The process of local institutional reform that different Western European countries underwent starting in the 1990s converged towards the direct election of mayors in Italy, England, Austria and Germany. Using a comparative analysis of those countries and including in it Belgium and the Netherlands (where the introduction of directly elected mayors is planned) and Greece and Portugal (with a long tradition of it), the article examines why countries with such different municipal traditions have tended towards the same institutional solution. The answer runs through the text implicitly: the institutional tradition of each country appears to have a greater capacity for explaining the outcome of the reform rather than isomorphism in the introduction of the direct election of mayors. Notes 1 In the Netherlands, a Royal Decree of the State Government in 1998 created the Royal Commission on Dualism and Local Democracy. This Commission organised an important research and consultation process into local government reform, which materialised in a white paper on the process of the reform of local political structures. This white paper is at the origin of the legislative change carried out with the approval of the Local Government Law of 2002 and of the proposed establishment of the direct election of mayors, which is still being debated (Arendsen, Citation2003: 11). In Belgium in July 2001, a special law was promulgated which allowed for the regionalisation of the new Belgian municipal law. The Walloon and Flemish regions and, to a lesser extent the Brussels region, can regulate the functioning of their local authorities autonomously. For the moment, however, a debate on changing the system used to elect mayors has only just begun in the Flemish region. On 17 July 2002, the Flemish government presented a bill which favoured the direct election of mayors. To these countries we can also add the cases of Ireland and Norway, which we have excluded from the object of the study. In Ireland, there were plans to establish the direct election of mayors in 2004. However, in the end, this did not occur. In Norway, pilot schemes have been carried out in several towns to analyse the introduction of this system. We have, however, limited our analysis to European Union member countries. Nevertheless, we should point out that all the German Länder have introduced the direct election of mayors, with the logical exception of the city-states of Berlin, Bremen and Hamburg. 2 In the case of Greece, we focused on the 900 demoi or urban municipalities, which, along with the koinotites or rural communities, make up the municipal map of the country. 3 According to Sartori (Citation1994), if a mayor is elected directly by a popular vote, he/she cannot be dismissed by a council vote during his/her mandate and leads the executive that he/she designates, this is a pure presidential system. A clear separation of powers between the executive and the legislature is established, which in parliamentary systems is impossible because the government is nominated and, sometimes dismissed, by a parliamentary vote. 4 The collective leadership model is characterised by the preponderance of a collegial executive organ formed by the mayor and the deputy mayors, which makes the executive decisions. On the other hand, the committee–leader form is characterised by the sharing of executive power between the mayor and the committees created by the council or the chief executive officer. 5 The exception is the Land of Hesse, where the magistrate continues to be the collegial decision-making organ, led by the directly elected mayor. 6 In the previous system, executive decisions were made by committees and subcommittees created by the Council. This was thus a collegial government system, which generated an ‘invisible’ leadership (John & Gains, Citation2005: 7) in which the political leaders were the town councillors who led the local political group. They were often unknown to the citizens, they were barely legitimised due to the low electoral turnout and did not have a salary or the formal powers essential for exercising executive leadership (John, Citation2001: 149). 7 In the first two, the mayor is elected directly – as in London according to the Greater London Authority Act 1999, which we have excluded from the article. However, in all the executive models proposed, the creation of an executive decision-making structure stands out, which is separate from the Council and controls the creation of policies and their implementation, with the configuration of this figure as a local political leader (John & Gains, Citation2005: 7). 8 The scope that local governments had for choosing one of the executive models contemplated in the Local Government Act 2000 was the result of political pragmatism, although the direct election of mayors was clearly one of the central government's aims, as seen from its introduction in London (John, Citation2001: 149). Despite this, two years after the law was passed, 81 per cent of the English local authorities had opted for the leader and cabinet system, which is like the parliamentary system, and 15 per cent for the modified committee system (Gains, Citation2004). Only 3 per cent – in other words, 11 local authorities – had introduced the direction election of mayors, in addition to the Greater London Authorities, which are governed by specific legislation. Of these 11, only one had opted for the mayor–manager system (Stoker, Citation2004: 4). 9 The only discordant element of this tradition in the Belgian and Dutch local political systems is the form of electing the mayor by a central authority. At present, in Belgium the mayor is nominated by the King after being chosen from among the elected candidates and after the municipal council has been consulted. Despite the fact that historically there has been no dissonance between the wishes of this organ and the proclamation by the monarch and that, therefore, this is an indirect system; the role of the King has a margin of discretion of certain importance. This is because, in the first place, at any time the King can ask for a new presentation of candidates for bourgmestre (burgomaster), and, secondly, although exceptionally, because – upon confirmation by the council – he can nominate any Belgian elector 25 years or over who is not a town councillor as mayor. In the Netherlands, on the other hand, the mayor is nominated by the state government after consultation with the municipal council and the King's commissioner in the province, from amongst a college made up of former échevins (deputy mayors) civil servants, etc. Unlike the case of Belgium, in the Netherlands, the mayor is a figure generally not linked to the locality. 10 In the Belgian case, this verification is reinforced by the fact that the bill establishes that the town councillor electoral system will no longer follow the Imperiali method and the d'Hondt system is to be adopted (Ackaert, Citation2005: 165), which is slightly more proportional. 11 In Austria, this election has to be carried out respecting the proportionality of the political composition of the council, whilst in Belgium and the Netherlands it is carried out by a majority vote and there are no plans for this to change. In most of the German Länder, the council has to vote for deputy mayors (Beigeordnete), who head sectorial departments of the local administration, subordinated to the mayor (Wollmann, Citation2004: 159). However, with the exception of the Land of Hesse, the local legislation does not provide for the existence of a local collegiate executive organ. 12 Nevertheless, the adoption of the vote of no confidence in presidential regimes has a long tradition in Latin American countries (see AAVV, 1992; Mainwaring & Shugart, Citation1997). 13 Sartori (Citation1994) does not share Shugart and Carey's (Citation1992) semi-presidentialism proposal, whereby, according to this author, the distinction between the premier–presidentialism and the president–parliamentary systems does not offer a great or discernible difference. An analysis of the institutional variables that characterise them offers us the opportunity to highlight the great hybridity of the forms of local government in the countries we studied.
No takes yet. Share an insight, caveat, or question.
Magre et al. (2007) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: