LAND AND URBAN DEVELOPMENT POLICIES IN A PLANNED CITY: ACHIEVEMENTS AND CHALLENGES IN BRASILIA, BRAZIL Ricardo Farret*

Abstract Urban planners and public officials agree that urban development – thus quality of life -depends to a large extent on the way urban land is managed. As such, property regulations, production and allocation of urban land are becoming the focus of a variety of experiences in many cities of the world which have in common a more active role of the public sector in implementing urban land policies. These policies have the triple role of (i) making the process of urban planning more flexible and responsive to social demands, (ii) making land to accomplish its social function, and (iii) recovering land valorizations "socially produced". Public ownership of urban land – once not used as a simple source of local funds - has been pointed out as a progressive tool to achieve these objectives, jointly with or instead of some traditional, cohercitive instruments such as land use controls and taxes.This paper analyses a unique experience in public ownership of land in Brazil, that of Brasilia, the planned 38 years old Brazilian capital city. Costly land use patterns in regard to the social reality of the city were identified. This is making the Government to reshape its land policy in order to use land as a capital asset to promote economic and social development.

Introduction There is a growing consensus among city planners and managers that urban development – and therefore the quality of life – depends, to a large extent and growingly, on how the land issue is dealt with. As objects of urban land policies, issues related to urban land ownership, production and allocation of urban land have been addressed in experiences carried out in many cities in the world in different socioeconomic scenarios. Usually, what these experiences have in common is the enhanced involvement of the public sector in land issues, as a result of the general understanding of the unique features of urban land, a heterogeneous asset made up of many other assets whose importance varies according to miscellaneous historical, cultural and economic factors. Therefore, the income elasticity of the land is determined by a combination of the various elasticities that characterize the demand for these features, which are hard to measure. The urban land market is thus marked by a remarkable set of unique features in relation to other markets, to which, by the way, it is strongly linked. Considered as a set of governmental decisions whose effects influence the allocation of urban land to different uses and social groups, the urban land policy plays the triple role of (i) facilitating the city planning process, making it swifter and more responsive to community demands; (ii) by making sure that the land performs its social function, ensuring greater efficiency and equity in its allocation; and (iii) recovering land rents embedded in appreciation in the value of real estate items directly or indirectly determined by actions of the public sector, and which are usually an object of private appropriation. For many years, public authorities have attempted to attain these objectives through coercive tools, such as controls over the use and occupation of the land, and tax instruments, the former being more related to efficiency and the latter to equity requirements. However, as problems derived from the disorderly growth of cities and the urban pathologies arising therefrom (social segregation and marginalization, shortcomings in the provision of urban services, fiscal crisis of the state, etc.) grew worse, local administrations were led to adopt more active tools to promote urban development precisely with resources raised through the land rent capturing tools. In Brazil, where this practice has not been widely disseminated yet, Brasilia represents an exception not only because it has been provided with a public agency charged with managing the urban land since before the city was established in 1960, but also – and perhaps most importantly – because of the evolution of the objectives, framework, and operational procedures of that agency, in tune with the approach adopted by the city

Government to land and urban development policies. In this context, a fragmented approach to the two processes (and respective institutional frameworks) is being forsaken and replaced by an integrated and more proactive approach to urban development. The basic approach of strictly controlling the use and occupation of the land – minimizing or even disregarding minimum efficiency standards in the allocation of the land, which until then was regarded as an abundant resource – is being gradually replaced by a more diversified approach, according to which the land is to be used more sparingly and taken advantage of as a tool for promoting economic and social development in Brasilia. This paper is intended to analyze these developments. From the outset, the transfer of the capital from Rio de Janeiro to Brasilia was based on land policy principles which in some cases were naive and in others quite daring. Section 1 analyzes the nature and unique features of the land and its market, as well as tools being adopted throughout the world to ensure greater efficiency and impartiality in its use, with emphasis on the public ownership of the land. Next, a brief historical account of the land policy in Brasilia, covering the few years after the city was established, is provided. The third section discusses the land policy in more recent years, when the public agency in charge of the issue dramatically changed its objectives and framework. Finally, some indicators are analyzed in order to evaluate the relationship of urban development and land policy in Brasilia.

The Urban Real Estate Market The nature of urban land The understanding of the urban real estate market is determined by how the land ownership issue is approached. It is a controversial issue, marked by many conceptual and methodological gaps. On the one hand, urban land, as an asset, has unique features such as immobility, indivisibility, and heterogeneity, etc. On the other hand, there are many agents involved in its production, marketing, and "consumption": legal property rules, the construction industry, financial agents and middle-persons (developers, builders, designers, consultants, etc.) whose interests, motivations, goals, strategies, alliances, and conflicts have not been so often studied in urban economics. Up till the 1970s, studies of how the urban space should be structured were predominantly based on the neoclassic economic approach, which regards the market, with its neutral and scattered nature, as the tool regulating supply and demand. According to that approach, the local production and allocation of urban land for different applications and at varying rates result from an aggregate demand based on a theoretically optimal economic behavior of consumers – individuals and, to a lesser extent, firms. The issue of how the urban space should be structured is thus depoliticized, reflecting "a naive conception of the behavior of consumers and institutions" (Dowall, 1978, p. 65). Any circumstances affecting the balance of that market are blamed on institutional obstacles and usually attributed to governmental actions derived from its regulatory power over the urban development. This scenario underwent changes as of the 1970s, particularly, but not exclusively, with the emergence of an approach focused on the behavior of agents and institutions engaged in the intra-urban structuring process in the fields of urban economics (Ball), urban geography (Harvey, Short), and urban sociology (Castels, Topalov). Doing away with the concept that the space is neutral, or merely a scenario where production and consumption relations develop, these authors showed that, in fact, it plays an active role in social processes and that its production and appropriation are therefore subject to the same conflicts and barriers that determine the allocation of the social product. It should be highlighted at this point that this view on the real estate market had, to a certain extent, already been proposed by Form, who said that "the concept of a free market, without any organization, where individuals compete impersonally with one another for the land should be forsaken," suggesting the need to take social aspects into account in the analysis of urban spatial processes (1954, p. 317). As opposed to neoclassic authors, Form believes that the real estate market is highly organized and dominated by a broad range of agents, institutions and social groups with clear-cut objectives and interests, including the State, although the latter plays a merely instrumental role.

In all these approaches, the formation of real estate prices ased on the appropriation of urban land rents both for sale and lease remains a substantive issue. It is not a new issue, as social scientists have been dealing with it since the days of the classic economists. Ricardo, for instance, asked two questions almost 200 years ago that remain largely unanswered. The first, how can a good – the land – whose existence was not brought about by any labor have a value? And the second, which is very much related to the previous one, is: what is the ethic justification for the private appropriation of this value? The neoclassic approach eliminated these ethic concerns by stating that rent is just the payment for a production factor, just like wages remunerate the labor and interest rates remunerate the capital. The land would therefore have an intrinsic productivity, which in the case of urban land would be determined by the inherent accessibility of each portion of space. The competition among consumers for space generates a gradient of rents that are naturally appropriated by proprietors in the form of sale or lease prices. Therefore, the rent would play the technical role of rationalizing the use of a resource in short supply. "Without it neoclassic authors would say that one could produce manioc in the downtown area of São Paulo" (Cunha and Smolka, 1978, p. 8). Urban political economists, however, far from believing that the land rent can be technically and conceptually explained in terms of an intrinsic productivity of the land, describe it as a "produced" good that is subject to the same laws governing the production, circulation, and consumption of any other good. Its unique feature lies in that, in the formation of its price, a factor is involved that is not physically linked to it, but rather to other areas of the city and incorporated into it: the appreciation in value derived essentially from actions of the public sector – changing designation of rural land into land for suburban development; increase in investments in services and urban improvements; zoning and other land use regulations such as building density; etc. Therefore, while according to the neoclassic thought land rents result from physical relations among things, from the standpoint of urban political economists they result from social relations among people (Farret, 1985). The urban land market According to the emphasis placed on one or another approach, the functioning of the urban real estate market has led to different views to deal with the above-mentioned ethic issues. The neoclassic approach, which assumes that the market is perfectly competitive and each participant can buy and sell limitless quantities of land and uses without affecting the price determined by the law of supply and demand, like any other good, already incorporates some institutional variables, such as, for example, the economic consequences of regulating the land market for prices and the legal-fiscal ownership regime. On the other hand, urban political economists have disseminated two market alternatives. According to the first one, the production and allocation of urban land should be determined by political-administrative mechanisms and not by the market, so as to avoid the private capture of any appreciation in real estate values, as in the socialist cities. The second one states that, in addition to traditionally accepted interventions of public authorities in the real estate market, such as zoning and tax arrangements, it must be based on positive aspects of the neoclassic approach, advocating the large-scale public ownership of urban land and, on a lesser scale, of housing, within a general market context. In this second alternative, the ownership of the real estate stock (including land) by the public sector would eliminate the speculative behavior that characterize the market through a planned (in time and space) supply in the form of sales, leasing, or temporary assignments. This is the case, among others, of land banks in Sweden and Holland; land nationalization in Great Britain; social rent housing in France; functional housing in Brasilia. In any of these alternatives, it is assumed that land policy should be associated to urban development goals such as: urban growth should be guided in its form and direction; urban services should be provided according to different spatial priorities; certain areas should be protected; large-scale infrastructure works should be anticipated and coordinated; etc. Because the market alone doesn't carry out any of these actions correctly, minimal coordination mechanisms must be established between urban planning and the real estate market, particularly in connection with its land and housing components. None of such alternatives are free of controversy. For instance, urban land banking is generally regarded as a "benign monopoly" that, in addition to being aimed at controlling land prices and fully or

partially capturing real estate valorizations, is also intended to guide the urban growth according to a desired pattern. Its critics argue that it is an unproductive capital, since requiring the allocation of large amount of money and considering it takes a long time for any actual gains to be produced, it could be used to meet more urgent urban social demands. Whatever legal system is adopted, the effectiveness of the land policy must be assessed in the light of objective criteria, like any other public policy. Three of these criteria are: a) better (enhanced efficiency and social justice in the city planning; b) control over the supply and the price of the land; c) community participation in the appreciation of real estate values socially produced.

Land Policy and the Challenges in Urban Development in Brasilia The Transfer of the Federal Capital The idea of transferring the capital of Brazil, first from Salvador to Rio de Janeiro, and to Brasilia, into the interior of the Country, has been in the Brazilian political agenda since the 18th century, and has been incorporated into different Brazilian Constitutions since 1891. Different motivations, some of them grounded and others unrealistic, nurtured the idea along all those years (Farret, 1985). A proposal prepared by a Federal Committee charged with defining the location of the new Federal Capital, which was refined through field studies in 1955, led to the selection of a 5,850-km2 area in the Brazilian central plateau for that purpose. Because the area was fully located in the state of Goias, the government of that state issued a decree, in 1955, expropriating and assigning it to the Federal Government as an area of public and social interest. After that proposal was approved, the Committee changed its role becoming the Committee for Planning the Construction and Transfer of the Federal Capital, which began to work in partnership with the government of the state of Goias. As of that moment, amicable expropriations of land areas included in the space selected for the new Capital began. In 1956, a public agency, the Corporation for the Urbanization of the New Capital -Novacap) was set up to plan, urbanize, and build the future Federal Capital. This corporation was also entitled to act as a real estate company, with full powers to buy, barter, alienate, rent, or lease real estate goods in Brasilia. In 1972, a new public agency was created, the Brasilia Real Estate Corporation – Terracap, to deal with real estate management, assigned to it by Novacap, which then focused exclusively on urbanization and building activities. The land policy Since the Constitution of 1891 was promulgated, a provision determined that the land in the Federal District would be expropriated and assigned to the Federal Government. Although the legislator did not specify the purpose of that provision, it is believed that the measure was intended to achieve two objectives: a) to finance the works required implementing the Capital with funds derived from the sale of land; b) to facilitate the planning and implementation process, because without expropriations there would be no judicial conflicts that could hinder the transfer of the Capital. Although the means (less than 2/3 of the land were actually expropriated, for example) and the ends were not fully achieved, the city government is provided with a unique tool for urban planning in relation to other Brazilian municipalities. Because it accumulated the functions of the various agents involved in the production of the built environment, acting as landowner, planner, developer, builder and, in some cases, even as financial agent and seller of real estate goods, which is a unique feature in Brazil, the Government of Brasilia seems to have disregarded, for a long time, the need to make sure that the land was efficiently used and left this issue out of its urban policy agenda, since "it did not feel the need to establish parameters for itself " (Malagutti, 1997, p. 68).

Urban land developments would be gradually created, planned and occupied, according to local needs, through a process of assignments to individuals or corporations by means of tenders, donations or loans for use. In the 1950s, it was proposed that the regime of private ownership of the land in Brasilia should be eliminated and replaced by the surface property regime, according to the British social model. The lease model would thus be adopted, ensuring to the State, the tenant of the land, permanent revenue from the users of the land. It should be highlighted that the Government of Brasilia, worried with a typical phenomenon in Brazilian cities, particularly after the mid-1950s – the proliferation of illegal land developments for the lowincome population – included a provision in Law, according to which urban tracts of land could not be subdivided by private developers, a provision that was maintained by Terracap: "real estate properties alienated by Novacap or Terracap in Brasilia shall be physically indivisible." Because of this legal provision, private developers are forbidden to subdivide and sell fractions of urban real estate goods, as opposed to the situation in the rest of the country, where any owner of a tract of land can subdivide it into lots, as long as their size and uses are in accordance with the municipal laws. An even more interesting aspect to be pointed out is that the legislator did not believe that Brasilia would continue to be an "island of urbanistic legality ", as neighbor municipalities in the state of Goias would not withstand the pressure for new developments capturing positive externalities generated by the implementation of the new Capital. Therefore, the law also imposed restrictions on rural properties covering less than 20 hectares within a 30-km strip outside and along the perimeter of the Federal District, which in practice meant that law encroached upon the state of Goias! This measure was intended to protect the so-called "green belt" of Brasilia, thereby ensuring that those areas would be used to supply food and other products to the Federal Capital. The provision, which was not applied in practice on grounds that it was unconstitutional (a final judicial decision on the matter is still pending!) was aimed at preempting – and therefore avoiding – the chaotic parceling of the land that, since the 1980s, has occurred in the outskirts of Brasilia, where settlements without any provision of infrastructure, social services and facilities, or jobs were established. As of 1960, when the Capital was established, the sale of lots by Novacap was sped up as the city was planned and in process of implementation. However, this was done without any urban spatial policy that could have made the use of the land in Brasilia more efficient and socially fair. The regime of public ownership of the land was not and is not, by itself, capable of ensuring these objectives, originally anticipated in the urbanistic planning of the so-called Plan Piloto area, the core center of Brasilia. Since the 1980s, there is a multiplication of irregular settlements of low-income and even middle-class groups in the outskirts of Brasilia, as a result of a growing migratory flow of people looking for jobs, which were often an illusion. Enhanced by emerging political populist schemes, the pressure of that migratory flow led to the large-scale donation of lots to the low-income population "as a tool to manage social conflicts" (Malagutti, 1997, p. 70). Many of the resulting developments were irregular, because the lots had no reliable title registration. During this period, the land policy was very much linked to the urban development policy adopted by the Government of Brasilia. First, the institute of the public ownership of the land was confined to strict limits, as land began to be managed as an asset to be assigned, sold, donated, or created according to the social or financial pressures faced by the government; second, as a consequence of the first, the "planning" process paradoxically consisted in efforts to adjust the organization of the land to its actual occupation; and third, the so-called satellite towns were established much more as a result of an urban planning strategy than of attempts to accommodate de facto occupations. As the process of implementing the Capital evolved and the city was more consolidated, the land policy began to be seen as an urban development tool. Perhaps its most important feature in the last decade is that its connection to urban development was enhanced in the broad sense. The treatment provided to land as input in elastic supply can be easily perceived in certain residential areas of Brasilia, particularly in Plano Piloto, South Lake, North Lake, and Mansion Sector, where we find low

and very low densities. These areas therefore require high and ongoing investments from the government and community at large, particularly to maintain the necessary urban infrastructure and facilities. The option for low housing densities and an extensive occupational model, which would, in theory, ensure a better quality of living for the population, led to a pattern of large-scale land consumption that turned out to be uneconomical and jeopardized the quality of living standards achieved in certain environments. Although we cannot say that there is a clear cause and effect relation between the abundant stock of land in the hands of the Government of Brasilia and its generous or even lavish use, it is clear that the land use model adopted has reduced the quality of living that had been originally planned and even achieved in some cases. This fact shows that the institute of public ownership of the land cannot, by itself, ensure more efficient land use patterns. It has to rely on a planning and management model that can make it more effective. In the light of those observations, and aware of the fact that land is a finite and scarce resource, the Government of Brasilia laid, in 1995, the groundwork for its urban development policy, which is based on four major measures: a) promotion of a planned occupation of the land, based on the rational use of the public non-built tracts of land and private vacant lots; b)decentralization of service activities through the establishment of new intraurban secondary centers; c) legalization of existing irregular settlements; and d) expansion and qualification of the supply of urban infrastructure and facilities in all existing settlements. The success of this policy will depend to a large extent of an adequate support of the land policy which, it should be pointed out, it is not just a matter of defining rules for a sound use of the land. These measures will not be effective as they should if they are not supported by an institutional framework capable of converting, in a swift and comprehensive manner, new economic and social demands into new configurations in the use of the land and in the built environment. The neoclassic assumption that this would be naturally promoted "at the right time, in the right place, and at the right price" by the many agents involved lacks empirical evidence in the real estate market. The complexity of the new economic and social demands facing the local governments, at large, are derived from (i) new determinations and flows of capital (both domestic and foreign), (ii) new positions of the government with regard to the creation of jobs and income, and the provision of goods and services, or (iii) changes in the profile of real estate markets, shaping a new environment where new and the existing agents try to change its organizational framework, objectives and strategies in relation to the production/renewal of the built environment. In Brasilia, specifically, the major factors affecting its urban development are: a) the mobility of the domestic and international capital (the latter either directly or in association with the former), for which the ownership of land is much more a source of accumulation than an asset, thus must be placed in the market and not kept in speculative portfolios; b) the restructuring of the domestic and international economies, with significant spatial impacts, as, for instance, the case of Mercosul, the South American Common Market; c) the social pressures for economic development, environmental quality, and urban services, simultaneously; d) the larger participation of the private sector in urban development which creates a demand for partnership, concession and privatization of state enterprises. With the enlargement of the real estate market as a result of new demands for economic activities and the emergence of new agents (for example, pension funds, real estate portfolios of domestic and foreign banks, the posture of large institutional urban proprietors, such as the University of Brasilia and the Federal Saving Bank (CEF), among others, significant changes in the institutional framework of urban development taken by the Government of Brasilia were to be expected, with Terracap taking the lead. On the other hand, the existence of many public agencies, directly or indirectly involved with the urban development planning of Brasilia – the Institute of Urban Planning of the Federal District (IPDF), the Institute of Housing Development (IDHAB), the Metro, the Secretariat for Industry, Commerce and Tourism – led, more than ever, to the need to establish mechanisms to promote the articulation not only among them but also among the urban development and the land policies.

A new institutional framework for the land policy To face this challenge, Terracap lead, in 1997, a substantive institutional reorganization, changing its role from a corporation designed for selling lots for residential, business, service, and industrial uses in order to feed the treasury of the Government of Brasilia, without any direct relation with urban development plans. In 1997, it was assigned a more important role to Terracap in connection with the urban development policy of Brasilia by turning it into a Business Development Agency. It was only the natural development of a company that intended to use its land properties in a more progressive way – not as currency to fund the government anymore, but rather as a resource base aimed at ensuring partnerships in economic undertakings with the private sector. As a result, Terracap began to play an important role within the Economic Development Program of the Federal District – PADES/DF, which is aimed at ensuring favorable conditions to enhance private investments and promoting economic and social development in Brasilia by encouraging industrial activities and the service sector, including tourism, thereby reducing unemployment rates and increasing the tax revenue. With that purpose in view, Terracap has been relying on the mechanism of granting concessions for the use of land with a purchasing option, where the remuneration derived from the use of the land is not linked to its price, but rather to the profitability of the undertaking. The percentage of this participation, as well as the minimum value, varies according to the type and size of the undertaking. The Government thus becomes a sort of a quotaholder whose public revenue comes from the private management of the undertaking. Therefore, although it is a real estate company, selling lots is not the main activity of Terracap anymore. Other strategic and social-oriented initiatives have been included on the company's agenda. First, acting in the legalization of both low-income irregular settlements, up to now involving a population of almost 500,000, and medium-income settlements, benefiting a population of about 110,000. Second, promoting incentives to economic and social development, through temporary assignments with a purchasing option to firms. Paying an occupation fee, variable according to the size of the firm (from 1,0% to 1,3 % of the market value of the land) and indexed on annual basis. If the deadline to implement the undertaken (from 12 to 48 months) is not met, the user has to pay a monthly fine to Terracap, contrasting with what the company used to do in the past, when it adopted the practice of buying the land back. Terracap gave up this practice because it was a "financial trap" for the government, since judicial decisions forced it to indemnify the defaulter for all the sums paid to the company, including improvements in the land, in indexed payments which were usually higher than the actual market price of the land, thereby benefiting the defaulter. Some of the undertaken under way are: the Shore Project, aimed at attracting investments in tourist and recreational undertakings on the shore of Brasilia's lake in the form of hotels, marinas, shopping malls, fairs, convention centers, cultural facilities, restaurants and bars; the Dry Dock Warehouse Complex; and the High-Technology Pole (Tecnopolis). Third, the promotion of a planned and efficient use of the land in strict coordination with IPDF. In order to increase de cost-benefit of urban services, including the new subway transportation system (metro), Terracap has given priority to the occupation of empty urban spaces and to the densification of urbanized areas. As a result, it has increased the supply of land, without burdening the public coffers. Among other undertakes, it must be pointed out the under construction residential area of Aguas Claras, with a population of 300,000.

Land Policy in Brasilia: A Preliminary Assessment The assessment of land policies is not a recurrent subject in the urban literature, even in countries with a long tradition in the use of more "progressive" tools (Australian Government, 1974; World Bank, 1978). This gap is even more remarkable in Brasilia, which has never specifically addressed such objective. This can be partly explained by the "abundant" availability of and by the political and social pressure of the demand for land, which have deviated the attention of the local government to the seriousness of the land use problem. However, it seems quite possible to make a preliminary assessment of the land policy in Brasilia, particularly in which concerns with public ownership of land, based, on the one hand, upon some of the objective criteria defined early in this paper, and, on the other hand, upon existing studies and reports concerning land use patterns and processes in Brasilia.

Despite being an open issue, and one that can hardly be measured, the first assessment criterion – "better urban planning" – can be analyzed in the light of some indicators. The first one refers to the compaction degree of the urbanized area in Brasilia. In this regard, Farret (1983) and Farret and Campos (1996) show how the city was implemented following the logic of the private speculative land developments, according to which more accessible areas are urbanized and sold after less accessible ones, leaving large empty urban spaces between them. As mentioned before, once the Government monopolized the land development process in Brasilia, a more compact pattern of land occupation was to be expected not only to reduce infrastructure costs but also, and mainly, to preserve the principles of a fair social distribution of population in terms of income, job location, and access to facilities, as emphasized in the Brasilia's master plan from the outset. This has not always been the case, once large distances (40 km in average), in the early stages of the implementation of the city – and for many years thereafter – have separated Plano Piloto – the high income area of the city - from the "satellite towns" where the medium and low income population live. It should be stressed that the pattern of a spread-out occupation, which resulted in one of the lowest rates of passengers per kilometer and, therefore, in an inefficient and expensive public transportation system, was gradually replaced, particularly as of the 1980s, by a star-like urban model along occupation and transportation corridors (Farret and Campos, 1996). The close links between the land policy and the general urban and housing policies represent another example of the inadequate use of land policy tools in the Federal District, particularly in the 1970s. Based on the assumption that by reducing the supply of public housing facilities the migration of low-income families to Brasilia would be discouraged, the local government (which at that time lacked any political autonomy and acted more or less as a federal agency) ended up generating a demand that led to an intensive process of land development in the region surrounding Brasilia, which, nowadays, with a population of over 500,000, represents a serious management problem for both the governments of Brasilia and the state of Goias where these new settlements are located (Farret, 1983). Urban density is another important planning indicator, because of its clear bearing on the costs for implementing and operating the urban infrastructure, particularly sanitation facilities and the road system. These services account for 21% and 42% of the total infrastructure cost, and for mean costs per hectare of US$ 20,000 and US$ 40,000, respectively. Moreover, the total cost of the entire urban infrastructure per domicile is of about US$ 11,000 (Mascaro, 1995). Because of the idea that low densities ensure a better quality of life has been widely disseminated, very low rates (50-80 inhabitants per hectare, in average) prevail in most Brazilian cities. In the city of Porto Alegre, for example, which has a population of two million, infrastructure costs are three times higher than in Paris, whose population is five times larger and where the per capita income is ten times higher than in that city. In Brasilia, the PEOT – Structural Plan for the Territorial Organization of the Federal District – showed, back in 1977, that "low population densities, particularly in peripheral communities, constitute the main urban design problems in Brasilia. (this) problem is first perceived at the global level... (where) the fragmented and spread-out occupation of the land creates distances between the different communities that constitute barriers to the interaction between the different parts of the city. This spatial segregation leads to poor and unequipped urban structures" (PEOT, 1977, pp. 296-297). At the intra-urban level, the PEOT revealed net, variable residential densities that nevertheless fell short of the rates that could ensure costs more in tune with the urban and social reality in Brazil. (Table 1)

Table1: Housing Density and Cost of Infrastructure in Brasilia Settlement Plano Piloto Cruzeiro Novo CruzeiroVelho, Nucleo Bandeirante Ceilandia,Taguatinga, Sobradinho, Planaltina, Lago Sul, Lago Norte Guara II Setor de Mansoes

Net Density (persons/sq.meter) 155 - 500 305 114 -187

Cost of Water and Sanitation (Cr$/m2/ per capita) 420,00 320,00 1.200,00

60 – 200

1.300,00

20 100 - 200 1,3 - 10

3.700,00 not available not available

Source: GDF (1977) Although observation shows that densities have been growing over the years both globally (through the establishment of new and the growth of several satellite towns) and in each town (through changes to higher occupation rates), with the first process prevailing over the second one, it is believed that the density rate is still below appropriate economical levels, including urban transportation services. Regarding the second criterion – control over land prices -, it could be said that if the Government of Brasilia, through Terracap, used the public land ownership tool effectively, land prices in Brasilia could have developed two alternative behaviors over the years. On the one hand, these prices could have increased, in relative terms, beyond the levels observed in other cities, which would evince the clearly monopolistic behavior of the government, similar to the behavior of a private monopoly. On the other hand, the government could have assumed a more social posture, offering land plots to the population at lower prices than those prevailing on the market, leading to a relative price curve lower than the one registered in other Brazilian cities during the same period. In this regard, Farret and Campos (1996) show that land prices gradients in Brasilia, Rio de Janeiro, and São Paulo were similar in the 1960-1991 period. Therefore, it seems that public land ownership, the focus of the land policy adopted in Brasilia, did not fulfil one of its main objectives – that of regulating prices - as prices varied in a similar way in places where that instrument had not been adopted. Finally, with regard to the third major objective, namely, the recovery of "socially created" land values, no specific calculations are available to compare the evolution of actual land prices on the market with those achieved by Terracap in tenders. However, rough calculations seem to indicate that both have increased, although the quantitative behavior of the two gradients is not known. After all, these considerations evince the need for studies about some specific aspects of the land policy adopted in Brasilia, particularly those related to the role played by the public land ownership, such as the determination of the opportunity cost of the stocked land in the hands of Terracap in comparison to the purchase of land on the market as the need arises; a cost-benefit analysis of the alternative tools of allocating urban land: sale of urbanized lots, leasing, and partnership in business, residential, and/or industrial undertakings; assessment of land policies as a price-regulating tool, based on some alternative models for the action of public authorities: control of the supply of urbanized tracts of land; direct price control; subsidized supply, so as to push prices down. These and other issues aim at establishing the actual advantages of the public ownership of land as a tool for efficiency and equity in urban development. Concretely, there is no evidence that the sheer and continuous transference of the public stock of land to the private sector be the correct policy to be adopted. It seems more likely that the land policy must be engaged in a continuous process of production and renewal of urban areas, which constitutes the essence of the urban development process.

References Australian Government (1974). Urban Land: Problems and Policies. Canberra: Department of Urban and Regional Development. Cunha, P. and Smolka, M. (1978). Critical Notes on the Relationship Between Urban Land Rent and Land Use. São Paulo: Fundap. (in portuguese) Dowall, D. (1978). Theories of Urban Form and Land Use. Berkeley: IURD, U.of California/Berkeley. Farret, R. (1983). Changing Residential Structure in a Planned City, PhD Dissertation, U.California/Berkeley. __________ (1985). Paradigms on Urban Spatial Structuring Processes. In R. Farret (ed), The City Space – A Contribution to Urban Analysis. São Paulo: Projeto/CNPq. (in portuguese) __________ and Campos, N. (1996). Signalizations of the Real Estate Market in a Context of the Public Ownership of Land: Urban Spatial Structuring in Brasilia, Annals of the II Research Network Seminar on Real Estate Market Dynamics and Urban Spatial Structuring of the Brazilian City. Brasilia: NEUR/University of Brasilia. (in portuguese) GDF (1977). Structural and Territorial Plan for the Federal District –PEOT. Brasilia: Government of the Federal District. (in portuguese) Malagutti, C. (1997). Illegal Urban Land Developments in Brasilia: Regularization or Exclusion?. Brasilia: Master thesis on urban planning, University of Brasilia. (in portuguese) Mascaró, J. L. (1995). Urban Infrastructure Systems in Brazil. Porto Alegre: Propur/UFRGS. (in portuguese) World Bank (1978). Urban Land Policy Issues and Opportunities. Staff Working Paper # 283. * Urban Planner, PhD University of California/Berkeley; Associate Researcher at the NEUR/University of Brasilia. E-mail: [email protected] Public houses made available to civil servants by the federal government as a fringe benefit, with a low, sometimes simbolic, monthly rent payment. It is part of the folklore of the city the ideology of spatial justice represented by the case where a senator and his driver woul live in the same neighborhood but, of course, in different types of housing. In fact these settlements are neighborhoods of Brasilia since they do not have political and financial authonomy. Brasilia obtained its political and financial authonomy in 1988.