1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE

BRİCK BY BRİCK Characterization of the Household Infrastructure Conditions of Metropolitan Area Municipalities over the two last census and its possible correlation with intra-metropolitan migration

TIAGO AUGUSTO DA CUNHA Doutorando em Demografia [email protected] Núcleo de Estudos de População (NEPO) Instituto de Filosofia e Ciências Humanas (IFCH) Universidade Estadual de Campinas (UNICAMP)

ABSTRACT This study takes, as a background, the construction and use of synthetic indicators of different dimensions of the built and infrastructure environment for comparative and analytic purposes, characterizing, as a case of study, these two dimensions of Campinas Metropolitan Area (CMA) municipalities members. Several studies about the migration issue, especially those about the intra-metropolitan flows, indicate two major reasons for the act of migrating, they are: a) the labor market and b) the land and housing market. It´s over this second reason that this study will look more carefully. One of the first assumptions is that improvements in access to land / housing, as well as the increase of the intrinsic quality of the infrastructure of the metropolitan land (proximity to the relevant public facilities and equipaments, water supply, garbage collection and sewage, electricity, etc.), can be correlated with the direction of population flows to certain locations. In fact, the assumption adopted here is that services and infrastructure opportunities "direct", even indirectly, population flows from certain municipalities to another ones, given that they [these opportunities] have repercussions on the land and housing prices.

1 Cities, nations and regions in planning history

INTRODUCTION

The real reasons/motives of intrametropolitan migration have been an important subject to several demographic and urban studies. The literature about the theme is more than consensual in pointing out at least two of them: a) the labor (and job opportunities) market, and b) the land and residential/housing market (CUNHA, 1994; RIGOTTI, RODRIGUES, 1994; JAKOB, 2003). The second main reason, here commented, will be the scope of the present article. However, the theoretical framework used in order to portray and discuss the processes and reasons behind migration intrametropolitan differs greatly. In this sense, the perspective here adopted - as suggested by Gottdiener (1990) - is the historical-structuralist one. Where public power is a central agent and catalyst of the inherent urban dynamics, either through a policy of strong performance or its opposite, namely, the pure and simple failure of them. The State heavy hand affects, even in an indirect way, the price of two elements: a) initially, the land and b) therefore the housing price. Perhaps two of the main reasons - or "The" primary reason (since they are inseparable) - of migration especially in a more circumscribed territorial clippings and whose road network is more strongly connected - a fact that would allow to have a residential unit in a particular locality and job opportunities in another one -as the metropolitan areas are. It is clear, that the adoption of indexes concerning about household infrastructure features is just one of many factors that somehow represent the land and housing price. And even he [price] is also one of the innumerable factors that can direct certain population groups to other locations inside the CMA (Campinas Metropolitan Area). Therefore, we do not want to take "the part for the whole". In fact, there is a vast path between the two extremities, availability and access to infrastructure and services on the one hand and migration flows on the other, however, both have their "contact surfaces".

1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE

"FORMAL" VERSUS INFORMAL LAND MARKET After all, the ownership matters

A first attempt to characterize CMA housing conditions shall consider the key problematic, in this case, established between formal and informal land and housing market. It is a first effort to expound the subject also based on an overview of the marketization of urban land process, specifically the metropolitan one, since it is perhaps the one that most routinely "experience" this constant harassment. Such distinction is extremely relevant, once the legal right to possession of land and/or domicile opens endless opportunities (but also constraints) to their owners. We could think in some of them: various and diverses opportunities of funds, access to basic infrastructure - water supply, sewage, solid waste and garbage collection, etc.. It's more than agreed that the urban space is a territory of land disputes and so for locations as defined by Villaça (2000). In this sense, some areas or zones are more appropriate by population specific strata (mainly according to the income variable). Studies such as Caiado and Pires (2006) or Cunha et alli. (2006) try to explain this clumsy segregation logic. In other words, the land/housing market was one of the occupation engines of certain areas, especially the peripheral ones. However, could it also be one of the main drivers that led to population "uprooting"/displacement from certain areas to anothers? Therefore, the assumption here adopted is that the land and house value according to their condition/situation - regular/irregular, formal/informal - would be a great incentive or [financial] constraint to intrametropolitan migration, i.e., this same residential and land valuing process was one of various population expulsion factors from certain places, but also in a pari passu way, it was responsible to direct this same population to others metropolitan zones. It is intrinsic to this point of view a kind of migratory selectivity, where the value of land is a cornerstone for understanding migration. Table 1 attempts to capture the CMA urban land regularization situation.

1 Cities, nations and regions in planning history

Table 1 – Regularization Situation – Urban Population, Campinas Metropolitan Area, 2007 Which is the regularization situation of your neighborhood? Situation N* Exp.. % Do not know 107 28126 1.8 Formal 3482 1375115 87.9 Informal 402 59138 3.8 Regulation in process 413 89310 5.7 Did not answer 35 12781 0.8 Total 4.439 1.564.471 100.0 Missing 1.463 570.055 24.8 Source: “Pesquisa Domiciliar do Projeto Vulnerabilidade” (FAPESP/CNPq). NEPO/UNICAMP, 2007. (*) Non expanded values. (.) Expanded values.

In general, most of the settlements of the region are properly legalized (approximately 88% of them). However, it is no insignificant the percentage of non legalized settlements, since 10% of them are in this situation, a cipher that worthy attention. When the absolute values are analyzed, we have the real dimension of the issue [and the liability generated during the last decades]. In this way, actually, approximately 150,000 individuals (148,228, to be exact) reside in an allotment under the "irregularity" branded. Probably there is significant internal differences among its members, since their realities, including habitation ones, greatly differ between each other. That is, even composing an administrative unity, their economic, social, historical, etc., characteristics are quite different (TABLE 2).

1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE

Table 2 – Regularization Situation according Municipalities – Urban Population, Campinas Metropolitan Area, 2007

Which is the regularization situation of your neighborhood? Municipalities Do not Regulation Did not Formal Informal Total know in process answer Americana 1.6% 98.4% .0% .0% .0% 100.0%

Artur Nogueira .0% 86.1% .0% 13.9% .0% 100.0%

Campinas 2.5% 81.3% 5.8% 9.0% 1.3% 100.0%

Cosmópolis .0% 100.0% .0% .0% .0% 100.0%

Hortolândia .2% 76.2% 13.8% 7.1% 2.7% 100.0%

Indaiatuba .0% 100.0% .0% .0% .0% 100.0%

Itatiba 5.2% 94.8% .0% .0% .0% 100.0%

Monte Mor 14.3% 85.7% .0% .0% .0% 100.0%

Nova Odessa .0% 100.0% .0% .0% .0% 100.0%

Paulínia .0% 100.0% .0% .0% .0% 100.0%

Pedreira .0% 93.1% .0% 6.9% .0% 100.0%

Santa Bárbara 1.0% 83.3% 2.6% 13.1% .0% 100.0% d'Oeste Sumaré .9% 96.3% .6% 1.8% .5% 100.0%

Valinhos .0% 100.0% .0% .0% .0% 100.0%

Exp. 28127 1375115 59137 89311 12782 1564472 Total N. 107 3482 402 413 35 4439 % 1.8% 87.9% 3.8% 5.7% .8% 100.0% Source: “Pesquisa Domiciliar do Projeto Vulnerabilidade” (FAPESP/CNPq). NEPO/UNICAMP, 2007. (*) Non expanded values. (.) Expanded values. - the emphasis is mine.

Based on the data displayed, it is interesting to note the situation of at least two municipalities, namely: a) Campinas and b) Hortolândia, given that both are those with the lowest percentage of legalized settlements of CMA. While both have characteristics and experience vastly different processes, at the same time they are articulated through numerous others relations (including population ones - migration and commuting, for example). On the one side, Campinas, the region core, seen by the environment (and beyond its suburbs/peripheries) as the "land of opportunities," since it holds a vast industrial park, as well as a strong trade and services sector. On the other, Hortolândia; city that was the target/destiny of intense and several migratory flows, mostly of them from its big brother, Campinas. There is, at least in this case and initially, a concentration process of informal settlements in Campinas doing front an extremely speculative land market, i.e., the occupation becomes a possible solution particularly in locations near work areas. That is, the occupation as an instrument of coping. But there is also another side, where the government interference is, at least, conniving, or worse, openly allows the construction of “uncertainty” neighborhoods. It seems that this is the case of large sectors of Hortolândia.

3 Cities, nations and regions in planning history

The data give evidence, helping to corroborate views more than discussed and agreed about spurious processes of segregation and formation of metropolitan suburbs and informal settlements (BONDUKI; ROLNIK, 1982, MARICATO, 2000). However, segregation is not a process restricted to Campinas or to its principal population redistribution axis. It also takes place in and toward Santa Bárbara D'Oeste, , or even in Arthur Nogueira, municipalities considered as others relevant axes of urban expansion and population redistribution of the region.

ABOUT CMA HOUSEHOLDS OVER 1991 AND 2000 CENSUS After all, a roof matters

To construct the indicator, makeshift and collectives households were excluded from the calculation, since both together represent just over 2% of all CMA households in 1991. The focus, in this regard, will be given mainly on the permanent private households, since besides they are immensely more numerically representative, they are also more representative according environmental and infraestructure conditions that here we are trying to minimally learn about. In summary, describing the permanent households we are apprehending almost all the reality of the region. Moreover, once they are not representative, the not exclusion of collective and improvised households from the proposed selection could generate a significant number of "missing", because most of the questions wouldn´t be applied to them, this fact could lead to distortions/biases in the indicators. From this criterion, according to 1991 Census, 472,701 permanent private households were selected. The definitions about household in 2000 have hardly changed compared to 1991 Census. In this Census [2000], 659,009 units were built. This is a considerable value, given that 186,308 new units were created, i.e., an increase of 39.4% in approximately 10 years. Another selection will be made here, only the urban households, for both periods of time (1991-2000), will be taken into account. This is due, initially, to the fact that is in this environment [urban territory] that most of the processes (and the mechanisms that promote them) that we minimally aim to look in fact occur with greater force. Therefore, for 1991 Census, 451,919 permanent private urban households were in fact analyzed. In 2000, 641,510 is the total amount (an increase of approximately 42%). But, what are the main residential CMA "hotspots" (TABLE 3)?

1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE

Table 3 – Private Permanent Urban Households Variation according CMA Municipalities – Absolute and Relative Values – Private Permanent Urban Households, Campinas Metropolitan Area, 1991- 2000

Private Permanent Urban Households Year New Households Municipalities 1991 2000 Variation (Absolute Values) (Relative Values - %) (Relative Values - %) Americana 39896 52364 31.3%

Artur Nogueira 3405 8272 142.9%

Campinas 218028 279076 28.0%

Cosmópolis 8246 11886 44.1%

Engenheiro Coelho  1828 -

Holambra  1082 -

Hortolândia  40370 -

Indaiatuba 22884 39792 73.9%

Itatiba 13350 18182 36.2%

Jaguariúna 4634 6994 50.9%

Monte Mor 5439 9043 66.3%

Nova Odessa 7892 11520 46.0%

Paulínia 7981 13628 70.8%

Pedreira 6773 9381 38.5%

Santa Bárbara 34238 46318 35.3% D´Oeste Santo Antonio de 2857 4057 42.0% Posse Sumaré 52883 52961 0.1%

Valinhos 15307 22237 45.3%

Vinhedo 8106 12519 54.4%

Exp. 451919 641510 42.0% Total N.* 45533 64213

% 100 100

Source: 1991 and 2000 Census. Own elaboration. (*) Non expanded values. (.) Expanded values. - the emphasis is mine.

The data are stunning. Almost all municipalities have increased their number of urban households far above those of the region (42%). The only exception was just the CMA most significant cities triad (at least in economy and demographic terms): Campinas, Sumaré and Americana. Their growth rates were much lower than those presented by the region. It should be considered, however, that they are cities with the most significant absolute values, unlike what happens with the municipalities whose increase was indeed remarkable. For example, it is necessary to relativize the fact that Artur Nogueira has more than doubled his total permanent urban households. However, its positive balance (in absolute terms) is much smaller than those presented by Americana or Campinas. Situation diametrically opposed is that experienced by Sumaré, as well as by its "youngest daughter": Hortolândia. From a superficial and somewhat decontextualized analysis, the data suggest that the creation of new urban households in Sumaré virtually stagnated. This is not, however, necessarily the truth. In 1991, the region that would become Hortolândia still belonged to

5 Cities, nations and regions in planning history

Sumaré, but in 2000, the municipalities were already separated. In this sense, Sumaré lost much of its urban permanent households stock to Hortolândia. So, Sumaré urban permanent household stock variation may have been much higher than that one presented. Thus, for a more precise analysis of the household situation in both cities, they deserve to be reassembled. In this regard, we will report here not only to Sumaré or just to Hortolândia, but to the region composed by Sumaré+Hortolândia. From this artifice the region Sumaré-Hortolândia is the only one of those here said as the most significant one (which still include Americana and Campinas) which has a remarkable growth rate (range of 76.49%) of permanent urban households. A clear vector of urban expansion and population redistribution, in fact, the most important one considering their absolute values. Figure 1 shows the same variation of permanent urban households according CMA municipalities over the decade in question.

Figure 1 – Private Permanent Urban Households Variation according CMA Municipalities – Absolute and Relative Values – Private Permanent Urban Households, Campinas Metropolitan Area, 1991-2000. Source: 1991 and 2000 Census. Own elaboration.

It is possible to conclude that today the municipalities bordering the CMA core (Campinas) are those who most increased its housing stock. We can infer from this finding those municipalities who most intensely grow (specially in population terms), so those who are most suffering speculative harassment. If this is indeed becoming a reality they are also the ones who are being target of several population flows, since it is very unlikely that this growth is only caused by natural/vegetative growth rates. Finally, it is interesting to note that the relative and absolute variations of permanent urban households were more intense precisely in those municipalities or vectors whose proportions of informal or in regularization 1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE process settlements are also more expressive, i.e., Sumaré-Hortolândia, Monte Mor and the North vector, culminating in Artur Nogueira and .

THE CONSTRUCTION OF A HOUSEHOLD ENVIRONMENT INFRASTRUCTURE INDEX After all, living well matters

The construction of an index and its adoption in the present study does not necessarily intend to draw up an unusual synthetic indicator, but, at least in this case, to use it to investigate possible correlations with CMA urban expansion and population redistribution. In this sense, the index here proposed is in fact very simple, trying to capture different dimensions of the urban permanent households environment, such as: water, sewage, garbage and electricity supply, essentially (TABLE 4).

1

1 Where: i IED = Índice de Infra-Estrutura do Entorno Domiciliar (Household Environment Infrastructure Index); RGA = Proporção de Domicílios Urbanos Permanentes atendidos por Rede Geral de Água (Proportion of Urban Permanent Households served by Water Supply Network); RE = Proporção de Domicílios Urbanos Permanentes atendidos por Rede de Esgoto (Proportion of Urban Permanent Households served by Sewerage); CL = Proporção de Domicílios Urbanos Permanentes servidos por Coleta de Lixo (Proportion of Urban Permanent Households served by Garbage collection); REl = Proporção de Domicílios Urbanos atendidos por Rede de Energia Elétrica (Proportion of Urban Permanent Households served by Electricity Supply Network).

7 Cities, nations and regions in planning history

Table 4 – Household Environment Infrastructure Index according Municipalities – Private Permanent Urban Households, Campinas Metropolitan Area, 1991-2000

Household Environment Infrastructure Index 1991 2000 iIED Municipalities RGA RE DE REl RGA RE DE REl Variation i IED i IED (%) (%) (%) (%) (%) (%) (%) (%) (% 1991- 2000) Americana 0.9660 0.9324 0.9707 0.9912 0.9651 0.9789 0.9403 0.9963 0.9996 0.9788 1.417% Artur 0.9953 0.9645 0.9959 1.0000 0.9889 0.9790 0.9780 0.9924 0.9975 0.9867 -0.369% Nogueira Campinas 0.9513 0.8533 0.9699 0.9949 0.9423 0.9733 0.8680 0.9890 0.9991 0.9573 1.595% Cosmópolis 0.9388 0.7122 0.9580 0.9945 0.9009 0.9870 0.9207 0.9950 0.9976 0.9751 8.239% Engenheiro - - - - - 0.9639 0.9687 0.9869 0.9962 0.9789 - Coelho  - - - - - 0.8780 0.8318 0.9806 1.0000 0.9226 - Hortolândia - - - - - 0.9727 0.0436 0.9948 0.9998 0.7527 - Indaiatuba 0.9851 0.8957 0.9865 0.9993 0.9667 0.9419 0.9194 0.9876 1.0000 0.9622 -0.458% Itatiba 0.9320 0.8942 0.9545 0.9984 0.9448 0.9911 0.9640 0.9976 1.0000 0.9882 4.594% Jaguariúna 0.9851 0.9198 0.9726 1.0000 0.9694 0.9794 0.9578 0.9939 1.0000 0.9828 0.545% Monte Mor 0.8768 0.4258 0.8651 0.9770 0.7862 0.9679 0.4512 0.9863 0.9978 0.8508 8.218% Nova Odessa 0.9743 0.9624 0.9811 1.0000 0.9794 0.9817 0.9786 0.9990 1.0000 0.9898 1.061% Paulínia 0.9414 0.8351 0.9545 0.9956 0.9316 0.9660 0.8571 0.9836 1.0000 0.9517 2.148% Pedreira 0.9579 0.9188 0.9489 1.0000 0.9564 0.9849 0.9635 0.9945 1.0000 0.9857 3.065% Santa Bárbara 0.9803 0.9742 0.9776 0.9964 0.9821 0.9936 0.9779 0.9925 0.9983 0.9906 0.859% d'Oeste Santo Antônio de 0.9328 0.0290 0.9279 0.9993 0.7223 0.9778 0.1903 0.9808 0.9936 0.7856 8.773% Posse Sumaré 0.9043 0.3137 0.9519 0.9932 0.7908 0.9759 0.7871 0.9867 0.9975 0.9368 8.400% Valinhos 0.8769 0.8422 0.9119 0.9976 0.9072 0.8963 0.8627 0.9833 1.0000 0.9356 3.134% 0.9076 0.7898 0.9204 0.9968 0.9036 0.9493 0.7676 0.9837 1.0000 0.9251 2.378% Total* 0.9469 0.7993 0.9641 0.9950 0.9263 0.9711 0.8238 0.9901 0.9991 0.9460 2.129% Source: 1991 and 2000 Census. Own elaboration.  Municipalities that did not exist until 1991. In fact, all of them were created in 1993. (*) There is no “missings”.

There are significant improvements in the infrastructure conditions and in so many services access in cities such as Santo Antonio de Posse, Cosmópolis, Monte Mor, Sumaré (all of them with a range of approximately 8%) and Itatiba (with 4.6%). Contrary to this progressive urban infrastructure qualification process are the municipalities of: Artur Nogueira and Indaiatuba (with negative i IED variation), and many others whose variation was almost stable. We must, however, consider the Sumaré outcome. Its significantly positive i IED was only reached from Hortolândia emancipation. So far is that Hortolândia has the worst CMA i IED of all, 75.27% to be more precise. In turn, Sumaré has one of the more intensive house conditions improvement of the region, approximately 18.5%. It seems, therefore, that the infrastructure investment was concentrated in Sumaré, considering that Hortolândia emancipation "purged" part of its troubles. In fact, this was one of the major reasons for the improvement of its i IED. That is, it would be immature to think that this living conditions increase 1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE experienced by Sumaré was solely due to these opportunities and services dissemination to a large number of households. Hortolândia, thus, continues to be the destination of most of those who can not afford the housing and living costs at Campinas or Sumaré (the CMA core) or in other towns where land and housing are more expensive. There is a certain novelty in the fact that Artur Nogueira compose another preferential destination target of this population group. Thus, very few cities expanded in a very intensive way their amount of new households units and have it done with quality, a fact that has a direct impact on land and housing value and, of course, on population flows that are directed to them. Not coincidentally, they are the same cities who most increased its habitation stock, but with a considerable irregular settlements percentage. What is more or less plausible is that the gap or mismatch between the Household Environment Infrastructure Index and the new urban households range between municipalities probably points out to typically urban phenomenon and processes, such as residential segregation (not to mention social segregation) between locations, i.e., to structural differences between them. And these (structural differences) function as the migration engines, especially the intrametropolitan one. From this argument another one is derived, namely, that the population that is directed to these spaces is different from each other (TABLE 5).

9 Cities, nations and regions in planning history

Tabela 5 – Net Migration according Migratory Modalities – Campinas Metropolitan Area, 1991- 2000

Net Migration according Migratory Modalities 1991 2000 Municipalities Total Intrametropolitan Total Intrametropolitan I. E SM I E SM I E SM I E SM Americana 18957 17932 1025 2002 11364 -9362 19079 17277 1802 5588 7204 -1616 Artur Nogueira 6123 903 5220 1213 356 857 7131 1896 5235 2264 423 1841 Campinas 92127 69129 22998 5046 25032 -19986 90539 89413 1126 9226 27712 -18486 Cosmópolis 6588 1515 5073 1324 717 607 5710 2805 2905 740 1171 -431 Eng.Coelho  ------2494 785 1709 303 330 -27 Holambra  ------1150 1114 36 463 731 -268 Hortolândia ------30585 5908 24677 13416 3276 10140 Indaiatuba 20478 4014 16464 1141 746 395 24245 7045 17200 2345 1260 1085 Itatiba 7696 3036 4660 277 645 -368 9989 4490 5499 640 584 56 Jaguariúna 4354 1912 2442 1433 1207 226 4922 1831 3091 1060 850 210 Monte Mor 7815 1923 5892 2161 994 1167 5784 2346 3438 1782 1149 633 Nova Odessa 7746 2796 4950 2692 1824 868 6465 2732 3733 2727 1609 1118 Paulínia 6584 1909 4675 2086 1209 877 8276 3032 5244 2307 1413 894 Pedreira 2218 1215 1003 283 373 -90 3473 1553 1920 542 263 279 Sta Bárb.dOeste 30624 3805 26819 8641 1068 7573 18044 10300 7744 4779 4718 61 Sto Ant.Posse 1954 1053 901 455 492 -37 1827 1040 787 421 474 -53 Sumaré 61266 6097 55169 20063 3031 17032 32124 13593 18531 10296 7165 3131 Valinhos 8819 3883 4936 2061 2094 -33 11671 4704 6967 4344 2354 1990 Vinhedo 5296 1842 3454 1079 805 274 7872 3539 4333 1300 1857 -557 Source: 1991 and 2000 Census. Own elaboration.  Municipalities that did not exist until 1991. In fact, all of them were created in 1993. .Immigrants. Emigrants. Net Migration (SM=I-E). - the emphasis is mine.

Not coincidentally, the municipalities with substantial net intrametropolitan migration are also those whose households stock were considerably increased. In fact, nothing more obvious, given that the construction of new units aims to give shelter to this substantial population contingent. See the case of Artur Nogueira, from a positive intrametropolitan net migration of 857 individuals in 1991; in 2000, under the same criterion, its balance is of 1841 individuals, and over the same period its Household Environment Infrastructure Index got worst. Another notable case is the one experienced by Valinhos. In 1991, its intrametropolitan net migration was negative (-33 individuals), indicating that this was a city where centrifugal forces acted strongly. In 2000, on the contrary, your balance is extremely positive, rising to 1990 intrametropolitan migrants. That is, from a repulsion area it became an attraction area, whose flows are strongly linked to Campinas and to the creation of closed condominiums (gated communities) for groups of middle and high income. Hortolândia had the greatest and the most significant positive CMA intrametropolitan net migration with the arrival of 10,140 new intrametropolitan immigrants. The second place in the ranking, at least in absolute terms, was Sumaré, once in 2000, 3,131 intrametropolitan immigrants 1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE landed there. However, this number was much more significant, once in 1991, its net migration was of 17,032 individuals. This fact leads us to believe that the perceived decline in its net migration was so abrupt, since most of its immigrants already were directed to what, in the future, would be Hortolândia. However, what was the CMA main repulsion area? With no doubt, Campinas. If in 1991, the long-distance immigratory movements "masked" the high emigration flows from Campinas, in 2000, this no longer occurs, once it emerges in this new scenario an own intrametropolitan dynamic. This was due not only to the mild slowing importance of long-distance migration flows but also to the intensification of emigration [return movements, mainly]. What is new here is the i IED incorporation. This suggests that the composition of the population flows differs greatly among themselves, pointing to structural inequalities among CMA municipalities, in a kind of analogy to what was proposed by Singer (1982), i.e., with his origin and destination areas. What is discussed here is that the land and housing market as well as another urban processes derived from there (recovery and land and housing speculation, essentially), are one of their main determinants. It is no coincidence, therefore, that in a country so unequal as and in a territorial clipping which concentrates a large part of urban poverty, the municipality that had the most remarkable positive intrametropolitan net migration (10,140) in 2000, also had the worst access to basic urban infrastructure (0.7527) conditions: Hortolândia – the nearest city to Campinas, the richest CMA city. Differences in the flows composition, mainly differences in income, can corroborate these inequalities. We take as an example, two municipalities whose intrametropolitan net migration were clearly positive: Valinhos and Hortolândia. While Hortolândia presented an i IED of approximately 75%, Valinhos, also in 2000, had a rate of approximately 94% and a variation between 1991 and 2000 of 3.13%. Almost a decade later (in 2007) approximately 71% of Hortolândia intrametropolitan immigrants had an average per capita household income of a wage, while all the Valinhos intrametropolitan immigrants earned around 1 to 5 wages. It is true that in order to illustrate how the infrastructure is crucial to the composition of urban land value and how this condition is essential for intrametropolitan migration, were chosen here the two extremes of CMA land and housing reality.

CONCLUSIONS

The present study has always been concerned about the possible correlations between land/housing market and spatial redistribution of population over the metropolis through its intrametropolitan population migration flows. In this

1 1 Cities, nations and regions in planning history sense, the construction of the Household Environment Infrastructure Index seems had been an essential tool to understand these possible connections. There is a sort of halo around the CMA core that starts in Monte Mor (South) and culminates in Artur Nogueira (North) one municipality whose range of new urban permanent households was higher throughout the decade and also one who had the highest percentages of irregular settlements or in regularization process. In other words, the growth, which encompasses both the urban sprawl as the population redistribution processes, of these municipalities are far from ideal, considering that they had not done with quality. Interestingly it was this group of municipalities who received more intrametropolitan migrants. With this, we do not want to lead the reader to believe that intrametropolitan population flows were responsible for the infrastructure deterioration of these cities. Indeed, far from this, they [municipalities] have historically been those who had always the worst indicators and indices. In this sense, the explanations escape from the simple causation relation, returning to historical-structural processes, where the speculation and urban segregation act strongly. Such observation only confirms the initial assumption of this study; that is, the i IED can be a tool for understanding historical and structural differences between internal origin and destination areas, where the land and housing market are one of the main determinants of intrametropolitan migration and as its selectivity.

REFERENCES

BONDUKI, N. J.; ROLNIK, R. Periferia da Grande : reprodução do espaço como expediente de reprodução da força de trabalho. In: MARICATO, E. A produção capitalista da casa (e da cidade) do Brasil industrial. São Paulo: Alfa-Ômega, 1982. CAIADO, M. C. S.; PIRES, M. C. S. Campinas Metropolitana: transformações na estrutura urbana atual e desafios futuros. In: CUNHA, J. M. P. (Org.). Novas Metrópoles Paulistas: população, vulnerabilidade e segregação. Campinas: Nepo/Unicamp, 2006. p. 275-304. CUNHA, J. M. P. (org.) Novas Metrópoles Paulistas: população, vulnerabilidade e segregação. Campinas, Núcleo de Estudos de População / UNICAMP, 2006. ______. Mobilidade populacional e expansão urbana: o caso da Região Metropolitana de São Paulo. Tese (Doutorado em Ciências Sociais), Instituto de Filosofia e Ciências Humanas-UNICAMP, Campinas, 1994. GOTTDIENER, M. A produção social do espaço urbano. 2.ed. São Paulo: EDUSP, 1997. (Coleção Ponta, 5). 1 5 th INTERNATIONAL PLANNING HISTORY SOCIETY CONFERENCE

JAKOB, A. A. E. Análise sócio-demográfica da Constituição do Espaço Urbano da Região Metropolitana da Baixada Santista no período 1960-2000. Campinas, tese de doutoramento, IFCH/Unicamp, 2003. MARICATO, E. Planejamento urbano no Brasil; as idéias fora do lugar e o lugar fora das idéias. In: ARANTES, O. B.; VAINER, C.; ______. A cidade do pensamento único: desmanchando consensos. Petrópolis: Vozes, 2000. RIGOTTI, J.I. e RODRIGUES, R. N. Distribuição espacial da população na Região Metropolitana de Belo Horizonte. In: ENCONTRO NACIONAL DE ESTUDOS POPULACIONAIS, 9., 1994, Caxambu. Anais... Belo Horizonte: ABEP, 1994. SINGER, P. Migrações internas: considerações teóricas sobre o seu estudo. In: MOURA, H. (Org.). Migração interna: textos selecionados. Fortaleza: Banco do Nordeste do Brasil S.A., 1980. VILLAÇA, F. Espaço intra-urbano no Brasil. São Paulo: Studio Nobel/ Fapesp/ Lincon Institute, 2000.

1 3