Sabani argues that the Ohrid Agreement impacted on primary education mainly through its provisions for decentralisation.211 Indeed, the agreement called for enhanced ‘powers of elected local officials’ and enlarged ‘competencies in ... public services’ including education. This decentralisation was expected to allow educational change, to depoliticise decisions over schooling reforms, and to produce policies reflecting more closely the needs of all citizens, including the members of smaller ethnic communi- ties.212 In fact, decentralisation did not alter the basic centralised structure of Macedonia’s education system. Rather, it facilitated the evolution of its schools from instruments for nation-state building to tools for cementing the mutually exclusive ethnic, linguistic and political communities in the state.
Education was one of the first responsibilities devolved to municipalities in 2005. Hereafter, municipalities became school founders and owners with the power to appoint school directors and pay for the salaries of teachers and supporting staff.213 Despite the official political discourse, Macedonia’s educational decentralisation was quite limited: the government could revoke unilaterally the powers of municipalities over education, and Skopje determined funding allocations, curricular contents and pedagogical approaches.214
Moreover, the impact of decentralisation was curtailed by the lack of local capacity among school directors, school boards and parent councils. Indeed, despite the creation of the ambiguous professional figures of municipal education instructor and municipal education officer, some mayors recognised their lack of local capacity and attempted to return their responsibilities to Skopje.215 Additionally, school appointments came to reflect the clientelistic networks of politicians: new mayors repeatedly dismissed school directors and staff to replace them with individuals affiliated to their political party.216
Furthermore, decentralisation has not depoliticised the financing of education: critics report that unclear procedures for the allocation of block grants translate into higher financing for municipalities ruled by majority parties.217 Similarly, procedures for the allocation of funding do not make the school network more adaptable to changing demography. Decentralisation may have made it easier to open Turkish-, Albanian- or Serbian-language classes. Yet decisions on whether to open new classes and build new schools are still in the hands of the Education minister. These decisions are dictated by political as much as financial concerns. Thus, Sabani and Lyon observe that the school network still reflects the demography of the 1960s and has not adapted to urbanisation or to the demographic growth of ethnic Albanians.218 As a consequence, almost a third of Macedonia’s primary schools have less than 20 pupils, while many urban schools are overcrowded and operate in shifts (with different groups of students attending school at different times of the day).219 Lyon proposes that this is because successive governments have deliberately avoided redistributing resources from underpopulated areas to overpopulated ones. Such a decision would be ‘understandably sensitive’: ethnic Albanians inhabit overpopulated regions, while ethnic Macedonians are the majority in underpopulated zones.220 Small schools in mountainous areas exemplify this problem, because the closure of a ‘monolingual’ school ‘means losing out to the other community’.221 Thus, rather than depoliticising education policy, decentralisation is contributing to the perception of schools as symbols of a community’s hold over a territory.
Finally, decentralisation has furthered an ethnic division of labour in education. Frequently, the ethnic Albanian deputy minister of education deals with Albanian-majority municipalities and Albanian-language schools, while the ethnic Macedonian education minister tackles issues in Macedonian-majority municipalities and Macedonian-language schools.222 To the public, the Ministry of Education appears ‘ethnically split’.223 Far from enhancing its legitimacy, this hampers public trust in the Ministry of Education: it creates perceptions of a marginalised Albanian deputy minister and discredits ministerial initiatives in the eyes of the Albanian community.224 However, the ethnic division of labour also serves to construct an ethnic Albanian elite as the only political referent for the ethnic Albanian community, and ethnic Macedonian politicians as the only referent for the ethnic Macedonian public. As such, it supports the short-term legitimacy and stability of consociational politics.