Policies of Space and the Space of Politics: The “Negotiated Expansion” of the Belo Horizonte Metropolitan Area¹

Abstract

The aim of this article is to analyse the process of expansion of the Belo Horizonte Metropolitan Area, created in 1973 with 14 municipalities, which today comprises 34 municipalities, making it the second largest MA in Brazil. After the redemocratization, more specifically after the promulgation of the State Constitutions in the late 1980s, several new metropolitan areas were created in the country, and all nine MAs instituted in the early 1970s increased their number of member municipalities, in a context of low prioritization of the MAs by the federal government. The article highlights the factors of a legal, institutional, symbolic and political/electoral nature that explain the expansion of the BHMA, also very present in other regions. It underscores the impact of the model of metropolitan management adopted after the 1989 State Constitution on the capacity to produce cooperative action in the metropolitan sphere. It also analyses the manner in which the dilemma of collective action for the management of the MAs in the state of Minas Gerais was sought to be overcome, by means of adopting a new institutional model, in the mid-2000s.

Keywords:
Belo Horizonte Metropolitan Area; Metropolitan management; Constitution of the state of Minas Gerais; Federalism; Intergovernmental relations

When the 1988 Brazilian Constitution delegated to the Brazilian federal states the autonomy to formalize and manage regional entities (metropolitan areas, urban clusters and micro regions), it not only enabled the country's metropolitan areas (MAs), as institutional phenomena, to multiply, but also the number of municipalities belonging to the already constituted MAs to increase. The first nine MAs had been created by means of federal legislation in the early 1970s, a number that was gradually increased when the State Constitutions were promulgated in 1989.² Today (March 2010), there are 35 MAs in Brazil (Observatório das Metrópoles 2010).³

From 1973, when Complementary Law no. 14 was created, until 1988, the Belo Horizonte Metropolitan Area (BHMA) kept its original dimensions, comprising 14 municipalities. After the promulgation of the Constitution of the State of Minas Gerais in 1989, the BHMA rapidly expanded, and today includes 34 municipalities. 20 new municipalities were incorporated as members of the BHMA from 1989, on 6 separate occasions, in a process that reflected: (a) the expansion dynamic of a metropolitan cluster itself, as a social/urban planning phenomenon; (b) the creation of new municipalities (6 out of the 20 new members); and particularly (c) the new political game created around the metropolitan phenomenon, arising from the aforementioned delegation of autonomy by the 1988 Federal Constitution to the federal states to institutionalize regional entities. As we will see in greater detail, this process of expansion was not exclusive to the BHMA.

In this process of expansion, which was mostly negotiated, the difference between the MA's “spatiality” and “institutionality” increased (Moura and Firkowski 2001), in the sense that the expansion bore little relation to the growth of the urban cluster and its dynamism. This was because the municipalities incorporated into the BHMA were not very integrated into the metropolitan dynamics and some were even predominantly rural. Put another way, the outer limit of the BHMA became even more distant from the metropolitan cluster's “urban sprawl”. It is worth pointing out that this expansion was implemented at a time when the MAs found themselves to be generally “orphaned from political interest” in the country (Ribeiro 2004) and in Minas Gerais, in spite of the innovative character of several aspects of the state Constitution.

The aim of this article is to analyse this process of expansion, highlighting the impact of the model of metropolitan management adopted after the state Constitution, and the political motivations behind the expansion, as well as the bargaining tools and the nature of the bargain. It is worth mentioning the non-existence of academic research specifically dealing with this issue, which is doubtlessly a large gap in the specialist literature. This piece of work also aims to contribute towards filling in this gap.

Although our focus is mainly on the experience of the BHMA, our work frequently seeks to compare it to that of other MAs in the country. As the development of the BHMA and its metropolitan management mechanisms is paradigmatic in many aspects (by the radical nature of its expansion and the innovative character of its institutions), this research will be of interest to all of those interested in the metropolitan issue in Brazil.

As we shall see, the expansion of the Brazilian MAs took place in a generalized manner, and in the majority of cases, widened the mismatch mentioned above. The academic bibliography on metropolitan management commonly highlights that the main dilemma – which does not only occur in the Brazilian MAs – concerns the difficulties in producing cooperation in order to overcome shared problems within the space of the metropolis – known as the dilemma of collective action. The phenomenon analysed in this article, that is, the increase in the number of municipalities that make up an MA, is thus central to understanding the efficacy of metropolitan institutions. If collective action is made difficult by an increase in the number of actors whose action must be coordinated, the greater the number of municipalities that make up an MA, the greater the difficulties in generating concerted action will be. As we will later see, the new model of metropolitan management adopted in the BHMA in the mid-2000s was implemented in a very specific context. However, the factors of a legal, institutional, symbolic and political/electoral nature that favoured the expansion of the BHMA in the recent past, which will be discussed in this paper, are essentially the same as the ones that led to the expansion of other MAs in the country.

With these objectives in mind, the article will be structured as follows. In the first section, we will seek to characterize the process of creation of the new MAs in Brazil after 1989 and the increase, in the same period, in municipalities pertaining to the first-generation MAs; that is, the nine instituted in the beginning of the 1970s. The federal policy on metropolitan space started to give way to a new policy on space that was predominantly under the auspices of the state. The second section will discuss the impact of the provisions of the Constitution of the state of Minas Gerais, more specifically the impact of the new model of metropolitan management adopted by the state on the behaviour of the Mineiro (native of the state of Minas Gerais) political agents, given it created incentives for expanding the BHMA. Here, the new post-1988/89 institutions, children of the past federal policy and moulded by a certain image of the metropolitan space, began to guide a new policy on space in the sphere of the state. The third section specifically discusses the motivations of the state's governors, state deputies and mayors to increase the number of municipalities that made up the BHMA, and highlights the bargaining tools and the nature of the bargain. In this section, it is a question of discussing the space occupied by politics in the process of expansion of the BHMA, to the detriment of the physical and urban planning aspects, which, in theory, should frame the metropolitan management space. We shall observe how regional politics develop in the microcosm of the state, bounded by a certain image and a new institutionality of the metropolitan space. The fourth and final section will discuss the proposals from the beginning of the 2000s to create new MAs in the state of Minas Gerais, which caused great mobilization to seek to supersede the second-generation, post-1989 model of metropolitan management. The final considerations briefly show how the new model of metropolitan management, implemented in the state of Minas Gerais and the BHMA in the mid-2000s, promoted a concentration of power in the state government and the more developed metropolitan municipalities, and put an end to a certain policy on space in the sphere of the state. It thus created another policy, still in the process of being structured today.

The “Metropolization of Brazil” in the Post-1989 Period and the Process of Expansion of the First-Generation Metropolitan Areas

In this article, our focus is on the institutionalization of metropolitan areas, in a process which, at least in Brazil, is not always closely related to metropolises thought of as urban clusters. In order to render the issue more precise, it is pertinent to initially present a definition of the so-called “metropolitan problem”. According to Alberto Lopes (2006, 139),

The specificity of the metropolitan [question] comes from the fact that the elements of space (environment, infrastructures, social subjects) keep a close, systematic and daily interdependence, manifested in a concentrated fashion, in a certain fraction of the territory, fragmented by the current politico-administrative division.

As we will see in greater detail, technical and urban planning criteria relative to this “close, systematic and daily interdependence” were not always decisive in the institutionalization of the MAs in the country. At times, they were also neglected, such as in the incorporation of new municipalities to the first-generation MAs. This term, “first-generation” MAs, refers to the nine MAs instituted by the federal government, still during the military regime in the early 1970s, among which was the Belo Horizonte MA.⁴ Put differently, in theory, only municipalities belonging to the metropolis’ urban sprawl, which transcend the municipal boundaries or have a high level of integration into the metropolitan dynamic in social and urban planning terms, should belong to the metropolitan area as an institution. However, it also makes sense to think that municipalities should be incorporated into an MA, if, for example, independently from the level of their integration into the cluster dynamics, they possess water reserves or are willing to house solid waste produced by the metropolitan municipalities in their territory. However, what Moura and Firkowski (2001) termed the “mismatch between the spatiality and institutionality” of the country's MAs could already be noticed since the first generation of MAs in the country,

As discussed by Moura and Firkowski (2001, 105), this mismatch seems to have been present in the very origin of the MAs in Brazil, as there is evidence to show that the military government's decision about which municipalities would belong to the nine first-generation MAs was not solely based on technical criteria. According to the authors:

The inclusion of certain cities in the group [of MAs] created at that moment showed signs of weakness in the conception of the criteria that guided the selection. Such weakness became evident in the case of [the MAs of] Belém and Curitiba. In it, the metropolitan dynamic showed itself, according to some authors, to be even less intense than ones in other areas of the same state, such as in the axis formed by the cities of northern Paraná. A similar situation took place in the case of Campinas and Brasília, discarded, among others, from the federal process of institutionalization. The fact that the nine [MAs] had their respective state capitals as their seat, and that political interests prevailed to the detriment, in some cases, of a qualitatively metropolitan issue, results from this process.

Thus, it seems possible to state that, initially, the greater the “mismatch between the spatiality and institutionality” of the MAs, the greater tends to be the incidence of political factors in the constitution of the MAs, to the detriment of technical or social/urban planning considerations. In the case of the BHMA, it has already been said that the municipality of Caeté, an original member of the MA, was only included because it was where the then state governor, Israel Pinheiro, hailed from (Fernandes 2004).

Table 1 shows a classification of the municipalities that initially made up the nine first-generation MAs. If we consider the data in the second column, the large variation in the number of municipalities of those MAs – between 2 (MA of Belém) and 38 (MA of São Paulo) – becomes immediately clear. The MAs of Belo Horizonte, Curitiba and Porto Alegre, with 14 municipalities each, occupied the post of MAs with the second greatest number of member municipalities.

Table 1
Spatiality and institutionality of first-generation MAs

First-generation metropolitan areas	Number of municipalities incorporated in 1973	Non-metropolitan municipalities incorporated in 1973	% Non-metropolitan municipalities incorporated in 1973
Belém	2	0	0.0%
Belo Horizonte	14	1	7.1%
Curitiba	14	4	2.6%
Fortaleza	6*	2	33.3%
Porto Alegre	14	0	0.0%
Recife	9	0	0.0%
Rio de Janeiro**	10	0	0.0%
Salvador	9*	2	22.2%
São Paulo***	38	4	10.5%
Total	116	13	11.2%

SourceMetrópoles (20042008
• ^*
  The municipalities of Maracanaú, from the Fortaleza Metropolitan Area, and Dias D’Ávila, from the Salvador Metropolitan Area, were included in 1986. /
• ^**
  The Rio de Janeiro Metropolitan Area was created in 1974, after the fusion of the state of Rio de Janeiro with the state of Guanabara. /
• ^***
  The Vargem Grande Paulista municipality was incorporated into the São Paulo Metropolitan Area in 1983.

Columns 3 and 4 in Table 1 show the number and percentage of “non-metropolitan” municipalities incorporated into the first-generation MAs, respectively. The idea is to gauge the mismatch between spatiality and institutionality from this indicator, which, as we will next discuss, should be seen as a proxy.

The “non-metropolitan” municipalities are those which have a level of “integration into the cluster dynamics” considered to be Low or Very Low, according to classification developed by the Observatório das Metrópoles (2004) presented in the document titled Análise das Regiões Metropolitanas do Brasil: Identificação dos espaços metropolitanos e construção de tipologias (Analysis of Metropolitan Areas in Brazil: Identifying metropolitan spaces and building typologies”). For a definition of the “levels of integration”, the following indicators were used: average geometric population growth rate 1991-2000; population density; number and proportion of people that make “pendular movements” (live in one municipality and commute to work in another); proportion of non-agricultural jobs; presence of functions specific and indispensable to the movement of people and merchandise, such as location of ports and airports and the capacity to generate income through the local economy, expressed by participation in the region's GDP. As a result of this, five integration categories were created: Very High, High, Average, Low and Very Low. We should reiterate that municipalities with a Low or Very Low degree of integration were considered to be “non-metropolitan”.

The indicators, however, were produced from data referring to the 1990s and the beginning of the 2000s. This being so, in truth, columns 3 and 4 from Table 1 show a classification that does not refer to the early 1970s, when the first-generation MAs were constituted. Given the difficulty of re-doing the calculations with data from the 1970s, and because we consider it is normal that the degree of integration of the municipalities into the metropolitan cluster should grow instead of shrinking as the decades go by, it seems possible to use this classification to at least signal, in some way, the existence of the “mismatch between spatiality and institutionality” since the beginning of the institutionalization of the MAs in the country. It seems quite likely, however, that low integration municipalities originally incorporated into the first-generation MAs increased their links with the metropolitan cluster. It is reasonable, thus, to expect that when these nine MAs were constituted, the mismatch between spatiality and institutionality was significantly greater.

This aside having been made, and bearing in mind the above considerations on the other motivations to technically justify the incorporation of municipalities not belonging to the urban sprawl into the MAs, let us now explore the data in Table 1 in greater detail.

First, it is worth highlighting that the data back up the comments made by Moura and Firkowski (2001) on the specificity of the Curitiba MA, which in Table 1 appears next to the São Paulo MA, as the first-generation MA with the greatest absolute number of non-metropolitan municipalities (four). Problematic as though it may be, this indicator generally suggests that criteria for defining member municipalities other that the technical ones were given much importance, as five out of the nine MAs that incorporated municipalities, to this day, have a Low or Very Low degree of integration (considered to be non-metropolitan). These are the MAs of Belo Horizonte, Curitiba, Fortaleza, Salvador and São Paulo.

Having made these considerations and presented the mismatch between the institutionality and spatiality of the first-generation MAs, let us now continue with an analysis of the “metropolization” process in the post-1989 period in Brazil, or, more specifically, of the institutionalization of the several new MAs in Brazil – or second-generation MAs – in order to then analyse the process of expansion of the first-generation MAs, also in the post-1989 period.

The creation of new MAs in Brazil after 1989

The nine MAs created by the military regime in the early 1970s still exist today, even if their capacity to promote intergovernmental cooperation in the metropolitan sphere suffered serious setbacks as early as the 1980s, with the economic crisis and the dismantling of the superstructure of support for urban development that had been articulated by the federal government during the military regime (Lopes 2006). Aside from the nine remaining first-generation MAs, another 15 were instituted in the country after 1989 (according to other sources, this number is much higher. See footnote 3). Table 2 shows these new MAs. It might be worth recalling that the nine original MAs, whose management model was reconfigured from new state regulations in the post-1989 period – as was the case of the BHMA –, today could also be characterized as belonging to the second generation, even if their “institutional memory” has had implications on their acting capacity.

Table 2
MAs instituted in Brazil after 1989

	State	Metropolitan area	No of municipalities	RM's year of implementation
1	Alagoas	Maceió	11	1998
2	Amapá	Macapá	2	2003
3	Amazonas	Manaus	8	2007
4	Espírito Santo	Grande Vitória	7	1995
5	Goiás	Goiânia	13	1999
6	Maranhão	Grande São Luís	6	1998
7	Maranhão	Sudeste Maranhense	8	2005
8	Minas Gerais	Vale do Aço	26	1998
9	Paraíba	João Pessoa	9	2003
10	Paraná	Londrina	8	1998
11	Paraná	Maringá	13	1998
12	Rio Grande do Norte	Natal	9	1997
13	São Paulo	Baixada Santista	9	1996
14	São Paulo	Campinas	19	2000
15	Sergipe	Aracaju	4	1995
	Total		152

Source

We can see that 12 of the 26 states of the Federation (almost half of them) instituted new metropolitan areas after the transfer of responsibility from the Union to the states enabling them to do so. Three of these states created two new MAs – Maranhão, Paraná and São Paulo. Paraná and São Paulo today have three MAs each, as they already had MAs whose poles were their respective capitals. These 15 MAs grouped a total of 152 municipalities, while the nine first-generation MAs were composed of 115 municipalities when they were created in the early 1970s. Today, after the post-1989 process of expansion of the first-generation MAs, their member municipalities total 188. Regarding the implementation dates of these new MAs, it is worth noting that: (a) the country's second-generation metropolization process began in 1995, with the constitution of the MAs of Greater Vitória and Aracaju. The six year gap between the promulgation of the new State Constitutions and the actual creation of the first of the second-generation MAs seems to suggest a certain lack of priority to the issue on the part of subnational political agents, who were still deeply steeped in the “autarchic municipalism” conception;⁵ (b) 11 out of the 15 new MAs were constituted between 1995 and 2000 and (c) the other four MAs were instituted between 2003 and 2007.

These brief considerations on the process of “metropolization” in post-1989 Brazil having been made, we shall now continue with an equally brief evaluation of the expansion of first-generation MAs during this period.

The expansion of the first-generation MAs after 1989

In order to later examine the expansion of the Belo Horizonte MA from a less regionalist viewpoint, from which we can appreciate the Mineiro experience based on more general comparisons, we will now present the expansion of the first-generation MAs in the period following the promulgation of the State Constitutions in 1989.

Table 3 presents the total number of municipalities incorporated into the nine original MAs after 1989, also highlighting the number and percentage of non-metropolitan municipalities that became members of those MAs. The definition of a non-metropolitan municipality is as per the previously made specifications.

Table 3
Non-metropolitan municipalities incorporated into first-generation MAs after 1989

First-generation metropolitan areas	Total number of municipalities incorporated into the MA from 1989	Non-metropolitan municipalities incorporated from 1989	% Non-metropolitan municipalities incorporated from 1989	Total number of municipalities today
1. Belém	3	1	33.3%	5
2. Belo Horizonte	20	10	50.0%	34
3. Curitiba	12	8	66.7%	26
4. Fortaleza	7	3	42.9%	13
5. Porto Alegre	17	5	29.4%	31
6. Recife	5	1	20.0%	14
7. Rio de Janeiro	6	0	0.0%	16
8. Salvador	1	0	0.0%	10
9. São Paulo	1	0	0.0%	39
Total	72	28	38.9%	188

Source

In the early 1970s, the São Paulo MA was – and still is today – the MA with the greatest number of member municipalities (37 in 1979 and 39 today). New members were added to all nine first-generation MAs after 1989. The Belo Horizonte MA, however, expanded the most, having incorporated 20 municipalities. The Porto Alegre MA incorporated 17 new members, while the Curitiba MA incorporated 12. All other MAs incorporated up to a maximum of seven members.

Regarding the degree of integration into the metropolitan cluster dynamic, it is initially worth recalling that the indicators here are more precise than those presented in Table 1, as they are less out of date. Having mentioned the indicators’ weakness, we should note that: (a) only three out of the nine first-generation MAs did not incorporate municipalities considered to be non-metropolitan: the MAs of Rio de Janeiro, Salvador and São Paulo; (b) the Belo Horizonte MA is the one that incorporated the greatest number of non-metropolitan municipalities (10), having been the one in which perhaps the technical criteria weighed the least when the decision to incorporate new municipalities was made; (c) when we consider the percentages, we find that all six MAs that incorporated non-metropolitan municipalities did so in at least 20% of cases; (d) more than half of the new members incorporated into the Curitiba MA and half of the new municipalities of the BHMA are today non-metropolitan; and (e) in percentage terms, Curitiba, not Belo Horizonte, was the MA that most incorporated municipalities of Low and Very Low integration.

Thus, it is evident that the 1988 Federal Constitution's act of assigning to the states the duty of caring for the management of their respective territories led not only to the creation of new MAs, but also to the expansion of all first-generation MA's. However, the motivations and reasons behind this process, as well as the degree of prevalence of political criteria over technical ones, should be sought in specific enquiries, which will be made in the following sections, but only for the Minas Gerais case and for the case of the expansion of the BHMA in more detail. It is worth noting, however, that the discussion that follows will emphasise the weight of the legal criteria and of the institutional impact (of the model of metropolitan management adopted), with special emphasis on the role of variables of a political nature.

Before we begin this discussion, though, we should draw the reader's attention to an interesting comparison, which seems to us significant, even if we take the deficiencies of the indicator we are using into account. A comparison of the “Total” rows of Tables 1 and 3, which deal with the first-generation MAs and their expansion, respectively, shows that at the moment of inauguration of metropolitan management in the country, 13 non-metropolitan municipalities were incorporated (11.2% of the total), while after 1989, another 28 municipalities (38.9%) considered non-metropolitan were added to the group of nine MAs. Even recognizing that the indicator used underestimates the number of municipalities with low integration into the metropolitan dynamics incorporated into the MAs in the early 1970s, it seems pertinent to suggest that with the transfer of responsibility promoted by the 1988 Federal Constitution, the issue of metropolitan management in the country became even more vulnerable to variables of a political nature, as the mismatch between institutionality and spatiality seems to have grown.⁶

It is also worth pointing out that, if it can repair the exclusions caused by the dynamic of expansion of metropolitan clusters itself, the incorporation of new members into the MA's institutional structure also inevitably implies adding a greater number of actors to the game of cooperation in the metropolitan sphere. If cooperation tends to be more feasible when a smaller number of actors are involved, an outstanding issue is the behaviour of actors who represent municipalities with low integration into the dynamic of the metropolis in these structures (see Faria (2008) on this important issue).

The Constitution of the state of Minas Gerais, the New Model of Metropolitan Management Adopted and the Expansion of the BHMA

The previously mentioned 1988 Federal Constitution's delegation of responsibility to the states to assume their territories’ management was, as far as the metropolitan issue is concerned, superseded by the 1989 Constitution of the state of Minas Gerais, in a clear – and, in some aspects, innovative – manner. In Article no. 42, the Mineira Constitution says that the state “can institute, by complementary law, metropolitan areas and urban clusters constituted of groupings of borderline Municipalities of the same geo-economic and social complex, to integrate the planning, organization and execution of public functions of common interest”.

In truth, in this respect, the Constitution of the state of Minas Gerais (CEMG) stands out from other State Constitutions in some important aspects. When she contrasts the treatment given to the metropolitan issue by several different Brazilian State Constitutions, Rovena Negreiros (1992) points out the following peculiarities in the CEMG's original draft: the state of Minas Gerais, as the state of Ceará, might have advanced in its purposes of regional organization by indicating in the constitutional text its concern “regarding the decentralization and deconcentration parallel to regional integration”. The CEMG (Negreiros 1992, 314) provided the adoption of “specific integration tools, from policies of planned deconcentration of economic development and of the sharing of community benefits and resources for compensating the effects of polarization”. Furthermore, both the CEMG (Negreiros 1992, 315) and the constitutional texts of the states of São Paulo and Ceará sought to outline the specificities of each one of the types of regional unit, even if generically, and defined “a group of factors that must be observed regarding the classification of the municipalities for each regional unit”.

As an advancement worth highlighting, Negreiros (1992, 315) also points to the fact that Ceará and Minas Gerais listed and defined the public functions of common interest, classifying them according to the different regional units. Azevedo and Mares Guia (2000, 135), however, state that

(…) only a low number of State Constitutions exactly define the functions of common interest of municipalities belonging to metropolitan areas. The one that shows up recurrently is that of the public transport/road network system addressed in the constitutions of the Federal District, Amazonas, Ceará, Goiás, Minas Gerais, São Paulo and Paraná. After ‘public transport/road network system’, among the most cited public functions of common interest are ‘water resources’, ‘land division/use and occupation’ (Federal District, Minas Gerais, Goiás and Amazonas) and ‘environment control’ (Federal District, Minas Gerais and Amazonas).

Another important specificity of the Mineira Constitution was the definition of the “Metropolitan Necklace”,⁷ configured – as a planning tool – as a potentially important mechanism for dealing with the impact of the process of metropolization on the surrounding municipalities. These impacts are relative, for example, to the intensification of the division of urban land, to the need for a reserve of water resources and the demand for public transport (Moraes 2001).

Regarding the institutional arrangement for metropolitan management, in its original draft, the Minas Gerais constitutional text proposed the creation of a Metropolitan Assembly that would have among its attributes “the regulatory normative power to integrate planning, organization and the execution of public functions of common interest” (CEMG, Article 45, Line I). However, in a general manner, the State Constitutions mostly strengthened the need for involvement in the management of the MAs by the “community and/or municipalities (Amazonas, Paraíba, Rio de Janeiro, Espírito Santo, Minas Gerais and Rio Grande do Sul), granting local governments a prominent role in the process of metropolitan decision-making” (Azevedo and Mares Guia 2000, 136). However, in the words of Negreiros (1992, 316), “from the point of view of the institutional arrangement, the management model that most advanced was that of Minas Gerais, whether because of its democratic character or the level of political articulation it suggests”. Furthermore, in Minas Gerais, as in Pará, the State Constitutions foresaw the creation of development funds (1992, 317). However, according to Azevedo and Mares Guia (2000, 136), in

(…) terms of financial support, the constitutions of the states of Paraíba, Minas Gerais and Espírito Santo are the ones that determine specific budget items and/or mechanisms of co-responsibility of the state and municipal governments, geared towards guaranteeing resources allotted to functions of public interest.

Faria (2008), recognizing the central role given to municipalities in the institutional arrangements characteristic of the second-generation's models of metropolitan management, considered it pertinent to denominate the modality instituted by the 1989 state Constitutions as being based on a “symmetrical hyper-municipalism”, “since the tendency in the sphere of the state was to not discriminate between different roles for member municipalities, according to their economic and demographic particularities and the type of place they occupy in the metropolitan dynamics” (Faria 2008, 56).

According to the provisions of the state Constitution, the Assembleia Metropolitana de Belo Horizonte (Ambel, Belo Horizonte Metropolitan Assembly) would be constituted of mayors, councillors appointed by their respective municipal chambers, a representative of the Legislative Assembly and a representative of the state Executive, appointed by the government. In this way, the BHMA's new institutional structure completely inverted the correlation of forces in the sphere of metropolitan management, giving ample primacy to the municipalities’ interests, in contrast with the state-biased emphasis of the Federal Legislation of 1973.

However, the new model, initially perceived as democratizing, would quickly reveal itself to be inoperable. As a possibly unexpected effect, the “hyper municipalism” of the Ambel produced new obstacles to intergovernmental relations in the metropolitan sphere, as demonstrated by several authors (see Azevedo and Mares Guia 2000; 2008; Faria 2008; Machado 2009; Mares Guia 2001). To summarize very briefly, what happened was that the Metropolitan Assembly directorates started to become dominated by coalitions of smaller municipalities, which opposed the metropolitan economic axis, comprising Belo Horizonte, Betim and Contagem, usually to the detriment of the state government's interest, which as we have seen, was very modestly represented in the Ambel. The reaction by the municipalities of the economic axis and the state government then started being one of emptying the Assembly. It is also worth recalling that, as the Fundo de Desenvolvimento Metropolitano (Metropolitan Development Fund) ended up not being regulated, and the financial issue was not resolved either, which meant, in practice, that the state government kept control of a significant part of the main tools of metropolitan intervention. Such a control was thus largely carried out outside of the institutional structure for metropolitan management created by the CEMG. The Ambel's weakness was such that it did not have resources to count on, not even for the maintenance of a technical-administrative body. Azevedo's and Mares Guia's (2000, 139) argumentation seems to us as concise as it is precise:

Why should the larger municipalities of the BHMA – Belo Horizonte, Betim and Contagem – and the state government be responsible for the near totality of resources of the said Fund, if formally they have such a modest influence in the decision-making process for the allocation of funds, and therefore irrelevant political gains? In such a situation, the elementary assumptions of the logic of collective action indicate that the behaviour of the state and these municipalities is exactly as expected, since the financial costs would be immeasurably higher than the possible political returns”.

According to Machado (2007), when the meetings of the Ambel did take place, they dealt with issues specific to certain municipalities, leaving the great metropolitan problems in second place. As was pointed out by Azevedo and Mares Guia (2008), the majority of mayors and councillors who participated in the Ambel remained firmly rooted to a localist attitude of exclusively defending municipal interests, and had difficulty adopting a regional view, vital to creating intergovernmental cooperation in the metropolitan sphere. At other times, municipal governments clearly bowed to the state government representative. Still according to Mares and Guia (2000, 138),

(…) when the sole representative of the state Executive participates in meetings, he/she inexorably takes the central position Although, in theory, the metropolitan decision process essentially depends on an agreement between municipalities, the state Government holds the control of a substantial part of the relevant metropolitan intervention tools (public services such as inter-municipal transport, water supply, sewage collection, electricity, building and maintaining roads, among others).

Another interesting fact is that the devices relative to metropolitan management contained in the state Constitution were only regulated in 1993, by means of Complementary Law no. 26. As highlighted by Machado (2007), this considerable time lapse can be thought of as more evidence of the little attention dedicated by the state government to the management structure of the BHMA, which, as we have seen, would prove to be inefficient.

In order to complete our idea of the deterioration of the management structure of the BHMA, as well as the Ambel's inoperativeness, we should also recall: (a) the weakening and subsequent extinction of the Plambel, the metropolitan planning body, in 1996 and (b) the incapacity of state agencies geared towards the urban issue to cooperate, that is, the difficulty with intergovernmental coordination in the sphere of the state, which also indirectly denotes the little interest of the state government in the metropolitan issue.

As he analyses the transaction costs for states to assume metropolitan management, Machado (2009) concludes that such costs tend to be quite significant due to the following summarized factors: (a) the high political cost for states to establish legal-vertical metropolitan management parameters without the consent of local authority leaderships; (b) legal controversy – currently being discussed in the High Federal Court – about the possibility of the state law for the creation of metropolitan areas making the integrated provision of public services of metropolitan interest compulsory; (c) the low resonance of the metropolitan issue in urban social movements; (d) the lack of support from the federal government to the possibility of states having central roles in metropolitan management and (e) the preponderance of the municipalist/decentralizing paradigm in the political, technical and academic fields.

In this manner, the author adds, the vertical-compulsory forms of organization of metropolitan areas tend to have high transaction costs for the state governments. This might be one of the explanations for the low effectiveness of the management of metropolitan regions created by state laws in the country. According to the analysis of Moura and her colleagues (2003, 52-53), the metropolitan areas formally instituted by the state:

(…) are not anchored in an institutional framework that actually structures their complex dynamics. They are, recognizably, spaces of economic and social expression, but not of Law, as they do not circumscribe territories able to normatize, decide or exert power, and are located in a hiatus between a municipality's autonomy – endorsed by the 1988 Constitution – and the Union's competency regarding management for development. (…) The realization of social and territorial pacts collides against the fragility of the complex legal-institutional environment of the regions, under pressure from hegemonies and political powers, and from politico-party disputes, which damage decision-making in the regional sphere .

In turn, as Souza (2003) analyses the trajectory of the institutionalization of metropolitan areas in Brazil, she argues that the states’ difficulties in assuming the metropolitan issue more effectively can be understood based on the notion of path dependence. To the author, the model of metropolitan management created by the military in the 1970s, as well as being stigmatized as a mechanism of authoritarian acting in metropolitan spaces, was incapable of generating lasting incentives for intergovernmental cooperation, as it did not generate a collective conscience or a sense of regional identity around the importance of metropolitan issues, either. Furthermore, the essentially technical character of the metropolitan agenda during that period “went against the grain” of the demands for more pluralist and decentralized public management formats during the redemocratization period. This might have caused the Union and the states to practically fall silent about issues relative to the management of metropolitan spaces, which would partly explain the decadence and/or extinction of the majority of metropolitan entities until then in existence.

Thus, the Brazilian states did not succeed in creating second-generation metropolitan management mechanisms capable of offering the state government a more significant acting space. Furthermore, the now prevalent metropolitan area conception may also have broken with the then current concept of MA in the 1970s, in which the political variable was of relative importance, given considerations of an economic and physical-spatial nature were given priority. The very creation after 1989 of several MAs in the same state could be a sign of rupture with the previous, first-generation MA model, and closer to the North-American model. In this model, qualifying an urban sprawl as “metropolitan” depends on the existence of a pole-city with at least 50,000 inhabitants, circumvented by urban counties; that is, locations identified by the Census as possessing a large expanse of a continuous area with high population density. Machado (2009) observes that the concept of metropolitan area in the United States is more important statistically than for intergovernmental relations.

Given this general outlook of low prioritization of the metropolitan issue in the country there arises an interesting question: if the metropolitan areas were, to some degree, “abandoned”, becoming “orphaned of political interest”, as per Ribeiro's (2004) incisive statement, one might ask why so many MAs were created in Brazil after 1989 (in truth, after 1995, as shown in Table 2).

The model of metropolitan management adopted after the Mineira Constitution and the governor's lack of concern are central elements to understand the expansion of the BHMA. The next subsection is dedicated to this question. In the last section of the article, we will discuss the attempts to create new MAs in Minas Gerais.

The expansion of the BHMA

The BHMA is today composed of 34 municipalities. As we have seen, out of the 24 MAs currently instituted in Brazil (see footnote 3), the BHMA is the second largest in number of member municipalities. Only the São Paulo MA comprises a greater number of municipalities (39). It is also worth noting that only the SPMA, the BHMA and the MAs of Porto Alegre (31 municipalities) and Curitiba (26) are constituted of more than 19 municipalities. Table 4 presents the makeup of the BHMA and some of the basic characteristics of the 34 municipalities that compose it today.

Table 4
Constitution and basic characteristics of the municipalities of the BHMA

(1)	(2)	(3)	(4)	(5)	(6)	(7)
Municipalities of the BHMA	Integration into the cluster dynamic	Year of integration into the BHMA	Resident population in 2000	Distance from the capital (Km)	Area (Km2)	Population density (2000) (inhabitants/km2)
Belo Horizonte	Pole	1973	2,238,526	0	331.9	6718
Contagem	Very High	1973	538,017	16	195.2	2748
Ibirité	Very High	1973	133,044	25	73.3	1812.3
Ribeirão das Neves	Very High	1973	246,846	15	154.6	1595
Santa Luzia	Very High	1973	184,903	12	234.4	788.1
Vespasiano	Very High	1973	76,422	14	70.3	1085.7
Betim	Very High	1973	306,675	30	346.8	875.4
Sabará	High	1973	115,352	17	304.4	376.3
Caeté	Average	1973	36,299	31	542.7	66.8
Lagoa Santa	Average	1973	37,872	22	232.7	162.3
Nova Lima	Average	1973	64,387	22	429.7	149.6
Pedro Leopoldo	Average	1973	53,957	24	291.9	184.4
Raposos	Average	1973	14,289	23	72	198.2
Rio Acima	Low	1973	7,658	35	228.7	33.5
Esmeraldas	High	1989	47,090	38	912.3	50.2
Igarapé	Average	1989	24,838	46	110.3	220
Brumadinho	Low	1989	26,614	44	634.3	41.9
Mateus Leme	Low	1989	24,144	53	303.4	79.5
São José da Lapa*	High	1993	15,000	13	48.8	307.6
Juatuba*	Average	1993	16,389	43	97.1	162.3
Sarzedo*	High	1997	17,274	31	62.1	277.6
Mário Campos*	High	1997	10,535	36	35.3	298.2
Confins*	High	1997	4,880	21	42.1	113.9
S. Joaquim de Bicas*	Average	1997	18,152	42	72.7	249.7
Florestal	Very Low	1997	5,647	51	194.9	28.9
Rio Manso	Very Low	1997	4,646	62	232.8	19.9
Matozinhos	Average	1999	30,164	32	253.6	118.6
Capim Branco	Average	1999	7,900	34	94.5	83.4
Itaguara	Low	1999	11,302	85	411.9	27.4
Nova União	Very Low	1999	5,427	42	172	31.6
Baldim	Very Low	1999	8,155	59	556.7	14.6
Jaboticatubas	Very Low	2000	13,530	40	1117.1	12.1
Taquaraçu de Minas	Very Low	2000	3,491	33	330.3	10.6
Itatiaiuçu	Low	2002	8,517	64	295.9	28.8

• ^(*)
  Municipalities emancipated from others that belonged to the BHMA.
SourcesMetrópoles (20042006Faria (2008

Observing the data presented in Table 4, the great heterogeneity of the BHMA becomes evident, such is the difference in the levels of integration of the municipalities into the cluster dynamic as well as in the resident population, the distance from the capital, the municipalities’ area and their population density. Some of these discrepancies must be pointed out: (a) 17 of the 34 municipalities of the BHMA (50%) had fewer than 20,000 inhabitants in 2000; (b) 21 of the member municipalities are located more than 30km away from the capital, and the one furthest away, Jaboticatubas, has an area 31 times greater than the area of the smallest municipality, Mário Campos; (d) Ibirité is the most densely populated (1,812.3 inhabitants per km2), and 14 out of the 34 member municipalities have fewer than 100 inhabitants per km2.

Regarding the heterogeneities in the member municipalities, we should also present data not present in Table 4. It is generally expected that the municipalities that make up an MA have a high degree of urbanization. However, if 16 out of the 34 municipalities of the BHMA had urbanization rates of more than 90% in 2000, according to data from the Population Census, six of them had rates below 60%, and in Nova União and Taquaraçu de Minas the rural population was greater than the urban one. If we take into consideration that the degree of urbanization in Brazil was of 81.2% of its population, we then note that 15 out of the 34 municipalities of the BHMA had urbanization rates below the national average in 2000 (Observatório das Metrópoles 2006, 19, Table III.2).

Such heterogeneities, to which surely must be added those relative to the degree of development of the municipalities and their capacity for administrative and political negotiation, imply a management model based on a “symmetrical hyper municipalism” – the assignation of the same weight in the management structure to deeply asymmetrical political units, which can be seen as an impediment to cooperative intergovernmental relations. On the other hand, it is also possible to think that, as the co-optation of the more fragile members is facilitated, this asymmetry can favour certain articulations in the metropolitan sphere. However, as we saw in our brief discussion on the Ambel, what was found in the specific case of the BHMA was a weakening of the structure of metropolitan management.

As for the “mismatch between spatiality and institutionality”, measured by the degree of integration of member municipalities into the cluster dynamics, we note that out of 34 municipalities of the BHMA, excluding the capital, which is the Pole, six show a Very High level of integration, six a High level, 10 an Average level, five a Low level and six a Very Low level. Therefore, according to the criteria adopted, 11 of the member municipalities of the BHMA (the five Low integration ones added to the 6 Very Low integration ones) can be considered “non-metropolitan”. All these 11 municipalities were incorporated into the BHMA from 1989, Rio Acima being the only exception. This Low integration municipality already figured among the 14 original members. As we previously indicated, 50% of the municipalities incorporated into the BHMA from 1989 (10 out of 20) can be considered to be “non-metropolitan”, which shows the weight of the political criteria on the decision to expand the BHMA.⁸ However, as far as the expansion by incorporating emancipated municipalities is concerned, it is important to highlight that this was the case in six out of 20 new members, as shown in Table 4.

At this point in our argumentation, we should point out that this great mismatch between spatiality and institutionality in the expansion of the BHMA is even more surprising when we recall the following provision by the Constitution of the state of Minas Gerais in Article 44:

• Art. 44 – The institution of the metropolitan area will be made based on an evaluation of the totality of the following objectively verified data or factors, among others:

  1. population and demographic growth, with a five-yearly projection;

  2. degree of conurbation and migratory fluxes;

  3. economic activities and development perspectives;

  4. polarization factors;

  5. deficiency of public services in one or more Municipalities, with implications for the region's development.

It seems evident that strict attention to constitutional precepts, which emphasise technical and objective questions, would have resulted in a much smaller expansion of the BHMA than the one that actually happened. Regarding the inconsideration of the technical/urban planning aspects, it is worth making the following suggestion: the incorporation of new municipalities into the BHMA accelerated from 1997, when 14 out of the 20 new members were added. Whether or not a coincidence, this phenomenon occurred after the extinction of the Plambel, in 1996, the state metropolitan planning body, which at the end of the 1980s survived with difficulty.

The next section of this article will focus on an evaluation of factors of a political nature that seem to have been responsible for a significant part of the mismatch previously referred to.

The BHMA'S “Negotiated Expansion”

Having so far highlighted the impact of legal and institutional factors on the expansion of the BHMA, and indicated the spaces made for the possible prevalence of political factors over technical/urban planning ones in the decisions regarding the expansion, let us now turn to a discussion of the motivations of the political agents, and of the bargaining tools and the nature of the bargain.

Initially, we must point out the existence of strong evidence that the party variable had a low impact on decisions regarding the expansion. As any decision on the expansion of the BHMA must be made in the sphere of the Legislative Assembly of the state of Minas Gerais (ALMG), on which the state governors usually have a strong influence, the following hypothesis can legitimately be raised: municipalities whose mayors are part of a coalition with the governor's support may have their demand to formally join the MA made easier. However, such a hypothesis seems not to stand. This is because a look at the party of the incorporated municipality's mayor and the governor's coalition, carried out in the case of the 20 new municipalities added to the BHMA (whose details will not be presented here due to lack of space), revealed the following: the mayors of only 10 of the incorporated municipalities belonged to parties that made up the governor's parliamentary support base during the decision-making process, whereas the mayors of nine of the incorporated municipalities did not belong to the governor's coalition (in the specific case of the mayor of São Joaquim de Bicas, of the Partido do Movimento Democrático Brasileiro (PMDB, Brazilian Democratic Movement Party), this correlation could not be established because only part of the party belonged to the “Todos por Minas” (“All for Minas Gerais”) coalition, which gave its support to the then governor of the Partido da Social Democracia Brasileira (PSDB, (Brazilian Social Democracy Party), Eduardo Azeredo).⁹

The process of expansion of the BHMA, as an institution, took place between 1989 and 2002, as shown in Table 4. During this period, Minas Gerais was governed by four different governos: Newton Cardoso, PMDB (1987 to 1991); Hélio Garcia, Partido das Reformas Sociais (PRS, Social Reforms Party) (1991 to 1995); Eduardo de Azeredo, PSDB (1995 to 1999) and Itamar Franco, PMDB (1999 to 2003). In each of the four governments, evaluating the degree of integration of the incorporated municipalities and whether or not the mayors belonged to the governors’ support base is interesting, as well as being an important element for our discussion on the space of politics in the process of expansion of the BHMA. It is initially clear that, in the administration of all four governors, municipalities whose mayors did not belong to the governor's coalition parties were incorporated. They were nine in total, as we have seen. However, out of these nine, only four were non-metropolitan, and all were incorporated during Itamar Franco's government (Nova União, Jaboticatubas, Taquaraçu de Minas and Itatiaiuçu).

During the administration of all four governors, municipalities governed by mayors whose parties belonged to the state government's parliamentary support base were also incorporated (three in the Cardoso administration, one in the Garcia administration, four in the Azeredo administration and two in the Franco administration, totalling 10). Non-metropolitan municipalities whose mayors were the governors’ allies were incorporated during the Cardoso administration (two), the Azeredo administration (two) and the Franco administration (two).

It must also be noted that all four governors promoted or did not succeed in obstructing the incorporation of municipalities considered to be metropolitan (with Average, High and Very High levels of integration): two in the Cardoso administration, two in the Garcia administration, three in the Azeredo administration and two in Franco's. Only in the Garcia administration were exclusively municipalities considered metropolitan incorporated (São José da Lapa and Juatuba, only the mayor of the latter being the governor's ally).

From an exclusively technical or urban planning point of view, the incorporation into the MA of municipalities with a good level of integration into the metropolitan dynamic is desirable, so that spatiality and institutionality can be adjusted. However, as we have seen, in the administration of three out of four governors, the exception being Hélio Garcia's, non-metropolitan municipalitie whose mayors belonged both to parties allied to the governor and opposition groupings were also added.

The patterns outlined in the previous paragraphs generally suggest the non-determining character of the variables utilized for the incorporation of a municipality into the BHMA, which are: (a) whether the municipality's mayor belongs the governor's supporting coalition and (b) the metropolitan or non-metropolitan character of the municipality incorporated. If such factors can have influenced specific decisions, sometimes contrary to expectations, the diversity of situations analysed must lead us to exploring other determining factors, such as the role carried out by the state deputies and mayors.

It must be noted that initially, the deputies who authored the complementary bills and amendments that added the BHMA's new municipalities all belonged to parties of the then governors’ support base, except those who initiated the process during Itamar Franco's administration. It is worth recalling, as we have seen above, that if in all four governors’ administrations, municipalities whose mayors did not belong to the governor's coalition parties were incorporated, only during the Itamar Franco administration were municipalities headed by non-allied mayors, also non-metropolitan, incorporated into the MA.

In order to advance our empirical examination of the motivations and interests of the political agents in the expansion of the BHMA, we also consulted: (a) the reports issued by the Comissões de Constituição e Justiça e de Assuntos Municipais e Regionalização (Commissions of Constitution and Justice and of Municipal Matters and Regionalization) for the bills that incorporated new municipalities into the BHMA,¹⁰ and (b) the archive of statements by state deputies in the meetings of the Plenário da Assembleia Legislativa de Minas Gerais (Plenary of the Legislative Assembly of Minas Gerais).¹¹

Regarding the role of the state deputies, we must initially point out that in some cases, such as that of the proposal to incorporate the municipalities of Itaguara and Itabirito into the BHMA, the initiative was taken to the ALMG without any previous articulation between the deputy who authored the Complementary Bill and the municipality's mayor. Such a fact indicates not only the possibility of autonomous action by the state Legislature, but also that the proposals of incorporation may express possible rivalries between the municipal Executive's leaderships and state deputies who were a majority in the municipality, or who intended to expand their sphere of influence.

Here we must expand our discussion on the motivations of local political actors. Initially, let us point out that the “dynamics that have a bearing on the metropolitan territory tend to direct positive externalities to the central areas, while draining the negative ones to the outskirts” (Lopes 2006, 141), where the smaller municipalities are concentrated. Because of this, for the negatively affected municipalities on the outskirts of the MA, it is important to apply for certain investments, also being demanded by municipalities clearly more integrated into the metropolitan dynamic, in a context of the recurring insufficiency of public service provision. Thus, in spite of the low engagement of governors in the metropolitan issue, discussed in this paper's previous section, an important motivation for the mayors to join the BHMA was undoubtedly the expectation of retrieving large-scale federal funding, as in the early 1970s. A certain institutional memory seems present here, as well as the fact that several government programmes have indeed continued to prioritize the metropolitan municipalities. The importance of these expectations is evidenced by the arguments recurrently employed in discussions engaged in during the analysis of the proposals of expansion in the sphere of the state Legislature.

Another factor must be considered, which seems to us to be of utmost relevance. Even when we recall that in recent years the Brazilian metropolises have become concentrations of a significant portion of the country's “social problems”, there is evidence that a conception of the metropolis as a place of progress has subsisted, as made evident by several debates that have taken place in the state Legislature. Such a conception can be inferred from the following example: by the BR 381 motorway, in Itaguara, a municipality of low integration into the cluster dynamic incorporated into the BHMA in 1999, and the most distant from the pole-city of Belo Horizonte, a road sign was placed saying “You are already in the Belo Horizonte Metropolitan area”.

On the other hand, as a counterpoint, a state deputy's manifestation on the acrimonious contention on whether or not the municipality of Itabirito should have been added to the BHMA must be recalled. Itabirito ended up being included and then removed from the BHMA, the process having been characterized by a dispute between the state deputy who authored the proposal on one side, and the deputy with the majority in the town and its mayor on the other. The latter two opposed the incorporation, having mobilized the municipality's population. In a statement at an ALMG plenary, the deputy with the majority expressed himself thus:

I have in my hands the letters of Mr Mayor [and of several other municipal leaderships] manifesting indignation and concern about belonging to and living with the problems of these real monsters that are the metropolitan areas, and that definitely would not reach the town of Itabirito (Diário do Legislativo 27/7/1989, 42, column 3).

However, when explaining the swelling of the BHMA after 1989, we should add another explanation related to more immediate interests to these motivations of a material and symbolic character, involving short-term political gains both for the municipal leaderships and for the governors, who have the potential capacity to mediate the expansion decided in the sphere of the state Legislature. It is the fact that formal incorporation into the BHMA implied an important benefit – regular timetabled inter-municipal transport linking the member municipality to the capital, a service provided by the Departamento de Estradas de Rodagem de Minas Gerais (DER-MG, Road Department of Minas Gerais), with a reduced tariff. As the interests in non-expansion were low, since the Fundo de Desenvolvimento Metropolitano was not instituted, and, as we have seen, both the government of the state and the municipalities of the economic axis were alienated from the directorates of the Ambel, the approval of incorporation proposals was easier.

Lastly, it seems evident that full comprehension of the process of expansion of the BHMA, and of the negotiations and bargaining involved in it requires not only tracing each one of the cases in question,¹² but also an attempt to unmask an intricate universe of crossed support, omission due to lack of interest and intergovernmental rivalries between the executive and legislative branches of two levels of government – state and municipal. Given the impossibility of carrying out this type of research, we believe, however, that our discussion about the impact of the BHMA's model of metropolitan management, the expectations, interests and motivations of different actors, even if relatively generalized and with topical explanations, will have at least clarified – even if along general lines – the policies of space and the space of politics in the BHMA.

As the search for superseding this second-generation model of metropolitan management – which produced an institutional paralysis and provoked the swelling of the BHMA – was also characterized by attempts to create other MAs in the state of Minas Gerais, the last section of this article will be dedicated to a brief discussion about these attempts.

The “Mitigated Metropolization” of the state of Minas Gerais

The state of Minas Gerais today has two instituted MAs (the BHMA and the MA of Vale do Aço, created in 1989). However, in the early 2000s, the creation of nine other MAs was proposed in Minas Gerais, a process that was aborted in the sphere of the state's Legislative Assembly. Thus, we consider it pertinent to state that a process of “mitigated metropolization” occurred in the state.

At the same time as it weakened the formulation of public policies of regional governance, the senseless trend of expansion of the Metropolitan Area of Belo Horizonte – combined with the obsolescence of the state metropolitan management system – also anchored a setting that favoured the reversal of the trend in the early 2000s.

The political movement that made the reversal of the trend possible was the presentation of numerous bills proposing the institution of new MAs in the state. The projects to create another nine new MAs in Minas Gerais (MAs of Vale do Alto Paraopeba; Vale do Rio Grande; of the Triângulo Mineiro (triangular-shaped region in the very western tip of Minas Gerais); Curvelo; Montes Claros; Caratinga; Governador Valadares; Juiz de Fora and Inconfidentes (Ouro Preto region) were incoherent with the local geo-economic and urban planning reality and were presented between 2001 and 2003.

As expected, such projects were presented by parliamentarians with a majority in those regions. The sudden interest in the creation of so many new MAs, in a kind of herding behaviour, also observed in Santa Catarina, was cause for great institutional concern on the part of the Legislative Assembly of Minas Gerais. State deputies at the time denounced the existence of what was termed “a fad”. A crucial aspect, which guaranteed the conditions for proceeding to a more all-encompassing discussion on the metropolitan issue, involving social actors of the most different types relevant in this arena, was the institutional prestige and tradition of the Legislative Assembly in organizing large-scale events for the discussion of issues on the agenda.

To start with, several technical preparatory meetings and five regional gatherings were organized between the months of August and November 2003, in five different cities of the state's hinterland. Wide mobilization prior to the event guaranteed a significant amount of participation by several social segments in the activities of the Legislative Seminar on the Metropolitan Issue in 2003. The participants were representatives from groups of organized society directly affected by the metropolitan issue and representatives from mayorships, as well as researchers and technicians linked to different spheres of government, which made possible the strengthening of the idea of the metropolitan issue as a “problem” to be dealt with in an institutional manner, with society's participation.

The state Executive, on its part, had designated technicians who were part of the staff of its recently created Superintendência de Assuntos Metropolitanos (Superintendency of Metropolitan Matters), linked to the Secretaria de Estado de Desenvolvimento Regional e Política Urbana (SEDRU – State Department of Regional Development and Urban Planning), active also in the preparatory meetings and regional gatherings. This intense process of meetings and debates gave rise to a large dossier called Documento Síntese das Comissões Técnicas Interinstitucionais (CTIs, Synthesis Document of the Interinstitutional Technical Commissions (ITCs)), containing analyses and dozens of proposals for the institutional reformulation of metropolitan management in the state.

The work done in the Regional Encounters diagnosed that in all the regions in which the institution of new MAs was intended to happen there lacked the basic characterizing elements of a metropolitan area, such as conurbation and the urban metropolitan scale. The advantages and disadvantages of a region becoming an MA were discussed. Among the advantages, the prerogatives for obtaining financial resources by municipalities formally defined as metropolitan were emphasised. However, the Seminar opened up a space and opportunity for technicians and academics to expose the outlook of complete paralysis of the metropolitan management of the state's main MA, that of Belo Horizonte.

During the Regional Meetings, a change of focus was advocated, and the following items were proposed: priority for the reformulation of the structure of the BHMA's management, and the adoption of alternative modelling for micro regions for the regionalization of municipal groupings that did not have “metropolitan area” characteristics. In November 2003, the final debates of the Seminar took place, with discussions, voting and approval of 184 proposals for the metropolitan issue in the state. Media coverage reported the presence of more than 700 participants in this final phase. It could be said that the Legislative Seminar, open to the participation of all of society and with a strong presence of academics, ended up serving as a channel for several proposals, considered progressive, for equating the metropolitan dilemmas.

In this way, political support was obtained for some basic premises that would guide the normative changes that followed. One change that stood out was the fact that in the new institutional format, the power of several actors could not be strongly asymmetrical in relation to the basic correlations of force in existence. As a result of this debate, sufficient political capital was gathered, based on Article 44 of the state Constitution, to support the archiving of all bills to create new MAs in the state,

The effects of this initiative, which reversed the trend of creating new MAs in the state, carried on in the following years. In 2004, Constitutional Amendment no. 65, which altered eight articles of the state Constitution, was approved. Such an amendment, inspired by those debates, institutionalized more rigid mechanisms for the creation of new MAs in Minas Gerais, thereby opening up a space for the institutionalization of a new state metropolitan management system for the BHMA, a process which counted on strong support from the state government.

Conclusion

In an address in special meeting no. 65 of 15/12/1999, in the Comissão de Assuntos Municipais da ALMG (Commission of Municipal Matters of the Legislative Assembly of Minas Gerais), which discussed the inclusion of new municipalities into the BHMA, deputy Irani Barbosa stated, in a sarcastic tone: “Mr. President, evaluating the bill after it having been voted, I would like to ask your Excellency if it is also possible to place the municipality of Ibiá (…) into the BHMA, which takes up almost the entire state. We are going to have to create the metropolitan area of the metropolitan area.”. As we saw in the previous sections, the swelling of the BHMA came about as a result of factors of different natures, such as legal, institutional, symbolic and political/electoral factors, which most probably could also be considered responsible for the expansion of the country's other MAs.

Having even given rise to joking manifestations such as the one cited above, the process of expansion of the BHMA was halted from 2003. With the alterations in the state Constitution, which created a new institutional apparatus for metropolitan management in Minas Gerais, the process of incorporation of new municipalities into the state's two MAs was also constrained. This because it became compulsory for bills for the institution or alteration of the MAs to be based on technical studies to assess the population, population growth, degree of conurbation and “pendular movement”, economic activity, polarization factors and lack of public services, reiterating the original text of the state Constitution of Minas Gerais. According to Ribeiro (2007, 7), such a measure meant “a constraining of political action and its dependence on technical-scientific knowledge”.

As it redistributed the power of different government actors, the design of the new metropolitan institutions of Minas Gerais clearly implied a “re-statization” of metropolitan management in the state, as well as the recognition of the differing weight of the municipalities of the economic axis of the BHMA (In the Ambel, which continues with representation from all 34 municipalities, the state today has 50% of the votes on the deliberations. The Deliberative Council is also based on the parity between state and municipalities in the deliberations, and the participation of two representatives of civil society is also instituted.). It is hoped that the new institutional, third-generation model will guarantee the engagement of actors with greater resources to make a concerted effort in the metropolitan issue.

Evaluating the legislative reforms that recently occurred in Minas Gerais, which, contrary to the country's current conjuncture, made possible for the state to assume a clearer leadership position in the Metropolitan Area of Belo Horizonte, Azevedo, Mares Guia and Machado (2008) argue that the fact that the bills in question were initiatives by parliamentarians and not by the state Executive was a determining factor for the success of these reforms. According to this analysis, the fact that the government of the state supported these proposals in the legislative arena without having had a central role in the discussions allowed the transactional costs in question to be diluted in the sphere of the Legislative, which allowed them to be superseded. Such authors also add that the good political relations between the then governor of the state (Aécio Neves, of the PSDB) and the then mayor of Belo Horizonte (Fernando Pimentel, of the Partido dos Trabalhadores (PT, Workers’ Party)) were also a determining factor for the new legislation to be made viable.

As we saw in Minas Gerais, overcoming the institutional paralysis resulting from the model of metropolitan management adopted after the promulgation of the 1989 state Constitution – the second-generation model – was possible thanks to the initiative by the state's Legislature as well as the Executive, in close alliance with the capital's mayor. In the search to supersede the dilemma of collective action in the metropolitan sphere, the relevance of the action by the architects of the policy of renovation of metropolitan management, as well as the selective incentives adopted, must also be highlighted. As a positive selective incentive, the injecting of resources – particularly by the state government – into large projects of metropolitan scope must also be pointed out. As a negative selective incentive, we can recall the threat of reducing the number of member municipalities of the BHMA made in the state legislative sphere.

Lastly, it is also worth making the term “negotiated expansion” clearer. This term is pertinent not so much due to the fact that the BHMA's expansion, in the period 1989-1992, involved bargaining that resulted in agreements between actors whose interest was explicit and whose gains were negotiated. What we saw was largely a business of votes, of the search for the expansion of the sphere of influence and prestige of local political actors. In this game, which should be based on the search for intergovernmental, intragovernmental and inter-sector cooperation, the disinterested, the omissive and the uninformed, as we have seen, played an important role.

Notes

• 1
  A previous version of this article was presented at the 33rd Annual Meeting of the Associação Nacional de Pós-Graduação e Pesquisa em Ciências Sociais (Anpocs, National Association of Social Science Research and Postgraduate Studies), Caxambu, Brazil, 2009 in the WG The metropolis and the social issue. We, the authors, would like to thank Clarisse Goulart Paradis for her competent and dedicated assistance in the research. We would also like to thank the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG, State of Minas Gerais Research Foundation), the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, National Council for Scientific and Technological Development) and the Fundo de Incentivo à Pesquisa da PUC-Minas (Research Incentive Fund of the Pontifical Catholic University of Minas Gerais), which, on different occasions, supported the research of which this article is a product. We must also thank Ronaldo Guimarães Gouvêa, for his attentive reading of the previous version, and the two referees of the BPSR.
• 2
  The Constitution of the state of Rondônia was promulgated in 1991.
• 3
  We must point out that there is not a consensus on the total number of MAs currently instituted in the country, which in itself is a significant fact, indicating the dynamism of the process of creation (and extinction) of the MAs in Brazil. This article, however, will work with a list of 24 MAs produced by the Observatório das Metrópoles (Observatory of the Metropolises) (2008) which, even though out of date, provides vital data for the discussion we will be making here.
• 4
  The nine first-generation MAs are the following: Belém, Belo Horizonte, Curitiba, Fortaleza, Porto Alegre, Recife, Rio de Janeiro, Salvador and São Paulo.
• 5
  However, when we ascertain the time gap between the State Constitutions and the expansion of the first-generation MAs, we have a more ambiguous panorama. This because two of the nine original MAs (Belo Horizonte and Porto Alegre) had already started their expansion process in 1989, and Salvador added a new municipality in 1990. The six other first-generation MAs expanded in 1993 (Rio de Janeiro and São Paulo), 1994 (Curitiba and Recife), 1995 (Belém) and 1999 (Fortaleza) (Observatório das Metrópoles 2004; 2008).
• 6
  This statement is further backed up when we perceive that in the case of the 15 MAs created after 1989, in Table 2, the level of non-metropolitan municipalities that compose them is also high. Available data referring to 11 out of those 15 MAs (excluding the MAs of Aracaju, Macapá, Manaus and Sudeste Maranhense) show that they all incorporated municipalities considered to be non-metropolitan, in a proportion of at least 11% of their members. In six out of those 11 new MAs, more than half of the member municipalities are considered non-metropolitan (Observatório das Metrópoles 2004; 2008).
• 7
  As well as Minas Gerais, the state of Santa Catarina is the only other state in Brazil that identifies “metropolitan expansion areas”, which in Minas are termed “Metropolitan Necklaces” (Moura et al 2003). The Urban Necklace of the BHMA is today made up of 14 municipalities: Barão de Cocais, Belo Vale, Bonfim, Fortuna de Minas, Funilândia, Inhaúma, Itabirito, Itaúna, Moeda, Pará de Minas, Prudente de Morais, Santa Bárbara, São José da Varginha and Sete Lagoas.
• 8
  In order to corroborate our word of warning on the frailty of the indicator used, it is worth pointing out that the municipality of Rio Manso – incorporated in 1997 and considered non-metropolitan, as it had a “Very Low” level of integration – is one of the “water tanks” of the BHMA. This was frequently mentioned in parliamentary debates when the proposal to incorporate the municipality was making its way through the Minas Gerais Legislative Assembly.
• 9
  When new members of the BHMA that were not recently emancipated municipalities are evaluated, we have a similar result. Seven of them had mayors belonging to the governor's support coalition when they were incorporated into the MA, whereas the other seven had mayors from parties that did not belong to said coalition.
• 10
  Available at: <http://www.almg.gov.br/index.asp?grupo=legislacao&diretorio=njmg&arquivo=legislacao_mineira>.
• 11
  Available at: <http://www.almg.gov.br/index.asp?grupo=atividade_parlamentar&diretorio=pronunciamentos&arquivo=pronunciamentos>
• 12
  Inclusion of 20 new municipalities, based on six different legal frameworks – the State Constitution itself, which determined the inclusion of four municipalities, and Complementary Laws nos. 26, 48, 53, 56 and 63.
• Revised by Priscila Moura

Bibliographical References

• Azevedo, Sergio, and Virgínia R. Mares Guia. 2008. “Atores e formato institucional: a Assembléia Metropolitana de Belo Horizonte”. In Metrópole: território, sociedade e política. O caso da Região Metropolitana de Belo Horizonte, org. Luciana T. Andrade, Jupira G. Mendonça and Carlos Aurélio P. Faria. Belo Horizonte: Ed. PUC Minas.
• Azevedo, Sergio, and Virgínia R. Mares Guia. 2000. Governança metropolitana e reforma do Estado: o caso de Belo Horizonte. Revista Brasileira de Estudos Urbanos e Regionais, no. 3: 131-46.
• Azevedo, Sergio, and Virgínia R. Mares Guia, and Gustavo Gomes Machado. 2008. “A prefeitura de Belo Horizonte e a questão metropolitana”. In Democracia participativa: a experiência de Belo Horizonte, org. Sergio Azevedo and Ana Luiza Nabuco. Belo Horizonte: Ed. Leitura.
• Faria, Carlos Aurélio Pimenta de. 2008. De volta a um futuro incerto: relações intergovernamentais e gestão metropolitana na BHMA. Revista Brasileira de Estudos Urbanos e Regionais 10 (2): 51-71.
• Fernandes, Edésio. 2004. Gestão metropolitana. Cadernos da Escola do Legislativo 7 (12): 68-99.
• Lopes, Alberto. 2006. “Gestão metropolitana no Brasil: da coerção simétrica ao voluntarismo sem modelo, em busca da responsabilidade política com resultados”. In Metrópole: governo, sociedade e território, org. Catia A. Silva, Désirée G. Freire and Floriano J. G. Oliveira. Rio de Janeiro: DP&A/FAPERJ.
• Machado, Gustavo Gomes. 2009. Gestão metropolitana e autonomia municipal. Dilemas das transações federativas Belo Horizonte: Ed. PUC Minas.
• Machado, Gustavo Gomes. 2007. Falhas seqüenciais nas reformas administrativas: o caso da Ambel. Belo Horizonte: PUC-Minas/PPGCS, mimeo.
• Mares Guia, V. R. 2001. “A gestão na região metropolitana de Belo Horizonte: avanços e limites”. In Direito urbanístico e política urbana no Brasil, org. E. Fernandes. Belo Horizonte: Ed. Del Rey.
• Moraes, Fernanda B. de. 2001. “Algumas considerações sobre a institucionalização das regiões metropolitanas e as mudanças ocorridas a partir da Constituição de 1988”. In Direito urbanístico e política urbana no Brasil, org. E. Fernandes. Belo Horizonte: Ed. Del Rey.
• Moura, Rosa, and Olga L. Firkowski. 2001. Metrópoles e regiões metropolitanas: o que isso tem em comum? Anais do IX Encontro Nacional da ANPUR. Rio de Janeiro, May.
• Moura, Rosa, Paulo Roberto Delgado, Marley V. Deschamps, and Nelson Ari Cardoso. 2003. Brasil metropolitano: uma configuração heterogênea. Revista Paranaense de Desenvolvimento, no. 105: 33-56.
• Negreiros, Rovena. 1992. Gestão urbana: as novas realidades constitucionais e as diversas abordagens estaduais da questão metropolitana. Anais do VIII Encontro Nacional de Estudos Populacionais 3: 311-30. ABEP: São Paulo.
• Observatório das Metrópoles. 2010. Regiões metropolitanas do Brasil. http://www.observatoriodasmetropoles.net/download/observatorio_RMs2010.pdf (accessed April 2011).
  » http://www.observatoriodasmetropoles.net/download/observatorio_RMs2010.pdf
• Observatório das Metrópoles. 2008. Lista atualizada das regiões metropolitanas atuais. 2008. http://www.observatoriodasmetropoles.ufrj.br/MunRM_RIDE_30.11.2008.pdf
  » http://www.observatoriodasmetropoles.ufrj.br/MunRM_RIDE_30.11.2008.pdf
• Observatório das Metrópoles. 2006. Como anda a BHMA. http://www.observatoriodasmetropoles.ufrj.br/como_anda/como_anda_RM_belohorizonte.pdf (accessed June 2008).
  » http://www.observatoriodasmetropoles.ufrj.br/como_anda/como_anda_RM_belohorizonte.pdf
• Observatório das Metrópoles. 2004. Análise das regiões metropolitanas do Brasil: identificação dos espaços metropolitanos e construção de tipologias. http://www.observatoriodasmetropoles.ufrj.br/produtos/produto_mc_1.pdf (accessed June 2008).
  » http://www.observatoriodasmetropoles.ufrj.br/produtos/produto_mc_1.pdf
• Ribeiro, Guilherme Wagner. 2007. As regiões metropolitanas em Minas Gerais: entre a dependência da trajetória e a capacidade de inovação. Belo Horizonte: PUC-Minas/PPGCS, mimeo.
• Ribeiro, Luiz Cesar de Queiroz. 2004. “Apresentação: as metrópoles e a sociedade brasileira: futuro comprometido?” In Metrópoles: entre a coesão e a fragmentação, a cooperação e o conflito, org. L. C. Q. Ribeiro. São Paulo: Ed. Fundação Perseu Abramo.
• Souza, Celina. 2003. Regiões metropolitanas: condicionantes do regime político. Lua Nova, no. 59: 137-58.

Publication Dates

• Publication in this collection
  27 Oct 2023
• Date of issue
  2011

History

• Received
  Apr 2011
• Accepted
  Oct 2011