DOCUMENT RESUME ED 337 863 EA 023 382 AUTHOR Wohlstetter, Priscilla TITLE Assessing Legislative Control of Bureaucracy: The Implementation Contract. PUB DATE 87 NOTE 50p.; Paper presented at the Annual Meeting of the Association for Public Policy Analysis and Management (Bethesda, MD, October 29-31, 1987). PUB TYPE Speeches/Conference Papers (150)-- Reports - Research/Technical (143) EDRS PRICE MF01/PCO2 Plus Postage. DESCRIPTORS *Educational Assessment; Elementary Secondary Education; *Legislators; *Politics of Education; *Program Administration; *Program Implementation; State Agencies; State Government IDENTIFIERS *State Legislatures ABSTRACT Findings from a study that explored the motivations of legislators to oversee and control program implementationare presented in this paper. Implementation is viewed as a contract between the legislative and executive branches with legislators acting as monitors who use oversight to control the agencies contracted to implement policy. A total of 57 elite interviews with legislators, committee staff, and legislative agency staff were conducted in 6 states--Arizona, California, Florida, Georgia, Minnesota, and Pennsylvania--to examine the constraints of legislative monitoring, the resources for legislative control of implementing agencies, and the extent to which legislaturesuse available resources to oversee implementation contracts. Principal-agent theory suggests that in all contractual reidationships information is distributed unevenly among the contracting parties, skewed in favor of the agent. Data from the states illustrate how legislatures drew on their institutional and politicalresources to enhance their access to information and their controlover contract implementation. Three tables are included. (27 references) (LNI) ** ***** ***************** ***** ************************************** **** Reproductions supplied by EDRS are the best that can be made from the original document. ********************* ***** ***************************** ***** *********** ASSESSING LEGISLATIVE CONTROL OFBUREAUCRACY: THE IMPLEMENTATION CONTRACT Priscilla Wohlstetter University of Southern California U5 DEPARTMENT OP IDUCATPON PERMISSION TO REPRODUCE THIS °A.( O eucewnel Reseertn and Improvement MATERIAL HAS BEEN GRANTED BY EDUCATIONAL RESOURCES INFORMATION CENTER ElliCt 11/1:nls ClOcurnent111819 been reproduced es received trees the perSOri Or Ct3913n,3100IN onometing 4 M..10r changes hive beenmod* 10 irriprO3 rIPP,OduChOr euelrfr bornisof vee or 001ruort$StetedindusdoCu ment do net necessar4v represent otirCial TO THE EDUCATIONAL RESOURCES OfR1 poS4ron or policy INFORMATION CENTER IERIC)" This paper was presentedat the 1987 Association for Public Policy Analysis and Management meetingin Bethesda, Maryland, October 29-31. The paper was written as part ofa research project at The RAND Corporation,sponsored by the Center for Policy Research in Education, andfunded by the U.S. Department of Education, grant numberOERI-6-86-0011. The opinions expressed are those of the author,not of the sponsoring institutions. 2 BEST COPY AVAILABLE ABSTRACT This study explores themotivations of legislatorsto oversee and control program implementation. Implementation is viewed as a contract betweenthe legislative and executive branches with legislatorsacting as monitors whouse oversight to control the agencies contractedto implement policy. Elite interviews in sixstates with legislators, committee staff, and legislativeagency staff examined the constraints that impede legislativemonitoring; theresources that can help legislatures control implementingagencies; and the extent to which legislaturesuse available resources tooversee implementation contracts. Principal-agent theory suggeststhat in all contractualrelationships information isdistributed unevenly among the contractingparties, skewed in favor of the agent. Data from the states illustratehow legislatures drewon their institutional andpolitical resources to enhance their access to information and theircontrol over contract implementation. INTROWCTION In enacting laws that authorize the creationof programs, legislatures initiate contracts foxservices with implementing agencies. The implementation contract isan explicit agreement between the legislative and executivebranches, under which each party promises to meet the conditionsset forth in the legislation. Agencies that accept implementation moniesfrom legislatures in effect are agreeingto the terms of the contract and confirming they are capable offulfilling its mandates. As is true of any contract, both partiesreceive some value for their participation. In the relationship between the legislature and an implementingagency, the former procures services while the latter receives monies. During program implementation thelegislature, as contract principal, attempts to exert controlover the agency or agent so that its actions will produce theintended results. One common aspect of all contractual relationships isthat information is distributed unevenly or asymmetricallybetween or among the contracting parties, skewed in favorof the agent (Moe, 1984). This complicates the principal's job ofmonitoring the contract. Agents are likely to be selective inthe information they provide to principals. Agents also can insulate themselves from principals, particularly if thereare multiple (sometimes competing) principals,as when several legislative committees share responsibility for monitoringa single agency. This paper explores the principal-agentrelationship in the context of state education reform. Principal-agent theory is an 4 2 Iconomic approach to the study oforganizations that has been used by political scientists to analyzethe behavior of public bureaucracies. In the language of the theory, one party,the principal, enters into a contractual relationshipwith another, the agent, with the expectation thatthe agent will pursue the principal's interests and objectives inan efficient manner, much as would the principal himself. Principals may contract with the agents because they lack specialized knowledgeor because the task is large and complex. Examples of principal-agent relationships in the publicsector include citizens-politicians, legislators-bureaucrats, and bureaucraticsuperiors-bureaucratic subordinates. Studies by political economistshave used principal-agent theory to explore the contractual relationshipbetween Congress and the executive and to explaincongressional oversight behavior (Moe, 1984; Weingast, 1984; McCubbinset al., 1987). Previous research has applied the theoryto only one type of agent--- federal regulatory agencies--- and most have been single case studies. The investigation reported here is distinguishedby its focus on service agencies at thestate level and by comparisons across states of principals and agents involved in implementing substantively similar contracts. The discussion that follows is basedon an investigation of education reforms in Arizona, California,Florida, Georgia, Minnesota, and Pennsylvania. The research, which was conducted in spring 1987, included 57 elite interviews withlegislators, committee staff, and legislativeagency staff with responsibility 3 for overL.eeing educationreforms enacted in the past fouryears. The sample was selectedto highlight various approachesto education reform (e.g., comprehensiveversus incremental); regional diversity; and differentstages of implementation. Some of the reforms are aimedat raising academic standardsfor students, while others have changedhow teachers are trained, recruited, and compensated inan effort to improve teacher quality. The paper is divided intothree sections. The first reviews institutional and politicalconstraints that may impede legislative monitoring ofcontract implementation by agencies. The second section examinesthe resources available to state legislatures that can aid theeffort to control and influence implementing agencies. The third sectionassesses the extent to which legislatures in thesix sample states made effectiveuse of available resources in overseeingimplementation of legislative contracts for education reform. There are recent indications(Aberbach, 1979; Rosenthal, 1981; Ethridge, 1984; andMcCubbins and Schwartz, 1984,among others) that the incidence oflegislative oversight of implementing agencies is increasing. Legislatures also have expanded their powers to controlagency decisionmaking (Ethridge, 1984; Council of StateGovernments, 1986; McCubbins et al., 1987). The present study suggests thatlegislatures draw on their institutional and politicalresources to enhance their access to information and their controlover contract implementation. t; 4 CONSTRAINTS ON CONTROL BYLEGISLATURES The ability of legislaturesto monitor and control agencies responsible for implementingprograms created by law is constrained by institutional and political characteristicsof the system of stategovernment, including the part-timeoperation of most state legislativebodies, the short timeperspective of legislators, the collectivenature of legislative mandates,and the fragmentation ofboth program implementationand control. As a result of this combinationof factors, information ishot readily available tolegislatures for oversightpurposes, and control of the implementationprocess often is neither sustained nor strong. Part-Time Legislatures Contract principals mustdevote considerable time and resources to the monitoringand control of implementingagents. With state legislaturesas principals, the difficulty of thetask is exac,rbated