Lessons on Public Housing From Singapore for Sao Paulo

In the mid-1990s Sao Paulo began a public housing program based on urban renewal strategies developed three decades earlier in Singapore, half a world away. Despite some modest success, the program was abandoned by Sao Paulo's government in early 2001. This raises questions about how south-south transfer processes can be make to work in practice, as discussed here by W.E. Hewitt, Ph.D, Associate Vice-President £¨Research£©and sociology Professor at the University of Western Ontario in Canada.

INTRODUCTION
In an era of instant communications, globalized commerce, and continuing advances in food production and disease prevention, housing remains one of the most basic human still unmet or largely compromised on a massive scale in many areas throughout the world. Housing deficits are especially acute in the major urban agglomerations of the developing world. Millions are forced to live cheek by jowl in downtown slums or in sprawling "suburbs"of precarious dwellings with few services and a minimum of infrastructure.

Though generally recognized as priority by developing world planners and politicians alike, there are precious few examples of national, regional, or local housing policies which have worked effectively to resolve this issue. This is true even in some of the more affluent developing regions, such as Latin America. Here, urban planning per se has evolved only recently, with policies directed specifically towards housing still rare. Those in place remain so seriously underfunded as to offer very little relief to the millions of families in need.

As Latin American municipal govern ments increasingly look outward in their attempts to seek solutions to the problems which affect them, change may, however, be on the horizon. With a now global trend towards south -south cooperation and information exchange, many Latin American cities have begun to look seriously at innovative housing policies and practices from other regions of the developing world as possible models for emulation in the domestic context.

One such borrowed "innovation"in the housing area was adapted by the city of Sao Paulo-Brazil's largest urban center -in the mid-1990s, drawing on Singapore's decades- long experience with urban renewal. Dubbed Projeto Cingapura £¨Singapore Project£©or PC for short, this program of public housing verticalization aimed at the city's growing housing deficit, estimated at over one million units. Given existing urban densities and the high cost of urban land, the program seemed especially well suited to meeting Sao Paulo's public housing needs. Despite its initial promise and successes, however, the PC provided for only a modest increase in available housing stock and was formally abandoned by the city government early in 2001.

Using data obtained from archival sources and field interviews conducted during the main phase of project construction, this study explores both the promise and the challenges which faced the PC during its short history. It seeks to develop a broader understanding than is currently evident in the literature with respect to the potential and especially the limitations of south -south technology and information transfer. The study clearly reveals that effective models for emulation in the housing area, while critical to the process, cannot in and of themselves ensure the success of the transfer process. Without political will, resources, and effective management and follow-up, even the best of innovations will have difficulty in adapting to foreign climes.

SAO PAULO'S URBAN HOUSING PROBLEM
Rapid urban growth since the 1960s £¨UNCHS, 1996£©has created a crisis of major proportion on the provision of suitable housing in most, if not all urban areas in the developing world £¨see Brabdt, 1980; Rodwin, 1987; Struyk, 1988; Werna, 1996£©. This situation has increased public awareness of housing problems, along with strident calls for action on the part of aid agencies and nongovernmental organizations to address the issue through national and local housing strategies. Some action plans have focused on the provision plans have focused on the provision of basic housing in single or multiple dwelling format, with concomitant investments in local services and infrastructure -usually targeted to the poorest populations. Another common strategy is slum upgrading by either contracted or selflabour to improve the quality of existing housing stock. Government sponsored housing finance schemes are also increasingly common, as are incentive schemes encouraging private sector developers to provide low -cost housing to those who may not otherwise be able to afford it £¨Brennan,1993£©.

However, the overall impact globally of government action on the housing front is remarkably limited. In Thailand, for example, the National Housing Authority produces fewer than 8,000 units per year, a paltry fraction of the total 250,000 units required. In Madras, India, less than one public housing unit is completed for every five needed £¨Brennan,1993£©.

Sao Paulo's recent experience in public housing reflects these dismal statistics. Between 1960 and 1991, Sao Paulo's population skyrocketed from 3.6 to over 9.6 million £¨PMSP, 1996£º49£©, owing mainly to large-scale in -migration from the impoverished northeastern regions of Brazil. By the 1970s growth rates in many parts of the city had reached 10 percent per year, with most migrants experiencing severe difficulties in locating adequate housing £¨Kowarick and Bonduki, 1994£©.By 2000, the housing deficit in Sao Paulo was estimated at approximately 1 million units £¨'Marta pretende regulizar', 2000£©. Substandard housing in some form or other occupied about 70 percent of Sao Paulo's area-approximately 1,500squae kilometers, or three times the size of Paris £¨Greenfield, 1994£º103£©. For example, the number of slum dwelling or favelas grew astronomically form a handful in 1973 to well over 350,000 in 1993. These residences are now home to some 2 million individuals, around 20 percent of the total municipal population. Residents of converted older homes and factories in Sao Paulo's inner core-known as corticos-are now estimated to number over 500,00 £¨PMSP, 1996£º146-148;'Corticos e favelas', 2001£©. For this segment of the population, living conditions are extremely precarious. Frequently whole families share one room, some-times without proper ventilation, electrical wiring, or plumbing. Rat and cockroach infestations are common.

Over the years, disparate attempts to deal with the housing crisis in Sao Paulo have been spearheaded by an alphabet soup of agencies established at the federal, stare, and local levels. These included a federal band £¨BNH£©which funded urban housing projects and low-interest loans to lower and middle-income home buyers; state-level cooperatives institute £¨INCOOP£©which helped build housing for state workers in a range of occupations £¨teachers, transit workers, etc.£©; a state-level development company £¨CODESPAULO£©for housing for low-income families and financing of slum upgrading projects; a collaborative private sector/state company scheme £¨COHAB£©to develop housing for limited -income families; and a municipally managed COHAB for public housing construction, which also funded self-help projects £¨"mutiroes"£©to upgrade substantdard housing.

Between 1965 and 1982, about 154,000 housing units were constructed or upgraded, mostly though Sao Paulo's COHAB £¨Shidlo, 1990£º76,81£©. Since the early 1980s, due to cutbacks at federal and state levels, the municipality's burden of public housing construction has increased still further £¨Shidlo, 1990£º77,82£©. Yes the number of units constructed by the municipality each year since the mid -1980s has been pitifully small, averaging less than 6,000 per year £¨PMSP,1996£º2,6; werna, 1995£©--hardly sufficient to even begin to meet the extant need.

More recently, some attempts were made to speed up public housing construction. During the 1989-1992 administration of leftist mayor Luiza Erundina, the municipal government invested in more cost-efficient self-help housing initiatives, known as mutiroes. The city provided funding directly to community groups, which in turn engaged local families to build new or renovate existing houses. Still, less than 8,000 units were built or improved during each year of the Erundina administration £¨Kowarick and singer, 1994; PMSP,1996; 2,6; Werna, 1995£©.

GLOBAL SEARCH FOR SOLUTIONS
With the election of right -wing mayor Paulo Maluf in 1992, Sao Paulo's municipal government began to seek more "radical"solutions to the city's housing crisis. Eschewing past practice, Maluf initiated an ambitious urban renewal plan borrowed from Singapore, a city-state with one of the world's most remarkable successes in urban renewal and housing construction.

Sao Paulo's search for solutions from other parts of the developing world -in this case, Asia- is part of an emerging global trend towards south -south technology and information transfer. Contrasted with more standard forms of north -south transfer £¨typical of most bilateral or multilateral aid strategies£©, south -south transfer is widely seen as offering opportunities for partner organizations and governments to interact as equals, and to directly share experiences leading to workable solutions of direct relevance and benefit to populations at risk. Increasingly studies have focused on information sharing and transfer in a variety of contexts, from environment £¨Clusener-Godt and Sachs, 2001£©, to trade £¨Brown, 1998£©and science and technology £¨Hassan,2000£©.

In urban development, Brazil's Institute de Administracao Municipal £¨IBAM-Institute of Municipal Administration£©has actively supported transfer of know-how and resources to urban governments in a range of developing countries. Since 1993, for example, IBAM has served as consultant to the Peruvian government regarding municipal affairs in the formulation of the country's new constitution; provided information to Cuba on the construction of low-cost housing; worked in a project to strengthen local government in Mozambique; and finally, helped set up a regional planning institute in Indonesia £¨IBAM Newsletters, 1993-95£©.

On a broader scale, another example of south-south transfer is the Mega -Cities Project, which links 18 of the world largest metropolitan areas in an effort to promote exchange of ideas and technological innovation (see Badshah and Lazar, 1995). In one case, methods used opportunities to reclaim and recycle waste products to sell for profit £¨to finance local neighbourhood improvements£©were transferred to both Bombay and Manila. In another case, an anti-littering strategy aimed at children £¨using cartoon characters£©which was developed in Bangkok, was later implemented in Rio de Janeiro.

Despite such examples, what we know about the process and critical outcome of south -south cooperation remains limited, with much of the literature to date focused on its more generic aspects and broad successes. This makes Sao Paulo's experiment with housing strategies developed in Singapore both intriguing and illustrative, because it reveals not only the promise, but also the complexities and the challenges associated with south -south exchanges.

The course of Sao Paulo's experience with the Singapore model, its successes, attendant frustrations, and failures, were of the project in 1995 to its end-point early in 2001. Research data were drawn from site visits to PC projects in 1997 and 2000, interviews with municipal administrators, PC project coordinators and residents during the main phase of project development in 1997, and a review of archival material and media reports.

THE SINGAPORE MODEL
Singapore's successes I slum removal and "verticalization"are now legendary. Its strategy mainly responded to critical urban overcrowding resulting from unprecedented levels of in-migration. Work on the strategy began in 1960 with the establishment of the Singapore Housing and Development Board £¨HDB£©. With generous government funding the HDB has overseen construction of approximately 700,000 flats, housing over 85 percent of Singapore's total population £¨SHDB,2002£©. Much of this construction was completed by the late 1970s and early 1980s. Since the 1990s the HDB has concentrated primarily on upgrading, applying new design concepts to deal with an increasingly sophisticated and upwardly mobile population and to position Singapore firmly as a global city £¨see Goh, 2001£©.

In its initial thrust into the housing market, the HDB actively planned and built housing blocks in "new towns", with a full range of urban services and infrastructure. During earlier phases of the project, one, two, and three room apartments were constructed, ranging in size from 23 to 53 square metres. During later phases, larger flats were introduced, largely to meet demand among middle-class residents. Initially offered only for rent, since the late 1960s, a home-purchase plan ha also been in effect, offering ownership to resident families at rates below real estate market prices £¨Wan, 1975£©.

Singapore's housing program has also traditionally included a strong social component. During the clearing and construction phases, residents of existing housing settlements were offered alternative accommodation. In addition, they were compensated for land which they had left behind. In 1964, the NHB paid our such compensation at a rate of ¡ç26.9 per square meter, an amount which has increased over the years. Residents also received a relocation allowance, and a rental subsidy of ¡ç15 per month for the initial three-year period of their tenancy £¨Wan, 1975£º13-14£©. In addition, at least 10 percent of space has typically been set side for small, labour -intensive businesses so residents can retain and expand commercial activities at home. Attractive landscaping and leisure space are also incorporated in every complex, helping provide a sense of "neighbourhood"£¨Yeh, 1975£º10£©.

By the time HDB's main construction initiatives were well underway and many were complete, resident satisfaction with Board efforts was already high. In a 1973 survey, As Yeh and Lee £¨1975£©report, well in excess of three-quarters of residents rated factors such as general living conditions, the social environment, access to shopping and services, and work as either satisfactory or acceptable, with just a minority of residents, a third or less, expressing dissatisfaction with children's playground facilities, noise, parking, and access to police.

FROM SINGAPORE TO SAO PAULO
During the Maluf administration£¨1992-1996£©, planners and administrators thought the Singapore model was especially applicable to Sao Paulo, owing to both the limited availability and high cost of urban land. Given the generally limited successes of previous public housing schemes, and the well publicized success of the Singapore strategy, a decision was taken to devote significant resources to a home -grown version of the project -Projeto Cingapura £¨PC£©. In 1994, some R¡ç7.5 million £¨then about US¡ç7.5 million£©was targeted for the PC, increasing dramatically to R¡ç6.75 in 1995, and R¡ç206.5 million in 1996. In 1997, the administration of Maluf's political prot¨¦g¨¦, Celso Pitta £¨1996-2000£©, planned to increase this by about half -to about R¡ç300 million -with supplementary funding from the Caixa Economica Federal £¨a government lending agency£©and the Interamerican Development Bank £¨SEHAB, 1997£©. If maintained at this level for the subsequent four years of the Pitta administration, or approximately 87 percent of the total 103,000 housing units promised by Maluf and Pitta, respectively, at the beginning of their terms £¨'Projeto Cingapura vai deixar,'2000£©.

In terns of operations, Sao Paulo's PC mirrored its better known Asian counterpart to a remarkable degree. Responsible to the municipal Housing Secretariat, day-to-day administration of the PC project was overseen by an appointed board of directors which did site selection and oversaw the tendering process for choosing construction £¨interview with PC Director, 1997£©.

Most apartment blocks were constructed in areas immediately adjacent to slum housing £¨typically favelas£©whose residents were to receive priority tenantship. Early buildings were typically low-rise, with larger buildings preferred as the project advanced. Apartment size has typically been about 40 to 50square metres, slightly less than Singapore residents got during early phases of the project. Once constructed, ownership of buildings passed to the municipal COHAB, which is responsible for collecting rents of R¡ç57.00 per month £¨plus a small condominium fee£©, guaranteed to tenants for a 25-year period. Although the Municipal Council passed a law allowing ownership of the units, to date none have actually been sold (interview with PC Director, 1997).

As in Singapore, a social component was been built into the project. Every new project has been assigned a social worker, whose initial task was to map the slum area and develop a list of residents. The social worker then assisted with the transfer of families to temporary settlements s the new buildings were constructed. Families also received information about the "dos"and "don'ts"of apartment living. Landscaping, sports fields, and leisure areas were also routinely incorporated in the layout of the development.

At the same time, in a major departure from the PE's progenitor, no provision was made for accommodation of small-scale businesses within the projects £¨interview with PC architect, 1997£©. This in turn has limited the ability of residents to earn income on site, as well s creating other difficulties.

OPPORTUNITIES AND CHALLENGES
In principle and structure, Projeto Cingapura has a good deal in common with its Singapore counterpart. It appeared poised to deliver in SO Paulo what so many proposals in the past has failed to do; affordable, quality housing for some of the poorest citizens. Residents of former slum areas would be provided quality housing near their former homes and at reasonable cost. Remaining in effect "on-site", they would stay close to their places of work in the city, in areas which most could scarcely afford through other means given the high cost of urban land.

Many residents of Sao Paulo's less affluent neighbourhoods appeared to welcome the initiative, even lobbying for subsequent phases of the project to be developed in their own areas. According to one respondent and former leader of a favela in neighbourhood known as Jardim do Arpoador, support for the PC was overwhelming. In an April 1995 plebiscite, 847 or 98 percent of residents voted to accept a proposal from the city to locate a PC apartment complex in their neighbourhood, with just 44 or 5 percent disapproving £¨interview with PC resident, 1997£©.

Those who benefited directly from the PC initiative have lauded the relatively short turn-around period between favela elimination, transfer to temporary housing, and construction of new buildings. For the Jardim Arpoador site, it was a mere 9 months from first announcement of project construction until the first families moved into their new units £¨interview with PC resident, 1997£©.

The "community"atmosphere envisioned by the project appeared also to be slowly taking shape, with residents increasingly organized to represent their needs, and leisure/sports areas developed on site for neighbourhood children. More than a few innovations designed to reduce costs for residents have also been implemented across the system. In 1996, the city announced that all buildings constructed from that point would include equipment for electricity form solar power, saving residents an estimated 45 percent on electric costs £¨'Cingapura ganha',1996£©.

At the same time, Projeto Cingapura has been beset be some serious challenges related to funding and unit cost, program management, resident quality of life and eligibility, costs of ownership, and local administration. Such problems are summarized below.

Funding issues
Program funding has remained perhaps the most serious challenge. Despite forecasts of significant financing £¨see above£©, the Maluf and Pitta administrations largely failed to apply the promised resources. Maluf, in particular, was stridently criticized in the public press for diverting some R¡ç1.7 billion destined for PC social services and housing to other purposes. Press reports claimed that this amount could have built over 100,00 public housing units, providing shelter for up to a quarter of the city's population in favelas £¨'Maluf tira,'1997£©.

Adding to this problem was the escalation of unit costs. Under previous regimes, the cost of housing units in Sao Paulo had been in the R¡ç15,000 range. During the Maluf and Pitta administrations, however, the cost of the PC units skyrocketed to nearly R¡ç25,000, further limiting the municipality's ability to produce the required number of units.

As a result of both factors, the numbers of PC units produced during the Maluf and Pitta administrations totaled only about 14,000. This is far from the nearly 100,000 projected, and only a fraction of that required to begin to deal seriously with Sao Paulo's housing deficit.

Program issues
These issues, widely reported in the local press, seriously affected the credibility of the program and its ability to attract funding from public and other sources. To begin with , there is evidence that the Maluf and Pitta administrations effectively used the PC to "play politics"as opposed to dealing directly with the housing needs of the poor. According to press reports, during early phases, the vast majority of PC apartment complexes were constructed in areas with high public visibility, thus serving political propaganda purposes. For example, it was revealed that of the first 41 complexes constructed, 16 were located near or on major roadways and clearly visible to passing motorists -by definition, middle -class voters. By contrast, in the area of the city with the most favelas, the Freguesia do O, not one unit had been built £¨'Freguesia do O.'1996£©.

The tendering process used to select construction companies for individual housing contain serious flaws, resulting in favoritism and potential corruption. In one widely reported case, a construction company which lost a bid publicly complained that the Housing Secretariat had illegally required that the successful company must be conversant with the objectives and requirements of one of the project's major and more recent funders -the Interamerican Development Bank. The company said this requirement favored larger development concerns with which the municipal government was already working £¨'Financiador diz,'1997£©.

Claims also arose that, during the early years of the PC, the Maluf administration turned a blind eye to developer attempts to save money and increase profits by using substandard construction materials and practices. In several cases, it was revealed that in place of solid concrete posts required by municipal building standards, some construction companies used hollow blocks filled with concrete. Construction companies argued that this should not pose undue risks, but the safety of residents was clearly put at risk by such illegal practices £¨'Prefeitura nao ve rescos,'1996£©.

Quality of life issues
As mentioned previously, levels of resident satisfaction in Singapore's various housing complexes were uniformly high during the mid-term phase of the urban renewal process in the mid-1970s. In Sao Paulo, however, once many buildings were occupied, residents began to identify serious quality of life issues. Living space was widely seen as inadequate, with one respondent lamenting that the units are "just the size of a matchbox"£¨interview with PC resident, 1997£©. At just 40 to 50 square metres, there is little room in the average PC apartment for a family of four. Some have even stated publicly that the PC units, with generally cramped quarters, "are worse than in the favela; one is piled on top of the other; it's a huge vertical favela"£¨'Euforia na favela,'2001£©.

Living in such high-density settle ments was a novelty for many residents used to more informal living arrangements. As social worker Eliete Barbosa explained, almost everyone had "to be taught the basics of apartment living; what to do with garbage, how to operate equipment, the responsibilities and duties of apartment dwellers"£¨interview with PC social worker, 1997£©. While the learning curve may accelerate ad former favela residents quickly occupy their new dwellings, conflicts inevitably have resulted, causing strained relations between neighgours.

For example, many residents were not aware of restrictions on keeping animals in the apartment complexes. Small animals such as dogs or cats are allowed. But goats or other large animals which they had raised in their former dwellings, sometimes to earn income £¨interview with PC social worker, 1997£©. This creates problems for other residents who must share their limited space with livestock, while removing animals raises the ire of owners who may be deprived of a principal source of livelihood.

Eligibility
Tensions within the PC apartment complexes have also arisen as a result of eligibility rules. Prospective residents have to produce evidence of a minimum monthly income of approximately R¡ç200 and register each family member £¨including children£©prior to assuming occupancy. Single men have not been allowed to occupy units, nor individuals with criminal records. In the early stages, however, such rules were not always communicated clearly to prospective residents, thus leading to disqualification of individuals who then had to seek accommodation elsewhere in the city, typically in another fafvela £¨interview with PC resident, 1997£©.

In some cases, friends or relatives moved in with tenants, who in turn then left the complex. This broke the rule that only the original favelados may occ8upy units, but little had been done to rectify the situation. In some cases, residents reportedly "sold"their apartments, though no such arrangement has been mandated, leading to a "black market "in public housing. In the end this may have hurt some poverty -stricken individuals most in need of subsidized housing £¨interview with PC social worker, 1997£©.

Another more generic issue related to eligibility, according to one respondent, is that resentment against those favelados lucky enough to obtain PC housing has grown within the surrounding communities. Many believe the favelados have been given "free"housing, which is not the case. Even those who know the costs of residency in the projects have felt that many favelados were "jumping the queue." The sense of "injustice"has been particularly acute for the so-called working poor who own their land and have managed to build rudimentary dwellings on their own over a period of years. Why bother trying to get ahead, many have felt, when one is more likely to get a brand new apartment by dong little or nothing. Such sentiment has obviously not created a positive climate between PC residents and their neighbours still struggling to make a life for themselves £¨interview with PC resident, 1997£©.

Financial issues and ownership
Up-front costs of apartment rental have been modest at R¡ç57 per month, but other costs have arisen for residents on tight budgets. For example, residents have to pay a "condominium"fee of up to R¡ç15 per month, as well all electricity charges. For low-income residents, such charges have left little for food, transportation, clothing, and other necessities of life.

While the rent charge is guaranteed for 25years, no such predictability exists making residents worry about new charges to come. As one explained, "they don't charge for water yet, but they likely will. Also, we may have to soon"£¨interview with PC resident, 1997£©.

In some complexes, resident concerns over rising costs have been translated into action. In April 2000, PC residents blocked a nearby arterial road to protest city plans to evict residents who were in arrears on rent £¨'Prefeitura ameaca', 2000£©. By August, nearly 58 per cent of all PC residents had failed to pay rent on time, some accumulating substantial back payments £¨'Devedores do Cingapura,'2000£©.

Such financial problems have been exacerbated by restrictions on commercial activities in the complexes. In the Singapore model, labour -intensive businesses were not only allowed but encouraged, and spaces for these were incorporated in the infrastructure of apartment blocs. In Sao Paulo, this was not done, in some cases eliminating possibilities for residents to continue earning income from self-employment £¨interview with PC architect, 1997£©.

The inability of residents to purchase their units may also be a factor. Although allowed by municipal law for some time now £¨Interview with PC Social Worker, 1997£©, no units have to date been offered for sale. Selling their units may be the only option the poorest families have to pay back rent owed and to attempt to create a new life elsewhere. But the city is caught on the horns of a dilemma. Any sales scheme could easily set up a speculative market situation that prices most poor families out of the running. Controlling prices is an option, but could be easily skirted though extra-legal "turn-key"arrangements between sellers and buyers.

Management issues
As previously mentioned, once construction was completed, ownership of PC buildings was transferred to a COHAB. Under the original plan, management was to fall to local resident associations, which would function much like condominium boards. Such boards would be responsible for day-to -day operations of the facility £¨e.g. maintenance and repair, common utilities, etc.£©and for resolving tenant disputes £¨interview with PC social worker, 1997£©.

This democratic model of governance would appear to be the exception rather than the rule. One building resident reported that the building committees were hardly democratic and, where existent, were largely appointed by the social worker assigned to the project £¨interview with PC resident, 1997£©. Social workers also reportedly were authoritarian about maintaining order, threatening to evict tenants who did not obey building regulations or engaged in disruptive behaviour. "it is the social worker who rules."claimed the respondent, with some decisions not made in the best interests of the tenants. According to the respondent, the social worker has been known to make exceptions with respect to tenancy. "The rules state that new people [i.e. people not living in the local favela are not supposed to be allowed in, but sometimes the social worker would allow them"£¨interview with PC resident, 1997£©.

Whatever management structure has been in place has been inadequate to deal with resident concerns. In several cases, residents have felt it necessary to resort to demonstrations and acts of civil disobedience to press their claims. In April of 1998, for example, approximately 500 residents of a PC apartment block in a northern neighbourhood of Sao Paulo blocked an arterial road to protest the lack of security, infrastructure £¨e.g. street lighting£©and regular bus service in their complex. In another neighbourhood in the western region of the city , a similar protest was mounted, again focusing on the lack of policing in the area. In both cases, vehicular traffic was severely impeded, and the protests drew strong and at times violent reaction from local police £¨'Moradores de Cingapura,'1998£©.

In August of 2000, similar protests were mounted across the city by residents who were in arrears on their monthly payments. They were reacting to letter from the Secretariat of Housing requiring them to pay the rent owed or face eviction. Rather than deal with these issues through local building committees, residents in at least four apartment complexes took to the streets "to burn tires, throw stones, and interrupt traffic." The protests only abated once the Secretariat agreed to review policy and procedures and undertake closer study of residents in difficult financial conditions £¨'Prefeitura ameaca,'2000£»'Para moradores,'2000£» 'Para governo,'2000£©.

CONCLUSION
While massive public housing construction by the Singapore government during the past four decades has almost uniformly been judged a success, the same can only be partially said of Sao Paulo's Projeto Cingapura. In theory, the project held out a great deal of promise, with strong potential to answer a critical need in a city with a large housing deficit and where the cost of urban land has skyrocketed in recent years. In reality, however, the funding applied to the project by the Maluf and Pitta administrations has fallen well short of projections- at one point in the hundreds of millions of dollars- as has the number of units constructed. Of the tens of thousands projected, in fact, less than 14,000 were delivered, and at a cost per unit far higher than other, more traditional housing strategies. Added to this are some very real problems in the project, from the tendering process and the quality of construction, to resident quality of life, tenancy and ownership, and management of the complexes.

It perhaps should come as no surprise that with the election of socialist mayor Marta Suplicy in 2000, Sao Paulo's housing strategy has taken a new turn. While vowing to complete approximately 1,000 unfinished Cingapura units, the municipal administration has formally announced that the "Projeto Cingapura will no longer be a municipal priority"£¨Projeto Cingapura vai deixar,'2000£©. Promising to apply over ¡çR3 billion to housing during its term in office, the Suplicy administration will focus instead on a housing strategy designed on a housing strategy designed to obtain maximum impact for minimum cost, and to involve project beneficiaties directly vious administration of PT mayor Luisa Erundina, the mutirao has become the cornerstone of Sao Paulo's housing strategy, with financial assistance to families in self =construction or upgrading of their own homes. Such assistance is estimated at between ¡çR11,000 and 15,000 per unit, as opposed to the per unit cost of a Cingapura unit in excess of ¡çR20,000 £¨Projeto Cingapura vai deixar, 2001£©.

Whether or not the Suplicy government is able to produce the 50,000 units it is now promising remains to be seen. What is more certain is that any pretentions which Sao Paulo city hall may have had in the past to transform South America's largest metropolis into the mirror image of Singapore have largely been laid to rest.

Do the problems faced by the PC in Sao Paulo imply that the potential benefits of south-south or lateral information and technology transfer in the housing area are limited? Not necessarily. But for such large-scale experiments such as this to work, a number of supports must be in place, not least of all sufficient public financing -always scarce in the developing world. Moreover, municipalities seeking to adapt innovations locally must ensure that management supports are clearly in place, and that attention is paid to critical factors including project integrity £¨especially transparency and accountability£©, resident quality of life, eligibility, and ability to pay , and local management capacity.