Migration and Urbanization in Brazil: Processes of Spatial ... - CiteSeerX

Migration and Urbanization in Brazil: Processes of Spatial Concentration and Deconcentration and the Recent Debate 1

Ralfo Matos

Rosana Baeninger2

To be presented at the XXIV General Population Conference International Union for the Scientific Study of Population Salvador, Bahia, Brazil, August 18-24, 2001

1

Full Professor and Researcher at the Institute of Geography at the Federal University of Minas Gerais. Full Professor at the Department of Sociology and Researcher at the Population Studies Center (NEPO) at the State University of Campinas (UNICAMP). 2

Migration and Urbanization in Brazil: Processes of Spatial Concentration and Deconcentration and the Recent Debate Ralfo Matos Rosana Baeninger

1. Introduction Internal migration and urbanization are key social processes behind the structural transformations promoted by the progress of modern capitalism in XX century Brazil. Besides, these processes also explain the occupation of the national territory and the recent dispersion and redistribution population movements within the country. This work attempts to examine, synthetically, this important aspect of Brazil’s contemporary socio-spatial development, vis-à-vis a review of the pertinent literature, exposition of controversial issues and basic data, trying to bring evidence for the population deconcentration process, with focus on the last 30 years. To do so, the first part of this text presents brief comments on the previous historical stage of demographic and urban-industrial concentration. From this background, a discussion about the subsequent stage is presented. It is important to stress that from this period on, the concentration trend began to wane, and new spatial processes of economic and population deconcentration emerged. Within this context, the discussion focused on the possible effects of economic deconcentration /reconcentration and population redistribution processes. Therefore, a discussion based on empirical data and recent studies on the rhythm and intensity of the deceleration of metropolitan growth became necessary, in order to best visualize the spreading of a new deconcentrated urbanization trend, which, in turn, promoted changes in the senses and directions of migratory movements. Thus, the last part of this study analyses the redistribution of population through the urban network, emphasizing the role played by non-metropolitan urban agglomerations and non-primate places during the recent process of urbanization and reorganization of population in the space.

2

2. Demographic and Urban-industrial Concentration in the Country The stages of emergence, expansion and consolidation of urban-industrial concentration in Brazil began at the end of the XIX century, lasting until around 19703. Based on Faria (1978), the path of the Brazilian urbanization can be divided in four different moments. The first, encompassing a longer period of time, from 1885-1929, with the subsequent ones being shorter: 1930-1955, 1956-1966, and 1967-1973. Throughout those periods, Brazil experienced different stages of economic development, in which the internal and international migrations constituted one of driving mechanisms behind the occupation of the national territory4. In fact, the dynamics of population spatial redistribution are linked, historically, to the structural transformations undergone by the Brazilian society, in which migratory and urbanization processes were important inputs. Within this context, the population movements were engendered according to the trajectories imposed by the transformations taking place in the national economy. Brazilian internal migration has been a key factor behind the redistribution of population in space since last quarter of the XIX century, with the end of slavery (Balán, 1974). The transition to waged work, along with the qualitative and quantitative contributions brought by international immigration, the transformations taking place the in productive system, prevalent from the end of the XIX century up to 1930, were important issues behind the structuring of South-Central Brazil (Cano, 1977). In fact, these changes unleashed the urban-industrial process that gained momentum afterwards (Lopes, 1973). Starting from the 1930’s, up to the mid 1950’s, the internal migratory flows became increasingly stronger towards the Southeast, South and Center-West regions, as a result of the import substitution industrialization; urbanization growth, most notably along Rio/São Paulo axis (Faria, 1983, Cano, 1977); and the expansion of the agricultural frontier in Paraná, Mato Grosso, Goiás, and Maranhão States (Martine, 1987). The Brazilian urban network up to 1950 was incipient in terms of number and size of cities. The densest and more dynamic urbanization was concentrated in the states of São Paulo and Rio de Janeiro, where the highway-railway systems favored the formation and development of an interlinked and integrated urban system. This situation, however, was changed in the 1950’s. Martine et al (1990) leave no doubts about the dynamism of the urbanization resulting from the high rates of natural increase among urban populations, and 3

See Cano (1977) and Faria (1978) about the process of economic and urbanization development in Brazil.

3

the impressive contributions of internal rural-urban movements. In fact, the growth rates of urban Brazil, between 1950 and 1960, reached the unprecedented level of 5,3% a year. This tremendous growth, in turn, led to an unprecedented rate of urbanization in 1960, when 45% of Brazilians lived cities, against 36% in 1950. During the 1960´s, the expansion of the Brazilian economy and the inertial effects of the demographic growth, derived from the mortality transition5, engendered several urban concentration cores. As a matter of fact, the period 1967-1973, the so-called ‘economic miracle’, represents a new stage for urbanization and migration. The maturation of the urban-industrial development process, along with the increasing presence of unionized workers, set the tone for the transformations in the Brazilian social and political structure. In this stage, short or long distance internal movements represented an effective means of social mobility (Faria, 1991; Dedecca, 1992; Martine and Pelliano, 1981). In spite of social exclusion, delinquency, and poverty among low-income populations, the economic development model in course promoted expressive levels of economic and social inclusion for many rural-urban migrants, especially in the more economically dynamic areas of Southeast Brazil, strengthening the links among urbanization, migration, and employment. Industrial modernization, along with the expansion of the highway system, the construction of Brasília, the beginning of the agricultural modernization, and the occupation of new resources frontiers6 accompanied that economic boom7. As a result, in 1970, for the first time, the urban population outnumbered rural one, 52 million against 41 million, respectively. At the same token, the exhaustion of certain agricultural frontier areas took place, adding fuel to the creation of new migratory trajectories8, which reinforced the population concentration pattern in favor of the ever encroaching cities (Martine, 1987). Concurrently,

4

Among the National Library studies the studies by Faria (1991), Balán (1974, Singer (1973), Lopes (1973, Lopes e Patarra (1975, Cano (1977), Graham e Hollanda (1980), among others, stand out. 5 The effects of demographic expansion, derived from the decrease in mortality rates, favor the onset of the “urban agglomeration”, as well as the process of interiorization of the urban network, as the population which survived the diseases typical of the rural Brazil, were increasingly moving into the cities, where the availability of medications and social infrastructure were expanding. 6 Prefigured in the rush to the Center-West and Amazon regions, privileging sectors like farming and mining (Martins and Camargo, 1984). 7 Cano (1985), studying the industrial movement and the regional concentration between 1930-1970, has shown the increasing concentration in São Paulo, when the national “periphery” also exhibited an accelerated industrial expansion, although at rates quite inferior to those of São Paulo. During the so-called “restricted industrialization”, São Paulo determined the form and the rhythm of the expansion, by consolidating the most diversified and dynamic heavy industry of Brazil. In the periphery, the chaining effects and a consistent expansion of internal markets were still absent, especially concerning the intermediate goods and capital assets. 8 See Brito (1997).

4

the expansion of large urban centers gained new spatial expression giving rise to metropolitan configurations9. On the other hand, the old ideals of interiorization of Brazilian development were becoming reality by the end of 1960’s. The changes in the productive structure, brought about by the “economic miracle”, and the materialization of strategic public and private investments, promoted the economic decentralization of Brazil, which, in turn, influenced the urbanization process, by expanding the urban network, and making the development of the interior more consistent and conspicuous (Negri, 1992). Thus, the path of internal population movements in Brazil, and the dynamics of urbanization prevalent in the 1930-1980 period promoted deep economic and social transformations, converting Brazil from an agrarian to an urban society. As a result, an intense and concentrated pattern of urban growth appeared at the more economically dynamic areas of Southeast Brazil, in spite of the industrial deconcentration trend inaugurated during the 1970’s.10

3. The debate over recent processes of spatial deconcentration, economic decentralization11 and redistribution of urban population12 The debate concerning the economic decentralization of Brazil is relatively recent13. Some studies suggest that the economic concentration, based on the industrial expansion of São Paulo State lost momentum and intensity in the 1970’s, which in turn led to an unprecedented deconcentration pattern. In fact, Amarante and Bondioli (1987) demonstrated that Southeast Brazil experienced a decrease in its share of the regional income between 1975-1985, promoting a reduction in terms of regional economic disparities. Martine and Diniz (1989) claimed that between 1981 and 1988, production rates of the transformation industry of São Paulo State grew at a smaller pace than the national average. At the same 9

Not by chance, the metropolitan regions were instituted in 1973 by the Federal Government Complementary Regulation (Lei Complementar) # 9, after all, nearly 70% of the internal migration of the 1970’s were directed to the metropolitan regions. 10 See Cano (1995), Diniz (1995), Pacheco (1998) and Negri (1996), among others. 11 Decentralization and deconcentration are not identical concepts. The first may refer to the act of creating or consolidating secondary centers by means of administrative, political and economical actions, which, do not necessarily materialize spatial relocalizations, nor encompass the production sphere of action, stricto senso. Such actions are usually conducted by different spheres of public power. The second concept, though similar to the first, refers more directly to the new localizations of human settlements and productive facilities originated from the departure of economic activities and populations from crowded and historically visible central spaces. 12 The Brazilian authors who focus the discussion of the economical deconcentration at the light of the urban and regional studies are privileged in this work. 13 The discussion about the decentralization of economic activities in industrialized countries, or those with a sound industrial base, is relatively new, and evidence of this process is pointed out by several scholars in recent decades, namely: Greenwood (1980), Linn (1981), Katzmam (1986), Redwood (1988), Richardson (1980), Townroe and Ken (1984),

5

token, São Paulo’s GDP data also demonstrated a sharp decline in the relative participation of the agriculture and industrial sectors, between 1970 and 1980, while witnessing a growing services sector. On the other hand, other scholars discussed the reach of the economic deconcentration phenomenon raising other arguments. Azzoni (1986:124-5) questioned if in fact the trend toward economic and spatial decentralization was widespread, or if it was the result of accommodating processes promoted by companies in response to the variations in locational costs, engendered by the fast urban-industrial growth, with limited space amplitude. The author, additionally, questions the validity of city-size as an indicator of agglomerative economies, introducing the concept of regionalization economies: “the region as an agglomerative field exercising attraction power on new industrial investments”. In this regard, cities pop up within the agglomerative field, taking advantage of satisfactory agglomerative advantages, without the need of covering the high locational costs of big cities, as in large urban conglomerates there is a trend toward the homogenization of services, transports, communications, etc. Haddad (1989), discussing the same subject, underlined some points that questioned the deconcentration thesis. Among his arguments, it is worth stressing the ones concerning the inertial resistances that would obstacle alterations in the concentrating locational pattern; the uncertainties associated with the presence of the necessary conditions for the sustained growth of peripheries; the still attractive externalities present around the large metropolis, especially at times of economic expansion, when new technological cycles would not spare the central atmosphere. Even Martine and Diniz (1989) questioned whether there was a significant loss in São Paulo, or if the deconcentrating trend was nothing but a natural outgrowth of the economic core area. Besides, it was uncertain if the deconcentration trend would persist, or if it was just a typical manifestation of the 1970’s ambiance, which took place in response to the heavy State investments of the time. Azzoni (1986:124) concludes that: a) the internal industrial movements of São Paulo State would be associated " to the variations in terms of cost and the locational advantages in each area "; b) the surveyed companies were quite condescending to decisions emanating from the central administrations located in the core area. The evidences would indicate that " far from constituting a sign of Reversion of the Polarization, the phenomenon observed in among others. Katzmam, for example, did not hesitate to state that, during the 1970’s, “for the first time in history, the phenomenon of the polarization reversal” had occurred. (Katzman, 1986:196).

6

São Paulo would represent more the spreading of industrial activities within the most industrialized area of Brazil, in a concentrated-deconcentration". Therefore, the trend could be conceptualized as a type of suburbanization of industrial activities in a broader region; c) the countless advantages of the central location (transportation, goods and services, stimulating industrial atmosphere, etc.) affect the companies in general, but most especially the smaller ones. On the other hand, with technological progress, several industrial sectors, characterized by spatially separated administration and production facilities, and by easy access to markets and services, overcome the spatial restrictions related to proximity to the core area, since the same attraction power has extended to larger areas. Diniz (1993), however, criticizes Azzoni’s methodological options, arguing that he could not have taken the State of São Paulo as a parameter for establishing comparisons with the rest of Brazil, as the large core are of Brazil, the greatest industrial concentration of the country was located within its metropolitan area. Besides, Azzoni failed to observe that the industrial spreading extrapolated its agglomerative field radius (approximately 150 Km), reaching a much wider area (extending for more than 1.700 Km), encompassing the central parts of

Minas Gerais State and going all the way to northeast Rio Grande do Sul.

According to Diniz (1993), the deconcentration only occurred among well-equipped selective and externalities-rich areas, comprising the metropolises of Belo Horizonte, Porto Alegre and extensive parts of São Paulo and Southern states, delineating what began to be called “polygonal development.” It is important to point out that the Brazilian industrialization experienced extremely high growth rates, especially during the 1950’s and 1960’s, period matched by the diversification of investments (with an expressive participation of foreign capital), which were channeled to the heavy industry of São Paulo (Cano, 1985). The large industrial concentration of São Paulo took place in a horizontal and vertical integrated fashion, being favored by the high levels of income concentration and by the expansion of the durable goods sector during the “economic miracle.” Cano (1985) also points that during the 19191970 period, no industrial stagnation took place among the Brazilian regions, although the rates of economic expansion of São Paulo State were higher than ones being experienced by the rest of Brazil. However, the author registered a decrease in terms of São Paulo’s intermediate, capital and durable goods sectors between 1959 and 1970, despite the expansion of the non-durable goods sector. Negri (2000) furthers the discussion about the determinants of the industrial deconcentration of the 1970’s and 1980’s, associating the phenomenon to several works, 7

programs, and policies fostered by the Federal and State Governments. The author relates, for instance, the pro-deconcentration policies promoted by the Federal administrations (especially those embedded in the II PND), that explicitly proposed several regional development goals by means of fiscal incentives. These policies, in turn, made the national periphery more attractive to investments at the time. The economic deconcentration within Brazil is certainly associated with the intervening action of the State, despite the failures of certain actions, which resulted in the worsening of regional unbalances. Certain public goals and investments produced, however, short and long-term impacts, like the project of interiorization of economic development embedded in Kubitcheck’s Plano de Metas and in the construction of Brasília in the late 1950’s. Such actions generated various multiplier effects, allowing the creation of new industrial areas, along with the metropolization of Brazil during the 1960’s. At the same token, these actions also led to the installation of industries in selective spaces of the interior of São Paulo State. It is also noteworthy the fact that the interior of São Paulo State presented expressive increments in its transport and circulation systems since the 1970’s. Thus, it becomes patent that the debate over industrial deconcentration, should involve a discussion about the diseconomies of agglomeration associated with large cities, in which the main industries are installed. Likewise, “geoeconomical” factors should also be present in the discussion about decentralization, especially those referring to the interregional articulation, which contributes to the diffusion of development and dispersion effects throughout the urban network. Moreover, it is essential to count on dynamic organizational structures - involving institutions and firms distributed throughout the national space - and on transportation and communication systems that favor the spatial proliferation, not only of industries, but also of services companies hooked up to the industrialization process. Redwood (1988), inspired by the Brazilian case, affirmed that decentralization may take place whenever the national urban population is concentrated in one or two large centers of countries with extensive territories, and holding a good number of secondary cities. The author claimed that, in Brazil, the “depolarization” would be linked to the concomitant process of industrial decentralization of São Paulo and Rio de Janeiro, vis-à-vis the increasing importance of regional metropolises and secondary cities, as alternative economic locations. According to the author, during the 1970’s the reduction in the polarization trend of the metropolis of São Paulo, both in the interior of the state and in the national urban system, indicated the beginning of an urban decentralization trend and a great 8

change in the population distribution pattern. When analyzing the Brazilian urban network, through population and industrial location data, he concluded that the deconcentration of Brazilian metropolitan areas took place, favoring adjacent areas and mid-size cities. Due to these transformations, the demetropolization thesis arises. Martine (1992), while analyzing data from the 1991 census coined the term. However, in a 1994 work, the author corrects himself, stating that the process represented much more than a decrease in the rhythm of demographic growth of large metropolises. Azzoni (1995), dealing with what he denominates “metropolitan sociospatial formation” , observes that the “demetropolization” (derived from the decrease of the growth rates of Brazilian metropolises) does not find support in empiric data (except in Rio de Janeiro, São Paulo and Belém). In the other cases, in spite of decreasing growth rates, the weight of metropolitan areas in terms of State’s overall population continues to decrease. Overall, the studies devoted14 to the possible impacts of the deconcentration of economic activities and the spatial redistribution of populations suggest the presence of: 1. expressive transformations concerning the State of São Paulo, indicating an incipient process of population deconcentration; 2. increasing urbanization of areas and states serving as deconcentration channels, namely Minas Gerais, Rio Grande do Sul, Paraná, and the Mid West; 3. expressive return migration flows to traditional migrant source states; 4. increasing intra-regional migration; 5. relative deconcentration of the Brazilian urban system, with an increasing presence of small and intermediate cities within the dynamics of urban agglomerations, especially among metropolitan ones.

Thus, one can concluded that the demographic deconcentration coincides with the economic deconcentration in several aspects. It is true that the deseconomies of agglomeration reach both economic activities and population, as well as governmental efforts, directed to the interiorization of development, may disperse economic activities and population. However, despite the fact that the 1970’s were characterized by the beginning of an industrial deconcentration process, one did not witness population deconcentration trend of the same magnitude. Would there be a possible gap between these two forces? Still, it is

14

See Negri (1996), Baeninger (1999), Patarra et al (1997),Pacheco e Patarra (1999), Cunha (1998 and 1999), Pacheco (1998), Martine (1994), IPEA/IBGE/NESUR (2000), among others.

9

true that within these “deconcentrations”, the participation of metropolitan peripheries as regional immigrants pull areas, and as areas of great urbanization expansion is indisputable. On the other hand, despite the investments directed toward medium cities in the period, their growth become noticeable in the 1980’s. In fact, only then the evidences of population deconcentration were clearer, when a strong reduction of metropolitan growth rates took place, especially among industrialized metropolises; areas where expressive contingents of returning migrants originated, heading to the interior of the states. The evolution of the urban deconcentration process in Brazil articulated with the predominance of the urban-urban type migratory movements, starting from the 1970’s. This phenomenon has directly contributed to the diffusion of the urbanization process to the rest of the country, through a more disperse and internalized urban network, in which an unprecedented expansion of many medium urban nuclei with important influence areas begins. Concomitantly, the intra-regional movements increase which, direct or indirectly, participate in the process of population deconcentration. Thus, one should consider the effects of the economic dynamics on new trajectories of population movements. Here, the discussion based on the industrial dynamics is insufficient to explain the intra-regional redistribution of the populations, notably in the North, Center-West and even in Northeast Brazil, areas characterized by remarkably different economic activities, such as tourism, services, agriculture, agroindustry etc. Based on the economic dynamics, the 1980’s and 1990’s, one no longer registers the same industrial deconcentration trend, typical of the 1970’s, despite the fact that “the first five years of the 1980’s were characterized by a great progress of the interiorization of industrial activities in São Paulo State” (Negri, 1996:216). Starting in 1985 and lasting until the early 1990’s, the process of industrial deconcentration lost momentum, due to the exhaustion of the investments cycles of the 1970’s. The industrial interiorization has certainly continued, but at a lower rate, given the loss of dynamism of the industry as a whole (Pacheco, 1998). Thus, the 1980’s raised new issues. Back then, the Brazilian industry demonstrated a performance far inferior than the 1970’s due to macroeconomic adjustment policies, the chronic inflation, and inconsistent economic approaches. These issues caused an absolute decrease in industrial output introducing changes in the industrial structure. During this period, a significant reduction in the participation of intermediate goods and capital

10

industries was witnessed, but, on the other hand, the sectors that managed to gain international markets, such as agroindustry, metallurgy, and footwear grew. (Negri, 1996). Therefore, the development model based on an intervening state had expired. Concomitantly, the financial resources that triggered the projects and investments of the previous decades became scarce; and the international financial flows changed in favor of developed economies. The financing difficulties experienced by the Brazilian State, and the foreign debt crisis provided more visibility to the fiscal crisis of the period, strongly decelerating the economy, and consequently, expanding the Brazilian social debt. Thus, the foundations of the urban poverty in metropolitan areas were set. Unlike the 1970’s, public resources for transport, housing, and sanitation became scarce; sectors that employed a large number of laborers and softened the social inequalities. Still, during the 1990’s, Brazil experienced an annual decrease of about 15,9 thousand jobs, which resulted in a dramatic decrease in industrial employment. According to Porchmann (2001:102) “(...) While in the decade of 1980’s the annual average evolution of formal employment was positive, 2,2%, in the 1990’s it was negative,–17,8%. The trade and service sectors experienced a positive performance, but the construction and transformation industries declined.” Coutinho (1995) points that the long economic crisis of the 1990’s was worsened by the historic iniquities of Brazil, which forced the local companies to introduce the so-called “defensive adjustments.” The modernization and administrative downsizing of companies; the compression of industrial production; the intense subcontracting; and the combination of economic recession and open markets culminated in the elimination of roughly 1,5 million formal industrial jobs between 1983 and 1993. Despite the fact that a large part of that quantum was transferred to subcontracting activities, several professional occupations simply disappeared from the job market. Eventually, the “structural tendency that presents clear indications of permanence is the restructuring of work, motivated by the introduction of technological innovations (...) and a decrease in the formal urban employment” (Coutinho, 1995:59-60). Within this vein, it is worthwhile resuming the debate questioning if the productive restructuring, by privileging more economically dynamic centers, would lead to an industrial reconcentration. According to Negri e Pacheco (1993), the new Brazilian industrial spatial arrangements, in tune with the international economic restructuring,15 shall find in São 15

This trend is marked by the geographical dispersion of industrial production; see Piore and Sabel (1984); Harvey (1992); Sassen (1988); Benko (1996), among others.

11

Paulo the most appropriate spaces for the evolution of this capitalist phase. Cano (1995) and Diniz (1995) also suggest a likely industrial reconcentration, especially the high tech ones, in São Paulo State. Based on the recent economic discussion, may one think about an articulation between the dynamics of economic deconcentration/reconcentration and population deconcentration/reconcentration? Before a supposed economic reconcentration, in course since the mid-1990’s, would there be a remetropolization of the population trend taking place? These are the questions that should permeate the debate on economic dynamics and population distribution during this decade. At any rate, it is reasonable to speculate that there would hardly be a reconcentration of population similar to the urban concentration of the 1930-1970 period. Firstly, because as the 1970’s illustrated, the population does not follow the geographies of economic activities linearly. Secondly, the new industrial paradigm is anchored in low levels of labor absorption, privileging the most qualified workers. Finally, the urbanization levels of areas outside the South-central axis absorb a large portion of the populations that potentially could migrate to the so-called historical areas of economical concentration, consolidating a relatively deconcentrated urban network.

4. Evidences of Deconcentrated Urbanization

It is unquestionable that a great part of the urbanization of later decades is associated with the multiplying effects derived from the “spreading” of the historical urban-industrial concentration in the Southeast. In fact, almost half of the urban centers of Brazil in the year 2000 were located in the region, among which the three major metropolitan areas. This industrial spreading promoted the densification of the Brazilian urban system and the interdependence and complementarity links among strategic cities located in the different parts of the system. Such nodal points articulate the urban network, synthesizing, on a spatial basis, long-term structural changes. These spaces constitute alternative locations for economic investments, favoring the appearance of new roles and different levels of economic specialization, through which goods, people and capital circulate. Moreover, these new spaces are not necessarily restricted to a pre-defined perimeter in the South-southeast, as Diniz proposes (1993).16 16

The author, in Diniz and Crocco (1996), when trying to evaluate the economic restructuration and the new map of Brazilian industrialization, agrees, in a reluctant way, that a relative geographic dispersion directed to most of the regions

12

In fact, it is more likely that the Brazilian urban system consolidates its structure based on several hybrid forms, among which a set of articulated spatial structures of dense development , where new opportunities for public and private investments may contribute to reinforce current tendencies, besides reinforcing the interiorization of the urban network . In this venue, Motta and Ajara (1999:14) suggest that “the several articulation and spatial integration types present among the urban centers, at the same time they express their insertion and the role they play in the productive structure, they reflect the several possible arrangements and engender a peculiar spatial configuration for each segment of the urban network.” The Brazilian urbanization, without fail, has impacted the process of population redistribution. In fact, the urban-industrial expansion made the urban centers responsible for 90% of everything that was produced. Consequently, this situation has promoted a vertiginous concentration process of people and activities in some of the largest metropolises on earth: São Paulo and Rio de Janeiro, respectively, the 3rd and 15th largest cities in the world. According to the preliminary results of the 2000 demographic census, 81,2% of the Brazilian population resided in urban areas, a figure that will probably reach 88% by the year 2025, according to projections set forth by the United Nations (IPEA, 2000). In spite of the stock of resident population among large metropolitan concentrations, it is important to stress that Brazil, throughout the 1990’s, continued to experience an important process of redistribution of population. This spatial organization, in turn, was characterized by the maintenance of the interiorization of population growth trend and the formation of new urban agglomerations (IPEA / IBGE /NESUR-UNICAMP, 2000). As a matter of fact, the study by IPEA/IBGE/NESUR-UNICAMP, indicate that the so-called “spreading” of the Brazilian urbanization is matched by the interiorization of the metropolization trend itself, as shown by the examples of Goiânia and Campinas (SP) , and by the non-metropolitan urban agglomerations of Cabo Frio (RJ), Ipatinga (MG), Itajaí (SC), among others. Moreover, the study identifies the presence of a significant growth among medium urban centers - those with population varying between 100 thousand and 500 thousand inhabitants. A tentative explanation for such changes resides in the process of subcontracting promoted by metropolitan urban economies, which has provoked the

and states had begun, notwithstanding the possibility of a geographical reconcentration in South-Center Brazil; in a region extending from Rio Grande do Sul to the center of Minas Gerais.

13

reduction of waged work and income levels, which, in turn, reflects an increasing inability to absorb immigrants. Thus, based on population studies, it is necessary to resume a more general discussion on the new population deconcentration trends, emphasizing the inflection of the metropolitan growth; the dynamics of urban population growth and the urban network; and the dynamics of the non-primate urban areas.

4.1–Deceleration in the Rhythm of Metropolitan Growth Despite the fact that Brazilian metropolitan areas have attracted and held large populations for several decades, between 1970-1980 a decrease in terms of their population growth began to be noticeable. In fact, the average rate of metropolitan growth changed from 4.7% per year, between 1960-1970, to 3.8%, between 1970-1980, presenting in that decade growth levels below the average Brazilian population (4.44%) Still, the migratory flows directed toward these metropolises revealed a population concentration pattern: 5,705,021 people17 headed to metropolitan areas between 1970-1980, corresponding to 22.5% of the national inter-municipal migration. In 1980 and 1990, however, the growth rates of Brazilian metropolitan population slowed down, reaching 1.99% per year between 1980/91 and 1.79% per year between 19912000. The deceleration trend undoubtedly had an inflection during the 1980’s, consolidating an incipient tendency observed during the 1970’s in which the Brazilian urban population grew faster (2.97% per year., from 1980-1991, and 2.42%, from 1991-2000) than the metropolitan areas (Table 1). The 1970’s, in fact, were characterized by an accentuated phenomenon of metropolization, especially in the regions of Belém, Fortaleza, Salvador, Belo Horizonte, São Paulo and Curitiba, which registered growth rates superior or close to the average growth rates for the urban population at the time. The absolute population increase among those regions between 1970 and 1980, was 8,3 million people, which corresponded to a 30% growth in the national urban population. The deceleration trend can be grasped based on 1980-1991 data. During this period, the rate of metropolitan growth, except for Salvador, Curitiba and Belém, which grew above the national average, the largest metropolitan areas - São Paulo and Rio de Janeiro - started 17

Refers to people who had moved from the municipality, at least once, in the decade. The intra-metropolitan migration is not included.

14

to present growth rates far below metropolitan average growth: 1.88% per year, and 1.26%, respectively, against 1.99% per year.

Table 1 Total Population and Growth Rates (% per year) Metropolitan Regions - 1970 - 2000 Metropolitan Regions

1970

TOTAL

Population 1980

Growth Rates (% per year) 1991

23,730,895 32,126,519 39,759,370

São Paulo Rio de Janeiro Recife Belo Horizonte Porto Alegre Salvador Fortaleza Curitiba Belém

8,139,730 6,891,521 1,791,322 1,658,482 1,574,239 1,147,821 1,036,779 821,233 669,768

12,588,725 8,772,265 2,347,146 2,609,583 2,285,140 1,766,614 1,580,074 1,440,626 1,021,486

15,444,941 9,814,574 2,858,147 3,436,060 3,038,792 2,496,521 2,307,017 2,000,805 1,401,305

URBAN BRAZIL 52,084,984 80,436,409 110,990,990 Source: FIBGE, Demographic Censuses 1980, 1991 and 2000.

2000 1970-19801980-19911991-2000 3.79 1.99 1.79 46,657,093 17,833,511 4.46 1.88 1.61 10,871,960 2.45 1.26 1.14 3,249,754 2.74 1.22 1.44 4,251,350 4.70 2.42 2.39 3,507,624 3.84 2.16 1.61 3,018,326 4.43 3.04 2.13 2,920,923 4.29 2.67 2.66 2,716,288 5.80 3.03 3.46 1,794,981 4.30 3.43 2.79

137,697,439

4.44

2.97

2.42

The results of the Demographic Census of 2000 suggest that the 1990’s advanced the deceleration of metropolitan process, especially in São Paulo (which registered from 19912000 a growth rate of 1.61% per year), Rio de Janeiro (1.14% ), and Porto Alegre (1,61% ), way below the national urban (2.42% ) and metropolitan averages (1.79%). The metropolitan regions of Recife, Belo Horizonte and Fortaleza, however, maintained the growth rates during the 1990’s quite close to those observed during the previous decade. Except for Curitiba, which registered an elevation in its rate of metropolitan growth (3.03% per year., in the 1980’s, to 3.46%, in the 1990’s), it can be observed in the other regions the continuity of the metropolitan growth decrease. The metropolitan Brazil of the 1990’s, therefore, grows less than it did in previous decades, and even less than the other urban areas of the country. Regardless of the fact that between 1970 and 2000, the resident population among nine old metropolitan areas has changed from 23,717,028 inhabitants to 50,164,717, the importance of the metropolitan deceleration process is revealed by the expressive losses of metropolitan populations in relative terms. In fact, in 1970, the national metropolitan population accounted for 45.56% of the Brazilian urban population, decreasing to 33.88%, in 2000 (Table 2). The largest decreases were registered in the metropolises of São Paulo State - which represented 15.6% of the national urban population in 1970 and 1980, and 12.9%, in 2000 -, and Rio de Janeiro,

15

dropping from 13.2% in 1970 to 7.9% in 2000. Fortaleza, Curitiba and Belém, experienced a slight growth on their relative share of the urban Brazilian population, despite not representing much in terms of the overall urban population (around 2%). These evidences already indicate that until the year 2000, even in the presence of some regional oscillations, a retraction in metropolitan deceleration, which could have been transformed into a possible metropolitan reconcentration was not observed.

Table 2 Relative Participation (%) in the National Urban Population Old Metropolitan Regions – 1970/2000

Metropolitan Regions

TOTAL São Paulo Rio de Janeiro Recife Belo Horizonte Porto Alegre Salvador Fortaleza Curitiba Belém

1970

1980

1991

2000

45.56

39.94

35.82

33.88

15.63 13.23 3.44 3.18 3.02 2.20 1.99 1.58 1.29

15.65 10.91 2.92 3.24 2.84 2.20 1.96 1.79 1.27

13.92 8.84 2.58 3.10 2.74 2.25 2.08 1.80 1.26

12.95 7.90 2.36 3.09 2.55 2.19 2.12 1.97 1.30

URBAN BRAZIL 100.00 100.00 100.00 100,00 SOURCE: FIBGE, Demographic Censuses 1980, 1991 and 2000.

Considering the metropolitan population within the urban population of various Brazilian States (Table 3), it is noticeable that between 1970 and 2000, there was an incipient process of metropolitan deconcentration. The metropolitan areas that lost more relative participation in terms of the total urban population of their respective states were Belém (accounted for 65.5% and dropped to 43.6%), Recife (from 63.7% to 53.7%) and Rio de Janeiro (from 87.2% to 78.8%). Conversely, with small oscillations, the other regions are maintaining in a fairly stable way their share of the total urban population within their states. Thus, within the state context, given the enormous resident population present at metropolitan areas, the deconcentration in course reflects very little in terms of macro-scale indicators. In the case of São Paulo, for instance, the Metropolitan Area of São Paulo accounted for 57% of the urban population in the state in 1970, dropping to 51%, in 2000.

16

Table 3 Relative Participation (%) in terms of State's Urban Population Old Metropolitan Regions – 1970/2000

Metropolitan Relative Participation (%) 1970 1980 1991 2000 Regions TOTAL 55.19 50.05 46.60 45.18 Belém 65.54 61.26 53.97 43.61 Fortaleza 58.24 56.22 55.43 55.07 Recife 63.73 62.02 56.58 53.70 Salvador 37.20 37.91 35.58 34.45 Belo Horizonte 27.37 29.05 29.15 29.07 Rio de Janeiro 87.17 84.61 80.45 78.79 São Paulo 57.02 56.72 52.69 51.65 Curitiba 32.79 32.21 32.28 34.90 Porto Alegre 44.31 43.52 43.43 42.20 Source: FIBGE, Demographic Censuses 1980, 1991 and 2000.

In attempting to understand the process of urbanization dispersion throughout the country, it is necessary to consider, besides the population trajectories of old metropolitan areas, the recent processes that involve the creation of new metropolitan areas (Brasília, Vitória, Campinas, Santos, Natal and Goiânia). Those new metropolises exhibited a growth rate of 5.83% per year in the 1970’s, dropping to 3.54% per year during the 1980’s, and 2.8% per year in the 1990’s. It is important to stress that these average growth rates are far superior than the ones experienced by the urban population Brazil, as well as by the old metropolises. While the old metropolitan regions are losing relative weight in terms of the Brazilian urban population, new MR are growing, and represented 7.78% of the total national urban population by the year of 2000. (Table 4).

Table 4 Metropolitan Population, Growth Rates (% p.a.) and Relative Participation (%) Old and New Metropolitan Regions – 1970/2000 Metropolitan Population Regions (MR) 1970 1980 1991 Old MR 23,730,895 32,126,519 39,759,370 New MR 3,236,578 5,703,896 8,360,908 Total MR 26,967,473 37,830,415 48,120,278

Growth Rates ( % per Year) 2000 1970-19801980-19911991-2000 46,657,093 3.08 1.96 1.79 10,720,429 5.83 3.54 2.80 57,377,522 3.44 2.21 1.97

Relative Participation in Brazilian Urban Population (%) Old MR 45.56 39.94 35.82 33.88 New MR 6.21 7.09 7.41 7.78 TOTAL MR 51.77 47.03 43.23 41.66 Source: FIBGE, Demographic Censuses 1980, 1991 and 2000.

17

Although New MR's concentrate a population (about 10,7 million, in 2000) lower than the Old MR's (around 46 million inhabitants), these areas are contributing, at national and state levels, for the deceleration of growth among the main national metropolises. The metropolitan area of Brasília, for instance, has been absorbing an expressive contingent of Northeast immigration, which would very likely had headed to São Paulo or Rio de Janeiro, as it happened in the past. The metropolitan regions of Campinas and Santos, within the state context, also represent areas of population deconcentration from the Metropolitan area of São Paulo to the Interior. Therefore, the understanding of the urban deconcentration and metropolitan deceleration processes are related to the conformation and distribution of populations within the Brazilian urban system.

4.2 - Distribution of the Population within the Brazilian Urban Network: the role of the nonmetropolitan agglomerations and non-primate cities

Considering the population distribution with respect to the Great Brazilian Regions, it is possible to observe the growing importance of non-metropolitan agglomerations and smaller places in the spatial reorganization of the population (Table 5). Thus, considering the metropolitan areas, urban agglomerations, mid-sized municipalities, and other locations– called here non-primate municipalities - it is noticeable that within the national context, the metropolitan areas concentrate 36.3% of the total population, the non-metropolitan urban agglomerations and the mid-sized municipalities agglutinate other 22.6%, and the other municipalities account for an expressive proportion of the total population (41.1%). These data suggest the presence of population deconcentration within the urban network, as Brazil is thoroughly urbanized, exhibiting 80% of its population dwelling in urban areas, the increasing importance of non-metropolitan areas indicate their potential for absorbing and retaining population. These new urban configurations, however, have been marked by the incorporation of expanding areas within the context of regional or local population dynamics, engendering new spatialities in the urbanization process. The nonmetropolitan urban agglomerations, for instance, contribute, on the one hand, to soften the urban growth of metropolises as they introduce a higher densification in the urban network, which, in turn, begins to play an important role in the internal redistribution process of the Brazilian population.

18

By the year 2000, there were 31 non-metropolitan urban agglomerations within Brazil, 14 of which were located in the Southeast (Table 5). The mid-sized municipal districts located outside metropolitan concentrations and agglomerations summed up 61 locations, indicating the importance of these entities, even in economically depressed areas such as the Northeast (with 12 urban centers of that order).

Table 5 Population according to the Brazilian Urban Network Morphology Great Regions – 1980/2000

Regions North MR Urban Centers Other Cities

Population 1980 1991 6623397 10030556 1021486 1401305 1456016 2536686 4145895 6092565

2000 12919949 1794981 3395356 7729612

Relativa Dis tribution (%) Growth rates (%per year 1980 1991 2000 1980-1991 1991-2000 100..00 100,00 100,00 3,85 2,85 15,42 13,97 13,89 2,92 2,79 21,98 25,29 26,28 5,18 3,29 62,59 60,74 59,83 3,56 2,68

Northeast MR UNMAG Urban Centers Other Cities

34812356 6248057 3149587 1560673 23854039

42497540 8487893 4736281 1875118 27398248

47679381 10227838 5829484 2324260 29297799

100,00 17,95 9,05 4,48 68,52

100,00 19,97 11,14 4,41 64,47

100,00 21,45 12,23 4,87 61,45

1,83 2,82 3,78 1,68 1,27

1,29 2,09 2,33 2,41 0,75

Southea MR ST UNMAG Urban Centers Other Cities

51734125 27264082 4519933 3424031 16526079

62740401 33282548 6023217 4403032 19031604

72262411 38103026 7025608 5132703 22001074

100,00 52,70 8,74 6,62 31,94

100,00 53,05 9,60 7,02 30,33

100,00 52,73 9,72 7,10 30,45

1,77 1,83 2,64 2,31 1,29

1,58 1,51 1,73 1,72 1,62

Sou th MR UNMAG Urban Centers Other Cities

19031162 4690126 2107761 1217767 11015508

22129377 6430352 2809354 1565820 11323851

25071211 8488232 3405332 1762832 11414815

100,00 24,64 11,08 6,40 57,88

100,00 29,06 12,70 7,08 51,17

100,00 33,86 13,58 7,03 45,53

1,38 2,91 2,65 2,31 0,25

1,40 3,13 2,16 1,33 0,09

Center-West MR UNMAG Urban Centers Other Cities

6801666 1337419 286227 544053 4633967

9418581 1963935 564771 885046 6004829

11611491 2935510 697340 1093816 6884825

100,00 19,66 4,21 8,00 68,13

100,00 20,85 6,00 9,40 63,76

100,00 25,28 6,01 9,42 59,29

3,00 3,55 6,37 4,52 2,38

2,35 4,57 2,37 2,38 1,53

100,00 35,12 11,35 10,10 43,43

100,00 36,30 12,00 10,64 41,05

1,93 2,21 3,42 2,84 1,18

1,61 1,99 2,24 2,21 0,98

BRAZIL 119002706 146816455 169544443 100,00 MR 40561170 51566033 61549587 34,08 UNMAG 11519524 16670309 20353120 9,68 Urban Centers 10892419 14821581 18043223 9,15 Other Citiess 56029593 63758532 69598513 47,08 Source: FIBGE, Demographic Census es 1980, 1991 and 2000; Population Count, 1996.

In different regional contexts, from metropolises to non-primate municipalities, the average rates of population growth were positive (see Table 6). This indicates the presence of a higher population retention power among various categories of municipal districts. While metropolitan Brazil grew, in the period 1991-2000, at a 2.0% rate per year; the nonmetropolitan urban agglomerations grew at 2.24%; the mid-sized municipalities at 2.16% per year; and the non-primate ones, at almost 1.00% .

19

Table 6 Distribution of Urban Centers according to the Great Regions in 2000 Regions

Metropolitan Areas

Urban Non-Metropolitan Agglomerations

North 1 Northeast 4 Southeast 6 South 5 Center-West 2 Brazil 18 Source: IPEA/IBGE/NESUR-UNICAMP (2000).

0 9 14 7 1 31

Urban Centers

11 12 23 10 5 61

Note: Those growth rates were quite diversified when considering the Great Brazilian Regions in 2000

Focusing on the North, it is noticeable that the non-primate municipalities experienced the highest rate of growth among the Great Regions (2.55% per year). This figure is higher than the metropolises of Southeast Brazil. Moura and Moreira (1998) call the attention for the emergence, in the urban system of the region, of non-metropolitan small municipalities, internalizing what has been called “new rurality”. Such places begin to have access to basic services and are now becoming service centers at local and even sub-regional level. However, the urban configuration18 of the area is marked by the influence of Belém and Manaus (NESUR-IE/UNICAMP, 1998a; Moura and Moreira, 1998), although the other state capitals are also worth mentioning, with growth rates above 3% per year for the 19912000 period. Therefore, one witnesses the incorporation of new regional spatialities: “new rurality.” In Northeast Brazil, the metropolitan areas have experienced growth rates similar to those of non-metropolitan urban agglomerations and to those of the mid-sized municipalities (around 2% per year in the 1990’s). Although, the non-primate municipalities accounted for 61,0% of the regional population in 2000, this participation declined, in favor of municipalities located within urban concentrations. This metropolitan concentration in the Northeast reflects the idiosyncrasies of the regional urbanization. The 1980’s witnessed an acceleration in the rural-urban movements, leading to a significant decrease in the local rural population (from 17,245,514 inhabitants, in 1980, to 16,721,621, in 1991, and dropping further to 15,575,505, in 1996). Part of the population that left the rural areas of the region between 1980 and 1996, was absorbed by the local urban areas, most especially by the regional metropolitan areas. Notice that the

18

The analyses concerning the urban network from each of the regions benefit from the results of the project “Characterization and Trends of the Urban Network in Brazil. IPEA/IBGE/ NESUR-IE/UNICAMP (2000).

20

growth rate of Northeast urban population was 2.5% per year between 1991 and 1996, far superior to the national average (2.1% per year). The recent economic growth of certain areas of the Northeast, especially those related to the petrochemical complex of Camaçari, the textile center of Fortaleza, the mineral-metallurgical complex of Carajás, the agroindustrial center of Petrolina/Juazeiro, the areas marked by modern grain agriculture, the tourism industry, and the export oriented fruit production increased and diversified the regional economic structure (Araújo, 1995; Pacheco, 1998), contributed to the expansion of tertiary activities and increased the population retention and attraction powers. Southeast’s metropolitan growth is the lowest among all Brazilian Great Regions. The growth rates are somewhat even across different urban categories (see Table 5), ranging from 1.51% per year to 1.73% , in spite of the predominance of non-metropolitan urban growth. A more homogeneous diffusion of the urbanization constitutes a strong characteristic of the region, a trait related to the emergence of new urban agglomerations, and the densification of urban areas, strategically located in the urban network. The historic concentration of economic activities in the Southeast, especially in São Paulo, Rio de Janeiro and part of Minas Gerais, made the Southeast the most urbanized region of Brazil (89,3% of its population lived in urban areas in 1996), encompassing the three most important national metropolises and the largest volume of urban population, 59,825,958 inhabitants in 1996. Within this context, the urban network is being redrawn, and new roles are being assigned to the old metropolises. In the Brazilian South, the growth of metropolitan municipalities (3.13% per year, in the 1991-2000 period) was higher than all the other urban categories, especially when compared with the non-primate municipalities (0.09% per year., in the same period). Its recent economic dynamism is impacting the regional urbanization, with the formation of development centers within all States. Within this context, the shoe pole in Vale dos Sinos, the mechanical industry within the metropolitan region of Porto Alegre, and in the northeast of Santa Catarina; the petrochemical complex of Triunfo, in Rio Grande do Sul; the oil refinery of Araucaria; and the industrial district of Curitiba, in Paraná are good examples of such development poles (Pacheco, 1998). The demographic recovery of the area, especially of Paraná State, and the new economic configuration had different spatial contributions, which in turn influenced the southern urban system. According to the results of a study by IPEA/IBGE/NESURIE/UNICAMP (1998:107) “the main characteristic of the southern urban network is the 21

combination of an increasing number of densely populated municipalities in urban agglomerations - spaces qualified as both population and economic concentration areas”, which, actually, diverted historical population shifts away from the regional metropolitan areas. The greater expression of the new socio-spatial dynamics of the Brazilian urbanization can be observed in the Center-West Region, with the formation of the Brasília’s hinterland19, and the process of regional metropolization. Beginning in the 1990’s, the population growth among this group of municípios represented 53.5% of the regional population, contrasting with 40,1% in the 1981-1991 period. In fact the growth rate among Center-Westerner metropolises (Brasília and Goiânia) reached 4.57% per year, between 1991-2000; the highest within the Brazilian urban network. Marked by the expansion, modernization, and urbanization of the agricultural frontier, the Center-West experienced a process of deruralization since the 1970’s, when, its urban population grew at 7.6% per year, rate superior to the ones pertaining to North Brazil for the same period (6.6% per year). Between 1980 and 1991, the urban population growth rate decreased, despite the fact that it still remained above the national average (4,3% versus 3,0% per year, respectively). This trend continued during the early 1990’s, when the Center-West presented an urban growth rate of 3,0% against the Brazilian average 2,1% per year. “This generalized movement towards the cities is characteristic of the economic dynamics of the region: the prevalence of an agricultural sector, highly capitalized and automated, focused on the production of commodities. In addition to not needing live work in a great scale, it conforms the property and the use of land according to its valuation logic, emptying the rural areas and provoking a migratory flow towards the large urban centers.” (NESUR-IE/UNICAMP, 1998:59) The growing agricultural activities, based on the grain/meat complex, polarized the tertiary and industrial sectors with important impacts on the regional urban configuration, as the Center-West registered 86% of its population living in urban places in the year 2000; second only to Southeast Brazil (90.5%).

22

5. Conclusions

Internal migration and urbanization were deeply intertwined phenomena in Brazil throughout the XX century. They have been key factors behind the process of spatial redistribution of populations and the development of a network of dense locations, in which the chief economic and social dynamics of the last 50 years and the next decades are disclosed. In fact, Brazil experienced expressive changes between 1950 and 1960, based on the increasing urbanization and on expressive rural-urban migratory movements, maintaining high expansion rhythms during the 1960’s and 1970’s, during the so-called ‘economic miracle’. However, a new stage in the process of territorial occupation began to be delineated, with the maturation of the urban-industrial development and the reduction of rural origin flows, despite the pattern of population concentration in ever growing remained important. Back then, the debate underlined the problems associated with the so-called ‘urban macrocephaly ' and the formation of ‘megalopolis ', due to the remarkable growth of metropolitan configurations during the 1970’s. Since then, studies attempting to understand the new directions of Brazilian urbanization have multiplied. The discussions about the interiorization of the economic development, demetropolization, industrial and population deconcentration from São Paulo and Rio de Janeiro, are marked by discrepant opinions. Some scholars emphasize the inducing power of public and private investment, disseminators of externalities and diverse multiplying effects, although restricted to a certain ‘agglomerative field’ within the Southeast. Others point the historic inertial effects concerning the facilities and productive investments already in place, that restrict changes in the concentrating locational pattern, barring economic-spatial deconcentration. Others still, understand that there is a concentrated-deconcentration process in course, developing among relatively well-equipped selective spaces, especially in mid-sized urban centers that have gained notoriety over the last 20 years. The debate will certainly continue in the years to come, probably gaining more consistency, as new studies arise. Therefore, several related themes that have multiplied over the last years deserve attention: a) the qualitative and quantitative importance of 19

It concerns the Integrated Region of Development of the Federal District and Entorno (Região Integrada de Desenvolvimento do Distrito Federal and Entorno) (RIDE). The Decree # 2710 of 08/04/1998 regulated the Lei

23

migratory

movements,

involving

metropolitan

areas,

non-metropolitan

urban

agglomerations and non-primate places; b) the migration returnees and their current socioeconomic characteristics, compared with the other internal migrants; c) changes in development pattern, starting from the waning of industrial deconcentration, economic and energy crisis and increasing poverty and exclusion at large metropolises; d) new possible dynamics of reconcentration of investments and urban facilities, based on high technology sectors; e) metamorphosis of the Brazilian urban network, outlining different interdependence bonds and complementarity among dense and strategically located cities. Actually, when the five Great Brazilian Regions are considered, it is noticeable that the Southeast exhibits both the smallest metropolitan growth and the densest urban network. Besides, it also represents the largest contingent of urban population in Brazil, home to new urban agglomerations, and the most important national metropolises. It is not surprising that the influence of the economic dynamism of Southeast Brazil reaches a vast areas the national territory, especially the frontier regions of the South and Center-West. In the South, the existence of a relatively old and stable urban network, with many small and mid-sized cities is a fact. However, the region has been marked by a recent expansion of its metropolitan areas, along with the consolidation of development centers in the coastal and western areas, being influenced by São Paulo State. A similar situations also occurs at a large portion of the Center-West, where new articulations give dynamism to the south and southwestern portions of Mato Grosso do Sul and Goiás States, to Brasília and its hinterland, Goiânia and Anápolis; all showing rates of urban growth well above the national averages. The North Region, although more distant from the more dynamic economic axis, also exhibits an expressive urbanization expansion, being characterized as a migrant pull region, due of the presence of large regional centers, Belém and Manaus, state capitals and other urban areas that are being restructured. Finally, with regards to the Northeast Region, a traditional population evasion area, it is important to observe that their metropolises and non-metropolitan urban agglomerations have also been experiencing growth rates above the national average, probably reflecting the emptying of their rural areas, the dynamics of regional urbanization itself in well determined economic spaces, and even return flows to the area. Based on this discussion, a question, for which there are no answers yet, arises: Will the productive restructuring, which has privileged dynamic and traditional urban spaces Complementar # 94 and authorized the Executive Power to create the Region (CODEPLAN/GDF).

24

culminate in industrial reconcentration? It does not sound plausible to bet on a reconcentration of the population, similar to the one occurring between 1930 and 1970. After all, the growth rates among national metropolitan population decreased substantially during the 1980’s and 1990’s. Recent census data indicate the maintenance of the metropolitan growth decrease trend, especially in São Paulo and Rio de Janeiro. Moreover, census data also indicate that the process of population deconcentration seems far from exhaustion. In fact, it is likely that the Brazilian population will continue deconcentrating during the next decades, despite the erratic rhythms and intensities of population growth, the economic geography of the places, economic crisis, and medium and long term macroeconomic dynamics.

25

References AMARANTE, L. A., BONDIOLI, P. A apropriação regional da renda nacional no Brasil; 1975-85. São Paulo em Perspectiva, revista da fundação SEADE, São Paulo, v.1, n.3, out/dez, 1987. ARAÚJO, T.M.B. (1995). Nordeste, Nordeste: que Nordeste? In: Affonso, R. e Barros Silva, P. L. (orgs). Desigualdades Regionais e Desenvolvimento (Federalismo no Brasil). São Paulo: FUNDAP/ ED. UNESP, p. 125-156. AZZONI, C. Indústria e Reversão da Polarização no Brasil. Ensaios Econômicos- IPE/USP, São Paulo, n.58, 1986. BAENINGER, R. (1999). Região, Metrópole e Interior: Espaços Ganhadores e Espaços Perdedores nas Migrações Recentes no Brasil – 1980/1996. Doctoral Thesis. IFCH/UNICAMP, Campinas. BALÁN, J. Migraciones e desarrollo capitalista en el Brasil: ensaio de interpretación histórico-comparativa. Migración y Desarrollo, CLACSO, 1974. BENKO, G. e LIPIETZ, A. O Novo Debate Regional: posições em confronto. In: Benko, G. e Lipietz, A (org.). As Regiões Ganhadoras – Distritos e Redes: os novos paradigmas da geografia econômica. Celta Editora, Oeiras, 1994 (p.3-15). BENKO, G. Economia, Espaço e Globalização: na aurora do século XXI. Editora Hucitec, São Paulo, 1996. BILSBORROW, R. E. (org.). Migration, urbanization and development: new directions and issues. New York : UNFPA/Kluwer, 1996. BRITO, F. População, espaço e economia numa perspectiva histórica: o caso brasileiro. Belo Horizonte : CEDEPLAR/UFMG-Faculdade de Ciências Econômicas, 1997. (Doctoral Thesis) CAMARANO, A.A. (1998). Movimentos Migratórios Recentes na Região Nordeste. . In: Encontro Nacional sobre Migração, Curitiba, Anais..., ABEP/IPARDES, p. 189-208. CANO, W. (1988), coord, A Interiorização do Desenvolvimento Econômico no Estado de São Paulo (1920-1980). São Paulo: Fundação SEADE, v.1. a v. 3 (Coleção Economia Paulista). CANO, W. Algumas implicações espaciais da Terceira Revolução Industrial no Brasil. In: GONÇALVES, M. F. (org). O novo Brasil urbano. Porto Alegre : Editora Mercado Aberto, 1995a. CANO, W. Auge e inflexão da desconcentração econômica regional. In: AFFONSO, R, SILVA, P. L. B. A federação em perspectiva: ensaios selecionados. São Paulo: Fundap, 1995b. p.399-416. CANO, W. Migrações, desenvolvimento e crise no Brasil. Campinas : Instituto de Economia/UNICAMP, 1996. (mimeo) CANO, W. Raízes da Concentração Industrial em São Paulo, 2 ed. São Paulo: TA Queiroz, 1977. CARVALHO, J. A. M., RIGOTTI, J. I. Análise das metodologias de mensuração das migrações. In: ENCONTRO NACIONAL SOBRE MIGRAÇÃO, 1998, Curitiba. Anais ... Curitiba: IPARDES/ABEP, 1998. CASTELLS, M. The Informational City, Information Technology, economic restructuring and the urbanregional process. Oxford, Basil Blackwell, 1989. CELADE (Centro Latinoamericano de Demografia). Dinâmica de la población en las grandes ciudades en América Latina y Caribe. Documentos Docentes, Santiago de Chile, 1994. COUTINHO, L. ‘O desenvolvimento urbano no contexto da mudança tecnológica”. In: GONÇALVES, M.F. (Org.). O novo Brasil urbano. Mercado Aberto, Porto Alegre, 1995, pp.: 41-62. CUNHA, J. M. P. (coord.). Projeto “Mobilidade e redistribuição espacial da população no Estado de São Paulo: características recentes, padrões e impactos no processo de urbanização”. Campinas : NEPO/UNICAMP, 1999. 273p. (Final Report) CUNHA, J.M.. (Des) continuidades no padrão demográfico do fluxo São Paulo/Bahia no período 1990/1991: qual o efeito da crise. In: ENCONTRO NACIONAL DE ESTUDOS DE POPULACIONAIS, 11., 1998, Caxambu. Anais ... Caxambu : ABEP, 1998. DEDDECA, C.S., BRANDÃO, S.M.C. Crise, transformações estruturais e mercado de trabalho. Campinas : CEDE/UNICAMP, 1992. (mimeo) DINIZ e CROCCO, M. Reestructuración Económica e Impacto Regional: el nuevo mapa de la indústria brasileña. In: Mattos, C.; Nicolás e Botero, D. Globalización y Territótio – impactos e perspectivas. Fondo de Cultura Económica. Chile, 1998. DINIZ, C.C. Desenvolvimento poligonal no Brasil; nem desconcentração, nem contínua polarização. Nova Economia - Revista do Departamento de Ciências Econômicas da UFMG, vol 31, n.11, Belo Horizonte, 1993. DINIZ, C.C. Dinâmica regional recente e suas perspectivas. In: AFFONSO, R., SILVA, P. L. B. A federação em perspectiva: ensaios selecionados. São Paulo: Fundap, 1995. p.417-429. EBANKS, E.G. Determinantes Socioeconómicos de la Migración Interna. CELADE, Santiago de Chile, 1993.

26

FARIA, V. Cinqüenta anos de urbanização no Brasil: tendências e perspectivas. Novos Estudos CEBRAP 29, São Paulo, p.98-119, mar.1991. FARIA, V. Divisão Social do trabalho, espacialização e crescimento urbano: o caso da macro-metrópole Paulista. In: Anais do II Encontro Nacional de Estudos Populacionais, São Paulo: ABEP, 1980. FARIA, V. O processo de urbanização no Brasil: algumas notas para seu estudo e interpretação. In: Anais do I Encontro Nacional de Estudos Populacionais. São Paulo: ABEP, 1978. FARIA,V. Desenvolvimento, Urbanizaçãoe Mudanças na Estrutura do Emprego: a experiência brasileira dos últimos 30 anos. In: Sorj , B. e Almeida, M.H. Sociedade e Política no Brasil Pós-64. São Paulo: Editora Brasiliense, 1983. FUNDAÇÃO IBGE. Contagem Populacional de 1996. Rio de Janeiro, 1997. FUNDAÇÃO IBGE. Pesquisa Nacional por Amostra de Domicílio. Rio de Janeiro, PNAD, 1995. FUNDAÇÃO IBGE/DPE/DPIS/DICAD/DEMOG. Movimentos migratórios segundo o Censo Demográfico de 1991 e Contagem da População de 1996. Rio de Janeiro : IBGE, 1997. (mimeo) GOTTDIENER, M A Produção Social do Espaço. EDUSP, São Paulo, 1993. GOTTDIENER, M. A teoria da crise e a reestruturação sócio-espacial: o caso dos Estados Unidos. In: Valladares, L. e Preteceille, E. (coord.). Reestruturação Urbana: tendências e desafios. Nobel/IUPERJ, São Paulo, 1990. (p.59-78) GRAHAM, D. e BUARQUE DE HOLLANDA , F.S. Migration, regional and urban growth and development in Brazil: a selective analyses of the historical record, 1972-1970. São Paulo, IPE/USP, 1971. GREENWOOD, M. Migrações internas nos Estados Unidos; uma revisão da literatura. In: MOURA, A. M. (Coord). Migração interna; textos selecionados. Fortaleza: Banco do Nordeste do Brasil, 1980, p. 733-77. 1980. GUIMARÃES, E. e LEME, H. (1997). Caracterização histórica e configuração espacial da estrutura produtiva do Centro Oeste. In: NEPO/UNICAMP. TEXTOS NEPO 33, Campinas. HADDAD, P. Economia Regional; teorias e métodos de análise, Recife, Banco do Nordeste do lBrasil, 1980. HARVEY, D. Condição Pós-Moderna. Editora Loyola, São Paulo, 1992. IPEA/IBGE/NESUR-UNICAMP. Características e Tendências da Rede Urbana no Brasil. Instituto de Economia-UNICAMP, 2000. KATZMAN, M. T. Urbanização no Brasil a partir de 1945. In: BACHA, E., KLEIN, H. S. A transição incompleta; Brasil desde 1945. Rio de Janeiro: Paz e Terra, 1986, p.177-229. LATTES, A. E. Population distribution in Latin America: is there a trend towards population deconcentration? In: Population, distribution and migration. New York : United Nations, 1998. LOPES, J.R. e PATARRA, N.L. Redistribuição regional e rural-urbana da população brasileira. Estudos sobre a população brasileira. Cadernos CEBRAP, 20, São Paulo, 1975. LOPES, J.R.B. Desenvolvimento e migração: uma abordagem histórico-estrutural. Estudos CEBRAP, São Paulo, n.6, 1973. MAGALHÃES, M. Movimentos migratórios na Região Sul: novas tendências. In: ENCONTRO NACIONAL SOBRE MIGRAÇÃO, 1998, Curitiba. Anais ... Curitiba : IPARDES/ABEP, 1998. MARTINE, G. A evolução espacial da população brasileira. In:AFFONSO, R. B. A., SILVA, P. L. B. (orgs.). Desigualdades regionais e desenvolvimento (Federalismo no Brasil). São Paulo: FUNDAP/Ed. UNESP, 1995. p.61-91. MARTINE, G. As migrações de origem rural no Brasil: uma perspectiva histórica. História e População – Estudos sobre a América Latina. ABEP/IUSSP/CELADE. São Paulo, 1990. MARTINE, G. e CAMARGO, L. Crescimento e distribuição da população brasileira: tendências recentes. Revista Brasileira de Estudos de População, Campinas, ABEP, 1(2), p.99-143, jan./dez.1984. MARTINE, G. e PELIANO, J.C. Migrantes no mercado de trabalho metropolitano. Série Estudos para o Planejamento, Brasília, Ipea, 1976. MARTINE, G. Migração e Metropolização. Revista São Paulo em Perspectiva, São Paulo, Fundação SEADE, 1987. MARTINE, G., CAMARANO, A.A., NEUPERT, R. e BELTRÃO, K. A urbanização no Brasil: retrospectiva, componentes e perspectivas. In: ENCONTRO NACIONAL DE ESTUDOS POPULACIONAIS 6, 1988, Olinda. Anais... Brasília: ABEP, 1988. MARTINE, G.. A redistribuição espacial da população brasileira durante a década de 80. Textos para Discussão 329, Brasília, IPEA, 1994. 43p. MARTINE,G. Processos recentes de concentração e desconcentração urbana no Brasil: determinantes e implicações. Documento de Trabalho 11, Brasília, Instituto SPN, 1992. 29p.

27

MATOS, C. de “Capital, Población y Territorio”. Seminário Distribución y Movilidad Territorial de la Población y Desarrollo Humano. Fundación Bariloche/CENEP/PROLAP. Bariloche, Argentina, maio, 1994. MATOS, R.E.S, Dinâmica migratória e desconcentração da população na macrorregião de Belo Horizonte. Belo Horizonte. CEDEPLAR (tese de doutorado), 1995. MATOS, R.E.S, “Questões teóricas acerca dos processos de concentração e desconcentração da população no espaço”. In: Revista Brasileira de Estudos Populacionais. São Paulo, 1995, p. 35-58. MERRICK, T. e GRAHAM, D. Population and economic developmente in Brazil, 1800 to the present. Baltimore, Johns Hopkins University Press, 1979. MOTTA, D. M. e AJARA, C. “Rede urbana brasileira-hierarquia das cidades”. Curso de Gestão Urbana e de Cidade EG/FJP-WBI-PBH-ESAF-IPEA, ago, 1999 MOURA, H. M. e MOREIRA, M.M. (1998). As Migrações na Região Norte em Período Recente: uma abordagem preliminar. In: Encontro Nacional sobre Migração, Curitiba, Anais..., ABEP/IPARDES, p. 138-188. NEGRI, B. Concentração e desconcentração industrial em São Paulo (1880-1990). Campinas: Ed. UNICAMP, 1996. NEGRI, B. e PACHECO, C. A . Mudança Tecnológica e Desenvolvimento regional nos Anos 90: da interiorização do desenvolvimento à nova dimensão espacial da indústria paulista (Relatório Final). Projeto Desenvolvimento Tecnológico e Competitividade da Indústria Brasileira. SCTDE/FECAMP/UNICAMP/IE, Campinas, 1993. NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 a). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: Norte. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 b). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: Nordeste. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 c). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: Sudeste. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 d). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: São Paulo. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 e). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: Sul. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL – NESUR-IE/UNICAMP (1998 f). Caracterização e Tendências da Rede Urbana no Brasil – Estudo Regional: Centro Oeste. Relatório 6. Campinas, outubro (mimeo). NÚCLEO DE ECONOMIA SOCIAL, URBANA E REGIONAL-(NESUR/UNICAMP). Projeto “Caracterização e tendências da rede urbana brasileira”. Campinas, out.1998. (Relatório 6) PACHECO, C. (1993). Evolução Recente da Urbanização e da Questão regional no Brasil: implicações econômicas para a dinâmica demográfica. Anais da IV Conferencia Latinoamericana de Población. Ciudad de México. PACHECO, C. A . Fragmentação da nação. Campinas : Instituto de Economia/UNICAMP, 1998. PACHECO, C. e PATARRA, N. L. Movimentos migratórios anos 80: novos padrões? In: ENCONTRO NACIONAL SOBRE MIGRAÇÃO, 1998. Anais ... Curitiba : ABEP/IPARDES, 1998. PACHECO, C.A. e PATARRA, N. L. Tendências da Urbanização e do Crescimento Populacional Brasileiro: População em Idade Escolar. Instituto de Economia/UNICAMP, 2000. PACHECO,C.. Os Aglomerados Urbanos de São Paulo: as novas realidades demográficas e ocupacionais da urbanização paulista. Relatório de Pesquisa. Terciarização e Precarização da Estrutura Ocupacional nos Anos 80. Projeto Urbanização e Metropolização no Estado de São Paulo. SPG/FECAMP,1992. PACHECO,C. A. Fragmentação da Nação. Instituto de Economia/UNICAMP, 1998. PANIZZI, W. Prefácio do Livro O Novo Brasil Urbano. Editora Mercado Aberto, Porto Alegre,1995. PATARRA, N.L. et al . Migrações, Dinâmica Urbana e Condições de Vida. Instituto de EconomiaUNICAMP, 1997. PIORE, M. e SABEL. C. The Second Industrial Divide. Nova York, Basic Books, 1984. PORCHMANN, Marcio. “São Paulo: a capital e o trabalho”. Revista República, São Paulo, n. 52, fev. 2001 PRETECEILLE, E. e VALLADARES, L. Introdução. In: Valladares, L. e Preteceille, E. (coord.). Reestruturação Urbana: tendências e desafios. Nobel/IUPERJ, São Paulo, 1990. (p.7-13)

28

REDWOOD III, J.. Reversion de polarizacion, ciudades secundarias y eficiencia en el desarrollo nacional: una vison teorica aplicada al Brsil contemoraneo. Revista Latinoamericana de Estudios Urbanos Regionales, Santiago, vol 11, n.32, dez. 1984 RICHARDSON, H. Economia Regional, Teoria da Localização, Estrutura Urbana e Crescimento Regional. Zahar Editores, Rio de Janeiro, 1975. RIGOTTI, J. I. R , CARVALHO, J. A. M. As migrações na grande Região Centro-Leste. In: ENCONTRO NACIONAL SOBRE MIGRAÇÃO, 1997, Curitiba. Anais ... Curitiba : ABEP/IPARDES, 1997. p.67-90. SANTOS, M. A metrópole: modernização, involução e segmentação. In: Valladares, L. e Preteceille, E. (coord.). Reestruturação Urbana: tendências e desafios. Nobel/IUPERJ, São Paulo, 1990. (p.183-191) SANTOS, M. et al. Território, Globalização e Fragmentação. Hucitec: São Paulo,1994. SANTOS, M. Técnica, Espaço, Tempo: Globalização e Meio Técnico-Científico Informacional. Editora Hucitec, São Paulo, 3 ª edição, 1997. SASSEN, S The Mobility of Labor and Capital. Cambridge University Press, Cambridge, 1988 SINGER, P. Economia Política da Urbanização. São Paulo: HUCITEC, 1973. STORPER, M. A industrialização e a questão regional no Terceiro Mundo. . In: Valladares, L. e Preteceille, E. (coord.). Reestruturação Urbana: tendências e desafios. Nobel/IUPERJ, São Paulo, 1990. (p.120-147) TASCHNER, S. P., BOGUS, L. M. M. Mobilidade espacial da população brasileira: aspectos e tendências. Revista Brasileira de Estudos de População, São Paulo, 3(2), jul./dez.1986. TOWNROE, P.M and KEEN, D.. Polarization reversal in the state of São Paulo, Brazil. REGIONAL STUDIES; Journal of the Regional Studies Association, Cambridge. V. 18, n. 1, Feb, 1984. VALLADARES, L. and PRETECEILLE, E. (coord.). Reestruturação Urbana: tendências e desafios. Nobel/IUPERJ, São Paulo, 1990. VÉRAS, M.P. Entre o Local e o Global: políticas urbanas, espaços e sujeitos da cidade mundial. Trabalho apresentado na XX Encontro Anual da ANPOCS, Caxambu, outubro, 1996. VILLA, M., RODRIGUEZ, J. Dinámica sociodemografica de las metrópolis latinoamericanas. In: Grandes ciudades de América Latina: dos capítulos. Documentos Docentes, Santiago de Chile, Centro Latinoamericano de Demografia-CELADE, Naciones Unidas-FNUAP, 1994.

29