INFORMATION TO USERS While the most advanced technology has been used to photograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. For example: • Manuscript pages may have indistinct print. In such cases, the best available copy has been filmed. • Manuscripts may not always be complete. In such cases, a note will indicate that it is not possible to obtain missing pages. • Copyrighted material may have been removed from the manuscript. In such cases, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, and charts) are photographed by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is also film ed as one exposure and is available, for an additional charge, as a standard 35mm slide or as a 17”x 23” black and white photographic print. Most photographs reproduce acceptably on positive microfilm or microfiche but lack the clarity on xerographic copies made from the microfilm. For an additional charge, 35mm slides of 6”x 9” black and white photographic prints are available for any photographs or illustrations that cannot be reproduced satisfactorily by xerography. 8707097 B e n s o n , G aye G ilb ert RATIONAL ACTORS AND ADMINISTRATIVE RULES: THE STATE OF MICHIGAN, 1972-1984 Ph.D. M ich ig a n State U niversity University Microfilms International LEGISLATIVE VETO IN 300 N. Zeeb Road, Ann Arbor, Ml 4B106 Copyright 1986 by Benson, Gaye Gilbert All Rights Reserved 1986 PLEASE NOTE: In all c a s e s this material has been filmed in the best possible w ay from the available copy. Problems encountered with this docum ent have been identified here with a ch eck mark V 1. Glossy photographs or p a g e s ______ 2. Colored illustrations, paper or print_______ 3. Photographs with dark background_____ 4. Illustrations are poor c o p y _______ 5. Pages with black marks, not original c o p y ______ 6. Print show s through as there is text on both sides of p a g e ________ 7. Indistinct, broken or small print on several pages 8. Print exceed s margin requirem ents______ 9. Tightly bound co p y with print lost in s p in e ________ 10. Computer printout p a g es with indistinct print_______ 11. P a g e (s)_____________ lacking w hen material received, and not available from sch ool or author, 12. P a g e (s)_____________ seem to b e missing in numbering only as text follows. 13. Two pages n u m b ered_______ . Text follows. 14. Curling and wrinkled p a g e s ______ 15. Dissertation con tain s p ages with print at a slant, filmed a s received 16. Other__________________________________ . S ' ________________________________________ University Microfilms International RATIONAL ACTORS AND ADMINISTRATIVE HOLES: LEGISLATIVE VETO IN THE STATE OF MICHIGAN, 1972-1984 By Gaye Gilbert Benson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Political Science 1986 ©Copyright by GAYE GILBERT BENSON 1986 ABSTRACT RATIONAL ACTORS AND ADMINISTRATIVE RULES: LBGISIATIVE VETO IN THE STATE OP MICHIGAN, 1972-1984 By Gaye Gilbert Benson The legislative oversight literature posits two models. The "runaway” model says legislators lack incentives to provide adequate oversight of bureaucrats. The "capture" model says adequate oversight already exists, and more overt oversight would be counterproductive. Empirically it is difficult to distinguish between the two; each predicts close to 100% success for administrative proposals referred to the legislature. This research poses a third, "constrained”, model in which overt institutional oversight mechanisms attract substantial legislator commitment, significantly lowering approval rates. Each model is tested against the 1972-1984 experience of Michigan, which has the U.S.'s most extensive and severe legislative review of administrative rule promulgation. Key data include 1,906 proposed Administrative Code changes reviewed by Michigan's Joint Committee on Administrative Rules (JCAR), meeting observations, interviews, and decision rules changes, including form of the legislative veto. Overt oversight stringency increased throughout the study period. More stringent mechanisms— type veto rule, majority requirements, review period duration— were uniformly associated with reduced approvals, eventually less than 70%. Review authority was centralized in and the Gaye Gilbert Benson almost exclusive authority of the JCAR which attracted high seniority legislators. Bureaucratic strategic behavior helped but did not guarantee success. Most previously withdrawn or disapproved rules were eventually approved in some form; adjusted approval rates exceeded 90%. Economic groups are regular review process participants with others * also quite visible. The process affords the public a final chance to influence administrative policy in a political arena perhaps more sensitive to their concerns. The committee frequently required the resolution of conflicts prior to taking action. Participants are forced to reach self-determined compromise which legislators then endorse, thereby minimizing electoral risks. Neither the "runaway" nor the "capture" theory fit the findings. Michigan's increasingly centralized and substantive oversight structure resulted in high levels of personal investment, greater legislative control over the promulgation of administrative rules, lower approval rates, and more "constrained" bureaucrats. To the memory of ray grandparents— I. Hep. G. and B. B. G., M. A. F. and P. W. P., and to my parents— H. P. G. and A. R. G., for their cornnitment to learning and the sharing of it. iii Acknowledgements The work presented in this dissertation reflects the concern, support, and interest of many people. It is only proper that they be given recognition. First, my thanks to Dr. Gary J. Miller, who chaired my dissertation ccnmittee and served as a member of my program guidance committee, and from whom I had the pleasure of taking two classes. Having come to the doctoral program from outside the discipline, it was my special good fortune to have Professor niller introduce me to the world of public administration as a subject of academic attention. I was soon captured by a field I had presumed to be dry and of little personal interest. As a menber of my guidance committee, he helped shape a substantial, dynamic program for me. As for the dissertation, he introduced me to the topic as something which cut across my various interests and one I might like to work on. It took only a small amount of preliminary research to convince me the topic was feasible as a dissertation, personally interesting, and inportant. The work which followed was enriched by his insight, patience, encouragement, and good humor. There is no amount of thanks which could repay him; I can only hope that my own relationships with students will be a credit to his good example. Second, my gratitude to all those persons in government and outside V who shared so willingly their time, expertise, and experience. Special thanks are due Representative Michael J. Griffin, who headed the Joint Conmittee on Administrative Rules during the time 1 was interviewing and attending hearings. His unfailing courtesy and cooperation were invaluable; without them this would have been a very different work, if even possible. Menbers of the joint conmittee staff, at that time special counsels Kenneth Sanders and Carol Smith and administrative assistant Coimey Klingbeil, also played a special role. Their active Interest and cooperation and Conney's excellent records, including the "Daily Status Report", were critical components of the project. Thanks also to Senator Ed Fredricks, alternate chairperson during the time I was doing my work and now head of the committee, and all the other current and former meirbers of the Joint Committee who shared their time with me. They all are, indeed, public servants, though they differ in their definitions of both the public and the best way to serve it. Seeing the world from their multiple perspectives was a fascinating experience and I thank them for making it possible. Likewise, icy encounters and extended discussions with other actors— bureaucrats and interested members of the public alike— expanded my understanding of the frustrations, rewards, and excitement of trying to make government work, xhe cooperation of these people broadened ray approach and enriched the results; ray thanks to each of them. Four other menbers of the Michigan State university faculty made special contributions to this dissertation. First is Dr. Ada Finifter, who chaired my program guidance conmittee and from whom I took classes in voting behavior and political socialization. The research requirements of those courses especially and her meticulous attention to detail helped prepare me for the dissertation. She helped me think of myself as a researcher, to identify interesting questions, and to pursue them carefully and honestly. For all of these things I am grateful, as well as her general support and continuing interest in my progress. Second, I would like to express my appreciation to urs. Jack Knott, Charles Press, and Joseph Schlesinger, members of my dissertation conmittee. Dr. Knott was especially helping in clarifying underlying theoretical Issues particularly in the early stages of the work. Drs. Press and Schlesinger on the other hand, veteran observers of (and sometime participants in) the Michigan political scene, helped me greatly in understanding this specific oversight environment. All were generous with their time and I thank them. I entered the doctoral program never having seen a computer. With the assistance of Harriet egislature (Same Chamber). 96 Movement out of JCAR, 1973-1981. 98 1.2. 1.3. 3.1 3.4 3*5 3.6 3.7 3.8 3.9 3.10 4.1 1972-1982. Total Transmittals and Overall approval Rates, by Department and Frequency, 1972-1982. x 121 xi 4.2 Department Approval Rates by Time Period. 122 4.3 Outcomes Aggregated by Average Frequency of Transmittals by Department, 1972-1982, and by Veto Period. 124 Per Cent of Rules Transmittals received by JCAR in Month rrior to Sumner Legislative Recess, by Year. 126 Subsequent Outcomes of *uiles Transmittals originally With­ drawn or Otherwise not Passed by the JCAR, by Veto Period, 135 Incidence of Resubmittal after unfavorable JCAR Action, by Time Period. 136 Outcome of Department of Natural resources Rules Trans­ mittals by Type of Rule, 1972-1982. 139 4.4 4.5 4.6 4.7 4.8 6.1 6.2 6.3 A—1 A-2 Nunber of Emergency Rules and Regular Rules Filed, by Year, 1972-1982. 155 JCAR Disposition of Rules Transmittals by Party Control (Legislature and Governorship): 1978-1983. 184 JCAR Disposition of Transmittals by legislative Party Control and Incidence of Inpasse, 1983-1984. 191 Nunber of Existing Rules Reviewed, by Department and Identifying Source (Internal Review or Public Complaint) by Reported Department Response as of July 1984. 196 Administrative Rule Proposals transmitted to the JCAR with Administrative Code Number, Dates Received and Acted Upon, Days Elapsed, and Outcome: 1977. 233 Annual NUrrber of Rules Trasnmlttals and Per Cent Approved for Twelve Department (including additional DNR breakdown) by Veto Period, 1972-1982. 237 LIST OP FIGURES Process for Promulgation of an Administrative Rule, State of Michigan: 1978-Present. Chapter 1 Introduction and Literature Review The research documented in this dissertation focuses on a particular point of intersection between the legislative and executive branches— the legislative oversight of administrative rulemaking. It examines incentives for, dynamics of, participation in, and the outcomes of rule oversight, using data from Michigan, the state which currently practices the most ambitious form of legislative oversight in the country. Special attention is given to the effects of two different forms of the legislative veto, the "negative" form, requiring negative action by the legislature to deny approval of a legislative rule, and the "reverse" form, where no rule takes effect without prior legislative approval. Three theories of legislative oversight are evaluated. The first two— "legislative capture" and "runaway bureaucracy"— the major competing theories within current academic literature regarding oversight incentives and implementation, offer contradictory and apparently irresolvably different explanations for what is seen as limited oversight. The third theory, that of the "constrained bureaucrat", is derived from observations of current oversight practice particularly at the state level, practice which has resulted in 1 2 strengthened state oversight mechanisms. The study takes us beyond the confines of the congressional model under which most oversight study has been done to a setting characterized by substantive legislative veto authority, centralized and conprehensive systematic review, and a series of increasingly more stringent decision rules which provide for the relatively controlled analysis of the effects of differing structural mechanisms. 1.1 Introduction Governments, from their inception, have promulgated rules and regulations, varying in subject, stringency, method, goal, and success, the existence of government rules and regulations nonetheless has been one of the certitudes of life. American political theory says the legislatures set major policy but, under the doctrine of delegated powers, that they may delegate authority to executive branch administrators to develop the details required for implementation. Consequently, rulemaking and rule implementation have become primary functions of governmental bureaucracies in this nation. At the federal level, bureaucrats decide such things as conditions for the conduct of interstate comnerce, allowable levels of certain air or water pollutants, and standards for reimbursement for Medicare service providers. At the state level, similar types of regulation occur as well as regulation in areas which are almost exclusively in the 3 states' province, especially the regulation of professions, in the state case which will be examined here, bureaucrats develop standards in many different policy areas— for the regulation of comnerce (e.g., comnercial fishing, real estate schools, horse racing, and consumer protection rules) and environmental and public health (air pollution control, disposal of diseased animals, elevator safety, fire fighting, and electrical code rules), control of the political process itself (mechanics of campaign finance reporting), and, as suggested, the regulation of professions (pharmacists, social workers, and nursing home administrators.) (See Appendix A for a sequential listing of new Michigan rules transmittals, including the above examples, during a typical twelve-month period.) The doctrine of delegated powers, however, inplies a continuing legislative responsibility for rules developed within the executive branch; thus, legislatures are looked to as the overseers of administrative activity in this and other areas. While growth of the administrative state in the latter twentieth century may have theoretically increased the inportance of such oversight, academic analysis of oversight incentives, mechanisms, and efficacy has been limited. At this point, the literature offers two contradictory perspectives on these issues— bureaucracies are characterized as either "runaway" or, alternatively, "captured." "Runaway" bureaucracy. The traditional view holds that legislative oversight is much lauded but little practiced. Legislator-constituent incentive structures focus on other aspects of legislative responsibility, leaving little motivation for investment of personal or institutional resources in the oversight function. Conventional 4 oversight mechanisms— appropriations, program audits, and investigatory hearings, for exanple— are believed to be cumbersome, sporadic, and largely ineffectual. Bureaucrats are left to pursue their own goals, essentially unchecked by their supposed legislative overseers. The result is the often discussed "runaway" bureaucracy and little hope for change.1 "Captured" bureaucracy. A more recent model, growing out of the rational choice perspective, asserts that a highly effective system of legislative oversight is already in place. Program approval and appropriations sanctions operating through the standing committees keep administrative agencies in line. Far from being runaways out of legislative control, bureaucrats in this model act so as to provide electoral benefits to their legislative overseers, who in turn, supply the resources for continuing agency existence. In this view, the "captured" bureaucrats are already doing what legislators want; more overt oversight would be redundant and counter-productive. This dissertation presents a course of research aimed at helping resolve the dilemma of these two competing and contradictory theories. 1. See David R. Mayhew, Congress; The Electoral Connection (New Raven: Yale University Press, 1974), pp.120-125; Morris S. Ogul, Congress Oversees the Bureaucracy (Pittsburgh: University of Pittsburgh Press, 1976), pp.181-182; Alan Rosenthal, "Legislative Behavior and Legislative Oversight," Legislative Studies Quarterly 6 (February, 1981): 115-116; Seymour Scher, "Conditions for Legislative Control," Journal of Politics 25(1963): 531. 2. See Morris S. Ogul, "Congressional Oversight: structures and Incentives," in Congress Reconsidered. Lawrence C. Dodd and Bruce Oppenheimer, eds., (New York: Praeger Publishers, 1977) pp.212-222; and Barry Weingast and Mark J. Moran, "The Myth of the Runaway Bureaucracy: The Case of the ETC", St. Louis: Washington University Center for the Study of American Business, Formal Publication No. 49, 1982. 5 It does so, in part, by proposing yet a third theory— that of the "constrained" bureaucracy, to be introduced in the next section. 1.2 The Research Problem Hie legislative veto of administrative rules represents a particularly interesting nexus in the legislative-administrative relationship. It allows direct legislative intervention in a delegated administrative function. Yet it has received relatively little systematic academic attention outside of the law journals.^ If, in fact, the legislative veto represents a notable point of influence over administrative decision making, we need to better understand the 3. It has been argued that: "Hie legislative veto can be viewed as a mechanism to help fill the void left by the decline of the delegation doctrinei.e., the increasing tendency of Congress to delegate authority without clearly specifying accompanying policy standards. See Harold H. Bruff and Ernest Gellhorn, "Congressional Control of Administrative Regulations: A Study of Legislative Vetoes," Harvard Law Review 90 (1977), pp.1372-73. 4. Hie bulk of the literature, most of it in law journals, has focused on normative issues— whether the legislative veto is constitutional, represents an undue intrusion of the legislature into administrative prerogatives, etc.— or presents basically atheoretical accounts of this or other control devices. See, for instance, Bolton (1977), Kaiser (1980), Schwartz (1978), and W&tson (1975). Standard texts in the field typically devote a few paragraphs at most to the legislative veto (e.g., Randall B. Ripley and Grace A. Franklin, Congress, the Bureaucracy, and Public Policy. Homewood, IL: Hie Dorsey Press, 1980, pp.74-75, and, at the state level, Charles Press and Kenneth VerBurg, State and Conminitv Governments in the Federal SvBtem. 2nd edition, New York: John Wiley & Sons, pp.241-242, 365-366.) A few enpirical studies have appeared in the last few years, generally in the political science literature, and will be discussed later in this chapter. 6 dynamics surrounding it and its substantive effects. Alternatively, if it represents a point of system redundancy or a counter-productive investment of legislative resources, we need to understand the incentives which support such an arrangement and the consequences of such a structure. At least two factors contribute to the failure to resolve the fundamental contradictions between the models outlined in the previous section. First, much of the discussion has been from an institutional perspective, centering on issues regarding the constitutionally appropriate roles of the respective branches in this particular 5 relationship. Even the nuch discussed 1983 Supreme Court in s vs. Chadha decision which invalidated most requirements of prior Congressional review or approval of federal level administrative rules and regulations was decided on formalistic, institutional grounds, left unaddressed the behavioral and theoretical conflict identified here.6 Its main findings were that Congressional review as it then operated violated two fundamental principles. First, it violated the requirements of legislative action by allowing a single house to sometimes exercise the (legislative) veto authority alone. Second, the legislative act of veto was never subsequently referred to the president for signature or executive veto; this lack of presentment, the Court found, violated rights of executive participation in the legislative process. This decision, while indeed addressing federal constitutional 5. Bolton, 1977; Kaiser, 1980; Schwartz, 1978; Watson, 1975. 6. See Imnigration and Naturalization Service v. U.S. Supreme Court, 1983. Chadha 103 SC 2764, 7 and formalistic, institutional, issues gave us no further understanding of the enpirical merits of different approaches. about the success of the veto as it had operated consequences might be of removing the tool. Ittold us nothing norwhat the Moreover, it had no direct bearing on legislative oversight at the state level where individual state constitutions determined the appropriate relationships between the branches. Second, even without the problem of the general institutional focus, there remains a serious problem in identifying either model in operation. Both models would lead us to predict administrative rule approval rates approaching 100%. In the first case, there is no effective control over the "runaway" bureaucrats andthey do as they wish; in the second, bureaucrats act in anticipation of legislator wishes and provide the rules preferred by their overseers. Because of this, as Weingast notes, an effective standing committee oversight system will be difficult to distinguish in process from an ineffective one. If COTndttee-based oversight works well and "require[s] the attention of Congress only when something goes wrong . . . then congressional "oversight* will appear sporadic, ad hoc, and without 7 systematic influence". The enpirical problem may be related to the limitations of a restricted data base. Enpirical studies to date have focused almost 7. Barry R. Weingast. "A principal Agent Perspective on Congressional-Bureaucratic Relations (with Enpirical Applications to the SEC'S recent deregulation of the Mew York stock Exchange)," paper prepared for delivery at the Fifth Carnegie Conference on Political Economy, Pittsburgh, Pennsylvania, June 10-11, 1983, p.10. Emphasis mine. 8 exclusively on federal level institutional structures and relationships. Yet throughout the 1970s, state requirements for the legislative review of rules were increasingly coirmon, either on a statute-by-statute basis or through establishment of procedures for routine review but the states had not limited themselves to the Q Congressional model. By 1982, forty-one states provided for some form of direct Q legislative oversight of administrative rule making. Of these, twelve states used a procedure which also required gubernatorial involvement while thirteen allowed disapproval of an administrative rule if sustained by both houses. One state (Oklahoma) allowed disapproval by one house only and six provided for disapproval by review conmittee (s) 8. For reviews of past and current state practices concerning legislative oversight of administrative rules see Keith E. Hamm and Roby D. Robertson, "Factors Influencing the Adoption of New Methods of Legislative Oversight in the D.S. States," Legislative Studies Quarterly VI:1 (February, 1981):135-138; Rich Jones, "Legislative Review of Administrative Rules: An Update," State Legislative Report— Legislative Management Series 7:4 (April, 1982) (Denver:National Council of state Legislatures) i R. Bradley Lambert, "Coitfnent— The Legislative Veto: A Survey, Constitutional Analysis, and Enpirical study of its Effects in Michigan," Wavne law Review 29 (Fall, 1982):92-95? David S. Neslin, "Quis Custodiet Ipsos Custodes?:* Gubernatorial and Legislative Review of Agency Rulemaking under the 1981 Model Act," Washington Law Review 57 (1982):672-675; and Bernard Schwartz, "The Legislative Veto and the Constitution— A Reexamination," George Washington Law Review 46 (March, 1978): 354-364. 9. According to Jones (1982), the states without specific provisions for legislative review of administrative rules included California, Delaware, Indiana, Massachusetts, Mississippi, New Mexico, Pennsylvania, Rhode Island, and Utah. 9 action alone. One state (Michigan) required prior approval of all rules by a joint review corrmittee with no action by the full legislature.*0 Moreover, of those forty-one states, twenty-five conducted initial review by special joint comnittees only. An additional four states used standing comnittees during legislative sessions and joint comnittees during the interims. In one state (Hawaii) the Legislative Auditor conducted initial, systematic review; standing comnittees might then give further consideration to questioned rules. Nine states used conventional standing comnittees only and one (Maine) used joint standing comnittees. Thus, 73% of those states practicing legislative oversight and 62% of all states diverged from the congressional model in the location of the review function. This expansion at the state level would not have been predicted by either of the major theories— neither by the "runaway" model, where the assumption is that incentives for such investments are lacking, nor by the "captured" model, because additional oversight mechanisms are presumed unnecessary and counterproductive. Neither would they readily account for the states1 choice of specialized rather than standing 10. Data in this and then next paragraph are extracted from Jones, 1982. [See text, and second table (unnuirtoered).] They differ from Hamm and Robertson (1981, p.137; data as of Decerrber, 1979) whose figures are, respectively: 34 states in all, and 11, 6, 1, 4, 1 in the individual categories. Five states listed by Hamm and Robertson as having no systematic or formal review subsequently passed review laws. (States included Alabama, Hawaii, New Jersey, North Dakota, Virginia and Washington. See Jones, p. 1.) Two other states (Arizona and New Hampshire) which give their legislatures only limited power in this regard are counted differently by the two sources. Other differences may be the result of additional statutory changes between the dates of the two surveys or artifacts of different aggregation rules. Of particular interest here, however, is the agreement regarding Michigan's place on the scale— the most extreme rule in both reports. 10 conrnittee oversight. Incorporating state data would substantially increase the range of variation coiipared to that observable in the Congressional setting. This factor alone could lead to significantly different interpretations of the dynamics of the oversight environment. A third model— "constrained" bureaucracy, state actions in adopting these new oversight measures seem based on a model of legislative-bureaucratic relationships which differs significantly from either of the models presented above. Bureaucrats are seen as less responsive to legislative wishes than desired , as acting on the basis of something other than legislator goals and benefits. This agrees with the "runaway" model and contradicts the "captured" model. On the other hand, state legislative action appears to support the view that greater control over bureaucrats is both desireable— now conflicting with the "captured" model— and feasible— disagreeing with the "runaway" model. Further, it is possible to devise oversight mechanisms which will justify and attract personal and institutional investments sufficient to accomplish the goal of greater control. These observations on state-level practice lead me to propose a third model of legislative-bureaucratic relationships— the "constrained" bureaucracy model. Table 1.1 summarizes the key assunptions of this and the other two models. 11 Table 1.1. Effectiveness and Feasibility of Legislative Oversight of Bureaucratic Activity: Key Assumptions of Three Models— "Runaway", "Captured", and "Constrained". Model of Legislative Oversight bv Characterization of Bureaucracv Runaway Captured Constrained Key Assumptions Reaardino. . . Bureaucratic goals More overt/direct legislative oversight Legisl. incentives/ structure for more oversight own legislators1 own ineffectual counter­ productive desireable lacking lacking feasible Clearly, no two of these models can be correct, whether any one is successful in uniquely explaining, let alone predicting behavior remains a matter to be submitted to analysis. 1.3 The Michigan Case The research reported here uses Michigan's experience in one area of legislative oversight— oversight of administrative rules over a thirteen year period beginning in 1972— to evaluate the three models previously discussed. Three factors make the Michigan experience especially suitable for such analysis. First, Michigan has the longest of the state histories of legislative supervision of administrative rulemaking. Explicit authority for legislative review dates from 1947. Systematic review of 12 all rules began in 1972; approximately 150 rules transmittals annually have been reviewed since then. This provides a substantial data base with which to work, not the case in all states. Second, since 1947, Michigan has employed a review structure different from that chosen by Congress. Congressional review of rules has been located within the standing comnittees, both authorization and appropriations. In Michigan, review m s assigned from the beginning at least part of the time to a joint committee— the joint Conmittee on Administrative Rules (JCAR) Since 1964, the JCAR has been given the full review responsibility; authorization and appropriations comnittees have been removed completely from the normal review process. As demonstrated, this location of the review process is similar to that in most of the states with legislative review, and formally divorces review from the structure and sanctions of the oversight mechanisms operative in the Congressional environment. The resulting concentration and specialization may create incentive structures and dynamics different from those in the Congressional setting from which most current theory was developed. Finally, over its nearly forty-year history of legislative review of administrative rules, Michigan has changed the process and its decision rules several times. In each case the result was increased 11. The conmittee was variously known as the Interim Conmittee, Legislative Conmittee, Statutory Conmittee, or Joint Conmittee on Administrative Rules in the early years, but since 1958 has most frequently been and is now called the "joint conmittee". To avoid confusion, the conmittee will always be identified in this paper as the Joint Conmittee— the JCAR. 13 stringency of oversight. A recently developing literature deals with the relationship between institutional factors and oversight incentives. 12 Michigan's internal changes facilitate the relatively controlled investigation of such relationships. The history of legislative review of administrative rules in Michigan can appropriately be broken into three major periods. The first, extending from 1947-1971 and preceding systematic review, will be termed the complaint period. It is reviewed briefly here simply to give some sense of the developments which led to current practice. The second major phase, from 1972-1977, begins with the first year of systematic review of all newly proposed administrative rules. It includes subsequent years during which rules took effect unless the legislature confirmed the negative action of the review conmittee. will be termed the negative veto period. decision rule was reversed. It Beginning in 1978, the New rules could not take effect without winning the prior affirmative vote of the reviewing conmittee. from 1979 on will be included in the reverse veto period. Years This dissertation focuses on the latter two periods. 12. See, for instance, William Lyons and Larry W. Thomas, "Oversight in State Legislatures: Structural-Attitudinal Interactions."American Politics Quarterly 10:1 (January, 1982), pp.117-133? Gary J. Miller, "Bureaucratic Compliance as a Game on the Unit Square," Public Choice 29 (1977) :37-51; Gary J. Miller and Terry M. Moe, "Bureaucrats, Legislators, and the Size of Government," American Political science Review 77 (JUne, 1983):297-322; and Alan Rosenthal, "Legislative Behavior and Legislative Oversight," Legislative Studies Quarterly VI:1 (February, 1981), pp.115-131. 14 Conplalnt period; 1347-1971. During this period, administrative agencies promulgated rules13 without any prior legislative review. To the degree that rules received legislative consideration, it was after they had taken effect. Legislative action was usually based on specific conplalnt; there was no systematic conmittee review. The Michigan legislature established the state's first administrative code through a 1943 act which called for the first time for all state administrative rules and regulations then in effect to be TA compiled and published. Two years later, in 1945, it attempted to create for itself a powerful role in the promulgation of changes or additions to the code— it passed a bill requiring prior legislative approval of most administrative rules and regulations.13 Governor Harry F. Kelly successfully vetoed the measure. In 1947, the legislature was more successful, if less radical. It passed legislation requiring all newly pronulgated rules to be transmitted to the legislature, where they would be referred to the appropriate standing comnittees.16 (Although not consulted in advance of the promulgation of rules, they would, presumably, at least be 13. Michigan has typically used the term "rules" for what are frequently referred to by other states and the federal government as "regulations" or "rules and regulations.” The term "rules” will be given precedence in this discussion, in keeping with Michigan general usage. 14. "Administrative Code Act", Michigan 1943 FA 88. 15. Michigan 1945 Senate Enrolled Act No. 69 (S.B. #123). 16. Michigan 1947 PA 35. 15 informed.) The legislature could, by concurrent resolution and without gubernatorial presentment, overturn any such rule. 17 Moreover, during the interims between sessions of the state's part-time legislature, rules were to be transmitted to a joint conmittee on administrative rules. This conmittee was empowered to meet between sessions and suspend rules until the legislature next met. This established a mechanism to control at least one administrative strategy— promulgating rules when the legislature was not in session— which might otherwise allow circumvention of oversight. Michigan now had a legislative committee whose sole responsibility was the review of administrative rules and the evaluation of their conformity to authorizing statutes. Two significant statutory increases in the committee's powers occurred during this early period. substantive veto authority. 18 In 1951, the JCAR was granted Rules suspended by the conmittee remained suspended unless reinstated by the JCAR or approved by concurrent resolution of the legislature. Conmittee votes during the interim to suspend a rule no longer required confirming legislative action to remain in effect. The legislature further strengthened the conmittee in 1964, authorizing it to meet year round and receive all newly pronulgated administrative rules. 19 This meant that all remaining review responsibility had been transferred out of standing comnittees and vested in the Joint Conmittee on Administrative Rules. This 17. Prior to 1947, binding legislative action against administrative rules would have required the regular legislative process, including presentment to the governor. 18. Michigan 1951 PA 9, at 24.78e. 19. Michigan 1964 PA 161. 16 increasing centralization and enhancement of the review function during this period had several notable aspects. First, the increased empowerment of the joint conmittee represented a substantial and almost unprecedented legislative deference to committee judgement. Prior to 1951, the JCAR's power was formally that of most other comnittees, i.e., advisory. The 1951 amendment, however, gave the conmittee substantive authority which carried over even after the legislature was back in session. This let the legislature exercise a substantial review function while minimizing the amount of detail with which the total body had to deal. Second, the eventual centralization of review in a single coimuttee signaled and furthered the development of a new kind of expertise in the legislature— an expertise in the administrative rules, their promulgation and application. Standing comnittees would concentrate on legislation; the joint conmittee, on the administrative rules necessary to fulfill the intent of the law. The cortmttee's expertise cut across previous divisions of authority within the legislature, including virtually all the traditional subject matter jurisdictions, but concentrated on an area which otherwise received very little legislative attention. The new division of labor recognized that the key interest of most legislators was legislation, but also manifested the legislature's increasing seriousness about exercising the right of oversight. It allowed a small group of legislators to begin to see administrative rules as something more than isolated phenomena and to react to them with a broader understanding of their place in a system. As early as 1959, the JCAR was calling for the development of a uniform system for 17 "promulgation of rules whether new or amendatory" and a "uniform 20 numbering system to be used by all agencies."'6 By 1969, it was advocating a general reworking of the state's administrative procedures act, including: 1} clear definitions of the terms "rule", "guideline", and "office practice"; 2) spelled out procedures for the making of rules; and, 3) designing future legislation so as to simplify the practice of administrative law, 21 They were at least partially successful in all three areas but approval of a rule generally still required only certification by the state's Legislative Service Bureau and the attorney general. Unless a complaint was raised, a rule was likely to receive very little legislative attention. The conmittee had little work to do. The committee chair reviewed recently promulgated rules and decided which the conmittee as a whole should review. hearing. There was an occasional JCAR members did not have to be particularly diligent, experienced, or well-informed. "Negative Veto" period: 1972-1977. In a single 1971 legislative act, Michigan's legislative review went from casual to systematic review, greatly increasing the responsibility and role of the Joint Committee on Administrative Rules. oo 20. "Report of the Joint Committee on Administrative Rules— 1958", State of Michigan, Journal of the Senate. April 15, 1959, pp.452,453. 21. "Report of the Statutory Committee on Administrative Rules for the Year 1968", State of Michigan, Journal of the Senate. March 13, 1969, p.393. The report also notes a survey showing that of 37 public acts acconpanied by a mandate to establish administrative rules, compliance was achieved in only 12 cases, "representing a percentage of 32%". 22. Michigan 1971 PA 171. 18 As beforer rules would be certified for form, legality and proper nunbering. Beginning in 1972, however, they would then would face a sixty day period during which the JCAR was required specifically to review them and prior to which they could not take effect. This act increased the costs for both the agencies and the legislature. Agencies now had to be prepared to satisfy the legislative conmittee as well as the attorney general and the legislative service bureau. On the other side, the legislative conmittee had to adopt procedures which tracked receipt of and action on proposed rules transmitted to it. (The resultant legislative record-keeping provided the basis for the quantitative analyses which follow in later chapters of this work.) The 1971 act provided for several possible outcomes under the new rules oversight provisions. If the conmittee did nothing, the rule took effect automatically at the end of the sixty-day waiting period. If the conmittee voted to approve a rule, it could then take effect once the governors office had had it for ten days. The conmittee might also vote to disapprove a rule, but that was not itself a binding action; for disapproval to take effect, the legislature must within 30 days pass a concurrent resolution of disapproval. (In this respect, the amendment, on the face, decreased the committee's power.) The amendment also authorized agencies to withdraw a rule prior to expiration of the sixty day review period. The bottom line, however, is that rules were routinely reviewed but took effect unless the legislature took negative action within the prescribed review periods. For this reason, outcomes under this amendment are characterized as occurring under the "negative veto". Although the above actions put Michigan far ahead of most other 19 states in legislative oversight of the rules promulgation process, legislators would find even this unsatisfactory. suggest two major areas of concern. Conmittee reports The first was the problem of agencies submitting rules "toward the close of session at a time when the Conmittee would not have an adequate opportunity to review the rules," 23 That, plus "the frustration of trying to get favorable consideration to [concurrent] resolutions disapproving rules caused most Ai Committee members to support [change]." Change there would be, substantial change. "Reverse Veto" phase: 1978-present. In 1977, the legislature passed an amendment requiring prior legislative approval of all administrative rules. Gov. William G. Milliken vetoed the measure, but contrary to the experience with Governor Kelly thirty-two years earlier, this time the legislature prevailed. 25 By a vote of 30-6 (of 38 serving) in the Senate and 74-5 (of 110 serving) in the House, the governor's veto was overridden. The vote represented the first time in twenty-six years that a gubernatorial veto in the state of Michigan had been overturned, and the only time it would happen in the fourteen years of the Milliken administration. The legislature was, indeed, serious about its role in oversight of the administrative rules process— no future administrative rule nor change to any existing rule would take place without the explicit approval of the legislature of the state of Michigan. This 23. "Combined Annual Report for 1976, 1977, 1978 of the Joint Committee on Administrative Rules," Joint Committee on Administrative Rules, State of Michigan, p.6. 24. Ibid. 25. Michigan 1977 PA 108. 20 represented the most stringent legislative oversight decision rule in the country. rule. The process had been reversed from that under the negative Outcomes in this period are described as occurring under the "reverse veto". Two other changes in the 1977 amendment were particularly inportant. First, Senate menbership on the conmittee was increased to five (from a previous three), matching the nunber from the House since 1970. (The implications of this investment of resources will be discussed in Chapter 3.) The interaction of the new veto rule, the added Senate menbership, and the concurring majorities rule, now meant that for new rules proposals to take effect, they must secure the affirmative vote of at least three Senators and three Representatives. negative votes from either chanber could block approval, 2S Three second, the comnittee could vote to extend the review period from the minimum 60 days to 90 days, thus potentially expanding the scrutiny given a particularly difficult or controversial rule. Thus, in 1983 when I began observations of the review process, it followed the pattern shown in Figure 1.1. There were four formalized points for public involvement or contact: initiation of a rule, at agency hearings, after agency modifications, and at the time of JCAR consideration. Rules proposals were required to be cleared with the Legislative Service Bureau and Attorney General's office prior to transmittal to the JCAR. Once approved by the JCAR, they could not take effect until ten days after the agency submitted them to the governor's 26. The requirement of concurring majorities dated from 1969. See Michigan 1969 PA 306, at 24.236. 21 office. Joint committee and full legislative action followed the lines indicated by the lower left hand portion of the diagram. Hie reader will note that there was no requirement that the conmittee submit disapproved rules to the legislature for further action. Only in the case the conmittee took no action (insufficient votes for either approval or disapproval) was legislative referral mandated. That system continues today. Michigan— Period Summary. Following Figure 1.1, the diagram of the current rules promulgation process, is Table 1.2. That table summarizes significant characteristics of the review process during each of the three periods. The first period, the complaint period, is offered for historical perspective only; it will be discussed only in passing in the remaining chapters. Details of the other two periods, in contrast, will be discussed at length in the chapters which follow, with a heavy focus on the differing effects of the negative and reverse vetoes. 22 M V 1 < .5 =3 s. 0“ g .=M lit a? 1 a/.a J < S 3 u ntlpfl prm iilg a fim , By statute, review operated through the Joint Comnittee on 71 Administrative Rules, with the possiblity of action by the full legislature under certain circumstances. Table 3.2 summarizes the joint corrmittee's disposition of all proposed administrative rules transmitted to it from 1972 through 1982. Table 3.2. JCAR Disposition of Rules Transmittals: Approved Withdrawn by agency . No action (1978-1982) Impasse Disapproved Total 1972-1982. N 1,305 272 3 10 27 1,617 % 80.7 16.8 0.2 0.6 1.7 100.0 d e l u d e s 76 "no action" cases (4.7% of total) from 1972-1977 which, under the negative veto rule, automatically took effect at the end of 60 days. Subsequently submitted to the legislature under concurrent resolu­ tion of approval. Source: Compiled from the "Daily status Report", unpublished record of the Joint Committee on Administrative Rules, State Legislature, State of Michigan, December 8, 1971 through December 31, 1982. The most common outcome of review by Michigan's Joint Committee on Administrative Rules is approval of the proposed rule. Of the 1,617 rules considered by the JCAR in the years 1972 through 1982, a total of 1,305 were approved at the committee level— 80.7% of all cases. Thus, four-fifths of all rules transmittals were approved, most with no recorded formal legislative intervention.^ On the other hand, 19.3% of the cases between 1972 and 1982 do 4. Some of the approvals are rules which had been previously withdrawn and were approved upon resubmittal. A later measure will take this into account, calculating an adjusted approval rate. 72 offer direct evidence of some level of legislative intervention. In this group are cases either 1) withdrawn by the administrative agency prior to comnittee disposition, or 2) voted upon but not winning conmittee approval. The more likely occurrence is withdrawal of a rule, a protective action taken by agencies after an unfavorable reception at a rule's initial JCAR hearing. The reader may have noticed that Table 3.2 carries nO category labeled "amended". The comnittee, from its beginnings in 1947, has been prohibited from amending administrative rules proposals. Should the comnittee disagree with certain portions of a rule, it nust make a decision— whether to accept the rule as a whole with imperfections, or deny the rule as a whole regardless of its residual merits. (In this respect it faces the dileima common to chief executives in this country who have general but not item veto powers.) This is not to say that the comnittee has no informal amendatory power. The threat of disapproval, if substantive, could provide incentive for an agency to withdraw a rule and resubmit it in a more acceptable form. Agencies withdrew the proposed rule in 16.8% of the cases (87.2% of transmittals not approved) 5 suggesting that the informal power may be considerable. Cases voted upon but not winning approval fall into three categories: outright disapproval under either veto rule, and "no action" (lack of sufficient votes for action, rather than no vote) and "inpasse" 5. In addition, comnittee pressure sometimes leads agencies to publicly clarify proposed interpretation and application prior to approval of a rule; the record fails to provide any information on intervention at this level. Eight months of observation of the current comnittee, however, leaves me confident that even with this type of intervention counted, the substantial majority of cases would correctly be described as involving no overt legislative intervention. 73 (lack of concurring majorities) under reverse veto. Outright disapproval of a rules transmittal, the outcome most likely to excite ccnment, has occurred only 27 times in the entire eleven years, for 1.7% of all cases. Instances of nno action" under the current reverse veto are quite rare— 0.2% of all cases.6 "Inpasse", which occurs as a separate outcome only under the current veto rule, accounts for only 0.6% of the cases. These three categories contained conprise just 2.5% of all cases (and 12.8% of transmittals not approved). While approval remins the most likely outcome of legislative oversight in Michigan, nonetheless, the 19.3% of the cases not approved probably have disproportionate importance. These cases usually demonstrate areas of relatively high public policy conflict within the dynamics of Michigan state politics. The handling of these rules will be of particular interest as we move to discussion of legislator, agency, and interest group interactions, especially.) (See Chapters 4 and 5, For the moment, however, discussion will focus on two other matters: 1) specifics of comnittee power as they affect the processing of rules transmittals and 2) the shift in the power balance between legislative and executive branches which occurred as a result of a change in decision rules affecting the comnittee. From the beginning in 1972 of systematic comnittee review of proposed administrative rules, the comnittee has had full approval power. There has been no statutory provision for routine intervention by the full legislature in the case that the joint committee approved a 6. "No action" under the earlier negative veto rule constituted automatic approval; those 76 cases were counted as approvals. See Table 3.2, note "a". i 74 rule or it was withdrawn by the sponsoring agency. With 80.7% of all cases approved by the comnittee, and another 16.8% withdrawn by the 7 agencies, only 2.5% of all transmittals had any potential for action O by the full legislature. Even then the type and extent of involvement varied in accord with the veto rule operative at the time. Under the 1972-1977 negative veto provisions, a comnittee disapproval had the force of a recommendation to the legislature; rules took effect unless the legislature subsequently passed a concurrent resolution of disapproval within thirty days. That provision tipped the balance between the branches in the direction of the executive. If the g legislature failed to act, the agency position prevailed. The JCAR considered 784 cases in the six years under the negative veto provisions; it voted disapproval only 11 times (1.3% of those cases). The legislature upheld the conmittee's vote in nine of the eleven cases. In the other two, the legislature failed to pass the required 7. Only one agency request to withdraw a rule has been denied by the comnittee, an occurrence regarded as something of a fluke by most observers and even menbers of the comnittee. (This occurred at the Deceirber 6, 1983, meeting. It was apparently in deference to the wishes of a Senate menber who wanted to be sure the agency got the message he did not wish to see another rule with similar content.) 8. Under Michigan 1947 PA 35, M.S.A. 3.560(7b), the legislature reserved "the right to approve, alter, suspend, or abrogate any rule pronulgated pursuant to the provisions of [that] act." In 1958, under PA 177, M.S.A. 3.560(12a), it further provided that any menber of the legislature could introduce a joint resolution or bill to "express the will of the legislature that (a specific) rule should be revoked or altered." If the agency failed to act in accord with such a sentiment, the legislature could abrogate the rule by legislation, with 1969 PA 306, M.S.A. 3.561.150, it was further stipulated that the legislature could, through the bill process, amend a rule. Action under any of these provisions is almost nonexistent. 9. As noted before, this was also true if the legislature's agent, the comnittee, failed to act. 75 resolution of disapproval and the rules took effect, as did the 76 transmittals on which the comnittee failed to act. Provisions adopted in 1978 with the reverse veto substantially shifted the balance between the branches in regard to the promulgation of administrativew rules. A comnittee vote of either approval or disapproval was fully effective without further legislative action. The comnittee considered 833 cases between 1978 and 1982 under this rule. 'Hie comnittee*s sixteen disapprovals during that time were themselves fully binding on the departments, "No action" cases (insufficient votes for approval os. disapproval), on the other hand, required the comnittee to introduce a concurrent resolution of approval.10 There were just three "no action" cases among the 833 handled between 1978 and 1982; the legislature passed one of these sets of rules and the other two died in standing comnittees.11 Before, the agency "won" if the legislature failed to act; under the new rule, the agency "lost" in that circumstance. The conmittee’s position was strengthened in the process and the balance of power shifted toward the legislative branch. In sumnary, a total of 1,617 cases were decided by the JCAR from 1972 through 1982. Imnediate approval was denied by the comnittee to 19.3% of all transmittals. Of those cases where the legislature had opportunity to act, it failed to sustain the decision of the comnittee in only 3 cases. Thus, initial quantitative analysis provides strong 10. The JCAR tries to avoid this outcome. TO do otherwise would give up comnittee authority and also require their colleagues to take on a burden of detail which most are not interested in. 11. An "impasse" (opposing majorities) had the same substantive effect as a vote of disapproval during this time; ten proposed rules failed to take effect because of such a comnittee outcome. 76 evidence of the degree to which the review function and authority are focused on and within the comnittee. The comnittee wields both formal and substantive power, power far beyond that of most legislative committees, in addition, changes in veto rule strengthened the relative importance of the legislative branch in the rule promulgation process, further strengthening the role of the comnittee itself. Interview data additionally support these conclusions. Members of the comnittee and staff, and agency administrators and interest group representatives closely associated with the rules promulgation process routinely state that other legislators, legislative comnittees, and the legislature as a total body are sinply not involved in the process. Exceptions are extremely rare. Even in the case of rules mandated by new legislation, the subject matter standing comnittees virtually never comiunicate with agency personnel or JCAR members or staff as the implementing rules are developed. 12 A current member of the JCAR offered one explanation of this behavior: "Partly it's that the standing committee people don't have time for this, but it's more than that. They think when the legislation is passed the battle.is over; business knows better. That is just one of the battles." This is not to say that one need study only the committee to understand the outcomes of the review process. Other chapters will demonstrate the role of a variety of actors and factors. The argument 12. ttiis is not because they are not informed of which rules are pending. By statute, the JCAR routinely informs all standing comnittees and all legislators of the rules to be discussed at each JCAR meeting. 13. Representative Virgil Smith, interview, Lansing, Michigan, June 20, 1984. i 77 here is merely that the comnittee, as opposed to the full legislature, is the operative unit in Michigan's legislative oversight of administrative rules and is at the nexus of legislative-executiveinterest group interactions relative to the promulgation of administrative rules. 3.2 Decision Rules and Outcomes The previous section has already discussed one way in which decision rules are inportant elements in the oversight environment, in this section, I will consider another way in which they are inportant, specifically relating three decision rule changes to the rate of rule approval. Although divided into two subsections, both are addressed to a single hypothesis: Hypothesis 3 s More stringent legislative oversight mechanisms will be associated with lower rates of rules approval. 3.2.1 Veto Rules and Majority Requirements Several references have been made to the differing effects of certain kinds of committee action (e.g "no action"), depending on the veto mechanism operative at the time. The next obvious question is whether those differences have any systematic effect on outcomes and inpacts of the review process. More specifically, does requiring the 78 legislature to give prior approval of proposed changes in administrative rules result in lower rates of approval? Table 3.3. JCAR Disposition of Cases— Approved or Other— by Type Veto, 1972-1982. Approved Other Column total % Negative veto (1972-1977) N % 729 (87.5%) 104, (12.5%) 833 (51.5%) Reverse veto (1978-1982) N % 576 (73.5%) 20fi (26.5%) 784 (48.5%) Row Total % 1305 (80.7%) _2I2 (19.3%) 1617 (100.0%) Corrected X = 50.27 p - .0001 YUle's Q = .434 Hie change from the negative to the reverse veto decision rule is clearly associated with change in transmittal outcomes. Table 3.3 2 yields a X significant at the .0001 level, and a Yule's Q of .434, indicators of a substantial relationship between the type of veto' mechanism and rule approval. The likelihood of rules not taking effect increases significantly when agencies are required to win prior legislative approval; it more than doubles under the reverse veto mechanism. At least two decision-rule factors help explain the observed variation: 1) the requirement of concurring majorities, operative under both veto mechanisms, and 2) the resulting differing effects of either a negative vote, a "pass" (abstention), or an absence under the two decision rules. During much of the history reported in Chapter 1, committee decisions were made by simple majority vote. Beginning in 1969, however, the statutory requirement of "concurring majorities" was added to the JCAH's decision making procedures. With the eight-member 79 comnittee (1972-1977), this meant that comnittee action would require agreement of at least two Senate and three House menbers; with the ten-menber comnittee (1978-1982), it required three Senators and three House menbers. This requirement interacted with the change in the veto rule in such a way that the effect of a negative vote was strengthened over time. Dnder the negative veto procedures (1972-1977), proposed administrative rules took effect unless the comnittee voted disapproval (and was supported by the legislature's adoption of a concurrent resolution of disapproval). During that period then, at least five properly distributed negative votes (of eight) were required to block a rule's taking effect (and even then, might be overturned by the legislature). Five House votes were insufficient to block a vote without the minimum two from the Senate for a concurring majority, Dnder this rule, the substantive effect of a "pass" or absence was that of a vote for approval, since the rule would take effect in the absence of sufficient votes against it. The reverse veto mechanism (1978-present) greatly changed these dynamics. Under the reverse veto rules take effect only with the support of concurring majorities. Now with five menbers from each chamber, a minimum of six votes, three from each chamber, is necessary for approval. If three meirbers from the Senate or three from the House vote against a motion to approve, a proposed rule will be blocked from taking effect, even if all other seven members favor the motion. This failure to take effect, however, would not constitute "disapproval" under current practice. Formal "disapproval" requires concurring majorities voting for a motion to disapprove. Should the motion to 80 disapprove also result in opposing majorities, the result would formally be known as an "inpasse". Although the name differs, the substantive effect is the same; having failed to get the necessary concurring majority, the rule does not take effect. Under this decision rule, then, a "pass" vote or an absence has the effect of a vote for disapproval, since it fails, to contribute to the votes necessary for a rule to take effect. Under the negative veto rule (1972-1977), then, a minimum of five votes of eight (62.5%) was required to block a rule; under the reverse veto (1978-present) as few as three of ten (30%) can achieve that result. The effect is to strengthen the individual vote which would deny the agency authorization to promulgate the rule under consideration. An expansion of Table 3.3 suggests that the agencies were fully aware of and took action to meet this greater possibility of a negative outcome. (Agency strategies will be discussed in Chapter 4). Table 3.4 shows that decreased approvals is largely the result of an increase in withdrawals. Table 3.4. JCAR Disposition of Cases— Approved, Withdrawn, and Other— by xype Veto, 1972-1982. Approved Withdrawn Other Column total % X2 = 51.79 p = .0001 Yule's Q = .428 Negative Veto (1972-1977) N % 729 (87.5) 93 (11.2) _J2. ( 1.3) 833 (51.5) Reverse Veto (1978-1982) N % 576 (73.5) 179 (22.8) _22 ( 3.7) 784 (48.5) Row N % 1305 (80.7) 272 (16.8) 40 I 2.51 1617 (100.0) 81 Outcomes within the time periods offer additional support for the hypothesis that approval rates will decline under more stringent veto mechanisms. Not only do the rates differ significantly between time periods, as demonstrated in Tables 3.3 and 3.4, but there is no overlap between time periods. Table 3.5 shows that approval rates under the less stringent negative veto rule, ranging from 79.9% to 93.3%, are never as low as they are under the reverse veto, where they vary from 71.4% to 75.9%.14 Table 3.5. JCAR Disposition of Transmittals (and Incidence of Subsequent Approvals by full Legislature) by Veto Rule and Biennium, 1972-1982. Negative Veto Rule Total transmittals Approved Withdrawn Disapproved (# leg. approvals) Reverse Veto Rule Total transmittals Approved Withdraw! No action Impasse Disapproved (# leg. approvals) 1972 N % 90 84 (93.3) 3 ( 3.3) 3 { 3.3) (0) 1973-74 N % 307 276 (89.9) 28 ( 9.1) 3 ( 1.0) (2) 1978*3 N % 162 118 (72.8) 40 (24.7) 1 ( 0.6) 1 ( 0.6) 2 ( 1.2) (0) 1975-76 N % 307 266 (86.6) 41 (13.4) 0 ( 0.0) (0) 1979-80 N % 315 225 (71.4) 73 (23.1) 1 ( 0.3) 7 ( 2.2) 9 ( 2.9) (0) 197713 N % 129 103 (79.9) 21 (16.3) 5 ( 3.9) (0) 1981-82 N % 307 233 (75.9) 66 (21.5) 1 ( 0.3) 2 ( 0.6) 5 ( 1.8) (0) fsome column totals vary from-100.0% by 0.1 due to rounding. T h e years 1977-78 were also a Michigan legislative biennium but are separated here because of the change in the decision rule. 14. Iftis also holds true if the percent approval is calculated by year, rather than biennium. Annual rates of approval range from 79.8% to 93.3% under negative veto; from 70.9% to 75.9 under reverse veto. 82 3.2.2 Duration of Coirmittee Review Period One of the arguments against legislative review of administrative rules is that it creates additional and unnecessary delay in getting regulations in place. Otiose favoring such oversight counter that any delay is a reflection of the need for review and brief in any case relative to the total time involved^in promulgating rules. This suggests two further avenues of investigation: 1) the degree to which Michigan's review process delays the promulgation of rules; and 2) whether delay periods vary by veto mechanism. The time elapsed between receipt of a rule and action by the joint comnittee varies considerably. Observed intervals range from no days at all to as many as ninety-two,*5 The mean interval for the entire eleven year period for time elapsed between receipt of a rule and JCAR action on the rule is 38.4 days. There is a priori reason to expect the mean interval between receipt of a rule and JCAR action will vary by veto mechanism. The act which originated systematic review by a joint committee of the legislature allowed up to two months for action. The time limit was increased, however, by the 1977 amendment which also created the reverse 15. Rules voted upon on the same day they were received were generally substitutes for a rule previously heard at least once by the conmittee. With modified language already worked out, "clearance" secured from the various interest groups, and the required certifications by the Attorney General and the Legislative Service Bureau in hand, the agency would appear at a committee meeting requesting permission to withdraw proposed rule "A", and submit proposed rule "B" for iirmediate action. If there were no complaints from any source, including members of the conmittee, the rule would likely be passed immediately. 83 veto. Beginning in 1978, the committee was authorized to vote a 30-day extension of the review period for any specific rule pending before it. Use of that authority would tend to increase the mean elapsed time. Table 3.6. Mean Time Elapsed between Receipt of Proposed Rule and JCAR Action, All Years, and by Veto Period. Time Period All years (1972-1982) Negative Veto (1972-1977) Reverse Veto (1978-1982) Mean Time Elapsed (in days) Std Dev N 38.4 33.4 43.8 21.4 17.9 23.4 1617 833 784 Difference of means test statistics (negative and reverse veto periods): t = 9.96 p = .0001 Table 3.6 shows a ten day difference between the mean intervals, a difference significant at the .0001 level, is elapsed time before the committee a predictor of a rule's eventual outcome? Using a three-step coding for "outcome''— approval, withdrawal, and other— and the actual number of days elapsed between receipt and action by the JCAR, the correlation coefficient over the entire time period is .2493 (p - .001). Within time periods, it is .1337 under negative veto and .2626 under reverse veto (both significant at the .001 level). The strength of the relationship is nearly doubled under the reverse veto, with time elapsed between receipt and action more strongly related to outcome under the reverse veto, but is relatively weak in both cases. This, however, is an overly stringent measure. Proposed rules are logged in by committee staff on a workday basis but the conraittee meets, at most, weekly, and sometimes only once a month, xable 3.7 gives 84 additional perspective on the relationship, using data grouped on a monthly basis. Table 3.7. JCAR Disposition of Rules Transmittals (in Per Cent) by Veto Period and Days Held by Committee. Outcome Necative Veto (1972-19771a Row N % 0-31 32-62 Approved Withdrawn Other Column N % x2 = p = Yule's Q = 92.1 7.9 0.0 83.2 14.2 2.6 (404) (429) 48.5 51.5 (729) 87.5 ( 93) 11.2 ( 11) 1.3 (833) 100.0 Reverse Veto (1978-1982) ROW N % 0-31 32-62 63-92 81.4 16.4 2.2 75.5 21.4 3.2 55.1 37.2 7.7 (269) (359) (156) 34.3 45.8 19.9 (576) 73.5 (179) 22.8 ( 29) 3.7 (784) 19.619 .0001 .407 100.0 37.643 .0001 .348 ^ r o m 1972-1977, the statute limited the review period to two months. The negative veto period grouped data show a statistically significant difference of 8.1% between rates of approval for those rules held for up to one month and those which took more than one month. Under the reverse veto, the difference between cases decided within their first month and the second is only 6.1%, but there is a dramatic decrease in approval of cases held into the third month. Approvals drop 20.4% between the second and third months; only 55.1% of cases going into the third month win approval. Although the cases categorized as "other"— those disapproved under either veto rule and no action or inpasse outcomes under reverse veto— are few, the change is in the expected direction; the percentage of these cases increases unidirectionally with time before the corrmittee. The third month of review does not cause rules not to be approved. 85 Instead, it reflects a certain level of controversy already having surfaced; "easier" rules, on the average, are taken care of more quickly. The relationship is probably interactive. On the one hand, those rules which are voted the extension of the review period are most frequently those about which there already has been conplaint, whether by JCAR meirbers or outside interested parties.*^ is usually a signal of a rule in difficulty, The decision to extend in addition, the longer a proposed rule is pending, the more opportunity people have to find something wrong with it and to mobilize forces for opposing it. Extension increases a rule's vulnerablity to opposition. 3.2.3 Decision Rules and Outcomes Summary Three measures of oversight stringency were reported here: 1) veto rule— negative or reverse (the latter requiring prior legislative approval); 2) majority requirements— sinple versus concurring; and 3) length of allowed review period. supported. In each case, the hypothesis was More stringent decisions rules uniformly resulted in lower rates of rule approval. 16. Some cases are granted an extension simply because the committee calendar has gotten overloaded, if not for these cases, the rate of approval in the third month would be even lower. 86 3.3 investments in the Review Function One issue raised in the literature is the relationship between institutional and individual investment in legislative oversight. This section looks at several measures in both categories, focusing at the individual level on those legislators who have formal responsibility for Michigan's rule oversight function— the menbers of the Joint Committee on Administrative Rules. 3.3.1 Institutional Investments The assignment of members and other legislative resources to any ooninittee necessarily represents an institutional investment, incurring both direct and opportunity costs. Level of staff support, committee workload, and patterns of committee assignments all serve as indicators of the institutional value of a given committee including that of a committee whose main responsibility is oversight. Staff support. When first constituted in 1947, the Joint Committee on Administrative Rules had three meirbers from each house, lacked its own staff, and met only a few times during the legislative interims. This generally low level of investment continued for a number of years. The only staff support the committee had was that provided through regular staff of legislators serving on the comnittee. The 1967 conmittee report contains the first record of separate 87 staff for the Joint Comnittee on Administrative Buies (JCAR), and the first record of formally "loaned" staff. Explicit institutional investment in the joint comnittee at that time included the comnittee1s own counsel and secretary, and regular assistance from one Legislative Service Bureau staff menber. 17 In 1972, the legislature again increased JCAR staff. With no ]0 reduction in Legislative Service Bureau staff, the joint comnittee was authorized to directly enploy a legal counsel, administrative assistant, comnittee clerk, and part-time secretary. This change was a direct result of the increased responsibilities placed upon the comnittee by the 1971 amendments to the APA. Doubling the JCAR staff provided the personnel to ensure the amendments had not been purely symbolic activity. The most recent adjustment in JCAR staff occurred in 1977. Since that time, the comnittee has enployed two attorneys (special and assistant special counsel, or two co-counsels) and an administrative assistant, an overall investment level roughly equivalent to the 17. The Legislative Service Bureau (LSB) was created in 1965. Although its major responsibilities revolved around the drafting and printing of bills it soon become involved in the rules process as well. Agencies would frequently send proposed rules language to the LSB for informal review prior to submitting it to public hearing. The bureau would review the form, classification, arrangement, and numbering of the rule. Once hearings were over, any necessary changes would again be made, and the rule resubmitted to the LSB, this time for formal approval. This regular processing by the Legislative Service Bureau is itself an indicator of increased legislative investment in rules oversight. See "Report on the Administrative Process in Michigan State Government," State of Michigan, Legislative Service Bureau, Vol. 3, No. 2 (Revised February 1983), p.6. 18. In fact, the Legislative Service Bureau increased in staff during this time. 88 immediately preceding period. Staff support provided for the activities of Michigan fs Joint Comnittee on Administrative Buies has thus exhibited each of the levels of institutional investment discussed by Lyons and Thomas in their 1982 cross-sectional study. In the JCAR's early years, as they report being the case in 1982 in Missouri, there was no formal direct staff support. Eventually the comnittee was given its own aide and secretary, but professional staff was "on loan" (from the Legislative service Bureau), similar to the pattern observed in Tennessee. The changes since 1972, with the JCAR directly employing both professional and clerical staff, represent a level of institutionalization and commitment similar to that they encountered in Florida. 19 If we were to assess the level of legislative oversight of administrative functions in Michigan solely on the basis of mean legislator involvement in oversight activity, this 19. Rich Jones' report for the National Conference of State Legislatures offers a state-by-state accounting of staff comnitments and in some cases, dollar costs as well. [See "Legislative Review of Administrative Rules: An Update," State Legislative Report Legislative Management Series 7:4 (April, 1982), first table (unnumbered).] Of the 14 states for which he had a direct dollar cost, Michigan ranked fourth. It was sixth of forty-one in direct staff commitment. The critical factor in the historical conparisons being reported here is the number and level of staff over which the JCAR has direct control. There may actually be more staff involved, individual legislators continue to use members of their personal staffs in various ways in meeting their responsibilities to the committee. There is also a member of the Senate Republican caucus staff who regularly reviews the rules and provides analyses to Senate Republicans on the committee, in addition, members sometimes temporarily "borrow" other committee staff over whom they have control and assign them JCAR-related duties. A recent example of this is assigning work relating to the JCAR subcommittee created in 1983 to an aide newly hired to the House Elections Comnittee staff, ‘Three members of the JCAR also serve on Elections, including the menber who serves as chair of both. 89 type of increased commitment would probably not be captured, and the level undoubtedly underestimated. 20 Comnittee Workload. Comnittee workload is yet another indicator of institutional investment in oversight. The time legislators spend in fulfilling their duties for any given comnittee necessarily reduces the time they are available to the chairber for other functions. Committees within the Michigan legislature vary greatly in the amount of work they do and the amount of time required of their members. They range from those which literally never meet within some legislative biennia to those which meet almost weekly when the legislature is in session. By this measure also, the Joint Committee on Administrative Rules reflects an increasing and substantial institutional investment, in the five years immediately preceding the implementation of the 1971 amendments, for instance, the committee met an average of 5.4 times per calendar year. Regular review changed that dramatically. The average number of meetings between 1972 and 1977 is 26.6. After the change in the veto rule, it increased again, to an average of 31.0 meetings per year, much higher than is the case for most Michigan legislative committees. Moreover, these are merely the formal committee meetings, the meetings during which they receive public testimony regarding specific rules proposals, debate, and take action on them. There may be numerous additional consultations with interested parties, including 20. This is particularly true because the number of legislators directly participating in the oversight of administrative rules is very small. Norms of legislative specialization in Michigan are such that it is usually only the members of the committee who are involved at any significant level. 90 other comnittee meirbers, beyond this time commitment. Meirbers of the comnittee routinely stated that they spend more of their week on JCAR-related work than has been the case for most other comnittees on which they have served. Seniority and Number of Meirbers. The final1indicators of institutional investment to be looked at are relative seniority and nunber of comnittee meirbers. Table 3.8 reveals that House menbers appointed to the JCAR begin in 1971 with a mean term of service just slightly higher than that of the rest of the House (3.80 and 3.62 terms, respectively). The mean for all other meirbers in the House remains quite stable throughout the entire period, but steadily rises for JCAR menbers; continuously more senior menbers were being appointed by the House to the JCAR. (The only exception is between the 1977-78 and 1979-80 biennia, when it remains unchanged.) By 1982, JCAR House menbers have an average of two terms (four years) more experience than do the rest of their colleagues. Conparing these data with those from two other House comnittees throughout the same period helps interpret the importance of these differences. Throughout these twelve years, even at the beginning, there is never as much as a one term difference between the means for House JCAR menbers and those appointed to the House Appropriations 21. Michigan's House and Senate appropriations comnittees enjoy the kind of power and status corrmon to their counterparts in many other states and Congress. Any bill requiring an appropriation is referred to these comnittees once favorably reported out of the subject matter standing comnittees. In addition, appropriations bills take precedence over other bills— they are placed at the head of the calendar each day and are given "preference in printing over other bills." See the Michigan Legislative Handbook. House Rule 42 (1971-1983), and Senate Rule 23,b (1971-1979) and 3.35 and 3.37 (1980-1983). 91 Comnittee. 21 The House Marine Affairs Comnittee, in contrast, has a period mean of only 2.55 terms, more than a term less than the House period mean, and over three terms less than either Appropriations or House JCAR menbers. This substantial variation between comnittees within the House underscores the extent of the House investment in the joint comnittee. There are at least two reasons to expect smaller Senate than House variation on this measure. First, Senators serve four-year, unstaggered terms; comnittee appointments are made at the beginning of each new Senate and tend strongly to be continued throughout that senate's four-year duration. There may be a few changes, due to death or resignation and special election replacement of members but, in general, Senate appointments remain stable by four year periods. Michigan Senate elections in the reported period resulted in new senates being seated in 1971, 1975, and 1979. The inpact is reflected in Table 3.8, which shows change in Senate JCAR menbers1 mean term of service only every four years. In contrast, five of the six House JCAR means vary from the previous biennium. In addition, a total of only 38 senators is distributed among seventeen comnittees, none of which has less than five menbers, and some of which have had as many as 13 (Appropriations, during part of this period). First term menbers serve on all comnittees; Appropriations is not excepted. This factor also tends to reduce senate variation in committee mean terms of service. 92 Table 3.8. Mean Terms of Service— Members of Joint Comnittee on Adminis­ trative Rules, All Other Legislators and Appropriations Canmittees (by Chamber), and House Marine Affairs Committee. 19711972 1973- 1975- 1974 1976 19771978 19791980 1981- Period 1982 Mean 3.80 3.62 4.80 3.46 5.00 3.59 5.60 3.50 5.60 3.80 5.80 3.80 5.11 3.63 Approp (16-18) Marine (5-11) 4.50 3.09 4.88 2.33 5.25 2.54 6.42 2.86 6.00 1.85 5.92 2.60 5.50 2.55 Senate (n=38)b JCAR (3,5)c Others (35,33) 1.00 2.57 1.00 2.41 2.00 2.04 2.00 1.74 2.20 2.20 2.20 2.11 1.63 2.18 Approp (8-13) 2.50 2.65 1.95 1.72 2.04 1.96 2.41 House (n=110) JCAR (5) Other (105) ^ouse terms are for two years; Senate terms are for concurrent four year periods. Between 1971 and 1976, all Senate appointees to the JCAR had prior legislative experience in the House; this measure includes only their service in the Senate. NUnber of menbers serving on comnittees changes over time. u Comnittee appointments are made at the beginning of each new legisla­ ture; they change little, however, unless it is also the seating of a new Senate. Senate menbers of the JCAR remained unchanged from 1971 through 1974 and 1979 through 1982. ^The number of Senate menbers was changed from 3 to 5 with the 1977 amendments to the Administrative procedures Act. 93 In 1971, one year before the beginning of routine legislative review, the Senate appointed three first tenners to serve on the JCAR; 22 the Senate JCAR mean term of service at that time was only 39% of that for all other senators. Those same senators continued through the next four years, a time of notable increase in JCAR authority and activity. In 1975, at the first real subsequent opportunity, the Senate appointed menbers with a higher mean term of service— double that of the previous four years and equal to the mean for all other senators. 23 Indeed, from 1975 on, the mean term of service of Senate JCAR menbers equals or exceeds both the mean for all other senators and those on senate Appropriations. The lower Senate JCAR period mean is solely the result of the low 1971-1974 mean. Despite chamber limitations, the Senate had increased investment in the joint comnittee, with the results slightly exceeding even Appropriations. An increase in Senate investment is also reflected in the change in the number of members serving. The 1977 amendments to Michigan's APA increased Senate JCAR membership from three to five of its thirty-eight menbers. This change is particularly evidence of the importance of the committee to the legislature as a whole and to the Senate, and the problems created by the multiple committee responsibilities of the small number of senators. Rep. Thomas j. Anderson, committee member from 1969-1982 and chair or vice-chairperson from 1975-1982, reported the 22. Although in their first Senate term, each had previously served in the Bouse. 23. None of the Senate members appointed in 1975 had previously served on the JCAR, so the difference is not simply a retention of the same menbers. 94 expanded Senate menbership was partially to make it easier to get a quorum. 24 under the negative veto decision rule operative at the time, if the comnittee failed to have enough votes to take action within the prescribed time period, the rule took effect by default. It was apparently easier to get 3 of 5 senators than 2 of 3 at any given meeting; expanding the menbership had the effect of strengthening the legislature's position relative to the executive. Summary. On each of the measures discussed above, the Michigan legislature has made a substantial and increasing institutional investment in the oversight of administrative rules. It has provided direct and increasing staff support to the Joint Comnittee on Administrative Rules, increased the responsibilities and workload of the comnittee, and the seniority and number of meirbers serving on the comnittee. 3.3.2 Career Investments Individual legislators make investments in comnittees, just as do legislative institutions. Legislators do not conpletely control the comnittees to which they are appointed, but individual preferences are at least considered as the leadership makes comnittee appointment decisions. How do individual legislators come to serve on the Joint Comnittee on Administrative Rules? Interview data suggest that most meirbers of the 24. Interview, Lansing, Michigan, August 1, 1984. 95 committee have requested the assignment, but not all. The rational actor perspective would imply that legislators prefer certain comnittees over others for the anticipated resultant benefits. These benefits are most usually assumed to be direct and personal, and for the legislator, to be interpreted in terms of contribution to re-election. Rosenthal's work, on the other hand, suggests there exists a particular type of legislator who seeks a less direct benefit and pursues the more generalized goal of inproving the functioning of government. This is not necessarily an alternative explanation of legislator behavior. Legislators interested in inproving government may well see membership within the legislature as an important vantage point from which to accomplish such a goal; as such, they would also have a major interest in re-election. Committee turnover has frequently been used as one indicator of committee status and desirability. That measure will be used here as one index of the career investment made by individual legislators in the Joint Committee on Administrative Rules. There are problems with such a measure, however, because it fails to distinguish between turnover resulting from exit to other committees and turnover resulting from failure to return to the legislature (or that chamber of the legislature). 25 Table 3.9 demonstrates the difference. 25. One might argue that failure to return to the legislature is evidence of poor choices in committee assignments; the legislator failed to make investments that returned adequate electoral benefits. That possibility will be dealt with only indirectly here. I will point out, however, that one of the seven former JCAR members who did not return to the state legislature and the comnittee did so because he was elected to the U.S. House of Representatives. 96 Table 3.9. Number and Percent of JCAR Menbers Returning to the Comnittee— Measured by a) Number of Seats Available and b) Number of Menbers Returning to the Legislature (Same Chamber). 19731974 A. Total JCAR seats 8 Menbers reapp'ntd 6 % of total seats 75.0 B. Previous-term JCAR menbers in legisl'tr 6 Menbers reapp'ntd 6 % of total possible 100.0 19751976 19771978 8 8 4 50.0 5 62.5 6 9a 5 55.6 4 66.7 197919B0 10 6 60.0 10 6 60.0 19811982 Period .Mean . 10 7 70.0 63.5 7 7 100.0 76.5 aExceeds number of seats due to a mid-term resignation from the commit­ tee in 1977. That member later returned to the committee. Most JCAR members who returned to the legislature returned to the committee, a mean of 76.5 percent. assignments. 26 Some, however, accepted other (See Table 3.10.) Four senators, for instance, exited to an exclusive appointment on the appropriations comnittee, including one to appropriations vice-chairj they were assigned to no other committees or leadership positions. One senator left in 1975 to become majority floor leader and chair the senate business comnittee. Only one departing senator did not move to a substantial new position. 27 26. This percentage slightly underrepresents the degree of comnittee return, due to the special case discussed in the note to Table 3.9. If 1977-78 were calculated on the basis of 5 returnees of 8 , the value for that year would be 62.5%, raising the period mean to 77.8%. 27. Sen. Donald E. Cooper, R-Rochester, in 1975, continued to serve on two other comnittees (corporations and economic development, and judiciary), dropped labor and the JCAR, and added the commerce committee to his assignments. The resulting package of comnittee assignments seems particularly in line with traditional Republican policy interests. The fact that this single exception was a minority party menber m y reflect the difference in irajority and minority party perceptions of oversight efficacy reported by Lyons and Thomas. 97 No House menber left the JCAR without obtaining a substantial position elsewhere, but the positions did not include the Appropriations comnittee. Instead, they moved to major leadership positions or to chair comnittees other than Appropriations. 28 leadership positions included House Democratic floor leader, and House minority leader. New comnittee appointments included chairing the comnittee on Corporations and Economic Development (as well as vice-chair of judiciary) for one meirber; chair of Environment and Agriculture (and vice-chair of judiciary) for another. 29 28. One interesting case not included in the above time period is that of a menber still on the comnittee in 1984. He reported he did not request a JCAR appointment but accepted it as an alternative, and a favor to the House leadership, to avoid a fight for chair of the House Comnittee on the judiciary. He wanted to try keep his JCAR seat even if the judiciary leadership were made available to him in the next session. As it turned out, he gave up the JCAR seat in 1985 to head the House comnittee on Economic Development and Energy, continued as majority vice-chairperson of taxation and menber of the House comnittee on the judiciary, and added labor. 29. ihe difference observed here between House and Senate movements out of the JCAR may indicate that the ladder to Appropriations is "longer" in the House, or it may indicate other differences between the career ladders in the two chambers. The greater variation in seniority, nurrber of menbers, and comnittees in the House may allow or cause the existence of several career ladders. Appropriations is surely at the top of at least one, but perhaps not all of them, although its prestige and power are generally regarded, by inside and outside observers alike, as very high. The simple explanation may be the correct one; Table 3.8 shows the House Appropriations comnittee to have a higher mean seniority than is the case for House menbers on the JCAR. I look forward to someone else's work on this question. 98 Table 3.10. Movement out of JCAR, 1973-1981. 1973 Returning legislators exiting from JCAR Exited to: Appropriations Chr., other comnittee Major leadership pos'n Other 1 221 1977 1979 1981 0 2 4 4 0 0 0 0 0 0 0 1 1 2 1 1 0 2 1 1 0 0 0 0 0 Total 10 4 2 3 1 includes all previous-term JCAR menbers continuing in legislature who did not return to the comnittee. The nunber is equal to the difference between lines 1 and 2 of Part B, Table 3.9, above. TJot necessarily exclusive appointments. A single individual may have been elected to a major leadership position and also appointed to one or more comnittees. This measure reports only the primary responsibility. Thus, although specific patterns vary somewhat between the House and the Senate, it is clear that individual legislators have made significant investments in legislative oversight in the state of Michigan, just as has the legislature as an institution. 3.4 Efficiency and Uncertainty Reduction 3.4.1 Joint Comnittee Menbers The work of the Joint Comnittee on Administrative Rules is somewhat different from that of most legislative comnittees. In writing bills, legislators are involved in broad policy scope within a specific subject area. On the JCAR, responsibility tends to be simultaneously broader and more detailed. Because the comnittee deals with literally every 99 department of state government, comnittee members are required to have (or develop) at least some understanding of policy implementation throughout the full range of state government responsibilities. This goes beyond the understanding required of many of their legislative colleagues. In addition, they are required to address a level of detail which goes beyond that of legislation, since that is the inherent nature of rules content. Obviously, in the legislative process, legislators are required to vote on matters from the full range of state policy. Often, however, they can base their votes on the work or recommendation of respected colleagues, especially those serving on the standing committee of jurisdiction; individual legislators raintain a fairly narrow base of expertise. Menbers of the JCAR must operate quite differently. They cannot defer to the judgement of their party counterparts on the respective standing comnittees because those colleagues in most cases lack knowledge of the intricacies of the rule promulgation and implementation processes; moreover, they are unlikely to have any position on the issues before the JCAR. This is at least partially a reflection of the fact that most rules appearing before the comnittee deal with precisely those details that the legislature at large had previously determined it did not want to deal with. In addition, many current rules proposals are revisions of rules previously promulgated under statutes passed years earlier; such rules are even less likely to capture the interest of current standing comnittee menbers. To be an effective menber of the JCAR, then, requires uncommon breadth and depth of involvement, even though it is formally a restricted involvement. In sharp contrast 'to most legislative comnittees, the Joint 100 Comnittee on Administrative Rules lacks the power to amend. The traditional "mark-up" sessions, with their attendant bargaining and log-rolling, are singly not a part of the JCAR review process. This is not to say there is no negotiation (some of that has already been suggested and more will be shown in the next chapter) but menbers are not supposed to be developing new policy. Instead, their responsibility is to assure that existing policy is being properly carried out, that the intent of the legislature is being met, that the rule is "expedient", and that the right of "the people" to participate in the rule pronulgation process has been assured. This significantly changes the role of the legislator and affects comnittee decision making dynamics. JCAR menbers and staff alike spoke of the difficulty new menbers have in learning to work in this framework. The Joint Comnittee on Administrative Rules disposes but does not propose, at least not formally. It nay require additional hearings, it may make clear to an offending agency the language it would find more acceptable, it may suggest that certain parties need to reconcile their differences, but formally, it can neither require nor substitute language. It exists, at least on the surface, in an either/or world. Despite their restriction, meirbers uniformly characterized the JCAR as a powerful comnittee. They variously described it as a mini-legislature, the last arbiter, and the place where the action is. At the same time, they all spoke negatively of anyone who would use the comnittee as a means to further policy goals they had been unable to achieve through the legislative process. Comnittee meirbers were unanimous in their response to a question regarding what they would do should a rule come before them promulgated under a statute which they 101 had voted against: if the rule were in accord with the law and the intent of the legislature, they would vote for approval. One should not be misled by the above for the matter may not be so sinple. Several past or current menbers indicated they would be very careful in the above circumstances that rules did not exceed the act and that they did the bare minimum. In addition, one recent menber of the comnittee reported an interesting conversation with a predecessor. The earlier menber said he could usually find something to hang his hat on if he wanted to vote against a rule, although sometimes one would be so well drawn that he would have to vote for it, despite inclinations to the contrary. Such corments suggest both that the process is not as formalistic as it nay first appear and that menbers are more stringent in scrutiny of rules based on policy with which they disagree. There was another way in which comnittee menbers go beyond the formalistic confines of legislative intent. statement of "legislative intent." Michigan statutues carry no The language of the statute is supposed to speak for itself; this adds to the discretion of agencies and the comnittee alike. Particularly in the case of proposals for revision of rules authorized under old statutes, where circumstances may have changed considerably, there could be considerable ambiguity concerning what the original intent may have been and how it might apply in the new situation. Such ambiguity makes it possible for the comnittee to be more responsive to current policy preferences, whether of legislative colleagues, interest groups, or the public at large. Comnittee menbers understand the policy defining aspects of administrative rulemaking and, as one said, want to have that last bit of influence into the process. 102 Meirbers are likely to identify only indirect linkages between their JCAR service and their constituents. In interviews, most of them indicated receiving very few contacts from their districts in regard to a rule proposal. One recent head of the comnittee recalled only two contacts from his district regarding the content of proposed rules and he voted contrary to the requested direction in one of those cases On the other hand, requests for assistance from district residents are frequently rule related, giving comnittee meirbers a broader perspective on implementation problems; this can be helpful in judging the merits of current rule proposals. In addition, although the comnittee usually deals with rules with statewide application, menbers may interpret them in a localized fashion. One menber, commenting on rules stipulating the drugs and supplies required to be carried on emergency vehicles, noted his constituents travel throughout the state; he wanted to be sure they received adequate assistance wherever they might be. 31 One current JCAR menber has departed from the norm and actively uses the committee to develop constituency linkages. He sends letters to or otherwise contacts persons in his district whom he believes might have reason to object to a pending rule. This simultaneously provides him coimunication with constituents, shows him to be "doing his job", and provides his own personal "decibel meter" for monitoring the degree of controversy associated with rules as they move through the review 30. The more active role of interest constituencies as compared to geographic constituents is discussed in Chapter 5. 31. Former Rep. Ernest Mash, R-Dimondale, speaking during regular meeting of the Joint Committee on Administrative Rules, State of Michigan, December 6 , 1983. 103 process. It has the potential of increasing the electoral benefits of oversight activity. To the degree that Michigan's institutional structure provides a framework within which to do this, it lessens and indeed may overcome many of the oversight disincentives identified by Rosenthal and others. Rosenthal argued that the legislature's oversight specialists will have three key characteristics: a desire to improve the performance of government and to increase their personal knowledge of how government works, and a commitment to strengthening the legislature as an institution. JCAR meirbers seem to have a fairly clear vision of government and its appropriate relation to those governed, although they differ in these visions. Virtually all interviewed former and sitting menbers of the comnittee enuciated a concern for making government work and saw themselves as the people's advocates in the process of reviewing the rules. This is true in two regards. First, the merest hint that an agency failed to conpletely oonply with requirements regarding notices to the public and full and open hearings will almost guarantee a proposal will not win approval until the allegation is resolved. Second, the comnittee considers not only conformity to statute but also the likely inpact on the public of the proposed rule. They have statutory support for bringing such concerns to the rule review process but they seem also to possess personal predilection for such an approach. JCAR menbers frequently see themselves and each other as hard workers who are interested in details. Veteran menbers of the comnittee were very critical of any menber who lacked commitment to keeping up with committee "homework." Cn the other hand, they spoke with respect of 104 and deferred to the specialized knowledge in certain areas of their various comnittee colleagues. (In the process, they demonstrate in a much smaller circle the same sort of collegial deference which operates on a larger scale in the legislature as a whole between comnittees.) A single unresolved question raised by a respected colleague at a hearing could cause postponement of action on a rule. On the other hand, menbers sometimes got a reputation for always being down on a certain agency; in that case, their opinion might be disregarded by their colleagues. There are two fairly sinple ways in which we can get at least some sense of these legislators' attitude toward the institution they serve, and serve in. Where you sit may indeed foretell if not determine where you stand on the issue of legislative veto; it certainly seemed to for former and current menbers of the joint rules committee. 32 Without exception, they expressed support for the legislative veto in general and saw Michigan's change to the more restrictive reverse veto as a necessary tool for keeping the executive branch agencies under control. As one meirber told me, the reverse veto was "the greatest thing since 32. An intriguing historical exanple of this principle is contained in the career of former Governor William G. Milliken. In 1962, as a state senator, he voted to override then Gov. John Swainson's veto of a measure which would have required prior approval of rules covering "vessels carrying passengers for hire." This was an institutional conmitment over a policy commitment, for he had earlier voted against the measure. Fifteen years later, in 1977, then himself governor, he experienced the only override of a veto in his entire fourteen year as chief executive. The issue?— the requirement of prior legislative approval of administrative rules, the reverse veto. The dispute was iruch publicized for it evoked institutional and partisan wrangling of a type which was rare during the Milliken years. For a quick review, see Senate Journal. 1977, State of Michigan, pp.1563-1565, containing Senators Cooper, Welborn, and Allen's statements, and Swainson's veto message of 15 years earlier. 105 sliced bread"— everybody should have it. A slightly different perspective on institutional coirmitment is attachment as it relates to political airbition. Schlesinger *s three types of airbition— discrete, static, or progressive— seem appropriate here. Of the total of 15 House meirbers on the JCAR between 1972 and 1984, only two have subsequently run for higher political office, one for state senate (unsuccessful) and one for Congress (successful). In contrast, of seventeen senators serving during that time, five have later run for higher office: two for governor, one for U.S. Senate, one for U.S. House of Representatives, and one for Michigan Court of Appeals. Yet another is reputed to have ambassadorial ambitions. To date, none has been successful. Senate members of the JCAR thus seem to be more progressive in their anbitions than are House members, but in both cases, the substantial majority appear to be static. Specifically how they conpare to other colleagues throughout this period on this measure I do not know but the earlier seniority data suggest they are at least no more progressively inclined and may, in fact, be more statically inclined, the durable workhorses of their respective chosen chambers serving on the JCAR in recognition of its strategic location as a committee of influence. 3.4.2 The Legislature In what ways is this legislature as an institution served by the arrangements which have been described here and in Chapter 1? First, by 106 ensuring administrative rule conformity to statute, the committee's review allows the legislature to continue to write statutes at a lower level of detail than would otherwise be necessary in order to accomplish policy objectives. This alone is a substantial contribution to institutional productivity and efficiency. Second, by considering the likely impacts of proposed rules and seeking to lessen them to the degree possible, the committee seeks to make state government less onerous on the people than might otherwise be the case. To the degree that it is successful, it lessens citizen conplaints and probably enhances re-election prospects for all legislators. The JCAR may also serve the legislature in a less obvious way by providing a check on its own decision making processes. While the conrnittee has had members representing a broad range of political opinion over the years, mean opinion among JCAR members is probably somewhat more conservative than is the case for the state legislature as a whole. Conrnittee members, other legislators, agency personnel, and interest group representatives alike agreed with this assessment, frequently even volunteering it. Host saw it as "natural". If the legislature as an institution wants effective oversight of administrative rules, then it must assign that responsibility to persons who have some skepticism concerning regulation. Asked why that function could not be provided by the standing committees, several respondents replied that those committees were known by everyone to be "stacked" by 33. This was in addition to the problem, already cited, of old statutes in which current standing committee members had no interest, and standing committee members' general lack of knowledge of administrative processes. 107 supporters of certain interests. 33 The iirplication was that the legislature looked to the JCAR as something of a check on the standing conmittees as well as the agencies and interest groups. Finally, the conrnittee serves as a .centralized bastion enhancing the legislature's ability to fend off the constant forays of the bureaucracy into the legislative, policy-making arena. The bureaucrats' guerilla tactics of old are well known and oft-discussed: losing a regulatory fight in the legislative arena, the desired bureaucratic objective would come back buried in twenty pages of rules. Being well versed in this lore, the members of the JCAR can be on the lookout for every such incursion into legislative authority and vigorously protect it from outlanders. As such, these hardworking warriors win a certain amount of recognition and respect from their colleagues for putting up with the detail most of them are perfectly happy to avoid. Tongue in cheek as this may be, it captures a certain underlying flavor of the Michigan review process. There is a competition, a tension, between the branches of government, a certain "we/they" attitude which colors the enterprise and affects conrnittee deliberations. 3.4.3 Staff Roles staff. Role Overview. JCAR menbers serve a variety of roles and perform a number of functions as legislators, in addition to running for re-election every two or four years. administrative rules is not easy. Responsible legislative review of Although some rules are as short as one paragraph and as relatively sinple a matter as regulating motor boat 108 speeds during certain times of day on a single 50-acre lake, others can range into the dozens of pages and deal with very conplex subject matters. Staff support can greatly increase the efficient use of legislator time by identifying key issues and by culling out rules of district or other special concern. Staff meirbers can advise legislators of related constituent canplaints. They may discuss issues with department personnel and affected groups and individuals, providing legislators information which will help clarify the consequences of a rule, especially as it relates to inportant constituency groups. They may advise them of developing controversy, and even contact potential supporters or opponents of the rule for testimony. And finally, but not necessarily least, as one legislator commented with gratitude, they can make legislators look good by providing relevant questions to ask when departments present their cases before the JCAR. There are at least three sets of legislative staff potentially influential in the decisions of the Joint Conrnittee on Administrative Rules. They include 1) staff hired directly by the joint conrnittee; 2) minority party caucus staff; and 3) individual legislators' staff. These people serve as information filters in a variety of ways. Joint Comnitfcee staff, in addition to cannon staff responsibilities of maintaining records and comnunications, JCAR staff also have explicit responsibilities relative to agency hearings and rule analysis. The APA requires the agencies to hold public hearings on proposed rules. According to the JCAR's annual reports, JCAR professional staff members attend these meetings on a "random" basis. In practice, however, they usually attend only hearings conducted in Lansing, the state capitol. In addition, staff members report they are more likely to go to hearings 109 of agencies they have "had trouble with" in the past. Attendance gives the staff direct information regarding public reaction to specific rules and at least some direct check on the agency's notice and hearing procedures. As noted earlier, there are three statutory bases for evaluating the content of a proposed rule: conformity to statute, expediency, and predicted fiscal inpacts. In addition, the process by which the rule was developed is itself evaluated. Attention is given to the question of whether the agency has complied with all procedural requirements, both generalized (APA based) and specific(unique to the rule's authorizing statute). These are all areas potentially incorporated by JCAR staff into the rule analysis they prepare for conrnittee meirbers.^ JCAR staff both formally evaluate and informally advise. Attendance at the required department-sponsored hearings and contacts from agency and interest group personnel give the staff forewarning of most rules which will be contested by other participants when heard by the conrnittee. This knowledge is passed to the chairperson and other conrnittee members, specifically preparing them for some of the more difficult issues. The latter is an especially inportant staff role. The JCAR legislators evidence a strong dislike for unanticipated controversy. They recognize that some controversy is unavoidable, but 34. The report of that analysis effectively, although not formally, constitutes a staff recomnendation. staff menbers insist they present an analysis, not a recomnendation. No one was able to offer an example of an approved rule which had not received a favorable staff review. Neither did a random sairpling of conrnittee records indicate this having happened. Note, however, that while staff approval may be operationally necessary for rule approval it is neither fomally necessary nor sufficient. 110 want to be forewarned. This appears to be simple risk aversion. If caught unaware, they may appear uninformed, make statements which will later cause them difficulties, or commit themselves to action which proves troublesome. They much prefer to have time to work out a position and prepare a response. Good staff support is essential in this regard. Minority Party Caucus staff. The joint committee staff, although preparing and distributing information to all members of the committee, is regarded by minority party conrnittee members as majority party staff. Indeed, Democrats controlled the hiring of JCAR staff from the beginning of systematic review in 1972 through 1984. Republicans frequently invested in additional staff. Senate Republicans for several years had a central caucus staff member who regularly reviewed proposed rules. The review was similar to that provided by the JCAR staff in that statutory authority was the first check point. If a rule passed that standard, implementation issues were looked at. Republican staffers are, perhaps, more likely to additionally focus on the question of whether there might be a less burdensome or intrusive method of implementing the policy. best way to inplement the rule? Was this the Would it give the state agency too much power, or more than necessary to implement the statute? The Senate Republican staffer was frequently another of the contact points for departments working their rules through the system. Concerns raised by this person could sometimes be resolved even before they were conveyed to the senators; if not, at least the agency knew what questions they were likely to raise during the JCAR hearings. House Republicans approached the rules process differently. For Ill most of this time, central staff was assigned by subject area and rule review was divided accordingly, in the judgement of several people close to the process, the result was a lack of understanding of the rules process and very little attention to rules. House minority members generally lacked the second staff review provided Senate minority members. Legislator. Staff. The other main staff support for JCAR merrbers was through their individual staff positions. First-term House menbers were provided one secretary; second-termers or conrnittee chairpersons, one aide and one secretary. Representatives chairing the largest conmittees (e.g., appropriations, judiciary, taxation) would, in addition, have attorneys as part of their staff. 35 Majority party senators were allowed five full-time positions, minority party senators got three. 36 A few JCAR menbers have assigned a staff menber the responsibility of reading and comnenting upon all rules; others rely much more on the Joint Conrnittee staff. Probably most comnon is for personal staff to review rules concerning an area of special interest to the legislator-all agriculture rules, for instance, or anything particularly affecting cities. While findings are generally given only to a single legislator, they are inportant. This is especially true because of the tendency of conrnittee menbers to defer to district-based concerns raised by another menber. 35. Central staff included an additional six conrnittee aides/clerks shared by the 13 House conmittees, a central research staff, and public affairs and communications personnel. 36. Senate staff included an additional eight general counsel positions, plus 37 positions for the majority caucus and 26 for the minority caucus and 3 to 4 additional people on hourly pay. 112 Suntrarv. JCAR, minority party, and individual legislator staff menbers all serve as collectors, filters, monitors, and transmitters of information important to the rule review process. As such, they are themselves an important element of the rules review structure in the state of Michigan. At the same time, from an electoral persepctive, good staff support can increase the legislator's review efficiency and greatly reduce the uncertainty attendant to the review process and decisions, thereby increasing legislator benefits. A structure with less staff support would seemingly be both less efficient and less productive, and probably more risky. 3.5-Summary This chapter has shown that Michigan engages in substantial rule oversight, with over 150 transmittals a year processed by the legislature. Review activity and authority are highly centralized, located in the Joint Conrnittee on Administrative Rules. The process is seemingly as open as any other government decision making process, perhaps more so than some. The conrnittee appears to jealously guard the public's right to participate in the rule development and review process, partially out of its own competitive relationship with the executive branch. Several hypotheses regarding structure and oversight outcome were tested. The evidence provided strong support for the argument that structure makes a difference, in each case, more stringent oversight 113 mechanisms— reverse as opposed to negative veto, concurring versus sinple majority decision rules, and longer versus shorter review periods— resulted in lower rates of approval for adminstrative rule transmittals. The evidence was also quite clear that legislatures and legislators can and will make substantial investments in the legislative process. The Michigan legislature has made substantial institutional investments of staff, time, seniority, and number of menbers in the oversight process. Enough senior individual legislators have chosen the committee over other assignments that the conrnittee mean terms of service exceed chanber means and approximate or surpass those of their respective appropriations conmittees. Moreover, seniority increased with increases in conrnittee authority. These results suggest a strong interaction between structure and incentive. The presence of a highly centralized conrnittee with substantive authority which surpasses that of virtually every other legislative conrnittee and substantial support appears to have attracted senior menbers who then invest the personal resources necessary for successful legislative oversight. The system is operating in such a fashion as to maximize benefits while minimizing attendant uncertainties. In the process, it supports the theory that there is value in more direct, overt, oversight. Contrary to the "runaway bureaucrat" theory, Michigan legislators seem clearly to have found it worthwhile to invest greater resources in the oversight of the development of administrative rules. Neither do the results support the legislative "capture" theory; not only are rule approval rates nowhere near 100%, they have declined over time. Bureaucrats appear indeed to 114 have been pursuing their own goals; greater legislative investment in oversight of the administrative rules process has resulted in modification of a significant proportion of bureaucratic proposals. From the legislative perspective, the result is at least partially "constrained'* bureaucrats. Chapter 4 The Executive Branch Executive branch perspectives on the issue of legislative oversight of administrative rules differ greatly from those of the legislative branch. Legislators see the right to monitor the application of delegated authority; governors and bureaucrats see an intrusion into executive authority— a violation of the separation of powers between branches of government, counterproductive limitation on administrative flexibility, the application of political values in areas which require technical expertise, and unnecessary delay. This chapter shows how executive branch actors have responded to and participated in the legislative review process and various ways in which they try to manage that process while minimizing intrusion and uncertainty. It works from the general hypothesis that departments will choose strategies which reduce uncertainty and maximize rule approval. Departments vary considerably in the extent to which they are involved in rules promulgation, in the degree to which they coirply with the Administrative Procedures Act, the types of benefits which they have to offer to legislators and other actors in the system, and the amount of controversy associated with the rules for which they are responsible. 115 116 4.L Bureaucrats_in_a Rational Choice World— Expertise and Negotiation Michigan's legislative review conplicates the life of the bureaucrat involved in rules promulgation. merit of the proposal become important. Factors beyond the technical Bureaucrats are questioned by * the joint conrnittee on the extent to which they have involved those to be regulated and are required to document the controversy they have encountered. Under Michigan's rules oversight provisions, if the rule stirs controversy, bureaucrats face disapproval of their rule if they do not become negotiators. Department personnel at several levels are involved in the rulemaking process, but rulemaking activity generally follows the common hierarchical bureaucratic pattern. at a fairly low level. Noncontroversial rules are handled The greater the difficulty in resolving a rule controversy, the higher it rises through the administrative structure, sometimes involving a director's office, more rarely, the governor's office. Drafting of rules and primary contacts with advisory groups and regularly participating groups are generally the responsibility of lower level personnel. These are also usually the people who receive public testimony at the required department hearings and who initially present their department's case at the hearings before the JCAR. still, rules receive considerable attention from upper level officials. Department directors monitor progress of rules and are required to sign them before they are sent to the joint Conrnittee. A rule that elicits controversy 117 may well trigger action from the department director *s office. One former director of the state Department of Labor, for instance, estimated he had personally taken action on one-fourth of the department's rules proposals.3, Directors do, although rarely, testify at Joint Committee meetings, and may also use their influence with members of advisory or policy-making boards whose approval may also be required. Considerable amounts of staff time can be involved in drafting proposed rules, preparing for and conducting public hearings on the rules proposal, and consulting with interest group representatives and legislators prior to Joint Committee action. Having to repeat any of these steps is costly, and thus, in the administrator's eye, to be avoided if possible. Administrators think about these costs in very concrete fashion, "well, said one agency representative of a disapproved rule: 2 that just cost the taxpayers $15,000. Another department representative reported that publishing notices of hearings for a single rule cost $9,000.3 There is substantial incentive for the bureaucrat to get it right the first time. There are various ways in which bureaucrats may attempt to manage the oversight environment. Process strategies may include the timing of agency transmittals, mobilization of support resources, invoking outside 1. William E. Long, former director, Michigan State Department of Labor, interview, Lansing, Michigan, July 18, 1984. 2. Gregory Lyman, Deputy Director, Department of Natural Resources, exiting a meeting of the Joint Conrnittee on Administrative Rales, Lansing, Michigan, February 7, 1984. 3. Dennis Hall, in-charge, Special Lands Program Section, interview, Lansing, Michigan, JUly 12, 1984. 118 authority, using the unacceptability of the status quo as a lever to gain acceptance of an otherwise unacceptable rule, making incremental changes which cumulatively result in nuch greater change, using emergency rules as a "trial balloon," etc. Allocational strategies, in this context, involve bureaucratic decisions relative to the content of rules which specifically provide benefits to legislators serving on the Joint Comnittee. Arnold analyzed decisions agencies made regarding district military employment, water and sewer grants, and Model Cities grants for their relationship to program and appropriations support, within the rules context, however, agency strategizing of this type need not be restricted to district benefits. The joint committee reviews rules dealing with all aspects of state government policy making and its menbers are potentially targets of interest group support or opposition from other than those "natural" to their geographic districts. It is possible for JCAR menbers to develop both geographic and interest group constituencies. Administrative strategies could include benefits targetted to either. Talking about strategies inplies several things about bureaucrats and their environment. First, it assumes bureaucrats can distinguish between factors contributing to and those detracting from a favorable JCAR vote. Second, it assumes that bureaucrats have some control over these factors. Third, there is an assumption of an incentive structure which somehow relates approval rates to rewards. If each of these things occurs, one might expect department success rates to improve over time, i.e., to show an increase in JCAR approvals, given fulfillment of the ceteris paribus assumption. For it to hold in this case, decision rules must remain the same; the level of controversy associated with the 119 issues mist remain approximately stable (although specific content may vary considerably); and either JCAR membership and staff, agency staff, and interest group personnel and balance mist remain stable or changes in these factors must be without systematic effect on outcomes. A lack of increased approval rates over time could be interpreted in several ways. The most obvious conclusion is that bureaucratic learning does not take place in the rules promulgation process. Bureaucrats may be unable to distinguish between factors favoring approval and those which hinder such an outcome, perhaps because department personnel change too frequently, or possibly because their universe is truly random. Either would violate the first assumption. On the other hand, they may clearly perceive patterns and factors in the process but lack the means to influence them. Under these circumstances, the second condition would fail. The third assumption would be violated if the reward system does not act— either by department, agency, or individual— upon information relating to rules success. These activities may be so rare or such a small part of total workloads that they are given little weight in any reward structures.4 As will be shown shortly (see Table 4.1), seven of the twenty departments had less than one transmittal per year over the twelve-year period. The importance attached by these or other departments to the success of rules transmittals remains to be seen. Obviously, the departments must meet certain legislative demands in 4. There is also the possibility that agencies feel they have to "give a few" to the joint committee, i.e., that the JCAR has to require a few changes from time to time if it is to justify its existence. If that is the case, this would further reduce the relationship between approval rates and rewards. 120 providing the means for iimplementing law. Where rules already exist, however, the status quo continues to operate if the proposed rule is withdrawn or approved. This would seem to lessen both internal and external pressure for rules approval as coirpared to areas in which there were no rules at all. H ypothesis 8; Approval rates will vary by department. This sinple hypothesis offers a starting point for investigation of differences between departments as a tool in understanding factors potentially influencing outcomes. If departments experience equal success in winning approval for their proposed rules, examination of any one department should be sufficient to understanding of agency-controlled factors affecting outcomes. If approval rates differ between departments, then a broader range of analysis is required. Outcomes do vary by department. As shown in Table 4.1, rates of approval range from 64.5% to 100.0%. There is also considerable difference by department in the nuirber of rules submitted. Seven departments, accounting for only 1 .2 % of all cases, averaged less than one transmittal per year over the eleven years. Two other departments— Labor and Natural Resources— accounted for nearly half of all transmittals (49.6%) and averaged more than two a month. 121 Table 4.1. Total Transmittals and Overall Approval Rates, by Department and Frequency, 1972-1982. Department LesE_.tai J~per_yeai a u Civil Service Executive Attorney General Civil Rights Military Affairs Corrections Mental Health 1-6 Der veat m : Transportation Treasury Management and Budget State Police State Social Services Education ,7.-12,per, year, .(2),;. Agriculture Public Health 13-24 per. year..(2)x Conroerce Licensing and Regulation More than 24 oer year (2) t Labor Natural Resources Total (all departments) Mean (all departments) Transmittals N % Per Ceit Approved 0.1) 0.1) 0.1) 0.1) 0.1) 0.3) 0-4) 100.0 100.0 100.0 50.0 100.0 60.0 87.6 18 24 29 37 38 48 69 ( 1.1) ( 1.5) ( 1-8) ( 2.3) ( 2.4) t 3.0) ( 4.3) 77.8 75.0 69.4 70.3 73.3 64.5 86.9 87 105 ( 5.4) ( 6.5) 87.4 67.6 154 185 ( 9.5) (11.4) 79.5 73.0 285 518 (17.6) (32.0) 85.3 87.4 1617 (100.0) 1 1 2 2 2 5 7 ( ( ( ( ( ( ( 80.7 122 Does the overall correlation between veto rule and outcome hold at the department level? Table 4.2 shows rates of approval for ten of the departments most frequently submitting rules. Table 4.2. Department Approval Hates by Time Period and Change Between. Department Negative Veto {1972-1977) N % appv. Reverse Veto (1970-1982) N % appv. Change in percent btwn periods Agriculture - 12.3 43 93.3 42 81.0 Commerce 60 81.7 94 - 5.1 76.6 Education 97.5 72.4 40 - 25.1 29 Labor 165 87.3 82.8 - 4.5 99 Licensing & Regulation 114 - 4.4 71 76.1 71.7 Natural Resources - 11.2 91.2 80 80.0 91 Public Health - 26.2 83.3 63 42 57.1 Social Services 70.0 - 12.9 20 28 57.1 State 15 80.0 23 60.9 - 19.1 Treasury 10 90.0 - 25.7 14 64.3 Total— ten depts. 86.0 - 12.6 586 74.6 557 £ Includes all departments averaging seven or more cases per year, plus those with 1-6 cases per year and 10 or more cases in each time period. In every case in Table 4.2, approval rates drop between veto periods. The differences are not uniform, however. They range from less than five points in the case of the Department of Labor, to more than twenty-five for Education, Public Health, and Treasury. While there seems to be a clear systematic effect, there appear also to be department specific factors influencing outcomes. Although approval rates dropped between time periods it is still possible they could increase within time periods, i.e., possible that bureaucrats would improve success after adjusting to a specific set of review parameters. This is not generally the case. The most conroon finding is decline of approval rates, even within time periods. A few 1 23 departments maintain relatively stable approval rates. Only two show positive trends within time periods: the Department of Social Services in both periods, and the Department of Natural Resources in the reverse veto period (non-local cases only). (See Appendix B.) Those cases will be discussed separately later in this chapter. This is not to say that experience has no value. Grouping the data on the basis of the frequency with which departments submit rules shows some evidence that departments with the greatest number of transmittals enjoy higher success rates. In Table 4.3, the only systematic difference across groups within the overall data is the substantially higher success rate (89.9%) of the greatest frequency group as compared to those of the other four (73.8% to 76.6%). Looking at the data within veto periods reveals something different, however. Under the negative veto, from 1972-1977, when effective negative action was much more difficult for the JCAR, there is no systematic difference across frequency groups. The range is from 79.0% approval to 91.7%, but the extreme values occur in the two highest categories; approval rates for the three lower frequency groups lie between these values. With the change to the reverse veto and the greater ease of effective negative JCAR action, the picture changes considerably. Approval rates for the four groups with the least transmittal experience range from 63.2% to 70.7%; for the two departments with more than twenty-four transmittals per year, the rate is 86.9%. In addition, although all approval rates drop between time periods, the difference is nuch greater for those groups with fewer transmittals. It appears that departments more frequently submitting proposals somehow utilize either 124 experience or some other resource in such a way as to disproportionately enhance their effectiveness. Table 4.3. Cutcomes Aggregated by Average Frequency of Transmittals by Department, 1972-1982, and by Veto Period. Average N of Transmittals Per Year • Qrerall (1972-1982): % approved ($ submitted) 75.0 (20) 73.8 (263) 76.6 (192) 74.3 (487) 89.9 (655) 80.7 (1617) 81.8 (11) 88.3 (111) 88.5 (87) 79.0 (214) 91.7 (410) 87.5 (833) 66.7 ( 9) 63.2 (152) 66.7 (105) 70.7 (273) 86.9 (245) 73.5 (784) -8.3 -4.8 ■ % approved (# submitted) Reverse Veto (1978-1982) % approved (# submitted) Change; -15.1 -25.1 -21.8 -13.8 ^ o r departments within categories and their individual totals, see Table 4.1. This is in keeping with Bruff and Gellhorn's finding that agencies submitting rules for programs which encountered periodic review were likely to be more successful than agencies submitting rules which had to be promulgated only once. If legislators believe they have only one opportunity to raise objections or take corrective action, they will be less inclined to be lenient or negotiate than in the case where there will be frequent other contacts on the same program. Obviously, there is more opportunity for this when a department is submitting twenty proposals a year than in the case where they submit only one. Individual departments differ greatly in rulemaking success, as previously shown (see Tables 4.1 and 4.2.) The data suggest that, rather 125 than using strategies to inprove their success rates, departments must work hard even to maintain current rates. Virtually every respondent who had an opinion about the reason for the change to the requirement of prior legislative approval cited the timing of agency rules transmittals as the cause. own report (1978) gives this explanation. The Joint Committee's In addition, several writers in legislative/executive relations or state government mention this as an administrative strategy. Such a strategy rests on the assunption that rules received near the end of a regular session or during the interim will not be acted upon or will be given only a cursory review, thus taking effect with a minimum of legislative interference; it also requires that rules can take effect without winning prior legislative approval. In the first time period, under negative veto, rules filed the month before the recess could take effect in one of three ways: 1) approval by the coirmittee prior to the legislative recess, 2) approval by the conrnittee at a special sunnier meeting, or 3) automatic approval as a result of no conrnittee action. If departments were using late transmittal as a deliberate strategy during this period, they were slow about it. Table 4.4 shows that for the first five of the six years (1972-1975), the proportion of transmittals in the month irnnediately preceding the summer recess was below the expected rate of 8.33% 5. "There were many instances where state agencies would submit rules to the Coirmittee toward the close of session at a time when the Conrnittee would not have an adequate opportunity to review the rules. Often state agencies would refuse to withdraw their rules to give the Conrnittee this opportunity . . . . [This] caused most Conrnittee menbers to support Senate Bill 609 of 1976." "Report of the Joint Committee on Administrative Rules," 1978, State of Michigan, p.6. 4 126 (one-twelfth of a year). But in 1976, the proportion of rules submitted in that month junped to 17.4%, and in 1977 was still at 10.9%. I doubt that legislators had these kinds of statistics. More likely, what they had was increased pressure in dealing with rules and, quite possibly, a growing resentment against what appeared to be a deliberate tactic to keep them from properly carrying out a legitimate duty. The bulge in pre-recess transmittals may indeed have been a factor precipitating the change requiring prior legislative approval of rules. Table 4.4. Per Cent of Rules Transmittals Received by JCAR in Month Prior to Sumner Legislative Recess, by Year. Year Negative Veto (1972-1977): 1972 1973 1974 1975 1976 1977 Period Total, % Reverse Veto (1978-1982): 1978 1979 1980 1981 1982 Period Total, % Annual Transmittals Transmitted During Month Prior to Recess: % of total 90 131 176 135 172 121 833 7 8 13 10 30 1A 82 7.8 6.1 7.4 7.7 17.4 10.9 9.8 162 157 158 166 141 784 21 15 18 23 22 99 13.0 9.6 11.4 13.9 1£*£ 12.3 aExpected value is 8.3% (one-twelfth of a year.) In the second time period, rules no longer took effect if the conrnittee failed to act— prior approval was required. behavior is less clear. Here, strategic Sending late transmittals to the conrnittee under these conditions could be counter-productive if the conrnittee 127 failed to meet early enough in the recess to consider such rules prior to the expiration of the allowed review period. could have benefits, On the other hand, it if there were some demonstrable urgency about the rules, the conrnittee might agree to take them up, but give them only superficial scrutiny. Sixty-six rules which were under consideration at special summer meetings or at the first meeting after the summer recess under reverse veto; fifty (75.8%) were approved, fourteen (21.2%) were withdrawn, and two (3.0%) were disapproved. Thus, outcomes for the pre-recess month during these years vary little from outcomes over the entire period (73.5% approved? 22.8% withdrawn? 3.7% other— see Table 3.4, p. 80.) While legislators may have felt cranped by the nuirber of rules transmittals salt to them near the close of session, it appears at least on a statistical basis to have had no significant inpact on outcomes. There are probably several reasons why the strategic timing of rules presentation is not a more successful tool for Michigan administrators. First, Michigan has an essentially year-round legislature.6 There are anple opportunities for the legislature, whether through the Joint Conrnittee on Administrative Rules or elsewhere, to discipline an agency which would blatently use such a device. Second, the review conrnittee is allowed sixty days in which to act and may vote to extend that an additional thirty days. This allows the coimdttee a fair degree of flexibility and reduces the time pressure the departments can exert. If agencies did adopt a strategy of aiming . 6. Summer legislative recesses for 1972-1977 averaged 64.7 days; during 1978-1982, they averaged 69.6. Source: Journal of the senate, state of Michigan, editions 1972-1982. 128 for that narrow "window" in the legislative calendar when the entire legislature might not sit for sixty days, the JCAR still is errpowered to act during the legislative interim, and does so. Moreover, should the JCAR fail to act under the current reverse veto provisions, the rule would die for lack of approval. When asked about the possibility of strategic timing, several administrators said they would not dare engage in such activity. On the other hand, departments can and do directly mobilize support for rules they have pending before the Joint Conrnittee. A first step in mobilizing such support is maintaining positive relationships with as many involved as possible. Departments do this in a general way through regular provision of services to regulated groups, through the lists they maintain (by law) for the notification of hearings, through public information canpaigns and general public service. are triggered by specific pending rules. More specific actions In some cases, the law requires the department to create an advisory group for the promulgation of implementing rules or even the periodic review of existing rules. In other cases, there may be existing groups which by tradition, law, or good politics are involved early in the drafting or negotiation of a proposed rule. Cooperative relationships established in earlier stages may become critical to rule approval by the JCAR. This is particularly true for a rule to which there has been opposition; even more so if the rule has been withdrawn and resubmitted. In the eight months in which I observed the conrnittee there were a number of occasions on which persons who had testified earlier in opposition to a rule later returned to testify in support of the revised language. People would make a 200-mile round 129 trip or more to say only: "I've met with the department and I'm satisfied. I support the rule." What are the dynamics supporting such a system? Certain benefits are available through departmental services or programs. Certain benefits are desired by the menbers of the public testifying. Legislators on the committee are reluctant to corrmit themselves to a vote on a rule as long as there is reconcilable conflict and time to do something about it. A person testifying earlier in opposition to a rule is apparently considered by the coimdttee to be continuing in opposition until it is told differently. In testifying against a rule, the person makes known a specific goal, presumably one not being provided for in the rule as proposed, in later testifying for the rule, the person implicitly confirms certain (frequently unstated) compromises and aids the department's ability to achieve the now-mutual goal. Each participant needs the other once an issue has been publicly raised at the JCAR level. Several agency administrators explicitly stated their dislike of surprises in the JCAR hearing setting. They at least want the issues, participants, and positions identified by that stage. They would prefer not to take a rule before the committee without "having their ducks in order"— no opposition remaining, and an adequate show of support. is not always possible. This In the event that a rule is necessary but conflict seems unavoidable, they will go to individual merrbers of the committee, explain the issues and problems, why the rule is needed, and where opposition is likely to come from and why. Administrators who do not do this and "lose" a rule before the JCAR are described by other actors as not having done their homework. 130 Losing a rule through failure to anticipate legislative objections costs more than the $9,000 for publishing of new hearing notices. can also require a major investment in staff time. It After losing a rule as a result of an unanticipated concern raised by a single menber of the coirmittee, one assistant division chief and the policy coordinator for his department contacted all menbers of the conrnittee individually, discussing the areas outlined above. In addition, since that rule disapproval was so unanticipated, the decision was made to routinely shepherd in the same fashion all rules pending before the conrnittee for awhile to make sure it did not happen again.7 Yet another department strategy is to invoke outside sources of authority as the basis for specific rule language. The Attorney General's office or the Legislative Service Bureau could be and often were cited as the basis for specific content. The attorney general or the Bureau could be and were cited as the reason for writing a rule in a certain way. For many years, the same was true of federal statutes and authorities. Now the departments are finding this a less productive strategy. The JCAR is challenging claims of requirements emanating from these sources. A representative of the Service Bureau and an assistant attorney general now are required to attend all meetings of the JCAR and may be called upon to substantiate any claim made to their authority. In addition, departments are required to document any claim made 7. Gregory Lyman, Policy Coordinator, Michigan Department of Natural Resources, interview, Lansing, Michigan, March 3, 1984. 131 regarding federal language being incorporated into or otherwise serving g as the base of a proposed Michigan rule. In addition, although difficult to assess, sequential incremental rules changes may be used as a device for accomplishing cumulatively greater change. Certainly, there are particular sections of the administrative code which appear numerous times as the identifying code for rules proposals. An almost reverse strategy, with possibly the same long-term result, is represented by cases in which an unsuccessful rule proposal was split into smaller segments > JCAR approval granted for the noncontroversial sections, and the more difficult sections pursued separately. A related strategy is to put a previously unsuccessful rule into a package of otherwise noncontroversial rules, while Joint Conmittee members tend to look extremely unfavorably upon such tactics, there are situations in which this may be acceptable. One of the more interesting such cases involved a definition of live birth, required by the new state public health code which took effect September 30, 1978. After several unsuccessful attenpts to get a definition approved by the JCAR as a separate rule, it was finally, two years later, included in a rule listing a number of other definitions relating to vital records, and q approved. The JCAR counsel was fully advised regarding the inclusion of this particular definition within the nine-page rule document, as 8. Several respondents agreed this resulted from objections to the tendency of some departments to substitute a Michigan "shall” for a federal "may”, potentially a matter of substantial regulatory inpact. 9. The case will be discussed in detail in Chapter 5, as an example of interest group participation. i 132 were the menbers. The Department of Public Health administrator who had coordinated this activity was one who believed in doing "homework” with the Coiwnittee.^ The regulatory status quo is more than a static fall-back reference position for actors in the process. The departments have the power to manipulate the status quo through the emergency rule procedure. Lanbert's example of the nuclear emergency preparedness rules is a case of precisely such activity.^ He had two concerns; 1) that it was not really an emergency and 2) that the review committee inappropriately held up approval when the regular rule came before them. demonstrate something more important, however. The case may By rescinding an existing rule at the same time it had promulgated the emergency rule, the department forced the committee to consider any related proposed rule from the perspective of a new status quo. Lairbert saw the case as an example of mutual abuse of the system; it might more productively be regarded as sophisticated agenda control maneuvering. Using department approval rates assumes that departments are the appropriate unit of analysis for detecting evidence of bureaucratic learning and administrative strategies. cases than in others. This may be more so in some The state Department of Public Health, for instance, has adopted the strategy of highly centralizing the development of rules. This includes having a single high-ranking staff member responsible for assisting department sub-units in the drafting of 10. George Van Airburg, State Registrar and former director, Michigan Department of Public Health, interview, Lansing, Michigan, July 16, 1984. 11. See Laxbert, pp.128-131. 133 initial language, reviewing for consistency with other department rules, negotiations with the assistant Attorney General, publication of hearing notices, preparation of all documents for Legislative Service Bureau, Attorney General, and JCAR action, etc. This would appear to maximize, on a departmental basis, any learning going on within the subunits. Public Health is one of the larger departments, however. Not all departments have such a person, although all departments eventually clear their rules through their director's office. Despite this strategy, Public Health, working in a number of areas which have excited public controversy, still has trouble with its rules. Approval dropped from 80% under negative veto (1972-1977) to 57.7% under reverse veto (1978-1982). It was not enough to be well-organized. 4,2 Rule Withdrawal and Resubmittal:. Effects on Outcomes In Chapter 3, the role of controversy in delaying action was discussed, we will look here at what happens to rules which experience the greatest amount of delay in the review process— those which are withdrawn. Administrators control the decision to withdraw, using this technique as a response to rules which experience controversy during JCAR hearings. Most withdrawn rules are resubmitted; many are eventually approved. Once again, though, overall outcomes are affected by the governing veto rule and the differing substantive impacts of minority opposition. (See Table 4.5.) 134 Under the negative veto, 72 cases (of an initial 748) were withdrawn after their first presentation to the JCAR. sixty (83.5%) of those proposed rules eventually passed, with varying degrees of modification, most of them on the second attenpt, but some only after the third or fourth transmittal, in twelve cases (16.6%), agency personnel gave up (in eight cases, after the first try; in four, after the second) and did not resubmit a rule which had received a negative hearing before the conmittee. of the seven rules originally disapproved under negative veto, five (71.3%) subsequently won approval. two were not resubmitted. The other All three rules which were disapproved cm their second try ultimately were approved in modified form. Of the rules eventually approved, 75.4% (49 of 65) made it on their second try; 21.5% on the third; and 3.1% on the fourth. The ultimate approval overall of 65 of the 79 cases (of an initial 632) previously unsuccessful under the negative veto means that 82.3% of all such cases finally were allowed adoption in some form under the negative veto. Clearly, agency persistence paid off. In the case of the reverse veto, forty-five (31.3%) of the rules originally withdrawn were dropped at some stage, although one not until after its fifth presentation to the conmittee. On the other hand, 68.7% (99 of 144) of rules originally withdrawn were eventually passed. Rules originally disapproved (including no action and inpasse) fared far worse. Ten of the fifteen (66.7%) were never resubmitted even once; another (6.7%) was not resubmitted after a second unsuccessful attenpt. Only three of the fifteen (20.0%) cases originally disapproved under reverse veto eventually won approval. The remaining case resulted in an inpasse on its sixth appearance before the conmittee. 135 Table 4.5. Subsequent Outcomes of Rules Transmittals Originally With­ drawn or Otherwise not Passed by the JCAR, by Veto Period. Original outcome: subsequent outcome tfegati.V5L.VetQ;. Out­ come 2 Out­ come 3 Out­ come 4 4 13 2 2 Out­ come 5 Outcome 6 Pinal Outcomes N % 1972--13.7.7. withdrawn fn«72) Not resubmitted Approved Withdrawn Disapproved Disapproved. _(n=7y Not resubmitted Approved Withdrawn Disapproved 8 45 16 3 (72) 2d 4 1 0 ( 7) 16.7 83.3 12 60 (72) (100.0) 28.6 71.4 2 5 1 ( 7) (100.0) 45 99 31.3 68.7 (144) (100.0) 11 3 73.3 20.0 1 (15) 6.7 (100.0) Reverse. Veto:. 1978-1982 Withdrawn fn=1441 Not resubmitted Approved Withdrawn Disapproved Disapproved fn=151 Not resubmitted Approved Withdrawn Disapproved 28 85 20 11 (144) • 10 3 2 0 (15) 9 11 7 4 7 3 1 1 1 1 1 1 1 1 aCases counted as "disapproved" in reverse veto period may include instances of inpasse or "no action." ^Sum of cases withdrawn and disapproved in one column equal total cases in next colunn. °Pinal disposition of cases in first column, regardless of number of times transmitted. Includes two sets of rules which took effect despite JCAR disap­ proval due to failure of the legislature to pass concurrent resolution of disapproval. 136 Overall, 56 transmittals (35.2%) unsuccessful under reverse veto were never resubmitted. Of those rules eventually approved, 86.3% (88 of 102) were approved on second presentation; 10.8% on the third; 2.9% on the fourth attenpt. The eventual success rate for all previously unapproved cases here is 102 of 159, or 64.2%. The outcome of initially unsuccessful rules differs, then, between time periods. Unsuccessful rules are much more likely never to be resubmitted under the reverse veto (35.2%) than is the case under the earlier negative veto rule (17.7%). Table 4.6 shows this to be a statistically significant difference (p - .01) of substantial strength (Yule’s Q * .421). Table 4.6. Incidence of Resubmittal after Unfavorable JCAR Action, by Time Period. Negative Veto (1972-1977) N % 65 (82.3) (17.7) _14. 79 (33.2) ...... Resubmitted Not resubmitted Column Total % Reverse Veto (1978-1982) N % 103 (64.7) (35.2) -5& 159 (66.8) . . Row Total N 168 _2a 238 % (70.6) (29.4) (100.0) X2 = 7.785 p = .01 Yule's O = .421 ^Cell values taken from "final outcome" column of Table 4.5. Administrators make the decision to resubmit a previously withdrawn rule. The lower likelihood of their choosing to do so under the reverse veto is one more indicator of the substantive inpact of the change in the decision rule. The fact that some rules are resubmitted, whether after a prior disapproval or withdrawal, and that at least some of them are approved 137 suggests one more measure of approval rate, an adjusted measure taking into account approvals as a proportion of "original cases." Calculated in this fashion, there were actually 748 cases between 1972 and 1977 12 (833 transmittals less the 85 which were resubmittals) and 632 cases between 1978 and 1982 (784 transmittals less 152 resubmittals). Of these cases, 729 and 576 were eventually approved in some form. This yields adjusted approval rates of 97.46% and 91.14% for the two time periods, * The adjusted approval rates are considerably higher than their unadjusted counterparts. Approval goes from 87.5% to 97.5% under the negative veto, from 73.5% to 91.1% under the reverse veto. This only underlines the earlier finding that the most common outcome of review is approval. Legislative oversight of administrative rules m y slow the bureaucracy but it does not paralyze it. Administrators are not guaranteed approval of their proposals; if they were, there would be no reason to have oversight. Neither are they summarily dismissed from further consideration if their proposals receive an unfavorable first hearing. Negotiation and bargaining are regular and important aspects of the rule review process in the state of Michigan. 12. The number of resubmittals includes all transmittals shown as other than "not resubmitted" in Table 4.5 "outcome" columns 2 - 6 ; n's exceed those in the "final outcome" column because some cases were resubmitted more than once. 138 A.3 Benefits. Publics, and Outcomes— Two. Examples Table 4.1 gave the approval rates for each of the departments of state government. The variation among them was considerable. This section looks at specific exanples from two major departments handling very different kinds of rules and existing in quite distinct public environments. The departments are the Department of Natural Resources(DNR) and the Department of Social Services(DSS). 4.3.1 Department of Natural Resources Given the level of controversy involved with environmental issues through the 1970s and into the 1980s, one might be surprised to find the Department of Natural Resources (DNR) with one of the highest overall approval rates. Arnold's distinction between local and general benefit policy may partially explain this outcome. Examination of the DNR cases reveals a set of rules accounting for almost three-fourths of DNR transmittals during this time which dealt with local control of boating (on individual small lake, river areas) or of hunting and firearms use in small (typically sub-township) areas. approval rate of 94.0%. (See Table 4.7.) These transmittals had an 139 Table 4.7. Outcome of Department of Natural Resources Rules Transmit­ tals by Type of Rule, 1972-1982* Tvoe of Rule Hunting/Watercraft Other N % N % Approved Other Column N % 346 22 368 (94.0) ( 6.0) 96 42 138 (72.7) (69.6) (30.4) ROW Total N % 442 64 506 (27.3) (87.4) (12,6) (100.0) X2 = 55.02 p = .001 YUle's Q = .746 The hunting and watercraft rules are undoubtedly the simplest example of local benefit cases in the Michigan rules system. The DNR has such cases in most of the state's counties over time, and is able to serve legislators and local interests alike by helping local groups reach consensus on these controls. The request for the rules is usually triggered by a situation of increasing traffic on a lake or substantial new building in an area which has been previously used for hunting. These rules are unusual in that the local unit of government votes on them before they are submitted to the JCAR. These rules are generally regarded as noncontroversial but they are not necessarily so. Opposition would most likely come from persons who are not permanent residents in either the hunting or water areas to be used. Unless they subscribe to a newspaper local to that area and read the legal notices they are unlikely to know the restrictions are being considered. Thus, the input which is likely to be received is in favor of the rule and directly related to a small, clearly delineated geographic constituency. Costs to the state DNR are only the expense of 140 local publication and those incurred in assigning one of it officers to the required meeting and perhaps one or two other meetings with local people. Posting of the restrictions is the responsibility of local officials, as is on-going enforcement in most cases. In the extremely rare event that opposition arises at the JCAR level, the department can alert the original petitioners, advising them of the need for their testimony at the appropriate committee meeting. These factors have combined to yield an approval rate on these rules of 94.0% for the eleven years. During the negative veto period, rates varied from 87 to 96 percent approval, with a mean of 95%; during the reverse veto period, they varied from 73 to 100 percent, with a 91% mean. (See Appendix B for annual data.) The rules process is much more difficult for the DNR in many other areas of its responsibility. Here, overall rates of approval are only 69.6% (Table 4.7)• The DNR non-local cases varied from 72 to 92 percent approvals during the negative veto period, with a mean of 80%, and 40 to 100% during the reverse veto period, with a mean of 60%. (See Appendix B for annual data— "DNRo".) Annual rates during the latter period showed steady inprovement through the five year period, the only such case of all the departments. While this offers at least partial support for a theory of applied bureaucratic learning, the year in which there was a 100% approval rate is the one with the lowest nunber of rules submittals for-the entire eleven year period. One might argue that what was learned was maximum possible avoidance. 141 The multiple atteirpts to revise part 4 of the Water Quality Standards are an exanple of problems encountered in non-local policy areas. 13 In 1976, the department proposed revisions to rules which had been promulgated in 1973. Despite considerable public controversy they presented the rules five times in less than seven months to the JCAR in 1978-1979. Each time the rules were withdrawn for further work. It was only after the fifth withdrawal that the Water Resources Conroission established an advisory task force to work with it on the rules. The rules were intended to regulate point-source discharges of toxic naterials into surface waters. Original controversy about the regulations was between an industry-related coalition led primarily by the Dow Chemical Conpany on the one hand and an assortment of environmental and public interest groups on the other. Industry wanted restrictions only on proven carcinogens; environmental groups preferred zero discharge of any substance which had not been certified harmless. During this time the groups in opposition were working separate from each other, each trying to use the various lobbying points to get its full preference enforced. Neither was successful; industry was faced with continuing regulatory uncertainty although still complying with the old rules, and the no-discharge interest groups knew that contaminated discharges were still occurring. The message from the Joint Conmittee had been clear— the groups had to reach a resolution themselves. 13* Information for this section was obtained from Joint Conmittee records and interviews; interviews with department and interest group personnel; and from "Proposed Rule 57 Comments and Summary," Water Resources Commission, Michigan Department of Natural Resources, August 3, 1984— a 298-page record and summary of the 1984 hearings on the rules. For a review of the history of the rules, see pp.2-3 of the "Comnents and Summary." 142 The advisory task force provided the first forum in which the opposing interests worked together. rules was ready. By June of 1980, a new package of These were transmitted to the state Water Resources Commission (WRC) for its approval, and then sent on to public hearings. Heavy controversy continued, despite the conpromises which had been reached through the advisory group. The WRC decided to concentrate on one rule in the package— Rule 57, that portion defining allowable levels of discharge of industrial wastes into the surface waters.^ It established a new advisory group solely for the purpose of reaching acceptable compromise on that rule. Proposed Rule 57 revisions have oscillated from the very broad, with wide discretionary powers to the department in enforcement, to very specific language which would require little interpretation. One of the problems was to devise a rule which would simultaneously allow the department to regularly incorporate into its enforcement program the most current technical information without necessitating recurring rule revisions while also providing a workable level of regulatory certainty for the regulated industries and municipalities. After monthly meetings of the task force and staff, tentative resolution of the problem was presented to the WRC in December, 1982. General language was adopted with specific standards to guide application in different cases. It took it until November of 1984 before the rules were resubmitted to the JCAR. They were approved in December. What had the state DNR done in the meantime regarding the discharge 14. Called "Rule 57" on the basis of its administrative code number: R.323.1051. 143 of toxic substances into the state's waters? cpite straightforward. Department policy was If there was a known carcinogen, even if it was not on the list included in the rules promulgated in 1973, it was be regulated— by guidelines or under the general authority of the existing Rule 57 language. According to the head of the department's representative to the Water Resources Commission there was no challenge to this system. People on both sides, he said, recognized we could not afford to randomly discharge carcinogens into the environment. 15 Participants were willing to recognize interim DNR authority even though they are unwilling to comnit themselves to a new rule. In a situation like this, the department faces well-organized, well-educated, opposing interests with resources to continue a battle over a period of years. Department personnel, both in public testimony and in private interviews, express recognition of and sometimes frustration at the dual publics in their environment— a public especially aware and supportive of high standards contrasted with the need of industry for regulatory certainty and competitive equality if not advantage. Environmental regulation is a highly politicized arena in the state of Michigan. This is just as true for rulemaking as it is for legislation. What is particularly interesting here is the refusal of the Joint Conmittee to make the policy decision and approve a rule which lacked consensus. Contrary to Bruff and Gellhorn's concern about deals cut by the legislative rule-approving body, the coirmittee returned the 15. Dr. Dennis P. Tierney, Michigan Department of Natural Resources, interview, Lansing, Michigan, August, 13, 1984. 144 decision to tfte public. groups, however. This is not merely an abdication to regulated In this case, the participants included active advocates for a broader public interest. Nonetheless, it could legitimately be charged that the joint conmittee did not fulfill its responsibility. The APA does not require public support of a rule, it requires only that it be in conformity with the statute under which it was authorized. Clearly, the system has been responsive to constituent concerns, as Lyons and Thomas, Ethridge, and Lairbert would predict, but in this case the result was long-delayed action. Political decision makers waited for a consensus preference which they could then endorse, suffering the least retribution possible. In this situation, the Joint Conmittee members looked more like risk averse rational actors than they did Rosenthal’s altruists or Fenno's good government types. the decision? Why suffer the consequences of making The generalized benefits provided by the DNR in this program totally lacked differential positive support for the legislators. There was nothing to trade. 4.3.2 Department of Social Services The Department of Social Services (DSS) has one of the largest budgets in Michigan state government, 16 generally the greatest number of enployees, relatively few rules, and twenty-one loose-leaf policy manuals. Respondents from the JCAR and its staff, from other 16. Social Services was second only to Mental Health from 1972-1979 and was first from 1980-1982. « 145 departments, and from interest groups all identified DSS as an agency often not in compliance with the state's APA. It was frequently mentioned as operating under guidelines rather than rules (a matter of controversy regardless of department). Explanations for this varied. Some respondents attributed it to the many requirements on the department which come from the federal level. Others saw it as a matter of professional orientations— professional "do-gooders" who feel no one else is qualified to tell them what to do. Still others analyzed failure to comply as a result of the public with which the department frequently deals. As one legislator saw it: "Those people don't know if they're rules, guidelines, or the Ten Commandments." One respondent thought the department had so nuch trouble getting rules approved in the past that they had just given up. These arguments have varying merit. The federalism explanation is not sufficient. Many departments of state government have a high degree of federal involvement, whether in dollars or shared jurisdictions. Exanples include, to varying degrees, the Departments of Natural Resources, Public and Mental Health, and Labor, Most respondents who originally cited the federal relationship as the explanation for the relative independence of the Department of Social Services recognized this similarity if questioned about it. The "professional do-gooder" argument is more difficult to assess. On those occasions during my observation of the conmittee that DSS personnel testified, this attitude did not seem to be present, but one would assume the department would be careful in its choice of representatives for that forum. On at least one occasion, a DSS representative specifically spoke in support of the review process. Appearing before the conmittee for 146 the third time on a transmittal completely updating the rules on children's camps, the assistant director of child welfare licensing stated: "This process works. I have really come to respect it. 17 The process he spoke of was a combination of specific statutory requirements and general APA requirements. The rules were promulgated under a statute requiring the department to create ad hoc advisory committees conposed of representatives from the state agencies and the organizations affected by the act. Although there was strong support when the rules came before the JCAR, there were also several specific objections. The committee withheld approval until most of those objections had been resolved. 18 In this case, the DSS professional gave recognition to benefits of the negotiatory process; in other circumstances, without a required advisory committee and organized outside groups, attitudes and results might be different. The majority of persons regulated by the Department of Social Services lack the kind of educational, informational, organizational, and economic resources utilized, for instance, by the camp operators in the previous example or by the state Chamber of Commerce in its systematic monitoring of rules promulgation and implementation. They also have a very different cost-benefit calculus and incentive structure. The individual welfare mother is unlikely to take on the 17. David Fitzgerald, Assistant Director of Child welfare Licensing, Michigan Department of Social Services, testimony before the Joint Committee on Administrative Rules, Lansing, Michigan, April 10, 1984. 18. After this third hearing, the rule was withdrawn, and a new rule transmitted, carrying all the modifications necessary to incorporate the compromises worked out in this latter stage. This resubmittal was approved almost imnediatedly. 147 department, let alone a conmittee of the legislature which she probably does not know exists. On at least one occasion, however, the state Welfare Rights Organization (WHO) testified before the conmittee. The encounter apparently was unsatisfactory to both the organization and the joint conmittee. The JCAR's previous special counsel reported that the VJRO criticized the conmittee for its 8:00 a.m. made it too difficult to attend. meeting time, saying it The conmittee chairperson reportedly responded that he got up at 4 a.m. to be there on time; they could too.19 There is outside evidence to suggest the person who thought current DSS noncompliance was a response to past lack of success may be right. Between 1972 and 1982, the Department of Social Services had the lowest overall rules approval rate of all the agencies. 20 Moreover, approvals dropped from 70.0% from 1972-1977 to only 57.1% from 1978-1982. Although I doubt the various departments have a precise measure of how the other departments are doing, and even lack such a measure for themselves, one suspects there may have been growing DSS frustration with the rules process. Critics of legislative review have frequently predicted administrative noncompliance as a response to legislative oversight which became too burdensome. Annual approval rates for DSS provide additional support for this 19. Kenneth E. Sanders, former special counsel to the Joint Committee on Administrative Rules, interview, Lansing, Michigan, September 30, 1984. The WRO corrplaint may have reflected problems with public transportation; if so, the conmittee interpreted it quite differently. 20. The only exceptions are two agencies who had such a small number of transmittals that they were discarded in much of the subsequent analysis: Civil Rights, with two rules, and Corrections, with five. 148 view. Social Services was the only department which showed increases of approval rates within both time periods. In each case, it finished in the last two years with 100% approval of rules transmittals. The findings parallel those of the earlier DNR example of inproved approval rates, however. The 100% success rates were in years in which either one or two rules were submitted, fewer than in any of the earlier years in their respective time periods. (For full data, see Appendix B.) What may have been learned is to entirely avoid the review process for all but the most certain winners. None of the explanations decisively account for why the Department of Social Services has been allowed by the legislature to so regularly skirt APA requirements while others, such as Natural Resources (with only a slightly higher 1978-82 approval rate for its non-local rules cases), follow it much more closely. It is my guess that it is a simple matter of lack of political incentives for such an exercise of legislative authority. DSS clients undoubtedly vote in nuch sraller proportion than do many other legislator constituent groups; they are also much less likely to render a complaint against "their" state agency than are teachers or automobile industry representatives, for instance. Even business interests, due to the need for regulatory certainty, lobby for environmental regulation. Lack of participation by DSS clients, in contrast, results in little threat of punishment for inaction and few rewards for action. 149 4.4 ihe governors.Mftjpc-frrtoc?. The governor of Michigan is an elected official, and head of the executive branch. Legislative oversight of administrative rules poses problems for the governor in two ways: one, as protector of the executive branch from undue legislative assumptions of power; and two, as manager of the executive branch. If increasing legislative authority means a reduction in gubernatorial authority, it lowers the ability of the governor to manage, and reduces potential electoral benefits. There is ample evidence that Michigan governors have been sensitive to legislative oversight of administrative rules as an issue in legislative and executive authority. There have been at least six related vetoes by four different governors. Vetoed legislation included 21 attenpts to require prior legislative approval of a specific rule of all rules, 22 or or to delay and increase uncertainty in the 23 implementation of rules . In each case, the governor was trying to maintain existing executive control over the administrative rule-making 21. Gov. John Swainson, 1962 ESB 1301, dealing with "vessels carrying passengers for hire," and requiring prior legislative approval of implementing rules. 22. Gov. Harry Kelly, 1943 SEA 69; Gov. Wn. Milliken, ESB 609, Jan. 18, 1977, and ESB 419, Aug. 5, 1977— all would have required prior legislative review of all administrative rules. 23. Gov. George Romney (by pocket veto), 1968 ESB 1374, requiring prior presentment to the Joint Committee on Administrative Rules, and Gov. Wm. Milliken, 1974 ESB 1064, suspending the 60 day review period limit when the legislature was in recess for more than 14 days. 150 process. Management of the executive branch is difficult under any circumstances in Michigan. The governor has a limited executive authority. government. There are currently nineteen departments of state Two department heads (secretary of state and the attorney general) directly elected by the people. A third department (education) is directed by an independently elected board which, in turn, appoints the state superintendent of public instruction. Five more departments (agriculture, civil rights, civil service, corrections, and natural resources) are headed by gubernatorially appointed bi-partisan cornnissions serving staggered terms (some as long as eight years) which, in turn, appoint the department directors. Thus, the governor has direct control over only eleven of the nineteen department heads. Even in these departments, control may be limited. OA There are many independent boards and conmissians assigned to these departments for budget, personnel and other "housekeeping" services only. 25 Clearly, Michigan provides an exanple of Hamm and Robertson's "weak" executive. As an elected official, incentives for gubernatorial involvement in the rules process are very similar to those discussed earlier for the 24. These departments are Coirmerce, Labor, Licensing and Regulation, Management and Budget, Mental Health, Military Affairs, Public Health, Social Services, State Police, Transportation, and Treasury. 25. There are over 250 independent boards and coirmissions within Michigan state government, ranging from the State Accident Fund Advisory Board and State Board of Accountancy to the Board of Veterinary Medicine, Water Resources Comission, and the workers' Compensation Appeal Board. See various editions of Elective, and Appointive State Officers. State of Michigan. Department of Management and Budget, State of Michigan, and current update, available from the Office of the Governor. 151 legislators. There is no constitutional limit on the nuirber of terms a Michigan governor may serve. Therefore, the electoral incentive has potentially as much force for the governor as for any state senator or 26 representative, Administrative rules, in general, are not a first level concern for a governor. Attaining major legislative and budgetary goals undoubtedly has higher priority. We would expect rules likely to draw gubernatorial attention to include those on which there is substantial disagreement between the department and outside interests as to what ought to be done, particularly if one of those outside parties is a past or potential source of gubernatorial electoral support. A group may take a rules concern to the governor's office, expecting pressure to then be applied on the department. A governor who can demonstrate executive control in such a situation may garner electoral support from the action's beneficiaries. drav&acks to such a scenario, though. There are possible It could result in minimum cooperation from agency personnel in the future, and it might produce opposition from interests who felt they had suffered as a consequence of gubernatorial intervention. If Fiorina is correct about interest groups liberally assessing blame but being stingy in granting credit, one would predict that the rational actor governor would prefer a rules pronulgation system which runs smoothly at the department level without 26. Michigan had only six governors in the forty years between passage in 1943 of its first Administrative Procedures Act and its most recent gubernatorial election, in 1982. Four of the six were re-elected at least once: Harry F. Kelly (1943-1946) and G. Mennen Williams (1949-1960) during the era of two-year terms, and George Roimey (1963-1969) and Wn. G. Milliken (1969-1982) under four-year term provisions. Williams was a Democrat; the other three, Republicans. 152 requiring this type of intervention. Nonetheless, there are times when the governor takes action on rules. Rules issues may become a part of a general reform package, part of a specific program advocated by the governor, or merely a small opportunity to assist a past or potential supporter. There are a number of ways beyond general advocacy of regulatory reform in which the governor might be involved in the rules promulgation and oversight process: 1) initiation of a rule change through an inquiry or request to an agency; 2) request that an agency consider certain factors or interests in preparing a rule; 3) involvement in negotiations between interest groups and agencies regarding rules content; and 4) intervention on behalf of agencies or those regulated as rules are under review by the JCAR. Michigan's governors in general have not had a high level of involvement in the process, particularly at the point of legislative oversight. JCAR members and staff alike report very limited contact from the governor's office regarding rules transmittals. Both the current and a former head of the conmittee, for instance, reported very little pressure from the governor's office. Contacts regarding rules proposals were described as rare, "very discrete and tactful," and with 27 no "improper involvement." The typical approach was a quiet message that a specific rule coming before the conmittee was very important to the department, with the request to give it careful consideration. 27. Rep. Michael Griffin, interview, Lansing, Michigan, January 4, 1984; former Rep. Thomas j. Anderson, interview, Lansing, Michigan, August 1, 1984. The language offers interesting insight into legislative perspectives on appropriate relations in this area between the branches of government. 153 Interviews with persons serving as legislative liaison to two different Michigan governors confirmed that departments are, in general, expected to manage their own battles in the promulgation of rules and their negotiation through JCAR approval. If a rule is perceived as essential, however, and is encountering problems, the legislative liaison may privately lobby members of the conmittee in an effort to increase support. Neither had ever publicly testified in support of a rule while on the governor's staff. This may represent, from the governor's perspective, a risk minimization strategy and is, perhaps, a source of what legislative actors are interpreting as "discreteness" in this context. Agency personnel may make a large investment in a given rule proposal and develop a strong commitment to a certain language or approach. This can make it very difficult for them to accept changes suggested in public hearings. One former legislative liaison reported that word from the governor's office sometimes provided the support needed for a department director to get the needed changes in language from a recalcitrant underling. Such gubernatorial involvement had the added advantage of allowing the director to avoid taking personal responsibility for requiring bureau personnel to act against their own preferences. The Michigan Administrative Procedures Act specifically provides for gubernatorial involvement in rules promulgation in two ways. First, it stipulates that once a rule has been approved (i.e., cleared by the Legislative Service Bureau, the Attorney General, and the Joint Conmittee on Administrative Rules) departments mist present it to the governor's office at least ten days before filing it with the Secretary 154 of State. The governor is not required to sign or in any other way acknowledge the rule in order for it to take effect. Neither is the governor prohibited from instructing the department to withhold the rule from filing. 28 This has never occurred and the provision seems to operate solely as a commnication device within the executive branch. The other APA language regarding gubernatorial involvement deals with a type of rulemaking which occurs outside of the processes discussed thus far. The Michigan APA provides for the promulgation of emergency rules. These rules, which may be promulgated for only a six-month period and renewed only once, do not go through the steps otherwise prescribed by the state's administrative procedures act. To promulgate an emergency rule requires only that: 1) the "agency finds that preservation of the public health, safety, or welfare requires the promulgation of an emergency rule without following the notice and participation procedures"30? 2) the agency states its reasons; 3) the governor certifies concurrence in the finding of an emergency; and, 4) the rule and supporting documents are filed with the secretary of state. Prom the beginning of routine rule review in 1972 through the end of 1982, a total of 212 emergency rules received the governor's signature and were filed with the secretary of state. This represents 28. The result would be for the rule not to take effect. 29. Capitol folklore has it that the provision is the result of an enbarrassment suffered by former governor Roimey. The governor had responded to someone coirplaining about a rule that the state had no such rule. He was incorrect; it had been filed the day before. 30. Michigan APA, section 24.248(1). 155 13.1% as many rules as those promulgated through the regular process during the same time period. Table 4.8. Year 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 Totals (See Table 4.8.) Number of Emergency Rules and Regular Rules Filed, by Year, 1972-1982. Type of Rule Emergency Regular N N 8 11 24 24 24 21 23 29 16 24 - . 8.. 212 Proportion: Emergency/ Regular 90 131 176 135 172 129 162 157 158 166 141 1617 8.9 8.4 13.6 17.8 14.0 16.3 14.2 18.5 10.1 14.5 5.7 13.1 Source for emergency rules: "Supplement to 1954 Michigan Administrative Code", Nos. 70-101 (1972-1980) and "Michigan Administrative Code 1979, Quarterly Supplement," Nos. 1-12 (1981-1982), coitpiled by the Legislative Service Bureau and published by the Department of Management and Budget, state of Michigan. The requirements of Michigan's APA are such that it is extremely unlikely that a processed rule will be promulgated within a year of its initial drafting; most estimates are that it is more likely to require two years. An emergency rule may be a means of quickly implementing a new program while awaiting approval of regular rules, it may also provide covering authorization for existing rules which have an expiration date but which have encountered delay in reapproval through the regular rules processes. Emergency rules may be adopted to protect the state from imninent federal sanctions, including the loss of funds * or federal enforcement of certain regulatory programs. They may also 156 reflect response to a newly discovered health hazard, such as in the PBB contamination case. Under these circumstances, outside observers might agree that an administrative "emergency" exists; it is less certain that the "preservation of the public health, safety, or welfare" is always involved. On the other hand, an emergency rule may represent strategic department behavior, seeing how a rule "flies", perhaps acclimating a regulated group to a new level of regulation prior to having to submit it to legislative review. It may also be a device to rescind old rules, forcing the joint committee to consider a new rules proposal from a different status quo base than would otherwise be the case. Host emergency rules do eventually become permanent rules. In the case of the Department of Labor, for instance, the department which filed 41 of the 212 rules (19.3%), all apparently became permanent rules. 31 Twenty-two of these 41 transmittals were actually the initial promulgation and extension of a single set of 11 rules filed in 1978 in response to federal Occupational Safety and Health Administration 32 standards. The MIGSHA standards were made permanent in 1979. Approval 31. Emergency rules are not printed in the Administrative Code, The supplements to the code carry only a notice of the promulgation of an emergency rule, the dates effective, department and division responsible, a three or four word description of the subject matter, and an address from which a copy may be obtained. Nonetheless, cross-referencing these data with conputer sorts of the information entered from the Joint Committee's "Daily status Report" made it possible to track subsequent action on at least 95% of the emergency rule cases. 32. The rules dealt with safety standards for machine guards and devices, abrasive wheels, floor openings or platforms, fixed ladders, fire exits, overhead gantries and cranes, powered industrial trucks, hydraulic power presses, mechanical power presses, metalworking machinery, and woodworking machinery. 157 Is not always easy, however. The Department of Public Health promulgated emergency rules relating to a definition of live birth, but it took four subsequent'attempts over a three year period to get a permanent rule through the Joint Committee. The departments vary greatly in the rate of emergency rules promulgation. In those first eleven years of routine rules review, among those departments filing at least five emergency rules, the proportion of emergency rules to regular rules varied from a low of 2.5% for the Department of Natural Resources, to 43.8% for the Department of Social Services, and a high of 71.4% for the Department of Mental Health. To what degree is the governor in control of or involved in the emergency rule process? Pour persons intimately involved with the process reported that the governor's activity is limited, but can be determinative when directed toward the agencies. 33 A former gubernatorial legislative liaison reported a point at vrtiich it seemed more and more emergency rules were being promulgated. (See Table 4.8.) They began*to scrutinize emergency rules more carefully, requiring 33. Informants here included William E. Long, former director of the Michigan Department of Labor, and earlier, legislative liaison to former governor William G. Milliken; iarry Tokarski, legislative affairs director for current governor James J. Blanchard; Whiter Wheeler, III, Chief of the Office of Legislative and Legal Affairs, Michigan Department of Health. 34. See Table 4.8. His service did, in fact, coincide with the peak year for emergency rules, 1979. Emergency rule promulgations have subsequently declined. 158 the departments to more fully justify their need and to explain why they had not used the regular rule-making process, 35 A current administrator serving as a clearing point for the last six years for all rules from his department cannot recall any case where the governor has refused to certify a rule. He reports his department makes very certain they have an emergency before approaching the governor's office for support. Respondents reported several cases where the governor's interest had been very important in the promulgation of an emergency rule, e.g., transportation of nuclear wastes, and PCB contamination problems— two issues which received widespread public attention. In the latter case, the governor appointed his chief of staff to coordinate and monitor departmental responses to and resolution of the problem. There were no examples offered, however, of an emergency rule action initiated by the governor's office. It seems that executive branch rulemaking activity, whether through the regular or the emergency process, remains basically a departmental function. 4.5 An Executive Branch Overview This chapter started from the hypothesis that departments would choose rules promulgation strategies which would reduce uncertainty and maximize approval rates. Rule-making was shown to be a hierarchical 35. The same respondent also suggested that the legislature may have begun to notice and to have questioned the governor about the increase in emergency rules. 159 activity, with non-controversial proposals involving lower level personnel, higher level personnel acting when necessary. Legislative review was discussed as an intrusion into administrative perspectives and routines relating to the rule-making process. Rule-making was shown to be a costly process in Michigan, one in which there are sizable budget incentives for administrators to be successful in the enterprise. . It appears that many Michigan administrators operate out of a strong professional orientation but find the system hinders them in multiple ways. Interest groups, especially business groups, use the rules oversight system as a device to achieve their own goals, often to the frustration of administrators. At the same time, agencies directly impacting a less organized public may be able to operate more independently, more directly out of professional norms, as in the case of the Department of Social Services. Substantial areas, although probably not the mjority, of the rule-making arena are inherently conflictual. Changes in the Michigan oversight system have expanded the opportunities for expression of those conflicts. Michigan administrators have had to accept increased political participation in and inpact on administrative rule-making processes. It was found that successful rule-making is very difficult for Michigan administrators. Rather than being able to utilize experience and employ strategies which resulted in improved approval rates over time, it appears that Michigan administrators engaged in rule promulgation activity are struggling in a highly uncertain and perhaps even hostile environment. All departments saw approval rates drop as a 160 result of the change to the requirement of prior legislative approval although they varied in the degree to which they were affected by that change. The two departments submitting the greatest nunber of rules per year were found to drop notably less than the others. One of those was the Department of natural Resources which was shown to be a special case, due to a large nunber of uniquely localized, generally approved rules. The utilization and viability of a series of possible strategies was investigated. Strategic timing of transmittals was found not to have been used systematically when it would have been most useful, and to be of limited usefulness now with a year-round legislature, possible 90-day review period, and the requirement of prior legislative approval. Agencies were found to engage in active mobilization and negotiation cairpaigns to enhance rules approval. Agencies offered as successful exanples by others actors in the process were recognized for the contacts they maintained with joint committee members and interest groups alike. Agenda manipulation strategies are engaged in to some degree by departments, but cautiously; there are ample opportunities for legislative retaliation if this becomes a regular pattern. Internal organizational structures have been devised in some agencies to facilitate rules promulgation processes, but have not resulted in high rates of approval. Looking at the kind of services provided by one such office (in the Department of Public Health), however, strongly suggests there would be an even lower approval rate without this investment of resources. Michigan administrative rule-making is a decentralized, fragmented activity from the executive branch perspective. Although Michigan 161 governors have acted a number of times to protect the executive branch from further legislative instruaions, they have been unsuccessful in stopping increased legislative oversight of administrative rules. With a strong system now in place, governors participate only infrequently, sometimes through the emergency rules process. Emergency rules, requiring the governor's signature, were found to be pronulgated 13.1% as frequently as were rules going through the regular public hearing and legislative review process and to serve a variety of functions, including preservation of the public health, avoidance of federal (frequently economic) sanctions, agenda manipulation by altering the regulatory status quo, and initial implementation of new laws. If this bureaucratic structure were runaway it should be enjoying a much higher rate of success in the presentation of rules, and should not have to invest so heavily in negotiation with other actors. On the other hand, if it were fully under legislative control it should be experiencing less conflict. Although there are certainly portions of the system which enhance rational actor legislator goals, the level of conflict associated with Michigan administrative rule-making strongly suggests a professionally-oriented civil service "trying to do its job" in a world only partly of its making, in other words, a constrained bureaucracy. Chapter 5 Die Public and Legislative Review Previous chapters have focused on government actors in the oversight process with only peripheral discussion of the role of the public. This chapter looks more directly at the role and dynamics of public participation in the rule promulgation and review process. The public, of course, is more than a monolithic entity acting on government. It is constituted of a multitude of individuals and groups among whom resources and incentives for participation in the process vary greatly. Chapter 1 posited member benefits as the primary goal motivating interest group participation in the processes and procedures surrounding the legislative oversight of administrative rules. Although most "public" participation in the review process is by groups, some participants may act as individuals. The incentive for them is also the possibility of attaining a certain benefit. For both, whether group or individual participants, the benefit may be monetary or more symbolic. In recent years, for instance, the Joint Committee on Administrative Rules has acted on rules ranging from those dealing with the definition of conmercial fishing zones to the definition of live birth. The first had clear economic inpacts for a certain sector of the state's 162 163 residents, the other had no apparent economic inpact on any party; both were among those cases which have involved nultiple hearings and multiple resubmissions, A major change in the Michigan administrative procedures acts over the last forty years has been the increasing definition of rights of citizens relative to administrative processes and procedures. By the beginning of the time period covered in this study, Michigan's Administrative Procedures Act specifically provided for public participation at several stages in the rules promulgation process. Participation might involve the filing of a request for a rule change, comnent in writing or at public hearing on the content of a formally proposed rule, or testimony at the time a rule goes before the Joint Conmittee on Administrative Pules. In addition, some statutes require the administering departments to establish advisory groups to assist in the development or periodic review of rules. Among the most notable of such requirements cure those of the Michigan Occupational Safety and Health Act (MIOSHA), under which more than seventy advisory coranittees have been created.^" Several statutes provide for financial support for 2 public participation in administrative proceedings. This does not yet mean there is equal participation or benefits for all parties affected by the promulgation of administrative rules in the state of Michigan, Legislative review of administrative rules occurs at the end of a 1, These advisory groups are in addition to the more than 250 state boards and coumissions referred to in Chapter 4, 2. Several acts, for instance, have specifically provided for paying travel expenses and/or a per diem for public participation in utility rate hearings by the Public Service Conmission and similar conpensation for the MIOSHA-related rules development advisory committees. 164 long chain of governmental decision making. Xt is a relatively unreported part of government activity— stories about administrative rules, particularly the legislative review thereof, rarely hit the front page or any other page of the general press. Some critics of legislative review have charged it is so far down the line that only the most conmitted will continue to this stage. Even those deeply coirmitted may be unable to participate at the level they would like because they have exhausted their resources at an earlier stage. In this chapter, I shall try to establish whether the legislative review process operates as an additional or an alternative point of influence for those seeking benefits through government action, and the inpacts upon participation and outcomes of the resultant finding. Mho among the public participate, using what resources, with what effect, and under what circumstances? 5.L Organization and Influence Both the process and the provisions of the Michigan APA are such as to encourage participation at an early stage in the rules promulgation process. An interested party may propose a rule, including changes to existing rules, even directly writing desired language, nuch as private sector participants write legislation-and then find sympathetic legislators to introduce it. Other participants may not become involved until the agency serves public notice that a new rule is pending. The Michigan APA requires the 165 departments to publish notice of hearings In at least three newspapers of general circulation. They are usually one of the Detroit dailies, an upper peninsula paper, and either the Grand Rapids or Lansing dailies.3 These notices are published in the same general format as other required pubic notices and are probably noticed infrequently by most menbers of the public, even if they regularly read a paper in which the notices are published.^ Probably more important in stimulating participation at the department hearing stage is the APA requirement that the departments maintain lists of persons and organizations wishing to be notified of such hearings. This allows organized groups to regularly receive notice of pending rules and to take action accordingly. This might include formal board action, notification to members through a regular newsletter or special bulletin, preparation of testimony, organization of a lobbying campaign with the agency, or even contacting related groups to check their positions or coordinate action. Private individuals may have their names placed on the notification lists and may pursue similar strategies. There are also opportunities for participation after the formal hearings and before presentation of a rule to the Legislative Service Bureau and the office of the Attorney General for clearance prior to presentation to the Joint Committee. The departmental public hearings frequently result in changes to the rule as originally proposed by the 3. Grand Rapids is the second largest city in the state; Lansing is the third largest and also the state capital. 4. A totally "unscientific" index of this problem is my own experience. After six months of casually watching for notice of such hearings in the Lansing state Journal, the only daily newspaper published in the state capital, my file holds clippings of exactly three such notices. 166 agency. Interested parties who maintain contact with the agency through this period may be able to additionally influence those changes. To what degree do these activities take place? considerably. It seems to vary The most easily documented participation is public comment, whether at the hearing or in writing, on a proposed rule. Since 1980, agencies have been required to file with the JCAR rule submittal a regulatory inpact statement which includes a listing of 5 those groups affected by the proposal. . Agencies also report the number of hearings held on the rule, and the number and identification of persons in favor and opposed to the rule.*’ This is also a conmon part of the content of testimony before the JCAR. it is not unusual that less than a dozen persons appear at the departmental hearing. times, although rarely, there may be hundreds. At other 7 Organizational resources are important determiners in a number of ways of the extent and influence of participation in the rules promulgation process. It is not unknown for an unaffiliated, private, individual to testify before the committee, but it is extremely rare. Host participation is clearly generated through organized group activity. Where do proposed rules come front? Many are the result of action 5. See Michigan 1980 PA 445 at 24.245(2) 6. Unfortunately, front a research standpoint, those records are maintained for only two years after the filing of a rule, the period during which an appeal on procedural grounds can be filed against a rule. 7. The 1983 hearings on rules dealing with state payment of compensation to farmers with FCB-contaminated silos are the most recent example of this higher level of participation. 167 by the national government. This is especially the case for the departments of Labor and Natural Resources (which are also the a departments with the greatest nunber of rules transmittals) and the Department of Social Services. Additional rules transmittals result from atteirpts within departments to solve a specific administrative or technical problem. Sometimes the proposal is an adaption of department policy to a new technology, as in two cases early in 1984 dealing with new motorcycle license testing procedures and changes in the breathalyzers used by the state police and other police agencies. Other changes may be atteirpts to streamline administrative procedures, reducing the nunber of forms or permits required to engage in a regulated activity. Yet another set may be generated by conplaints of regulated parties about conpliance difficulties; the problem is not necessarily one of a wish to evade regulation, but to make it more workable. Such conplaints may be registered by individual businesses, but they are much more likely to come through one of the state-wide organizations who have staff whose responsibility it is to monitor the rules process and its inpacts. Among the most effective in this regard are the Michigan Manufacturers Association (MMA) and the Michigan State Chamber of Conroerce. Both the MMA and the state Chamber regularly have a staff member in attendance at JCAR meetings, in addition to monitoring proceedings, this person maintains a personal relationship with legislators and 8. Together they accounted for 49.6% of all transmittals from 1972 through 1982. 168 department personnel alike. The position also carries responsibility for keeping menbers informed through regular newsletters of pending rules which might be of interest and coordinating menber contacts with JCAR menbers and hearing testimony if they believe it necessary. Few volunteer organizations can afford to maintain this degree of attention and only one private individual within my eight months of observing the Q committee had a level of participation which even neared this. The Joint Coranittee sees itself both as protector of legislative prerogatives and guardian of the people in their rule-related dealings with the executive branch. Originally formally authorized only to act on the basis of the conformity of the proposed rule to the confines of the statute on which it was based, and then "expediency", recent amendments to the act have also given the committee information and authority allowing it to act on the basis of the fiscal impacts of proposed regulation. The committee has always acted on more than these formal criteria, however. When asked what the primary responsiblity of the canmittee was, the four-time head and fourteen-year member of the committee responded first with "fair to the public." He next talked about intent of the statute, 9. That single case was the mother of a child who had died at a summer camp, she was tenacious in keeping informed of the progression of the new rules package regarding summer canp regulation and appeared at every related JCAR hearing, in addition, she individually contacted both department and committee staff and JCAR members. She achieved major portions of the changes she desired regarding means of notification of illness but the committee agreed with the department in not requiring camp directors to notify parents of all health conplaints. Several JCAR menbers had been camp counselors and remembered homesick campers who got well with a little more attention and involvement in camp activities. 169 and making sure the bureaucracy was not doing something just for its convenience and as a result, inconveniencing the public.'1'® Whatever the legalities, this attitude as protector of the public against bureaucratic power and self-interest has been a part of joint comnittee considerations throughout the time of systematic review.11 This suggests several bases for interest group lobbying of the joint committee. Opposition to a proposed rule could charge lack of conformity to the underlying statute, an unfair burden on whomever or whatever was to be regulated, or agency convenience which would inconvenience the public. Sometimes all three are charged. Support for a proposed, on the other hand, might cite careful conformity to the statute, evenhandedness in application, and a service to the public. The cover form which accompanies a rule to the comnittee upon initial transmittal usually claims at least two of these factors, and frequently all three. Groups obviously vary in their ability to demonstrate any of these factors. Arguments regarding the underlying statute are more likely made by paid lobbyists than private individuals. Arguments about the balance of agency convenience and public service are most convincing when made by persons with access to information and/or experience within the administrative structures. Private individuals may charge such 10. Thomas J. Anderson, menber of Michigan House of Representatives, 1966-1982, interview, Lansing, Michigan, August 1, 1984. 11. Rep. Anderson served on the comnittee for his entire time in the legislature, and headed it in alternate sessions from 1974 until he left the legislature in 1982* This attitude was shared by current members of the committee, as demonstrated in interviews and numerous public consents at comnittee meetings. 170 problems but be overpowered by the professionals. The basic issue of fairness is the one most likely invoked successfully by those with less information or sophistication. A witness who can document uneven application or the potential for such can almost always force change in a rule, in recent years, language in a rule which provides for enforcement "at the discretion of the director" has been almost uniformly rejected by the committee. It should be noted, however, that comnittee responsive may be based on something more than sensitivity to fairness alone; discussion in earlier chapters would suggest that this responsiveness would also have utility as a further risk minimization strategy. 5.2 A Few Words on Campaign Contributions How does one gain access to the comnittee? Traditional means advocated in civics classes on "how to influence your legislator" are relevant in the rules oversight process, perhaps particularly so because of the committee's concern for fairness. That is not all that seems to count, however. Although I did not attempt a systematic investigation of campaign contributions to menbers of the joint comnittee, several things are readily apparent from a cursory look at recent data. 12 First is the balance between district and non-district based contributions. 12. Michigan's state campaign finance reporting began in 1978. Members 171 receive contributions from persons in their geographic constituencies but these are most likely relatively small contributions from individuals or the county Democratic or Republican committees. It appears from initial examination of the financing records that interest group constituencies, i.e., groups based outside the legislator's district and with no apparent direct or distinctive relationship to the district, are more important sources of caupaign♦funds for members of this comnittee. This finding mirrors legislators' experience regarding rule-transmittal contacts. As reported in Chapter 3 , almost all inputs are from interest groups, not individual district constituents. Given that balance, the second notable finding is the breadth of interests which contribute to JCAR menbers' re-election, in 1978, the first year in which Michigan required the reporting of campaign contributions, JCAR menbers, and especially the chairperson, received contributions from FACs from a broad range of professional and occupational groups (from chiropractors and lawyers to contractors and petroleum jobbers), labor groups and major enployers (United Auto Workers to General Motors), private financial institutions and credit unions, and utilities. Not all contributions were from economic interests, however; several members of the comnittee also received contributions, for instance, from Michigan Right to Life. The list of contributors in the first year of reporting alone— totaling over 150 groups— mirrored the wide range of issues which come before the joint committee, both economic and noneconomic. As one looks at the record over time, most organizations seem not to be rewarding legislators for individual votes so much as buying access. There are numerous cases where a given member of the committee 172 is receiving contributions from what are generally regarded as likely opponents on many regulatory issues. Indeed, one menber received on the same day contributions from the united Auto Workers, the Ford Motor Company, and the Michigan Automobile Dealers' Association. 13 As discussed in Chapter 3, individual comnittee members "specialize" in certain rule-making areas, such as agriculture, labor, mining and logging, highway patrol, education, or urban issues. Many such interests are district or occupationally related, but not all. The existence of these areas of rule specialization enhances both the development and potential influence of "interest group constituencies" for individual menbers of the comnittee. There appears to be some variation by party, but there are groups which contribute to virtually all menbers of the committee. There are many other groups which contribute to only a small nunber of legislators, perhaps no more than ten or fifteen, but include among that nunber, menbers of the JCAR; those contribution patterns would be of particular interest in a systematic analysis. Such limited evidence tells nothing about how contributions to members of this committee differ from those who are menbers of other committees, does not address the possibility of contributions made because of membership on another committee rather than one's presence on the JCAR, and reveals nothing regarding the size or frequency of contributions. (Some groups made pre- and post- primary and general election contributions; others contributed only once.) That work 13. Campaign finance reporting documents, Rep. Virgil Smith (D-Detroit), May 4, 1984, on file with the Secretary of State, Lansing, Michigan. 173 remains for a separate effort, Nonetheless, it appears to tell us that interest groups find the meirbers of the JCAR worthy of their financial attention. To the degree this is the case, it offers additional evidence of the pivotal influence of the conmittee. 5.3 Economic Groups— Federalism. Faction and State Regulation James Madison warned us nearly two hundred years ago of the dangers of faction and proposed a solution— the federal structure.'1'4 The federal structure poses several problems, however, in the context of state regulatory policy. Madison believed that the concerted action of interest groups (factions) could best be controlled by requiring them to conpete against each other in the national arena. He was pessimistic about the ability of state legislatures to resist the pressures of such groups. He recognized the potential problem of small interests looming large in the more limited geographical domain. While Madison saw federalism as largely a preventive device, he also gave some support to it for the flexibility and opportunity for experimentation which it would introduce into the system. (He was somewhat cynical about it, however, saying that at least a mistake made in one state would not have to be suffered by all.) Regulatory action at the state level represents an opportunity for citizens to choose 14. James Madison, The Federalist. Nos. 10 and 51. 174 differing levels of government oversight of various activities, but not necessarily more than would be the case if there were only federal regulation. Regulated interests may, in fact, use the differences between the states as a means to ratchet all the states into lower levels of regulation than might otherwise be the case. In the eight months that I observed Michigan's joint Comnittee on Administrative rules there were a number of occasions on which the level of regulation in other states was specifically introduced into comnittee discussion. Sometimes a representative of an industry group raised the question; other times it was a member of the joint committee. Issues ranged from environmental standards to the number of hours truckers could be on the road without resting to the factors taken into account in authorizing utility rates. Industry representatives were most frequently asking for uniformity of practice between the states. Mot only was regulatory certainty important for them, as noted earlier, so was regulatory uniformity. There were two major reasons for this. First, many of the companies or industries were involved in business in a nunber of states. Differing rules between states simply complicate compliance, resulting in higher costs. 15 Businesses located completely within Michigan had a related but different concern. They feared more stringent regulations in Michigan would result in noncompetitive prices for their own products. The continuing slow recovery of Michigan's economy and the general public anti-regulatory mood have caused the committee to regard such 15. No one complained that a Michigan standard was too low and asked for a more stringent standard as a way to achieve uniformity in regulation; this argument always went in the other direction. 175 arguments carefully. Fear of losing Michigan businesses or business for Michigan companies is given greater credence than even two years previously under these circumstances. In the process, it has increased the force of anti- or reduced-regulatory arguments in the JCAR voting decision. Parties to such disputes variously interpret the participation of others and themselves. The Michigan Manufacturers' Association and State Charrber of Council representatives, Department of Labor personnel, and even JCAR menbers, for instance renarked during joint comnittee hearings that one reason for the success of rules promulgated under lg MIOSHA is the existence of the many ad hoc advisory groups. These groups, required by the statute, have equal labor and management representation. a many such groups. labor representative, however, thought there were too So many people were required that it made it difficult to find enough labor people to fill the slots. Moreover, even when on the comnittee, labor representatives could be overshadowed by the "fast-talking, highly educated" management representatives. The solution suggested by this respondent was a reduction in the nunber of groups, allowing labor to educate a few representatives to operate in this arena more effectively. 17 16. Success in this context means lack of conflict at the comnittee level, and relatively high rates of approval. 17. This respondent preferred not to be identified. 176 5.A. Npnegongnric-GigupsrTtae aranple. Much of the interest group literature as well as discussion thus far here has focused on economic interests as prime motivators of participation in regulatory activity. It is not always so, however, and one of the more interesting exanples in recent Michigan history has been the case of rules defining a live birth. Michigan enacted a new public health code in 1978. One of its requirements was a definition of live birth. An emergency rule definition was promulgated that fall pending approval of a permanent rule through the regular rule process. There was no controversy concerning the emergency rule and no opposition expressed at the public hearing conducted by the Department of Public Health prior to transmittal of the rules proposal to the Joint Committee. The Public Health rule was scheduled first on the JCAR agenda on May 29, 1979. The first sign the director had of trouble was the announcement by the comnittee chairperson that the rule was being moved to the end— they had had a nunber of conplaints. The rule was withdrawn. Conplaints had been from opposite sides of an issue that was a recurring source of polarization in Michigan politics. Both Michigan Citizens for Life and the Michigan state National Organization for Women chapter had taken their concerns regarding the rule to the legislators. In each case, it appeared that the groups had learned of the proposed rule only after the public hearing. 177 After the rule was withdrawn, department personnel met a nunber of times with representatives from the interest groups. Five months later, they reached language acceptable to both groups and resubmitted the rule to the Joint Comnittee. In an interesting departure from the comnittee tendency to approve cases on which there was department and interest group concurrence, the comnittee voted twice but was unable to arrive at a concurring majority either to approve or disapprove. It appeared to be a case where at least some legislators were not fully confident that the parties were truly in agreement regarding the proposal. their normal risk averse behavior, they waited. Exhibiting The second presentation of the rule resulted in an inpasse. After still further meetings, and intense lobbying of legislators, the rule wait to the comnittee a third time. Once again, they were unwilling to adept the rule and it was withdrawn. Finally, two years later, with the full knowledge of the interested groups and JCAR members, the rule was adopted after being incorporated in a nine-page set of other public health code definitions. One of the particularly interesting aspects of this case was the initial contact with the comnittee. The groups simply did not know about the department action in time to participate there; thus, the JCAR became their alternate contact point with the rules promulgation process. Having used that access and gained entry to the process, they then became involved in regular negotiations through and with department personnel. Neither of the participating groups had an economic interest motivating their participation. Nonetheless, they followed the issue for months, indeed several years, participating in negotiatory meetings 178 with department personnel, lobbying JCAR menbers, and attending comnittee meetings. 5.5 Interest Groups in Perspective The most successful lobbying is carried out by those who operate in both parts of the rules oversight system— being involved with administrative promulgation as well as legislative review. 'Hie normative issue regarding "the second bite at the apple" raised by Bruff and Gellhom and others is a serious one. Does this system unfairly advantage certain participants over others? The answer depends, in part, on one's perceptions of legislatures, public administration, and the proper place and role of public participation in governmental decision making. Bruff and Gellhorn seem to have taken as their reference point a a relatively value-neutral public administrator acting on the basis of the record developed through public hearing and document submittal. The Michigan legislative review system clearly goes beyond that, incorporating all those elements of private contact and extended side negotiation which were causes of their concern. What is Interesting is the direct participation of public administrators in that process. Whatever the degree to which public administrators may have been able to act independently in the past, the review process has made them even more open to public perusal of and participation in administrative decision making. Interest group participation has frequently caused a department to 179 modify an initial position. Legislators' attitudes toward this outcome is that it is probably appropriate. They look to lobbyists and other private sector participants for information on the inpacts and implications of rule implementation. A former speaker of the state House had called "arguments given by the lobbyists °the means by which 18 we educate ourselves."1 Lobbyists are most likely to achieve their goals if they work both sets of participants. The two parts of the system serve different purposes and only to a limited degree are alternatives for each other. The review comnittee does not write rules. Although there has been controversy over that at times, the fact is that the comnittee operates from the basis of the rule which has been placed before it by the department. Thus administrative actors set the initial content of the agenda. Groups may seek to amend that in the context of joint committee review but, in general, the basic framework is established. To be a part of the construction of that framework could frequently be advantageous to an interest group. At the same time, the legislative review process offers participants both an additional and alternative means of participating in the process. If they have missed the administrative development, as was the case in the live birth definition rule, they can still have substantive effect by entering the process at the JCAR level. In addition, the fact that the JCAR substantively intervenes means that those hearings offer the opportunity to inject into consideration 18. "Bless the Lobbyists, Ryan says," John B. Albright, Tanging state Journal. Deceirber 16, 1982. 180 certain values which the public administrators may not have wished to consider. These are not necessarily matters of right or wrong choices but different ones. The legislative actors, by virtue of different experience and orientation, are sometimes willing to consider factors which the administrators are not. The majority of rules pass, but a significant nunber only after substantial negotiation between interested parties. Successful negotiators usually work with agency people in the development of a rule, and then are involved in contacts with legislators to assure its passage. Some do, indeed, get a second bite at the apple under those circumstances, but the alternative seems to be to let none have any. Worthwhile input would be lost if that were the alternative adopted. Who then is in charge? process? Do the interest groups really control the Is there room for professional autonomy among civil servants within this context? Perhaps the bureaucrats do indeed merely offer up whatever the legislators want. Evidence from this chapter, brief though it is, still finds bureaucrats operating as independent actors and further supports the finding of earlier chapters. Administrators do play a key role and have special agenda advantages in this setting. (We will see more of this in the next chapter in the discussion of the review of existing rules.) For example, the agency needed a live birth definition and it got one, even if legislators were uncomfortable going on the line on the issue. Nevertheless, interest groups and private menbers of the public are inportant participants in the negotiation process and while not sole determiners of either rule content or outcome, are additional sources of constraints upon the bureaucrats. Chapter 6 Legislative Oversight, 1983-1984: Continuity and Change The 1983-84 biennium brought several changes potentially affecting the legislative oversight of administrative rules in the state of Michigan. To begin with, for the first time since systematic legislative review had been instituted, the legislative session started with one party in control of both the legislative and executive branches. Second, there were substantial changes in the membership of the Joint Conmittee on Administrative Rules, first, because of the greatest turnover in its modem history and then, because of further changes resulting from Senate recall elections in the fall of 1983. Finally, the conmittee was given a new mandate— the systematic review of existing rules. These factors allowed preliminary testing of hypotheses addressing the effects of institutional, party, and ideological factors on approval rates. 181 182 Oversight under Same Party Control— Legislature and Governorship With the same party in control oC the legislative and executive branches, one might hypothesize that policy direction would be harmonized and a greater percentage of administrative rules would be approved by the legislature. If this were the case, several conditions would have to be met: 1. substantive gubernatorial authority over administrative agencies, and utilization of that authority; 2. gubernatorial interest in administrative rules; 3. general agreement between governor and legislature regarding the proper approach to and content of administrative rules. On the other hand, inproved approval rates would result only if party identification were more important than institutional perspectives and loyalties in determining orientation toward rules proposals and processes. Put sinply, the question is this: when the Democrats control the legislature, which counts more— that the governor is a Democrat or that the governor is governor? More formally: Hypothesis 5. Approval rates will be higher under same party legislative-executive control than under split party control. The election of Governor James Blanchard in 1983 was a momentous occasion in recent Michigan history. For the first time in twenty years, Democrats controlled both the office of chief executive and both houses of the legislature. Despite the generally good relationships between former Republican governor William Milliken and the legislature, 183 many were hoping for a new day in executive-legislative relations and the realization of Democratic policy goals. There were voices of caution, however. Michigan was still suffering from the general recession and the specific problems associated with the decline of the American automobile industry; the new governor would not be able to completely reform state government and programs into a new Democratic mecca. Concern for the inpact of change on the Michigan business coimunity was an important couponent in the evaluation of any new proposal. During 1983, a total of 152 proposed rules were transmitted to the Joint Committee on Administrative Buies. Table 6.1 shows a result opposite from that predicted. Rather than increasing under same party legislative and gubernatorial control, approval rates actually declined. Prom 1978-1982, under the same (reverse) veto rule but split party control, a total of 73.5% of rules proposals were approved. In 1983, under same party legislative-executive control, the approval rate dropped to 70.7%. The relationship lacks statistical significance, however, at even the 0.10 level. 184 Table 6.1. JCAR Disposition of Rules Transmittals by Party Control (Legislature and Governorship): 1978-1983+ Outcome Party Control (Legislature and Governorship) Split Same 1978-1982 1983+ N N % % Approved Other Column total % 576 208 784 (73.5) (26.5) (82.4) 118 49 167 (70.7) (29.3) Row N % 694 257 •951 (73.0) (27.0) (17.6) (100.0) X2 = ,737 p = .40 Yule's Q « .070 aIncluded in the 1983 cases are 15 cases decided in 1984 before the OCAR's Senate meirbers were changed. How ought this be interpreted? A new governor, particularly when assuming office after a long period of opposite party control of the executive branch m y initially have only limited control of and input into the administrative rule-making apparatus. Michigan, as is true of many American state governments, has a professionalized civil service of many years standing. Although a new governor has the right to name some department heads and a certain number of second level administrators, the overwhelming majority of state workers remain in place regardless of changes of governor. State personnel most directly associated with the development of administrative rules are below those personnel levels immediately affected by a change in governors. On the other hand, a new governor under these circumstances is likely to be advocating a number of new policy goals; to the extent that policy program is enacted, new rules will be necessary. Areas of high gubernatorial interest could reach legislative review within the first twelve months of a new 185 administration but the bulk of the rules reaching the conndttee during 1983 were undoubtedly "in the pipeline" prior to the beginning of the Blanchard administration. Overall, these factors suggest caution in interpreting these results. Despite these considerations, several indicators would favor using this evidence as the basis for rejecting the hypothesis. Prior to being elected governor of Michigan, James Blanchard had served as a Michigan assistant attorney general and twice as a Michigan member of the D.S. House of Representatives. As an assistant attorney general, he had reviewed administrative rules. As a U.S. representative, he testified in favor of their legislative review as well.'*’ The new governor had the knowledge and experience to use the rules promulgation process to advance policy goals should he see it as necessary or advantageous. Although the new governor expressed concerns about administrative 2 rules, especially as they affected the state's business climate, the governor's office actually paid little attention to specific rules. The only rule receiving explicit gubernatorial attention in Blanchard's first year was an emergency rule relating to FCB-contaminated silos. 1. He testified in Congressional hearings that he had advised and represented, among others, the Departments of Agriculture and Commerce and numerous licensing boards in their rulemaking activities. See "Hearings Before the Subcomnittee on Administrative Law and Governmental Relations," Corrmittee on the judiciary, D.S. House of Representatives, Serial No. 30, 1975, p.162-165. 2* See "State of the State Address", Hon. James J. Blanchard, Governor, State of Michigan, January, 1984, pp. 21, 23. 3. Larry Tokarski, State of Michigan, See also "state of Governor, State of Director of Legal Affairs, Office of the Governor, interview, East Lansing, Michigan, January 23, 1984. the State Address", Hon. James J. Blanchard, Michigan, January, 1984, p. 38. 186 While the evidence is at least suggestive that institutional perspective is more inportant than is shared party control in determining the outcome of legislative review of administrative rules, intervening political events made it at least temporarily impossible to expand the data base for further testing of the hypothesis in the Michigan setting. Senate majority change in 1984 resulted in split party control within the legislature, ending same party legislative-executive control. 6.2 The Joint Committee on Administrative Rules: Party and Ideology The Joint Comnittee on Administrative Buies underwent important memberships with the beginning of the 1983-84 legislature. 1981-82 JCAR members returned to the legislature. JCAR to head the Bouse appropriations committee. Only four One of those left the The other three members returned, maintaining an adjusted career investment level similar to that of previous years. Nonetheless, with seven new members, the committee experienced its greatest turnover since the beginning of systematic review. One of the members not returning to the legislature was Representative Thomas Anderson, a 9-terra member of the committee and for the previous four terms either chairperson or alternate chairperson. 4 4. The head of the delegation from the chamber not in charge during a given biennium was known as the alternate chairperson. 187 Appointments to the JCAR in 1983 generally were in keeping with a more conservative regulatory climate. The new head of the comnittee, a Democratic representative who had served on the comnittee the two previous terms, is generally recognized and freely characterized himself as one of the more conservative legislators of his party. The Republican members from the House, although not the most conservative of g their colleagues were well to the right of the House majority. One of the other House members could be described as a moderate Democrat, and one as liberal. 7 The Senate membership was even more conservative. At least four of the five senators would be regarded as conservative representatives of their respective parties. The only senator returning to the comnittee, a Democrat, was one of the most anti-regulatory members of that body, at Q least as it applied to the Department of Natural Resources. The two Republican senators were recognized as among the most conservative of 5. Rep. Michael Griffin, R-50, Jackson, interview, Lansing, Michigan, January 4, 1984. 6. Reps, Ernest Nash, R-56, Dimondale, representing a rural and suburban district adjacent to Lansing, the state capital; and Charles Mueller, R-83, Linden, from a largely rural district adjacent to Flint. 7. Reps. Dennis M. Dutko, D-25, warren, in urban Macomb County, just north of Detroit; and Virgil Smith, Jr., D-10, representing an inner-city area of Detroit. 8. Sen. Joseph Mack, D-38, Ironwood, in the upper peninsula, an area with heavy forestry and mining interests. 188 their colleagues. The other two senators, Democrats, would probably best be described as moderate and conservative respectively. g The appointment of the Republican senators was particularly interesting. At the beginning of the 1983-84 legislative session, there was a stand-off for leadership of the Senate Republican caucus. It was resolved when one of the senators vying for the position went to the other contender, asking what he would take in place of the leadership of the caucus. The answer was simple: appointments to the Joint Comnittee on Administrative Rules for both himself and another conservative senator. As the senator explained in an interview: "I wasn't interested in who got the biggest sofa; I was interested in policy. Some people didn't like that. They like, to spread the conservatives out so they can't change anything. The deal was made but greater change was yet to come.'*"** At the beginning of the 1983 legislative session, the state senate had a 20-18 Democratic majority but in the fall of 1983 two Democratic senators were recalled and replaced in special elections by Republicans. Thus, the Republicans assumed control of the Senate. For the first time in Michigan senatorial history, there was chamber wide mid-session comnittee realignment. Democratic members were removed from committees 9. The two Republicans were Sens. Ed Fredricks, R-23, west Olive, representing a two-county rural area on Lake Michigan; and Alan Cropsey, R-30, Dewitt, representing three rural counties north and east of Lansing, the state capital. The Democrats were Sens. Jerome T. Hart, coirmittee vice-chair, D-14, Saginaw, representing an industrial city and its surrounding county; and Michael O'Brien, D-5, Detroit, a middle to upper-middle class area of northwest Detroit. 10. Sen. Ed Fredricks, interview, Lansing, Michigan, March 8, 1984. 11. Hie senator appointed with Senator Fredricks was Senator Alan Cropsey. 189 and additional Republicans appointed with the net result being Republican chairs and majorities for all Senate conriittees. The Senate delegation to the Joint Coirmittee on Administrative Rules underwent the same change, with three Republicans and two Democrats now being appointed, reversing the party balance. Senate alternate chair of the comnittee, a Democrat, left. The 1983 According to several sources, this was because he would be bumped from that position by the new Republican majority and thus would not chair the JCAR when that position rotated to the Senate in 1985. In addition, since the senior Democratic Senate member of the comnittee lost his position as head of the comnittee on natural resources during the reshuffle, he exerted chamber seniority, gave up his seat on the JCAR, and took a seat on the appropriations comnittee. 12 That left room for one new Democratic senator as well as the new Republican. The new Democrat appointee, one of the most conservative senators, had been forced off appropriations in the reshuffle. 13 The new Republican appointee, a decidely conservative former state representative, was one of those elected in the special elections following the recalls The potential effects of these changes were multiple. It has already been demonstrated that outcomes are significantly related to changes in the veto rule. Prior testing of that relationship, however, 12. This was Senator Joseph Mack, then in his fourth term on the JCAR. 13. Sen. Gilbert DiNello, D-26, East Detroit, representing Mt. Clemens and St. Claire Shores as well, a mixed district with working class, lake/resort, and middle to upper-middle class areas. 14. Sen. Kirby Holmes, R-9, Utica, representing Sterling Heights and northern Macomb county, a rural and industrial area north of Detroit. He had served earlier in the state House of Representatives.) 190 was under conditions of same party legislative control. Now, however, for the first time since the beginning in 1978 of the requirement of prior legislative approval, agencies were dealing with a comnittee with split majorities: the Senate majority was Republican; the House majority, Democratic. If party is an important conponent of JCAR meitber voting decisions and the parties differ significantly on issues upon which they are voting, approval rates should decline under conditions of split legislative control. Moreover, one would expect to more frequently find a lack of concurring majorities between the two delegations and the incidence of iirpasBe should increase. On the other hand, if ideology is more important, it is more difficult to predict the likely effect of these comnittee changes. The comnittee delegations from both chairbers are relatively conservative, perhaps reducing the partisan differences and the likelihood of impasse, but not conpletely. There could still be disagreement between the delegations as to where to draw the line on regulatory activity. Hypothesis 6. Approval rates will be lower and impasse higher under split party legislative control than under same party control. Table 6.2 shows a decline in overall approval rates, as predicted, but the difference of 1.8% is "significant" at only the .50 level. Looking at impasses alone, the difference is tantalizing but not necessarily significant. The change from no impasses in 1983 to three in 1984 is in the right direction but with so few cases is inconclusive. The evidence at this level fails to support the hypothesis that split party legislative control will result in significantly lower rates of approval and higher inpasse rates. 191 Table 6.2. JCAR Disposition of Transmittals by Legislative Party Control and Incidence of Impasse, 1983-1984. Transmittal Party Control of Michigan Senate and House Same (1983+) Split (1984-) N N % % Approved Other Column Total % • Impasse 118 49 167 (70.7) (29.3) 84 38 122 (55.8) 0 ( 0.0) (68.9) (31.2) Row N % 202 87 289 (69.9) (30.1) (44.2) 3 (100.0) ( 2.5) X2 o 0.60 P = .50 YUle's Q « .043 alnclndes all 1983 cases plus 15 from 1984 decided prior to the seating of the new Senate mentoership. uata for 1984 are those cases decided after 2/21/84, the date the new Senate delegation was seated. In 1983+, 44 of these rules were withdrawals; in 1984, 35 were with­ drawals. There is one bit of evidence which may clarify the picture somewhat. At its August, 1984, meeting, the JCAR processed twenty-one transmittals rescinding 195 existing Public Health rules. 15 Subtracting the ensuing twenty-one approvals from the other 1984 approvals (84) results in an approval rate of only 62.4%.. That approval rate set 15. This is a good example of the problem of unit of analysis. When introduced by JCAR Chairperson Griffin at the meeting, they were referred to as 195 rules and are again so identified in a letter to the governor. (See Appendix B— letter to the Honorable James J. Blanchard, from Michael J. Griffin, chairman, Joint Comnittee on Administrative Rules, State of Michigan, January 10, 1985, p. 1.) Individual transmittals may contain any number of "rules," Indeed, I never encountered a formal definition of the minimum-unit constituting a "rule"; one which would probably serve for most purposes is "any separately numbered subsection of the Administrative Code." In this work, the individual transmittal package has been used as the unit of analysis, regardless of the number of "rules" it might contain. 192 against the 70.7% approval in 1983 results in a X2 of 2.064, with a probability of less than .20 (Yule's Q -.185). This suggests that in cases which lack prior consensus there m y be a party effect but, if so, it is apparently quite weak. Overall, these data do not support a solely party-based interpretation of the outcomes of rule review. To the degree that party influence is found, it appears to be more a function of underlying ideology. Yet another 1984 change in conmittee membership offered a further example of the multiple influences on the conmittee. After several unsuccessful attempts to pass an administrative rule dealing with auto exhaust emissions, Michigan was in danger of losing its federal highway monies. Conservative Democratic Senator DiNello, appointed to the conmittee in January of 1984, following the recalls, had been less than diligent in meeting conmittee responsibilities and, further, was against the rule. In an unusual move mixing party, ideology, institutional, and budgetary factors, DiNello was removed from the conmittee. His seat was taken by Senate Majority Leader William Faust. The auto emissions rules passed and Michigan's federal funds were protected. Thus, even as the conmittee sometimes served as a check on other legislative processes, the legislature could, and indeed did, act as a check on the committee. If the JCAn itself acted in a way clearly outside of legislative intent or necessity, changes could and would be made. 193 6.3 A New Responsibility— Reviewing Existing Rules Ideology and a generally more conservative public opinion regarding regulation resulted in a major expansion in 1983 of JCAR function and responsibility. Reflecting the general concern about Michigan's continuing recession, the legislature directed the JCAR to begin systematic review of existing rules of the four state departments with the greatest direct regulatory inpact on Michigan businesses and "business climate": the departments of Labor, Commerce, Natural Resources, and Public Health. In March of 1983, the JCAR constituted itself as a subcommittee for the purposes of conducting that review. The purpose was to identify those rules which were "obsolete, unnecessary, duplicative or unduly burdensome to business and industry in Michigan."16 The joint committee instituted a three phase review process which w eis very similar to that used in the promulgation and review of proposed rules. Public hearings in six different cities were held over a period of five months. Those testifying identified rules which they thought ought to be rescinded and others for which they requested modification, sometimes with specific recommendations as to the form of modification desired. The four departments were requested to do a similar internal 16. "Administrative Rules Review of the Departments of Commerce, Labor, Natural Resources and public Health: Preliminary Report," Joint Comnittee on Administrative Rules, state of Michigan, January, 1984, p.i. 194 review for rules which could be rescinded or modified. The next phase called for the subcoirmittee staff (the JCAR assistant special counsel) to prepare a compilation and analysis of the conplaints and reccmmendations which had been node at the public hearings. This process identified a number of issues, some of a very specific nature and others of more general scope. The departments were requested to prepare a written response to each of the issues raised and to present the response to the JCAR. In January, 1984, the comnittee began hearing the reports from the departments. The dynamics of the subcommittee meetings were particularly interesting, Hiis process was operating in addition to that of regular review of proposed rules. During the first five months of 1984, the comnittee was meeting twice a week most weeks, although not all members were always in attendance at both meetings. It was unusual, in fact, to have more than four members of the committee at a subcommittee meeting at any given time (although the four might vary through the course of a meeting) j frequently only the chair and vice-chair of the conmittee would be present for the full duration of a subcommittee meeting. Hie subcoirmittee took no votes nor any other kind of official action on the matters it was hearing, but the lines between the subcommittee and the JCAR were very short. Those testifying seemed well aware that they had to respond in a fashion cognizant of the full membership and authority of the committee. Department responses were of several kinds. In some cases, they agreed with the industry complaint that a rule should be rescinded or 195 modified and in those cases, most gave the subcommittee a timetable foe the appropriate action. 17 This was usually presented in terms of when the comnittee could expect to have before it a proposed rule change dealing with the issue. Agencies which failed to volunteer such information were inevitably questioned as to when they expected to act. There were other cases in which the department did not expect to act in conformity with the issue raised in the hearings. occur for several reasons. This could The agency might argue that it disagreed with the complaint, that the rule was in conformity with the statute, was being fairly enforced, etc. On the other hand, it might be in agreement on the issue, but argue the lack of personnel to do whatever was requested, or lack of statutory authority. Cases where the agency agreed with the complaint but lacked statutory authority to act were referred to the governor's office. 18 Table 6.3 summarizes the conplaints received and their original treatment by the agencies. 17. A rescission of a rule does not differ in treatment under the APA from any other proposed change to the Administrative Code, simple agreement was not a sufficient basis for removing, the rule from the code although it might not be enforced in the interim. 18. Gov. Blanchard appointed a special committee to review these conplaints in particular. The committee will eventually recommend corrective legislation as they see appropriate. 196 Table 6.3. Number of Targetted Existing Rules, by Department and Identifying source {Internal Review or Public Conplaint) by Reported Department Response as of July 1984. Comnerce IR PC Requires leg. 6 0 Will rescind 0 80 Will amend 1 147 Already amended 0 99 TO conns,bds,etc 0 0 Further review 0 0 Resolved 2 0 Not resolved 0 7 Department Natural Public Resrcs. Labor Health IR PC IR PC IR PC 0 0 0 4 1 3 195 0 0 0 0 0 55 25 115 11 304 3 0 0 0 0 0 0 0 0 17 0 0 3 0 13 0 14 0 14 0 13 0 4 0 5 0 24 0 36 0 0 Source Totals IR PC 14 0 0 275 40 621 0 99 20 0 41 0 0 24 0 67 Row Total 14 275 661 99 20 41 24 67 * 1201 1201 Column total 115 82 55 82 499 26 995 206 326 16 Dept, total 525 197 137 342 Dept/source % 16.4 28,.5 11 .4 43 .7 82.8 17.2 it Committee staff classified an additional 28 conplaints as miscellan­ eous. Eight related to other departments? twenty, to "use of guide­ lines, directives, policies, and inter-office memorandums in lieu of administrative rules; staff interpretations of statutes/rules/guide­ lines; the administrative rules process and the organization of the Michigan Administrative Code." See "Administrative Rules Review of the Departments of Commerce, Labor, Natural Resources, and Public Health: Preliminary Report," Joint Comnittee on Administrative Rules, State of Michigan, January, 1984, p. iii. Source: Extracted from "Administrative Rules Review of the Departments of Connerce, Labor, Natural Resources and Public Health: Interim Status Report," Joint Comnittee on Administrative Rules, State of Michigan, JUly, 1984, pp.3, 10-12. 197 All cules ace not equal and certainly not those in Table 6.3. it shows that 82.8% of the cules narked for attention were identified by the departments themselves and only 17.2% were identified by the regulated businesses and industries. Those figures are misleading. In many cases, rules identified by the departments were rules not being enforced anyway. Those identified by the public participants, however, were uniformly rules currently being enforced; as such, they were sources of friction in the system. Interest groups made a substantial investment in the review of existing rules. They organized testimony at the department hearings throughout the state, kept merbers informed through a variety of newsletters and personal contacts, had industry groups which screened and reduced conplaints to a core minimum, and followed up with careful monitoring at the JCAR level, providing additional testimony as necessary. As indicated above, departments varied widely in their responses to public instigated conplaints and there were a series of actors through which some changes would have to be processed, even after department personnel and industry representatives had come to agreement on a response. Testimony at one subconxoittee meeting demonstrated one public participant's concerns about the problem of getting the necessary responses from these multiple actors, ttie director of industrial relations for the Michigan Manufacturers Association (MMA) told the joint committee: "We feel it is incumbent upon this committee to convey to the department and bureau heads our sense of urgency about this 198 [requested rules changes.] It takes so long, and rules are just avoided at all costs. We have to get the S. Martin Taylors and the Bernie Lennons, people at that level, involved. I ’m not trying to speak for the bureau people, but this is critical, . . . For MESC, none of the four commissioners were here today, nor any of the Board of Review. Any of the agreements we reach can be nixed by the autonomous commissions. It really rests with Martin Taylor— if he wants it, it will happen." The head of the joint conmittee responded that all of the issues would be "nailed down, one way or another, if not resolved before by the parties." 22 Pressed, he acknowledged: "We won't get them all; if we can bat .333 . . . [we will be doing well]." 23 The MMA representative replied they knew that but hoped for action on at least the items on which there was department and industry agreement. At its August 14, 1984, summer recess meeting, the Joint Committee voted approval of twenty-one Public Health rules proposals, rescinding 195 obsolete rules identified through the department's internal review. (See Table 6.3.) Responding to a question from the conmittee alternate chairperson, the department representative confessed his personnel had not even known they had the right under the rules to inspect barber shops, Other rules rescinded were equally obscure or obsolete. Nonetheless, the committee coranended the department for its cooperation 19. Respectively, the director and deputy director of the Michigan Department of labor. 20. MESC— the Michigan Employment Security Comnission, one of the independent commissions. 21. David Zurvalec, Director of industrial Relations, Michigan Manufacturers Association, testimony at meeting of subcommittee of Joint Comnittee on Administrative Rules, Lansing, Michigan, April 30, 1984. 22. Rep. Michael J. Griffin, meeting of Joint Conmittee on Administrative Rules subcoirmittee, Lansing, Michigan, April 30, 1984. 23. Ibid. 199 and expressed the hope that the others would take note. first formal action resulting from the new process. This was the While rescinding obsolete rules may seen at first glance purely symbolic it also has substantive effect. others. By removing such rules it increases the iirpact of Nonetheless, these were the easy rules to deal with. Others would prove more difficult. Numerous transmittals on through 1984 involved rule rescissions but the joint comnittee and apparently some of the public as well aided the year with a strong sense of frustration regarding the process. In a January 10, 1985, letter requesting the governor's assistance, JCAR chairperson Michael Griffin noted that although 65% of the business identified concerns had been positively responded to, the process had, OA "in other respects, . . . been utterly frustrating". He cited several ways in which the departments were being less than fully cooperative: burying small conplaints in large packages which would take much longer to handle, not meeting scheduled submission dates, delays in the meetings of statutory advisory groups, and gaps in internal department connunications. An attached letter from the director of regulatory affairs for the Michigan Manufacturers Association was even stronger. It focused particularly on the failures of the Department of Natural Resources, noting among other conplaints the DNR's penchant for dismissing a complaint in its sunsiary report to the JCAR simply by 24. For the letter, see Appendix C. 25. Included in Appendix C. 25 200 saying it "disagreed" and indicating no intention of any further consideration or action. The new JCAR responsibility was no more easily discharged than the old. Subcommittee work required additional investments by comnittee members and the institution alike. Members had to spend more time reviewing conmittee materials and additional supplies resulted in a $6,500 deficit in the first six months of the new review. The House and Senate leadership agreed to split the cost and to do the same thing with the next year's anticipated deficit. The comnittee notified the Department of Social Services (DSS) it was next and began to receive related materials from that department. The Department of Education was also informed that it was slated for 1985. Although those reviews are beyond the scope of this chapter and even of the dissertation, a few comnents are appropriate. Why did the Department of Social Services (DSS) finally become the focus of action? Rosenthal identified "legislative climate" as one of the conditions for substantive legislative oversight. Here there appears to be a favorable intersection of public opinion and committee membership. The heavily conservative comnittee would be taking on a currently popular target, fueled at least partially by several anti-tax ballot attempts and the recent recall elections. There is a priori reason to expect the dynamics and perhaps outcome of review of Social* Service and Education rules to differ from that of the Conmerce, Labor, DNR, and Public Health reviews. There was 26. See "Final Report: Administrative Rules Review of the Departments of Comnerce, Labor, Natural Resources and Public Health", Joint Conmittee on Administrative Rules, State of Michigan, January, 1985. 201 testimony at one subcommittee meeting that some businesses were afraid to appeal DNR decisions out of fear of reprisal; that fear is undoubtedly even greater among DSS clientele. The process is unlikely to bring forth conplaints from individual Social Services clients r even if notices were to be included with monthly checks. One JCAR menber felt the problem of review of DSS or the Department of Education was that only the employees could really tell what the problems were, and they probably would not. 27 The attitude of most current JCAR members is less favorable to DSS clients than it is to interests regulated by, for instance, the Department of commerce. Opinion is not monolithic, however. One of the more liberal members thought it would be interesting to see if welfare regulations could be as nonburdensome as they were trying to make business regulation, in contrast, a more conservative member would like to see the current system completely disbanded, substituting block grants to the counties who would then devise their own social welfare programs. Whether comnittee members will use the review when it reaches them as an opportunity to effect policy changes remains to be seen. It should be noted before leaving this chapter that comnittee staff changes in 1984 and 1985 had a bearing on several of these issues. The junior staff counsel had been assigned the responsibility of managing the review of existing rules. In 1984, the senior special counsel, who had worked eleven years for the comnittee was fired. One criticism of his work was that he had relied too heavily on the agencies. This may have signalled a conmittee desire to take a more independent, more 27. Sen. Ed Fredricks, interview, Lansing, Michigan, March 8, 1984. 202 activist, perhaps even more adversarial stance relative to the departments. If so, that may account for some of the frustrations expressed in the conmittee chairperson's letter to the governor. Unfortunately, from the comnittee's perspective, it soon lost its now new senior special counsel and thus began the substantive review of the new departments with completely new professional staff. The process was slowed tremendously. 6.4 Assessing the Changesr-Sunmarv Change was the most immediately apparent feature of government in Michigan in 1983; it was even more so for the Joint Committee on Administrative Buies. For the first time since the beginning of systematic legislative review of administrative rules in Michigan, the governor and the legislative majority were of the sang party. The comnittee had new leadership and seven new menbers; the legislature as a whole, and the conmittee in particular, were generally recognized as more conservative than had been the recent case. Then 1983 recall elections further scranbled the conmittee membership and, indeed, resulted in split party control within the JCAR and the end of shared party control of the legislative and executive branches. Of what consequence have these factors been? The 1983-84 Joint Conmittee members felt they had been given a mandate to even more carefully scrutinize administrative rules than had their colleagues in the past. They were also directed to begin a 203 comprehensive review of existing rules, a responsibility given to none of their predecessors. At the same time, with a governor of the same party as the legislative majority in 1983 there was the possibility of increased accord between the two branches and improved approval rates of newly proposed administrative rules. The data show continuing decline in approval of administrative rules proposals throughout this time. Front a mean period high of 87.5% under split legislative-executive control and negative veto in 1972-1977 to the next high mean of 73.5% approval under reverse veto from 1978-1982, approval declined to 70.7% in 1983 and again, to 68.9% or even 62.4% in 1984. Ideology seems to be interacting with the veto decision rule to make it increasingly difficult for Michigan administrators to gain legislative approval of the rules to implement statutes. Same observers are concerned this situation will result in significant adjustments in the system, either through greater reliance on administrative alternatives (such as rule-making by adjudication, or greater use of administrative guidelines) or by increasing the incentive for opponents of the veto to challenge it in court. Comnittee members recognize the need for some constraint in their activities, and a proper balance of oversight and administrative authority. Yet they also joke in committee hearings about the number of rules they have or have not approved at a given meeting, about keeping up their average. The time period of shared party control of the legislative and executive branches was so brief one cannot draw substantial conclusions about the impact of such control on rule outcomes. Most administrators drafting rules were persons hired into the civil service well before 204 Gov. Blanchard arrived on the seme. The dynamics of the Michigan political scene have yet to provide us the opportunity for a good test of competing institutional and party explanations of review outcomes. The dynamics of interest group influence were especially apparent during this time period in the review of existing rules. Substantial group resources were invested in the process and the JCAR used for leverage in obtaining desired changes. With Michigan rules approval down to 68.9%, it is even more difficult to maintain either the "runaway bureaucracy" or the "legislative capture" argument. It is easy to feel some sympathy for the apparently "constrained" bureaucrats who invest considerable time and professional commitment in drafting and shepherding proposals through the system only to lose them at the last step in the process. Legislators apparently continue to find the review process worthwhile. The conmittee has continued to attract senior members from the Bouse, and has attracted, even for it, an unusually conservative contingent from the Senate. The comnittee was provided full financial support for pursuing the review of existing rules of the first four departments. Rates of approvals in 1984 may prove to have been a reflection of a period of transition while administrators began making adjustments to a more conservative environment. They may yet have the opportunity to demonstrate bureaucratic learning in the terms described in Chapter 4. If the approval rate continues to drop, however, there would be cause for substantial concern about the health and benefits of the entire system. If major segments of the administrative apparatus eventually routinely get less than 60% of their rule proposals approved 205 they m y well seek more cost effective means of providing the framework for program implementation. Chapter 7 SUnmary, Conclusions, and Suggestions for Further Research Three competing theories of legislative-bureaucratic relations within the legislative oversight context were posed in the first chapter— runaway, captured, and constrained. Chapter 2 presented a series of hypotheses designed to evaluate those three theories within the context of an exceptionally strong test case— Michigan, the state practicing the most extensive rule review current in the U.S. . Analysis incorporated a rational actor perspective, but was not based on a strict rational choice model because it was already clear there were noneconomic values operative in the system and uncertainty was a major element in the decision environment. The first two theories were from the existing literature, in the older one, the bureaucracy was assumed to be out of control— "runaway." Legislative oversight was seen as irrportant to reassert control and make the bureaucracy more responsive to the legislature, the people's representatives. Unfortunately, according to this model, there is little institutional or personal incentive for practicing the needed oversight. The newer model assumed, on the contrary, that a highly effective but rarely noticed system of legislative oversight is already in place 206 207 and that bureaucrats act as they do In order to provide benefits desired by the legislators. The bureaucrats have been "captured". The standing coniDitteesr through which oversight is practiced, wield program approval and appropriations powers sufficient to bring balky agencies back into line in the rare event one should stray from legislator preferred paths. Legislator incentives for oversight activity are those resulting from the benefits associated with membership on the standing committees. The above theories present a major problem in testing for they predict very similar results. In the first case, bureaucrats are likely to get what they want because the legislature lacks sufficient incentive to exercise control. In the second case, bureaucrats are likely to provide what legislators want as a condition for their agencies' and their own survival. In either case, bureaucratic proposals should enjoy legislative approval nearing 100%. While this theoretical debate had been going on in the academic world something very interesting was happening in the real world of state government. There was a substantial increase in state adoption of overt oversight mechanisms, neither of the existing models provided an explanation for this phenomenon, in addition, 73% of all states practicing a form of legislative oversight utilized a system which diverged from the congressional model in the location of the oversight responbibility. The operating assumption on which state action was based appeared to be one that said that more overt oversight was needed (contradicting the "captured" model) and could be efficacious (contradicting the "runaway" model). That assumption became the basis of the third model, which would be labelled the "constrained" bureaucrat 208 model. Research was based on data concerning patterns of participation in, influence on, and outcomes of legislative oversight of administrative rules in the state of Michigan, It assumed that rational actors, whether viewed as institutions or individuals, would not make substantial investments in an enterprise which yielded few benefits. It examined Michigan's rules oversight system from the perspective of the legislative, executive, and non-governmental actors through a thirteen year period during which there were substantial changes in decision rules, partisan institutional relationships, and public opinion. A nurrber of the hypotheses posed in Chapter 2 have received substantial support, but not all. Ibis chapter summarizes those findings and discusses their consequences for the larger questions. 7.1 The Findings in Review Chapter 3 and and parts of Chapter 6 were devoted to the legislative branch. They examined several issues and presented multiple measures addressed toward several hypotheses: 1) that there was substantial visible, direct legislative oversight of administrative rules and that it was centered in the Joint Coinnittee on Administrative Rules; 2) that institutions and individuals would make substantial investments in the review process; 3} that more stringent oversight mechanisms would result in lower rates of approval; 4) that individual JCAR member votes would vary in accord with conformity to authorizing 209 statute, personal voting history on the authorizing statute , regulatory ideology, and political party; 5) that approval rates would be higher under same party legislative-executive control than under split control; and 6) that approval rates would be lower and iirpasse higher under split party legislative control than under same party control. The first substantial findings of this research were the degree to which Michigan's legislative oversight of the promulgation of administrative rules is a separate and visible process and the centrality of the Joint Committee on Administrative Rules (JCAR) within that process. All changes to the state's administrative code— rescissions, amendments, additions of new sections, or conplete recodification— are formally transmitted to the JCAR, placed on a published agenda, and required to undergo a hearing before the conmittee in an open meeting. This is in addition to prior requirements of agency publication of and hearings on proposed rules. Hie joint conmittee handled over 1,600 transmittals in the first eleven years studied; in only three cases did the legislature fail to uphold the committee's decision. This pattern continued in the 1983-1984 data presented in Chapter 6; during this latter period the conmittee was never overturned. The obvious next question was whether this was merely pro forma review or a substantive exercise of legislative authority and decision making. The conmittee was not merely rubber-stanping the bureaucrats' proposals. Approval rates at the beginning of the study period, 1972, were 94.4%; by 1984, they had declined to 68.9%. This decline in approvals was associated with the adoption of increasingly stringent oversight rules. Three aspects of oversight stringency were considered: 210 negative versus reverse veto, simple versus concurring majorities, and shorter versus longer allowed review periods. On all three measures, the more stringent rule was positively associated with significantly lower rates of approval. Chapter 3 also documented a substantial pattern of increasing institutional investments in the oversight process. There have been direct and increasing staff support to the Joint Committee on Administrative Rules, expansion of the responsibilities and workload of the conmittee, and increases in the seniority and number of members serving on the conmittee. The JCAR started (in 1947) with only the staff support of individual member legislators, then had professional support from the Legislative Service Bureau, and finally, by the 1970s, had its own professional and clerical staff. It moved from review of proposed rules on a complaint basis (prior to 1972) to systematic review with negative decisions requiring legislative confirmation (1972-1977), to systematic review with full decision authority vested in the committee (1978-1984). In addition, in 1983, the committee's mandate was further expanded with directions to begin the systematic review of the existing rules of all agencies. From the beginning of systematic review in 1972, membership on the conmittee has shown increasing and high seniority, and membership has been increased to five from each chamber. In short, the Michigan legislature provided the institutional investments and structural incentives necessary to give substance to its strict oversight mechanisms. The effectiveness of these institutional investments was reflected in individual legislator responses. Not only were senior members of the legislature assigned to the committee; they chose to stay there. Rate 211 of retucn of corrmittee members and patterns of movement out of the conmittee showed the JCAR to be an attractive, inportant conmittee from the individual perspective. Only one returning JCAR legislator in the entire time period left the conmittee without assuming a major leadership position or an appropriations conmittee assignment. The ways in which the oversight process operating through the JCAR provides institutional and individual benefits is conplex. A first problem is that of training new JCAR meirbers into their new role. They no longer have the latitude of the legislative "markup" sessions; indeed, they lack any formal power to amend. Informally, however, the threat of a negative corrmittee action gives JCAR members' questions and suggestions a power not formalized in the statute. Learning to function in this new environment is sometimes frustrating for both legislators and the staff who advise them. Four relatively simple models of individual JCAR menber voting behavior were posited. The legal model was easily rejected. Rules proposals were required to have been certified by the attorney general's office before being transmitted to the JCAR. On rare occasions the JCAR counsel differed from the attorney general's office regarding the sufficiency of statutory authority but the incidence of these cases was far below the rate of withdrawals and disapprovals. Conformity to statute was necessary for rule approval but not sufficient, or virtually all rules would have been approved. The legal model failed to explain individual votes. The personal history model, hypothesizing that the menber's vote on the authorizing statute will predict vote on inplementing rules, was rejected more for lack of relevance than for contrary evidence. The 212 problem here was that most rules proposals are based on old statutes on which current legislators never voted. On the other hand, interview data suggested the personal history question has been an issue for some time within the legislature, despite the limited number of cases involved. In the case where they may have voted on the statute, JCAR members expressed a conraitment to acting in accord with "the will of the legislature." At least a few current and former menbers confessed, however, that they would probably be more stringent in their review of rules which represented a policy stance with which they disagreed. Unfortunately, testing of the regulatory stance and partisanship hypotheses was hanpered by the fact that individual voting records are continuously purged; individual records are maintained from only the most recent twenty-four months. From interview data, however, it was clear that the corrmittee has attracted a disproportionate number of conservative legislators, particularly in recent years. A seat on the conmittee may well be an effective way of dampening the thrust of a generally more liberal legislature. At the same time, there may be institutional benefit in making this conmittee more conservative than the legislature's mean. Some respondents suggested the conmittee serves as a check on the legislative process as well as the administrative process. There are constraints, however. If JCAR members are overly zealous, systematically and doctrinal rely rejecting administrative proposals, the executive branch may adept other means of fulfilling its regulatory responsibilities. This hypothesis requires further study before definitive statements can be made about these relationships. At the individual level, however, party influence was difficult to separate from ideology and district-based factors, a not uncommon problem in this 213 type of research, further confounding testing of these hypotheses. The original statement of the hypothesis may well represent the real world, with a coirbination of these several factors rather than any single one of them being determinative. Examination of the influence of party on an institutional basis was inconclusive. The year 1983 is the only time in the entire thirteen year period in which there was same party control of the legislative and executive branches. Conparing that year to earlier years under reverse veto (1978-1982) actually showed a slight decline in approvals, from the earlier average of 73.5% of rules approved to 70.7% in 1983, a difference "significant" at only the .40 level. Partially because this was the first year of government under new party control of the executive branch, however, it was argued that this was not yet a sufficient basis for rejecting the hypothesis that approval rates would be higher under same party legislative-executive control than under split control of these branches. A closely related hypothesis was that split party control within the legislative branch itself would lower JCAR approvals and increase the incidence of iitpasse. Split party control within the legislature occurred in 1984 and affected the conmittee starting in mid-February. Compared to the previous year, the result was a very small decline in approvals, from 70.7% to 68.9%, significant at only .50. Even adjusting for the rescissions processed by the committee, the adjusted approval rate was 62.4%, with the difference having only a .20 level of significance. Obviously the first half of the hypothesis fails on the basis of these data. The other half, however, was tantilizing. There were no inpasses in 1983, but three in 1984, too small a number for 214 meaningful statistical testing. As split pacty legislative control continues, a clear picture may emerge, but for the time being, the results are inconclusive. It may be that weak party effects exist, interacting with ideology. Legislative norms of deferral to colleagues' district-based concerns seem to extend to the JCAR, providing at least one potential source of readily identifiable benefit tb serving on the conmittee. Individual legislators were allowed by the conmittee to shape rules in policy areas of specific concern to their constituencies. Unless there were substantial department need to pass a given rule, the conmittee norm appears to be to require the departments to meet the concerns of any menber of the conmittee. To suppose this is the entire story however would be to greatly oversimplify the real situtation. JCAR members reported rarely being contacted by district constituents regarding pending administrative rules; contacts were much more likely from what I called "interest" constituents. Individual members of the committee are likely to specialize in certain policy areas, often but not necessarily and certainly not completely reflecting district based interests. To the degree that this is the case, they would presumably also be the target of specialized interest group attention. A brief examination of canpaign contributions appeared not to support the idea of specialized interest group contact, at least not on this basis. Legislators serving on the conmittee saw it as highly powerful, a mini-legislature which acted on everything. The breadth of action was reflected in canpaign contributions, which came from virtually every major interest group in the state of Michigan. Interest 215 groups with high regulatory profiles were likely to contribute to several members of the committee and not necessarily those of only one party or a single regulatory stance. The pattern here is more in keeping with one of buying access than it is of concentrated attention to the election situation of a single candidate. Given the generally high seniority of committee members, perhaps this is not surprising. If they return to the legislature (and they frequently are from relatively safe districts), they are also highly likely to return to the Joint Committee. If not, the evidence is clear that they will go to another position of substantial influence. Thus, individual JCAR menbers attract contributions from a wide range of contributors often encompassing conflicting interests. One key way in which the JCAR operates to increase benefits and reduce risks for institution and individuals alike is through its norm of coerced consensus. During recent years, if testimony at the JCAR hearing reveals unresolved disputes between department and regulated public the relevant parties will be instructed to "get their act together" and return with an acceptable alternative. Committee members justify this on the basis of their statutory authorization to review rules for "expediency". its effect is to force participants to reveal their true preferences, and to reach self-determined compromise on those preferences. From the legislator's perspective, this operates as a highly effective strategy for reducing uncertainty— they need only endorse self-generated compromise. It would be difficult to construct a more successful risk averse strategy. JCAR members fit fairly well Rosenthal's portrait of oversight specialists. They enunciated goals of making government work and liking 216 to learn the details of programs; they also liked getting that last bit of influence over policy. They tended to be among the workhorses of their respective chambers. House members were more clearly static in their political ambition than were Senate members, but in both cases, the subsequent careers of all members who have served on the JCAR at any time since 1972 show them overall to be more statically than progressively ambitious. The JCAR gives them a place to exercise substantial influence within their chosen situations. Chapter 4 shifted to discussion of executive branch actors. Departments varied considerably in their success with the oversight process— approval rates varied by as much as twenty-five percentage points. They were alike, however, in that all departments saw their approval rates drop with the imposition of the reverse veto, the requirement that rules have the prior approval of the legislature. The departments averaging the greatest number of proposals over the years, however, dropped the least. Experience seems to have at least some value in this environment, so there is at least some support for a notion of bureaucratic learning in this context. It was interesting in this regard to see what happened to rules which were originally rejected, whether withdrawn or disapproved. Under negative veto (1972-1977), 83.% of such rules (n = 79) were eventually passed in some form; only 12.7% were never resubmitted. Under reverse veto (1978-1982), 68.7% were eventually adopted, with 24.9% never resubmitted. Two conclusions were drawn from this. First, negotiatory skill is an important part of the Michigan administrator's competence; persistence pays off. Second, the more severe oversight rule is making it more difficult for administrators to achieve success, whatever their 217 skills. The rules promulgation process is expensive. It involves staff time, costs of newspaper publication of notices, and ccmnunications charges at a minimum. transmittal. Publication alone can cost over $9,000 per rule Losing a rule, even if only a withdrawal, is thus to be avoided if at all possible. As a result, agencies engage in a variety of strategies in an attempt to achieve as high a rate of success as possible. They mobilize and coordinate public testimony before the JCAR to the degree they can, provide centralized processing within the department in some cases, "do their homework" with committee members and related staff, exercise care in selecting who represents them before the JCAR, increase the number of hearings held prior to transmitting a rule to the joint committee, refer to outside authorities, and sometimes, although usually cautiously, engage in agenda manipulation through the promulgation of emergency rules. On the other hand, timing of submission of transmittals— the most frequently suggested strategy both in the literature and by informants trying to explain the reasons for the change to the reverse veto— seemed not to be occuring on any systematic basis; to the degree that it was, in was in the direction opposite to that which would be predicted. A greater percentage of proposed rules were submitted near the end of the spring session or early in the summer recess under the reverse veto than under the negative veto. Now, with the virtually year-round legislature, the power of the committee to meet during legislative interims, and the ability since 1978 to extend the review period to 90 days, it is difficult to construct what would constitute strategic timing in the Michigan case. 218 Lack of strategic timing notwithstanding, the key fact remains: bureaucrats do indeed pursue strategic activities. This was interpreted as further support for the argument that Michigan's legislative oversight of rules is substantive and that the designated review conmittee, the JCAR, is a force with which bureaucrats must reckon as they seek to advance professional-technical goals and standards. It appeared that when departments could target legislator benefits they enjoyed greater success than in cases where there was great controversy and only generalized benefit to be realized. The Department of Natural Resources transmittals showed high approval of noncontroversial, localized rules (the watercraft and hunting control cases) but repeated difficulty with controversial, generalized rules (the water quality/Rule 57 case). In the situation where programs lack differential positive support, the departments have nothing to trade; the legislators' tendency is to wait until a consensus emerges for their confirmation. The result could be multiple withdrawals of a rule. Gubernatorial involvement seemed highly strategic, limited to those cases simultaneously most likely to encounter difficulty and most inportant to the departments. The governor, and the departments, however, have several alternative courses of action should this particular part of the rules process fail them. Emergency rules can be pronulgated, they can attenpt to operate by guidelines, or can move to adjudication. The relative rate of emergency rules promulgation varied considerably. They ranged from 8.9% of the number of rules processed through the JCAR in 1972 to a peak of 18.5% in 1980 to a low of 5.7% in 1982. While obvious needs for emergency rules exist, not all rules 219 promulgated through that special non-legislative review process seemed to fit the category. These cases were particularly interesting for their strategic elements, in cases where emergency rule promulgation simultaneously repealed an existing rule, the status quo position for further JCAR consideration was affected. Even though emergency rules could be promulgated for only a six month period, with one renewal allowed, this, it was argued, was one of the ways in which bureaucrats could exert greater agenda control in the face of the normal legislative review requirements. The recent review of existing rules has shown the resilience of departments in resisting public desire for change. The letter by the JCAR chairperson {and its attachment from the Michigan Manufacturers Association) to the governor requesting his assistance in securing department and independent commission cooperation was an especially interesting exanple of the complexity of the process, the continuing independence of the agencies, and the use of the system by well organized groups. (See Appendix B.) Chapter 5 showed the legislative review process to be an additional arena for lobbying rather than an alternative, with rare exceptions. An organization deliberately bypassing the earlier stages was likely to be caught by the committee and rebuffed. On the other hand, the conmittee would send a proposal back to the department if they were satisfied that someone's concern had not been sufficiently considered by the agency. By being an additional lobbying point more than an alternative, legislative review has clearly raised the stakes in participation in the pronulgation of administrative rules. Critics who fear this merely increases the "special interests" control of government have cause to be 220 concerned. Cn the other hand, the joint conmittee is not responsive only to special interests, if by that, one means large economic interests, slightly more to the point might be a concern for "single interest" influence, whether the right-to-lifers, the Michigan United Conservation Clubs, or the individual parent who had a daughter die in a state-licensed stunner camp, a determined, single interest representative is likely to have an impact on the content of rules processed by the Joint Conmittee on Administrative Rules. Despite the preceding statement, the preponderance of participation at the JCAR level is by well-organized, well-financed, and, most frequently, economically based interest groups. Business groups have several benefits available through participation in the oversight process. One result may be greater regulatory certainty. uniformity. Another is Yet a third is possible advantage compared to conpetitors in other states. These are in addition to the frequent argument that less regulation is an economic "good" on general principles. Major business umbrella groups such as the Michigan Chamber of Commerce and Michigan Manufacturers Association have their representatives in regular attendance at JCAR meetings and in frequent contact with conmittee members, staff, and agency personnel alike. Others maintain less frequent but equally intensive contact when a specific need arises. Bruff and Gellhorn, and Lanbert's concern that certain groups will get "the second bite at the apple" is well worth attention. The Michigan legislative review system is obviously open for full blown political influence. To isolate it from that, however, would return it to the realm of more closed decision making with presumably greater weight placed on professional-technical standards, values, and input. 221 The opening up of the process does indeed mean that some get a second bite at the apple; the alternative seems to deny it to everyone. The Michigan experience contradicts both models posed at the beginning of Chapter 1. Levels of institutional and individual investment in the rules promulgation oversight function in Michigan are substantial, whether measured by seniority, workload, or staff support. The "runaway" model cannot account for this. The second model asserts that the bureaus are already under legislative control. If that were the case, one would expect legislative intervention in rules promulgation to occur very infrequently— rules should be almost routinely approved with very few being denied. Here, however, even under negative veto, there was documented legislative intervention in approximately fifteen per cent of the cases. Under the reverse veto, that figure sometimes exceeded thirty per cent. Even this under reports the degree of legislative intervention for, as pointed out, JCAR documents do not record committee-directed "lay-overs" or negotiations on "administrative clarification," both frequent occurrences. It would also be extremely difficult to defend a rational actor perspective as the sole explanation for conmittee voting patterns and decision outcomes, unless one is willing to accept an expanded model in which non-economic, non-personal, benefits and goals are recognized. Concepts of the public interest or the proper role of government do play a part in these decisions. Many cases could probably be predicted on more conventional economic, or constituency benefit bases, but not all. The model best supported by the findings was that of the "constrained" bureaucrat with its belief in the need for and efficacy of increased oversight. Bureaucrats seemed indeed to have been pursuing 222 something other than legislator goals, evidenced both by the rates of withdrawal and disapproval and the proportion of denied rules which were never resubmitted. The enplacement of more stringent oversight mechanisms, especially the requirement of prior legislative approval of administrative rules changes, was accompanied by lower rates of approval, regardless of whether each transmittal was measured as a separate unit or resubmitted cases were tracked and evaluated only on the basis of their eventual outcomes, ttie use of the more stringent oversight mechanisms made it more difficult for bureaucrats to do what they wanted and easier for legislators to achieve their own ends. The system has resulted in legislators directing bureaucrats into more "acceptable" behavior, eventually achieving adjusted approval rates of over 90%. The continuing rate of personal and institutional investment in this system indicates that this state legislature, at least, has not found more overt mechanisms of control counterproductive. It appears that legislative and bureaucratic incentives are interacting in a fashion which is responsive to the public. The driving force of the legislative (electoral) imperative shapes the professionally-oriented bureaucratic (survival) decision environment. The review process exercised through the Michigan Joint Conmittee on Administrative Rules offers non-governmental actors a forum unique for its intersection of these interests. 223 1.2 Where Next? Suggestions for Further Research Three major problems remain for resolution as a result of these findings. First, there is at least one alternative major explanation which needs to be systematically tested. Secondly, there should be systematic investigation of the influence of various factors on the individual JCAR menber voting decisions. Finally, a specifically predictive model needs to be developed, testing in a comprehensive fashion the relative contributions of the multitude of factors probably affecting outcomes. This could be done both at the individual and conmittee levels. This study found a substantial relationship (ganroa = .428) between the type of veto rule and outcomes, with approval declining under the more stringent legislative oversight mechanism. Shifts in the balance of regulatory ideology among JCAR merrbers could also account for this difference. Construction of a regulatory index score for each JCAR menber through time could provide the basis for testing such an alternative. Such an index might be based on votes on relevant legislation; its greatest problem would be maintaining comparability over time. This or a separate measure might incorporate information on interest group endorsement of candidates, but may not be available for all JCAR members nor on a sufficiently corrplete basis. I suspect the influence of regulatory ideology is actually interactive with change in veto rule. Change in the rule in nichigan 224 took place in the middle of a legislative session. No members of the conmittee changed at that time, but approvals dropped 7.1%. (See Table 2 3.5.) v-ontingency analysis of the difference yields a X of 1.91, significant at the .20 level, and a gamma of .193. While the 20% level is generally regarded as unsatisfactory for hypothesis testing of this type, people involved in real world politics would probably think these were great odds. On the other hand, even if we were willing to accept this as a strict test, the relationship is fairly weak at .193. The veto rule is obviously only one influence on outcomes. Statistical analysis in this research was conducted entirely at the aggregate level. More systematic investigation of determiners of individual voting decisions could add greatly to our understanding of these phenomena. _Data-need to be collected on individual districts, more on caitpaign contributions and endorsements, regulatory ideology (as suggested above), perhaps indicators resulting in business/commerce and union support indexes, personal votes on authorizing statutes and rules transmittals, and even the size of the majority in the member's chanber on recent authorizing statutes. Collection of the type of data suggested above would allow the creation and testing of a comprehensive model of outcomes of legislative oversight of administrative rules. Individual level data could be used to predict individual votes as well as aggregated to predict the committee outcome of a given rule. Probit testing would be appropriate in this context since the 225 dependent variable would be a sinple dichotomous variable— a yes or no 1 vote, or approval against all other outcomes. *robit analysis would yield information not directly (Attainable through cross-tabulations. It would allow the simultaneous consideration of the effects of several variables on the individual voting decision and the outcome of the JCAR votes on proposed rules. In addition, because probit estimators are coirputed in the context of their contribution to the probability (range "0" to "1") of a given outcome, the relative weights of a specific 2 variable can be corrpared across time periods, such comparisons would allow us to more clearly interpret the simultaneous effects of several variables than would be possible through contingency analysis even with a series of controls. It would make it possible to test the degree to which the included variables account as a group for observed transmittal outcomes and to see how their relative influence may vary under different conditions, most specifically, under conditions of negative and reverse veto, in addition, models generated from the first two time periods could be tested against data from the third time period, where changes in factors other than the veto rule may prove to have significant inpact. Obviously, substantial work remains to be done. I believe it worth our while. 1. Actually, in the corrmittee vote probit model, it might be possible to predict withdrawals separate from other non-approvals. 2. See Aldrich and Cnudde, 1975, and Pindyck and Rubinfeld, 1981, Chapter 10, for discussion of the merits and interpretation of probit estimators and predictions. 226 1.3 Extending the Findings Michigan's review process, and changes in that process, demonstrate a system of multiple influences, with different actors in charge in different portions of that system and under different circumstances. „hat are the key elements of the Michigan system which seem to account for the observed results? First, the system is highly centralized, maximizing benefits for those legislators who are predisposed to participate in this type of activity. The rules oversight function is formally separated from other standing coirmittees and traditional forms of review, allowing legislators to specialize in this unique role. Second, the conmittee has been invested with substantial power and resources, increasing the value of serving on it and linking personal motivations and institutional incentive structures. Both institution and individuals gain as a result, with the institution satisfying oversight values while requiring little personal effort by most of its members. Third, Michigan's full-time, year-round legislature means legislators are able to invest more of their time in legislative activity than would be the case for part-time legislators. Given that oversight takes second place to legislation for most legislators, increasing the total time available increases the likelihood that oversight will get some portion of it. An additional effect is 227 legislator year round physical availability for regular meetings. They are mich more likely to meet if already in the capital than if they have to be called together from throughout the state. Next, and an additional critical governmental factor, is the presence of a long established, independent civil service. Some may see this as an obstacle, others as the only saving element in the entire picture, but at a minimum, it provides the balance which allows the system to work, something easy to lose track of in the focus on legislative oversight. Michigan's professional civil servants do indeed bring their own, independent, values, goals, and standards, their own experience and expertise to the rules promulgation process. Without it, the choices would be a truly "captured" bureaucracy or possibly machine-type political disbursement of resources. Finally, an undoubtedly important outside factor is a public generally supportive of regulation although in recent years it has reflected the growing conservatism observed nationwide. The conmittee is responsive to these changes, particularly so as a result of the legislative mandate to review and reduce the "burdensomeness" of existing rules, Jais alone must account for some of the decline in approval rates, although there is no direct way of measuring it. I prefer to think of it as an interactive effect with the greater legislative control being not only an expression of institutional values but also representative behavior in response to public demand, xhe result is a more difficult existence for bureaucrats. What do these results and factors suggest for other states? First, current political opinion in most areas of the country would 228 seem conducive to greater legislative oversight. There is no way to predict how long that environment will be maintained. If state legislators are serious about asserting greater responsibility in this area, this would seem an ideal time. Second, states are increasingly professionalizing their public employee structures, xo the degree that they do so they will create one element of the balance necessary for this system to work. States without such a system in place face a much more difficult task. It is my belief that to accomplish the same quality output without such a staff would require much more legislative input that anyone is likely willing to give, whether from institutional or individual perspectives. This factor may make the Michigan level of legislative oversight inappropriate in same states. Finally, looking at the legislative factors, I believe the centralization of the Michigan system to be the single most important factor in its effect. By removing the oversight responsibility from the traditional standing conmittee location, it has provided a sufficient concentration of resources and incentives to attract a few predisposed members of the legislature to specialize in this unique role, simultaneously accomplishing personal and institutional goals. In a state in which oversight is currently a very minimal level activity, it may be possible to initiate this type of oversight centralization with minimal disruptions of existing relationships, the other hand, if in a state where some standing committee based rules oversight occurs, transition to a centralized system might be accomplished through dual committee assignments, initially drawing oversight committee members 229 from those who are currently most active in oversight activities in the old coniQittees. to the degree that existing conmittee chairs or other conmittee mentoers are unwilling to relinquish any area of jurisdiction, however, there will he problems with even this strategy. I have not attempted to assess the degree to which Michigan's system relies on the current year round schedule. As demonstrated in Chapter 1, in the early years of the Joint Committee on Administrative Rules, Michigan held a part-time legislature; the conxnittee was enpowered to meet and suspend rules during legislative interims. Throughout the era of systematic, routine review covered here, however, Michigan had a fulltime legislature. Could such a review system be maintained under conditions of a part-time legislative schedule? conditions were met. Possibly, if certain If substantial institutional resources and significant substantive authority were invested in a centralized conmittee it still might be possible to attract the requisite level of personal investment necessary to make such a system functional. 1.4 The "Bottom Line"— a Final Word Clearly, it is possible to create a system of legislative oversight which will attract sufficient personal investments to make a difference. In the Michigan case, increased severity of legislative review requirements resulted in lower rates of approval of administrative rule proposals. Not all observers would agree this is to 230 be desired. What is rrty opinion of the «ichigan process of legislative oversight of the promulgation of administrative rules? It further opens administrative processes to the public in an arena where their voice has more weight, even if not all participate equally. v*iven the apparently necessarily increasing role of administratively determined regulation in the lives of American citizens, I think this is important. Despite the problems this poses for those who lack resources to sustain both the legislative battle and the rules promulgation battle, x think the net result is positive. Wie ensuing rules are not necessarily those which cormitted professionals in each of their respective fields would like to see, nor even those I would prefer. Nevertheless, I believe that overall the state has more workable rules, which in the real world usually means better ones. I believe Michigan is better off for its increased investment in overt legislative oversight of administrative rules. APPENDICES APPENDIX A APPENDIX A Buies Transmittals: 1977 The following table shows all rule transmittals received by the Joint Comnittee on Administrative Buies during 1977 , the last year of review under the negative veto rule. (List includes a case received on Deceirber 30, 1976, but numbered by comnittee staff as the first 1977 transmittal.) The final column, labelled ”0", is the outcome column. Outcomes in that year are: 1 2 3 7 9 Approval— no action Approval, comnittee vote Withdrawal Disapproval Other Among the 1977 cases are exanples of: a. Approval through lack of comnittee action, e.g.r is 002, 018, 038. b. Approval by comnittee vote, e.g., #s 003-004 , 006-009. c. Agency withdrawal of a rule, e.g., #s 001, 005, 010-011. d. Besubmittal of a withdrawn rule, e.g., 4026 (code #285.817.1+), submitted one day after #019 was withdrawn, and #039 (code #285.627.1+) submitted the same day #027 was withdrawn. e. Disapproval of a rule, e.g., #s 067, 093, 100, 103+. NUnber 103+ is a rare occurrence of a transmittal split subsequent to its submittal to the JCAR. Transmittal "103" was approved, "103+" was disapproved. 231 232 f. Resubmittal of a previously disapproved rule, e.g., #115, code #436.1101+, a resubmittal of #067. This transmittal came in too late to be resolved in 1977 and died (outcome code "9" — other) under stipulations implementing the new reverse veto amendments. Resubmitted as #015 in 1978 (not shown), it passed. Cases designated "H-" (see #041) are examples of localized hunting restrictions; those identified by "W-" (see #012) are localized controls on watercraft use. « 233 Table A—1. Dept Tms # Administrative Rule Proposals Transmitted to the JCAR with Administrative Code Number, Dates Received and Acted Upon, Days Elapsed, and Outcome: 1977. Subject DNRS 001 Betsie River Natural R Zoning PbHlt 002 Mininum stds, hosps-maternity PbHlt 003 Program match requirements Agric 004 Mutuals DNRO 005 Oil & gas operations DNRO 006 Polychlorinated biphenyls Comrc 007 LCC-declaratory ruling Labor 008 Agricultural tractors Labor 009 Farm field equipment labor 010 Agric powrd industrial trucks Labor 011 Head protctn equip,ag operatns DNKL 012 W-Comnerce Lake, Oakland Co DNRL 013 W-Galien R,New Buffalo Hrb,Ber DNRL 014 W-Galien Rv, Berrien Co LcReg 015 Pharra Bd-regstrd pharmcst exam DNRS 016 Betsie River Natural R Zoning Comrc 017 DCC-beer -.nonreturnable contnrs Labor 018 Plumbing Code Agric 019 Qrtr hrs brdrs awrds,suplnnts Labor 020 Electrical code Agric 021 Prevntg spread of brucellosis Agric 022 Dispsl bruc'll's expsd animals Educ 023 Direct student loans labor 024 Barrier Free Desgn Bd-gen rls State 025 Campaign finance reporting Agric 026 Qrtr hrs brdrs awrds,suplmnts Agric 027 Care nrsry stock,sales outlts LcReg 028 Real estate schools State 029 Campaign finance reporting Agric 030 Quarter horse racing DNRL 031 W-Gun Lake,Barry Co Agric 032 Landscape Archtct Bd-exam prcd DNRO 033 Oil & gas operations DNRO 034 General-obsolete DNRO 035 Cleaning agents,water condtnrs Educ 036 School social worker Educ 037 Legislative merit award prgrm Educ 038 Teacher certification code Agric 039 Care nrsry stocks,sales outlts Administrative Date Code No. Rec'd Acted DE 281 325 325 431 299 299 436 408 408 408 408 281 281 281 338 281 436 408 285 408 285 285 390 125 169 285 285 338 169 431 281 285 299 299 323 340 390 390 285 31 1051 4151 51 1101 3301 1971 45101 45301 45201 46101 763 711 711 474 31 1627 30701 817 30801 123 156 1601 1001 1 817 627 2601 1 71 708 901 1101 1 1173 1001 1501 1125 627 c c c c c c c c 43 2 3 c c lc c lc 1 c c c lc lc c c c 8 4 c c c c c lc 12 1 1 1 2 2 2 2 2 2 2 2 f 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 5 5 5 5 30 12 17 26 1 4 11 14 14 14 14 28 28 28 28 28 3 10 10 11 14 14 15 28 29 30 31 31 1 7 13 14 18 18 18 2 4 6 12 2 3 2 2 3 3 3 3 3 3 3 3 3 4 3 3 3 5 3 4 4 4 5 5 4 4 5 5 4 4 5 5 5 5 6 6 6 7 5 22 54 14 62 15 29 15 20 9 36 1 25 1 18 22 36 8 22 21 37 21 37 22 22 22 22 21 52 29 29 29 29 29 26 10 61 29 19 5 25 5 22 5 22 3 59 3 36 1 3 5 6 12 42 26 56 5 4 26 19 17 34 17 33 3 15 17 29 17 60 7 36 7 34 6 61 17 5 o 3 1 2 2 3 2 2 2 2 3 3 2 2 3 2 2 2 1 3 2 2 2 2 2 3 2 3 3 2 2 2 2 2 2 3 2 2 1 2 234 Table A-l (cont'd.). Dept Tens # labor DNRL Comrc DNRL DNRL DNRL DNRL State Educ DNRL EdUC HighW Comrc DNRO Agric Agric MilAf DNRO DNRO LcReg Comrc Comrc labor Agric Labor AtGen PbHlt Comrc DNRO Labor Corr Labor labor LcReg Educ DNRO DNRO State DNRL LcReg DNRL DNRL DNRL MilAf Comrc 040 Ski Area Sfty Bd-ski lifts 408 041 H-flilford Twp,Oakland Co 317 042 PSC-Michigan gas safety code 460 043 W-Hardwood Lake,Ogemaw Co 281 044 W-Lake Montcalm,Montcalm Co 281 045 W-Healy Lake,Manistee Co 281 046 W-Lake Fenton,Genesee Co 281 047 Dealer designation 257 048 Driver education 388 049 H-Highland Twp,Oakland Co 317 050 Schl dist pupl acct,dstrb aid 340 051 Drainage assessment 280 052 LicpC-attire,conduct on-premise436 053 Wilderness & natural areas 322 054 Commercial feed 285 055 Premium altnnts to fairs,shows 285 056 Rental of armories & grounds 32 057 Cleaning agents,water condtnrs 323 058 Flood plains & floodways 323 059 Physcl Therapy Reg Bd-gen rls 338 060 Stds rate filng physcn prof ins500 061 Liq Con Com-licens 'g qualfctns 436 062 Automotive service operations 408 063 Food Inspctn Div-last day sale 285 064 Fire fighting 408 065 Consumer Protection Act 14 066 Control of tuberculosis 325 067 Liquor control Commission 436 068 Air Pollt'n Control Com-gen rl 336 069 Elevator Sfty Bd-existg instln 408 070 General rules 791 071 Elevator Safety Bd-general rls 408 072 Errply Sec Com-publ partp, rntgs 421 073 Nursing Home Admstrtors-exams 338 074 Use schl bus t m s p senior ctzn 340 075 Conmercial fishg-yellow perch 299 076 Conmercial fishg-closed seasns 299 077 Special farm vehicle permit 257 078 W-Vhlley&WiIdwood Ucs,OaklandC 281 079 Pharm Bd-cntrld sbstncssregtrn 338 080 W-Round Lake,Oakland Co 281 081 W-Lake 28,Mecosta Co 281 082 W-Galien R&New Bufflo Hrb,Ber 281 083 Dept org & gen functions 32 084 Securities Bur-debt management 451 Subject Administrative Date Code Ho. Rec'd Acted DE 0 101 163 14008 765 759 751 725 21 312 163 2 1 37 49 635 811 3 1173 1311 1131 901 1101 17201 554 17401 51 898 1101 28 8205 1101 8149 351 2811 231 815 1075 51 763 3131 763 754 711 11 1201 cd 42 r c 14 8 6 14 c c 40c c c a c lc 3c lc c ' c c c c lc c c c c c c c c 54 52 11 2 c c 5 16 5 16 5 19 5 23 5 23 5 23 5 23 5 23 5 24 5 24 5 26 6 1 6 3 6 8 6 14 6 14 6 16 6 21 6 23 6 24 6 30 7 6 7 8 7 13 7 21 7 21 7 21 7 22 7 22 7 28 7 28 8 3 8 8 8 11 8 12 8 18 8 18 8 18 8 22 8 25 8 30 8 30 8 30 8 31 9 12 6 7 6 21 7 7 6 21 7 5 6 21 6 21 7 7 6 28 6 21 6 28 6 28 8 3 8 2 8 2 8 2 8 2 8 21 8 2 8 2 7 7 7 22 9 8 8 2 9 21 9 21 9 21 9 21 9 21 9 28 9 27 9 27 10 4 9 27 9 27 9 27 9 27 10 4 10 4 10 25 10 4 10 4 10 4 10 25 11 10 22 36 49 29 43 29 29 45 35 28 33 27 61 55 49 49 47 61 40 39 7 16 62 20 62 62 62 61 61 62 61 55 57 48 46 40 40 47 43 61 35 35 35 55 58 2 2 2 2 3 2 2 2 2 2 2 2 1 2 2 2 2 1 2 2 2 3 1 2 2 2 2 7 2 1 2 2 3 2 2 2 2 2 2 2 2 2 2 2 3 235 Table A-l (cont’d.). Dept Tens # DNRO PbHlt DNRL LcReg DNRL LcReg State LcReg LcReg MgtBd Comrc Comrc SocSv SocSv LcReg LcReg State PbHlt LcReg LcReg Comrc DNRL DNRL DNRL DNRL DNRL DNRL DNRL LcReg LcReg Comrc comrc PbHlt 085 Public access stanp program 086 Health maintenance organiztns 087 H-Summit Twp,Jackson Co 088 NUrsing Home Admstrs-cont educ 089 E-Bagley Twp,Otsego Co 090 P h a m Bd-cntrld sbstncs-sched2 091 Motor vehicle service & repair 092 P h a m Bd-cntrld sbstncs-schld4 093 Osteopthc Med & Surg-pbl rcrds 094 Lottery Bur-millionaire party 095 PSC-motor carrier safety 096 Ins Bur-nonprofit hosp srv crp 097 State Housing Dev Auth-gen 098 Stt Hsg Dev Auth-dev fund g m t 099 P h a m Bd-cntrld sbstncs-schldS 100 Real estate schools 101 Licensing vehicle brokers 102 Supplying water to the public 103 Dentistry Bd-dentistry rules 103-+Dentistry Bd-dentistry rules 104 PSC-transptn of migrant wrkrs 105 W-Arnold Lake, Clare Co 106 W-Cranberry Lake, Clare Co 107 W-Bellew lake,Benzie Co 108 W-Bellew Lake, Gr Travers Co 109 W-Lake Lancelot,Gladwin Co 110 W-Bronson Lake, Benzie Co 111 W-Galien R,Berrien Co 112 Medical Practice Bd-pbl rcrds 113 Med. Practice Bd-anphetamines 114 Debt management 115 Liquor Con Com-lcnsg qualfctns 116 Supplying water to the public Subject Administrative Date Code No. Rec'd Acted DE 0 299 325 317 338 317 338 257 338 338 432 460 550 125 125 338 338 257 325 338 338 286 281 281 281 281 281 281 281 338 338 451 436 325 c 981 6101 c 138 2c 2841 c 169 1 3116 c 111 3123 141 c 201 c 16101 c 1 c 101 c 151 3125 2601 c c 181 c 371 c 201 4115 c 1 718 11 718 12 710 7 728 9 726 1 710 6 711 3 2391 2303 c 1201 c 1101 c 10101 9 a 9 9 9 10 10 10 10 10 10 10 10 10 10 10 10 10 10 plO slO 10 11 11 11 11 11 11 11 11 11 11 12 12 15 21 26 28 10 13 14 18 18 20 26 27 28 28 28 28 20 30 31 31 31 1 1 1 1 1 1 1 1 1 30 6 7 10 11 11 11 11 11 12 11 12 12 12 12 11 11 12 12 12 12 12 12 12 11 11 11 11 11 11 11 12 12 12 99 12 25 16 10 15 15 15 14 15 15 13 13 13 15 15 6 13 28 6 13 13 29 29 29 29 29 29 29 29 15 15 15 99 13 40 56 46 50 36 33 61 28 5B 54 48 47 18 18 39 46 61 36 43 43 59 28 28 28 28 28 28 28 44 44 15 99 6 2 3 3 2 2 2 1 2 7 2 2 2 2 2 2 7 1 3 2 7 3 2 2 2 2 2 2 2 7 2 2 9 2 Source: Conpiled from "Daily Status Report", unpublished daily log of Joint Comnittee on Administrative Rules, Legislature, state of Michigan. APPENDIX B APPENDIX B Rule Approval by (Selected) Department by .ear and Veto Period, 1972-1982 The following table shows success rates for cases from 1972 through 1982 (the period of split party legislative and executive branch control) for those departments which averaged at least two cases per year over the eleven year period. Departments are listed in order of the total nuirber of rule transmittals submitted, those with the fewest being first, .he Department of Natural Resources (DNR) had the greatest nunber of cases and is thus last. DNR data are given for: a) all cases, b) local cases (local hunting and watercraft regulations), and c) all other cases* For discussion, see Chapter 4, page 122 and subsequent references. 236 237 Table A-2. Annual Number of Rules Transmittals and Per Cent Approved for Twelve Departments (including additional DNR breakdown) by Veto Period, 1972-1982. Negative Veto Period if Reverse Veto Period Dept- 1972 1973 1974 1975 1976 1977 Total A197B 1979 1980 1981 1982 Total Treas 1 100 3 67 2 100 2 100 1 100 1 100 10 90 MgtBd 1 100 1 100 3 100 1 100 1 100 1 100 State 0 na 0 na 2 100 5 60 2 100 StPol 4 75 2 50 0 na 1 100 ScSrv 3 33 3 100 6 50 Educ 12 92 7 100 Agric 6 100 PubHl 3 67 14 64 1 100 12 50 21 52 8 75 5 40 5 40 23 61 7 57 4 75 9 67 7 71 29 69 11 6 I 50 6 17 13 77 1 100 2 100 28 67 6 67 1 100 12 58 4 75 6 100 29 72 45 93 If 12 If 75 8 75 7 86 5 80 10 90 42 81 6 67 42 83 f f 5 60 10 70 18 44 22 68 8 38 63 57 16 81 13 62 60 82 11 23 f 74 15 87 21 71 19 79 16 81 94 78 7 71 24 75 16 69 71 76 If 30 11 53 21 62 23 91 23 83 17 71 114 71 39 97 25 44 53 100 15 67 165 87 If 14 If 79 30 87 14 79 29 79 33 85 120 83 70 90 89 90 48 96 48 92 45 87 328 91 If 58 % 85 43 74 30 73 43 79 16 100 190 81 16 94 53 94 71 94 36 97 38 97 31 90 245 95 f 48 11 94 23 96 18 83 25 84 11 100 125 91 12 92 17 77 18 72 12 92 10 70 14 77 83 80 11 10 * 40 20 50 12 58 18 72 5 100 65 60 3 67 4 100 1 0 3 33 8 100 If 0 % na 2 50 6 50 6 83 15 80 11 1 If 100 4 75 1 100 0 na 8 75 2 1 1 100 5 80 1 100 2 100 20 70 4 100 5 100 5 100 7 100 40 98 u f 8 100 5 100 5 100 9 89 12 93 3 100 9 100 5 60 8 88 11 82 Comrc 4 100 4 100 6 83 17 88 LcReg 5 100 9 89 10 70 Labor 20 100 13 92 DNRall 28 93 (DNRl) (DNRO) 1 f • APPENDIX C APPENDIX C Letters to: 1* The Honorable James J. Blanchard, Governor (from Michael J. Griffin) 2. The Honorable Michael Griffin, House of Representatives (from Deborah K. Hoover, Michigan Manufacturers Association) For discussion, see Chapter 6, p. 238 199 , and Chapter 7, p. 219. 239 THE LEGISLATURE WL mrs -Jttamx tMUIC VM,m*amEr UVNSING, MICHIGAN v* * » JOINT COMMITTEE ON ADMINISTRATIVE RULES atMtmiw m vtanama CMX4 SWltt tittMTvccuicBuma comm in—■IthiffftASSflMtf mmsttuam n*nooLm.m m u ttm a m numb amirum J anuary 10, 1985 The Honorable Janes J. Blanchard Governor of Michigan _ State Capitol Building Lansing, Michigan 48909 Dear Governor Blanchard: In March of 1983, at the request of-Speaker Gary Owen and Senate Majority Leader William Faust, the Joint Comlttee on Ads1n1stat1ve Rules began a comprehensive review of the administrative rules of the Departments of Commerce, Labor, Natural Resources and Public Health to determine which administrative rules of these departments were obsolete, unnecessary, duplicative or unduly burdensome to business and 1ndustry 1n Michigan. After nearly two years of work, this review has come to an end. Same of the results, are impressive— the Departments have responded positively to over 65% of the business complaints and have made commitments to amend or rescind the neces­ sary rules. These coonltments have all been submitted to the Joint Connlttee with a time schedule which provides projected promulgation date. In addition, as part of the review, each department did an Internal audit of their rules to identify those which were burdensome or unnecessary* Hundreds were identified. The Department of Public Health has already rescinded 195 obsolete rules. In other respects, this-review has been utterly frustrating. It was never the Intent of the Joint Comnittee to turn the review process Into a confrontation with the state departments. Rather, 1t was hoped that the executive and legislative branches would work closely with one another 1n an effort to Improve Michigan's business climate. This cooperative effort 1n many cases has simply not occurred. Particularly discouraging was our last Coemlttee meeting on December 17, 1984, where each of the department's presented a status report of the Issues raised regarding their department's administrative rules. Several areas of concern discussed at this meeting Include: I. Some of the complaints raised by business and Industry In which the Department's agreed to stake changes In rules could have been handled expeditiously. Instead, the Departments (particularly the Natural Resources Department) made these proposed rule changes part of larger, more controversial rule packages that may not be ready for rule promulgation for a year or longer. 240 The Honorable Janes J. Blanchard - 2 - January 10, 1985 2. Although each department has comaltted to a time certain for submission of aaendnents or rescissions to rules, these tine schedules have not been followed. Instead, new schedules have been made with "amended" dates. 3. There-have been lengthy delays In the meetings of several advisory groups who _ nust consider rule changes before they go thraughPthe rulemaking process 44ueto an Inability to get a quorum of- other Internal difficulties). This has postponed the discussion and resolution of many Issues for nearly the full two years. 4. In some Instances, there has been a lack of coomunlcation between a department liaison and the heads of agencies within his/her own department regarding preparations needed to respond to the issues and make appearances before the Joint Court ttee. In addition tothese concerns, I received on December 10, 1984, a letter from the Michigan Manufacturers Association which 1 have enclosed. This letter points out more frustrations regarding this review and their sense of disappointment. Because of the situation outlined above, 1 have asked Barbara McLeod and Pete Plastrlk of your Cabinet Council to assist us 1n tying up the loose ends. This State needs a reduction 1n red tape - not excuses and Inexcusable Inaction. He have worked diligently over the past two years and would like to see the fruits of our labor. This cannot be done without the cooperation and assistance of the executive branch. We urge your active participation in.persuading these departments to resolve outstanding issues and to follow through with their commitments in a timely manner. I look forward to your response. SIncerely, Chairman MJG:jra enclosure cc: Speaker Gary M. Owen Senator William Faust Members of the Joint Committee 241 M December 20, 1984 ITIichigan manufacturers association 124 East Kalamaioo Street- msiounMi k l M k CXCCUTIVI V K t PRtStKK T m a te CHIUMUM ■L a n s in g , M ic h ig a n 48933-2182 * P h o n e : A r e a C o d e 517 372-5800 - The Honorable Michael Griffin Chairman, Joint Committee on Administrative Rules Capitol Building Lansing, Michigan 48901 U m M I. Ilillai Dear Kike: __ VICE CHIUH I M I believe it important to convey to yon the reaction of HMA to the committee's follow-up meeting with the state agencies regarding the rules reviev project which you initiated over two years ego. TKUURER twice* k n l M . OlttCTMS Im. a LLt* MkUCM Meteel i m m a F m t P. Cap» . *• r f r n M n f C ara. t w n L n r* Ton, your committee, the MHA cosasitteea, and many ochers, all worked very hard and sincerely to identify important areas of concern in current rules and bring them to the attention of the agencies. Ua applaud your efforts and your sensitivity to the burdensome regulatory climate which we-are often faced with in Michigan. Manaatt Maul Pra - ata Mf» L.M M Siwtf. lac. Mae a, I h H h m TeecJia «aaa a Co. H IM NaaUac OttnH . StMf/Bnfe r m i h n Ea. I tK.MHk Tha O a n u ea -O tH i In e Ci. C ln a la a i la a a t i . M e a ta e t ia a i l ila ta n Carp. IWn« C*M V. Eraaa* CMai H a law ia t Ca. a a n r JL la a a a a a _ D aacM a M M Ca. T tm lac M MHa Ckartaa T. Uaafa* Tin UtfMa Ca, M M aa U HdOalar „ Cafeat traduct* Ca. Td fata* a . UMceaU ' e. Ira* Laaa U n * * ,' lac. ML O e la M rt w . H a w n SlatfMae laaaaulaa akwiaihaat ITcaaaie L ftf a a t* Wahl art Ca**. (P v a ri L tc ia iaa Saala* Ptwar Ca**. MmkifM MIN l a a fmr* Mcwr t u p l e 0t«fe*n « > * n j. f tl u a KTiKlwM c « * . ItM M h u m * ilk *c.nin< M MA Imlug However, in witnessing the response of the agencies to