Between the Integration and Accommodation of Ethnic Difference: Decentralization in the Republic of Macedonia

Abstract

The decentralization process in the Republic of Macedonia has been widely regarded as a success story by regional and international actors alike. It is frequently considered a suitable non-territorial model of ethnic conflict management that can be replicated elsewhere. By increasing the number of competences administered at the municipal level, in addition to replicating the central government's system of consociational power-sharing locally, the reforms seek to provide local, culturally diverse communities with greater control over the management of their own affairs and resources. This paper will begin with a theoretical discussion of how municipal decentralization may offer an institutional solution for managing and preserving cultural diversity within unitary states. It will seek to position Macedonia's decentralization reforms within the ongoing theoretical debate between integrationists and accommodationists, and will offer some initial observations on how the reform's implementation thus far have diverged from the original intentions of the Ohrid Framework Agreement. [PUBLICATION ABSTRACT]

Headnote

The decentralization process in the Republic of Macedonia has been widely regarded as a success story by regional and international actors alike. It is frequently considered a suitable non-territorial model of ethnic conflict management that can be replicated elsewhere. By increasing the number of competences administered at the municipal level, in addition to replicating the central government's system of consociational power-sharing locally, the reforms seek to provide local, culturally diverse communities with greater control over the management of their own affairs and resources. This paper will begin with a theoretical discussion of how municipal decentralization may offer an institutional solution for managing and preserving cultural diversity within unitary states. It will seek to position Macedonia's decentralization reforms within the ongoing theoretical debate between integrationists and accommodationists, and will offer some initial observations on how the reform's implementation thus far have diverged from the original intentions of the Ohrid Framework Agreement.

Key words: Decentralization; cultural pluralism; consociationalism; Macedonia; accommodation; integration.

(ProQuest: ... denotes non-US-ASCII text omitted.)

The decentralization process in the Republic of Macedonia1 has been widely regarded as a success story by regional and international actors alike. It is frequently considered a suitable non-territorial model of ethnic conflict management that can be replicated elsewhere, such as in neighbouring Kosovo. The reforms, which form part of a more comprehensive peace process defined by the Ohrid Framework Agreement (hereafter, Framework Agreement) of 2001, offer limited autonomy to Macedonia's ethnic communities, in particular the ethnic Albanians. By increasing the number of competences administered at the municipal level, in addition to replicating the central government's system of consociational power-sharing locally, the reform process seeks to provide local, culturally diverse communities with greater control over the management of their own affairs and resources.

The purpose of this paper is to position Macedonia's experience with decentralization within the ongoing theoretical debate concerning how states deal with ethnic difference. It will begin by outlining how the devolution of power to sub-national units such as municipalities may offer an institutional solution for managing and preserving cultural diversity within a unitary state. The paper will then position this discussion within the wider theoretical debate between integrationists and accommodationists regarding how states should manage cultural pluralism. A review of Macedonia's decentralization process, a rather weak form of self-government in comparison to strategies adopted by other former Yugoslav republics, will then follow. Elements of the reform will be identified as belonging to either accommodationist or integrationist approaches to managing ethnic difference. Finally, the decentralization reforms defined by the Framework Agreement and subsequent legislation will also be compared with the reality of their implementation. Particular attention will be paid to the controversial process of territorial reorganization in 2004 and the effectiveness of consociational power-sharing arrangements adopted locally.

This paper will argue that accommodationist and integrationist strategies - i.e. those that either recognize ethnic difference or seek to weaken it through the promotion of a common (civic) public identity, for managing ethnocultural diversity are not mutually exclusive. Nowhere is this more apparent than in post-Ohrid Macedonia and its "complex" design of decentralization which incorporates many consociational techniques at the local level. Despite decentralization's offer of limited local autonomy to territorially concentrated groups, the reform is principally a mechanism for integrating ethnic communities into unitary state structures. However, inconsistencies at the implementation stage have altered the delicate balance decentralization was intended to represent between accommodationist and integrationist approaches to diversity management at both the state and local levels.

1. Defining decentralization

The focus of this paper is only one of a variety of self-governance arrangements available for the management of ethnopolitical conflict: decentralization to local or municipal government. Other common forms of self-governance arrangements not discussed here are federalism, territorial and personal autonomy. I define decentralization as a process by which central-local relations are restructured in a unitary state, through the devolution of competences from the national to local levels of government (GTZ, 2006: 5; Braathen et al., 2008: 3; Grasa and Camps, 2009: 21). The process is understood to be an effort to share power vertically between central and local government in the following three dimensions: political, administrative and fiscal. Political decentralization refers to the transfer of political authority to the local level through the establishment of elected local governments and procedures aimed at increasing the participation of citizens and civil society in local decision-making. Administrative decentralization involves the devolution of functional responsibilities to municipalities, along with their bureaucratic structures but, importantly, without removing their accountability to central government. Fiscal decentralization entails the transfer of financial authority from the central to local level and is essential if institutions are to exercise public policy functions autonomously.

Whilst decentralization can represent extensive devolution to local governments, it is important not to lose sight of the fact that these self-governing entities, unlike federal units, remain ultimately subordinate to the authority of central government. Unitary states may possess national and sub-national tiers of government, explains Pippa Norris (2008: 168), but in the case of conflict, 'the national government remains constitutionally sovereign so that executive decisions and laws passed by the national legislature cannot be overruled by lower units'. Although governments of unitary states have the authority to delegate powers downwards, sub-national units have no right to these powers which can, in principle, be withdrawn through new central legislation without any need for local consent (Schou and Haug, 2005: 10).

The lack of a constitutionally entrenched guarantee, which requires the consent of all government tiers to authorize amendments, is one of the key attributes of decentralization which differentiates it from other forms of territorial self-governance, principally federalism (Rothchild and Hartzell, 2000: 261; Coakley, 2001: 299; Wolff, 2010: 10). Other characteristics of federalism which are missing in decentralized states are: bicameral legislatures, where the second chamber represents the sub-national level centrally and may compensate smaller units through over-representation; independent mechanisms of judicial review, with supreme arbitral authority to settle disputes over the constitution and intergovernmental relations; and exclusive legislative and judicial powers. Unlike federal systems of governance, in decentralized unitary states sovereignty is not divided. Decentralization also differs from autonomy, another form of self-governance, which can be granted on either a territorial or non-territorial basis. The principal characteristic of autonomy which distinguishes it from both decentralization and many types of federalism is its explicitly ethnic nature (Young, 1998: 60). Ruth Lapidoth (1997: 174-5), a leading expert on autonomous arrangements, defines territorial autonomy as 'an arrangement aimed at granting a certain degree of self identification to a group that differs from the majority of the population in the state, and yet constitutes the majority in a specific region'. It involves the creation of separate governance structures through which members of a specific ethnic community exercise control over their political, economic, social or cultural affairs (Hadden, 2005: 34). Importantly, however, and akin to decentralization, autonomy (both territorial and personal) is normally a feature of an otherwise unitary state (Wolff, 2010: 5).

2. Decentralization as a tool for mitigating ethnopolitical conflict

One of the most frequently cited arguments in favour of decentralization as a tool for managing ethnopolitical conflict is its ability to satisfy the demands of spatially concentrated non-majority groups for limited autonomy over their own affairs and resources. The creation or enhancement of sub-national units can allow a group which is a minority at the state level, but constitute a majority locally, to exercise governmental power in ways that would otherwise be foreclosed if the whole country was one undifferentiated territory (Horowitz, 2007: 958). Importantly, however, it does so whilst maintaining the territorial integrity of the unitary state. In situations where there is no alternative to preserving a state's territorial integrity, conclude Weller and Wolff, carefully designed self-governance regimes can provide the institutional structure that offers sufficient space for non-majority groups to experience genuine self-governance (Weller and Wolff, 2005: 270).

Decentralization's promise of limited autonomy also provides greater scope for promoting minority rights and the protection of their cultural identity. In this respect, devolved competences relating to education, culture and language policy are of particular importance. By granting non-majority groups greater control over their own destinies, decentralization is believed to instil a greater sense of security within these groups that they will not be subject to discriminatory practices and unwanted intrusions in the future (Manor, 1999: 97).

Decentralized institutions not only facilitate better access to political structures and state resources for ethnic minority and opposition groups, they can also restrain the monopoly central government and majority-dominated political parties have on state power. The distribution of power to different levels of government facilitates a system of checks and balances that sets limits on central government if it attempts to overstep or abuse its powers. 'Power dividing', notes Roeder (2005: 62), balances decision-making against the 'dangers of tyranny' by a single majority and, according to Horowitz (1991: 217), 'makes hegemony more difficult to achieve'. Since decentralization increases the number of arenas in which there are political prizes to be won, the opportunities for previously disenfranchised groups to participate legitimately in government are enhanced. Local institutions may become 'training ground[s] for democracy', suggests Manor (1999: 49, 85); relieving dangerous frustrations by providing aspiring politicians with additional openings into the political system. Providing alternative sites of power and patronage can also disperse political conflict away from the centre, thereby lowering its stakes (Sisk, 1996: 54; Young, 1998: 32; Roeder, 2005: 52; Treisman, 2007: 14). 'Proliferating the points of power', confirms Horowitz (1991: 226; 2000:598), makes politics 'a much more diffused game'; transforming it from a "zero-sum" contest into one where different groups can each control some part of the governing apparatus.

Experts have argued that the creation of decentralized institutions can also promote social cohesion locally, in addition to facilitating a more balanced relationship between majority and minority communities at the state level. In ethnically heterogeneous regions, self-government can create non-violent platforms for interethnic and intergroup discussion relating to local issues and allocation of resources. Such 'learning laboratories', notes Zoë Scott (2009: 16), can help local politicians build trust among groups and acquire political and conflict resolution skills that can be used in different social arenas. Local government may also serve as an incubator for small ethnic parties that choose to form inter- and transethnic alliances (Treisman, 2007: 245). If, as Dawn Brancati (2008) suggests, decentralization does indeed help proliferate regional and ethnic political parties, it may be difficult for one group to dominate locally, and incentives for forming interethnic coalitions may be enhanced (Horowitz, 2007: 962). 'Training in compromise is important in divided societies', concludes Horowitz (2007: 960); particularly before local politicians rise to the national level, where more complex and delicate issues of national policy may need to be resolved.

Decentralization is an institution that, if properly designed, can potentially benefit both political minorities, who can be induced or compensated not to secede from the state, and political majorities, who gain countrywide acceptance of state institutions (Lake and Rothchild, 2005: 122). When threatened with secessionist attempts, federalism, or even partition of the state, decentralization can appear an attractive way of preserving state integrity and of integrating separatist movements (Litvack et al., 1998: 108; Hannum, 2004; Schou and Haug, 2005: 18; Braathen etal., 2008: 15). Importantly, decentralization to local governments, as opposed to regional or federal institutions, can also be employed to deliberately fragment the local power bases of secessionist groups into smaller, weaker, non-politically significant units (Young, 1998: 32; Horowitz, 2000: 646; Crook, 2002: 300; GTZ, 2006: 7; Ejobowah, 2008: 257). In Uganda for example, Schou and Haug (2005: 25) claim that the decentralization reforms were never intended to assuage ethnic nationalism, but rather to cut across and fragment important geopolitical areas. Similarly in Indonesia, decentralization was used to disperse power and fragment different ethnic groups into 301 local governments and 19 states (Ejobowah, 2008: 239). While the boundaries of local units may be drawn to allow an ethnic group that constitutes a majority locally to govern (for example in Kosovo), it is more common that territorial designs ensure no one community dominates locally. Either way, the limited size and scope of the local units ensures the political influence of territorially concentrated ethnic groups is diffused. In this way, ethnic conflict management techniques differ significantly between governments in unitary and federal states. Many federal governments, for example Switzerland and Belgium, believe accommodating national minorities holds the key to stability and, as a result, explicitly incorporate them into state design. In contrast, unitary state governments apply the exact opposite approach and often use decentralization as a tool for eroding ethnic identity and solidarity (Schou and Haug, 2005: 32).

3. Accommodation or integration: 'the enduring debate in conflict regulation'2

The ongoing debate between whether states should accommodate or attempt to integrate the ethnic differences of citizens demonstrates a fundamental normative disagreement over the mechanisms of interethnic cooperation. Each approach proceeds from different assumptions and epistemological positions regarding the durability and malleability of politically mobilized ethnic identities. Integrationists, notes Richard H. Pildes (2008: 175), focus primarily on the long-term normative vision of the state; while accommodationists are (allegedly) more concerned with the immediate, short-term pressures states face. Both approaches translate into a much broader set of policy options with regard to constitutional design in divided societies than the familiar Lijphart-Horowitz debate has generated (Choudhry, 2008b: 27). In the following section I will briefly summarize the principle differences between the accommodationist and integrationist approaches to managing ethnic difference. I will then position the various arguments concerning decentralization's perceived ability to manage ethnic conflict within this theoretical debate.

In general terms, accommodationists promote dual or multiple public identities and advocate equality with institutional respect for difference (Choudhry, 2008b: 27). They assume ethnic identities in segmented societies are resilient and not susceptible to short-term transformation. However, they do not necessarily believe identities are primordial and fixed (Bertrand, 2008: 209; McGarry et al., 2008: 52). Accommodationists seek to ensure each ethnic group has the public space necessary to express its identity, make its own decisions in areas of critical importance, and to protect itself against the majority (McGarry et al., 2008: 42). The result is the design of public policy which permits the institutional expression of differences in the public sphere, such as minority language rights. Consociational techniques advocated by Arend Lijphart (1975; 1977; 2008) and others (McGarry and O'Leary, 2005; McGarry et al., 2008; etc.) are examples of approaches to accommodating cultural pluralism (power-sharing executives; proportionality; segmental autonomy, territorial or corporate, along ethnic lines; mutual veto rights among groups; and arbitration mechanisms). Centripetalism, advocated principally by Donald Horowitz (1991, 2000, 2002, 2007), is another example of how states can accommodate ethnic difference, albeit at the integrationist end of the spectrum. The political incentives this approach advocates to encourage intergroup cooperation assume that the existence of ethnic political parties are inevitable (Choudhry, 2008b: 27).

Integrationists, by contrast, believe political instability and even further conflict are a consequence of group-based partisanship in political institutions, since they empower elites that have a vested interest in maintaining these social divisions. They reject the idea that ethnic difference should necessarily translate into political differences, and instead argue for the possibility of a common (civic) public identity (Choudhry, 2008b: 27). As McGarry, O'Leary, and Simeon note (2008: 73), integrationists advocate such an approach even when ethnicity is served as the basis of political mobilization, since they believe ethnic identities are seldom as long-standing or as deep as supporters of accommodation suggest. Accordingly, integrationists support constitutional strategies which promote a common public identity which transcends, cross-cuts and minimizes ethnic cleavages, without (importantly) demanding ethnocultural uniformity in the private sphere. Examples of such strategies include common state institutions, "ethnically blind" public policies, the promotion of individual rather than communal rights, the design of mixed or non-ethnic territorial entities, and electoral systems which encourage the formation of pre-election coalitions across ethnic divides (Sisk, 1996: xi).

An excellent example of how a state's desire to either accommodate or integrate its non-majority communities can have direct implications on the design its decentralization process ultimately takes is the creation of sub-national units. The approaches differ concerning whether the borders of decentralized units should be congruent with ethnic divisions or "cross-cut" them. Advocates of accommodation, with their emphasis on ethnic representation, stress that territorial boundaries should coincide as much as possible with ethnic or other group cleavages, resulting in the creation of ethnically homogenous units. 'Clear boundaries between the segments of a plural society', writes Lijphart (1977: 88), 'have the advantage of limiting initial contact and consequently of limiting the chances of ever-present antagonisms to erupt into actual hostility'. In contrast, whilst not all integrationists support territorial autonomy, when they do so it is primarily for classical Madisonian reasons: to proliferate points of power, to disperse ethnic communities across different heterogeneous units, and to emphasize cross-cutting cleavages across group boundaries (Lipset, 1960; Elazar, 1966; 1992; Horowitz, 2000: 598; 2007: 964; McGarry, 2008: 703; Wolff, 2010: 1). Heterogeneous units, integrationists claim, may generate incentives for establishing interethnic coalitions locally and serve as a training ground for mutual tolerance (Manor, 1999: 85; Norris, 2008: 163; Pildes, 2008: 19).

Table 1 illustrates how the arguments regarding decentralization's perceived ability to manage conflict can be positioned within the integrationist-accommodationist debate on managing ethnic difference. It demonstrates that decentralization designs may recognize and accommodate ethnic diversity in a variety of ways (enhanced local autonomy, minority rights, etc.), particularly if territorial boundaries are drawn to coincide with ethnic cleavages. However, and in contrast to other, more extensive forms of territorial self-governance (federalism, regional autonomy), decentralization's offer of only limited local autonomy within an otherwise unitary state, along with its potential to fragment the local power bases of secessionist groups, is principally a mechanism for integrating local (ethnic) communities into state structures.

A review of Macedonia's decentralization process will now follow. It will demonstrate that the reform's design, an example of "complex power-sharing" which combines territorial self-governance with local power-sharing mechanisms, seeks to simultaneously integrate non-majority communities into unitary state structures while accommodating ethnic diversity locally. In doing so, the reform demonstrates how, in the context of institutional design in culturally and ethnically plural societies, accommodationist and integrative approaches need not represent a dichotomy.

4. Decentralization in the Republic of Macedonia: an integrationist approach to diversity management with elements of accommodation

In its Basic Principles, the Framework Agreement signed in Ohrid declared: 'The development of local self-government is essential for encouraging the participation of citizens in democratic life, and for promoting respect for the identity of communities' (Art. 1.5).4 Certainly, the Framework Agreement called for the transfer of 11 new municipal competences, along with a revised law on local government financing to ensure sufficient resources and fiscal autonomy; 'a cure against federalization', remarked one ethnic Albanian politician at the time (PER, 2003: 11).5 However, despite common perceptions suggesting the contrary, Florian Bieber (2005: 116) notes how decentralization was framed to conform with European standards, especially the principle of subsidiarity, rather than facilitating fully-fledged self-government for the ethnic Albanian community. Ermira Mehmeti (2008: 73), then spokeswoman for the ethnic Albanian party Democratic Union for Integration, remarked how Lijphart's models of non-territorial corporate autonomy and segmental autonomy along ethnic lines were abandoned during the negotiations in Ohrid because of the government's insistence on preserving the unitary character of the state. Given the Macedonian government's substantial and persistent fear that decentralizing power to sub-national levels would facilitate ethnic Albanian secession of the western part of the country, it is worth examining precisely what impact decentralization has had on the unitary nature of the state.

Whilst Macedonia's decentralization process does indeed represent extensive devolution of political, administrative and fiscal responsibility to the municipalities, it does not, as in federal states, constitute a division of central government authority. As the Framework Agreement clearly points out in its 'Basic Principles', 'Macedonia's sovereignty and territorial integrity, and the unitary character of the State are inviolable and must be preserved. There are no territorial solutions to ethnic issues' (Art. 1.2). Municipalities in Macedonia remain ultimately subordinate to the authority of central government, even though, significantly, their right to the devolved competences is constitutionally entrenched (Constitution of the Republic of Macedonia, 2001: Art. 115).6 Some of these devolved competences, in addition, remain limited and are not held exclusively by the municipalities. Macedonian municipalities do not, for example, enjoy legislative powers; and with regards to primary and secondary education only have responsibility for the maintenance of school buildings and the payment of staff salaries, rather than set policy or influence the development of curricula (Law on Local Self-Government, 2002: Art. 22.8; Wolff, 2010: 24).7 The municipalities are also not directly represented at the central level, as federal units would be in a bicameral system. The only mechanism available to municipalities for influencing central government policy (apart from utilizing informal political party channels) is the municipal association, known by its o Macedonian acronym "ZELS".8 However, despite its energy and boasting full membership of all 84 municipalities plus the City of Skopje, ZELS is frequently marginalized by central government because of its non-governmental organization status.

The original design of the territorial reorganization, irrespective of the manner in which its implementation was carried out, illustrates the government's fear of promoting territorial disintegration and a desire to fragment the political influence of the ethnic Albanian community. Unlike examples of regional autonomy, the ethnic nature of Macedonia's municipalities is not explicitly recognized. Even forms of personal autonomy for specific territorially dispersed communities are conspicuously absent from the Framework Agreement and subsequent legislation (Daftary and Friedman, 2008: 287; Sulejmani, 2008: 161). In contrast with neighbouring Kosovo, Macedonian municipalities comprising a significant minority ethnic community do not enjoy any special asymmetrical status (UNSC, 2007: Annex 3, Art.4; Sulejmani, 2008: 152). All 84 Macedonian municipalities have been granted the same competences, regardless of local demographics, and the reforms have been framed within the context of the Council of Europe's subsidiarity principle, rather than as a mechanism for facilitating enhanced autonomy of ethnic communities. Whilst non-majority communities comprising at least 20% in a particular municipality can use their community language in an official capacity locally - as the former Minister of Local Self-Government, Rizvan Sulejmani (2008: 144) points out - these community languages are not in autonomous official use, but must be used in addition to the Macedonian language. It is worth emphasizing that the recognition of community languages comes as a function of demographics, rather than as a symbolic recognition of their equal status with the state (Macedonian) language. Nowhere in the Framework Agreement is the use of the "Albanian" language, for example, specifically mentioned. Such an arrangement is at risk of causing friction in the future if a situation arises where local demographic changes mean a particular community no longer meets the required 20% threshold.9

The absence of any regional tier of government, in addition to the sizable number of municipalities which survived the territorial reorganization (contrary to the advice of local experts), also illustrates how anxiety over devolving power to territorially concentrated ethnic groups influenced the particular design of Macedonia's decentralization. Given the territorial concentration of ethnic Albanians in the north and west of the country, it is of no surprise that the government adopted decentralization reforms that would disperse limited and conditional authority to more than 80 non-ethnically defined sub-units. Ethnic Albanians currently comprise a majority in 16 municipalities and represent at least 20% of the local population in a further 13 (including the City of Skopje) (Statistical Office, 2005). This would not have been the outcome had the government endorsed the creation of regional units or indeed an autonomous region. The fact that municipalities are allowed to cooperate and form joint institutions, but are not allowed to merge with adjacent municipalities (Law on Local Self-Government: Art. 14), further supports this view (Bieber, 2008: 34). Unlike in neighbouring Kosovo, Macedonian municipalities are also unable to receive financial assistance from neighbouring (kin) states (UNSC, 2007: Annex III, Art. 11).

With regard to the accommodation of ethnic diversity at the local level, the Framework Agreement replicates most of the consociational provisions foreseen nationally.10 Of these provisions, greater use of national languages and symbols has had most impact on the daily lives of citizens. Under the terms of the Agreement and subsequent Law on the Use of Languages (2008), the Macedonian language remains the official language in the country and is used for international relations (Framework Agreement: Art. 7). However, any other language spoken by at least 20% of the population locally is also recognized as an official language within municipalities. Previously, the threshold for official use of community languages in Macedonian municipalities was 50%, although this was rarely respected and was indeed annulled by the Constitutional Court in 1994 (Caca, 2001: 152; Caca, 1980: 96). With regard to languages spoken by less than 20% of the population within a municipality, the Agreement also allows the possibility for their use as official languages, but the decision to do so remains at the discretion of the local authority (Constitution of the Republic of Macedonia: Art. 6.6). It is worth noting that the 20% threshold means that Albanian is the only language other than Macedonian granted recognition at the state level. The Agreement and subsequent Law on the Use of the Flags of the Communities (2005) also regulates the use of community emblems, such as the flying of community flags in front of local public buildings if that community constitutes a majority within a municipality (Framework Agreement: Art. 7). Previous legislation on the use of flags was also repealed by the Constitutional Court in December 1998 (Helsinki Committee, 1999; Ragaru, 2008: 5).

The consociational principle of proportionality within municipal administrations, councils and committees is another crucial accommodationist mechanism incorporated into the decentralization design (Framework Agreement: Art. 4). However, the Framework Agreement does not call for strict ethnic quotas in Macedonia's public administration; state institutions are not obliged by law to employ a certain percentage of a particular ethnic group. This is in contrast to other consociation-inspired political settlements, such as Bosnia's Dayton Agreement (OHR, 1995). Special voting procedures, sometimes referred to as "double-majority" or "Badinter-majority"11 voting, are also envisaged to ensure greater consensus in decision-making within municipal councils. They operate along similar lines to Arend Lijphart's concept of minority veto; however they differ in that the right to use them is not given to a particular community. Instead, certain municipal acts 'cannot be approved without a qualified majority of two-thirds of votes, within which there must be a majority of the votes from those claiming to belong to non-majority communities' (Framework Agreement: Art. 5). This procedure represents a much weaker protective mechanism than comparable veto powers in other former Yugoslav 12 republics since its use is restricted to decision-making only in specific areas (Marko 2004/05: 709). Finally, the Framework Agreement re-established Committees for Inter-Community Relations in municipalities where at least 20% of the local population belongs to a certain non-majority community (Law on Local Self-Government: Art. 55). Municipalities may also establish committees in areas where local communities comprise less than 20%, although this decision is at the discretion of the municipal council. Comparable with Lijphart's notion of arbitration mechanisms, their role is to enable institutional dialogue locally between different ethnic communities and to settle disputes affecting intercommunity relations, such as decisions requiring the use of special voting procedures.

To summarize, despite granting local communities enhanced autonomy over the management of their own affairs and resources, as in most decentralization processes, the reforms in Macedonia predominantly represent an attempt to integrate non-majority communities (particularly the ethnic Albanians) into unitary state structures. Macedonian decentralization does this by ensuring that municipalities remain ultimately subordinate to the central government, have no explicit ethnic identity, and are sufficiently numerous to fragment the power base of non-majority communities across multiple local units. In doing so, decentralization intends to balance the more consociational/accommodationist aspects of the Framework Agreement foreseen at the state level.13 However, in a departure from more commonly designed decentralization processes, the reform's combined use of consociational techniques locally suggests that Macedonian decentralization seeks to simultaneously accommodate ethnic diversity at the municipal level.

5. Decentralization in the Republic of Macedonia: discrepancies in implementation

The process of territorial reorganization during 2004, when municipal boundaries were redrawn to create 84 municipalities from the previous 123, illustrates how discrepancies between the de jure provisions and de facto implementation of decentralization have affected the delicate balance the Framework Agreement intended to strike between the integration and accommodation of ethnic difference at the state level. According to of the Law on Local Self-Government, municipal boundary changes should take into account ethnically neutral factors, such as economic, geographic and infrastructure features; whilst the Council of Europe recommended that the population of a municipality should not be less than 5,000 citizens (Marko, 2004/05: 713; PER, 2004: 18). The resulting 84 municipalities, 14 of which have populations of less than 5,000, suggest that instead the reorganization was based largely on political and ethnic compromises and was more concerned with accommodating group needs rather than integrating them into state structures (Friedman 2009: 217; Siljanovska-Davkova, 2009: 112).

The most controversial boundary changes were those affecting the municipalities of Struga and Kicevo, which were enlarged to ensure Albanians became the majority in both municipalities.14 The City of Skopje was also enlarged to ensure the Albanian population reached the 20% threshold required to make Albanian an official language in the capital. A candid remark made by an ethnic Albanian politician sums up the governing coalition's motives at that time: 'We want to maximize the number of municipalities where Albanians make up 20% of the population [...] and we want to bring Albanians into connection with the urban centre; the Macedonians want the opposite [...] ' (ICG, 2003: 20). The controversial manner in which the territorial reorganization was carried out meant that decentralization, whilst originally intended to be an ethnically-blind process, immediately became ethnicized. The national referendum on whether to accept the proposed municipal boundary changes in November 2004 raised ethnic tensions to a level not experienced since the cessation of hostilities in 2001. It succeeded in branding the process as being tantamount to ethnic cleansing and, to a certain extent, has tainted the implementation of decentralization ever since (Petrov, 2004). As a consequence of difficulties experienced during implementation, therefore, the initial emphasis the Framework Agreement places on the integration of ethnic diversity through decentralization has in practice been replaced by a more accommodationist approach.

Poor and inconsistent application of the power-sharing arrangements envisaged locally provides another example of where inconsistencies during the implementation of decentralization have impaired realization of the Framework Agreement's objectives. In this case, however, the ability of decentralization to accommodate diverse groups locally, and in particular the smaller ethnic communities15, has been weakened. Whilst greater use of community languages locally has meant Albanian has now acquired official status in 29 of the 85 municipalities (including the City of Skopje), the Turkish language has only become a working language in four, Serbian in one and Romani in one. Albanian is also in use in a further 17 municipalities where the community constitutes less than 20% locally, Turkish in another four, and Vlach language in one (OSCE, 2009: 55). Therefore, while the Albanian language has been recognized (formally if not in practice) in over half of Macedonia's municipalities, community languages other than Albanian are in official use in only ll.16 The geographic dispersal of the Turkish, Roma, Serbian, and Vlach communities throughout Macedonia means that in most municipalities they fall well below the 20% threshold required to benefit automatically from community language rights. Their lack of numeric and political strength locally also means that very few municipalities have chosen to voluntarily recognize these languages.

The equitable representation of all communities within the public administration and enterprises is another accommodative measure where progress has been both slow and inconsistent. The European Commission's progress reports on Macedonia for 2009, 2010 and 2011, for example, consistently remark how the representation of the smaller communities in the civil service, particularly the Turkish and Roma, remains low (EC, 2011: 20; EC, 2010: 22; EC, 2009: 21). Similarly, the limited use of special voting procedures in municipal councils further supports the impression that the accommodation of Macedonia's ethnic communities locally is far from effective. The practice is 'rarely respected on the local level', and in many instances local councillors are unaware of its compulsory nature for specific topics (ADI, 2006; IRIS, 2006: 14).17 Finally, the effectiveness of the Committees for Inter-Ethnic Relations established in over 30 multi-ethnic municipalities also raise doubts regarding the equitable accommodation of non-majority group needs (CDI, 2007; Forum, 2008; OSCE, 2009). One of the criticisms frequently directed against these committees, most notably by the European Commission, is that 'their role is still largely unknown by the public' (CDI, 2007: 9; EC, 2010: 21). An influential United Nations programme in Macedonia also observed how they are 'generally found to convene for the sake of demonstrating that they have done so, and they rarely provide advisory, preventive or reactive recommendations' (UN, 2010: 5). Membership of the 18 committees in particular has been the focus of much criticism.18 Smaller communities, notably the Roma, are often unrepresented and, when they are, they are frequently outnumbered by more dominant local communities, notably Macedonians and ethnic Albanians (ZELS, 2010).

6. Conclusion

This paper has positioned Macedonia's experience with decentralization within the ongoing debate concerning how states deal with ethnic difference. After discussing the arguments concerning decentralization's perceived ability to manage ethnic conflict, the paper examined the design of Macedonia's reform and located it in the debate between those who advocate for integration as a strategy for managing ethnic and cultural pluralism in multi-ethnic states and those who support a more accommodationist approach. Discrepancies between the de jure provisions and de facto implementation of Macedonia's decentralization were also considered, along with its impact on the accommodation and/or integration of non-majority communities at both the state and local levels.

Macedonia's experience with decentralization questions whether the perceived dichotomy between integrationist and accommodationist approaches is the correct way to consider the various options available in the constitutional design of divided societies. As Alan Pattern (2008: 92) has recognized, a dichotomous way of thinking may be too simple to do justice to the problems under consideration in culturally plural societies. This simplification may lead to a serious distortion of the possible solutions to problems raised by diversity and the ways in which real-world conflicts should be understood. In spite of representing alternative strategies for managing ethnic diversity, the differing approaches to accommodating and integrating pluralism are not mutually exclusive and can be combined in different ways. Nowhere is this more apparent than in post-Ohrid Macedonia and its "complex" design of decentralization which incorporates many consociational techniques.

Like many decentralization designs, the Macedonian reform's offer of limited local autonomy within an otherwise unitary state, along with its intention to fragment ethnic communities across multiple local units, is principally a mechanism for integrating local communities into state structures. In doing so, decentralization seeks to balance the more accommodationist aspects of the Framework Agreement foreseen nationally. Significantly, the inclusion of local power-sharing mechanisms suggests that the reform also intends to accommodate ethnic diversity in a way many other decentralization designs do not. However, discrepancies between the de jure provisions and de facto implementation of the reform in Macedonia have led to contradictory results. Ethnification of the territorial reorganization in 2004 has meant that, for many, decentralization has become more about accommodating local group cleavages than promoting integration within the unitary state. Similarly, poor and inconsistent application of power-sharing arrangements at the municipal level has resulted in only the partial accommodation of some ethnic communities at the expense of assimilating others. In both instances, discrepancies at the implementation stage have altered the delicate balance decentralization was intended to strike between accommodationist and integrationist approaches to diversity management at both the state and local levels. Failure to address these discrepancies in the future may have far-reaching consequences for the equilibrium and ultimate success of the wider peace agreement.

Footnote

Notes

1 Hereafter referred to as "Macedonia". It should be noted that the "Republic of Macedonia" is the constitutional name for the state; however this name is the subject of dispute with neighbouring Greece. Consequently, the Republic of Macedonia is currently recognized under its temporary name the "Former Yugoslav Republic of Macedonia" ("FYROM") in its dealings with international organizations, such as the United Nations and European Union.

2 This replicates the title of a book chapter written by John McGarry, Brendan O'Leary and Richard Simeon (McGarry et al., 2008: 41-90).

3 Referring to Arend Lijphart and Donald L. Horowitz, leading advocates of consociational and centripetalism approaches to managing ethnopolitical diversity.

4 Official Gazette of the Republic of Macedonia: Framework Agreement. Skopje: Government of the Republic of Macedonia, 2001.

5 Those competences specifically referenced in the Framework Agreement were: areas of public services, urban and rural planning, environmental protection, local economic development, culture, local finances, primary and secondary education, social welfare, and health care (Art. 3.1).

6 Official Gazette of the Republic of Macedonia. Constitution of the Republic of Macedonia. Skopje: Government of the Republic of Macedonia, 2001.

7 Offical Gazette of the Republic of Macedonia: Law on Local Self-Government. Skopje: Government of the Republic of Macedonia, 2002.

8 ... (Association of the Units of Local Self-Governance of the Republic of Macedonia).

9 A census was planned to take place in 2011. However, given the significance of local demographics, the process became highly politicized and was later postponed due to a dispute arising between Macedonian and ethnic Albanian political parties over data collection procedures. It remains to be seen whether local demographics have altered significantly since the last post-Ohrid census in 2002.

10 It is worth mentioning that Macedonia is not fully consociational at either the state or local levels, according to Lijphart's criteria. Grand coalitions and segmental autonomy do not feature at either level, although traditionally in Macedonia multi- ethnic coalitions established at the parliamentary level. Multi-ethnic coalitions are also common within municipal councils.

11 Named after the French judge Robert Badinter.

12 Special voting procedures can be used when deciding on the following areas: culture, use of languages, education, personal identification, use of symbols, and issues pertaining to local self-government (Official Gazette, 2001b: Art. 5.2).

13 For example, greater use of non-majority languages; proportionality in parliament, the constitutional court, public administration, army and the police; special voting procedures in parliament; and the establishment of a parliamentary Committee on Inter-Community Relations.

14 Boundary changes to Kicevo were later postponed and in 2012 they had still not been implemented.

15 For example the Turks, Roma, Serbs, Bosniaks and Vlachs, which according to the 2002 Census constitute 3.85%, 2.66%, 1.78%, 0.84%, 0.48% of the national population respectively.

16 Municipalities do not receive additional state resources to fund associated costs, such as the salaries of translators or interpreters. While larger municipalities can more easily meet the expense of hiring additional members of staff, smaller less affluent municipalities cannot. In many instances, expensive simultaneous interpretation equipment donated to the municipalities by the international community remains unused.

17 Fieldwork undertaken by the author to selected multiethnic municipalities in June 2010 and 2011 found that the situation had not changed significantly since 2006.

18 The procedure for selecting committee members varies. However, in most cases candidates are nominated by political representatives in municipal councils.

References

References

Association for Democratic Initiatives (ADI). Power Sharing -New Concept of Decision Making Process in Multicultural Municipalities. Gostivar: ADI, 2006.

Bertrand, J. 'Indonesia's Quasi-Federalist Approach: Accommodation Amidst Strong Integrationist Tendencies'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 205-232. Oxford: Oxford University Press, 2008.

Bieber, F. 'Partial Implementation, Partial Success: The Case of Macedonia'. In New Challenges for Power-Sharing: Institutional and Social Reform in Divided Societies, eds. D. Russell and I. O'Flynn, 107-122. London: Pluto Press, 2005.

Bieber, F. 'Power-Sharing and the Implementation of the Ohrid Framework Agreement'. In Power Sharing and the Implementation of the Ohrid Framework Agreement, 7-40. Skopje: Friedrich Ebert Stiftung, 2008.

Braathen, E., and S.B. Hellevik. 'Decentralization, Peace Making and Conflict Management: from Regionalism to Municipalism'. Journal of Peace, Conflict and Development 12 (2008).

Brancati, D. Peace by Design: Managing Intrastate Conflict through Decentralization. Oxford: Oxford University Press, 2008.

Brown, K., P. Farisides, S. Ordanoski, and A. Fetahu. Ohrid and Beyond: A Cross- ethnic Investigation into the Macedonian Crisis. London: Institute for War and Peace Reporting, 2002.

Bunce, V., and S. Watts. 'Managing Diversity and Sustaining Democracy: Ethnofederal versus Unitary States in the Postcommunist World'. In Sustainable Peace - Power and Democracy after Civil Wars, eds. P. Roeder andD. Rothchild, 133-158. Ithaca: Cornell University Press, 2005.

Caca, G. The Constitutional Development of the Socialist Republic of Macedonia. Skopje: Graflchki zavod Gotse Delchev, 1980.

_____'Status and Rights of Nationalities in the Republic of Macedonia. In The New Macedonian Question, ed. J. Pettifer, 148-166. Basingstoke: Palgrave, 2001.

Choudhry, S. Constitutional Design for Divided Societies: Integration or Accommodation? Oxford: Oxford University Press, 2008a.

_____'Bridging Comparative Politics and Comparative Constitutional Laws: Constitutional Design in Divided Societies'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 3-40. Oxford: Oxford University Press, 2008b.

Coakley, J. The Territorial Management of Ethnic Conflict. London: Frank Cass, 2001.

Community Development Institute (CDI). Committees for Interethnic Relations: Establishment, mandate and existing experiences. Tetovo: CDI, 2007.

Crook, R.C. 'Decentralization and Good Governance'. In Federalism in a Changing World: Learning from Each Other, ed. R. Blindenbacher, 299-333. St. Gallen: Mass Market, 2002.

Daftary, F., and E. Friedman. 'Power-sharing in Macedonia?' In Settling Self- Determination Disputes: Complex Power-Sharing in Theory and Practice, eds. B. Metzger, M. Weller and Ν. Johnson, 265-306. Leiden: Brill, 2008.

Ejobowah, J. B. 'Integrationist and Accommodationist Measures in Nigeria's Constitutional Engineering: Successes and Failures'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 233-257. Oxford: Oxford University Press, 2008.

Elazar, D.J. American Federalism: A View from the States. Springfield: Crowell, 1966.

_____ Exploring Federalism. Tuscaloosa: University of Alabama Press, 1992.

European Commission (EC). The Former Yugoslav Republic of Macedonia 2008 Progress Report. Brussels: European Commission, 2008.

_____ The Former Yugoslav Republic of Macedonia 2009 Progress Report. Brussels: European Commission, 2009.

_____ The Former Yugoslav Republic of Macedonia 2010 Progress Report. Brussels: European Commission, 2010.

_____ The Former Yugoslav Republic of Macedonia 2011 Progress Report. Brussels: European Commission, 2011.

Forum - Center for Strategic Research and Documentation. Commissions for Inter-Ethnic Relations in Municipalities of Macedonia. Skopje: Forum CSRD, 2008.

Friedman, E. 'The Ethnopolitics of Territorial Division in the Republic of Macedonia'. Ethnopolitics 8:2 (2009): 209-221.

Gesellschaft fur Technische Zusammenarbeit (GTZ). Decentralization and Conflicts -A Guide. Eschborn: GTZ, 2006.

Grasa, R. and A. Camps. Conflict Prevention and Decentralized Governance. Barcelona: International Catalan Institute for Peace, 2009.

Green, E. 'Decentralization and Conflict in Uganda'. Conflict, Security and Development 8:4 (2008): 427-450.

Hadden, T. 'Integration and Autonomy: Minority Rights and Political Accommodation'. In Power Sharing: New Challenges for Divided Societies, eds. I. O'Flynn and D. Russell, 30-44. London: Pluto Press, 2005.

Hannum, H. 'Territorial Autonomy: Permanent Solution or Step toward Secession?' In Facing Ethnic Conflicts: Towards a New Realism, eds. A. Wimmer, R. J. Goldstone, D. L. Horowitz, U. Joras, C. Schetter, 274-282. Lanham: Rowman and Littlefield Inc., 2004.

Helsinki Committee for Human Rights of the Republic of Macedonia. Annual Report for Year 1999. Skopje: Helsinki Committee, 1999.

Horowitz, D.L. A Democratic South Africa? Constitutional Engineering in a Divided Society. Berkeley: University of California Press, 1991.

_____Ethnic Groups in Conflict. London: University of California Press, 2000.

_____'Constitutional Design: Proposals versus Processes'. In The Architecture of Democracy: Constitutional Design, Conflict Management, and Democracy, ed. A. Reynolds, 15-36. Oxford: Oxford University Press, 2002.

_____'The Many Uses of Federalism'. Drake Law Review 55 (2007): 953-966.

Institute for Regional and International Studies (IRIS). The Process of Decentralization in Macedonia: Prospects for Ethnic Conflict Mitigation, Enhanced Representation, Institutional Efficiency and Accountability. Skopje/Sofia: IRIS, 2006.

International Crisis Group (ICG). Macedonia: No Room for Complacency. Skopje and Brussels: ICG, 2003.

Lake, D., and D. Rothchild. 'Territorial Decentralization and Civil War Settlements'. In Sustainable Peace - Power and Democracy after Civil Wars, eds. P. Roeder and D. Rothchild, 109-132. Ithaca: Cornell University Press, 2005.

Lapidoth, R. Autonomy. Washington D.C.: United States Institute of Peace, 1997.

Lijphart, A. The Politics of Accommodation: Pluralism and Democracy in The Netherlands. London: University of California Press, 1975.

_____ Democracy in Plural Societies: A Comparative Exploration. London: Yale University Press, 1977.

_____ Thinking About Democracy: Power Sharing and Majority Rule in Theory and Practice. Abingdon: Routledge, 2008.

Lipset, S.M. Political Man: The Social Bases of Politics. Portsmouth: Heinemann, 1960.

Litvack, J., J. Ahmad, and Bird, Richard. Rethinking Decentralization. Washington D C.: World Bank, 1998.

Manor, J. The Political Economy of Democratic Decentralization. Washington D.C.: World Bank, 1999.

Marko, J. 'The Referendum on Decentralization in Macedonia in 2004: A Litmus Test for Macedonia's Interethnic Relations'. European Yearbook for Minority Issues 4 (2004/05): 695-721.

McGarry, J. 'Power-sharing Theory: Lessons from the Complex Power-sharing Project'. In SettlingSelf-Determination Disputes: Complex Power-Sharing in Theory and Practice, eds. B. Metzger, M. Weller and Ν. Johnson, 691-720. Leiden: Brill, 2008.

McGarry, J., and B. O'Leary. 'Federation as a Method of Ethnic Conflict Regulation'. In Power Sharing: New Challenges for Divided Societies, eds. I. O'Flynn and D. Russell, 263-296. London: Pluto Press, 2005.

McGarry, J., B. O'Leary, and R. Simeon. 'Integration of Accommodation? The Enduring Debate in Conflict Regulation'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 41-88. Oxford: Oxford University Press, 2008b.

Mehmeti, E. 'Implementation of the Ohrid Framework Agreement'. In Power Sharing and the Implementation of the Ohrid Framework Agreement, 67-88. Skopje: Friedrich Ebert Stiftung, 2008.

Norris, P. Driving Democracy: Do Power-Sharing Institutions Work? Cambridge: Cambridge University Press, 2008.

Office of the Higher Representative and EU Special Representative. 'The General Framework Agreement for Peace in Bosnia and Herzegovina' (1995). OHR.

Organization for Security and Co-operation in Europe Spillover Monitoring Mission to Skopje (OSCE). Decentralization Survey 2009. Skopje: OSCE, 2009.

Pattern, A. 'Beyond the Dichotomy of Universalism and Difference: Four Responses to Cultural Diversity'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 91-110. Oxford: Oxford University Press, 2008.

Petrov, T. The New Territorial Reform of the Local Self-Government. Unpublished, 2004.

Pildes, R.H. 'Ethnic Identity and Democratic Institutions: A Dynamic Perspective'. In Constitutional Design for Divided Societies: Integration or Accommodation? ed. S. Choudhry, 173-201. Oxford: Oxford University Press, 2008.

Project on Ethnic Relations (PER). Macedonia's Interethnic Coalition: The First Six Months. Mavrovo: PER, 2003.

_____ Macedonia's Interethnic Coalition: Solidifying Gains. Mavrovo: PER, 2004.

Ragaru, N. Macedonia: Between Ohrid and Brussels. Paris: CERI (Centre d'Études et de Recherches Internationales), 2008.

Roeder, P. 'Power Dividing as an Alternative to Ethnie Power Sharing'. In Sustainable Peace - Power and Democracy after Civil Wars, eds. P. Roeder and D. Rothchild, 51-82. Ithaca: Cornell University Press, 2005.

Roeder, P. and D. Rothchild. Sustainable Peace - Power and Democracy after Civil Wars. Ithaca: Cornell University Press, 2005.

Rothchild, D., and C. Hartzell. 'Security in Deeply Divided Societies: The Role of Territorial Autonomy'. In Identity and Territorial Autonomy in Plural Societies, eds. W. Safran and R. Maiz, 254-271. London: Frank Cass Publishers, 2000.

Schou, ?., and M. Haug. Decentralization in Conflict and Post-Conflict Situations. Oslo: Norwegian Institute for Urban and Regional Research, 2005.

Scott, Z. Decentralization, Local Development and Social Cohesion: An Analytical Review. Birmingham: Governance and Social Development Resource Centre, 2009.

Siljanovska-Davkova, G. 'Legal and Institutional Framework of Local Self- Government in the Republic of Macedonia'. Lex Localis - Journal of Local Self-Government 7 (2009): 107-127.

Sisk, T. Power Sharing and International Mediation in Ethnic Conflicts. Washington D.C.: United States Institute of Peace, 1996.

Statistical Office of the Republic of Macedonia. Census of Population, Households and Dwellings in the Republic of Macedonia, 2002. Skopje: Government of the Republic of Macedonia, 2005.

Sulejmani, R. 'Consensus Democracy and Power-Sharing in Macedonia'. In Power Sharing and the Implementation of the Ohrid Framework Agreement, 131- 164. Skopje: Friedrich Ebert Stiftung, 2008.

Treisman, D. The Architecture of Government: Rethinking Political Decentralization. New York: Cambridge University Press, 2007.

United Nations Programme to Enhance Inter-Ethnic Dialogue and Collaboration (UN). Results of a Participatory Assessment of National and Local Capacities for Strengthening Inter-Ethnic Dialogue and Collaboration. Skopje: United Nations, 2010.

United Nations Security Council (UNSC). Comprehensive Proposal for the Kosovo Status Settlement. New York: United Nations, 2007.

Weller, M., and S. Wolff. Autonomy, Self-governance and Conflict Resolution: Innovative Approaches to Institutional Design in Divided Societies. Abingdon: Routledge, 2005.

Wolff, S. 'Approaches to Conflict Resolution in Divided Societies: The Many Uses of Territorial Self-Governance'. Ethnopolitics Papers 5 (2010).

Young, C. Ethnic Diversity and Public Policy. New York: St. Martin's Press, 1998. ZELS (Association of the Units of Local Self-Governance of the Republic of Macedonia). Doracakpraktikpër punën e komisioneve për marrëdhënie midis bashkësive (Practical Handbook for the Work of the Committees for Minority Relations). Skopje: ZELS, 2009.

_____ Membership Data for Municipal Committees for Inter-Ethnic Relations. Skopje: Unpublished, 2010.