Externalactorsplayedakeyroleinencouragingandenablingethnicaccommodation in most cases examined. They were both directly and indirectly involved in the
7 Gretchen Helmke and Steven Levitsky, 'Informal Institutions and Comparative Politics: A Research Agenda', Perspectives on Politics, 2 (December 2004): 725–40.
8 See Lauth, Gryzmala-Busse and Luong, and Helme and Levitsky on this, as well as the discussion in Chapter 2. policy processes in each country. In Bosnia, the High Representative directly participated in the policy-making process by establishing the reform commissions, appointing their chairs and members and setting the rules of procedure. Paddy Ashdown chose which of the PRC conclusions to follow, while Miroslav Lajčak came up with his own proposal for police reform after previous versions had not been accepted by the RS parliament. The High Representative was appointed as the final arbiter of the Dayton provisions and their meanings. Their role in Bosnian politics is to ensure that Dayton agreement is respected by all sides and any resistance to it is removed. With such powers at their disposal, the HRs is an important actor in the policy process in Bosnia, and their behaviour affects the actions of the domestic actors.
Even though he did not refrain from putting pressure on RS politicians to initiate reforms, during military reforms in Bosnia Ashdown largely fulfilled his role in a neutral fashion, which enabled domestic politicians to find a compromise. To the extent that he had a preference, he represented the interest of the wider international community, on behalf of whom he was appointed as HR. Similarly to NATO representatives, for example, he spoke about the need to centralise and cut the size of the army. However, during police reform, he was perceived by RS politicians to be acting in a biased manner, first by refusing to instruct the PRC to use consensus and then by insisting on a reform proposal that the Serb leaders found unacceptable. Along with a lack of coordination between him and the EU in meeting the requirements for successful police reform, this inhibited compromise instead of enabling it. Later on, when Miroslav Lajčak tried to amend the quorum rules in the BIH Parliament, the HR's reputation suffered further. His actions to relax decision-making rules were seen as an attempt to scrap veto mechanisms that had given minority groups protection from outvoting. The HR's failure in this case to act as a credible and unbiased factor in the policy process further contributed to the failure of police reform.
In Macedonia, there is no similar figure to the High Representative, so the direct involvement of external actors in the policy process is more limited. Most of the external input in domestic politics is provided through the indirect, advisory and facilitating efforts of international organisations and the ambassadors of powerful countries and organisations, such as the US and the EU. In both countries, statements, criticisms and visits from high-ranking international officials tend to receive a lot of domestic attention and can affect the behaviour of domestic politicians. This indirect influence of external actors has generally proved helpful in encouraging ethnic accommodation.
NATO and EU officials used membership conditionality to elicit greater willingness for accommodation among the political leadership. In Macedonia, when the initial law on local government was debated in 2002, its adoption was set as a condition by EU and NATO representatives for holding the donors' conference that would raise funds for post-conflict reconstruction. In 2004, the EU firmly took the government's side in supporting the new territorial organisation and encouraging people not to vote on the referendum over municipal boundaries. In both instances, conditionality resulted in the expected outcomes, although in the case of the referendum it was aimed at the population rather than the politicians. In Bosnia, conditionality was applied in both police and military reform, but was only successful in the first case. The difference between the two cases was that with military reform NATO had a clear and consistent set of conditions, which neither changed over time nor were they tailored specifically for the country's accession to the Partnership for Peace programme. The EU had no prior criteria for police restructuring and introduced its three conditions specifically for Bosnia. It eventually went back on all of them, demonstrating that the conditions applied were neither consistent nor fair. Credible and consistent conditionality therefore produced better results in encouraging accommodation, as the rewards and punishments administered to political elites were more predictable and reliable.
The case of police reform in Bosnia also illustrates the limits of conditionality as an instrument for propelling domestic reforms. Contrary to the claims of the literature on the 'transformative power' of EU and NATO, conditionality can sometimes fail to trigger the desired transformation in the receiving state.9 The case of Bosnian police reform shows that this process can run in both directions and instead of EU or NATO integration efforts working to de-ethnicise domestic issues, ethnic divisions can spill into EU and NATO integration, making them ethnically divisive issues. RS politicians certainly resented the role that the EU played during the repeated police reform attempts. Their trust in EU integration and their commitment to the EU integration process has waned as a result. That can undermine the commitment to EU and NATO membership as a shared goal among all groups in society, and can remove a powerful lever that those organisations have at their disposal. Moreover, ethnicisation of EU or NATO integration removes a significant potential cross-cutting cleavage in politics that could otherwise re-frame policy debates in terms other than gains and costs for ethnic groups.