‘1 I O "3:" {$31.11;} ' I; "’Ll'h'sz )H W @363 ”I?“ I?) O' 1. ”'5'” in; 61%: 3’52“" JI'OOP" ' ”it '.\‘.‘ 5:9 OL'IOpi J“ “rink: {O WC; O1: ’1“ ,f\ \OO‘ '"j' git S'KI]; ., W ." UH my 1m, {E ' V . '11.! 'I! V ‘f’ 't ' NO” 9“ka OO’, 1.11%}! 6'! 1;!” O‘.'.|\ 10-}359; 'fi‘r '3'); f’ 2 "UL! | s 33:1,“ ‘u A“ a 3"." '1: ,. (‘- "-' . ‘fzr'P'E \oIO HAL-3;} .7, Si‘tik' fi‘Tfi'J " . ”Egg?! '3‘" 3'! . v'{' 'V "'5“: 3;?!“ 'R'le‘ ’1' 11':"'f'O"" ' A w s “A “'wuahfléh‘”? s’ ' 'H.‘I'.’7‘,',"';3"' O . ' "-"3-'.5'1'4O’I‘n"1"3"' :' O YOOflOié' OOIRI'EOOOH‘HI'I O .- - JWIM, I '1‘“ L: v0 . *5 Of-f :5th [E's-I" 4. 'O;"'§u"liJ"O; '3'" '- "':"".1l!”‘~§ O'M'OIOOEM “IL!" "'IJ""""""" ‘ .', ‘Tm “H“ LO" OOH!" ' I (:33: f7 “I '+ ' 1"" - - .a \ ‘m'fl' _ ‘L w. Rib—‘12:?“ h "9"" 'F"'?§L§:"fi"' , g ' :‘f 4 Jam ' 5 O5" lidfiaéi'glfififillf I“? .35! t m ‘: . g :{V‘P‘ “' . '-' ' o J ' : o. it if; .. a... .. ... . 131:: .~- ,_.. 0* , gfi ; - -: . : :V -_ ‘v; ‘I ' I; ‘Io'? -. :2: - ' : ’ ' " " $43,. “ ' ' ’ : >:r:"-':« -‘i;:5:3~-i.~':1 ‘ ,. 4 .‘J f:.,' v A y a“! 7 {WOO «’2': O2}? "’ 5345313“, I“: '13., “ '3 '5‘. J. fill ":71 2"'"}'O'}1"'h L' A'HOH "‘ k V" c _ £1“: , 'XW‘O’ . 7'. .' HO'IO V H“ OH“ “'8": O: .11,“ I 1,13} ‘ ~:: {L' 'Oufl ' O": I!;"' u. .‘du' ; e .1'V? -4? :“33‘ . bi: 3.1:"; w‘. . “fl . 2:1! . '. :5 ‘ ‘o "I ' , 47,-. . 515.3- I r... .ml 0 v ‘ If c 'J i u ’7 u. A .- v . :i’fiv «ffic. ~ » .‘ ti . ‘ ‘O .. no ', DO 79. . . It A M-. XOI‘ ”~I'O'h; It'l’1tg'; v::‘. u 11:“ v: 33H: ‘2}; '51: ,. thn' r Ur” J3 .1 u. I -‘¢ . ‘ ~-..... 101 '... -‘-_' o . L l (3", " "r" -. 7‘ '1‘ ' ?A . "1 Jr . . "S Y ‘ n‘ f. ~ ~ ‘{ fi.‘ 1,“: n ' * ,-. .1 ' r .‘fi — :. PO , N A I. s ' . - ,:.~ .- .L' . I. O ' z; ".}'.ij-‘..r. . - v ‘1 Y" :fi‘i‘t‘ ' 7. - -‘ ‘ ,‘V. t '. u. 3! I'm . .r-sxm'z ‘ ‘v 'V' "3"- 9- O; 1“‘.;-"‘": 5m . nw'firvfi‘fi": :‘ 1“,”, t ‘1'. M14? " Qfiéfi‘fzflfi ' é~3w30 L"‘*\ ”'5‘ '1‘"? rum Jig QJr‘N'" ‘ “‘3' '32:! ‘1 " f [flaw/5, “I"; " 4,3? 5%.; . 1.23%. ‘O: . m ‘ .‘ .J -‘— ' ,-I \I\‘ :::.#‘£"O ”“4" 0.: :']OHO. 3— .1 JWO‘O' “1‘59.“ ' If}; hm'Jum‘bfaiq'p." V 3' . I‘ulz, {fig'if'f' ”g?- , ‘_~ 3 flfi$'.':., Oh -‘fih" ['zh‘l 591w“ 774., ' l‘Tgb 4 $371 a, -'r (x ' w' " " '1‘; infant}: ' .M '31": ' -f‘ hr He ' my; 0 I; JR?! :1# "POOL, “bun. '1 Rh {‘3' "u‘A-O ‘ \‘ {OM 1mg}; "L" 'EgnLv-MWJWILMJ" :‘ g'll'WR“? “HO'WB‘MLM. “flier I llBRARY Michigan State University This is to certify that the dissertation entitled CONFLICT OVER RULES: THE STRUCTURE OF COMMUNITY POLITICS presented by Allen Bronson Brierly has been accepted towards fulfillment of the requirements for Ph.D. degreein Political Science 34 557/111 [7 fi/gfi MM m6 Q r prof r Date jjm/ $1,. / ‘/ {577 J MS U is an Afl'mmm've Action/Equal Opportunity Institution 0-12771 hhhhhhm 3 1293 01085 5686 TV‘ESI.) RETURNING MATERIALS: Place in book drop to LJBRARJES remove this checkout from n your record. FINES will be charged if book is returned after the date stamped below. CONFLICT OVER RULES: THE STRUCTURE OF COMMUNITY POLITICS by Allen Bronson Brierly A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Political Science 1987 Copyright by ALLEN BRONSON BR I ERLY 1987 ABSTRACT CONFLICT OVER RULES: THE STRUCTURE OF COMMUNITY POLITICS by Allen Bronson Brierly Federalism is a complex political economic structure that influences private and public resource allocation. The role of federalism has not been well stated because of time underdevelop- ment of federalismflthegry. This dissertation pro- vides a coherent theoryMOf federalism with applica- tions to economic development policy choices and growth rates. The dissertation assesses these top- ics in Chapter Two by reviewing the relevant norma: tive social choice theories of federal structures. Chaptef"Threéydescribeswtfie historical development of metropolitan competition under state home rule induced fragmentation. Chapter Four provides an direct analysis of community tax price competition. Chapter Five analyzes the latest state efforts to compete with differentiated deVelopment policy packages."Chapter Six links economic growth rates to Eganizational activity within state "political systems. Thus I begin with a structural framework, explaifi the competitive behavior observable within this framework, and conclude with some analysis of long term consequences for economic performance. FOR MY MOTHER iv ACKNOWLEDGEMENTS I wish to express my appreciation to my disser- tation committee comprised of Professor Gary J. Mil- ler, Professor Charles Press, Professor Jeffrey Hill, and Professor Lawrence Martin. I am indebted to these individuals for their time, patience, and cre- ative energy which provided the impetus for this research. In many instances I either failed to com- prehend their criticisms or I stubbornly refused to take the suggested actions. Special thanks to the late Professor Phillip Marcus and the Center for the Study of Industrial Redevelopment for his financial support (of Chapter Five) at the early stages of this dissertation. I would also like to Professor Gary J. Miller for his financial support (of Chapters One, Six and some continuing research) material during the middle developmental stages of the dissertation. Finally, I would like to thank Professor Joseph Dar- den and Professor Bonnie Morrison of the Urban Affairs Program for their financial support (of Chap- ters Three, Four, and parts of Five) during "crunch" time at the later dissertation stages. My research could not have been initiated or completed without generous support from these organizational represen- tives. There are several kind individuals who read and provided constructive criticisms and encourage- ment of various chapter drafts. I am grateful to Professor Ira Polley, Professor Kenneth VerBurg, Pro- fessor James Morrow, Professor Richard Feiock, and Margaret Gillis. Professor Richard Feiock is also co-author of a recent version of Chapter Six, and his kind contributions of references, data, and opera- tional definitions greatly improved the content of Chapter Six as it appears in the dissertation. Despite all of this good help and advice, the remain- ing judgemental errors are my responsibility alone. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES A FEDERAL SYSTEM EQUILIBRIUM A REVIEW OF FEDERALISM THEORY THE HOME RULE REFORM PARADOX vii ix 10 18 74 A THEORY OF IMPERFECT COMMUNITY COMPETITION 132 STATE ECONOMIC DEVELOPMENT POLICY CHOICES COMMON INTEREST ACCUMULATION AND SUBSYSTEM POLITICS BIBLIOGRAPHY vi 173 224 272 LIST OF TABLES Vote Percentage in Favor of Detroit Annexation Constituency and Year 1980 Suburban Census Data Stratification of Cities Contiguous with Detroit Income Class Homogeneity From 1950-1980 Detroit Tax Abatements Suburban Industrial and Commercial Development A Central Core City in a Rural County Wayne County Tax Exemption Percent Oakland County Tax Exemption Percent Macomb County Tax Exemption Percent Property Tax Bill Incentives for Business Estimated State Development Policy Choices Model Comparison with Goodness of Fit Results The Effect of Organizational Schlerosis on Economic Growth The Effect of Organizational Schlerosis and Unionization on Economic Growth vii 88 118 120 124 149 151 157 162 163 164 181 200 217 250 252 6.3a The Effects of Organizational Schlerosis and Subsystem Politics on State Economic Growth 261 6.3b Table 6.3b 262 6.4 Comparison of the Olson and Subsystem Models 266 viii LIST OF FIGURES 5.1 A State Development Policy Typology 176 5.2 A Classification of Development Policy Choices 186 ix CHAPTER ONE A FEDERAL SYSTEM EQUILIBRIUM Federalism is a complex political economic structure that influences private and jpublic resource allocation. The central role of federal- ism has not been well stated because of the underdeveIOpment of federalism theory. This dis- sertation provides a coherent theory of federalism with applications to specific areas of subnational policy concern, namely, economic development policy choices and economic growth rates. --‘> _.._-—_.... . By focusing upon a single constitutional arrangement within a political system we are able to understand the causal relationships between preferences, institutions, and policy outcomes. This somewhat traditional approach Ii; justifiable because, according to William Riker, "we have already learned from the (impossibility results) that we cannot expect to find equilibria of prefer- ences, but we may be able to find equilibria gener- ated from a given subset of preferences by particu- lar institutions." (Riker:20) This is an effective 10 11 research strategy for explaining and predicting complex empirical variables like economic develop- ment policy choices and growth rates. The value of ci,t_t,m ’ #11! _d//,,c~“111 m— theoretical understanding for the interpretation of complex real world phenomena cannot be overstated. From a: traditional institutional.perspective federalism assigns rights and responsibilities to various individual and incorporated agents within a political system. The patterns of relationships between the national, state, and local government are ordered by federalism. Federalism as a struc— ture is an important component to the general checks and balances adopted by the framers of the United States Constitution. From a modern political economy perspective, the mere existence of any political" rule affects resource allocation. Thus, procedural biases play a determinate E239 by influencing both the politi- cal processes and outcomes. Federalism is no exception to the general idea that structure biases outcomes; procedural biases create strategic incen- "—.._¢ tives to manipulate agendas and misrepresent pref— ‘..——. erences . 12 A federal structure induces strategic politi- cal competition by defining vertical and horizontal intergovernmental relationships. To summarize the basic theory of federalism presented in the disser- ation, consider the following logic: i) a federal structure induced equilibrium exists in an community system, if consumer-voters preferences meet standard conditions and if the community system acts in a multi-dimensional local public good space in such a way that each dimension is under the jurisdiction of a particular commu- nity; this implies l l {ii:,homogeneity of preferences for rules (like a federal structure) can generate stable patterns of group choices in the absence of preference homogeneity over (economic development policy) out- comes; w—m, iii) procedural stability generates strategic incentives to either a) manipulate rules to gener- ate outcomes or b) misrepresent preferences for the rule; iv) structures influence the direction and magnitude of strategic intergovermental 13 competition; v) vertical intergovernmental relationships involving the nation to city, nation to state, or state to city agents generally tend to be a) non- competitive and b) asymmetric with respect to intergovernmental bargaining positions; vi) state to city relationships tend to be integrative in time sense that states legally dom- inate cities. States regulate the degree of local competition by controlling a) the entry of new com- munities and b) the boundary decisions of existing communities through constitutional home rule provi- sions; vii) horizontal intergovernmental relation- ships tend 11) by highly competitive. Competition is induced by the degree of institutional fragmen- tation existing“ under federal structural arrang- ments; (:3???) state to state relationships focus upon non-price competition in the form of policy differ— entiation; currently, states utilize economic development policy’ packages in their efforts to ‘~___ \‘w- 7 -~— ——._._. .1... 1m and»; — —— ---—-L __ . differentiate the local public goods and services 14 available to firms. By differentiating the set of government incentives confronting firms, states hope to gain a strategically competitive advantage over other states; ix) city to city relationships involve tax- H price competition. Local competition for resources affects fiscal (taxes and subsidies) and regulatory (zoning) policy outcomes. Local competition is regulated by the entry barriers states utilize to incorporate and maintain community jurisdictional boundaries. Whenever states fail to actively regulate com- munity entry, local competition within metropolitan areas involves fierce tax, price competition for resources. Perfect competition among communities is not theoretically because 1) some services have economies of scale and 2) land resources are lim- ited in metropolitan areas and 3) the right to pro- vide local public services within a geographical area conveys a monopoly position to the agent(s) responsible for supplying the public good(s) in question. Even though perfect tax-price competi- tion is an unrealizble goal, some have argued that 15 more structural fragmentation induced competition is desirable. Given the imperfect nature of local intergov- ernmental competition, this implies states should incorporate as many community jurisdictions within resource constraints. There are clearly some effi- cieny gains to be had on the demand-side by increasing community competition. But there is an upper bound to these efficiency gains from increas- ing the number and scope of state and local poli- ties. The Los Angeles and Detroit areas are probably two metropolitan polities which have come close to exhausting competition induced efficiency gains. My belief is that producer efficiency for those local services exhibiting a high degree of publicness is not well served by metropolitan tax price competi- tion. There are some gains in consumer efficiency that generate production inefficiences. Allocative efficiency requires the correct level of community competition. I believe that destructive competition is likely to occur at fairly low levels of intergovernmental competition. 16 The causal relationships between the degree of pub- licness, institutional. fragmentation, enui commmu- nity competition and private and public sector resource allocations are crucial for understanding subnational economic problems. Yet these important variables have not been precisely modeled and for- malized within a structural theory of federal political economy. The dissertation begins these questions in Chapter Two by reviewing the relevant normative M social choice theories of federal structures and '\ then proceeds to interpret the descriptive politi- cal science literature within ‘a positive game _‘-__ A .- - ,‘ .._._._.. “.m-H‘, theoretic framework. Chapter Three describes the historical development of metropolitan competition under state home rule induced fragmentation. Chap- ter Four provides an direct analysis of community tax price competition. Chapter Fize analyzes the latest_state efforts to compete with differentiated development policy packages. And Chapter‘géx links economic growth. rates 11) organizational activity within state political systems. Thus I begin with a structural framework, explain the competitive 17 behavior observable within this framework, and con- clude with some analysis of long term consequences for economic performance. To conclude, this dissertation follows the basic logic of the proceeding argument. Means effect ends. Since: the ends are biased by the means utilized, agents attempt to influence the adoption, maintainence, and reform of institutional means. The specific institution in question (fed- eralism) has been undervalued as a means for stabi- lizing macro political system outcomes and micro policy outputs. A federal political system influ- ences public and private resource allocation by setting the limits of intergovernmental competition resulting from the formal division of power between national and subnational agents. CHAPTER TWO A REVIEW OF FEDERALISM THEORY I. Introduction Federal System Stability Under' a federal system, national government power is dominant but not dictatorial. National officials can use rules, regulations, and resource transfers to influence state and local policies. National authorities can dominate state and local policy processes and outcomes within federal struc- tural arrangements, but they cannot generally impose decisions upon subnational governments. State and local officials have numerous oppor- tunities to check and balance national authority. Community officials can utilize their independent electoral "soap-boxes" to articulate policy demands running counter to national priorities and to sup— port policies at variance with other communities. Community governments have some political indepen- dence under a federal structure, which is limited by the degree of community tax base competition and 18 19 the scarcity of transfer resources. A federal structure provides diffuse opportu- nities for influencing policy processes and out- comes. Individual political entrepreneur's and coalitions organize around the policy opportunities created knr a federal structure. Federalism thus provides both a mechanism for generating policy outcomes as well as a means for revealing individ- ual preferences. The stability of a federal system depends upon the stability of policy processes and outcomes under ea federal structure. The dynamic stability of a federal system depends upon the continual adjustment and formation of coalitions of minori- ties. Stable intergovernmental relationships, between national and subnational governments, rein- force existing policy outcomes and maintain incre- mental policy processes. Complex federal relation- ships enable government to allocate services and to redistribute resources. In the dynamic context, a federal system functions as a stable policy adjust- ment process. Federalism thus influences both short run policy outcomes and long run policy 20 processes. Community Policy Outcomes in a Federal System Federalism influences community policy out- comes by pmoviding multiple electoral and appoin- tive political forums. By furnishing numerous electoral opportunities, 2“: becomes unlikely that either a single global majority will exist over all forums or that a single local faction will be able to dominate all forums at any point in time. Fed- eralism creates opportunities for local factions and broad based coalitions of minorities to form. Short run policy outcomes may clearly reflect the importance of certain political leaders and their followers, but the long run stability of a federal system does not depend upon the stability of a single forum or the existence of a global majority. Long run system stability depends upon the constancy of intergovernmental relationships and the relative dominance of factions across all forums. If ea unique global faction exists, then that faction can enforce its preferred policies 21 comes across forums, no matter how diffuse the policy making process. But if global factions do not generally exist, and if inter-governmental relationships exhibit consistency, then policy processes and outcomes will tend to be stable under a federal structure. The instability of a majority rule unitary government, attempting to provide services and to even out resources, was well understood by state politicians designing the United States Constitu- tion.[1] Instead of a unitary parliamentary gov- ernment, they designed.aa structure separating and dividing authority which has checked and balanced extreme possibilities of liberal democracy. The mere existence of a federal structure influences the exercize of power by elective offi- cials, voters, and organized groups in the pmdicy process. Yet by checking the' possibility of a global dictator, federalism regulates time ability of elective officials, voters, and organized groups to unilaterally implement preferred policy alterna- tives. The purpose of dividing power with a fed- eral structure is to create a stable and complex 22 political system, while reducing the possibility of radical policy outcome changes or tyrannical policy processes. Thus a federal division of power gener- ates short term policy outcomes and stabilizes long term decision-making processes.[2] Why a Federal Structure? State anui local governments are increasingly designing policies to attract resources. Faced with declining tax bases and employment opportuni- ties, communities are competing to provide jobs and increase their tax bases. At the same time, for a variety of reasons, the availability of intergov- ernmental grants is decreasing. Those remaining transfer programs are also facing cutbacks. Inter- governmental relationships in general (and resource transfers in particular) have become overloaded, as community (factor) demands have outstripped national resource transfer supplies. Community competition to expand own resources is increasing at the same time transfer resources are decreasing. Increasing community competition and 23 decreasing transfer resources both limit subna- tional policy processes and outcomes. Community officials appear more limited in their ability to deliver services and to engage in redistributional activity. Community service needs are growing, while community and transfer resources are declin- ing. With the decline in scope and size of inter— governmental grants, transfer mechanisms are becom- ing less important. The remaining transfer mecha- nism appear overloaded as need exceeds resources. With these limitations upon community policies and overloaded intergovernmental relationships, we might ask ‘whether a federal structure is worth keeping? William Riker posed this important ques- tion concerning the existence of a federal struc- ture over two decades ago. [3] To answer Riker's challenge to federalism theory, this chapter pro- vides a theoretical account of the premises for evaluating a federal structure. Many researchers generally assert that struc- tural reforms are neccessary to improve policy performance. Yet "proposals for reform of an insti- tution are likely to yield unintended consequences 24 if their architects do not consider how people condition their decisions on those institu- tions."[4] If the purpose of intergovernmental relationships is to provide a stable division of power, federal structural reforms to improve policy outcome performance will redistribute power between national enul subnational leaders, constituencies, and groups. Some reforms may have unintended destabilizing effects. The theme of this chapter is thus to answer Riker's challenge by examining the basic components of a federal structure and their consequences for empirical policy' performance. The: chapter links deductive structural axioms with empirical policy hypotheses in order to provide a federalism theory reflective of modern policy problems. We do so by first describing three reasonable conditions to impose on a federal structure these represent criteria for evaluating intergovernmental relation- ships. Upon examinination we find these conditions logically inconsistent. Federal structural incoh- erence is a consequence design intergovernmental relationships simultaneously satisfying all three 25 conditions. The second section provides two linkages. First, we argue that structural incoherence deter- mines both the structural constaints limiting com- munities and. the norganizational complexity' over- loading intergovernmental relationships. Second, the community limitation and intergovernmental overload hypothesis statistically explain observ- able policy performance measures.[5] These empiri- cal hypotheses are explanations of the same federal policy performance failures. The community limita- tion hypothesis explains policy failure at the micro-federal system level, while the intergovern- mental overload hypothesis is consistent with macro federal system failure. Both hypotheses account for the current state of incoherence describing modern federalism theory and policy performance. After summarizing the various lines of feder- alism theory into analytical conditions, and link- ing them to pmesent empirical hypotheses, we con- clude by synthesizing the results into a federal theory reflective (n3 modern structural processes and policy outcomes. 26 II. A Useful Federalism Theorem An Overview of Model There are three basic conditions or postulates which have been advanced to justify a federal structure. However research indicates that no fed- eral structure or set of intergovernmental rela- tions exists jointly satisfying these three condi- tions.[6] The first axiom is citizen sovereignty.[7] When. a collective choice environment contains a locational component, citizen sovereignty or free individual locational choice has been hypothesized to generate an optimal allocation of resources. Under this allocation, the domain of individual preferences is restricted, because a system of diverse communities forms which is internally homogeneous.[8] The runMxn: of diverse individual preference combinations equals the number of commu- nities which have formed in equilibrium.[9] The second condition is local autonomy.[lO] Local autonomy' refers to the sovereignty' of the community population. Local autonomy implies a 27 system of communities have a right to local self government. Or, put another way, community pOpula- tions have local jurisdictional responsibility spanning a widely differentiated set of public goods and services. Without some degree of local autonomy, a federal structure cannot exist. The third and final condition relates to administrative efficiency.[ll] A optimal federal structure requires administrative efficiency, where community population boundaries and the span of local jurisdictional responsibilities coincide. Sometimes called perfect correspondence, othertimes named fiscal equilivalence, the administrative efficiency criteria covers both process and outcome waste. Postulate 1: Citizen Sovereignty Arrow himself postulated the non-imposition or citizen sovereignty condition. This condition has sometimes been referred to as non-coercive choice.[12] If individuals, as 21 group, prefer something, then the given collective consumption 28 bundle should 1x3 provided. Individual choices are somehow meaningful because their preferences over alternative consumption bundles are translated directly into actual provision. In a locational, territorial, or geographical context, the non-imposition condition refers to non-coercive individual location decisions. These nonimposed. or free individual location decisions are aggregated into collective group decisions within a federal system. In fact, the two rein- force each other: a federal system protects the right to free location choice and non-imposed col- lective location choices are probably only possible under a federal structure. The locational exit signals of firms and indi- viduals greatly limit community power, to the degree firms and households are mobile. Free loca- tional adjustment processes, such as “voting-with- feet”, result in a general locational equili- brium.[13] In this general locational equilibrium, the domain of individual preferences is restricted. Individuals sort themselves into internally homoge- neous communities. 29 The Tiebout "voting-with-feet" hypothesis was first articulated in "A Pure Theory of Local Expen- ditures." [14] In the Tiebout model, location Choices are revealed preferences for public goods. Tiebout wrote, "if consumer voters are full mobile, the appropriate local governments (will) form, whose revenue and expenditure patterns are set from consumer voters (collective) location choice."[15] UnObservable consumer preferences for public goods are revealed through search process. Once all consumers reach their ideal location, the group choice is efficient. Because social choice problems often revolve around public goods provision, the Tiebout model offers both a social choice process and a public goods allocation. James Buchanan and Charles Goetz summarize the public goods solution by stating that "Tiebout demonstrates that so long as local govern- mental units are appropriately assigned the task of providing certain public goods and services, and so long as individuals retain the freedom of personal migration choice among jurisdictions, there are efficiency generating (social choice) processes at 30 work, despite the publicness of the goods pro- vided.”[l6] If individual location choices result in a stably efficient group allocation choice, then it would appear that exit signalling is a superior mechanism to majoritarian voice mechanisms. Tiebout claims that the "solution, like a general equilibrium solution for a private spatial economy, is the best that can be obtained given preferences and resource endowments.”[l7] In a very recent article, this conjecture is proven false because "Tiebout's notion of an equilibrium does not have the nice properties of general com- petitive equilibrium, except under very restrictive assumptions."[l8] These assumptions include commu- nity officials following profit maximization goals, perfect intergovernmental competition, and usage of exclusionary devices. The Tiebout voting with feet process solves the free rider problem, if local public goods are essentially treated as non- divisable private goods. In some respects, the Tiebout model of feder- alism is a more interesting solution to the voter's paradox. The Tiebout process resolves the voter's 31 paradox by creatimg a federated community system. The partitioning rules inherent in a federal struc- ture are neccessary to enforce a Tiebout outcome. For instance, Fisch argues that ea Tiebout outcome is stable when exclusionary rules a utilized to regulate resource entry and exit.[19] Zoning, incorporation, annexation, and charter rules all function as community level exclusionary devices for regulating destabilizing resource mobility. These exclusionary devices enforce inter- nal community public goods allocations, while reducing the destabilizing effects (n3 mobile resources on the community system. Exclusionary devices structurally impose sta- bility on the community system as a whole. Tiebout suggests "while the solution may not be perfect because of institutional rigidities, this does not invalidate the importance" of exit options as a social choice process.[20] Exclusionary devices create incentives for individual preference revela- tion, and they generate policy outcomes by frag- menting metropolitan polities. The Tiebout ”voting with feet process achieves 32 pareto optimality by grouping individuals together in polities of homogeneous tastes; in the extreme, it satisfies Gerald Kramer‘s severe condition for consistent majority rule decisions, that all indi- viduals (within a community) have identical indif- ference maps, through the imposition of a silent unanimity rule."[21] Free location choice thus generates a public goods allocation when a federal structure exists to impose a system of communities on populations, and then enforce the subsequent collective location decision. A federated community system influences indi- vidual location decisions. Revealed location pref- erences are affected by metropolitan fragmentation. A simple model shows how individual's reveal pref- erences for wealthier communities.[22] Fragmenta- tion creates strategic preference revelation incen- tives because individual's do not face perfectly competitive community systems. Metropolitan fragmentation also influences local public goods outcomes. Community residents may strategically utilize exclusionary devices to create entry or exit barriers, in order to maintain 33 a community's tax base. Fragmentation also gener- ates a degree of community competition for resources. Community' competition :UT metropolitan areas often involves a large number of cities; in these situations, private property and income are shielded from redistributional activity by inter- community competition. The intimate connection between free location choice (for both firms and households), low redis- tributional activity, private property rights, and federal structural arrangements-~in the form of exclusionary devices--cannot be denied. Collective decisions resulting from free individual location choice are unstable without. a federal structure comprised of zoning, incorporation, annexation, and other rules enforcing local public good outcomes. Conversely, a the existence of a federal structure protects and insures property rights and citizen sovereignty or freedom to choose locations on an individual basis. 34 Postulate 2: Local Autonomy Federal structures also allow for a degree of local jurisdictional autonomy, or autonomy in the decision making process. Autonomy is defined spe- cifically with respect to partitions of (homoge- neous) individual (preferences) into communities matched. with partitions of public good packages into local jurisdictions. When individuals have a limited right to self government, then the span of community decision making autonomy matches their jurisdictional control. Most rights are delegated constitutionally or through. some legislative process. For instance, cities are granted local autonomy by state consti- tutional provisions covering home rule, and by legislative statutes governing local charters. The right to local self government is not a natural or constitutionally protected right; it::h3 a plenary or property right. The legal research discussing the differences between property and natural rights is quite volu- minous.[23] Suffice to say, local charters began 35 as titles to land and natural resources by the King of England. Local charters remain plenary rights. Indeed, Kenneth Shepsle "conceives of the associa- tion between (legislative) committees and jurisdic- tions as a kind of property right."[24] The right to tax resources within a commu- nity's boundaries is a similar collective property right. Without tax resources, the postulate of local autonomy is essentially meaningless. Yet the "point of origin nature of (local) revenue gener- ation gives municipalities a property right to all the resources they can squeeze into their corporate boundaries, while similarly benefiting all those cities that can squeeze out of their boundaries those groups of individuals who put a strain on municipal resources through their redistributional demands."[25] The right to tax assessed value, or point of cmigin revenue generation, is a central part. of local autonomy. The postulate of local autonomy implies a granting of property rights as a precondition for decentralized public good alloca- tive decisions. Suppose local public goods are treated like a 36 standard (consumer-consumer) externality problem. Then an optimal pattern of community property rights solves the joint externality problem. The standard solution, due to Coase, states that "if costless negotiation is pmesible, rights are well specified, and redistribution does not affect mar- ginal values, then (1) the allocation of resources will be identical, whatever the allocation 1f legal (charter) rights, and (2) the allocation will be efficient so there is no problem of externality." [26] In other words, if local autonomy is well defined under a federal structure, then the local public goods problem is solved. Furthermore "proposition two holds even if redistribution does affect. marginal values; and, proposition one holds even if there are transaction costs, provided these are independent of the pat- tern of legal liability."[27] The Coase theorem provides an interesting justification for local autonomy, such as home rule, within a federal structure. A federal structure thus also serves to allocate local public goods and to (re)distribute private and public sector resources. 37 Private property is shielded from redistribu- tional activity' at. the local level by point of origin revenue generation. Public sector resources are limited by those contained. within community boundaries, and by the degree of competition for mobile resources at the local level. Local auton- omy is a two edged sword. On one side resource limitations are pdaced (n1 community jurisdictions redistributional activity (therefore protecting private property from eminant domain), at the same time local jurisdictions are guaranteed an indepen- dent tax base. Since states determine home rule charters (or community rights) they determine the degree of local autonomy. If transaction costs are zero, and resources are mobile, then Coase model poses an interesting alternative solution to the local pub- lic goods revelation problem. In the Tiebout model, individual location decisions ultimately generate a system of local community jurisdictions allocating public goods. But by treating local public goods as externalities, we find that the individual public good demand revelation problem is 38 less significant than federal structural arrange- ments governing the rights and responsibilities of community jurisdictions. Postulate 3: Administrative Efficiency The final desirable criterion for a federal structure concerns generating an optimal pattern of communities and jurisdictions.[28] Recall that communities represent population groupings, or partitions of the total population by preferences; also remember that jurisdictions are partitions of the local public policies into individual and pack- ages of commodities. The principle of perfect correspondence defines exactly what we mean by administrative efficiency. In a classic work, Fiscal Federalism, Wallace Oates argued "the optimal form of federal govern- ment to provide the set of public goods would be one in which there exists a level of government for each subset of the population over which the con- sumption of a public good is defined; this would be sufficient to internalize the ‘benefits from the 39 provision of each good; such a federal structure of government, in which the jurisdiction that deter- mines the level of provision of each set of public goods includes precisely the set of individuals who consume the good, I shall call a case of perfect correspondence in the provision of public goods."[29] Given partitions of total population into communities and the public policy space into juris- diction, perfect correspondence can be defined one of two ways. Perfect correspondence exists if the set of communities and the set of jurisdictions are homeomorphic or if a function (like a federal intergovernmental relation) exists which is a1 one to one napping of communities onto jurisdictions. Perfect correspondence, as a criterion for adminis- trative efficiency, thus requires 1) partitions into communities and jurisdictions plus 2) the existence (NE a federal structure relating popula- tions to policy responsibilities. The perfect correspondence definition of administrative efficiency is derived from George Stigler's argument that "representative government 40 works best. the closer the government is to its consumer-voters, and that subsets of people within a country have the right to vote for themselves different kinds and amounts of public ser- vices.”[30] This idea of a right to local self government or free right to select a location is intrinsic to the argument for an efficient decen- tralized federal structure. An optimal federal structure must satisfy some criteria for administrative efficiency. I have picked the perfect correspondence principle because it subsumes the other two arguments for postulates one and two. Citizen sovereignty and local autonomy both result in an Optimal federal structure. Yet they do not describe a precise institutional design condition. Perfect correspondence describes both the constituent elements of a federal system (ie. partitioning rules defining communities and juris- dictions), and the fundamental intergovernmental relationship(s) governing a federal structure. 41 A Federal Structural Dilemma No federal structure exists satisfying post- ulates one, two, and three.[31] The logic underly- ing the federal dilemma is related to Sen's famous liberal paradox.[32] The method of proof is quite similar. Batra and Pattanaik place four conditions upon a federal structure. First, they presume a restricted profile of individual preference order- ings. If consumers sort themselves out into commu- nities, this condition is easily satisfied. Sec- ond, they postulate the existence of a set of com- munity welfare functions. Third, they presume a set of partitions dividing the pOpulation and policies. Fourth, the perfect correspondence principle is augmented with a variation of a unanimity (or weak pareto) rule. They proceed to show that no social decision function, in this case a federal struc- ture, can simultaneously satisfy all four condi- tions. But how does this result imply the three post- ulates? First, free location choice implies a 42 restricted set of individual preferences. Of course, preferences could be restricted for reasons besides the Tiebout hypothesis. But the citizen sovereignty condition is more general than the axiom utilized by Batra and Pattanaik, (and it was also used by Arrow); yet in the location context it generates a similar equilibrium profile restric- tion. Postulate one is more general (and I would argue more normatively desirable) than the axiom necessary for proof. Second, local autonomy or the decisiveness of community jurisdictions is directly embodied in Batra. and IPattanaik's minimal federal structure. Sen minimal liberalisnn conditicul grants individ- ual's autonomy or right to privacy in on some issues. The existence community welfare functions stems from a degree of local autonomy, where iden- tifiable preference groupings partitions themselves into communities. Finally, if we require perfect correspondence of community jurisdictions, we essentially guaran— tee the existence of intra-community jurisdiction unanimity. Since individuals Ihave :freely' sorted 43 themselves into most preferred communities, and local jurisdictional rights and responsibilities have been appropriately assigned, the expectation is that public good allocations will satisfy strong or weak pareto rules. Thus, no federal structure jointly satisfies the following desirable normative criteria: citizen sovereignty, local autonomy, and administrative efficiency. Clearly, there are other desirable criteria which might be applied to a federal structure. The three postulates justifying a federal structure presented here are only necessary, but not suffi- cient, conditions justifying a: federal structure. But these postulates represent various lines of reasoning (and research) supporting a federal sys- tem of government. Furthermore, they are often utilized implic- itely' or explicitely' as evaluative criteria for recommending federal structural reforms. Taken together, these evaluative criteria are inconsis- tent for judging the relative merits of a federal institutional design. To summarize, I believe, it is when we adopt structural reforms designed to 44 implement these various postulates, that we gener- ate the incoherence, inconsistency, constraints, and overload which are empirical facts describing modern federalism. If we strive for an ideal (or optimal) federal structure, by satisfying these three principles through structural reform, then our federal structure must necessarily fall short of the posited goals, by virtue of their conflic- tual nature. III. Empirical Hypotheses In this next section, I revieW' the recent empirical hypotheses describing modern federalism. The purpose is to show' the linkage ‘between the analytical result presented above, and to illus- trate the connection between the intergovernmental overload and community limitations hypotheses. Not only are these empirical hypotheses two heads of the same coin, but they are also plausible consequences from striving to achieve the preceed- ing normative standards. 45 Community Policy Limitations Community politics is limited politics.[34] A federal structure induces stable community policy choices by limiting the number of alternative strategies and payoffs available to subnational (state and local) governments and their constitu- ents. To quote Paul Peterson, "local politics is not like national politics; by comparison with national politics local politics is most limited; there are crucial kinds of public policies that (subnational) governments cannot execute; they cannot make war or peace; they cannot issue passports or forbid out- siders from entering their territory; they cannot issue currency, and they cannot regulate imports or erect tariff walls."[34] Community politics is limited because a federal structure limits commu- nity decision making power (policy) with intergov— ernmental relationships (transfers). Community policy limitations stem from these formal legal constraints. The United States fed- eral structure limits formal community power. Some 46 policies are simply' prohibited. at the community level, or they are explicitely assigned to national authorities. American federalism structurally constrains community decision processes. The federal structure limits the policy goals community government's pursue. National fiscal resources are 1) scarce and 2) distributed by intergovernmental transfers. Federalism limits community policy performance and output levels by prioritizing grant monies. Grant monies are stra- tegically utilized to create local policy incen- tives. The Structural Constraints Hypothesis The federal structure constrains policy alter- natives to a feasible set of alternatives. National officials utilize rules and resources 11) define a feasible set of policy priorities. A federal structure can thus be used as both a carrot (grant monies) and stick (constitutional provisions) to implement. national policy' goals. National offi- cials determine the contextual domain of community 47 policy strategies and the range of payoffs associ- ated with various policy choices. The structural constraints hypothesis can easily be phrased in game theoretic terms. Both rule and resource limitations structurally con- strain community policy choices. "The federal structure operates to limit choices" I) by imposing some community choices, 2) by limiting the range of alternative policies, and 3) In! determining the incentives facing' community’ decision. makers. To summarize, note that 1) rules of a game can be used to impose outcomes, 2) a limited strategy space occurs when relevent alternatives are separated from irrelevant alternatives and when feasible alternatives are defined, and 3) incentives define payoffs associated with choices.[35] A federalism structurally constrains community policy choices in the player space, the strategy space, and the pay- off space.[36] The player space refers to the number of actors, and the rules of game governing those actors. If players are not sovereign, then an outcome may be imposed or forced. A player's (or 48 community official's) choice is limited the sense that a rule may simply force the selection of a given policy alternative or outcome. A forced choice situation is a game where a solution is imposed upon the players.[37] By limiting the player space, or violating the principle of citizen sovereignty, the rules of «game constrain player choices. Policy alternatives are constrained by limit- ing strategy spaces. Transfer resources constrain policy choices by influencing the number, quality, quantity, and mixture of local public goods and services. Some policies running counter to national policy goals may be deemed irrelevant, by reducing the relative amount of national resources committed to providing a local public good. In other circumstances, declining national resources may reduce the amount of grant monies available for all local public goods, such that many policy alternatives are rendered infeasible. National resources constrain community policy strategies by creating incentives and limiting the feasible alternatives. 49 When local resources are scarce, communities compete for available transfer resources. Decreas- ing own and transfer resources have lead to increasing community competition. Declining local revenue bases, declining intergovernmental grant monies, and increasing local public good demands, increase the degree of community competition. High community competition implies low or constrained community power. Community power, or policy making autonomy, is clearly limited by payoffs. Thus, federalism defines the sovereignty of the player space, the strategy space, and the pay- off space. National legal powers force (n: impose some community policy choices on subnational offi- cials. The scope of community policy agenda is clearly shaped by federal regulations and intergov- ernmental relations. Finally, the relative incen- tives for various policies is greatly influenced by intergovernmental transfers or grant monies. 50 A Synthesis of Empirical Federal Hypotheses Community political leaders make choices within a federal structure. This structure con- strains community power. Structural constraints exist in the form of rule and resource limitations. These limitations are enforced through (increas— ingly overloaded) intergovernmental relationships. Intergovernmental relationships constrain players' sovereignty, the feasible alternative policy strategies, and the payoffs associated with policy outcomes. Each intergovernmental constraint serves to structurally limit community policy choices. At the micro level of the federal system, we find community power limitations. The federal structure induces these limitations through rule and resource constraints. These constraints create and provide for restrictive community policy lati- tude. Furthermore, constraints are increasingly used by national officials to induce community competition. When community competition increases, national officials are best able 11> promote their goals by strategically utilizing intergovernmental 51 grant monies. At the macro level of the federal system, we find that these intergovernmental relationships are increasingly overloaded or strained. Community competition for transfer resources, and the stra- tegic usage of transfer resources as incentives, have lead 11) more complex intergovernmental rela- tionships. The previous hypothesis works from the bottem of the federal ladder up, by' describing micro-community policy incentives. We now turn the organizational complexity hypothesis, which works from the top to the bottem of the federal struc- ture, by focusing' on the complex. organizational processes required for distributing transfer resources to communities. The Intergovernmental Overload Hypothesis Federalism is :3 complex: organizational pro- cess. Some authors maintain that there is "over- whelming evidence to suggest that the current intergovernmental (transfer) system suffers from an overload," as indicated by the increasing state and 52 local dependency on federal aid, increasing spend- ing mandates, and increasing pre-emption of local autonomy through federal regulatory activity.[38] Proponents of the intergovernmental overload hypothesis include David Walker, George Hale, Mar- iam Leif Palley, and other research published by the Advisory Commission on Intergovernmental Rela- tions.[39] The Advisory Commission on Intergovernmental Relations proposes four areas of federal system failure. They are administrative failures such as red tape pmeventing efficient implementation, demand failures sucht as those affecting’ citizen participation and evaluation of programs, and political failures such as those involving local autonomy and accountability. The Commission con- cludes that a federal system failure is occuring which requires structural reforms designed to rationalize intergovernmental relationships, in such a way as to reduce the incoherence and incon- sistency resulting in system overload. Of course, :1 rationalizing structural reform to satisfy these conditions is not possible, 53 because no federal institutional design exists to correct. these: systenl failures. Intergovernmental relationships, and in particular resource trans- fers, have become increasingly complex as communi- ties have become more dependent upon transfer resources. And as own community resources decline and service demands increase, competition for these decreasing transfer resources is increasing. Thus pent up resource demands are overloading the trans- fer system. Increasing micro—community competition results in macro intergovernmental transfer systehl over- load. Macro federal system failure is a consequence of stably increasing community competi- tion, increasing public goods demands at all levels of our society, and decreasing public sector resources at the same time the size and scope of government is increasing. Yet organizational pro- cess models of federalism tend to argue that this failing system has no discernable decision making structure. If our federal system is unhinged, and the federal system needs reform, then it is by no means clear how failures can be corrected. 54 The Failure of Organizational (Systems) Theory Modern federal theory has not progressed beyond Morten Grodzins organizational process model of intergovernmental relationships.[40] Paul Pet- erson suggests, "however apt and appealing the marble cake analogy may be, comparing federalism to a structureless piece of pastry is in the end non- theory; the metaphor suggests flux, change, and complexity when the purpose of theory is to iden- tify simplicity, pattern, and order."[4l] At the core, a federal system is a stable distribution of legal power among independent governments. Organizational process models fail to provide a coherent theory federalism. "The essence of federalism is a stable (institutional) relationship among structures of government."[42] Even though empirical intergovernmental relationships "have been characterized as a (complex) process of shar- ing and exchange," the federal structure itself has changed far more gradually.[43] Given the system failure and general struc- tural irrationality conveyed by organizational 55 theory, one would expect to find a highly unstable disorderly system. Clearly, we find structural changes taking place as the Supreme Court reinter- prets the Constitutional division of powers. As an organizational process, the federal structure imposes stability because the "modern federal theory has replaced the older (dual federalism) concept that each level of government must pursue the functions appropriate 11) it with the flexible idea that the national government can and does exercixe any function performed by state and local government."[44] Modern complex federalism places an emphasis upon overlapping, sharing, and concurrent exercises of functional policy responsibility. The organiza- tional complexity of modern intergovernmental rela- tionships is evidence for the successful adaption of our federal system to emergent problems. Organizational theory fails to provide guide- lines for interpreting how federalism should work. If federalism must satisfy administrative effi— ciency, local autonomy, and (citizen sovereignty, then all federal system must fail. The failures in 56 our federal system today are that there is too much destructive competition among state and local com- munities, and too nmch strategic usage of limited transfer resources as a means for achieving national policy goals. The system is failing because of tremendously increasing service delivery expectations at all levels of government, the decreasing willingness to pay for those services at local levels, the increasing degree of competitive behavior of state and local officials in the grantmenship game, and the decreasing amount of transfer resources being made available by national to local officials. The system is failing because of strategic usage of decreasing resources, by trying too much with too little, not because the system has become too orga- nizationally complex. ' Empirical Structural Incoherence This overuse of organizational complexity denies the fact that federal structure in place today is exists because the old dual federal 57 structure failed to solve civil rights conflicts, to enhance stabilization policy goals, or to pro- vide local public services at the levels demanded by community governments and their constituents. In response to the failure of dual federalism to solve modern policy problems, the federal structure has become much more complex and in many ways more difficult to understand. But this section does however provide a clear link between the community limitations and intergovernmental overload hypothe- sis to the same general phenomena: increasing competition for scarce transfer resources. The prior account of federalism: provides a coherent overview of analytical and empirical fed- eralism theory. The structural incoherence of modern federalism has been described with a number of colorful metaphors. This emphasis on federal organizational behavior, focused on intergovernmen- tal administrative complexity and macro-systemic failures. At the macro level, researcher's hypoth- esize that the modern federal system is overloaded, while micro-level community limitations are fcmnd at the local level. 58 IV. A Political Economy View of Federalism Models Fiscal Federalism Models Fiscal federalism is a complex system of intergovernmental resource transfers. These resource transfers include both national and subna- tional tax and expenditure policies. By concen- trating on tax and spending policy, fiscal federal- ism provides a largely economic federalism theory. But to a great extent, national fiscal policy determines community policy limitations with resource transfer programs. The fiscal federalism literature in economics has evolved within a specific neo-classical welfare tradition. This ("first best") welfare tradition has two important consequences for fiscal federal theory. Fiscal federalism presumes national authority for tax and spending policy and the existence of a single national welfare function, so that distributional policy is a national function. Crucially, economic theories tend to represent federalism as a unidirectional decision making structure where national authorities have primary 59 authority. This primary authority stems from for- mal rules and a belief that only the national gov- ernment should make welfare judgements. A welfare function contains all the relevant datum necessary to set policy goals. Policy goals include equity, efficiency, and stability. Fiscal federalism models generally assign equity and sta- bility goals to national officials, in the form of a single welfare function. Both the distributional weighting of individuals, and the interpersonal comparisons necessary to determine weights, are believed national government functions.[45] National officials are assigned the following important policy making tasks: they must determine the relative priorities of policy goals; they must assign distributional rankings and aggregate indi- vidual preferences according to the policy goals set; and, they must enforce the redistribution of resources, with interpersonal and intergovernmental transfer payments, in order to maximize social well being. Thus, redistributional policy is exclu- sively a national function, because it requires a highly complex operation of goal setting, 6O evaluative comparisons, and enforcement mechanisms to be successful. Criticism of First Best Economic Theories Fiscal federalism is deficient in two impor- tant respects. As a conceptual theory of federal- ism it is deceptively simple. This simplicity comes at a price. The price is that it assumes primitive terms, such as a national public interest (or public goods allocation) and a single welfare function. These primitive terms are not generally refutable. The first problem suggests governments attempt to pursue policies which are in the public inter- est. Yet a single public interest does not exist in a general context. In those few environments where some public interest is clear, there are often competing policies for achieving the public interest. Thirty' years of social choice theory research has proven that even if an unambigous public interest exists, a democratic process may not exist to find it. First best fiscal federalism 61 is deficient because aggregate well-being is poorly defined and prone to multiple interpretations and approaches. The underlying concepts defining fiscal feder- alism are deceptively simple, in contradistinction to the organizational process approach. Yet first best theories rest on contradictory values. Since these values are problematic, fiscal federalism supplies a very limited federal structure. National officials are solely responsible for designing and implementing virtually all policies. All equity policy goal setting responsibility and redistribu- tional activities are given to national officials. Yet equity goals and redistributional policies are pursued and implemented by community officials. Community level redistribution is less effective, and far less elaborate, than national equity goals for many reasons. However, some redistributional (and stabilization) policies are undertaken at the community level. Given increasing community competition, and decreasing transfer resources, we would also expect that decreasing redistributional activities at the 62 community level. Also, bargaining versus unilat- eral dominance is rule, rather than the exception to intergovernmental relationships. The simplicity of (first best) fiscal federal models run counter to the existing strategically competitive environ- ment. Experience dictates a political economy inter- pretation of federal structures, sensitive to real world competition and strategizing to obtain resources. A. political federalism: theory should allow for equity and efficiency decisions to be set by national and subnational officials. A Political-Economic Model of Federalism Social choices produced In? fiscal federalism are determined by national authorities whereas political federalism produces social choices through intergovernmental bargaining. The inter- governmental bargaining process is an elaborate system of community jurisdiction service delivery. Through those shared and concurrent powers, a fed- eral structure distributes legal powers in order to 63 supply increasing public good demands. Even though local autonomy exists under polit- ical federalism, community policy discretion is limited, since political federalism specifies a game theoretic, rather than economic decision theoretic, context. Community decision making is limited because national authorities dominate national-subnational bargaining relationships. The intergovernmental bargaining process has rules which favor national agents, national agents have more resources, and local actors compete with each other. Community officials may set equity goals. But there is a penalty for straying from national priorities. Other communities will bene- fit by attracting resources from natioal authori- ties and by using incentives to attract resources from a community which attempts to set an indepen- dent course of action. In contrast, fiscal federalism severely restricts local autonomy in order to set equity goals, and to prevent destructive community compe- tition. Redistributional and stabilization poli- cies are considered primarily national functions. 64 The national government priorities are represented by the existence of a single social welfare func- tion. Political federalism allows for multiple com- munity welfare functions. Each community may establish equity, efficiency, and stability goals consistent with the constitutional division of power. This intergovernmental bargaining process is shaped by national grant monies and community competition for transfer resources.[46] Political federalism consists of competitive or non-cooperative bargaining relationships. Each local government sets relative standards, makes interpersonal comparisons, assigns distributive rankings, and then enforces policies at the commu- nity level. The role of welfare judgements also differ in fiscal and political federalism models. In fiscal federal models, national autorities determine the public interest. The national government has pOpu- lar sovereignty to determine the public interest from individual values directly, rather than via community polities. Once public good demands are 65 expressed, national authorities design and imple- ment policies. Modern political federalism is a kind of com- pact cn: social contract: theory. Communities set policies which correspond to competitive and stra- tegic: political realities. Vertical intergovern- mental relationships tend to allow the higher part- ner to dominate the lower by having rule and resource advantages. Horizontal relationships tend to be non-cooperative as equals tend to compete. In the extreme, horizontal relationships are destruc- tively competitive. National officials aggregate both. community ‘welfare functions, and individual preferences. The nature and form of this aggrega- tion mechanism is quite complex, given the enormous difficulty of coordinating strategic behavior, community competition, service delivery, and redis- tributional activities. 66 V. Conclusion The Purpose of Institutions Institutions serve many purposes. Institu- tions determine how individuals reveal preferences for policy alternatives. And institutions effect collective policy outcomes. The existence of cer- tain institutions, such as federalism, has a stabi- lizing effect upon both state and local political processes and outcomes. A federal structure per- forms this stabilizing function by dividing power through vertical and rmuizontal intergovernmental relationships. The fundamental social choice problem is to fairly aggregate individual preferences into stable group choices. Direct aggregation mechanisms, like pure majority rule, cannot be relied upon to pro- duce consistent group choices. Pure majority rule institutional arrangements have proven to be unst- able for translating diverse individual preferences into multidimensional policy choices. In certain situations, indirect aggregation mechanisms can generate minimally consistent group 67 choices. New institutionalism is the study of how contextual factors, like structures, rules, and procedures affect individual demand revelation and the stability of group choices. By structurally imposing group choices in situations where direct majority preference exists, we can partially solve Arrow's paradox. Of course "institutional features...do nothing to mitigate the (nonexistence) results of Plott, Cohen, McKelvey, Schofield, Schwartz, and Slutsky; it is still true that the majority preference rela- tion is ill-behaved, and this instability underlies and affects ultimate (group) choices."[47] In the absence of stable preference induced equilibrium, institutions temper the cyclicity of pure majority rule. Structures solve the voter paradox by relaxing (or violating) Arrow's seldom cited nonimposition or citizen sovereignty condition. Representation systems, veto hierarchies, and other organizational structures all impose stability' on political adjustment processes and outcomes. But the nonex- istence of social welfare functions under general 68 conditions, and the global instability of pure majority rule, pose fundamental contradictions for democratic institutions, whose rmumative signifi- cance cannot be solved totally by representative imposition of group choices. Federal System Stability A federal structure is a complex set of inter- governmental rules and resource transfers mechanism for aggregating diverse community interests into public policies. The purpose of federalism is to provide a stable institutional framework for making political choices, in environments where community interests are unclear, individual demands are div- erse and changing, and where all interests are often times conflictual. By structurally dividing power, a federal system reduces the instability and scope of direct collective choice rules. A unitary structure, where a single heterogeneous national community has global policy authority, is an unstable means for determining group choices. A federal structure 69 induces stability in? creatingr semi-autonomous (homogeneous) communities, and by transferring resources in order to influence local jurisdic- tional policy choices. Federal system stability comes about because a federal governmental structure divides, checks, and balances the formal political power of elective officials and citizen-voters. In this sense, a federal structure imposes stability by limiting the sovereignty of (national and community) leaders and their constituents ability to unilaterally or directly translate individual preferences into group policies. A federal structure induces sta- bility by limiting the scope of conflict through intergovernmental checks and balances. CHAPTER TWO ENDNOTES Diamond, M. "What the Framers Meant by Feder- alism," in Robert A. Goldwin (ed.). A_Nation of States. (Chicago: Rand McNally, 1974). Shepsle, K.A. "Institutional Arrangements and Equilibrium in Multidimensional Voting Mod- els," American Journal of Imditical Science, 1979, 23, pp. 27-59. Riker, W.H. Federalismu Origin, Operation, Significance. (Boston, Mass.: Little, Brown and Company, 1964), p. 137. Shepsle, K.A. and B. Weingast. "Structure Induced Equilibrium and Legislative Choice," Public Choice, 1981, 37, pp. 503-519. Peterson, P.E. City limits. (Chicago: The University of Chicago Press, 1981). Wright, D.S. Understanding Intergovernmental Relations. (North Scituate, Mass.: Duxbury Press, 1978), p. 121. Batra R.N. and P.K. Pattanaik. "On Some Sug- gestions for Having Non-binary Social Choice Functions," Theory and Decision, 1972, 3, pp. 1-110 Arrow, K.J. Social Choice and Individual Val- ues. (2nd. ed. New York: Wiley Press, 1963), pp. 28-30. 7O 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 71 Kramer, G.H. "On a Class of Equilibrium Con- ditions for Majority Rule," Econometrica, March 1973, 41, pp. 285-297. Mueller, D.C. Public Choice. (New York: Cam- bridge University Press, 1981), p. 128. Tresch, R.W. Public Finance: A Normative Theory. (Plano, Texas: Business Publications Incorporated, 1981), p. 598. Oates, W.E. Fiscal Federalism. (New York: Harcourt. Brace .Jovanovich, Inc., 1972), pp. 34-35. Frohlich, N. and J. Oppenheimer. Modern Political Economy. (Englewood Cliffs, N.J.: Prentice-Hall Foundations of Modern Political Science Series, 1978), p. 22. Tiebout, C.M. "A Pure Theory of Local Expen- ditures," Journal of Political Economy, Oct. 1956, 64, pp. 416-424. Ibid. p. 416. Ibid. p. 424. Buchanan, J.M. and C.J. Goetz. "Efficiency Limits of Fiscal Mobility: An Assessment of the Tiebout Model," Journal of Public Econom- ics, April 1972, 1, p. 26. (25-43) Tiebout (1956: p. 424) Bewley, T.F. "A Critique of Tiebout's Theory of Local Public Expenditures," Econometrica, 49, p. 739. Fisch, 0. "Optimal City Size, The Economic Theory of Clubs, and Exclusionary Zoning,” Public Choice, Winter 1975, 24, pp. 59-70. Tiebout (1956: p. 424) 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 72 Mueller, D.C. Public Choice. (New York: Cam- bridge University Press, 1981), p. 129. Miller, G.J. Cities by Contract: The Politics of Municipal Incorporation. (Cambridge, Mass.: MIT Press, 1981), p. 39. McCarthy, D.J. Local Government Law. (2nd. ed. St. Paul, Minn.: Nutshell Series of West Publishing Company, 1983). Shepsle (1984) Miller (1981:p. 130) Layard, P.R.G. and A.A. Walters. Microeco- nomic Theory. (New York: McGraw-Hill Company, 1978), p. 192. Ibid. (1978: p. 192) McGuire (1974) Oates (1972: pp. 34-35) Stigler, G. "Tenable Range of Functions of Local Government,” Federal Expenditure Policy for Economic Growth and Stability. (Washing- ton D.C.: Joint Economic Committee, Subcommit- tee of Fiscal Policy, 1957). Batra and Pattanaik (1972) Sen, A.K. Collective choice and Social Wel- fare. (San Francisco: Holden-Day Pub., 1970), pp. 87-88. Peterson (1981: p. 3) Peterson (1981: p. 4) Peterson, P.E. "A Unitary Model of Local Tax- ation and Expenditure Policies in the United States," British Journal of Political Science, 1979, 9, pp. 281-314. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 73 Friedman, J.W. Game Theory with Applications to Economics. (Oxford: Oxford University Press, 1986), pp. 23-24. Miller (1981: p. 165) Walker, D.B. Toward a Functioning Federalism. (Cambridge, Mass.: Winthrop, 1981), pp. 8-9. Hale, G.E. and M.L. Palley. The Politics pf Federal Grants. (Washington D.C.: Congressio- nal Quarterly Press, 1981). Grodzins, M. The American System, ed. by D. Elazar. (Chicago: Rand McNally, 1966). Peterson (1981: p. 13) Ibid. (1981: p. 13) Hamilton, C. and D. Wells. "Viewing Federal- ism as a Complex Political Economy: The Macro Social Functions of FCfleralism," Paper Pre- sented at the Annual Meeting of the Sourthwes- tern Social Science Association Meeting at the Dallas Hilton, March 18-21, 1987. Walker (1981) Tresch (1981: p. 597) Ingram, H. "Policy Implementation Through Bargaining: The Case of Federal Grants-in- Aid," Public Policy, Fall 1977, 25, pp. 501-526. Shepsle (1979: p. 59) CHAPTER THREE THE HOME RULE REFORM PARADOX 1. Structural Reform Strategies Progressive Political Strategies The political strategy of reform is to play upon some paradox which an incumbant coalition can not resolve under a new set of rules without inter- nal conflict. Structural reformers of any era attempt to construct a set of rules in the form of "a trap so clever that no matter how the opponent responds, the response itself would give a future victory."[1] Losing coalitions can either introduce new issues or change the rules of the game in order to divide winning coalitions. Internal strife among the winners will then lead to a new winning coalition forming thus upsetting status quo insti- tutions and policies. Revolutionary local institutional changes took place during the progressive reform era. Progres- sive municipal reformers sought to gain control of local government from political party machines. 74 75 Radical institutional changes 'were .necessary' for the reformers to upset the status quo. The pro- gressive reformers' political strategy was to change institutional arrangements in order to place party machines at an electoral disadvantage. The widespread institutional changes of the era are a testimony to the progressive reformers' political success. Examples of progressive reforms included 1) shorter ballots 2) primary elections 3) non partisan elections 4) at-large legislative constituencies 5) civil service criteria 6) appoin- tive local executives 7) improved budgeting and accounting’ systems, enui most importantly, 8) expanded state provisions for local self govern- ment. While many of these reforms were instituted, progressive candidates often failed to win elec- tions. But this ladk of electoral success cannot detract from the reformers lasting impact on commu- nity politics. The progressive reformers were not adept can- didates. Their amateur View of political leader- ship and institutions precluded personal success. But they were highly effective political 76 strategists because they changed the rules govern- ing local politics. Progressive reformers struc- tured the rules of community politics so they could upset the status quo party machines and begin implementing policies. Progressive reformers were able to build a winning' coalitjtui by' one, fostering inter-ethnic group conflicts, thus dividing the immigrant voter bloc; and two, by asserting that party machines could not "maintain public order, protect private property, and improve the urban quality of life"--platforms that mobilized middle class voters to join the progressive coalition.[2] Expanded state provisions for local self gov- ernment, home rule, created a dilemma for party machines because home rule increased local auton- omy. Increasing autonomy benefitted all local governments, including those: cities dominated by party machines. However party machines were depen- dent upon the state legislature for enforcing their policies over areas not controlled by the machines. Once all communities had greater local autonomy over internal affairs, no central city could 77 dominate across jurisdictions. Party machines depended upon immigrant voters. The fragmentation resulting from home rule created political opportunities for immigrants by increas- ing the number of cities. With home rule immigrant voters had more service options available. Frag- mentation also provided opportunities for ethnic leaders and voters to exercise direct influence over local government. This enabled some communi- ties to achieve an equal position within existing party' machines while others ‘were able to (assert independence. Prior to home rule, party machines utilized selective incentives and exclusionary devices to form winning state legislative, central core city, and metropolitan area coalitions. But the combina- tion of a organized urban middle-class and diverse ethnic groups, created the potential for winning reform coalitions to emerge. Progressive reformers played upon both intra-city and inter-community differences by formulating successful election and administrative reform proposals under the aegis of home rule. 78 The machine era is normally described as cor- rupt, yet administratively and electorally stable. However no unitary party machine "captured" local government. Party machine leadership and organiza- tional structures were always changing. But party machines did take advantage of existing rules and resources to win in an unstable environment. "It (was) not that the foreign born voter... preferred to vote...blindly (for) some politi- cal faction. If (the immigrant voter) had too often proved to be the tool of the exploiter, it (was) largely because a system of partisan (electoral rules) had given the latter a redoubtable position."[3] If uniform immigrant support for party’ machines had existed, progressive reforms would not have had widespread voter appeal. Progressive Influence on Community Organization The governmental structure in existence prior to home rule favored vertically integrative politi- cal organization. Furthermore, given the horizon- tal concentration (or central city monopoly provi- sion) of local services, immigrant voting blocs, and special legislative acts were necessary to 79 maintain integrative organizations. Party machines were successful because they were integrative orga- nizations. For the progressive organizations to begin winning, they needed to reform structures in order to produce a different political environment. "And this is what political strategy is all about, structuring the world so you can win."[4] The new structures created an environment with the follow- ing characteristics .being’ important 1x) coalition formation: large numbers of local service options, middle class voters, and general law charters. Once home rule produced a new environment, fragmented progressive style local organizations were required to win elections and to implement policies. Thus the progressive strategy was suc- cessful even though progressive candidates did not always win. Party machines did continue to win elections, but policies were made within progres- sive structures. The ‘progressive (coalition used home rule to create an environment favorable to their style of political organization. Thus progressives attacked party machines from 80 two directions. First, home rule reduced access to outside resources (i.e. state money) and limited direct state intervention in local decisions. Sec- ond, home rule allowed reformers to attack machines within a community by changing both the local elec- toral and service rules. The following is a histor- ical review of home rule instituted by the progres- sive coalition in Michigan. II. The Origins of Metropolitan Fragmentation Special Act Incorporation: 1805-1908 From 1805 to 1908 the Michigan State Legisla- ture had the power to pass special legislation affecting local governments. City incorporation acts were one form of special legislation. In "Michigan's first constitution, in effect from 1835 to 1850, (there was) no mention of the organization and incorporation of cities. During this period no cities and only five villages were incorporated; Detroit had been incorporated in 1815 by a special act of the territorial legisla- ture."[5] In the period from 1850 to 1895, under a 81 new constitution, 89 cities and 297 villages were incorporated by special acts of the state legisla- ture.[6] The party machines in Michigan were influen- tial from around 1875 to 1908 under the 1850 Con- stitution. Party machine influence was bolstered by special act legislation. Special acts bene- fitted the party machines by permitting state monies to be allocated on a case by case basis for local projects. Progressive reform organizations began, in part, as a protest to the corruption associated with direct state legislator involvement in local projects. Home rule altered this relationship between state legislators and cities by augmenting local autonomy. With the expansion of local auton- omy cities were better able to resist annexation. The Threat of Annexation: The Delray Proposal In 1905, the City of Detroit annexed 22,000 people from unincorporated territories: 9,000 resi- dents of Delray Village, 7,000 in Woodmere Village, 82 and 6,000 peOple from Springwell Township.[7] The area contained over ten million dollars worth of property, including twelve factories and brick yards. The area lay southwest along the Detroit river. The Delray annexation involved a reform mayor in Detroit and a suburban party machine. Progres- sive reformers supported. annexation :hi order to improve local services to the middle-class Woodmere area and 11) break the Delray machine. Detroiters also favored annexation as a means to increase the industrial tax base. Suburban communities per- ceived the annexation as a threat to their politi- cal independence and economic resources. For 1905 the annexation of 22,000 people was a dramatic political event. The case luni a greater symbolic than practical effect. The annexation signalled. Detroit's proclivaty' for' expansion. Given the uncertain future for progressive rule in Detroit, and the potential suburban tax base losses associated. with. industrial strip» annexation, the Delray case revealed the credibility of Detroit's annexation strategy. After Delray, suburban cities 83 premised their incorporation decisions upon a cre- dible annexation threat. In a: front page story titled, "Delray Failed Them" The Detroit Times described the wild elec- toral scene: "In vain did the 'antis' of Delray fight against the odds established in Woodmere. The river town was expected to pull a majority of 400 against the annexation to offset the known predilections of Woodmere and Springwells (township), but this anti-majority dwindled to 79. "Rah for Jimmy Burns, and to hell with annexation," yelled one of the 2,000 in the middle of Woodmere road with his burden of beer and sorrow. Confiding habitues of the road houses spoke feelingly of "Jimmy" Burns and his decaying power in the pastures. "Just six more months boys, whooper up," (as) all along the west end pike that joins the three villages could be heard the merry round of the slot machine."[8] Between the unincorporated territories of Woodmere Delray and Springwells Township, the combined vote was 1124 to 610 in favor being annexed to Detroit. Both Detroit residents and the combined vil- lage population were allowed to vote on annexation. Delray and Woodmere village and Springwells town- ship residents' votes were added together. Hence a faction within a single community could pass (or reject) annexation proposal for several 84 jurisdictions' residents ‘within sum affected area under the 1850 Constitution. Annexation to increase the industrial tax base created strife within the progressive coalition. Suburban progressives feared the precedent set by Detroit expansion. Suburban communities faced tax base loss, and the prospects of being dominated by political leaders and cmganizations responsive to Detroit constituencies. Furthermore, the suburban leadership all faced a loss of political influence. Annexation by Detroit would have eliminated a num- ber of local township supervisor, village presi- dent, and small city mayor jobs. The Delray annexation divided urban progres- sives from small town and suburban progressives. The Delray incident mobilized the structural reform wing of the progressive coalition. Political lead- ers on all sides began calling for home rule to limit future annexation attempts. After the Delray annexation experience, the 1908 Constitutional delegates. provided for three separate voting constituencies on annexation pro- posals. First, the residents inside a: proposed 85 annexation voted separately. If residents are contained in three local political units (i.e. a village or township), then three independent elec- tions are held for residents within a proposed annexation. Second, residents in the city propos- ing annexation vote as a unit. Third, the resi- dents inside a political unit not included but affected by a proposed annexation must also vote. The 1908 Constitutional home rule provision limited annexation by requiring annexation proposals to be approved by all three separate constituencies. An annexation fails if one constituency rejects the proposal. The 1908 voting rule would have changed the Delray annexation outcome. First, Delray and Wood- mere village and Springwells township residents' votes would have been counted separately. Since a majority of Delray residents voted against the proposal, they would not have been annexed to Detroit. Second, the remainder of Springwells township residents outside the proposed area would have had to vote on the annexation because Delray and Woodmere village residents were contained 86 within Springwells township. Even though Woodmere village anui Springwells township residents inside the proposed annexation area voted to join Detroit, Springwells township residents outside the area could have vetoed the whole annexation by voting against the proposal. After the 1908 Constitution, proposed city annexations had to obtain majorities in three separate voting constituencies. The 1908 Constitutional Reform Strategy In the 1908 Convention, progressive reform organizations changed the rules of local politics to put party machines at a strategic disadvantage. The progressives passed articles which were vigo- rously' protested by party machines because they placed machine organizations at strategic electoral and policy disadvantages. Progressive officials gained an advantage because they controlled local government in many suburban areas, and virtually all the rural local governments in the lower penisula. Machines won local elections in a few of the largest cities, 87 while progressives dominated in a large number of small cities and townships. Progressive (republican) reform organizations benefitted from increasing home rule. They were given independent autonomy to make policy decisions at the local level, at the same time governmental fragmentation decreased the influence of partisan machines. Local officials and Republican party members still benefit from home rule. Greater autonomy and fragmentation :us generally tx> the Republican Party's advantage because they control a larger number of small cities and a majority of the (par- tisan) elected township positions. Passage of home rule at the 1908 Constitutional Convention was not neutral ‘with. respect tx> political organization's electoral and policy strategies. Central Core Annexation Votes from 1905-30 Detroit's legal boundaries .have run: changed since 1926 when Detroit tried to annex Macomb County residents. Table Iifil describes Detroit's 88 annexation votes from the Delray proposal in 1905 to the cross-county Hazel Park Village annexation prOposal of 1926.[9] The percentage favoring annexation are broken down for city and township voters residing inside or outside the territory proposed for annexation. TABLE 3.1 VOTE PERCENTAGE IN FAVOR OF DETROIT ANNEXATION BY CONSTITUENCY AND YEAR YEAR DETROIT% TOWNSHIP INSIDE% OUTSIDE% 1905 65 1915 82 68 *82 3O 83 87 83 83 83 86 83 86 1917 83 83 59 *87 3O 63 87 79 80 83 65 62 87 86 77 87 84 76 1918 86 78 64 1922 71 90 56 1926 67 85 70 *-- 78 32 Average .82 .74 .64 Deviation .06 .18 .13 Skewness -.44 -.41 -.35 Kurtosis .33 .27 .39 89 Thirteen of the seventeen proposals occured from 1915 through 1918 during World War I. The average Detroit support for annexation. was 82%, fully thirteen of the fifteen Detroit elections were over 80% in favor of annexation. In these fifteen elections, Detroit residents' favorable support for annexation was all within plus or minus 6% of the 82% average. Consistent Detroit support for annexation does not fall until the 1922 and 1926 proposals invol- ving fringe villages located far from residential Detroit areas. The simple correlation between Detroit annexation vote's and time is equal to -.64 indicating moderately strong negative downward trend. Detroit support for annexation dropped to 76% and 67% because the proposals required higher costs for extending services to residents in Red- ford Village and Hazel Park Village. The land between Detroit and these two villages was largely rural with few services or industrial and commer- cial activity. The less consistent support for these proposals was caused by city resident's unwillingness to expand into high service cost and 90 low property value territories. Voter support for annexation within the terri- tories differs slightly from the Detroit prefer- ences revealed. The average support for annexation was 74% for residents within a proposal area. Hamtramck Village residents rejected Detroit annex- ation proposals in 1915 and 1917. In the 1917 election, Hamtramck voters rejected the proposal by a 70% margin while the rest of Hamtramck Township approved the detachment by a 63% vote. The stan— dard deviation for territories being annexed was four times larger than the average deviation for the Detroit ‘votes, indicating :1 less consistent support for annexation by affected territory vot- ers. The average level of support among township voters for detachment was 64% with a standard deviation of 13%. The number of voters residing in a township, but not included in the proposed annex- ation, was usually very small. Voters within a proposal normally lived within a a subdivision constituting a small part of the township land. Voters within a proposal were essentially the first 91 suburban residents in metropolitan Detroit while detachment voters lived in small (agricultural) villages. Further statistical evidence suggests that only minor differences in city, suburban, and small village resident support for Detroit annexation existed. A nonparametric test for sample indepen- dence is appropriate given the lack of information about individual voter's preferences for (or against) annexation. The Kruskal-Wallis test for sample independence is equal to 15.22 which is approximately' equal to :1 chi-square distribution with two degrees of freedom with a probability value of .0005.[10] The observed K-W test value rejects a null hypothesis that the annexation vote results are drawn from statistically independent populations. Voting occured under legally indepen- dent. constituencies jyet 13m: aggregated electoral outcomes indicate a degree of underlying preference homogeneity supporting Detroit annexation. In only one situation did an area favor annex- ation while the remainder of a township voted against. detachment. Hazel Park. Village approved 92 Detroit's annexation proposal 85% to 15% while the rest of Royal Oak Township voted against detachment 68% to 32% in 1926. Since this annexation proposal cut across Wayne into Oakland County, Royal Oak Township resident's feared the potential future ramifications of cross-county annexation. The Hazel Park Village defeat and the subsequent Warren Township (Macomb County) court challenge ended further cross-county Detroit annexations. The 1922 Gratiot Township and the 1926 Redford Village annexations extended Detroit's boundaries III a northeastern and northwestern direction to their present locations. Both of these areas have recently attempted 1x) detach. themselves from Detroit in order to be joined with the City of Harper Woods and Redford Charter Township. During this period of Detroit annexation, Highland Park became the first suburban Detroit village to incor- porate in 1917 under the 1908 Constitution. Ham- tramck Village incorporated as a city in 1922. Thus, within a decade the strategic impact of home rule limited Detroit boundary expansion. Highland Park and Hamtramck responded to Detroit's 93 annexation threat by defensively incorporating, Warren. Township: prevented. future: cross-county Detroit annexation, and time last Detroit boundary expansions occured with lowered levels of support. Defensive Suburban Home Rule Incorporation: 1918-68 From 1918 to 1940 sixteen cities incorporated on Detroit's boundaries. Twelve cities of these cities incorporated in Wayne County. This series of municipal incorporations effectively blocked off all future Detroit boundary expansion. From 1945 to 1968 thirty eight more suburban cities incorpor- ated. At the same time the City of Detroit's pOpu- lation decreased from two to nearly one million residents. The post World War II housing supply expansion increased development of commuter municipalities. With the housing supply expansion, new communities could incorporate their property tax base and pro- vide low cost local public services. Suburban incorporation granted these "municipalities a prop- erty right to all the resources they (could) 94 squeeze into their corporate boundaries, while similarly benefiting all those cities that (could) squeeze (nu: of their' boundaries those. groups of individuals who put a strain on municipal resources through their redistributional demands."[11] These second tier municipalities were both industrial and commuter suburban communities. None of these municipalities shared boundaries with the central core City of Detroit. Second tier munici- pal incorporations surrounded the older (first tier) suburbs coterminuous ‘with..Detroit. Second generation suburban incorporations reflected a desire to block strip annexation threats from older suburbs and a need to provide more services to fringe metropolitan residents. The 1908 Constitution had encouraged municipal incorporation by providing low entry barriers to potential cities and by providing restrictive bar- riers to existing cities. Existing cities could not easily manipulate their territorial boundaries to reflect service demands and delivery areas. Home rule did grant local self government to all general purpose municipalities. Municipal entry 95 was not highly regulated so that community forma- tion became a local rather than state legislative decision. Thus home rule encouraged large scale defensive incorporation. and limited all existing municipal changes by assigning each city a property right-annexation veto. III. State Regulation of Metropolitan Fragmentation The 1949 Charter Township Act Until 1949, Detroit expansion in northeastern and northwestern directions was still feasible. After all, strips of both Gratiot and Redford town- ships had already been annexed into Detroit. Fur- ther Detroit encroachment would have involved vir- tually all of the remaining township areas. Urban township residents' all across the state demanded some rule to protect these areas from being "swal- lowed up" in either a single large proposal or a series of small proposals. Township officials and residents also were increasingly concerned about the structural limita- tions constraining township government in all areas 96 of the state. The Michigan Township Association lobbyied the Legislature for structural reform. The legislators' response was the 1949 Charter Township Act. The 1949 Charter Township Act had very little initial impact upon Detroit metropolitan fragmenta- tion. Warren Township became the only township in the whole state to approve "charter" status in 1949. Only three charter townships were in existence before time 1961 Michigan Constitutional Convention. But the charter township act did serve a fundamental purpose: township residents could deliver more services and charge higher taxes to pay for them. Townships could charter themselves in response to the threat of strip annexation. But without a means to satisfy resident service demands at low prices, they could not hope to compete against existing municipalities in an annexation election. The three urban townships adjacent to Detroit all took action to prevent further annexation. Gratiot township incorporated as the City of Harper Woods, and Redford and Warren Townships became 97 charter townships. The Charter Township Act increased the number of choices available to unin- corporated residents. The Charter Township Act was intended as a way to modernize urban township ser- vices yet the intent was to bloc munipal consolida- tion in metropolitan areas. Home rule chartering for townships was an incremental reform procedure for slowly altering local governmental structures. Urban townships were supposed to use the Act to change from rural to urban governments. The State Legislature intended the chartering process to create a addi- tional halfway point between rural township and home rule urban city government. The long term effect has been to incorporate all urban townships as (minimal) fiscal policy actors without requiring them to go through a state regulated incorporation process. The Charter Township .Act (provided the option of defensive incorporation to any (formally unincorporated) township residents threatened by strip annexation. At the time of the Act, there were 1,200 townships in Michigan but only 100 con- tained populations over 5,000. Most of the heavily 98 populated townships were located in metropolitan areas. The State legislature strengthened the Charter Township Act prior to the 1961 Michigan Constitu- tional Convention. The degree of fiscal policy discretion and organizational flexibility was increased with the 1960 revision. In the Detroit metropolitan area many townships had already become incorporated suburban municipalities. But most second tier cities had not followed township bound- aries: for instance, Ecorse township became the cities of Ecorse, Melvindale, Lincoln Park, Allen Park, River Rouge, and Woodhaven. According to the 1960 census, 200 of the 1200 townships in Michigan contained over 5,000 resi- dents. Between 1960 and 1977 18 townships became chartered under the revised Act. The remaining urban townships in the metropolitan Detroit area are all charter townships. During the 1970's Red- ford Township became the largest "minimal" city -township- government in the country with over 80,000 people. The Charter Township Act of 1949 and it's 99 subsequent revisions provided a compromise. Charter townships can offer more local public goods and charge higher tax prices than rural (unincor- porated) townships. And these tax prices and ser- vice offerings purportedly cost less than what residents would pay if they became a city. Town- ships "which choose this form of government do not actually' adopt. a charter as cities or counties may do. The Act itself is a charter, and it requires the same organization of gov- ernment as for general law townships. Charter townships do have additional taxing powers and may levy up to five mills without a vote of people...They have a better credit rating for borrowing, the same ordinance-making power as cities, and more flexibility in administra- tion. The more urban townships in Michigan have opted to become charter townships because this increased flexibility facilitates provi- sion of the urban kinds of services required by their constituents."[12] In a single act the Michigan legislature incorpor- ated a whole third tier of minimal cities. By passing this law, the state legislature institu- tionalized a continuoum of home rule which subsequently increased metropolitan fragmentation. At this point it is worthwhile to note that a highly differentiated metropolitan governmental 100 structure had evolved in Michigan's urban areas. Laizzez-faire community entry luul produced first, second, and third tier cities. The effort to increase inter-community competition by increasing structural fragmentation.‘was a: stunning' success. The first tier consisted of areas blocking central core expansion; the second tier defensively incor- porated to block the threat of first tier strip annexation and to provide services; the third tier involved no formal incorporation, but was motivated by a desire to protect existing township boundaries and increase services without increasing taxes. The cumulative impact of unrestrictive state regu- lation (including the state's recognition of town- ships) produced aa fragmented metropolitan govern- mental structure. The State Boundary Commission Act in 1968 The 1908 Constitutional home rule provision had decentralized community formation decisions. Enforcement of state entry barriers limiting munic- ipal incorporation was very low. Low entry 101 barriers and decreasing community formation costs enabled many suburban communities to take advantage of the home rule tax and service options. Home rule incorporation lead to a proliferation of com- munities in metropolitan areas. The State Legisla- ture responded to calls for metropolitan reform by passing the Boundary Commission Act. The Boundary Commission Act established an agency to regulate new community formation and to arbitrate existing community boundary adjustments. The Commission's primary function was to regulate incorporation rather than annexation decisions. The Act stated that the Boundary Commission "must review and approve petititons proposing the formation of a new city or village. If the Commission finds that the petitions are in order, it must then set a date for a public hearing to be conducted in the affected area within 180 days. Thereafter, on the basis of a number of criteria provided in the act, the Commission may take one of three actions. It may: 1) deny the petition for an election on incorporation to be held, thus terminating the case; 2) approve the petition and order the election to be held; 3) revise the boundaries of the proposed incorporation and order an election on the incorporation in accordance with the revised boundaries." [13] The principal focus was on new municipal 102 incorporation proposals. The Boundary Act granted the Commission full responsibility for implementing (strict) regulations governing incorporation deci- sions. But the Commission was not given the same authority over annexation decisions. The Boundary Commission was cast in an advisory role, in the sense that it could provide technical information or arbitrate non-binding solutions. The Boundary Commission thus regulates new .incorporations and provides advice on annexation decisions. By 1968, most boundary disputes involved charter township incorporation and annexation deci- sions. Third tier boundary disagreements affected fringe residents and township officials dispropor- tionately. Given the status quo advisory role of the Boundary Commission, it is perhaps not surpris- ing that township residents perceived that the Commission. was ibiased against them sh: favor' of second tier municipal incorporations. The Commis- sion had more control over community entry rather than. community' boundary' decisions. The Boundary Commission strictly limited charter township incor- poration into cities at the same time existing 103 cities began proposing strip annexations of indus- trial and commercial land developments in charter townships. The legislation's intent seemed to favor strict regulation of incorporation while providing only limited oversight of annexation. The idea was to limit the number of new cities and provide a mechanism for arbitrating the large num- ber of metropolitan annexation disputes between second tier suburban cities and the minimal city- charter township fringe areas. The first case brought to the Boundary Commis- sion involved three urban townships bordering the City of Flint. Mount Morris, Burton, and Flint townships all wished to incorporate. The City of Flint had annexed large strip of property cutting Flint township into two discontiguous pieces. Flint had also strip annexed parts of Mount Morris and Burton townships. The Boundary Commis- sion report stated that 104 "annexation by the city of Flint has been highly selective. The only territories which the city of Flint has been able to annex are primarily high value commerical and industrial properties in which there were no residents. In spite of several attempts to annex terri- tory of diversified land uses and in spite of affirmative votes by the voters of the central city, each such annexation proposal has been rejected by the voters of the suburban commu- nities." [14] The Boundary Commission report included exten- sive service benefit and cost analysis associated with incorporating a new city in the Flint area. The Commission redrew the potential community boundaries before it allowed residents to vote on community boundaries. The Commission rejected one proposed incorpo- ration, and revised the potential boundaries for the other two. One of the two revised petitions was defeated by voters under the redrawn bound- aries. Voters approved the incorporation of Burton township into the City of Burton under the Commis- sion revised boundaries. All three proposals were altered by the Boundary Commission before they were submitted to community residents for electoral approval. In this first controversy the State 105 Boundary Commission provided explicit cost, bene- fit, externality, and discounting criteria for regulating community incorporation decisions. After Burton's incorporation there were no major incorporations from 1968 until 1977. Some villages were incorporated within rural townships during this period but these decisions were inconsequential for metropolitan structures. Of course most primary and secondary defensive incor- porations had already taken place in the metro- Detroit and outstate metropolitan areas. Since the Boundary Commission could IKM: retroactively regu- late community formation or boundary adjustment decisions, many controversial incorporation and annexation decisions were not subject to Commission oversight or regulation. Between 1971 and 1984, "the Boundary Commis- sion processed 294 requests from cities that wanted to annex land, and it approved 182, or more than 60 percent of them."[15] Many existing community strip annexations did not require a vote because less than 100 residents were affected. The most difficult; political conflicts ibrought. before the 106 State Boundary Commission involved second tier Detroit suburban strip annexations of third tier urban charter township> commercial and industrial property. Harrison and Shelby charter townships in Macomb County, Avon and Commerce charter townships in Oakland County, and Brownstown charter township in Wayne County represent the develOpment of a third tier (n3 Detroit metropolitan fragmentation. These five charter townships represent a dispropor- tionate amount of the industrial and commercial growth in time metropolitan Detroit area. Present boundary conflicts are between the second tier of suburban communities incorporated from 1945-1968 and township residents "incorporated" under the 1949 Charter Township Act. The 1978 Charter Township Revision From 1949 to 1960, three urban townships "incorporated" themselves as charter townships. Many other townships incorporated as (parts) of cities in the post-war wave of suburban community 107 formation. By 1977, there were 21 chartered town- ships. In 1978, the state legislature revised the charter township act in response to the Michigan Township Association's lobbying the state legisla- ture for a procedure to sidestep the Boundary Com- mission. The Township Association appealed for relief from the boundary decision process because the restrictions limited the Commission's authority to new incorporations. The Boundary Commission tended to favor second tier annexations of township property; the Township Association argued that strip annexation of valuable township land conti- nued with minimal Commission oversight. Given the advisory role of the Commission, it was difficult for the Commission to oversee annexation decisions. At the very least, the Commission did not have the formal authority to regulate annexation. In practice, the Commission appeared to favor contigu- ous city annexation over the Charter Township pro- vision. Under the 1978 law, charter townships have some although, not complete, immunity against annexation. In a 1985 State Supreme Court deci- 108 sion, the Court ruled against a charter township in an annexation case. While the issue is still un- resolved, it seems that townships cannot continue to utilize the township chartering rule like an incorporation rule. The township: chartering rule does not appear to grant urban township residents an annexation veto. After the 1978 revision 21 more townships became charter townships.[16] There are presently 80 Charter townships. These charter townships supply a greater range and variety of local public goods, at higher tax-prices and costs than unchar- tered townships. But township supplied local pub- lic good costs, prices, quantities, and product differentiation are lower than comparable municipal incorporation service levels. Charter townships provide metropolitan fringe residents with another option between city and rural township status. The charter township home rule option greatly increases metropolitan fragmen- tation by de facto incorporating all of the remain- ing third tier fringe residents. The Charter Town- ship Act institutionalizes minor general purpose 109 jurisdictions. These minimal cities provide fringe residents with a home rule option in addition to annexation or incorporation. At the same time, these charter townships compete with central cities and first and second tier incorporated suburban communities for resources. IV. Phases of Metropolitan Structural Fragmentation Phase 1: 1805-1908 The Regulated MonOpoly Structure Detroit. was the dominant supplier of local public goods from 1805 to 1908. Detroit community boundaries contained a high percentage of the metropolitan population. The City had primary responsibility for supplying local public goods to metropolitan residents even though very few local services were available and those services avail- able were offered in very limited supply. The early central core dominance period con- stituted a regulated monopoly structure. The regu- lated monopoly' period fostered a: non-competitive vertically integrated metropolitan community 110 structure. In 1805, Detroit was granted its' mono- poly position under the NOrthwest Territory Ordi- nance. Detroit's charter was redesigned by the territorial Michigan legislature in 1815. The regulated monopoly structure was continued under the 1835 Constitution which had no provision for the incorporation. of general purpose cities and villages. The 1850 Constitution specified that the state legislature shall provide for the incorporation and chartering of cities and villages. But from 1850 till 1908, the state closely regulated local public goods monOpolists through special act legislation. Special acts regulated Detroit policies an; if the city were a public utility corporation. All city governments were treated like special administra- tive districts where the supply decisions were subject to direct state legislation. By 1908 in Michigan it was clear that the special legislative act-- monopoly regulation sys- tem -- had very high transaction costs and that party machines had taken advantage of the city's monopoly supply position. Horizontal and vertical 111 community competition achieved the progressives efficiency goal to limit rents availabLe to local political organizations. In this manner progres- sive municipal reformers were very similar to the corporate ”trust busting" progressives at the national level. Home rule provisions were like the Sherman Anti-Trust Act because they intended to limit monopoly supply power by both increasing horizontal (metropolitan) fragmentation and decreasing ‘vertical (state 11) city) integration. Thus the logic of breaking the monopoly power of large firms extended to the breaking up of verti- cally integrated non-competitive party machines. Phase 2: (1908-1950) The Oligopolistic Structure The home rule concept was implemented by gen- eral law incorporation and chartering rules. This second phase of institutional development was marked by an increasing number of communities and by an absence of state entry barriers. A large number of first tier and a few second tier suburban communities incorporated in order to prevent 112 central city (strip) annexation by taking advantage of the negligible state regulation of community formation. Community formation was encouraged with minimal artificial barriers. Under low entry barriers and credible annexa- tion threats, community proliferation induces a strategic form of metropolitan competition. Stra- tegic interdependencies between contiguous local governments are caused by the fact that benefits and costs spillover jurisdictional boundaries. Existing cities strategically compete by strip annexing valuable commercial and industrial prop- erty while potential cities defensively incorporate resources. Visable patterns of metropolitan fragmentation appear from 1917 to 1930 as new municipalities took advantage of low entry barriers in response to the success of Detroit's annexation proposals. Stable long term metropolitan fragmentation patterns emerged from the strategic behavior of central cities and first tier suburbs during this period. The strategic bilateral annexation and incorpora- tion decisions affecting central city and first 113 tier suburban city play a large role in determining fragmented metropolitan structures. The small number of communities directly com- peting leads to interdependent central city and first tier suburban public goods supply and bound- ary decisions. Bilateral metropolitan competition causes the strategic design of local service deliv- ery areas. Oligopolistic competition exist in spillover contexts where interdependencies causes existing and potential communities to strategically define their' boundaries. Strategic' community competition generates fragmented Oligopolistic metropolitan structures through 21 bargaining pro- cess over spatial fiscal policy areas. Phase 3: (1950-1987) The Monopoly Competition Structure In the Detroit area a less strategically interdependent more competitive environment begins to emerge after the second and third tier suburban communities and charter townships incorporate. These incorporated communities compete with local 114 property taxes in cmder to attract households and firms. The degree (n3 intra-metropolitan competi- tion remains imperfect because of the type of goods being provided by local government. A monopolistic competition structure describes an environment where a large number of communities are imperfectly competing to provide local public goods and ser- vices. In this environment the central core city may still be a dominant actor in metropolitan service delivery. But the degree of competition among suburban communities and between the suburban com- munities and the central core city is quite sub- stantial. A monopolistic competition structure is literally a situation with a large number of mono- polists. The structural analysis can be modified to allow for differences in the relative size of monOpolists. The dominant service delivery posi- tion of any community is limited by the large num- ber local public good suppliers, regardless to the differences in sizes. The incorporation of townships via the Charter Township Act is a primary cause of increasing 115 metropolitan competition. Like the Lakewood Plan in metrOpolitan Los Angeles the Charter Township Act had the effect of immediately creating a large of cities in the economic growth areas of metropo- litan Detroit. These minimal cities are in a posi- tion to compete with central cities and older (first tier) suburban cities since they can consti- tutionally guarantee permanently low tax-prices for services because township millage rates are manda- torily limited to nominal tax rates lower than city millage rates. Charter townships can also fend off hostile strip annexations inf nearby incorporated suburban communities seeking to augment their tax base. The The Charter Township Act and the constitutional home rule provision for townships passed in 1961 increased the degree of metropolitan fragmentation. This structural fragmentation greatly expanded the number' of alternative local service delivery options available to metropolitan residents. The proliferation of communities under home rule incorporation and chartering procedures restructured the nature and degree of intra 116 metropolitan. competition. Today' Detroit. and the older first tier suburbs are keenly competing with second and third tier Cities and townships to pro- vide jobs and local services. The development of an. institutional arrangement. (like :charter town- ships or Lakewood Plan cities) in metropolitan fringe areas is crucial for understanding the increasing degree (n? (monopolistic) community competition. State home rule options cause the direction and magnitude of this fragmentation induced compe- tition. The home rule arrangement originated in a conflict between party machine organizations and progressive reform organizations vying for the control of municipal government. Progressive fram- ers of constitutional home rule provisions intended to divide the "top and bottem" party machine coali- tion by increasing the number of forums. Middle "class representatives who did not share in the spoils but had to pay for them through property taxation complained bitterly about graft and pro- vided the political base for the mmnicipal reform movement.”[l7] The net result was to institute a 117 political structure favoring metropolitan fragmen- tation by the creation of a large number of inde- pendent general purpose local governments. V. Fragmentation Induced Metropolitan Stratification Inter-Community Stratification Fragmented metropolitan structures correlate with stratification patterns inside and among com- munities. Home rule reinforces income stratifica- tion and racial segregation in! institutionalizing collective location choices. Class stratification is maintained within metropolitan political struc- tures accompanying home rule induced fragmentation. Table 3.2 describes the income and racial sorting in the five largest cities outside Detroit within Wayne County according to time 1980 Cen- sus.[18] The five major suburban cities are col- lectively less than one percent black (Total A). Charter and rural townships are excluded from the non-Detroit Wayne County total and total black population calculations. These five cities contain 46%, of the total (non-Detroit) Wayne County 118 TABLE 3 . 2 1980 SUBURBAN CENSUS DATA CITY NAME TOTAL POPULATION TOTAL BLACK Livonia 104814 108 Westland 84603 2200 Taylor 77568 1266 Dearborn 90660 83 Dearborn Hts. 67706 63 A Total 425351 3720 Ecorse 14447 5675 River Rouge 12912 4218 Inkster 35190 19994 Romulus 24857 4333 Highland Park 27909 23443 Hamtramck 21300 2751 B Total 136615 60415 population but only 6% of the total (non-Detroit) black population. The relevant comparison is thus the degree of stratification among blacks living in incorporated Wayne County suburbs. The second half (total B) of Table 3.2 describes the suburban ‘Wayne County' cities with higher proportions of black residents. The average percentage black is 44% in these five cities. Ecorse, River Rouge, Inkster, Romulus, (and the Detroit enclaves of) Highland Park, and Hamtramck 119 contain 15% of the total population and 91% of the total black population of the non-Detroit incorpor- ated Waynce County population. Table 3.3 profiles the fifteen cities and two townships adjacent to the City of Detroit. Table 3.3 includes two Macomb County cities (Warren and East Detroit), four Oakland County cities (Hazel Park, Ferndale, Oak Park, and Southfield) and a township (Royal Oak Township). According to the 1980 Census Detroit is 63% black with a $13,981 median income. For the con- tiguous cities the average percentage black is 13%. Clearly a wide range exists, from 84% to .1%, among these first tier suburban municipalities. The standard deviation for these communities is 24%, and the average median income is $22,013 with an $8,219 standard deviation, reflecting this wide range. Yet Detroit is socio-economically isolated from most of its incorporated neighbors and these fragmented incorporation patterns appear dynami- cally stable. Racial and income differences between Detroit and contiguous communities. exists. Perhaps even 120 TABLE 3.3 STRATIFICATION OF CITIES CONTIGUOUS WITH DETROIT CITY NAME Wayne County % BLACK MEDIAN INCOME Detroit .63 13981 Highland Park .84 $10298 Hamtramck .13 11432 Harper Woods .001 21436 Grosse Pointe Farms .002 38400 Grosse Pointe Park .002 29644 Grosse Pointe Woods .001 32434 City of Grosse Pointe .01 31514 Dearborn .001 22135 River Rouge .33 14242 Redford Township .002 24746 Oakland County Southfield .09 26451 Oak Park .18 22215 Ferndale .005 17592 Hazel Park .001 17409 Royal Oak Township .581 8836 Macomb County East Detroit .001 21578 Warren .002 23857 more striking is the degree of segregation among blacks living in suburban communities. While home rule was not passed with an intent to segregate blacks from whites in suburbs, home rule 121 institutionalizes a system of political units cor- related with racial segregation. Racial isolation remained constant between 1950 and 1980 and, if anything, slightly increased while relative income stratification actually decreased. Intra-Community Stratification A simple indicator of internal community stra- tification is a an income score. An income score measures the relative dispersion of the income distribution within a community. The income scores used here equal the sum of the squared proportions of each city's families within each income class.[l9] The procedure for determining an income score is quite simple. First, find the income categories dividing the whole metropolitan areas income into roughly three (upper, middle, and lower) income classes. Second, given the income categories for the whole metropolitan population, determine how many residents fit in each category for all commu- nities. Third, determine the proportion of commu- nity residents in each income "class" within the community then square and sum the proportions. 122 Forth, notice that if a: community's pOpulation is equally divided into the three categories then it's income score will be equal to .33 reflecting the same division as the metropolitan population. If all of a community's residents fall in one category the income score equals one. A community is stratified to the degree of income concentration in one of the categories. If the income distribution is evenly mixed within all communities in a metrOpolitan area, then the degree of metropolitan stratification would be low. Per- fect stratification occurs when each community has an income score of one. Perfect dispersion occurs when each community has the same degree of disper- sion as the metropolitan area or .33. The metropolitan area score always equals .33 because we have selected categories dividing the income groups into three classes. If we presume that a metropolitan areas are heterogeneous with respect to income distributions, and if presume that communities are subsamples of the whole metro- politan population, then the difference from .33 tells us hOW' much more homogeneous a community 123 income distribution is relative to the metropolitan (base) income distribution. The income score measures also provides a indication of the relative changes affecting commu- nities over time. Increasing income levels make any specific income cflassifications irrelevant in very short periods of time. Since the index starts with categories selected for the for the whole metrOpolitan area, the relevant comparison is within a community with respect to the whole metro- politan population, not just a simple internal distribution measure. Even though a memropolitan income base may be changing over time, income scores are not contingent upon absolute income levels. The indicator measures community disper- sion relative to the metropolitan distribution. For instance older wealthy or "exclusive" suburbs will have income scores near one. They will tend tx> stay (upper income) homogeneous even though absolute income levels are constantly increasing. Income scores essentially identify outlier income distributions. Table 3.4 above presents the income score 124 results. Each cell contains the number of cities in that cell. The results for 1950 can be obtained by reading across, similarly the results for 1980 are read down. I distinguish between heterogeneous and homogeneous communities with an arbitrary income score cutoff of .37. Those communities with income scores greater than .37 are considered homogeneous and they are further broken down by which income category they are homogeneous. TABLE 3.4 INCOME CLASS HOMOGENEITY FROM 1950-1980 1980 Heterogeneous Homogeneous Low Upper 1950 Het. 2 3 1 Hom. Low 1 Middle 1 [Hazel Park] Upper 24 1 10 For example Hazel Park was a homogeneous middle income suburban community in 1950. By 1980 Hazel Park is a heterogeneous community with an 125 income score less than .37. There were six heter- ogeneous communities in 1950 whereas by 1980 there are twenty seven. The results in Table 3.4 are exactly the oppo- site of those found for the Los Angeles area. [20] The existing cities in metrOpolitan Los Angeles and the newly incorporated cities became relatively income homogeneous over time. All of the new Lakewood Plan "mdnimal cities" resembling charter townships were upper income homogeneous. Only ten cities remained income homogeneous in the Detroit area. Hamtramck went from being upper income homogeneous in 1950 to low income homoge- neous in 1970 while Royal Oak township remained low income homogeneous in both years. Unlike the City of Los Angeles, Detroit. became more homogeneous (poor) over time. Only Riverview went from a rural income heterogeneity to an upper income homogeneous community out of the 43 Wayne County cities in Table 3.4. Over half of the cities (24) went from upper income ihomogeneity 1x3 income heterogeneity during this thirty year period. Unlike Los Angeles communities the cities in 126 the metropolitan Detroit area are increasingly becoming less homogeneous with respect to income distribution differences. Furthermore the central core city is becoming increasingly homogeneous (or isolated) with respect to the income distribution and racial composition of the metropolitan popula- tion. The existence of fragmented metropolitan governance has, at the very least, institutional- ized collective location choices tending to rein- force patterns of class stratification. VI. Dynamic Coalition Formation and Structural Reform Strategies Structures are not neutral with respect to the participants. Institutional design features affect political organizational "flow" charts. A success- ful organizational structure in.