If Territorial Fragmentation is a Problem, is Amalgamation a Solution? An East European Perspective

Introduction

Since the 1990 political breakdown in Central-East Europe (CEE), the lowest level of local territorial organisation has changed in several countries. Territorial fragmentation has been a common tendency in the Czech Republic, Slovakia, Hungary, Macedonia, Slovenia and a few other countries. Quite often, it has simply been a reaction to an earlier consolidation imposed by the respective communist governments in an undemocratic manner, without public consultation (for example the situation in former Czechoslovakia or Hungary).1

After 1990, decentralisation and paradigms of local autonomy have often been understood as ways of legitimising separate local governments in almost every settlement unit, even the smallest village. Attempts to create (or to keep) larger territorial jurisdictions have been seen as violation of local autonomy. As a result, in several countries there have been a significant proportion of very small-scale authorities, many of them with fewer than 1,000 residents. One may find anecdotal examples of the villages of Bidovce (Czech Republic) or Prikry (Slovakia) with populations under 10 citizens.2

A summary of the fragmentation process is provided in Table 1. In some cases the process was rapid and relatively short, completed a year or two after the political changeover, while in other countries it continued throughout the 1990s or even longer.

Table 1  . Process of territorial fragmentation of the municipal tier in selected Central and East European countries after 1990

Country	Number of municipalities		Increase in the number of municipalities (in %, 1989 = 100)
	After the fragmentation process
Bosnia	109	143 (1995)	131%
Croatia	115	543 (1998)	472%
Czech Republic	4104	6230 (1993)	152%
Hungary	1368	3133 (1992)	229%
Macedonia	30	123 (1995)	410%
Romania	2948	3190 (2008)	108%
Slovakia	2694	2875 (1998)	107%
Slovenia	62	210 (2006)	339%
Source: Author's calculations based on Horvath (1999), Kandeva (2001) and Swianiewicz (2002), and Kreci & Ymeri in this issue.

On the other hand, there are also examples of countries that have maintained territorially consolidated systems (such as Yugoslavia/Serbia, Montenegro, Bulgaria, and Poland) where the median size of a local government unit is much larger, and none have fewer than 1,000 residents. But, in general, the phenomenon of territorial fragmentation of the lowest tier has been widespread. Figure 1 provides an illustration of this phenomenon in individual countries of the region.

Figure 1. Distribution of municipal countries according to their size in the fragmented systems of Central. Eastern and South-Eastern Europe.

Very soon after 1990, territorial fragmentation came to be considered one of the major barriers to decentralisation and effective functioning of the local government system. With a different degree of intensity, such voices could be heard in Albania, Macedonia, Moldova, Hungary, Slovakia, Czech Republic, Latvia, Estonia, Ukraine, Armenia, Azerbaijan, Georgia, and perhaps in some other countries too. In a different context, a similar discussion was conducted in Poland, where the size of upper tiers of the sub-national government –powiat and województwo– was discussed. Parallel debates also took place in Bulgaria, Montenegro and Poland, where bottom-up pressure to split existing lager municipalities was occasionally pushing towards increasing territorial fragmentation.

But in spite of the fact that the territorial consolidation reform is often presented as a recipe for various problems, in fact, territorial reforms have not usually been implemented. Problems with implementation are typical not only of Eastern and Central Europe. Baldersheim & Rose (2010, forthcoming) quote Norway as an example of a country where territorial consolidation has been continuously discussed for over 15 years, but nothing as yet has been done about it.

Why is Territorial Fragmentation a Problem?

Perhaps the question in the title of this section might be reformulated: is territorial fragmentation a problem at all? The debate concerning this issue is not a unique CEE phenomenon. It has also been conducted in the US and Western Europe, especially in relation to the organisation of local governments in metropolitan areas. Also, in a wider context, the optimal size of local government is one of the most commonly discussed issues. Baldersheim & Rose (2010, forthcoming) quote publications by Dowding et al. (1994) and Boyne (2003), which both include an overview of the empirical search for an optimal local government size. The former includes 65 references, while the latter as many as 190, and – as Baldersheim and Rose note –‘it is probably just a tip of an ice-berg’. A comprehensive summary of the theoretical discussion may be found in Sharpe (1995) and Keating (1995), while an econometric interpretation is given in King (1984). An overview of the theoretical discussion as well as some of the empirical studies presented in the context of East and Central Europe may be found in Swianiewicz (2002). There is no need to present the whole discussion once again, so in this article I will very briefly recapitulate the most classic arguments.

The arguments in favour of territorial consolidation leading to creation of large sub-national jurisdictions may be summarised as follows:

	There is economy of scale, which makes it possible to provide services more cheaply (or more effectively) in the larger local government units. The most straightforward evidence of this rule has been presented in the context of municipal administration expenditure; see Swianiewicz (2002) for evidence from Poland, the Czech Republic, Hungary and Bulgaria. As a result the excessive proportion of small local budgets is often devoted to administrative overheads.
	Larger local governments have more capacity for a wider range of functions, so territorial consolidation would enable more services to be allocated to the local level.
	Since large local governments may perform more functions, it is more likely that citizens will participate in local politics (Dahl & Tufte, 1973). According to this interpretation, consolidation helps to promote local democracy.
	A territorial organisation with large local governments results in fewer income disparities among municipalities, so there is less pressure on horizontal equalisation systems, which may be politically sensitive and/or costly for the national budget.
	Large local governments may be more effective in planning and economic development policies.
	In a territorially consolidated system it is easier to reduce the problem of free-riding, i.e., situations when provided locally services are used by residents who live (and pay local taxes) in another jurisdiction. In other words, consolidation helps to reduce the mismatch between administrative boundaries and the catchment areas of services.

Let's take a closer look at the second argument concerning the relationship between territorial consolidation and the scope of functional decentralisation, since it often plays a crucial role in discussions conducted in countries of Central and Eastern Europe.

As we can see in Figure 2, there is a positive relationship between the mean size of municipality and share of municipal spending in relation to GDP.3 The most noticeable diversion from the relationship is Portugal, with very large local governments, but a narrow set of functions performed by municipalities. But in other countries the rule saying that more functions are allocated to municipalities with more territorially consolidated systems is very clearly observed.

Figure 2. Relationship between territorial consolidation and functional decentralisation.

However, it needs to be mentioned that the same rule is not necessarily followed by the countries of Central and Eastern Europe. Of course, the amount of municipal spending is just one of several indicators of functional decentralisation, certainly not the most important one, but easiest to use for international comparisons. There are equally significant indicators, such as discretion in deciding on local revenues and spending allocation, number of national norms and standards binding local governments in service delivery, extent of formal and informal influence national administration has over local decisions, etc. Nevertheless, the graph in Figure 3 is meaningful – there is no relationship between the size of municipalities and the size of municipal finance. There are countries (such as Azerbaijan and Armenia) in which territorial fragmentation co-exists with the minimal role of municipal budgets, or others, such as Poland, in which the large size of basic sub-national jurisdictions corresponds to a wide scope of locally delivered public functions. But there are also opposite examples – Macedonia represents a much lower and Hungary a much higher share of spending compared to predictions that can be made based on the size of municipalities.

Figure 3. Relationship between territorial consolidation and functional decentralisation in CEE countries.

According to a previous statement (Swianiewicz, 2002: 302), ‘in Central and Eastern Europe, it is quite clear that the level of territorial consolidation has had some importance on a municipal level, and yet it has not been a decisive factor in functional decentralization. Several other factors … such as political determination to stick with the decentralization agenda have played much more important roles'. Indeed, territorial consolidation makes it possible to allocate more functions to the local level, but it does not guarantee decentralisation. It may happen (as we will see in the example of Georgia) that the territorial reform introduced under the banner of strengthening local governments is not followed (or accompanied) by municipalities being allowed to take on new responsibilities. On the other hand, countries such as Hungary decided to devolve a wide variety of functions to the local level in spite of their territorial organisation. As a result, they are experiencing various problems and are looking for solutions to make delivery of these functions viable in the ase of very small jurisdictions.

But advocates of territorial fragmentation also have powerful arguments on their side. The defence of small local jurisdictions is usually based on democracy-related arguments. In small communities the link between citizens/voters and local representatives is closer. There is more mutual trust, the flow of information is easier and, as a result, local governments are more accountable to local populations. There are several studies indicating that citizens of small municipalities are more satisfied with local government performance. Although, as studies by Hajnal (2001) and Borecky & Prudky (2001) on Hungary and the Czech Republic show, this rule does not concern tiny municipalities with fewer than 1,000 citizens. In such cases perhaps the low functional capacity of the local government overshadows the ‘sociological beauty’ of a small community, and negatively influences popular perceptions of local government performance. In most countries, citizens' interest in local public affairs, reflected in the turnout in local elections, is usually higher in small municipalities (for evidence of this in CEE countries, see Swianiewicz, 2002).

To sum up, while consolidation is usually promoted on the grounds of the efficiency or functional capacity of local governments, fragmentation is often defended on the grounds of local democracy. In a popular perception there is a trade-off between democracy and economic efficiency. In that sense there is no universal answer to whether consolidation is better than territorial fragmentation, or vice versa. The answer depends on the values of local governments which are at the top of political agendas. If decentralisation is seen first of all as a tool to ensure better provision of public functions – as it is, for example in Scandinavia or the UK, see Goldsmith (1996) and also in Poland, see Swianiewicz (2003) – then consolidation is more understandable. But if the focus is on the ‘natural’ perspective of the local government being an emanation of the local community (as traditionally it has been in Southern Europe), then there is a case for defending territorial fragmentation.

However, a recent research report by Mouritzen (2008) challenges this simple conclusion, even if earlier research by the same author – e.g., Mouritzen (1989) – promoted the idea of a ‘trade-off’ between the two systems. Mouritzen (forthcoming) concludes that empirical evidence suggests that the (negative) effect of the size of local government for democratic performance is nowadays less significant than 25–30 years ago. Speculating about possible reasons for this change he mentions:

	Local bureaucracy that acts in a more impersonal, professional way across the country, regardless of the size of the local government;
	An explosion of information, which is easily available everywhere;
	Homogenisation of daily lifestyles.

Assuming that Mouritzen is right in his conclusions, at least two important questions are brought forward:

If homogenisation, the information explosion and the professionalisation of local administration reduce the impact of local government size on democratic capacity, can they also reduce the positive correlation between the size and ‘system capacity’ (dependent, for example, on the assumption that bigger municipalities have better access to human resources)?

Are the same factors explaining the decreasing role of the size in democratic performance also significant in countries of Central and Eastern Europe? Perhaps there are important differences between Western and Eastern Europe, as well as variations among countries of the region in our focus. In general, the level of professionalisation in local administration is much lower in Eastern than in Western Europe; see Campbell & Coulson (2006) comments on attempt to implement New Public Management (NPM) in Eastern Europe. In some Eastern Europe countries, the factor Mourtizen calls ‘homogenisation of lifestyles’ is also much less pronounced, while the differences between capitals (or major cities), and small, remote towns and villages, are still very much visible.

There is also a range of empirical studies, referring to the ‘trade-off’ between system capacity and citizens' effectiveness, which try to seek a ‘balance point’ between the beauty of democratic performance in small communities and economy of scale that can be achieved only in larger local governments. Looking for an optimal size of local jurisdictions resembles the search for ‘the philosopher's stone’ and leads to very different results, depending on assumptions and indicators used in the model. For example, in a Polish study, an index of the local government performance was constructed on the basis of variables reflecting democratic performance, service delivery and the ability to promote local economic development (Swianiewicz & Herbst, 2002). The result is presented in Figure 4. The conclusion was that the ‘optimal’ size for the local government is around 30,000 citizens.4 Unlike in the logic of Mourtizen's argumentation, if we go above this figure, the marginal loss in democratic performance is higher than the marginal gain in economic efficiency. Obviously this result should not be taken too literally. The shape of the ‘performance curve’ may be different if other assumptions are made or other variables are taken into account. But regardless of the precise numbers, the most important conclusion is that moderate rather than extreme solutions should be promoted. Both extreme fragmentation (as we know from several countries) and extreme territorial consolidation – as the case of England seems to suggest; see Copus (2010) – might be a problem.

Figure 4. Index of performance and size of local government in Poland.

Concluding, we may say, that although the arguments of economy of scale or democratic deficits are not fully convincing, there is a series of studies5 suggesting that jurisdictions below 3–5,000 inhabitants cannot be given significant public sector responsibilities and such fragmentation will make functional decentralisation difficult and costly. Another formulation of the same conclusion would be that territorial consolidation is about subsidiarity a rebours– the subsidiarity argument is usually read in a way supporting more functions allocated to lower levels. But we may also look from another angle. The subsidiarity principle is about keeping functions as low as possible, supporting small communities in their provision and transferring to the upper tier those functions that cannot be implemented effectively at a local level. Through territorial consolidation we make more space for a number of responsibilities that may be reasonably managed on a municipal level.

Territorial Consolidation Reforms: Why do They Happen and Why So Rarely?

As argued above, the case for consolidation reforms is usually based on economic efficiency, and they are more likely to take place in countries with a primarily functional perspective on the role of local governments. An additional argument, which has recently been used in some European countries, mentions needs arising from the European integration. This is a two-fold phenomenon. It either takes place in a direct way, in which pro-reformers (not quite correctly, but sometimes effectively) argue that a certain shape of sub-national institutions is a requirement of the European Union.6 Indirectly, it is argued that the challenge of competing with regions/cities in other countries of the uniting Europe requires stronger local government institutions. But even if the arguments appear strong, reform is usually very difficult to implement. There are several (complementary) explanations of this difficulty.

First, there is a degree of inertia among the existing institutions, plus there are also actors interested in defending the status quo. As Baldersheim & Rose (2010, forthcoming) conclude: ‘once institutions are established, they set limits to the future choices that are available’. This resistance may be explained on a rational choice level. The consolidation reform would mean a reduction in the number of available political posts (mayors, councillors, etc.). This may mean losing job security or prestige. As Paddison (2004: 34) states bluntly: ‘local (municipal) elites are unlikely to vote for territorial suicide’, though the resistance to change is very common.

Second, the autonomy of small villages is often seen as important by local communities, even if, in practice, this autonomy is more symbolic than real (because a small local government is not capable of performing any significant functions). And indeed, reform may lead to potential negative side effects, which are feared by the local population, and whose importance may be highlighted by opponents of the reform, trying to cover up their own selfish interests (see the previous paragraph) with arguments calling for protection of local communities. These negative side effects may include:

	Accessibility of local administration. In peripheral regions of Central and Eastern Europe, where transport infrastructure is poorly developed and Internet communication is still a future dream, it may be an even greater problem than in well-developed regions of more affluent countries. The physical distance to the town hall makes the life of citizens in remote villages more difficult.
	The identity of local communities. In small towns and villages the town hall (a little like a village school) is not only a place to deal with administrative registers, licences etc., but it is also a local community centre, which provides a feeling of local identity. Liquidation of the local administration may mean not only more difficulty in gaining access to services, but also damage to local social life.
	Fear of not being represented. The local community may be afraid that its voice will not be listened to, or may be disregarded in a distant municipal centre, when it comes to decisions regarding, for example, local investments. Sometimes the fear may be irrational, but sometimes the domination of the central town over the enlarged municipality may be a real issue.
	Conflicts among different parts of the new (amalgamated) municipality. A typical example of a situation that may result in such a conflict is between the core city of an agglomeration and a small suburban town that is being annexed by the central city.

Those negative side effects may of course be addressed by the reform in order to reduce their adverse impact. The office may be organised in a way that will minimise problems related to the distance between the place of residence and the town hall location. There might be special provisions in the electoral law that would prevent overall domination of one part of the municipality (proportional elections systems with voting for party lists may be especially dangerous from the point of view of the representation of small rural villages). There might be special provisions for sub-municipal governments to help represent village interests and promote local identities.

The Politics of the Reform Implementation

Why are some reforms successfully implemented while others fail before they are undertaken or bring about more chaos and conflicts than positive effects? First of all, for the reasons specified in the previous section, it would be naïve to expect that a reform proposal (even if prepared in the best possible manner) will not meet with opposition. As Paddison (2004: 25) notes: ‘it is almost a law of local boundary restructuring, that there will be powerful forces intent on maintaining the status quo’. Paddison also formulates three suggestions that should help in successful completion of the reform:

1. Local government restructuring should meaningfully address local preferences and needs. Paddison opposes the planning paradigm arguing for very large local jurisdictions, which may be difficult to accept for other (social, political) reasons: ‘Local governments should be sufficiently small so that they are able to be responsive’ (2004: 27).

2. Local government revision should be conducted in a fair way, as a transparent and relatively accessible process. Here, Paddison draws our attention to the process of social consultations. It should be reminded that this issue is a formal requirement mentioned in the European Charter of Local Government, so it is obligatory for each country which has ratified the Charter.

3. The territorial reform should be a compromise. It should not be wholly decided by either central or local political elites.

The issue of local consultations is worth an additional comment. The Council of Europe recommends a local referendum, but it also allows other forms of consultations (especially if the local referendum is not allowed by the national legislation). According to the European Charter of Local Governments the result of the consultation is not legally binding for the decision-maker (government, Parliament), but in an obvious way it builds political pressure. But as Copus' (2010) discussion of the UK case shows, it is also possible that the final decision made by the government clearly opposes the popular will expressed in the consultations.

The successful implementation of the reform depends on many factors. For Baldersheim & Rose (2010, forthcoming) the most crucial ones are:

	The ability of the reform promoters (‘policy entrepreneurs’) to frame policies ‘lifting policy ideas out of a “primeval soup” of competing possibilities’.
	The ability of: Policy entrepreneurs to form advocacy coalitions. Opponents to form veto alliances (as we noted before, they would be first of all local politicians, especially from those local governments which may lose their independent status as a result of the territorial reform).
	The institutional context, including the possibility of various stakeholders participating in the decision-making process.

Most frequently, advocates of the reform are recruited from the community of experts or from central governments. But central level politicians must not be treated as a homogenous group – it is those politicians whose political capital is based on support in the regions who often promote regional interests and are likely to defend the status quo rather than the reform (Meligrana, 2004).

There are two additional factors that are perhaps worth looking at in the context of Eastern and Central Europe. The first one is the existence of one specific, powerful actor, namely the international community of donors. In both Macedonia and Georgia – two cases described later in this issue – this factor played an important (in Macedonia perhaps even decisive?) role. Experts brought by foreign donor programmes are not involved in local interest networks, so it is easier for them to formulate radical recipes, and they are often keen to test theories which would be much more difficult to try out in their home countries. The special role of the Council of Europe (CoE) is perhaps worth of more attention. For several countries of Eastern Europe membership of the CoE is the main symbol of their European identity. Through the ratification of the European Charter of Local Governments it imposes the obligation of some decentralisation reform and exposes legislation in those countries to the scrutiny of CoE experts. Although the Charter is not very specific in its stipulations, there are several examples suggesting that this provides one of the most significant incentives for decentralisation reforms. Even if the specific recommendations of CoE experts are rejected (as in case of Georgian territorial reform – see relevant paper in this volume), the impact of the CoE on initiating the debates on decentralisation is hard to overestimate.

The second specific factor is the ability of the reform advocates to act quickly, while the ‘window of opportunity’ is open. As one of the main authors of the Polish decentralisation reforms – Michał Kulesza – rightly stated:

The most favourable moment during which the central bureaucracy (which is usually defending its position) is weak enough to allow any substantial changes is usually very short. The question of time is crucial. If the reformers are not ready to present their concepts and their particulars exactly when it is needed and possible (from the viewpoint of the political situation), then a proper time is probably over … young democracies do not like big structural reforms, which hit economic and political interests of many parties and groupings by destroying their positions and mechanisms present in the functioning of the state, economy and politics … The main factor guaranteeing success is maintaining the fast pace of work. (Kulesza, 2002: 204–205)

The history of the (never implemented) Ukrainian territorial reform devised after the Orange Revolution in 2004 is an excellent illustration of how a political opportunity to implement the reform may be missed (Swianiewicz, 2006).

Territorial reforms are sometimes implemented in two stages. The first one includes adoption of the criteria that should be fulfilled by each local government unit (such as a minimal population threshold, existence of certain local institutions etc.) and allowing for bottom-up re-organisation of local governments based on these criteria. Usually these bottom-up initiatives are additionally stimulated by incentives in the form of funds or other privileges available for ‘front runners’. The Danish case discussed later in this issue is an excellent example of a successful application of such a mechanism. In vast majority of the cases the reform was implemented in this phase, allowing local governments to choose their own specific way of making new arrangements. But the Greek case suggests that this stage may also be a failure. The difference lies perhaps in a consensual versus confrontational political culture, as well as in the belief of local governments that the central level is determined and ultimately able to impose the reform.

Another option in the two-stage reform process is when in the first stage local governments are obliged to co-operate through various local multi- or single-purpose associations (or communities). This helps to build links and trust among actors, before the second stage in which compulsory territorial consolidation is implemented.

Why Consolidation Reforms, If Small Local Governments May Co-operate on a Voluntary Basis?

Some of the opponents of territorial consolidation reforms claim that the problem of small scale may be solved more easily through the voluntary co-operation of local governments. It is definitely an alternative option, but the question is, is it a realistic and effective way to cope with the negatives of territorial fragmentation? A full discussion of issues related to inter-municipal co-operation (IMC) would go well beyond the scope of this article, but it is perhaps useful to make a few related comments.

First, contrary to what advocates of such a solution suggest, bottom-up IMC does not happen very often and is not unproblematic. Actually, the political difficulty of undertaking voluntary co-operation is often similar to that raised by boundary re-drawing reforms. It requires compromises on the particular interests of the individual municipalities involved. Local leaders need to agree on co-operation that will sometimes affect their personal ambitions. Trust among involved actors is a necessary condition for voluntary co-operation to function effectively. It is telling that wherever we find good and plentiful examples of co-operation, they have been supported by a strong set of incentives provided by national or European policies. The nature of those incentives, however, is such that sometimes it may be debatable whether co-operation is still bottom-up and voluntary or rather, imposed by external actors. It has been so in France and Finland – two European cases most frequently cited as positive examples of IMC.7 To some extent it is the same with the recent development of Hungarian and Czech inter-municipal bodies (see Illner, 2010; Pfeil, 2010). Swianiewicz and Lackowska (2007) convincingly prove that relying on bottom-up voluntary co-operation by local governments in Polish metropolitan areas was a failure, i.e., resulted in very poor, insufficient co-ordination of metropolitan functions, or (in some cases) no coordination at all.

But there are also potential organisational and democratic problems related to IMC. In some countries, establishing single-purpose local governmental associations is very difficult, or even illegal, and a favourable national legal framework is not always there.

The joint provision of functions, although it frequently brings financial benefits, requires transaction costs, which may be identified with a complicated organisational-managerial setting. Complex inter-communal arrangements, including the necessity for debating issues by councils of involved local municipalities, may also slow the decision-making process.

The one negative side effect is that co-operative arrangements sometimes make accountability towards local citizens and transparency of decision-making process more problematic. Decisions made by individual municipal councils and mayors have direct electoral legitimacy and are controlled through local civic society institutions (local media, NGOs, individual citizens having access to council meetings and documents produced by local governments), even if this control is often far from perfect. On the other hand, decision-making in local government associations or other co-operative arrangements implies only indirect electoral legitimacy and is less visible and less accessible to the civic society.

Discussing probably the most often quoted French example, Wollmann (2008) identifies several shortcomings of inter-communal cooperation. He mentions among others:

	Duplication of personnel in communes and communautés, which sometimes means additional costs instead of expected savings.
	Overlap of functions between communal and inter-communal bodies, contributing to service delivery problems instead of relieving it. Quoting French reports, Wollmann describes it as a ‘chaotic system of partnerships … in which everybody seeks to seize the entirety of functions’ (2008: 7).
	Deficit of democratic legitimacy, which weakens control and increases probability of corruption.

Also, describing German IMC, Wollmann indicates weak points, which make the territorial consolidation still discussed in some German Länders, in spite of current IMC arrangements (see also Wollmann's discussion of the German case later in this issue). The French case of IMC is discussed extensively in a separate chapter in this volume. The wide acceptance of IMC as an alternative to amalgamation reform starts with the failure of any attempts at territorial reforms in France. Part of the reason perhaps lies in the French tradition of cumul des mandates in which a large group of mayors and local councillors also hold influential posts in central level politics (as MPs or even ministers). They form a very strong lobby, opposing any changes in territorial boundaries and protecting the territorial status quo. On the other hand, the successful delivery of local services in France would probably not be possible without the privatisation of key network services (at least water and sewage) that took place in the 1980s and 1990s (Lorrain, 1997), which allows large public utility companies to serve many small municipalities.

This is all not to say that inter-municipal co-operation is a bad solution. Just the opposite, in fact, as in some cases where political opposition to the territorial reform is too strong, it is perhaps the only realistic option. But it would be naïve to see it as a painless remedy with no negative side effects.

About the Articles in This Issue

The first wave of European territorial consolidation reforms took place in 1950s, 1960s and 1970s, and was rooted in the economy of scale paradigm stressing that better quality local services may be delivered more cheaply in larger local government units. This first wave involved several countries of Western Europe (Scandinavia, the Netherlands, the United Kingdom, Germany) but was also visible in Central-Eastern Europe, where it was implemented without any democratic debate (Poland, Czechoslovakia, Hungary). However, we do not intend to look at the historical examples of the territorial reforms.

Instead, in this volume we try to present experiences gained from major territorial reforms of the municipal tier that have been implemented in Europe during the last decade. In addition to examples from West European countries, which are most frequently discussed in academic literature, we also bring in a specific perspective of East European countries (especially Macedonia and Georgia). East European countries considering territorial reforms often look at models from Western Europe, so putting together cases of Denmark and Germany with those of Macedonia or Georgia is quite interesting. In subsequent country-case articles we try to answer questions related to all stages of policy process: policy-making (the debate preceding the decision), decision-making, policy implementation, and assessment of the outcomes. More precisely, the questions addressed by individual articles include:

	What were the most significant arguments for the reform and possible arguments against it raised in the debates? Were the debates limited to a narrow group of politicians/experts or did they raise interest of the general public (media, citizens' groups, etc.)?
	What was the process of their implementation (including consultations with local governments and the general public)?
	Where the goals of the reform achieved? Were there any negative ‘side effects’? In particular: Is there any evidence of the impact of the reform on the cost of service delivery? What was the perception of the reforms by the general public? What are the changes in citizens' level of interest in local public affairs and local government accountability towards the citizens? Were there attempts (and were they successful) to protect the interests and feelings of identity in small communities that lost their status of autonomous local governments? Are there any provisions for sub-municipal government units in individual local communities?

The examples presented concern both reforms for which the starting point was a relatively consolidated system (Denmark and perhaps also Macedonia fall into this category) and countries that were already territorially fragmented before the reform (Greece, East Germany, Georgia).

In actually implemented boundary reforms (discussed in following articles of this issue) arguments raised by initiators and advocates of the changes, referred to theoretical arguments mentioned in the previous paragraphs. The most frequent motive was an attempt to strengthen the capacity of local governments to effectively deliver a broad scope of public functions. In Greece, the specific variant of this motive concentrated on the ability to absorb EU structural funds by local authorities. Macedonia is perhaps the only exception to this rule, since ethnic policy was a dominant driving force there (interestingly enough, similar ethnic argumentation was on the agenda few years before when the fragmentation of over 30 big municipalities into over 120 smaller was introduced), although in official documents the capacity of municipal government units was mentioned as well.

It is very striking that during the last decade in spite of vigorous discussions conducted in many countries, actual municipal boundary reforms were implemented in very few cases only. The disproportion between the intensity of discussions and actual implementation is very striking, especially in Central and Eastern Europe. Moreover, the only two implemented reforms – Georgian and Macedonian – raise some doubts whether they may be classified as typical territorial consolidation reforms. In case of Macedonia, we witnessed not so much an amalgamation of municipalities, but rather (motivated mainly by ethnic policy) partial reversal of the territorial fragmentation trend observed a few years earlier. Thirty big municipalities ‘inherited’ by Macedonia after collapse of Yugoslavia were first replaced by 120 smaller units, so it is questionable if the following reform introducing 84 municipalities may be called a typical consolidation reform. In Georgia the position of local governments before 2006 was so weak that one may question if the recent reform should be interpreted as boundary restructuring or rather, as a gentle attempt to create real local self-government from scratch. One might also recall the case of Lithuania, where very large municipal government units (over 60,000 inhabitants on average) were created in the second half of 1990s. But it was an element of the creation of the new territorial system introduced instead of extremely centralised Soviet structure, rather than a typical municipal boundary reform, similar to those implemented in Denmark, Greece or other countries.

Discussing factors influencing success or a failure of territorial reforms, I want to indicate three additional elements that are seen clearly in several chapters of this volume. The first one is the role of external forces, which may stimulate change within the political system. These external forces may be of a different nature. In Germany or Greece they might be related to the discourse of international competitiveness within integrating Europe and especially to the absorption capacity of EU structural funds. In Macedonia the territorial reform was to a large extent a follow-up of conditions (or at least strong suggestions) made by foreign donors (especially of IMF). It is not in conflict with the thesis formulated in the relevant chapter of this volume that, if we look at details, we discover that the logic of the reform itself was primarily ethnic politics, and had little to do with concerns for efficiency or local democracy. In a weaker form, the territorial reform was also advocated by foreign donor organisations in Georgia.

Second, it may be no accidental that Nordic countries have adjusted boundaries more easily (and to some extent more rationally) than others (Denmark in this issue may provide an illustration of this claim; earlier reforms in Sweden and Finland fall into the same category). It should be reminded that – like anywhere else in Europe – local governments in these countries are financed primarily by local taxes, imposed on local citizens (with no legal regulations imposing ceiling rates, but leaving tax rates entirely in hands of local councils). In such circumstances, the cost/tax arguments have a direct link with voters, so political support for reforms bringing savings may be easier than in the countries where the cost of service delivery is externalised to the national transfer or tax-sharing systems.

Third, prospects of the territorial reform very much depend on the way local government and its role are perceived in the country. I mention this in the introductory section of this chapter, when I discuss two different traditions (functional and ‘natural’) of the way in which role played by sub-national government in modern states is seen. France and England, both discussed in this article, provide two opposite, extreme cases of the public discourse in this respect. In France territorial amalgamation seems unimaginable, so it is necessary to look for alternative solutions (such as inter-municipal co-operation, perhaps combined with privatisation of many communal services). The situation in England is just the opposite – territorial reforms are introduced ‘too easily’, the technocratic discourse of service provision seems to dominate totally over democratic arguments, and the system, which for a long time has been by far the most territorially consolidated in Europe, seems unable to stop further and further attempts to amalgamate territorial (no more ‘local’) units. In France the idea of ‘liberty of communes’ is a non-negotiable prerequisite. Any decision on change of territorial boundaries has to be accepted by the municipalities concerned. In England, the government may introduce any change relatively easily, even if it ignores the will of the local community expressed in public consultations (see examples quoted in Copus, 2010). What is fascinating is how two opposite ideas (‘amalgamation is impossible’ and ‘amalgamation is easy and should be done’) have dominated discourses and seem to leave no alternative solutions. Both of them are of course rooted in a different paradigm of what local government is about, but these are usually silent assumptions (unspoken in the public debate, and often not even consciously made by the involved parties). France and England provide extreme cases, but perhaps the same line of analysis may be successfully used to explain variation among other countries. For example, in my earlier analysis (Swianiewicz, 2003) I was trying to show how different territorial organisation in Poland (territorial consolidation) as opposed to the Czech Republic, Slovakia and Hungary (territorial fragmentation) are inter-related with a different perception of the role of local governments in a contemporary state.

In the previous sections we discussed potential negative side effects of the territorial amalgamation reform. It is interesting to notice that these were very rarely considered by authors of the reforms. Or, at least, they rarely introduced this topic into political agenda, demonstrating what measures might be adopted to minimise the potential losses. Perhaps the topic only played an important role in the Danish reform. The report suggesting that the reform would not have negative impact on local democracy (which would be expected on the basis of ‘traditional theories') strengthened the determination of the government to implement the planned change.

What was the role of the public opinion in the described reforms? Certainly diverse. In Georgia the reform was simply ignored (or unnoticed) by local communities. In Macedonia, citizens had an opportunity to express their opinion in a national referendum over the reform. In Denmark the role of public consultations was surprisingly low. One could say that the reform, introduced shortly after national elections in which territorial boundary restructuring was not an important issue, has taken everyone by surprise – a bit surprising in the Scandinavian consensual political culture. In Greece the consultation stage led to a weakening of the plan of amalgamation (in spite of initially planned reduction of the number of municipalities from over 5,000 to 500, eventually 1,033 new units were created), but it was local politicians rather than local public opinion that played an active role in opposing more radical consolidation. Hlepas suggests that relatively passive attitude of local communities was a result of the fact that the reform would strengthen capacity of local administration, and that goal was very much meeting citizens' expectations. Therefore, local politicians who opposed the change had difficulty mobilising public opinion in their support. In general, taking into account the provisions of the European Charter of Local Governments, the importance of consultations and open, public debate has been astonishingly low. However, one should perhaps mention that in some of the failed reforms (such as Ukrainian attempt in 2005) the negative reaction of the local communities expressed in the public consultations played an important role in maintaining the territorial status quo.

This collection of articles started with a clear assumption that we were interested in cases of the imposed municipal boundary reforms. However, the picture presented in individual articles suggests that notions we use in our descriptions are often fuzzy. The difference between bottom-up voluntary change and reform implemented in a top-down manner is not very sharp. The Danish case provides an excellent illustration of that. The first stage of the reform assumed bottom-up proposals of boundary changes made by local governments, but was based on the criteria formulated in the top-down manner. Moreover, the failure to formulate a bottom-up proposal was to be followed by a top-down intervention. A similar semi-voluntary stage of the reform was attempted in Greece (where it was a failure) and in Germany (where it was a relative success in some of Landers, such as Sachsen). Inter-municipal co-operation in the form of communautés in France, or micro-regions in Hungary, Czech Republic or Slovakia provides another example (Bucek, 2008; Hertzog, 2010; Illner 2010). Inter-municipal structures are supposed to enable an effective provision of services by very small municipalities and provide an alternative for territorial amalgamation. Those inter-municipal institutions are created voluntarily, in a bottom-up manner, but incentives or even pressure imposed by the central government can be so strong that the change has a clear element of coercion.

Another diluted concept is a distinction between territorial boundary reform and the bottom-up co-operation among municipalities. Passing more and more functional responsibilities (sometimes even together with a power of taxation) to inter-municipal institutions is sometimes a method for avoidance of political problems related to territorial amalgamation reforms, but at the same time it means the marginalisation of small municipal governments. De facto the result of the process may be the creation of a new, larger and stronger local government unit, while the ‘old, small municipalities’ formally remain, their role mainly symbolic, with a minimal significance for service delivery. So instead of politically risky amalgamation, one may marginalise a small municipality, passing real powers to the new, larger unit. It is one of the possible interpretations of the tendency observed in France, but the logic of change may lead in the same direction the Czech Republic, Slovakia or Hungary. The current economic and fiscal crisis may bring back discussion on territorial consolidation in these countries and strengthen the tendencies described above.

Notes

1 For a wider discussion of this process see Swianiewicz (2002).

2 Fragmentation in CEE countries has not gone as far as in France, where there are municipal jurisdictions that have no permanent residents at all.

3 The size of municipalities in the United Kingdom are not comparable to any other European country, so the UK has been excluded from the figure and related calculations.

4 An optimal size of a local unit has been an issue of debates for many years. It is fascinating to notice that the size indicated by the Polish study very much reminds the ideal city described by Plato: ‘The number of our citizens shall be 5,040 – this will be a convenient number, and these shall be owners of the land and protectors of the allotment’ (Plato, Laws, Book V, trans. B.V. Jowett). Plato was counting heads of households only. So, taking into account the size of their families and slaves, he meant a city of around 30,000 inhabitants.

5 See, for example, a review of such studies in Sharpe (1995); see also the wider discussion in Swianiewicz (2002).

6 As Baldersheim & Rose (2010, forthcoming) suggest: ‘“Requirements from Brussels” may sometimes be a convenient excuse for national policy makers to rush through unpopular reforms.’ A similar observation on the reforms in Central Europe has been made by Hughes et al. (2003).

7 For an excellent discussion of the French case and an alternative point of view see Hertzog (2010), also Borraz & Le Gales (2004).