se‘s‘hemted 290 their initial requests (0.5934 + 0.05). But from the vantage point of the executive an equally strong, nega— tive relationship is evident (-O.509O + 0.05). Why such a striking difference exists is unknown, except that logical interpretations of the structure of municipal budget influence could be made of either of these two different patterns. Department heads, when they are free of direct executive supervision, do not perceive themselves free of informal influence. Even when they possess leeway and the absence of evaluations to min— imize competition they do not perceive themselves free of anticipated reactions and behind the scenes contact and preferences, and do not prepare their budgets in complete isolation from the exercise of some form of executive influence. However, the executive does not perceive this to be the case. He sees, the supervision of department as a component of his more explicit in- formal influence over them. Conclusion The structure of municipal budgetary influence is more complex, more varied, more subtle than has previously been portrayed. The executive is not simply dominant over both the department heads and the legis- lature. Breaking down the concept of interpersonal influence into three separate elements reveals a more segmented pattern. The budgetary system is neither completely nor totally participant heads, and It woc compartment making stag process of 291 completely centralized in the hands of the executive, nor totally dispersed among all three authoritative participants. It is evident that both the department heads, and the legislature possess considerable influence. It would appear that the structure of influence is compartmentalized, into two, and not the three decision making stages as originally described. The first is the process of budget preparation and the second is that of budget adoption. Budget preparation involves the interactions between the department heads and the executive as initial department spending requests are reviewed by the executive. The result of these decisions is the presentation of the "administrative" budget to the legislature as both the department heads and the execu— tive come to agree upon a single set of expenditure figures. The adoption of the final appropriations ordinance by the legislature is the process of decision making as the legislature reviews the budget submitted to them. It involves the interactions with both the department heads and the executive. The process of budget preparation is divorced and insulated from the subsequent decisions of the city council, as is the adoption of budget ordinance isolated and separated from decisions that went before. The departments heads and the executive are in an alliance vis-a-vis the spending preferences of the legislature. She executi spending le cipal legis support the in. support act indeper the executi can achieve 292 The executive is more inclines to support the expansion of spending levels, albeit a modest one, than is the muni- cipal legislature. So, the department heads assist and support the executive in his dealings with the legislature in support of whatever increases is recommended. They act independently, but not to undercut the position of the executive, but in unison with him, as the only way they can achieve their goals of spending growth. '. \T I hatter n H The n: process is choice empj among vari ferent app; Examined u answers t‘n TWO m identified ”it” ape lie i Chapter Nine: Cognitive and Evaluative Mechanisms of Choice Introduction The next component of the municipal budgetary process is the cognitive and evaluative mechanism of choice employed in selecting one course of action from among various alternatives. This is an entirely dif- ferent approach to the decision making process, than examined up to this point as it looks at why each actor answers the particular budget problem he is faced with. Two models of this decision making process are identified: the "synoptic" and the "incremental." The former approach is composed of the following elements: 1) A comprehensive overview of factors relevant to a decision; 2) Clarity of definition of social objectives; 3) A means-end approach to policy; 4) Deliberate and explicit choice among policies; 5) A calculation and minimization of costs; 6) Reason and cooperation rather than arbitrariness, coercion and conflict; and 7) A unified decisiOn making process for decisions Ir. this 83” aproblem. i then ranks text, a com sible alter asi-milar e native, the :‘ejective :' hrthp decision he the formula step is the process the 13 f0 llowe: “hip. . runs that Mr: n DPOgr: . . 3.3 T 294 that are highly interdependent.l In this synoptic model, the decision maker when faced with a problem first establishes the goals to be achieved and then ranks them according to their relative importance. Next, a comprehensive and exhaustive search for all pos- sible alternatives to obtain them is undertaken. After a similar evaluation of the consequences of each alter- native, the single course of action that maximizes the objective is selected. Such a series of steps emphasizes the rationality of problem solving. Systematic analysis identifies specific means to agreed upon ends. Arthur Smithies shows how this rational-comprehensive decision making model is specifically translated into the formulation and adoption of the budget. The first step is the "determination of policy objectives" by a process that is essentially outside of the arena. This is fOllowed by "planning-- the preparation of alternative 2 Plans that will further particular policy objectives . . . and "PrOgramming-- the selection among alternative plans . . ."3 This last component is most crucial. for it is k 1. Charles Lindbloom, "Decision Making and Taxation and Expenditures," in Public Finances: Needs, Sources, and Utilization: A Conference, National Bureau of Economic Research (Princeton: Princeton UniverSity Press, 1961), p. 298. 2. Arthur Smithies, The Budget Process in the United States (New York: McGraw Hill and Company, 1968), p. 23. 3. Arthur Smithies, The Budget Process, p. 23. here that 1 and cents : The : the j ture: of a comp cisi of t The budget. execution Smit that are i expenditur sees and c 0: the Sign lo 295 here that plans and programs are represented in dollars and cents figures: The formulation of the budget is an extension of the programming stage. Programs become expendi- tures programs and, thus, are expressed in terms of a common denominator, money. They can be compared with each other . . . with greater pre- cision than is feasible in the earlier stages of the process. The budgetary process is then logically completed with the execution and the audit stages. Smithies then goes on to present the principles that are intended to guide the process of making specific eXpenditure choices. These clearly emphasize the explicit- ness and comprehensiveness of the means-end calculation of the synoptic decision making approach. They are: l. Governmental objectives chould be as clearly and explicitly defined as poss1ble; 2. Alternative policies should be explicitly regarded as alternative means toward the achievement of objectives; 3. Specifically, expenditure decisions should be made explicitly and deliberately in the light of all the objectives they are intended to achieve; 4. In the interests of a rational comparison of alternatives, final expenditures deCiSions should not be made until all claims on the budget can be considered; 5. Revenue and expenditure decisions should be deliberately coordinated; 6. For each expenditure, some systematic and deliberate appraisal of benefits and costs should be made; and 4. Arthur Smithies, The Budget Process, p. 24. Howe decision m government of how spe ficials ac follow the process. h‘hiCil is m ’3" the pro than capa all conoei ‘9 CODSeq Edi only e belord the Capacities Emblem of Values is {Blues in :1 3Wires 296 7. Policy making, including budgetary policy making, should achieve a unified policy.5 However, this rational-comprehensive model of decision making is less an empirical description of governmental budgeting than it is a normative prescription of how spending decisions should be made. Public of- ficials according to past empirical research do not follow the tenets of such a cognitive and evaluative process. Instead they adhere to an "incremental" approach, which is more suited to the complexity and uncertainty of the problem solving situation. The limitations of human capabilities precludes a comprehensive search for all conceivable alternative means to achieve a given goal. At the same time it precludes an exhaustive analysis of the consequences of alternative policies. Such an effort not only entails a cost in time and energy, but are beyond the limits of available information and intellectual capacities of both man and machine. Furthermore, the problem of first establishing and then comparing social values is often insurmountable. There is always multiple values in any policy and conflict among goals that make an abstract "social welfare function" almost impossible to attain. The separation of means and ends is an artificial distinction as one is established through 5. Charles Lindbloom, "Decision Making," pp. 297-298. the other. by availab Cons the increm have been procedures identifica consequenc one specif These mini smell numb situation. is charact l. 5. EI::_________________________________* 297 the other. The objective to be achieved is often shaped by available means. Consequently, the synoptic model is replaced by the incremental one. A set of "aids to calculation" have been established that serve as standard operational procedures, that provide a set of strategic steps in the identification of alternatives, the analysis of their consequences, and the criteria for the selection of one specific policy from among a set of alternatives. These minimize debate over values, and identify only a Small number of variables to be considered in every choice situation. As summarized by Charles Lindbloom, this model is characterized by the following components: 1. Only that limited set of policy alternatives that are politically relevant, these typically being policies only incrementally different from existing policies; 2. Analysis of only those aspects of policies with respect to which the alternatives differ; 3. A view of the policy choice one as in a suc- cession of choices; 4. The marginal [italics in original7 values of various social objectives and constraints; 5. An intermixture of evaluation and empirical analysis rather than an empirical analysis of the consequences of policies for objectives independently determined; and 6. Only a small number out of all the important relevant values.6 6. Charles Lindbloom, "Decision Making," p. 306, The decisic and exhauS‘ course of 2 functional limited nu: :he margin: serial, an: Thee I__——’ 298 The decision maker does not undertake a comprehensive and exhaustive search and evaluation of every alternative course of action, but instead simply supplies the in- formational and analytical requirements to only those limited number of alternatives and to only those limited number of consequences that differ in relatively small degree from the present policy. Decisions are made on the margins of existing state of affairs and are partial, serial, and remedial adjustments to current activities. These two different approaches to decision making and their direct relevance to the budgetary system has been perceptively analyzed in the writings of Allen Schick, who has categorized them into three budget orientations: control, management, and planning. Each of these are marked by a distinctive cognitive framework and a distinct standard of evaluation that represents the incremental and synoptic decision making models. These are not examined as they are present in municipal government and then affect expenditure outputs. Control The control orientation is the traditional mode Of governmental budgeting and represents the incremental model of decision making. As defined by Schick, it is, 'H . . the process of enforcing the limitations and conditions set in the budget and in appropriations, and of securing imposed by following . How cell the cedu ine ona execution COElements . the“, :TO” 299 of securing compliance with the spending restrictions 7 imposed by central authorities." It focuses upon the following issues: How can agencies be held to the expenditure ceilings established by the legislature and the chief executive? What reporting pro- cedures should be used to enforce propriety in expenditures? What limits should be placed on agency spending for personnel and equipment?8 While it is explicitly described as pertaining to the execution and audit stages of budgeting, its essential components are just as applicable to the adoption of the appropriations ordinance. A control orientation incorporates the relevant features of an incremental model of decision making by its cognitive attention upon the line-items of expenditures and its decision rule of a comparison and minimization of the yearly increase in spending. Line—items of expenditures are the inputs of governmental activities-- the costs of purchases. The different kinds of items bought have been classified into different account categories. Depending upon their Specificity, the number of lines of expense extends from a few broad groupings such as wages and employee 7. Allen Schick, Budget Innovation in the States, (Washington D. C.: The Brookings Institute, 1971), p. 4. 8. Allen Schick, "The Road to PPB: The Stages of BUdget Reform," Public Administration Review, vol. XXVI, no. 4 (December, 1966), p. 246. orchases. one of the . O T"ommon A»; ”K“ 54. ~En810r rule fol] Q ,_l h l: (D I‘eaSe ChoiceS’ Often. t 300 benefits, supplies, equipment, maintenance and repair, etc., to literally hundreds, if not thousands, of specific purchases. The amount of money spent for each and every one of these individual line-items is the kind of in- formation available in an incremental approach to decision making. The budget is a dollars and cents statement of the costs of governmental purchases. It is only a series of sums of how much money is to be expended. The program and policy commitments reflected in current Spending are not reconsidered in the annual budget process. Instead, existing appropriations are accepted as the legitimate base of funding levels and from this starting point budget decisions are made. Next year's budget is based upon the present, as this year's budget is based upon the past. The future is merely an extension of the past. Correspondingly, the evaluative rule followed is to minimize the yearly increase in costs. Continuity and consistency in expenditure levels is its hallmark. The decision making process has been simplified by concentrating upon the cost of government's purchases and upon the narrow range of spending differences from one year to the next. So, it is the size of annual increases that is the primary consideration in spending choices, and not the absolute amount of appropriations. Often, this all—pervasive fact is lost sight of. It is only the 1: mint in I EPpI‘Opria- elli'ifllpi e S cision ha” 35 govern: Ric ‘i‘ppropl‘ia The red ati inq nor OI“ Sid con 301 only the percentage change in spending levels from one point in time to another that is under discussion and not the absolute amount of the difference. As Davis, Dempster, and Wildavsky write: It is evident, for example, that decision makers in the budgetary process think in terms of per- centages. Agencies talk of expanding their base by a certain percentage. The Bureau of the Budget is concerned about growth rates for certain agen- cies and programs. The House Appropriations Committee deals with percentage cuts and the Senate Appropriations Committee with the question of whether or not to restore percentage cuts.9 Incrementalism, is by far the dominant empirical description of the budgetary process. In almost every past study, it is presented as the way in which the appropriation ordinance is prepared and adopted. Three examples of department, executive, and legislative de- cision making on the national, state, and local levels Of government are provided below. Richard Fenno describes the decisions of the House Appropriations Committee in these terms: The Committee perceives its oversight and budget reducing tasks as essentially incremental oper- ations. When the Committee makes its annual inquiry into agency appropriations, it does not normally range throughout the length and breadth of agency activities. Just as the agency con- siders much of its appropriation to be beyond controversy, so, too, does the Committee act on . 9. Otto A. Davis, M. A. H. Dempster, and Aaron Wildavsky, "A Theory of the Budgetary Process," American Eplitical Science Review, vol. LXVI, no. 4 (December, 1966)! p0 5300 this to t prev year crea n We Thou Separtment that exci new ful- some revf two our} peh< glaz 0011 Fri; sum neX' 3993 do: are the dit exp It abo the the And he \Lh rega Cmm Greg Ins 302 this assumption by restricting its purview to those budgetary increments granted in the previous year and requested for the coming year. "We only have the time to study in- creases and new functions," said one member. "We don't go into existing functions."lO Thomas Anton describes a similar process in the Departments at the level of State Government: . . . it is less than surprising to discover that questions and discussion focus almost exclusively on institutional requests for the new biennium. But such requests become meaning— ful-- and questionable-~ only in comparison with some other standard. In the context of the budget review situation, this standard is provided by two other columns on the various budget forms: current appropriation and estimated total ex- penditures for the current biennium. A Quick glance by Deputy Director Merten at these three columns enables him to compare the present appro— priation against the sum now being spent and the sum which the agency would like to spend in the next biennium. Where the latter two sums are approximately the same, or where the new requests do not exceed the present appropriation, they are passed over quickly-- frequently in seconds. Whenever the new requests exceed current expen- diture by a considerable amount, however, some explanation is requested by the deputy director. It is precisely this process of raising questions about the increased sums of money requested for the new biennium that constitutes the heart of the budget review in Mental Health.ll And finally, the same incremental process is described with regard to the executive on the municipal level by John Crecine: Instead, the mayor perceives the budgetary problem _ 10. Richard Fenno, The Power of the Purse: Appro- priations in Congress (Boston: Little, Brown and Company, 1966). p. 318. ll. Thomas J. Anton, The Politics of State Expen_ diture in Illinois (Urbana: University of Illinois Press, 303 as a continuing one that must be dealt with periodically (yearly). He perceives this year's budget to be basically similar to last year's with a slight change in resources avail- able (new revenue estimates) for dealing with a continuing set of municipal problems (police and fire protection, urban renewal, public works, transportation) augmented by a small number of partial solutions to old problems. In this context, a logical way to proceed in solving the complex budgeting problem is to take last year's solution (current appropriations) and to modif it in light of the change [italics in original7yin available resources and change [italics in original7 in municipal problems and their available solutions. to obtain this year's solution. This, of course, means that the budget is a slowly changing thing, consisting of a series of marginal changes from previous budgets. Very small portions of the budget are reconsidered from year to year and consequently, once an item is in the budget, its mere existence becomes its reason for being in succeeding budgets.12 Analysis A series of questions were asked to department heads, the executive, and the legislature to measure the adherence to an incremental budget orientation. First of all, a normative assessment of their belief that: "The best guide in deciding what to request, recommend, or appropriate this year is what was spent last year." This is not meant to provide a description of how these three decision makers actually go about making spending choices, but their adherence to the 12. John P. Crecine, Governmental Problem Solving: A Computer Simulation of municipal Budgeting (Chicaso= Rand McNally, 1969). p- 41- :n ”l 4'. is what WE V I .fi) W I}! 304 prescriptive tenets of an incremental approach as the preferred way of budgeting. However, only to a slight extent do they agree that the best guide for the future is what was done in the past (see tables 9.1, 9.2, and 9.3). The next aspect of the incremental budget orien- tation is its informational base: "The attention paid to the purchases-- the line‘items of expenditures," of the municipal budget makers. In terms of the department, this question was asked twice: once for their own preparation of requests, and the other time for their meeting with the executive. Then it was posed, more generally, a single time to the executive and the legis- lature. In all four questions, with only some minor differences, decision makers report paying great atten- tion to the cost of the items bought by government, throughout the three decision making stages. This cognitive framework is persuasive, for in no single case is there less than some focus upon this cognitive input, while in many instances the highest rank-- category of very great attention is indicated. The next two questions measure the application of incremental evaluation rules. The first was the importance, "In deciding what to request, recommend, or appropriate this year, how much was spent last year." The second was the extent to which: "If a request, Jul C II\ II\ :r\ writ Ir: . . . . I: . . II I I . S .1 O O 6 56 3 36 3 0 6 0 >6 1 7 e \u .............. - - W 4 n .u 0 2 O \I. ‘i O 0 l 2 l 1 l 0 l l 0 “V9 n I 15L a ' vvl/i .11- 7| 2 A) , ”w. ffila fhv 7 «UFV C/ “U \II fi/h 1.11. .149: Q» to m: J AAIV «MU .nItU D. U 1.1.. O 0 0 A, U NI s \l ‘ n l~ l ~||~ Wk.“ U \ .4 TABLE 9.1 INCREMENTAL BUDGET ORIENTATION (DEPARTMENT HEAD) Best Attention Attention Importance Cut Mean Clty Guide Internal Meeting If More 01 0.00 3.66 3.00 3.44 1.66 2.33 02 2.00 3.33 3.66 3.00 1.33 2.65 03 0.66 .00 3.66 2.66 1.33 2.47 04 1.50 4.00 2.66 2.33 1.50 2.40 05 1.66 2 66 1.00 3.33 0.66 1.87 06 0.33 4.00 3.00 4.00 1.00 2.47 07 0.33 4.00 2.00 2.66 0.33 1.87 08 1.66 2.00 2.16 4.00 1.00 2.17 09 2.33 3.66 3.33 2.66 1.50 2.70 10 1.00 3.33 3.66 2.33 1.66 2.40 11 1.66 2.33 3.33 2.33 1.66 2.27 12 1.00 3.50 2.00 3.00 2.00 2.30 13 0.50 3.00 2.66 3.66 0.00 1.97 14 1.66 3.90 2.00 2.66 2.66 2.40 Mean 1.16 3.32 2.70 3.00 1.31 2.31 St. U m 4 o O V N c O 0\ \n c an N c> . kn \o c: O\ a) o no V Dex). \ n a 0.. r2; DJ \| \J l.l. t (,J/O 7flhv «(U/ 0 \ll. 2 A).MHV. . uslv AU‘ M/le mVJan a NHV 0 0 U \vlos \ls \l- 1 !|\ u .4. Nb 1 1|— V 306 TABLE 9.2 INCREMENTAL BUDGET ORIENTATION (EXECUTIVE) City ggige A$§§$§ign Importance ggt Mean More 01 1.50 3.00 4.00 0.50 2.25 02 0.66 3.33 3.33 0.33 1.92 03 2.00 4.00 3.00 1.50 2.62 04 0.00 3.50 3.50 0.00 1.75 05 0.66 3.66 2.33 1.00 1.92 06 0.66 4.00 3.25 1.00 2.23 07 1.50 3.50 3.00 2.50 2.62 08 0.75 3.00 3.25 1.25 2.06 09 0,33 3.33 2.66 1.00 1.83 10 0.50 2.50 3.00 1.50 1.87 11 2.00 3.50 3.25 2.00 2.69 12 1,50 2,50 3.00 1.00 2.00 13 0.50 3.75 2.00 0-50 1.69 14 1.50 3.50 3-00 0.50 2.12 Mean 1.00 3.36 3-04 1°04 2-11 St. Dev. 0.64 0.47 0.49 0.68 0.43 “Viv h . 0 TV a 0 v :1 sly. .t/N .lxltttiy Ell/AV 7 «Hut. AU/ 0 l rl/N at} NV Au» I‘IU We a J 0 DAM O 0 AIIU» OJ» All» flu. n.U \lls l ‘1'. al‘ ‘1 .. ._ \.\U \ 307 TMEE9J INCREMENTAL BUDGET ORIENTATION (LEGISLATURE) City gfiige A$E§¥Eign Importance Cgt Jean More 01 1.50 2.37 2.75 1.00 1.91 02 0.25 2.40 2.60 0.50 1.43 03 0.57 3.14 2.85 0.71 1.82 04 0.33 3.00 3.33 1.33 2.00 05 1.40 1.75 3.14 0.25 1.64 06 1.25 3.57 3.85 1.75 2.61 07 1.00 2.83 3.66 1.50 2.25 08 1.66 2.20 1.60 0.66 1.53 09 3.10 3.00 3.20 3.10 3.10 10 2.00 2.66 3.66 0.00 2.08 11 1.25 2.75 3.00 1.00 2.00 12 1.25 3.25 3.25 1.25 2.25 13 1.25 2.85 3.85 1.33 2.32 14 1.25 2.00 3.57 0.50 1.83 Mean 1.15 2.70 3.17 0-91 2-05 St, Dev. 0.05 0.51 0.60 0.52 0.27 01‘ I‘GCOZDIH divergenc the first a bachvar. Expenditu: 30 the st: 317-: as p. the third idheFe to all propo CPDOSed b; The 308 or recommendation was near the level of current appro— priations it was approved; otherwise, it was reduced." For some unexplained reason, there is a significant divergence in the responses to these two questions. In the first case, all three participants reply that such a backward looking glance was of great importance making expenditure decisions for the next year. This attests to the strong presence of an incremental spending cri- terion. However, at the same time, all three decision makers report that only to a slight extent do proposals for increases automatically trigger a negative response from those reviewing the budget; while proposals that are consistent with existing levels are accepted readily. This is crucial, for it is not so much that yearly change is examined, but that deviations from the base are opposed. But, as partially evident in the statistical analysis of the third chapter, municipal decision makers do not adhere to such an incremental process of evaluation. Not all propOSals for spending increases are automatically Opposed by the executive and the legislature. The index of the separate questions indicate that an incremental budgeting is present to some extent. This has to be considered a low absolute level of response, given its pervasiveness in past descriptions of the budgetary process. Letting public officials speak for themselves, indicates a far weaker application of incremental decision 1 observers an increm an increa is what '1 evaluativ 309 decision making than reported by their academic observers. Relationship to Budget Behavior Very simply, the more decision makers adhere to an incremental orientation to budgeting, the less of an increase in spending levels should be apparent. This, is what incrementalism is all about. As a cognitive and evaluative process of choice is presumably results in smaller yearly increases in expenditures. 0n the level of departmental decision making, this is, indeed, the case. The more department heads follow such a process, the less of an increase in funds is requested (—0.3274). A concern with the costs of pur- chases, and a comparison of yearly change does have the direct and immediate effect of inhibiting the submission of expansionary requests. However, and most significantly, a similar connection between incremental budgeting and the end—products of executive and legislative decision making does not have the effect of holding the line on yearly increases. The executive does not recommend any less of an expansion in spending (0.0747); nor does the legislature approve a smaller growth in final appro- priations as a result of an incremental review of the budget (—0 . 0220). Furthermore, incrementalism does not display its assumed effect upon the change in spending levels from he stage or. the bu adherence the execu ieoartmen noose la. alter man so ts ‘tha 310 one stage to the next. Larger cuts are not imposed on the budget proposals of others as a consequence of adherence to this budgetary orientation. Neither does the executive make greater reductions in initial department requests (0.0286); nor does the legislature impose larger cuts in executive recommendations (0.1972) after making a comparison of the yearly increase in costs than if they do not apply this evaluation rule. The overall consistency of the five statistical correlations is marked, and quite conclusively indicates that far too much attention has been placed upon the concept of incrementalism as both an empirical description of the budgetary process and as an explanation of expen— diture outputs. It is evident that the effect of adhering to such a decision making process does not produce its intended effects. Neither in the direct connection between the decision making process and its eXpenditure end—products, nor in the statistical inter- relationships among the different spending outputs identifies the presence of incrementalism. Similar expenditure decisions are made without examining the line—items of expenditure and without a yearly com— parison of costs, while utilizing such a cognitive and evaluative process is no guarantee than smaller yearly Change will be produced. In spite of all that has been written, an incremental budget process does not provide igoortanc v seen Witst is define approved in o‘esig programs, curenent budget pr 0‘: goverr W. . LA ‘ .«thy ( ‘41 Mial‘ a T‘ :em“ 311 a fruitful explanation of municipal budgeting. The importance and consequences of incrementalism have been vastly over—emphasized. Management The second orientation to budgeting, that stands ambiguously between incremental and synoptic decision making processes is called management, and is defined by Schick as, ". . . the programming of approved goals into specific projects and activities, the design of organizational units to carry out approved programs, and the staffing of these units and the pro- curement of necessary resources."13 In short, the budget process is now used, not to control the costs of government, but to manage the operations of govern- ment. The unit of analysis is activity categories, as expense accounts are rearranged into functionally related groupings of work to be done. This is a fundamental Shift in information, from the input costs of the items bought to the outputs of government-- the level and quality of services and programs provided by these dollar and cents purchases. Thus it is often called Performance Budgeting. As Allen Schick goes on to write, 13. Allen Schick, "The Road to PPB," p. 244, it is con Wha per is a t duc activitie “How ef f E their avc conduct < Spend is the serv.‘ that is - nle‘l’e‘lOpez prOdUCtii llantity Similari is That 1814318 a Emilial b \ 14 312 it is concerned with the questions of: What is the best way to organize for the performance of authorized activity? What is and what should be the cost of performing a unit of work? What is the maximum pro- ductivity derived from a unit of input?l4 The evaluative criteria of expenditure choice is now the efficiency and economy of agency operations-- how activities are carried out. As Arthur Smithies writes: "How effective are budgetary procedures in achieving their avowed objective of economy and efficiency in the 15 conduct of government operations." The decision to spend is based on the work to be accomplished. It is the service rendered and not the things to be acquired that is the standard of evaluation. Consequently there developed measures of accomplishment, especially the productivity of the delivery of services. These examine the relationship between the cost and the amount of goods and services produced, that identified and assessed the quantity and quality of governmental activities. The similarity between incremental and performance budgeting is that they both accept the base of existing spending levels and policy goals as beyond the scope of the annual budget review. Even though the management orientation 14. Allen Schick, Budget Innovation, p. 5. 15. Arthur Smithies, The Budget Process, pp, 8-9, examines and outpL the purpc essential It accept while ati went. It to expend h’ow lion Sper Peoresent budgetary is itselj theSe WC 5093 Star getiné’ at IIIIIIT___________________________________________________‘DE7 313 examines the more complex relationship between inputs and outputs, than the simple rate of spending increase, the purposes of the efficience and economy concern is essentially the same as the traditional control orientation. It accepts the current state of affairs as a given, while attempting to maximize the performance of govern— ment. It may serve just as much as a negative orientation to expenditure choices, as does incrementalism. However, at the same time, the reorientation from costs to services should not be minimized and does represent a different cognitive map, which may have a direct effect upon spending choices. In addition, the performance focus represents the application of systematic analysis to budgetary decisions that were previously absent and which is itself the hallmark of a synoptic process. Thus for these two reasons, the management orientation to budgeting does stand for a first step away from incremental bud- geting and as Schick so aptly phrased it, it is on the road to PPBS. Reports of a management approach to budgeting are limited. The existence of an efficience and economy criterion in governmental budgeting has not been ex- tensively identified. Part of this reason is that whatever performance orientation to governmental policy making takes place outside of the formal boundaries of the budgetary process. It is more associated with other decision making areas, that occur in other times and other places ti one excet Thomas Ar the budge statement The do: her of v02 00: pa: pr: Opt 314 places than the explicit three stage sequence. The one exception to this pattern worth quoting is from Thomas Anton's study of Three Cities. He writes that the budget manual in one city contained the following statement that exemplifies the performance orientation: The basis of a sound budget is the work to be done and the service rendered. This compre- hensive work programming includes consideration of such factors as: 1) service standards 2) volume of work 3) work methods 4) units of cost and 5) results to be achieved. Each de~ partment head is Specifically charged in budget preparation with analyzing his entire department's operations with a view of effecting every economy possible . . .16 However, this is only a prescription of what the executive wants the department heads to do, as there is no evidence that this is, in fact, the process by which budget decisions are made. Analysis The budget orientation of department heads, the executive, and the legislature were similarly measured by three sets of questions that measure its noramtive, cognitive, and evaluative presence within municipal government. It is a common ethos, that municipal government is a service delivery organization, where business . . 16. Thomas J. Anton, Budgetingin Three Illinois Cities, Commission Papers of the Institute of Government and Public Affairs (Urbana: University of Illinois Press, 1964). p. 11. practices question business uanagemer the norei executim Tanager ; ethos, Z‘Ssponse attestin: mt all llllllllllIIIIIIII-::r———————————————————*—————————————————————~e+-e~r 315 practices and procedures are applicable. Thus the first question posed, was the belief that: "A city like a business should be run on the basis of professional management techniques." Agreement with this normative value extends from a great extent for both the department heads and the legislature, to very great extent on the part of the executive. The very strong agreement with the norems of performance budgeting on the part of the executive is expected considering the prevalence of the manager form of government and its embodiment of that ethos. However, there is no apparent difference in responses from mayor cities or other decision makers attesting to the prevalence of these values and through- out all municipal government. The next element of the performance orientation to budgeting is its cognitive inputs. Again two questions were asked to the department heads, where one question was sufficient for the executive and the legislature to measure the extent they: "Paid attention to the activities (services) of government." In all four cases, the output side of spending was the focus of great attention. This is a strong cognitive orientation of budget makers as there is not a single city where less than some attention is paid upon this kind of information, and the range of variation is narrow, only extending along the upper ends of the scale. The iiiegemer rating 8; of: "Peri request, izportanc coeratior ease, the performer five ran} Th5 index of the: thre the 58am absolute 316 The final two questions probe the presence of a management orientation as an evaluative criterion in making spending choices. The first is the importance of: "Performance-productivity data in deciding what to request, recommend, or appropriate." The second is the importance of: "Efficiency and economy of government operations in making budget decisions." In the former case, the department heads and the executive report performance measures as being of great importance, as the legislature indicates it is of very great importance. All three decision makers then report that the efficiency and economy criterion is of some importance-- the third of five ranked alternatives. This series of questions together as an composite index of the presence of a performance budgeting, reveals that three decision makers, adhere to a great extent to the features of a management orientation. This is a high absolute score, that is in marked contrast to the general absence of past descriptions of this approach to budgeting. It should be made clear that the description of this budgetary approach derives from the response of the decision makers themselves. There was no effort to independently corroborate the actual utilization of Performance measures. The description of what occurred depends entirely upon the perceptions of those participating and what they perceive to be the process of making budget choices. responses :his to ‘: eaters p: :enageme: It has upon n‘oiguoo S"Footie it has 11 “limitud Provide 317 choices. Assuming the validity and reliability of the responses (and the consistency of the responses suggest this to be the case), it is clear that municipal decisions makers perceive themselves far more concerned with management questions than others have seen them to be. Relationship to Budget Behavior The specific impact a performance orientation has upon the two measures of expenditure outputs is ambiguous. As it represents elements of both the synoptic and incremental approach to decision making, it has uncertain consequences upon the direction and magnitude of change in spending levels. It could provide a positive impetus because of its use of systematic analysis and quantitative data to support increased ex- penditures. As Allen Schick writes, in support of this effect: Agencies are forever playing the "numbers game" to their advantage. They invent staffing form- ulas, minimum standards of performance, grading systems, and similar seemingly scientific measure- ments in order to obtain more money for recreation, libraries, education, or some other "neglected" concern. They revel in demonstrating that they are understaffed or underfinanced by comparison to their peers. They develop measures to show how much more money is needed in order to meet minimum standards. And, of course, the standards are raised once the coveted level is reached.l7 On the other hand, it could represent a negative orientation 1?. Allen Schick, Budget Innovation, p. 172. to spend. and em lature S \ . 18 13‘ ‘ S felt ‘ I" A 318 to spending as a result of its efficiency and economy criteria. As Donald Borut writes of such consequences: Better program data is a two-edged sword. It can be used to justify the necessity of ad- ditional funding and [italics in original7 and it can be used by a sophisticated advocate to verify the validity of budget reductions [italics mine/. In a word, greater information means greater control for the stronger of the two adversaries. Department heads may prefer not to have program data, and at the critical moment, when a reduction is to be made, stand on their "reputation as professionals" . . .18 The overall pattern of the five correlations indicate that the management—performance orientation to budgeting is a technique to hold the line on annual expenditure growth. For the department heads (-O.2132) and the legislature (-0.0772) it functions as a constraint upon the annual expansion of spending levels. The more these two actors follow such an approach the less the budget either first requested or finally appropriated expands. On the other hand there is a substantial positive asso— ciation between the executive management orientation and the growth of his own recommendations (0.4627). However, the executive cuts departmental requests for increases more as a result of the application of such a cognitive and evaluative framework (0.5528 + 0.05), as the legis— lature similarly reduces executive recommendations for 18. Donald J. Borut, Implementing PPBS "A Practi- tioners Viewpoint," in John P. Crecine, ed., Financin the Metropolis: Public Policy in Urban Economics (Beverly Hills: Sage Publishers, 1970). Po 300- exoansim W549) balancer ii expendh 319 expansion as a consequence of adherence to a management orientation (0.2950). The impact of the management-performance orientation is further evident in its connection to the executive's and the legislature's negative budget role. It serves as a reinforcement for their particular orientation to budgeting, and as the mechanisms to implement such ob- jectives. Thus the more the executive employs such cognitive and evaluative rules in evaluating departmental requests, the more he evaluates their requests as padded (0.4549) and the more he perceives his own role as the balancer (0.3176). The performance orientation supplies him with the data to eliminate excessive and wasteful spending requests in order to balance the budget. There is a similar orientation to the legislature's oversight role (0.4198). The more they are oriented to affecting economics and efficiency is in the operation of govern- ment the more extensively they review the executive's recommendatiOns with this aim in mind. It is clear that the negative orientation to expenditure increases is represented in the twin values 0f efficiency and economy. It is not the simple comparison of costs of items to be purchased and their incremental change that is the orientation of budget makers, but the more complex relationship between expenses and the amount of services they buy. That is the tool to inhibit the amual g1 Th that rep: is calle- for a P1. the util :0 set t the, ". the eval 1.3011 the 320 annual growth of municipal spending. Planning The final orientation to budgeting and the one that represents the synoptic model of decision making is called Planning and is identified with the movement for a Planning Programming Budgeting System. This is the utilization of the budgetary process as the vehicle to set the agenda and goals of government itself. It is the, ". . . process of determining public objectives and the evaluation of alternative programs."19 It focuses upon the widest range of issues: What are the long-range goals and policies of the government and how are they related to particular expenditure choices? What criteria should be used in appraising the requests of the agencies? What programs should be initi- ated or terminated, and which expanded or curtailed.20 The unit of analysis is also upon outputs, but in terms of the programs' objectives of government, than the activity itself, and thus it is often called Program Budgeting. The criteria of choice are how effectively expenditures achieve the objectives sought. As Arthur Smithies writes: "How effectively are expenditure decisions related to policy objectives they are intended to secure?"21 19. Allen Schick, Budget Innovation, p. 5. 20. Allen Schick, "The Road to PPB," p. 246. 2l. Arthur Smithies, The Budget Process, The deci: avhieved :ethods ‘ ends cal is evide: Th 35' compa relat insider 55 beyon 15 31% i- ‘fkbe a, Huh/v 321 The decision to spend is based on the goals to be avhieved and the systematic analysis of alternative methods to reach them. Funds are allocated according to the purpose to be accomplished. An explicit cost/ benefit analysis is undertaken as the deliberate means- ends calculation of the synoptic decision making process is evident. This orientation to decision making becomes clear by comparing it to the two other approaches. First, in relation to incremental budgeting, planning does not consider spending levels and the objectives it represents as beyond the scope of the annual review. It is not the cost of what has been done that is examined, but programs as a means of deciding upon future goals that is evaluated. As Allen Schick writes: In process politics the contestants tend to View the options from the perSpective of their estab- lished positions (existing legislation, last year's budget, the "base," etc.). Theirs is a retrospective bias. Budgeting is treated as the process of financing existing commitments and of creating some new commitment (the incre- ments). System politics tends to have a prospective bias; budgeting is regarded as the allocation of money to obtain some future value (the outcome or objective). This year's budget, in system terms, is an installment in buying that future.22 It is in this same way that planning differs from 22. Allen Schick, "System Politics and System Budgeting," Public Administration Review, vol. XXIX, no. 2 (March/April, 1969). P. 138. 322 management. One can be concerned with the efficiency and economy of governmental services and utilize the measurement and analytical techniques without eval— uating the objectives of the activity being analyzed. One has to consider the purpose of the program, as well as its performance. The effectiveness of the activity in relation to its objective and not its efficiency and evonomy is examined: As a policy device, program budgeting departs from simple engineering models of efficiency in which the objective is fixed and the quantity of inputs and outputs is adjusted to an optimal relationship. In PPB, the objective itself is variable; analysis may lead to a new statement of objectives . . . but from the planning per— spective, the all-important thing surely is not the work or service to be accomplished but the objectives or purposes to be fulfilled by the investment of public funds.23 Past research has generally found that the budgetary process is not used as a means for program and policy planning. The budget is not utilized as the means of making independent policy choices aside from the marginal review of spending levels. Instead it is largely the fiscal translation of decisions arrived at prior to and outside of the budget making process. Examples are evident at all levels of government. Thus Aaron Wildavsky writes Of the national government that: 23. Allen Schick, "The Road to PPB," p. 250. *U C) 'U ’C‘1 (D l‘“ L) C. '1‘ '4 r» Hl—A’fi‘ (I) 1‘ (AF )1) Claw. . m . F‘s 323 Budgeting is treated as if it were non-programmatic . . . What it does mean is that they View most of their work as marginal, monetary adjustments to existing programs so that the question of the ultimate desirability of most programs arises only once in a while.24 0n the state level, Thomas Anton provides a further example of the absence of a program content in departmental budgeting: Departments and institutional policies [italics in original7, in other words, are determined elsewhere, through procedures distinct from the process of budget preparation . . . program and policy innovation normally do n9: [italics in original7 originate in the process of budget preparation . . . Significantly, such major ad- justments-- whether increases or occasional decreases-- are not generated by the roles analyzed here. Normally they are initiated through different systems of roles and merely reflected in this system . . . Action by this system came after these plans had been formulated and approved, and took the form of a ratification of proposals conceived and implemented by others . . . In each of these cases, involving major adjustments in existing expenditure patterns, the budgeting and appropriating systems was only formally involved: it did not initiate, it did not plan, it did not negotiate; it merely ratified actions taken by other roles.25 Finally, on the local level, the research of Robert Eyestone comes to a similar conclusion: . . . the budget is not in itself a useful instru- ment for policy leadership. When policy change is required, these city councils adopt extra- ordinary procedures that enable them to make 24. Aaron Wildavsky, The Politics of the Budgetary Process (Boston: Little, Brown, and Company, 1964), p,‘60, 25. Thomas J. Anton, State Expenditure in Illinois, pp. 74, 199, 200, and 201. A heads, 1 to team I ' 4 .:.ng Wl l -4- hide on accomplj ail‘eemer approac? — i 324 budgetary decisions markedly different from those they would normally make . . . The difficult decisions requiring policy leader- ship are made outside the framework of "normal" budgeting.26 Analysis A similar set of questions asked to department heads, the executive, and the legislature were designed to measure the presence of synoptic budgeting. Begin- ning with beliefs that: "Budget decisions should be made on the basis of the policy objectives sought to be accomplished. All three decision makers report great agreement with the normative prescriptions of a planning approach to municipal decision making. Next, the cognitive component of budgeting: "The attention paid to programs-- the policy goals of govern- ment," was probed. In all five questions there is great focus upon this kind of information. The next two questions assess the existence of a planning approach to evaluative rules. First, it was the importance of the: "Effectiveness programs in attaining goals in deciding what to request, recommend, or appropriate." The second question was the relative importance of: "Planning the future objective of government in the course of budgeting." Again all three decision makers 26. Robert Eyestone, The Threads of Public Policy: fig§tudy of Policy Leadership,(Indianapolis, Indiana: Bobbs ’Ierrill Publishers, 1971). pp. 156-157. general] was of g Th decision to budge magetar authorit the exis V CD, A ‘ \..| NM- «1111 'r ‘ns 1: i- 325 generally indicate that such an evaluative standard was of great importance in making expenditure choices. This series of questions reveals that municipal decision makers greatly adhere to a planning orientation to budgeting. To a great extent the budget is used as a tool to establish the future. This too, is a surprisingly high absolute score, considering all that has been written in the past of the failure of governments to utilize the budget as independent policy making arena. Again, it must be emphasized that this description of the municipal budgetary process relies solely upon the replies of authoritative decision makers. This does not indicate the existence of a Planning Budgeting System, for only one city follows the organizational features of program categories and memoranda, multiyear forecasts, cost/ benefit analysis, etc. It is purely the subjective perspective of these actors of the extent they employ the budgeting process for the purposes of planning. Relationship to Budget Behavior The synoptic planning orientation to budgeting, as the conceptual opposite to the incremental model of decision making is expected to produce the opposite pattern of ex- Penditure outputs. Thus, as it was assumed, but not empirically corroborated, that incrementalism has the effect of stable spending levels, the program orientation would be associated with larger change in expenditures. 326 But beyond this logically deduced relationship to outputs of budget decisions, the essential meaning of the concept itself suggests this end result. Its future orientation, would tend to produce larger annual shifts than derive from incrementalism with its focus on the past and performance budgeting with its concern for the present. As Allen Schick writes: A budgeting process which accepts the base and examines only the increments will produce decisions to transfer the present into the future with a few small variations. The curve of government activi- ties will be continuous, with few Zigzags or breaks. A budget making process which begins with objectives will require the base to compete on an equal footing with new proposals. The decisions will be more radical than those made under incremental con- ditions.27 Then as Robert Eyestone specifically writes of municipal budgeting: A negative budgetary style implies strict limits on city spending of any kind: hence, councilmen using this style should be found in low spending cities. Programmatic budgeting implies that policy decisions may be directly implemented in the budget without regard to previous or expected levels of spending: hence, programmatic councilmen should be found in cities where expenditures are changing more frequently than in cities where spending levels are remaining stationary.28 But as Eyestone finds little empirical connection between a programmatic orientation and his particular measure of expenditure outputs, the present study similarly 27. Allen Schick, "The Road to PPB," p. 49, 28. Robert Eyestone, Threads of Public Policy, p. 910 327 uncovers little support for the assumed effect a planning orientation has upon municipal expenditure outputs. Neither the department requests (-0.0088), the executive recom— mendations (0.1549), nor the legislature's final appropriations (-O.2472). The budget does not increase any faster as a result of the application of this budgetary approach. There is also no connection to the amount of cuts that are imposed from one budgetary stage to the next. The executive does not reduce departmental requests any more or less as a result of his adherence to such an approach (0.1154), nor does the legislature's application of a program planning approach result in any effect upon their review of executive recommendations (0.0033). It is quite clear that the programming and planning approach does not represent a positive impetus to municipal expen— diture outputs.29 Perhaps this empirical finding, that a planning orientation to budgeting has no apparent consequences for the content of spending choices partially explains the failure of PPBS to be implemented. Not only must dis- content and dissatisfaction with existing structures and 29. Even if legislative budget behavior were arranged along a scale of the absolute amount of change there is no consistent connection to a planning orien- tation. It is 0.2975 to the absolute amount of change in executive recommendations, but -0.1246 to the absolute amount of yearly change in spending levels. f— _i 328 procedures of budgeting be present, but what is to be gained from such an innovation must also be made evident. But, reformers could not identify how spending outputs would be effected, for as it is now evident, such an approach has no direct and concrete impact upon spending decisions. Conclusion All too often the discussion of incremental and synoptic budgeting has been marked by polemics, as either supporters of the status quo or proponents of the PPBS reform have presented these two modes of budgeting as dichotomous and alternative decision making processes. Either one follows all the steps of a synoptic model of decision making or all of the practices of the incremental approach. There is no middle ground, all or nothing, * either/or choice are forced upon the budgeter. However, as James Barber writes: "The alternatives are not incre- mentalism versus synopticism, but more-or-less incrementalism 3O versus more—or-less synopticism." They are less jointly exhaustive and mutually exclusive categories as they are "ideal" types along a continuous distribution. As Allen Schick writes of the greater similarity between them . 30. James Barber, Power in Committees: Experiments gpgthe Governmental Process (Chicago: Rand McNally, 1966), p0 350 329 than usually assumed: On all other counts, however, the difference between the incremental and the analytical methods is considerably narrower than is usually realized. PPB does not force an all-or~nothing choice. The alternatives are always at the margins, and the increments can be either large or small. PPB does not require that every program be compared with all others, nor does it mean that everything must be decided all at once. PPB does not require zero base decisions. Only a portion of the budget can be analyzed during a single cycle, with most of the programs continuing according to standard incremental procedures.31 And to this view the present research most strongly agrees. In a temporal sense almost all decisions are incremental changes from an existing situation. Spending choices are almost never determined de novo either from one year to the next, or from one stage to the next; but are made in reference to what took place earlier. But even more importantly, the application of one set of values, one source of information and one set of evaluation rules does not preclude the adherence to another. Just because, decision makers pay attention to the line- items of expenditure does not necessarily imply that they ignore service and program categories. Just because they desire the minimum deviations from the base of existing Spending does not mean that they are also not concerned with the efficiency and economy of governmental spending 31. Allen Schick, Budget Innovation, p. 202. 330 operations, or the effectiveness of spending. The absolute responses to the questions measuring these orientations among the three budget participants re- veals that they are not alternative decision making practices, but complementary ones. Furthermore, the statistical interrelationships among them clearly in- dicate that one approach does not replace another: Incremental/ Incremental/ Performance/ Performance Program Program Department Head 0.1352 0.4242 0.3077 Executive -0.0506 -0.0165 -0.2220 Legislature 0.0066 0.0956 -0.1692 There is essentially no connection among the three budget making orientations. Except for the two cases of departmental decision making. adherence to one specific budget making approach is not related to following another. It neither precludes nor accompanies the utilization others. A different set of reference points are provided and another Standard of evaluation exists. Most significantly, a DrOgram orientation is not a substitution of incrementalism, or vice versa. The new does not push out the Old- Chapter Ten: External Pressure Introduction The analysis of the part played by individuals and groups from the community in the municipal budget process is important for two reasons. First of all, it serves as a linkage for demands arising from the environment to be transmitted to public officials. The impact of the environment in the determination of expen- diture outputs cannot be explained by aggregate measures of social and economic conditions. Instead, their direct and immediate connection to the budgetary process must be Shown. The involvement of this fourth set of actors Provides the concrete behavioral mechanisms for such a relationship to be made. Secondly, the study of interest groups, has long been a favorite approach to the explanation of the POlitical process and public policy. While claims for the efficiency of these extra-governmental participants are often exaggerated, the study of the municipal budget Process would not be complete without including them Within the analysis. Past research has reported different situations of 331 their invc study of r 01“ em acti sued by ag on their 1 the l Supp firm: unifi cula' 3a'lid Capt Of four 0 Mi 2: mi tive SUCC ' I re conclu (although 332 their involvement. Thus Aaron Wildavsky writes in his study of national budgeting that the, ". . . cultivation l of an active clientele" is a ubiquitous strategy pur- sued by agencies to affect the decisions made by others on their requests. Such considerations, he writes: . . . are found everywhere and at all times in the budgetary system. The need for obtaining support is so firmly fixed a star in the budgetary firmament that it is perceived by everyone and uniformly taken into account in making the cal— culations upon which strategies depend.2 David Caputo reaches a similar conclusion in his study of four cities. Thus he writes: During each stage of the budgetary process, a minority of the municipal group representa- tives attempted, usually with some degree of success, to influence the budgetary process . . . to change budgetary allocation deCiSions.3 He concludes that, ". . . municipal group representatives (although a minority) who participated played significant roles."l‘L However, John Crecine's study of municipal 1. Aaron Wildavsky, The Politics of the Budgetary Process (Boston: Little, Brown and Company, 1964), p. 64. 2. Aaron Wildavsky, Budgetarerrocess, p. 65. 3. David Caputo, "Normative and Empirical Impli- cations of Budgetary Processes," paper prepared for delivery at the 65th annual meeting of the American . P01itical Science Association, Los Angeles, California, 1970. pp. 11-12. 4. David Caputo, "Normative and Empirical Impli— cations," p, 16, I I \ .eups J A. V ma orr all D..0 e C." EUd X f A... I]. O p 1.4 an v.1. AL .1. S 39 CM 5 . :Pnu n . a»... A. » though 333 budgeting describes a different situation, where interest groups are relatively unimportant. He writes, in state- ments throughout his work that: The presence of external influence or pressure in the political sense was not detected in the budget formation process . . . influentials play a very minor role in the formulation of municipal operating budgets . . . and play a very minor role in the gross allocation of governmental resources . . . budgets in municipal governments are reasonably abstracted documents bearing little direct relationship to specific community pressures . . . theories that assume the city budget is the result of some kind of external event do not prove to be consistent with the process uncovered in this study. Concepts like "influence" . . . "interest group," politicians, and elite do not prove to be very useful in understanding the process by which budget level decisions are made."5 In examining the part played by this fourth set of external actors in municipal budgeting, a demarcation has to be made between their actual participation in each of the three formal budget stages and an assessment of their influence by authoritative decision makers. Even though this distinction has not been made in past research, these are two separate dimensions that do not necessarily go together. Thus Crecine writes that: Effective cues from the environment are not [italics in original7 received from community interest groups either because they are not articulated or because the cues, taken in toto [italics in original7 add up to an 1nfeas1ble 5. John P. Crecine, Governmental Problem Solving: iiggmputer Simulation of Municipal Budgeting (Chicago: Rand McNally, 1969), pp. 189, 191, 192, 216. SI 'these at '1. 334 set of demands on the system . . .6 These are two different explanations for the ineffective- ness of group demands on spending outputs. They are not equivalent. The first phrase, "they are not articulated" points to their failure to contact public officials to express their budget preferences. They are simply non- participants in the municipal budget process. On the other hand, the second phrase "add up to an infeasible set of demands," describes a situation where group demands on the budget are made, but are unable to affect author- itative expenditure deicsions. They are present, but too weak to influence budget outputs in the direction of their own objectives. A fourfold contingency table can be established between the participation of these external actors and their impact. Groups can or cannot be present and they can or cannot be important. First the extent of their activity and then their perceived influence will be assessed. Finally the actual relationship between external pressure and spending outputs in each of the budget stages is examined. Group Participation Department heads, executives, and members of the legiSlature were presented with a list of fifteen kinds 6. John P. Crecine, Governmental Problem Solving, P- 192, of inte Persona ihis qu o: the an inpu mount sinners for a ; involve hen may "N’t Meme t( Within 595211" We 335 of interest groups, and asked whether each of them: Personally contacted you about items in the budget." This question more exactly measures the breadth of group participation than its quantity.7 It is an indication of the scope of potential community interests that make an input into the budget process, more than the actual amount of group activity. It would have been too cumbersome to ascertain the frequency of such contact, for a precise measure of the extent of each group's involvement. But the present question does reveal how many different types of groups in the community come to the attention of governmental decision makers. The mean number of groups that contact department heads is less than three (see table 10.1). This is certainly a small number and represents only 17 percent of the possible number of groups in existence. In only four cities are more than four groups mentioned as having made some spending demand upon the departments. Although there are extensive differences among the cities, from an average of less than one to eight, there are only three cities where more than three groups are active within the first stage of departmental budgeting. The Specific groups contacting at least one department head in a city are (see table 10.2): neighborhood, homeowner, 7. Robert Eyestone and Heinz Eulau, "City Councils end Policy Outcomes: Developmental Profiles," in J. Q. Wilson, City Politics and Public Poligy (New York: John Wiley and Sons, 1968). PP. 37-66. )1. »/~ A). 15 .HJ/nt. 7 «ad xU/ Atty ‘1‘ n/s at. 14 0 my flu fly. 1.: AU 01» AM» AV .1. )x .1‘ .1. \l\ 516% ,exlr' 336 TABLE 10.1 EXTERNAL PRESSURE DEPARTMENT HEADS # # of Eval. Eval. City o Group Group Ind. Mean Contacts Influence Infl. Infl. 01 5.33-2.66 0.66-1.33 2.00 1.33 1.83 02 4.00—2.00 0.33-0.66 1.33 0.33 1.08 03 3.00-1.50 0.33-0.66 1.66 1.33 1.29 04 8.00-4.00 0.33-0.66 2.50 1.50 2.16 05 2.00-1.00 2.00-4.00 2.00 0.66 1.92 06 1.33-0.66 0.33—0.66 1.00 0.33 0.66 07 1.00-0.50 0.00-0.00 0.66 0.00 0.29 08 2.00—1.00 0.66-1.33 1.66 1.00 1.25 09 2.00-1.00 0.00-0.00 1.00 0.00 0.50 10 0.66-0.33 0.00—0.00 0.66 0.40 0,33 11 1.33-0.66 0.33-0.66 1.00 1.00 0.83 12 2.00—1.00 0.00-0.00 0.00 0.00 0.25 13 1.66-0.83 0.66-1.33 1.00 0.50 0.92 14 1.66-0.83 0.00-0.00 1.33 0.60 0.69 Mean 2.57—1.28 0.40-0.80 1.27 0.64 1,00 St. Dev. 1.99-0.99 0.53-1.06 0.65 0.52 0,62 h TABLE 10.2 KINDS OF GROUP CONTACT Department Executive Legislature Head Appear—Contact Neighborhood, Homeowners and Taxpayers 13 9 9 13 Business 11 12 9 13 Civic and Service 10 ll 7 11 Municipal 1 Employee Unions 10 12 9 ll Blacks 7 8 7 10 338 and taxpayer—— thirteen cities, business—— eleven, civic and service-- ten, unions of municipal employees-- ten, and blacks—- seven. The presence of the first three types of groups are usual interests associated with the study of urban politics. But the extensive presence of municipal employee unions and black organizations are less expected and attest to the changes of the past several years that have not escaped these cities and the budgetary process. A similar pattern of interest group participation is described by the executive. 0n the average, just less than three groups (see table 10.3) make some budget demand. This too indicates a low level of involvement and repre- sents less than 21 percent of the potential number of organized groups present in the community. Differences among the cities are large, extending from an average of none to eight, but in eight cities, executive's report contact with less than three groups. The pattern of the type of group's involvement is similar. Business organ— izations and municipal unions (see table lO.2)-- twelve cities, civic service associations-— eleven, neighbor- hood groups-— nine, and then black organizations—— eight. Interest group participation in legislative decision making was measured by two separate questions, as a demarcation between formal and informal contact with the city council. Group representatives can appear TABLE 10.3 EXTERNAL PRESSURE EXECUTIVE # . # of . Group Ind. City of Group Mean Contacts Infl. Influence Infl. 01 2.33-1.16 1.50 0.00—0.00 0.50 0.79 02 0 oo—o.00 0.66 0.00-0.00 0.00 0.17 03 8.00-4.00 1.50 2.00—4.00 1.00 2.63 on 1.00—0.50 1.00 0.00—0.00 0.00 0.37 05 3.66—1.83 1.33 0.33-0.66 1.00 1.21 06 5.75-2.87 0.66 1.00—2.00 0.33 1,47 0? 1.50—0.75 1.50 0.50—1.00 1.50 1.18 08 5.00-2.50 0.75 0.25—0.50 0.25 1.00 09 3.66-1.83 1.66 0.33-0.66 1.66 1.47 10 3.00-1.50 1.00 1.33-2.66 1.00 1.54 11 2.25—1.12 1.00 0.50—1.00 1.00 1.03 12 2.50—1.25 1.50 1.00—2.00 1.00 1.44 13 1.50—0.75 1.25 0.25—0.50 1.00 0.88 l4 1.50—0.75 1.00 0.00—0.00 0.50 0.56 Mean 2.87—1.49 1.17 0.54-1.07 0.77 1.12 st, Dev. 2.17—1.08 0.34 0.60-1.20 0.51 0.61 340 before the whole legislature at the time of the official budget hearing; or they can contact individual councilmen outside of the public legislative chamber. A similar distinction is made by Richard Fenno, who not only notes these two different methods of interaction, but explicitly indicates the superiority of informal relationships as a point of access: . . . demands, however, are communicated through more effective channels than an appropriations hearing. And the same is true for demands of clientele groups. For one thing, informal com- munication outside the hearing is infinitely more effective . . . The political intelligence concerning support and opposition which members take in from outside witnesses at public hearings amounts to very little over and above that which they take in privately.8 The breadth of group involvement in the third stage of legislative budgeting reveals significant differences between them. The mean number of groups that appear before the council is just two, while on the average almost twice as many personally contact councilmen (see table 10.4). One half of the legislature's report less than three informal meetings with interest groups; ex- tending from the single city where all councilmen agree that they did not have any group contacts, to another City where the average number is eight. On the other hand, the number of representatives of groups appearing 8. Richard Fenno, Jr., The Power of the Purse: Appropriations in Congress (Boston: Little, Brown and Company. 1966). p. 351. ': I 341 TABLE 10.4 EXTERNAL PRESSURE (LEGISLATURE) - J # Group # Of Ind. City of , Group Mean Contacts Of Appear Infl. Influence Infl. 01 2.83-1.72 1.50—1.43 1.25 1.12-3.86 0.75 1.80 02 6.00-3.64 1.66-1.58 0.60 0.80-2.76 0.60 1,84 03 1.60—0.97 4.20-4.00 1.42 0.71-2.45 0.85 1.92 04 2.80-1.70 1.40-1.33 1.00 0.66-2.28 0.66 1.39 05 2.00-1.21 1.66-1.58 1.20 1.12—3.86 0.60 1.69 06 4.60—2.79 1.83—1.74 0.75 0.43-1.48 0.50 1.45 07 3.40-2.12 1.20-1.20 1.25 0.83-2.86 0.50 1,59 08 6.60—4.00 2.75-2.62 1.66 0.00-0.00 1.00 2.06 09 4.20-2.55 3.20-3.05 2.10 0.20-0.69 2.10 2.10 10 5.20-3.15 1.33-1.27 2.00 1.16-4.00 1.33 2.35 11 3.75—3.27 1.00—0.95 0.75 1.00-3.45 1.75 2.03 12 0.00-0.00 0.00-0.00 0.75 0.50-1.72 0.50 0.59 13 2.66-1.61 2.66-2.53 0.50 0.59-2.03 0.50 1.43 14 1.40-0.85 2.33-2.22 0.00 0.28-0.97 0.50 0,93 Mean 3.37-2.11 1.91-1.82 1.13 0.67—2.32 0.13 1.65 St . Dev. 1.85—1.17 1.05-1.00 0.53 0.36-1.25 0.38 0.48 T‘ to H 499' 9V“- .ul‘. .r “~v |__———__" 3A2 before the legislature during the public budget hearing, reaches a high of only four, and in nine cities less than two groups were present at this time. In ten cities, more groups made budget demands outside of the public hearings than in it, while in only two cities do more groups communicate their desires to the legislature through formal means than by informal means. In two cities the number is the same. Relying upon the public appearance of interest groups before the whole legislature considerably under- estimates the extent of their activity within the third stage of the municipal budget process. Interaction occurs through more informal social contact outside of the legislator's public role than in the structured, official points of decision making. As Betty Zisk writes of this mode of access: "The local 'group struggle' . . . appears 9 It appears that interest to be relatively intimate." group leaders themselves understand this difference be~ tween these two points of entry as these two mechanisms Of group participation are not statistically related to each other (0.0538). There is a disjuncture in the articulation of group demands to the city council. Ex— tensive informal relations are not associated with high 9. Betty Zisk, Local Interest Politics: A One Way Street (Indianapolis Indiana: Bobbs Merrill Publishers, 1973 . p. 28. 343 contact inside the hearing room. Interest groups making personal contact with individual members of the city council, do not make a further appearance before the legislature as a whole. Apparently this is unnecessary. Having one avenue of approach, subsequent activity is superfluous. Political parties in eight cities have informal relationships with individual councilmen, but in no city do they make a formal appearance before the legislature as a whole. Civic and service organizations display both forms of participation in five cities, but in six cities they make only informal contact (see table 10.5). But not all groups are so situated. Some community interests do not have the advantage of private entree to council members. Black groups have informal and formal contact in seven cities, but in only three cities are they involved only ‘ informally. Unions of municipal employees have informal interactions in only two cities, although in nine others they display both forms of contact. These last two groups need an additional mechanism to be heard, that apparently other segments of the community find unnecessary. Group Influence Having examined the participation of interest groups in the budgetary process, now the extent they are per- ceived by authoritative decision makers as being influential is investigated. This is not the statistical analysis 3m. TABLE 10.5 DISJUNCTURE OF GROUP CONTACT Both Informal Only Contact Formal Only Appearance Political Parties -— 8 -- Business 9 5 -- Neighborhood, Homeowners and Taxpayers 9 4 1 Civic and Service 5 6 2 Blacks 7 3 -- Municipal Employees' Union 9 2 345 between these measures of the part played by groups and actual expenditure outputs, but the subjective evaluation of the impact interest groups have upon the budget behavior of the three authoritative decision makers. The first question posed was a general assessment of the extent each actor's own, "budget decisions reflected the influence of community interest groups." Uniformly, only slight influence was attributed to this form of external pressure. This is certainly a low level of perceived group strength. While there is variation in their group importance, in few cities are groups considered to possess more than some influence in the budget process. A second question follows this general evaluation by asking for a precise specification of the names—— kinds of groups: "Who were particularly important in affecting your budget decisions." The responses to thiS' question corroborate the lack of group influence. The mean number of groups listed, was less than one. De— partments reported an average of less than one half, the executive of just one half, and the legislature an average 0f two thirds. In other words, most of those interviewed, did not point out a single specific interest group as having especially affected budget decisions.lo 10. The total number of specific kinds of interest groups that were mentioned by each set of actors, as Particularly important in making their budget decisions —___’—__T 1.11.111 11 346 As mentioned earlier, a demarcation between the activity of groups and their impact upon budget choices is made. The relationship between these two features shows that perceived influence partially follows from actual involvement. For the departments (0.6220 + 0.001) and the executive (0.6484 + 0.001) influence, strongly follows from participation. The more contact is made, the more groups are assessed to be important. The low level of attributed influence derives from the low level of group demands. But in the case of the legislature this relationship does not occur at all (—0.l824). Even separating the two different measures of group contact, there is no apparent relationship to the measure of influence it is-— 0.0968 to informal contact and 0.6679 to formal appearance. So in the case of department and execu- tive perceptions, the barrier to group influence is their absence from the decision making process. They are considered umimportant, because they are simply not heard from. When they do make a demand, they appear to be listened to and are evaluated as influential. However, 10. (continued) are: . Dept. Exec. Legis. Neighborhood, Homeowners 6 4 8 and Taxpayers Blacks 4 3 13 Unions of Municipal Workers 2 5 14 BuSiness l 1 11 Civic and Service - 2 6 Other 2 1 7 IIIIIIIII:_____________________________———————————————‘—”‘S=““""”—*E’ 347 in the case of the legislature, the barrier is not the failure to make their voices heard, but the legislature's insulation from the clamor of their activity. Having either informal and formal access to the city council is no guarantee of success, as the municipal legislature appears to be immune from external pressure. The last question asked to all three decision makers to round out the concept of external pressure was the, "influence of individuals from the community upon budget decisions." While this is not intended to represent the presence of a social and economic "power elite," budget demands can be articulated by individuals who are outside of organized interest groups. However, this does not appear to be the case. All three actors, uniformly assess them to have less than slight impor— tance in affecting their own budget choices. Combining the value of the separate questions into an index of the part played by external participants in budgeting, indicates that for the departments and the executive slight group and individual contact and influence is reported. In terms of the legislature, Some involvement and pressure is present. The municipal budget making process is not extensively connected with its environment. Through the articulation of budgetary demands by individuals and groups from the community. (r; (u m a», w 1, 348 Yearly Change in Expenditure Levels Regardless of what authoritative decision makers say, the impact of external pressure upon the actual spending outputs of the three budget stages as determined by the statistical association between them, is examined now. The activity and perceived influence of groups and individuals from the community is assumed to be an impetus for increased spending. These fourth actors serve as an input demand for more and better service, and consequently higher expenditures for the particular departments that represent their policy preferences: "People want more services, not less." John Crecine provides support for this interpretation of the specific direction group influence has upon budget outputs: All interest groups . . . feel that some depart- ment's budget should be increased. The pressures transmitted to the council concerning the operating budget are of one kind-- those advocating increases in the mayor's recommendations.1l These demands for program and expenditure expansion are conveyed not only to the city council, but to the department heads and executives as well: "All groups which made requests desired specific improvements in my department . . . Each group requested additional services 11. John P. Crecine, Governmental Problem Solving, p, 101. ‘ IIIII[IIII:______________________________________________——‘E:==:§é'—‘—_ii7 " 349 from this department for their particular group . . . Generally they ask for the inclusion of some project in my budget before I send it to the council." However, the data collected in the present research setting this does not demonstrate a systematic connection between group demands and expenditure outputs. In two of the three measurements of increase in spending, there is no apparent statistical relationship between them. The expansionary budget requests of departments does not derive from the external source (—0.0462); nor does the final appropriation ordinance grow any faster, as a result of this form of environmental pressure (—O.1253). Only in the case of executive recommendations is a moderate correlation reported. That is, the size of increases submitted to the legislature for review, and involving the fourth budget actor is 0.2055. Why the executive is any more susceptible to the entreaties of groups is unknown. If anything, it is expected that the popularly elected legislature would be more open to group pressure, but this is not the case. So on the whole this fourth set of actors is not able to affect the budget choices 0f authoritative decision makers. They are unsuccessful in obtaining expenditure increases for the departments they support. As Betty Zisk writes, although of a different measure of expenditure outputsz Per capita expenditure levels, in short, do not depend primarily on political demands , . . The 350 political translation process-- at least in terms of group activity-- is probably not relevant to the final level of expenditures for a given community . . .12 Such a conclusion can be extended to the present research. The growth that does occur in appropriations cannot be attributed to the influence of individual and groups from the community. The presence of external pressure is not directly translated into higher spending levels. Change in Expenditure Levels from One Stage to the Next Assuming that interest groups act as a positive impetus for spending, their presence within the budgetary process serves as a constraint upon the budget cutting behavior of executive and the legislature. Richard Fenno writes that: To the degree that outside witnesses appear, Committee members feel they will be put under that much more pressure to increase appropriations. The more they can keep outside witnesses at arms length, the easier will be their pursuit of budget—cutting.13 However, there also does not appear to be any consistent relationship between external pressure and spending decisions. In the case of the executive's review of initial depart- ment requests, the correlation (-O.2050) is barely high 12. Betty Zisk, Local Interest Politics: A One Way Street (Indianapolis, Indiana: Bobbs Merrill Publishers, 1973)) p0 1370 13. Richard Fenno, Power of the Purse, p. 341. €110 ”PW “.1 Q ‘4‘ a.-.\ I-\ 4.! .n .l. 61 0 h “D . S X e 111‘ in 5!. r . L +1. ,1 -1 .AU 1L nu a: d» FL 0 . 0... 0 at. m0 u - WM my... .NJ. w. L. \ s a .r 351 enough to reach the standard of a moderate association; but in the legislature's cuts, executive recommendations are not inhibited by the amount of pressure exerted on them by community-wide actions and actors (0.0420). Again there is little reason to expect the executive to be more affected by pressure from the environment than the legis— lature; but, in any case, the inconsistency of the relationship does little to corroborate the assumed connection between the influence of individuals and groups from the community upon municipal spending outputs. The explanation for the overall absence of an association between the involvement and influence of external participants and municipal expenditure outputs, revolve around the two dimensions of participation and in- fluence previously identified. Groups and individuals can be present, but ineffective in changing the behavior of authoritative decision makers. It is this situation, that seems to be described by Crecine, as he writes that public hearings are held by the city council, ". . . where interested parties plead their case and usually for in- creased expenditures for some purpose . . ."14 So groups do have the opportunity to at least make formal demands upOn the budget. But he then later concludes that, ". . . the 14. John P. Crecine, Governmental Problem Solving, pl 35. 352 council ignores specific or direct pressure."15 Budget demands are articulated, but ignored. Examining each of these dimensions in turn, several explanations can be offered for both the absence of participation and the lack of influence of groups and individuals in the budgetary process. Beginning with their involvement, external actors are simply not involved in the municipal budget process. They do not articulate demands, so their voice in support of spending increases is not heard. The particular questions that measured this aspect, showed that, indeed, few groups participate in any of the three budget stages. The level of external contact is low. Three specific explanations for the lack of activity can be offered. First, as the community is apathetic about govern- ment and politics in general, it should come as no surprise that there is little interest in the particular area of budgeting. Consequently, few contacts are made with PUblic officials to communicate particular spending preferences. Then, individual and group representatives are discouraged from participating in the budgetary process because they lack the capabilities to effectively and successfully affect budget decisions. The budget, full of figures, may be too complex, confusing, and difficult 15. John P. Crecine, Governmental Problem Solving, P. 219, 353 to understand. Community actors do not have the time, information, or expertise to decipher the mass of dollar and cents figures, the different account cate- gories, and the different funds to know what is going on to be able to make a specific budget demand. Finally, lack of involvement is evident, because groups and in- dividuals do not appreciate the significance of the budget for obtaining what they want from government. The connection between the determination of how much is going to be spent on each department and the amount and quality of services is not perceived: "I don't think the general public realizes the significance of the budget. That it is a statement of what programs will be carried out and that it is the central point to participate in city government . . . The budget document per se doesn't mean too much to individuals in the com- munity. They see it as a formality. They don't make their complaints in terms of budgets." So, the municipal budget process is isolated from external pressure. There is a low absolute level of organized activity specifically aimed at influencing budget decisions. There is a lack of contact between the public and municipal government: "The problem is that there is no direct link and contact with citizens. I'm not in a position to know and I‘m not sure the council is either to know what the public thinks is important." ———— 354 The lack of a connection between external pressure and budget outputs is the result of the absence of channels of communication. If the volume of demands were in- creased, these actors apparently would have more impact upon actual spending choices. If they were more vocal, they would be more effective. A second interpretation for this same set of statis- tical associations comes from a far different perspective. Demands are articulated, but are ineffective in changing the spending choices of public officials. The municipal budget making system is insulated from external pressure. Again three specific explanations for this situation are identified. There is first of all a general antipathy to the activities of interest groups. This has been previously reported by Betty Zisk as an underlying orienta- tion of public officials, acting as a, ". . . filter through which group efforts to influence public policy must pass. How accessible officials are to groups will depend in part on these predispositions."l6 As department heads said: "Our department is run by the charter, it's non-political . . . Our budget is prepared on the basis of the needs of the department rather than by outside influence and pres- sure . . . I won't put up with them. They are aware of that. I make it clear that they can't change things one 16. Betty Zisk, Local Interest Politics, p. 3. 355 iota." As two councilmen further commented: "We try to operate without being indebted to any pressure group or individual. There is no reason for them to be there. The budget is not their responsibility. We do not budget for outside groups and individuals." External actors are present and do articulate demands, but their voice is rejected. A second response, from the departments and the executive, is that external influence is transmitted through other channels, notably, the city council. Thus members of the executive branch of government, hide behind the policy/administration dichotomy and their "non-political" appearance to ward off interest group pressure. Departments heads passing it off to the executive and the legislature, state: "Not on my level. The place to put political pressure is on the manager and the council." Then the manager, in turn, forwards it to the City council. These are policy questions that belong to the legislature and this is the accepted and known road of accessibility . . . People tend to contact the city council if they have concerns about the budget." Finally, it may be that individuals and groups from the community and indeed, powerless to affect budget cut- Puts. They make a demand, but are ignored as they are too Weak to influence decision makers. As Betty Zisk writes: H . . . for most councilmen, the group struggle takes place 356 on a one way street [italics in original7 . . . rela— tively insulated and apolitical councilmen, where the search for 'right answers' prevails, regardless of the occasional clamor outside."l7 As described by decision makers, "The budget process is autonomous and enclosed. It is an internal affair not exposed to exter- nal group pressure . . . The budget is too far removed from the political process." This may come about because pressure at the time of the legislature's public budget hearing is too late: "By this time decisions are pretty well made. What is said at this point will not make much difference, it's too late to do anything . . . Historically budget decisions made by the council are final-— not open to change at this time." Or this powerlessness may derive from alienation from government: "People can't fight City hall . . . There is a lack of confidence in city government. People believe that 'none would pay any attention to them,’ that they can't Change a thing, that it is useless to do anything about it . . . There is a general attitude that the city will do what it wants to do anyway . . . They know it is going to pass whether they attend the hearing or not." Several explanations have been discussed for the 1?. Betty Zisk, Local Interest Politics. p. 143. 357 lack of group activity within the budget process and/pr the inability to influence spending outputs, but another explanation can be identified. It is not that groups and individuals are uninvolved or powerless, but that the direction of their influence is different from what has been assumed. They do not serve as a pressure for spending increases, but instead express demands to limit the size of annual growth. They act more as a revenue economizer, than as an expenditure maximizer. As a result of the concern for property tax rates, this situation exists. Consequently, demands are articulated, not to expand program and expenditure levels, but to hold the line on expenditures and thus hold the line on taxes. John Crecine provides some evidence that this may indeed, be the direction of interest group pressure. He writes that: ". . . [ihi§7 model suggests a convergence of voter and business interests helps to 'hold the line' on tax increases and hence tends to limit the total amount of resources available for government allocation."l8 He goes on to write that: The only case where external influence could be considered as imposing a decisional constraint is in the revenue estimate. Most systematic pressure from the business community concentrates on keeping tax rates constant and not on partic- ular expenditure items.l9 18. John P. Crecine, Governmental Problem Solving, p. 192. 19. John P. Crecine, Governmental Problem Solving, p, 92, |7——__—’U 358 By keeping the tax rate constant, the rate of expenditure growth is constricted. It will be recalled that business, homeowner, taxpayer, and neighborhood associations are the most common types of community interests represented within the municipal budget process. Both can be expected to articulate demands to contain the size of the budget: "They wanted the budget cut so that increases in taxes would not be necessary . . . they asked me to take a long hard look at demands for more money. To cut out government waste . . . don't spend any more money . . . just to say don't raise taxes." Only the influence of black organi- zations and unions of municipal employees can be unequivocably assumed to be favorable to spending increases. Since there is no way the present research can more rigorously test this hypothesized impact of external pres- sure, the absence of any uniform statistical connection with spending decisions suggests a multi-dimensional influence on government. There is both a positive and a negative thrust to group pressure. Thus far, the impact of individuals and groups from the community upon spending outputs has been examined as they directly participate within the three stages of the budget process. However, external pressure can be exercised through other mechanisms that do not directly involve decision making on the budget. First of all, community influence is exercised through elections. Both 359 the council and the mayor, plus specific bonds and referenda are affected by the voting public. Unfor- tunately this aspect of the budget making process was not examined in the present study. A second mechanism for external influence to be felt upon expenditure decisions is through the entire year and through the entire political policy making process, not just in the few months when the official budget document is prepared and adopted. The present conception of the budgetary process may be too abstracted and too divorced from other decision making processes that impinge upon it. Budgeting is more than a seasonal activity and may be less autonomous from other areas than the highly compartmentalized three stage model suggests: "As a general rule grOups do not concern themselves with the budget preparation and adoption. They are more prone to protest or express an opinion during the year . . , When specific items are presented at different times during the year . . . When specific items come up, then people make known their desire for services . . . Agitation for programs goes on throughout the year. Interest doesn't just focus on the budget . . . Throughout the year they talk to the manager and the council, but not specifically at budget time. They don't come at that time and say 'put this in the budget,‘ They ask for this program or that program over the year and we keep track of it.“ 360 Conclusion In summary, there are four possible situations that describe the part played by this fourth set of participants to the municipal budget process. These revolve around the two dimensions of participation and influence previously identified, and which is presented in the following contingency table: Participation Yes No Yes Pro or Con to Other mechanisms Expenditure (elections, dif- Influence Expansion ferent arenas) No Insulated Isolated Budget Budgetary Process Process FIGURE 10.1 EXTERNAL PARTICIPATION AND INFLUENCE First there is the situation where activity is reported and influence is evident as either support or Opposition to spending increases is evident. On the other hand, there can be no involvement and no influence as individual and groups from the 00mmunity are absent. The municipal budget process is isolated from its environment. Then there is no direct participation, but influence upon Spending outputs is evident through alternative channels 0f access. Finally, external actors are able to articulate 361 demands, but fail to change the budget behavior of authoritative decision makers. The municipal budget process is insulated from external pressure. Chapter Eleven: Decision Makers' Attitudes Toward the Environment Introduction A second mechanism for environmental conditions to cross the boundary line of the political system and enter into the budgetary arena is through the attitudes, beliefs, and values of authoritative decision makers. The subjective orientations of the department heads, the executive, and the legislature to the social and economic characteristics of the community can at times be more impor— tant than what those conditions objectively may in fact be. These perceptions serve as the intervening variable between) the inputs of social and economic environment and the political response of expenditure policy outputs. As Heinz Eulau and Kenneth Prewitt write: An environmental challenge calling for a policy response has the expected effect only if it is perceived by policy makers as constituting a problematic situation. Unless environmental challenges are experienced as problems, policy responses are not likely to be forthcoming.l l. Heinz Eulau and Kenneth Prewitt, Labyrinths 9: Democracy (Indianapolis, Indiana: Bobbs Merrill Publishers, 1973), p. 561. 362 1, ,, 363 Two kinds of environmental inputs are especially relevant to the study of governmental budgeting. The first is the amount of resources that are available through taxation. These provide the upper limits of the ability to act. Without money, government cannot spend. The second is the "needs" of the community. These are environ— mental conditions that generate demands for specific policies. Both of these are measured in terms of the orientations of public officials. What is interpreted as the available resources of government depends upon perceptions of the wealth of the community and appropriate tax levels; while what is perceived as the needs of the community and governments responsibilities in meeting I them depends upon the policy preferences and spending priorities of public officials. Much has been written of the fiscal plight of municipal governments across the country and the cities included within the present research are no exception to this general condition. If anything, the specific circumstances at the time of the study might have exacer- bated this situation. 1970 was a period of economic slowdown and the accompanying strike at General Motors might have produced an unusually severe financial crunch on these cities. In any event, it is necessary to examine the impact of this environmental condition upon the budgetary process. Money, certainly, is a significant 364 factor in making expenditure decisions. As John Crecine writes, ". . . the revenue constraint plays a key role in the budgetary process."2 The financial position of city government is des— cribed by the concept of Resource Capability. This is the amount of wealth present in the community that is potentially available to government through taxation. This is more exactly defined by Robert Eyestone as: "The maximum amount of income a city may expect yearly when it makes an effort to tap all available income sources . . ."3 Thus cities can be scaled from a situation of scarcity and poverty on one end to abun- dance and riches on the other. The problem inherent in this concept is how to determine if such an effort, "to tap all available income , sources,"u is being made. Is it an objective measure of the revenue system, or is it a subjective interpretation of budget actors? Eyestone examines these two alternatives and his conclusion provides the starting point for the present analysis. He identified six statistical indicators 2. John P. Crecine, Governmental Problem Solving: AgComputer Simulation of Municipal Budgeting (Chicago: Rand McNally, 19697. p. 179. 3. Robert Eyestone, The Threads of Public Policy: Ag§tudy of Policy Leadership (Indianapolis, Indiana: Bobbs Merrill Publishers, 1971), p. 30. 4. Robert Eyestone, Threads of Public Policy, p. 30. 365 of the financial condition of the municipality, such as the market value of property, the dependence on the property tax, and so forth only to find that they are not correlated with his measure of expenditure policy outputs. It is the subjective orientation of decision makers, and not objective features of the economic environment that is most important. As he and Heinz Eulau write: In this process of adjustment and control through appropriate policies the city's resource capabilities seem to_play only a limited part [italics ming/. It appears that policy-makers' willingness to tap city resources in order to adopt appropriate policies is a crucial component of the policy development process.5 What authoritative decision makers interpret the resource capability of the city to be is more significant in making spending choices, than what resources may actually exist. The assessment of the availability of money depends, above all, on the level of governmental activity desired. If satisfaction and acceptance of a low level of services is expressed, then the lack of funds is not a problem. On the other hand, if a higher level of programs are preferred, the limited supply of resources would then constitute a problem. The amount of wealth existing 5. Robert Eyestone and Heinz Eulau, "Policy Maps Of City Councils and Policy Outcomes: A Developmental Analysis," American Political Science Review, vol. LXVII, no, 1 (March, 1968), p. 143. 366 in the community that can be taxed by government, is not a self—evident condition. As Eulau and Prewitt write of this: Resource potential does not guarantee that governors will respond to environmental pressures and steer their city appropriately. Neither does the immediate availability or needed resources always develop efforts to deal with challenges . . . Policy makers respond to environ— mental pressures less in terms of resources available than in terms of their willingness to generate and mobilize them.6 Therefore, it is through the attitudes and values of the department heads, the executive, and the legislature, that the environmental input of resource capability affects the municipal budgetary process. The concept of resource capability can be broken down into two analytically distinct components. Beliefs about the financial environment can come from two dif- ferent sources: the actual inadequacy of the city tax base, and citizen demands for low taxes.7 Presently these are called the Expenditure Constraints and Revenue Consciousness, and are examined in turn. Expenditure Constraint The actual adequacy of the tax base means that there is a limited amount of money available to City government. 6. Heinz Eulau and Kenneth Prewitt, Labyrinths of Democrac (Indianapolis, Indiana: Bobbs Merrill Publishers, 1‘973 1 p. “'68. 7. Robert Eyestone, Threads of Public Policy, pp, 59-724’ | C. E . T. C .1. D. S . .L .1. .nu QC X e A» l. .1. nu; :1 n... .0; if... no . o4. AL 4.“ V1 0 _ y a: Go A... d L .1 I; no wlu mm 9 VJ .L 20 «t n“ . v Q. L. . u... . Fl. 0 0 w- o. .1. Q n . a . a. a. ... .. m. .. .. .. .. AL A» Cy Ax. .Fv 1. .1. »~ .A. .«n 367 The supply of resources existing in the community is constricted. There is not sufficient wealth in the community to support governmental expenditures, so that even a rich city, with a high tax rate, faces a shortage of funds to finance the activities of city government. In short, the yearly rise in the price of purchases, and the salaries and fringe benefits of employees, outstrip the annual growth the revenues provide. As Arnold Meltsner and Aaron Wildavsky describe this situation in Oakland: The main problem city officials try to solve every year is simply to find sufficient revenue to maintain their current payroll. Their problem is that the city's budget increases at a faster rate than the tax base. Fiscal atrophy is the city's chronic malady.8 Even without an expansion of the level of on-going programs, or the implementation of new ones, the costs of government are expanding faster than its resources. The dimension Of the expenditure constraint describes the scarcity or munificence of the financial environment in which spending Choices are made. RM —_ 8. Arnold J. Meltsner and Aaron Wildavsky, "Leave City Budgeting Alone: A Survey Case Study and Recommenda- ion and Reform," in John P. Crecine, Governmental Problem SOlving , p, 324. 368 Analysis Perceptions of expenditure constraint were measured by two questions (see tables 11.1, 11.2, and 11.3). The first asks for an assessment of the: "Effect the city's financial position had upon budget decisions." Responses to this open ended question were coded into five categories that extend from the absence of this environmental impact to its very severe presence. In the first case, no limitation of resources is perceived and no effect felt within the budgetary process. In the next category there is a closer review of the budget, as a more detailed and extensive examination of spending proposals occurs as a result of the tightening financial situation: "A great deal more care is being exercised in looking at the budget . . . It made the city council take a closer look at everything." Decision makers are more cautious in making expenditure commitments; but, as of yet, no adverse impact on appropriations is evident. The third category pinpoints such a deleterious effect on spending levels as larger reductions are imposed as a consequence of the perceived inadequacy of funds to support expansion: "Everything was cut as much as possible . . . We had a larger cutback this year than last year because funds did not increase in accordance with the amount of departments requests." Cuts are larger than they would have otherwise been; however, there is still TABLE 11.1 RESOURCE CAPABILITY (DEPARTMENT HEAD) Expenditure Constraintz/ Revenue Consciousness City Rev./Exp. . Who Set Tax Taxes Effect First Mean Pays Rate Too Mean High 01 2.33 1.00-4.00 3.16 1.00 4.66-1.33 1.33 1.22 02 1.33 1.00-4.00 2.66 1.66 3.66-2.33 0.33 1.44 03 2.00 0033—1033 1066 1000 3000-3000 1.33 1.78 04 1.50 1.00—4.00 2.75 2.00 3.33-2.66 0.00 1.89 05 3.00 1.00-4.00 3.50 2.66 2.33-3.66 1.33 2.22 06 2.66 0.50-2.00 2.33 1.00 2.33-3.66 2.33 2.33 07 1.00 1.00-4.00 2.50 1.66 5.00-1.00 1.33 1.33 08 4.00 1.00-4.00 4.00 3.50 5.00-1.00 0.00 1.50 09 1.00 1.00-4.00 2.50 3.00 4.66-1.33 0.00 1.44 10 3.66 0.66-2.66 3.16 0.60 5.00-1.00 0.00 0.50 11 2.00 0.66-2.66 2.33 1.00 5.00-1.00 1.00 1.00 12 3.00 1.00-4.00 3.50 0.50 4.00—2.00 0.50 1.00 13 3.33 0.50—2.00 2.66 3.50 4.66-1.33 0.50 1.78 14 0.00 0.00-0.00 0.00 1.33 2.00-4.00 1.00 2.11 Mean 2.03 0.74-2.96 2.62 1.74 3.90-2.10 0.78 1.54 St. Dev. 1.29 0.33-1.33 0.96 1.04 1.12-1.12 0.68 0.52 TABLE 11.2 370 RESOURCE CAPABILITY (EXECUTIVE) Expenditure Constraint// Revenue Consciousness City Rev./Exp. . Who Set Tax Taxes , Effect First Mean Pays Rate Too Mean High 01 1.00 1.00—4.00 2.50 1.50 1.33-4.66 0.50 2.22 02 1.33 0.66-2.66 2.00 1.66 2.00—4.00 1.00 2.22 03 0.50 1.00—4.00 2.25 1.50 2.00—4.00 0.00 1.83 04 0.00 1.00-4.00 2.00 0.00 2.50-3.50 0.00 1.17 05 2.00 1.00-4.00 3.00 1.00 2.00-4.00 1.33 2.11 06 3.00 1.00-4.00 3.50 0.00 2.75—3.25 0.66 1.64 0? 0.50 1.00-4.00 2.25 1.00 1.00-5.00 1.50 2.50 08 4.00 1.00—4.00 4.00 1.75 1.00-5.00 0.75 2.50 09 2.75 0.33-1.33 2.04 3.00 2.33-3.66 1.00 2.55 10 2.66 0.50—2.00 2.33 1.00 2.66-3.44 1.50 1.94 11 2.25 1.00—4.00 3.12 1.00 2.50-3.50 0.50 1,66 12 3.00 0.50—2.00 2.53 1.00 2.50—3.50 2.50 2.33 13 2.66 2.00—4.00 3.33 2.75 1.33-4.66 1.25 2.89 14 2.00 0.50-2.00 2.00 1.50 2.50-3.50 3.00 2.66 Mean 1.97 0.82-3.28 2.63 1.33 2.03-3.97 0.85 2.16 St. Dev. 1.16 0.26-1.26 0.65 0.84 0.62-0.62 0.85 0.47 371 TABLE 11.3 Y+ T I .L.) HE R AU D.T AA CL S E1 mm UL O/I\ S E R Expenditure Constraint// Revenue Consciousness City Mean Taxes Too Set Tax Rate ow ha WP n a e M m. mt //S .r v.1 9F R .t C e f :1 my High 00830553000505 muesoiRibnlorqiannUOIK/7. .............. 31202103131212 ndnuzj7kunu7kunu9LDnuéib 56115012827512 ....... O I O O O I O iaeeaeieazzzzz .. . .. n2nuzJ31Unu310nu1LRiUcJK/ L4hwRZ8anuRibn2n/925sbn/ O O O O O 0 C ....... 234a2o¥441h43fi4oj313o233 5ZUn¥annu57OnuOZUKJOLD 22562076805257 .............. n21llflnaoalrl1l3iuoanfl 18e392510h4RfOn0018e3956 ,6n9922h4115iDmflqu7L4qx O O O ........... A29231192722nJ113floa9T1 nunZUnunZUnunYOnunZOAUno 0Y4nuniznun00nunionun60 ..... O O O O O O O O O amumaawwauaamm (JOZUnunZUnunZUnUn0510nu 76008500505705 0 O O O O ......... 0A011120nulflnulf0nu110 n/nuqiunuRibnunVOHDnu5il AJKH4HDAO9TlnunZZn(K