Resumé

Restructuring of central and local government relations has occurred in a number of countries during the 1980s, and has been a striking phenomenon even in such different countries as Poland. Sweden and the US. In the first part of the article a resource framework is presented. of resources are identified: constitutional-legal, regulatory, financial, political and professional resources. It is argued that local-covernment disposal of such resources provides a potential for local discretion. The applicability of the framework is illustrated by reference to a study of Poland, Sweden and the US. focusing on the restructuring of central and local government relations in these countries during the 1980s, The conclusion is that decentralization has occurred in all three cases, especiallv in terms ot reculstorv and other non-financial resources. However, in Sweden and the US this trend has been counterbalanced by centralization of financial control. In Poland the post-communist revolution has paved the vvav tor a restoration oi tne pre-war local self-covernmcnt svstem. Democratic elections at the local level are seen as crucial to break down the old, centralized partv-statc svstem. and to reintroduce a dual local government svstem. In all three cases economic pressure, related to the individual countries" positions in the world capitalist system, seems to have been the main driving force beind the decentralization processes. However, the torms. magnitude and effects of decentralization must be seen in relation to the specific political economies and histories ot the three countries. Thus, to understand the causes, mechanisms and ettects ot central—local government restructuring in depth one has to relate the resource approach to a broader theoretical framework.

Introduction

For a long time research on local government was rather untheoretical and non-comparative, consisting mainly of case-studies of specific countries. Summarizing the state of research Rhodes (1980) concluded that "the comparative study of local government, in its widest sense, has not existed and does not exist". During the 1980s, however, there has been a rapid growth in research on local government and central-local government relations, research which has also touched upon general theoretical themes such as the crisis of the welfare state, problems of legitimation and power

(Goldsmith & Villadsen 1986; Dente & Kjellberg 1988; Sharpe 1988; Reade 1989). We have also been presented with the first more serious attempts at systematic comparative analysis of local government in different countries. Comparisons have been made between West European countries (Page & Goldsmith 1987) as well as between West European countries and the US (Elmore et al. 1986). However, cross-national comparisons of centrallocalgovernment between capitalist and socialist/post-socialist countries remain relatively rare (Regulski & Regulska 1988: one recent exception is Regulski et al. 1988). For that reason one could easily agree with Harloe (1981) that comparisons between Eastern countries and Westerncapitalist would be of some value even if they are of a purely descriptive kind.

As a consequence of the dramatic developments in 1989 the whole issue of central versus local government in Eastern Europe now appears in a radically different light. Thus, to take Poland as an example, during the 'socialist' period it did not even make sense to speak of local government in the Polish system. When attempting to apply a theoretical framework developed within the Western context to a political system as different as Poland, it resulted in 'seeking and examining non-existing phenomena' (Mokrzycki 1982, 47). Today a broad movement of local citizens' committees emerged across the country, a Ministry for Local Government Reform has been created, and democratic communal elections were held in May 1990, i.e. two years earlier than previously scheduled (Foundation in support for local democracy 1989).

However interesting the introduction of local self-government in a previously state may be in itself, the intent of this article is to present a framework suitable for comparison of central-local government relations in different countries. Using a 'most different systems approach' (Meckstroth 1975), the objective is to see if it is possible to observe corresponding processes of central and local government restructuring in different cultural and political contexts. Since 'decentralization' in different forms has been on the agenda in different countries at least since the end of the 19705, one might ask if there are causes, mechanisms and effects of a general non-systemic kind behind this development or if the similarities are merely superficial and actually represent underlying developments which are quite different.

The aim of the article, in short, is to formulate a conceptual framework, suitable for cross-national comparative analysis. Drawing upon Rhodes (1980, 1981, 1986a, 1986b), a framework is presented which may serve as an instrument for broad comparative description and analysis. After discussing the framework critically, the article gives some empirical illustrationsdrawn a comparative study on the restructuring of centrallocalgovernment in Poland, Sweden and the United States. In

conclusion causes and mechanisms of the decentralization processes in the
three countries are discussed.

The Resource Framework

In 1980 the state of research on central-local government relations was summarized in a 48-essay volume entitled International Handbook on Local Government Reorganization. Contemporary Developments (Rowat 1980). In the concluding chapter the editor argues:

Before much progress can be made in comparative studies, a way must be found to measure or at least estimate the relative power of local government in the overall system of a country. Since the process of decision-making is central to the exercise of power in any political system, one must therefore study the role of local government in this process. One must ask, W^hctl s Hare docs loccil &o v eminent have in the total of ull Governmental decision- tnci kin @ in a country? In this connection, one must also ask to what extent local governments are able to influence decisions made by senior governments, either state or central. A full analysis of the strength of local government must take both of these questions into account. (Rowat 1980. 601)

Obviously there is no simple way of accounting for local government strength. However, as a starting point it seems unavoidable to take the first of Rowat's questions at its face value: What share does local government have in the totality of all governmental decision-making in a country? The question refers to the scope assigned to local government in the overall political system. How is one to measure that scope? Sharpe (1988) argues that local-government growth should be measured in terms of several indices, such as functional scope, power, personnel, expenditure, outputs and outcomes. However, some of these measures of government are 'difficult to conceptualize, let alone measure', and that is why. according to Sharpe, 'the generally accepted measure of government growth has become largely confined to expenditure' (1988. 365). Yet even this measure is not without problems. Should public corporations producing for the market be included? Could not an inefficient government increase its expenditure without any real increase in that government's outputs? On the other hand, could not efficient governments increase their outputs without any real increase in expenditure? Bearing these and other possible ambiguities in mind, one can nonetheless agree with Sharpe that the expenditure method of measuring government may be useful as a general point of departure for comparative analysis of central-local government relations in different countries.

Even so, estimating local government's share of total public expenditure or employment in a country does not in itself say anything about the degree of discretion exerted by local authorities. Local government may well have responsibility for a broad range of functions, but may at the same time

be strongly supervised by central government. On the other hand, local government may have responsibility for only a few functions but still have a good deal of discretion in this limited area. So, taking local government's share of total government growth as a point of departure, we have to take our analysis one step further. It is necessary to look at the distribution of public expenditure and activities on specific functions.

A question has been raised whether there is an inherent logic in the distribution of functions between the two levels of government. In their 'dual state thesis' Cawson and Saunders (1983) suggest that there does exist a functional division and imply that central government bears the main responsibility for social investment ('production') whereas social consumption the privilege of local government. The thesis seems to have some value as a heuristic device for 'identifying deviations from an abstract pattern of relations' (Page & Goldsmith 1987, 162-163). However it seems too crude to describe, much less explain, cross-national variation in centrallocal relations. All attempts at a more sophisticated functional division of governmental activities have to face the problem that most policy fields are multi-functional. In terms of the dual-state thesis, for example, housing policy may be classified as both investment and consumption.

In a less pretentious way it seems necessary to differentiate between various policy areas in order to search for similar patterns or striking differences between countries. Central-local government relations in a country may vary a lot from one policy area to another making sweeping generalizations about this relationship misleading (Page & Goldsmith 1987, 161). In Sweden, for example, it appears that educational policy has been more centralized and regulated than cultural policy - in spite of the fact that the former is a much more predominant part of local expenditure than the latter. Thus, a more or less sophisticated functional division of governmental activities on central and local levels may serve as a link to an analysis focusing on specific policy sectors.

Before analysing such sectors, however, there is a need for a more general analysis of local-government resources. Without legal, financial and other resources, local governments would be reduced to extensions of central governments unable to act according to their own priorities. But it is important not to look at resources too rigidly as they have a 'subjective, volatile transmutable character'. They may be used as weapons to accomplish goals, which may in turn be looked upon as 'prizes' and add to the original resource base as weapons for the next round of the 'power game' etc. (Rhodes 1986b, 12) These 'games' played by central and local governments may be either of a zero-sum or positive-sum kind. An increase in central-government resources does not necessarily imply a corresponding decrease in local-government resources. And resources may be used both

for competitive struggle and coalition building (Bacharach & Lawler 1980,
107).

Resources may be classified in a number of ways. For the purpose of crossnational it seems superfluous to construct very sophisticated classification schemes. Drawing upon the classifications presented by Rhodes 1981, 1986a), it is possible to distinguish between five sets of resources: (1) constitutional-legal, (2) regulatory, (3) financial, (4) political and (5) professional. The boundaries between the different categories can be discussed, and there may be other types of resources which ought to be taken into account. All the same, the categorization includes the major resources which seem to be relevant to an analysis of central-local government

Constitutional-legal resources pertain to the range of functions formally devolved to local government under terms of the Constitution or by parliamentary A high degree of local autonomy in this dimension does not necessarily in practice correspond to a high degree of local autonomy. Central government may have restricted local autonomy through a number of supervising and controlling mechanisms, bringing us to another kind of resources.

Deriving their legitimacy from the constitutional-legal framework of a state, regulatory resources pertain to the scope and strength of rules used by the central government in order to regulate and control local government. Following Hancher & Moran (1989. 130) regulation may be defined as 'the making and enforcement of legal and administrative rules'. Deregulation, of course, is the mirror image of regulation, meaning 'the explicit alteration. amendment or abolition of a set of rules'. Hancher & Moran also give a more abstract, theoretically founded definition of the concepts of regulation and deregulation, but the simpler formulations fit our purpose well enough. One should observe that deregulation does not always mean the abolition of rules. It could also mean substitution or systematization of rules. In the last case deregulation is 'a kind of regulatory reform, the goal being to make regulation more efficient" (Hancher & Moran 1989. 131). In at least one instance the term 're-regulation' has been used to denote this phenomenon, i.e.

instruments designed to achieve more effectively and efficiently the same behavioural
changes sought with the original regulatory scheme (Hanf lc)84.lc)84. W).

Deregulation is not only a 'central feature of the political economy of advanced capitalist nations' (Hancher & Moran 1989. 136) but also a striking feature in the East European countries, and this even before 1989. Indeed, regulatory resources, roughly corresponding to what Rhodes once called hierarchical resources (Rhodes 1980. 573). is a dimension that

cannot be dismissed when analysing central-local government relations in
a country.

Financial resources refer to the funds raised by local governments from taxes, service charges, central grants and borrowing. This dimension is intimately connected to the foregoing as grants from central governments are often accompanied by detailed rules governing their use. But central grants are not always conditional. They may also be 'block grants' not limiting local discretion as the former do. However, the right to impose taxes on its citizens is probably the basic financial provision for localgovernment

Political resources refer both to the access to public decision-making structures and the right to build public support conferred on representatives by the legitimacy deriving from election. An important aspect of access involves the capacity of local governments in a country to form a 'national local government system', manifested in national local-authority associations define 'the national role and the state of opinion in local government as a whole' (Dunleavy, as quoted in Rhodes 1986a, 23). But if local government is to mean 'government' and not just deconcentrated central state administration, it also has to reflect territorially defined local constituencies where elections have more than symbolic significance for the people, not just a democratic facade behind which authoritarian top-down governing is hidden. A situation where local government councillors have access to the national decision-making structures without being legitimately anchored in civil society may imply a strong central government impact on local government rather than the other way round. This may well have been the case in many East European countries, and it vividly illustrates the need to take both kinds of political resources into account.

Professional resources refer to 'the possession of people, skills, land, buildings, material and hence the ability to act directly rather than through intermediaries' (Rhodes 1986b, 17). It also has to do with the capacity to collect and process data. This capacity in turn requires a staff of personnel representing knowledge and expertise. Professionals have an intricate position, being able to lend their competence to either of the two governmental levels, and also having the capacity to pursue their own interests.

In reality it may be difficult to separate the different sets of resources from each other, and, as has been said before, the list is not necessarily exhaustive. For the purpose of broad cross-national comparison, however, it can be seen as a fair starting-point. The resource perspective may also be useful as an instrument in assessing the direction of the development of central-local governement relations within a country. Indeed, time-series analysis of the central-local government relationship within a single country seems to be a necessary precondition to cross-national analysis of this

relationship (Wolman 1988a). What at the surface may look like a general decentralization process may by resource analysis be revealed as a much more complicated phenomenon. Central government may have an interest in presenting itself as a promoter of decentralization, yet not always act in accordance with its word.

Talking about resources, it is also important not to overlook the fact that
there is normally a basic asymmetry in central-local government relations:

The relationship between central and local governement is not so much a variety of 'different' types of relationship, but rather one relationship between a constitutional superior and subordinate in which the centre has a differential disposition and ability across different services and activities to set parameters varying in their tightness and consequently varying in the discretion they permit local government actors. (Page, as quoted in Rhodes 1986 a.)

Thus, the picture of two equal partners bargaining with each other is too idyllic. The centre usually has the privilege of being able to alter the rules of the game unilaterally. Changes in central-local government relations occur all the time but within the former's preference of rule interpretation. Of course, the situation is more complicated in federal states, but principally the argument will hold even there.

Discretion

The concept of 'discretion' has been defined as 'The room for decisional maneouvre possessed by a decision-maker' (Rhodes 1981. 108). viewing 'decisional' in its broadest sense and encompassing the whole of the policy circle, not just its formal decisional stage. Discretion is a matter of degree and even a local authority subjected to the closest supervision by a central department will have some degree of discretion. The degree of discretion, moreover, may clearly vary beetween the different resource dimensions. A local authority may have a large degree of discretion in legal terms yet at the same time be strongly supervised financially by central government.

The degree of discretion may also vary according to services. Thus,
examining central and local government relations in seven West European
countries, the editors conclude:

the seven countries: local discretion appears to vary more between services than between
states, and it is impossible to state that any one country has a more restrictive regime than
another. (Page* Goldsmith 1987. 161)

They even go as far as stating that "it is not possible to come to a general conclusion, even an impressionistic one. about the degree of discretion in service delivery that local authorities have in practice" (Page & Goldsmith 1987, 158). This conclusion is a strong argument in favour of a shift in focus to the policy content of central-local government relations. Exploring the

evolution of particular policy areas over time and examining the relations between the actors and interests involved may be a way of penetrating the formal aspects of central-local government relations and pave the way for conclusions as to the degree of local-government discretion. However, considering the earlier absence of studies comparing these relations in different systems, there is indeed also a need for initial comparisons of a more general kind. It is here that the resource framework may be most useful as a conceptual framework.

Limitations of the Framework

When used to analyse central-local government relations in Britain, a number of criticisms have been raised against the 'Rhodes' framework', including the criticisms noted by Rhodes himself (Rhodes 1986a). The main conclusions drawn by Rhodes after his self-critical scrutiny of the resource framework point forward to a shift in the conception of the object of study:

The phrase 'central-local relations' suggests a bias towards the analysis of institutional relationships. Such analysis does not always provide an adequate account of policy systems. To focus on policy communities is to assert the primacy of policy networks and policy content over the relationships between particular types of institutions. The phrase centrallocal is, therefore, an inappropriate definition of the subject. 'Intergovernmental theory' with its emphasis on fragmentation, professionalism and policy networks is more appropriate. (Rhodes 1986a, 28)

In considering this conclusion it should be emphasized that it pertains mainly to the analysis of central-local government relations in one country, namely Britain. The framework proved too superficial to be able to grasp these relations in depth. However, the framework is still useful to compare the development of central-local government relations in different countries in more general terms. Without such a basic comparison, it does not seem meaningful to dig more deeply into specific policy areas searching for 'policy networks', 'policy communities' etc.

My conclusion, in short, is that the criticisms directed at the Rhodes' framework do not reduce its value as an instrument of basic cross-national analysis. This conclusion should be seen in relationship to the general state of the art in this long-neglected, although currently expanding, branch of political analysis. Thus, there is not an ocean of alternative approaches to choose from. When Page and Goldsmith in 1987 published their book Central and Local Government Relations. A Comparative Analysis of West European Unitary States, they noted in the preface that

although there are many excellent studies of local government and intergovernmental relations in a variety of countries, there have been few attempts at a systematic comparison of the kind necessary to achieve the kind of understanding we believe to be desirable (Page & Goldsmith 1987, vii; my emphasis).

And they also stated that 'comparative description, albeit on a relatively large scale, is a modest objective, yet one that has largely eluded students of central-local relations' (Page & Goldsmith 1987, 9). Although the resource approach may warrant criticism for its crudeness, the words of Page and Goldsmith must be borne in mind.

Illustrations

To illustrate the applicability of the resource framework some examples may be provided from a comparative study of central and local government relations in three radically different countries, i.e. Poland, Sweden and the US. The choice of countries compared reflects a most-different systems approach. The intent of this study has been to see if there are non-system specific or analogous differences in the development of central and local government relations in countries that are radically different with respect to their political systems. For obvious reasons the richness of descriptive detail in this article has to be low. An attempt has been made to single out those changes in resource dimensions in each country that highlight the restructuring of central and local government relations that has occurred during the 1980s. Because of the nation-specific histories surrounding these developments, it is not meaningful to structure each case-study section in an absolute identical way. In the concluding section the results of the casestudies brought together analysing the more or less parallel processes that seem to be at work behind the restructuring of central and local government relations in the three countries.

The Case of Poland

Before 1980 the local government's share of the total state budget in Poland as in other socialist countries of Central Europe amounted to between onefifth one-third (Zawadzka & Zawadzki 1980, 368; Regulski 1989, 437). In 1983 the total volume of local-authority expenditure exceeded 884 milliard zloty, i.e. one-third of all public expenditure. About 60 percent of the local expenditures were located at the province level, while 40 percent related to the communes (Regulski et al. 1988, 110-111). Economic activities, and health care accounted for about one-third each of local spending (Regulski 1989, 439).

Thus, as regards local government's share of total public spending. Poland did not differ much from many West European countries in the middle of the 1980s, falling in between Britain with 26 percent and Sweden with 42 percent (Page & Goldsmith 1987, 157). Nevertheless, according to Mokrzycki it 'made no sense to speak of local government in the Polish

system¹ under communist rule. It was a 'non-existing phenomena' (Mokrzycki 1982, 47), perhaps most vividly illustrated by the faet that the local councils did 'not control any executive bodies¹ but had to 'act through the state administration' (Regulski 1989, 432). Consequently, one of the main objectives of democratic reform in today's Poland is the 'territorial reorganization of the state. This, however, cannot be achieved without fundamental restructuring of the current Soviet-like local government' (Foundation in support for local democracy 1989). Looking more closely at developments in Poland before the Solidarity government took power, however, one discovers that even then there were decentralist changes occurring. Thus, it would be wrong to say that current developments represent a transition from total centralization to a mirror picture of decentralization.

Constitutional-legal and regulatory resources. - The basis of the 'socialist' government system in Poland was established in 1950. Since that time this basis evolved slightly and was amended by several acts. However, the institutional basis that was laid by 1950 remained largely the same until the end of the 1980s. Thus, local authorities were elements of the state system. The property they managed was state property, local ownership did not exist, and local budgets were part of central-state budgets. The concept of the socialist state implied that all activities, including those of the local authorities, had to be subordinated to the central authority. Consequently, central government had the power not only to supervise and control the local authorities, but also to assign tasks to them and to interfere in the way they performed them.

The nationalized means of production were organized in sector-divided, state-owned enterprises. These enterprises had more means at their disposal than did local authorities. Thus, the territorial principle of organization took second place to sectorized centralism. Although responsible for the comprehensive development of its areas, and the satisfaction of the needs of its residents, local authorities lacked the resources for taking initiatives and action. Local government was responsible for a lot of things but had no resources of its own. The internal organization of local government was a reflection of the previously mentioned system. Thus, the local executive bodies were subordinated in two ways - to the appropriate council and to the central authority. The local chief officers were first and foremost acknowledged as civil servants of the state (Regulski & Markowski 1988; Regulski et al. 1988, Chapter 3).

Thus, in constitutional-legal terms, local government in 'socialist' Poland was subordinated to central government in all respects. Although from time to time this subordination was modified somewhat, the decisive power was firmly kept in the hands of the central state, and ultimately the

communist party. However, from the beginning of the 1980s there were
hints of what was going to become the radical anti-communist revolution
by the end of the decade.

Financial resources. - In 'socialist' Poland industry was understood as the main instrument for the development of the socialist state. Thus, a majority of state resources was directed at sectoral industrial organizations, while local governments had at their disposal a much smaller share of resources. For example, a research project found that in about twenty smaller towns in Central Poland more than 70 percent of day-care centres, more than 60 percent of cultural amenities and nearly 40 percent of all kindergartens were built and run by industrial firms (Regulski 1989, 436). From time to time the industrial priority was challenged, but the local councils never got the financial opportunity to run their own urban policy. The municipal budgets were an integral part of the general state budget, and the collection of budgetary incomes (taxes, profits made by state owned companies etc.) was done through centrally controlled tax offices. For a long time the provinces played a strategic role as instruments of central-government financial policies. Between 1981 and 1983, however, the communal share of non-central-government spending increased from 19 to 39 percent, representing an enlargement of communal responsibilities. The communes were nonetheless "still subordinated financially to the provincial administrations" 1989, 439).

In 1985 29 percent of local expenditures was covered by para-budgetary funds, e.g. retirement fund, foreign-debt repayment support fund, science and technology fund, and family-help fund. The number of such funds increased from 10 to 48 between 1970 and 1986, and their importance grew steadily. Formally their use was tightly controlled and regulated by the central state, but they were widely critized because they allowed important sums to be collected outside of overall control and to be used for fragmented and not always important goals' (Regulski 1989, 441). Thus, in spite of the formal restrictions these funds seem to have potential as a local-government resource.

Political and professional resources. - The "socialist' Constitution assured the Communist Party the leading political role in Poland. The listing of the candidates for the local elections was in the hands of the Party and other bodies subordinated to it. This meant that all councillors were Party members, or at least that they had the Party's approval. The main task of the local councillors was to implement the Party's national programme. Thus, in practice the local councillors had a stronger identification with the monistic political structure at the national level than with the local interests. According to Rhodes' framework access to the national political system is

one important dimension of local government's political resources. In the case of socialist Poland the situation was rather the other way round: through the local councillors the central Party apparatus had access to local government.

Consequently, the legitimacy of local government was weak, and only a limited number of people, 'local patriots', who were emotionally linked to a town or region, were ready to sacrifice their own time to opt for local development. On the other hand, local managers of state-owned enterprises, doctors, lawyers, architects and other professional groups based their careers on sectoral and professional organizations. On the whole, few people looked upon local government as a basis for action to meet local needs (Regulski 1989).

In search of a new local government. - Since about 1978 Poland has been plagued by a severe economic crisis. One effect of this crisis was the introduction of a policy to restructure the relationship between central and local government in favour of the latter. After stormy discussion the Parliament in 1983 ratified a new administrative structure concerning the position of the local authorities: the Act on City Councils and Local Self- Government. Earlier, local government could make policy only on issues which were explicitly placed within their competence. Under terms of the new Act it was accepted that local government could make decisions on all issues which are not the explicit juridical competence of higher administrative In fact, this meant a recognition of the principle of local self-government.

The local budget was also strengthened, as every local unit now got a guaranteed income from grants for the next five years. Thus, grants not used in a particular year could be reserved for the next year. 'Excess' financial resources no longer had to be returned to the central-state budget. The introduction of the Department of Treasury at the regional level and the Chamber of Treasury at the local level meant that they took over some financial functions from the Association of Enterprises. The Chambers and Departments of Treasury had a strong relation to the central Ministry of Finance which continued to control its management of local authorities' finance. However, that did not exclude a potential for a higher degree of local-government discretion.

According to the 'socialist' order, the Local Executive Officer and the Regional Executive Officer were the tentacles of the Prime Minister. The new Act changed these relations a little, as the People's Council could require the Local Executive Officer to resign. Even so, the local authorities were still mainly incorporated into a highly centralized administrative structure.

The 1983 Local Act meant that the uniform central state power was

diluted a little, introducing some 'decentralized responsibilities' with less control by higher administration. People's councils were also given more discretion in the field of local economic planning and budgeting and they were also given more discretion as regards the appointment of local officers. Still the financial resources of local government remained weak (Goverde & Markowski 1986).

In constitutional-legal, regulatory, professional and even financial terms local governments were given a little more resources, but the central state did not loosen its political sovereignty. The legitimacy of the Communist Party and the central state system had totally vanished, and the increase in local-government discretion was too small to give it a legitimacy of its own. Decentralization did not help much as long as political power was concentrated in the hands of the central state (Regulski et al. 1988, Chapter 3; Regulski 1989).

The need for a radical local-government reform was raised during the Round Table talks between Solidarity and the Communist government in the spring 1989. Thus, Lech Walesa in his introductory speech mentioned four fields which were crucial to break up the old system: (1) access to mass media and to information sources; (2) freedom in establishing new associations; (3) independence of judges from political and administrative pressures; (4) local democracy.

Professor Jerzy Regulski. who was Solidarity's chief negotiator on localgovernment
summarized the demands raised upon the government.
They aimed at:

(1) the abolition of the constitutional principle of uniform state power: the
local council should represent local society only, and be released from
hierarchical dependencies;

(2) a new democratic electoral law;

(3) recognition of communal legal entity and ownership rights;

(4) a stable and controllable system of supplying local budgets, free of
arbitrary decisions by state administration;

(5) limitation of state interference in local affairs to control of the legality of
communal decisions only, and abolition of all administrative regulations
concerning communal tasks;

(6) transfer of local state administration to communal control:

(7) freedom to establish intercommunal associations both local and goal
oriented, and national, to represent local interest in central government:

(8) the right to judicial appeal against decisions of the state administration
(Regulski 1989:442).

Taken together these demands would imply an immense increase of local
government discretion. No wonder then that the communist government

found some of the demands unnegotiable, i.e. numbers 1, 4, 6 and 7. W. Sokolewic, who co-chaired the talks on behalf of the government, acknowledged the Solidarity demands as an attempt to build an alternative political system competing for power with the ruling party. He said the first step would be to capture the power on the communal level, then to take provincial administration under political control, and finally to create a union of local governments as a pressure group on the national level. Although this was not the outspoken goal for Solidarity, the effect of a radical local-government reform might well be 'the extinction of the nomenclatura system at all levels of power', as expressed by the parliamentaryleader Solidarity, Bronislaw Geremek (TT-Reuter, Warszaw, as quoted in Dagens Nyheter, 18 January 1990).

As a result of the parliamentary electoral campaign in spring 1989 a nationwide social movement grew up, expanding under the auspices of the Commission for Local Self-Government, which is a part of the Lech Walesa Citizens' Committee. More than 3000 citizens' groups emerged across the country. The committees were socially and politically broader than the working-class based Solidarity, and thus had a stronger potential as future local-government councils.

In order to initiate the process of local-government reform, the Senate passed a resolution on 29 July 1989 to introduce local self-government in Poland and to undertake appropriate legislative measures. The resolution includes all eight points which were mentioned above. The radicalism of the resolution and the speeding-up of its planned implementation imply that local government reform has now become strategically urgent. Quoting the Foundation in support for local democracy (1989, 2):

The totalitarian centralized state has taught people how not to participate in local institutions. Furthermore, the existing system destroyed the understanding of the purpose of local selfgovernment, its tasks and mechanisms of functioning. The basic knowledge of selfgovernance to be regained. It is necessary to prepare local communities for democratic elections. It is important to show what local self-government should mean under democratic conditions, what the real limits of its powers are and how such a body should work (my emphasis).

This also means a return to a dualistic system of local administration
with two overlapping structures, the state administrations and local selfgovernment.
local self-government was

an important part of public life in Poland until the Second World War. Before 1918 it was one of many ways in which the national and economic interests of the Polish population were protected. Between 1918-1939 it promoted the development of towns and villages, public facilities and institutions etc. It was also a scene of sharp political struggles, which resulted in 1933 in its independence being limited. . . (Regulski et al. 1988, 67)

Resources and discretion. - Looking at the development of central-local

state relations in Poland from the resource perspective it is self-evident that the period after 1980 represents first a half-hearted, then a dramatic shift in a decentralist direction. However, it was not until the uniform state power was broken in 1989 that constitutional-legal, regulatory, financial and professional resources will have a fair chance to be used in favour of local se//-government. As long as the local governments were integrated into the centralized state structure they could get no public legitimacy, the degree of local-government discretion was extremely limited and decentralistreforms this framework were useless. In that sense the political resource dimension was crucial.

Whether a new system of local self-government will emerge, as aspired by the Solidarity government, remains to be seen. Less than 50 percent of the electorate took part in the local-government elections in May 1990, which might be interpreted as a remaining distrust in the local political system in Poland. How to handle the old party/state 'nomenclature" at the different levels of administration also seems to be an issue of conflict within Solidary (Bujak 1990). On the other hand the government has initiated an ambitious programme to promote the building of local self-government and democracy in the country.

The Case of Sweden

Sweden well illustrates the thesis that the development of the welfare state since 1945 'has tended to mark a process of decentralization rather than centralization of the modern democratic state' (Sharpe 1988, 369). Together with the other Scandinavian countries local governments in Sweden have a higher percentage of expenditure and work-force than most other countries, i.e. 42 and 54 percent respectively in 1984 (Page & Goldsmith 1987, 157).

There are two kinds of local government in Sweden: municipalities (kommuner) and county councils (landsting). The 23 county councils are responsible primarily for health care, including the provision of hospital services, and certain types of education and vocational training. The 284 municipalities are responsible for a large number of services: provision of infrastructure, housing, education, social welfare, leisure activities and culture. There are no hierarchical bonds between the two kinds of local governments.

Constitutional-legal and regulatory resources. - Local self-government in Sweden has a strong legal foundation, and finds its place even in the ornamental paragraph of the 1974 Constitution. Based on the principles of the Constitution, a more detailed regulation of local self-government is given in the Local Government Act of 1977, which replaces a number of

laws with antecedents going back as far as 1862. The Local Government
Act regulates such things as organizational structure, finance, taxation and
citizens' rights of appeal.

Constitutionally, local government has a dual character, being simultaneously 'antennae' of central government and a 'voice of the periphery' (Miliband 1969, 52-53). Thus, there are two fields of local-government administration - facultative and mandatory administration. In the first field local governments are free to make their own decisions; in the second centrally decided laws regulate what local governments ought to do and must not do. The political history of Sweden illustrates that the latter field has grown faster than the former, and about three-quarters of municipal expenditure relates to mandatory, often closely regulated tasks. These tasks can be seen as a reflection of social-democratic and social-liberal welfare ambitions as regards social, health and educational policies.

Financial resources. - According to the Constitution municipalities and countries are allowed to impose taxes on their citizens, although Parliament may enact laws specifying the forms of taxation. Taxes, in the shape of a proportional local income tax, account for a stable share of overall resources (slightly over 40 percent) for the municipalities and about 20 percent more for the counties. Central-government grants have increased their share of total incomes to more than a quarter for the muncipalities and about 10 percent or less for the counties, while net borrowing has decreased, now playing a seemingly insignificant role, especially for the counties (Murray 1985, 310). The Local Government Act contains few detailed rules regulating of local finances. This is well in accordance with the ideology of local self-government. In practice, however, central government tried to influence local governments in a number of ways and has taken action on matters such as general economic policy, planning, categorical grants and agreements on tax rates. In April 1990 there was even a decision taken in Parliament to prohibit the increase of local taxes.

Political resources. - Traditionally local government in Sweden has been carried out by lay people, directly at town meetings or, in the cities, by local councils. Indeed it was not until 1953 that the representative system became mandatory. And even towards the end of the 19505, there were still municipalities without any employed administrative staff. On the other hand the number of representatives was drastically reduced with the amalgamations in the 1950s and the 1970s (Strömberg & Westerståhl 1984, 54). The political system in Sweden is deeply penetrated by party politics and the central-local dimension cuts through the minds and actions of activists within each of the political parties, continuously provoking internal conflicts, at least in a period of financial restrictions. A political resource

base of special interest is the 'national local government system" (Rhodes 1986a, 1986b) manifested in Sweden by the Association of Local Authorities and its county-council counterpart at the regional level. These associations represent an important set Of political as well as professional resources.

Professional resources. - Competence is a necessary condition for decentralization become something more than decentralization within central government, i.e. deconcentration. Since recruitment and advancement within the local administrations have become strongly sectorized and meritbased, administrators of today represent great professional competence. Whether this competence is a local-government resource versus state power, a central-state resource versus local-government power, or a resource used mainly to promote the self-interest of the bureaucracy itself, may be disputed. It probably varies from field to field and from time to time. When local administration as a whole is threatened - as in times of public expenditure cut-backs - it can mobilize considerable force via different channels to defend its position. One such channel is the labour unions, who organize practically all employees within the public sector.

Decentralization and control. - Since the late 1970s several deregulation and decentralization programmes have been initiated and implemented in Sweden (Eländer & Montin 1990). Thus, a number of discretionary laws have been enacted replacing older ones which regulated in detail the activities in question. Examples are the acts of social services (1982), health and medical care (1983), conservation of natural resources (1987) and planning and building (1987). According to the free commune reform some local authorities have been allowed to organize their activities and to use state grants with a greater degree of freedom. With few regulations from central government, they were free to develop a pattern of goods and services which was adapted to local needs.

Another reform which seems to imply increased local-government discretion the sub-municipal reform. By 1988, about 30 local governments had divided their territories into subunits governed by sub-municipal or sub-county councils. These councils were granted authority for one or several policy areas, mainly the 'soft' ones such as culture, leisure, primary education and social services. Presented in the early 1980s, this submunicipal was believed to be a very important instrument indeed to revitalize local self-government.

Whether or not an economic crisis existed, political elites began to define the situation in these terms at the end of the 19705. Strategies were developed to decrease the speed of public-expenditure growth. Central government negotiated with the national associations of local government on voluntary agreements not to raise taxes. Central government also

became more restrictive regarding grants to local government. At the time, local government developed strategies to minimize the effects of central government's restrictive policies. More business-like methods of managementwere for example. Through more speculative handling of money, local governments began walking a thin line between what is legal and what is illegal behaviour. Ideas were borrowed from the private sector in order to give more economic responsibility to officials at the local level in the service delivery system. Many local governments developed an internal competition-system, and wages began to be settled on a more individual basis than before, changing the wage-system from an equalityorientedto open competitive one.

However, privatization has not yet occurred extensively. Instead Sweden has been witnessing a growth of para-governmental or semi-private institutions, as well as locally. These institutions have often been set up for specific and temporary purposes, thereby representing an element of flexibility in the complex political and administrative system, and they often incorporate both public actors and representatives of various interest groups. Forms of public-private partnership have developed, for example in the local-government industrial policy sector. This local industrial policy process could be described as a kind of 'democratic corporatism' with continuous political bargaining between local governments, firms, interest groups, regional state bureaucracies and political parties (Olsson 1989).

Further, there has been an increase in the number of local-government companies, foundations and trusts. The rationale for these kinds of institutions that they are not subject to the same kind of publicity and political accountability as more traditional public organizations. They are therefore considered to be more suitable for meeting new demands from the public. Proceedings can also be faster and the entire process more business-like. In sum, the development which has been outlined illustrates an increasing interdependency between central and local government agencies and private institutions (Gustafsson 1987).

In the beginning of the 1980s the Swedish municipalities and counties found themselves in a new situation. The rate of increase in local-governmentspending in local tax rates became considerably lower, partly as a belated response to the restraining efforts made by the central government since the middle of the 1970s (Ysander & Nordström 1985). During the period 1980-85 central grants declined in real terms by 0.8 percent per year. This decline was even greater than in many other welfare states (Wolman et al. 1988). Changing the grant system has been explicitly stated as a strategy to strengthen the financial control of the public sector in general and the local-government sector in particular. However, this is not just something going on behind the process of decentralization and deregulation. These latter programmes are viewed as important instruments

to increase efficiency and to reduce costs in the local-government sector. Considering the local governments' large share of the public sector. it is no wonder that central government tried to shift responsibilities for crisis management downwards.

Resources and discretion. - According to the conceptual framework used in this article local-government discretion is a concept which summarizes the potentialities given by different sets of resources. Thus, for a long time local government in Sweden has had a strong constitutional-legal foundation, backed up by a set of laws regulating local-government relations to central government and to the citizens. It has its own fiscal rights. Its political organization is based on direct and proportional elections making local government somewhat more qualified than deconcentrated state administration. Finally, its employed administrators and field workers represent a strong professional competence. Taken together these factors represent a set of resources giving local government a strong potential for discretionary action.

Recent local-government reforms in Sweden have affected the resource dimensions in a somewhat contradictory way. On the one hand they have been announced as radical decentralization and deregulation reforms seemingly increasing local-governmental discretion and definitely doing so in a legal and regulatory sense. On the other hand they are also an expression of increasing central-state control over local-government expenditure. message from central to local government may be summarized like this: yon are freer than before to do as vou want but within a narrower financial framework.

The Case of the US

The US local-government structure is rather complicated, with central government, state governments, counties, municipalities, townships and towns, school districts and special districts. When discussing central-local government relations in a general, cross-national perspective, however, it is appropriate to focus upon the relationship between the federal state and the lower levels of government. Regarding the scope of local (non-federal) government, its share of total public expenditure corresponds to that of Poland, but is somewhat smaller than in Sweden. Thus, when President Reagan took office after 1980. state and local expenditure stood at 37 percent of total public expenditure, compared to 35 percent four years later (Wolman 1988b, 431).

Constitutional-legal framework. - The principle of local autonomy is a
central tenet of the 1789 US Constitution. The powers originally delegated

to the federal state were clearly specified, e.g. including the collection of taxes, the regulation of foreign commerce, the coining of money and the declaration of war. According to the Tenth Amendment to the Constitution,adopted the First Congress in 1789, 'the powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people'. Despite this seemingly clear-cut division of power, the Constitution also contains the 'implied powers' clause (Article 1, Section 8), which has allowed the Congress to extend its power into areas unknown and unanticipated by the authors of the Constitution. The ambiguity inherent in these two constitutional principles has given rise to a 200-year-long debate among federalists (centralists) and anti-federalists (decentralists) over its interpretation.'The is constantly and incessantly debated, characterized by periodic swings of the pendulum in response to broader social, economic, and political forces and events' (Lake & Regulska 1989).

Until fifty years ago the accepted legal and practical principle of federalism that under a federal system of government, 'each level of government exercises its assigned powers independently of the other' (Peterson 1984, 223). However, both in doctrine and in practice the Roosevelt started a development of extensive public-sector growth that implied a growing overlapping of functions and authority between the different levels of government. This extensive use of the 'implied powers' clause was not seriously challenged until President Ronald Reagan launched his alternative vision of federalism. Before highlighting this 'New Federalism', it is useful to look a little at the political and financial dimensions of the US case in a historical perspective. Due to the ambiguity in the Constitution mentioned previously, it is appropriate to look at these dimensions together, to see how the Constitutional principles have been used in its federal context.

Financial and other non-constitutional resources.' - Throughout US history the constitutional structure of central-local government relations has remained fairly constant. What has changed, and developed radically, over time has been the allocation of other resources. Centralization and decentralization have been means to accomplish other goals rather than ends in themselves: 'U.S. history traces a fluctuating process of centralization decentralization in response to economic and political crises' (Lake & Regulska 1989).

The 'New Deal' announced by President Franklin Roosevelt at his inauguration in 1933, and lasting until 1939, engaged the federal governmentin and activities that were unprecedented in American history. In Roosevelt's first 'hundred days' in office, Congress passed legislation that imposed federal regulation and control of banks, industry, agriculture,

labour and social welfare. This massive expansion of central-government functions signified a remarkable degree of centralization, and a correspondingdegree involvement into the local government and private spheres.

Thus, President Lyndon Johnson's 'Great Society' could build on the New Deal for its ideological, political and legal justification. The two main goals of the Great Society programme were: (1) to improve the quality of American life through environmental regulation, crime prevention and the like; and (2) to improve minimum living standards for those in poverty, through job training, education, health services, income support and other means. It was the role of the federal government to command this War on Poverty.

The principal mechanism for implementing the federal goals was the grant-in-aid. a direct grant from federal to local government aimed at a particular programme objective. Limited in scope earlier, these grant programmes exploded under the Great Society. The implication for federal-state-local was a strong centralization of initiative, resources and discretion in the federal government that surpassed even the New Deal. Local discretion was strongly subordinated to national goals. By categorical grants the federal government determined the priorities of the local governments. Indeed, the latter's scope and number of functions increased, but at the price of a lower degree of autonomy vis-å-vis the federal government.

The attack on poverty by extensive federal programmes eventually provoked The fragmentation of grants led to duplication and overlap, for example, so that local officials were confused as to what programmes were available for what purposes. Application procedures were complicated, time consuming and burdensome. Local concerns were deemed of lower priority or were ignored altogether. Detailed federal regulation decreased local discretion and paved the way to local opposition, and to counterattacks from the Republican Party with its "New Federalism'.

When President Richard Nixon came to office in 1969, he was determined to reform intergovernmental relations in a decentralist direction. Federal grants were redesigned from categorical to block grants and a system of Federal Revenue Sharing was introduced. Both innovations were designed to be used by local governments for any purpose. Yet the implications were confusing. On the one hand, local discretion increased substantially with the elimination of central administrative control over those funds. On the other hand, once every local governmental unit was a recipient of federal funds, local dependence on the continuation of those funds grew significantly.The towards less regulated federal grants was continued under Presidents Ford and Carter, although the federal role as such grew

continuously. Thus, federal grants to state and local governments grew
from 7.6 percent of the federal budget in 1960 to 15.5 percent in 1980.

The new federalism. - According to President Ronald Reagan:

the centralized federal government had become overloaded, assuming more responsibility than it could handle. Accountability between voters and government had hemorrhaged, and, too often, state and local officials were judged more by their grantsmanship in Washington than by their responsiveness and effectiveness in governing. (Williamson & Foods 1986, 11)

Reagan's vision of federalism has been summarized in three phrases: 'separation of powers', 'devolution of responsibilities to governments that are closer to the people' and 'less spending by all levels of government' (Peterson 1984, 222-223). There seems to be a relative consensus that the third of the three goals was given the highest priority by the Reagan administration:

So there is an irony to the Reagan record on intergovernmental relations. The administration set before the nation a highly ideological program of federalism reform, many of whose recommendations lay outside the mainstream of conventional thinking. The formal New Federalism never got off the ground. The principle of radical devolution of program authority was rejected by Congress, the states, and the electorate. However, the same principle inspired a devolution of management responsibilities (program implementation) that the states have seized upon and that has helped rationalize the operation of intergovernmental (Peterson 1984, 218)

The conclusion just quoted pertains to President Reagan's first period in
government, but it seems to be confirmed also by studies concerning the
second period (MacManus 1988; Wolman 1988b; Lake and Regulska 1989).

Since Reagan also advocated large increases in military spending, his cuts were heavily concentrated in domestic social programmes to state and local governments. These cuts were cloaked in a heavy rhetoric of decentralization. In his inaugural speech he initiated a debate on federalism, aiming to clarify the relationship between levels of government. According to the President's National Urban Policy Report (1982) it is the 'policy of this Administration to return maximum authority and discretion over the use of resources to State and local governments . . . State and local governments have amply demonstrated that, properly unfettered, they will make better decisions than the Federal Government acting for them' (as quoted from Lake & Regulska 1989, 28).

It should be noted that decentralization in accordance with the President's interpretation of federalist principles meant two somewhat contradictory things. It did not only enhance state-government power with respect to the federal government, but with respect to local governments as well. Direct federal-local links were broken, and federal aid was restructured from local to state governments. From the local governments' point of view this rather meant a centralization (Wolman 1988b, 26).

The real effects of President Reagan's policy on central-local government relations were double-edged. On the one hand, Reagan continued President Nixon's policy of redesigning grants from categorical to block grants. On the other hand, Reagan eliminated two large programmes introduced by the Nixon administration for distributing federal funds to local governments for community development and job training. According to Lake and Regulska (1989, 29) 'the negative impact of Reagan's budget cuts in withdrawing financial resources from local governments far outweighed whatever beneficial impact may have been provided through his rhetoric of decentralization'. In Wolman's words: 'grant reduction decreased local autonomy by diminishing the resources actually available to subnational level governments' (Wolman 1988b, 429).

Resources and discretion. - The above-mentioned ambiguity written into the Constitution, 'the dual federalism' (Cole & Taebel 1986, 6). resulted in a gradual and incremental specification of State and local powers. However, the Great Depression resulted in a massive federal intervention in local affairs. Thereby local government did not necessarily lose functions, but those became intertwined with the federally imposed functions. Local governments got more resources, but they also became more dependent on federal grants. During the 1980s these grants were partly drawn back by the New Federalism under President Reagan. In the same vein the transformation of grants from categorical to block grants meant a transfer of regulatory resources in favour of the local governments.

Federal fiscal assistance to state and local governments 'declined significantly' Reagan (MacManus 1988) making it 'more difficult for subnational governments to support existing spending levels" (Wolman 1988a, 429). Politically, the decrease of federal programmes and grants resulted in less dependence on federal grants by local governments. In a way this could be seen as an increase in local-government political resources. The change in the grant system also meant a deregulation in favour of local-government discretion, although 'deregulation' in general seemed to be more beneficial to the business interests (Conlan 1986. 37-38). this 'gain' did not outweigh the financial cuts. Thus, the states and the local governments were somewhat freer to do what they wanted, but with less financial resources. Finally, in terms of professional resources the decrease in federal grants to local governments may have caused some losses in competence of the latter, as the grants were generally ear-marked, presupposing the recruitment of specialized professionals.

Conclusions

Poland, Sweden and the US are three strikingly different countries within

distinct political, economic and historical settings. Still, recent developmentsof government relations show some interesting parallels. Developments appear to be motivated by similar causes and seem to have similar outcomes for local governments. The central governments in the three countries from the late 1970s to the early 1980s have responded to crises with a devolution of responsibilities to local governments. However, this devolution has mainly affected regulatory and other nonfinancialresources, nothing to local financial resources. On the contrary, in the last respect decentralization has had the effect of increasing local government's dependence on financing from other sources than the central state. Local governments have tried to mobilize new resources, e.g. through property and sales taxes in the US, through a local wage tax in Poland (Lake & Regulska 1990), and through privatization and different paragovernmental arrangements in Sweden. To a certain extent these measures have been able to counterbalance a decrease in central-state grants. Generally speaking, however, devolution of responsibilities from central to local governments is closely related to diminishing local-governmentfinancial Thus, decentralization in all three countries seems to provoke different forms of privatization.

In the countries compared the pressure for decentralization was mainly economic, although it was tightly linked to political and ideological reactions overcentralization. Of course, in Poland the last aspect was much stronger, as the whole political system had lost all its legitimacy. Current developments in Poland might result in a restoration of the prewar system that is firmly established in Sweden and the US, although the deep distrust of the Party-dominated local-government system may be a decisive hindrance to such a development. As argued by Regulski:

Bad experiences in the past have given evidence about the inefficiency of local policies and the helplessness of councillors. In effect, the population is restricted in its involvement in local affairs and very little attention is paid to councils' activities. People have forgotten what an active and autonomous council can achieve (1989, 434).

Indeed, the political developments in Poland are dramatic compared to the relatively modest changes that have been going on during the 1980s in the two other countries. Nevertheless, it is striking how the direction of development has been very much the same in all three countries.

Regarding the earlier extremely centralized party-state system of Poland, decentralization strongly affects all resource dimensions in that country. On the other hand, in Sweden and the US, decentralization along mainly the constitutional and regulatory dimensions is counterbalanced by increased central-state control of financial resources. This is a good illustration of the thesis that centralization and decentralization seldom occur as political goals in themselves, but are more often means to achieve other ends. As

Wolman puts it in a comparative study on Thatcher's Great Britain and Reagan's US: 'the relative degree of centralisation/decentralisation is thus a political tactic rather than a political objective' (Wolman 1988b. 427). At least in the case of Poland 'tactic' is a misleading word, as the proposed local government reform would rather imply a radical restructuring of central- and local-government relations. Indeed, it would be more correct to talk about strategically motivated rebirth of local self-government.

Thus, the approach applied in this article has been able to demonstrate at least three points. First, it proved useful to compare central-local government relations in three countries representing three radically different The purpose was not to confirm the obvious differences between the countries, but, on the contrary, to see if the current restructurings central-local government relations had something in common. Second, the analysis focused on the development of central-local government and did not contend with overall comparisons of the past or the present. Thus, the examination of these relations did not limit itself to patterns that prevailed prior to the late-19705. Indeed, the focus was very much oriented towards analysing recent development of these relations. Third, a multi-dimensional framework was used, which made it possible to give a nuanced picture of the processes at work, not forcing the conclusions into trite statements such as 'Poland is more centralized than the United States'. This may be true as regards one or some of the resource dimensions, but the framework helped to qualify such a conclusion.

As for the explanatory power of the framework, it did not claim to explain anything in itself. It was rather used as an analytical basis paving the way for explanation of similarities and differences between the three countries. The relative crudeness of the analytical framework should be seen in relation to the earlier absent, or meagre, efforts in this branch of political science. One has to start somewhere.

The interepretation used to explain the causes for central-local government in the three cases can be related to three fundamental dimensions. First, the economic pressure on the three central governments is obviously connected with the position of each country's economy in the international capitalist order (Lake & Regulska 1990). State budget deficits in Sweden and the US and large foreign debts in Poland have been triggers to initiate strategies with deregulation and decentralization as central elements. Second, in all three countries there have also been ideological pressures for decentralization coming from below, although in Poland the pressure for economic decentralization has been stronger than for political decentralization. Third, pressures for decentralization could also be related to interorganizational 'ungovernability' emanating from 'demand overload" (cf. Held 1987, 229-242).

In sum, the approach has been able to demonstrate that three countries

with radically different political systems have been confronted by similar problems, and that in spite of these differences they have reacted in similar ways. It has also been able to sketch a general causal interpretation. However, the approach obviously has its limits. To be able to examine the causes, mechanisms and effects of the restructuring of central-local government relations in depth, one has to take the analysis further along at least two paths. First, one has to relate the framework more closely to an analysis of the countries' positions in the world capitalist order. Second, one has to pursue historically oriented, comparative case-studies, including closer examination of specific policy areas (cf. Page & Goldsmith 1987, 161).

---

NOTES

* This paper emanates from a research project on the Impact of Political Systems on Urban Policy: A Comparative Analysis of Poland, Sweden and the United States, including Robert W. Lake and Joanna Regulska, Rutgers University, US; Jerzy Regulski Wlodeck Kocon, University of Lödz and Polish Academy of Sciences, Warszaw, Poland; Berth Danermark and Ingemar Eländer, Örebro University, Sweden. The Swedish part of the project has been sponsored by the Swedish Council for Building Research. Thanks to Jon Pierre and an anonymous referee for valuable comments on an earlier version of the manuscript.

1. This section will draw heavily upon Lake & Regulska (1989).

REFERENCES

Bacharach, S. & Lawler, E. 1980. Power and Politics in Organizations. London: Jossey-Bass.

Cole. R. L & Taebel, D. A. 1986. The New Federalism: Promises, Programs, and Performance',
The Journal of Federalism 16, ?—lO.

Conlan. T. J. 1986. 'Federalism and Competing Values in The Reagan Administration',
Publius: The Journal of Federalism 16, 29-47.

Dagens Nyheter 1990. 18 January.

Dente. B. & Kjellberg. F.. eds. 1988. The Dynamics of Ins'.itutional Change. Local Government
in Western Democracies (Sag« Modern Politics Series Vol. 19). Sage:
Beverly Hills and London.

Elander. I. & Montin, S. 1990: 'Decentralization and Control: Central and Local Government
Relations in Sweden', Policy and Politics 18, 165-180.

Elmorc, R., Gustafsson. G. & Hargrove, E. 1986. 'Comparing Implementation Processes in
Sweden and the United States', Scandinavian Political Studies 9, 209-233.

Foundation in Support for Local Democracy. 1989. Wspolna 52/54, atp. 31. Warszaw, Poland
(mimeo).

Goldsmith, M. & Villadsen. S., eds.. 1986. Urban Political Theory and the Management of
Fiscal Stress. Aldershot: Gower.

Goverde. H. & Markowski, T. 1986. 'The Structure of Government in Poland: Some Aspects of an Organizational System in Change', in Goverde, H., ed., Cross-national Comparison between East and West. The Case of Local Government and Urban Government and Urban Development in Poland. Nijmeegse Planologische Chaiters no. 22. Geografischen Planologisch Instituut Kath, Universiteit Nijmegen, Holland, 35-52.

Gustafsson, G. 1987. 'Deregulation as Decentralization or Privatization - Changes in Swedish Central-Local Government Relationships". paper to the European Consortium for Political Research in Amsterdam 11-15 April 1987. Umeå: University of Umeå, Department of Politics (unpublished).

Hancher, L. & Moran. M. 1989. 'Introduction: Regulation and Deregulation'. European
Journal of Political Research 17. 129-136.

Hanf. K. 1989. 'Deregulation as Regulatory Reform: The Case of Environmental Policy in
the Netherlands', European Journal of Political Research 17, 193-207.

Harloe, M. 1981. 'Notes on Comparative Urban Research', in Dear. M. & Scott. A., eds..
Urbanization and Urban Planning in Capitalist Societv. London and New York: Methuen.
179-195.

Held, D. 1987. Models of Democracy. Cambridge and Oxford: Polity Press.

Lake, R. W. & Regulska, J. 1989. "Changing Central-Local Government Relations in the U.S.' US Monograph prepared for the Research Planning Group on the Impact of Political Systems on Urban Policy: A Comparative Analysis of Poland. Sweden and the United States, meeting in Warszaw. Poland. 10-20 June 1989. New Brunswick: Rutgers University. Center for Urban Policy Research and Department of Geography (unpublished).

Lake, R. W. & Regulska. J. 1990. 'Political Decentralization and Capital Mobility in Planned
and Market Societies: Local Autonomy in Poland and the United States". Policy Studies
Journal 18.

MacManus, S. 1988. 'Transferring Functions among Local Governments and the Private Sector in High Growth Areas: Divestiture Strategies to Control Public Sector Expenditures'. paper prepared for the International Conference on Intergovernmental Decentralization sponsored by the US Advisory Commission on Intergovernmental Relations and Pion Publishers. Washington DC, 22-24 February 1988. Tampa. Florida: University of South Florida, Department of Public Administration and Political Science (unpublished).

Meckstroth. T. 1975. ' "Most Different Systems" and "Most Similar Systems": A Study in the
Logic of Comparative Inquiry". Comparative Political Studies 8. 132-157.

Miliband. R. 1969. The State in Capitalist Societv. London: Weidenfeld & Nicolson.

Mokrzycki, E. 1982. 'What to Take into Account when Comparing? The Problem of Context". in Niessen, M. & Peschar, J., eds.. International Comparative Research. Problems of Theory, Methodologv and Organisation in Eastern and Western Europe. Oxford: Pergamon Press, 45-56.

Murray, R. 1985. 'Central Control of the Local Government Sector in Sweden'. in Gramlich,
E. M. & Ysander. 8.-C. eds., Control of Local Government. Stockholm: The Industrial
Institute for Economic and Social Research, 195-345.

Olsson, J. 1989. 'Forhandlingsekonomiska tendenser i den lokala nåringspolitiken - exemplet Sverige" (Negotiations in Local Economic Policy - The Case of Sweden). in Nielsen. K. & Pedersen, O. K.. eds.. Forhandlingsokonomi i Norden. Charlottenlund. Denmark: Juristog Forlag, 271-295.

Page, E. & Goldsmith, M.. eds.. 1987. Central and Local Government Relations. A Comparative
ofWest European States (Sage Modern Politics Series Vol. 13). London:
Sage.

Peterson, G. E. 1984. "Federalism and the States. An Experiment in Decentralization'. in
Palmer, J. L. & Sawhill. I. V.. eds.. The Reagan Record. Cambridge. Mass.: Ballinaer
Press, 217—259.

Report, SB:18. Gavle: The National Swedish Institute for Building Research.

Regulski, J. 1989. 'Polish Local Government in Transition', Envinmment and Planning C 7.
423-444.

Regulski, J. & Markowski. T. 1988. 'The Position of Local Authorities in the Socialist Countries of Eastern Europe". in Goverde. H.. ed.. Cross-national Comparison between East and West. The Case of Local Government and Urban Government and Urban Developmentin Nijmeegse Planologische Chaiters no. 22. Geograrisehen Planologisch Instituut Kath. Universiicit Nijmegen, Holland, 21-34.

Regulski, J. and Regulska, J. 1988. 'Urban Policy and Political Systems", in Goverde. H..
ed., Cross-National Comparison between East and West. The Case of Local Government

and Urban Government and Urban Development in Poland. Nijmeegse Planologische
Chaiters no. 22. Geografischen Planologisch Instituut Kath. Universiteit Nijmegen,
Holland. 5-20.

Regulski, J., Georg, S., Jensen, H. T. & Needham,B., eds., 1988. Decentralization and Local
Government. A Danish-Polish Comparative Study in Political Systems. New Brunswick and
Oxford: Transaction Publishers.

Rhodes, R. A. W. 1980. 'Developed Countries', in Rowat, D. C, ed., International Handbook
on Local Government Reorganization. Contemporary Developments. London: Aldwych
Press, 563-581.

Rhodes, R. A. W. 1981. Control and Power in Central-Local Government Relations. Aldershot:

Rhodes, R. A. W. 1986a. '"Power Dependence". Theories of Central-Local Relations: A
Critical Reassessment', in Goldsmith, M., ed., New Research in Central-Local Relations.
Aldershot: Gower, 1-33.

Rhodes, R. A. W. 1986b. The National World of Local Government (The New Local
Government Series 26). London: Allen & Unwin.

Rowat. D. C. 1980. 'Concluding Remarks', in Rowat, D. C, ed., International Handbook
on Local Government Reorganization. Contemporary Developments. London: Aldwych
Press, 594-603.

Sharpe, L. J. 1988. 'The Growth and Decentralisation of the Modern Democratic State',
European Journal of Political Research 16, 365-380.

Strömberg, L. & Westerståhl, J. 1984. The New Swedish Communes. A Summary of Local
Government Research (Research Reports 1984:1). Gothenburg: University of Gothenburg,
Department of Political Science.

Williamson, R. S. & Foods, B. 1986. 'A New Federalism: Proposals and Acheivements of
President Reagan's First Three Years', Publius: The Journal of Federalism 16, 11-28.

Wolman, H. 1988a. 'Decentralization: What It Is and Why We Should Care', paper prepared for presentation at International Conference on Intergovernmental Decentralization, Washington 22-24 February 1988. Detroit: Wayne State University (unpublished). Wolman, H. 1988b. 'Understanding Recent Trends in Central-Local Relations: Centralisation in Great Britain and Decentralisation in the United States', European Journal of Political Research 16, 425-435.

Wolman, H., Goldsmith, M. & Kousgaard, P. 1988. Local Government Fiscal Condition: The
Capacity for Action. Detroit: Wayne State University (unpublished).

Ysander, 8.-C. & Nordström. T. 1985. 'Local Authorities, Economic Stability and the Efficiency of Fiscal Policy', in Gramlich, E. M. & Ysander, 8.-C, eds., Control of Local Government. Stockholm: The Industrial Institute for Economic and Social Research, 347-98.

Zawadzka, B. & Zawadzki, S. 1980. 'Socialist Countries of Central and Southeastern Europe',
in Rowat, D., ed.. International Handbook on Local Government Reorganization. London:
Aldwych Press, 363-374.