Housing public-private partnerships in Latin America: lessons from Mexico and elsewhere

In the early 1980s, Latin American governments abandoned direct production of public housing in favour of the enabling housing markets approach. They heavily subsidized private developers and mortgage lenders to go down-market in an attempt to reach low- and medium-income households. But, as illustrated by the case of Mexico, this approach has performed poorly. A new housing entrepreneurship, led by innovative public-private partnerships, is now emerging to satisfy the mass market. National and state governments in Latin America first responded to the challenge of accelerating urbanisation and growing slums by developing housing units and financing these projects directly. Large government agencies, often called Institutes of Housing,¹ contracted private developers to build these units. National governments mobilised funds for financing these projects through mandatory salary taxes on formally-employed workers. The housing finance agencies often doubled as pension/social security schemes to ensure that long-term savings would match fund long-term investments for mortgages. However, Latin America's public housing systems produced relatively few units of inconsistent quality, often in poor locations, frequently lacking basic urban services and land title. They were also highly expensive as builders had little incentive to keep their costs down. Governments typically contracted private builders with cost-plus-inflation pricing to carry out these projects. Patronage and political affiliation heavily determined the choice of builders and beneficiary households. Families paid back very little of the true costs of production of these units. As public housing production fell further and further behind demand and slum formation, government housing agencies turned to the less expensive option of upgrading slums, and site-and-services projects.

The shift to enabling housing markets

These serious drawbacks, as well as the global shift to markets in the early 1980s, persuaded major donors and Latin American governments to abandon the direct production of public housing in favour of enabling housing markets. The seven main enabling instruments were outlined in a key World Bank document (Mayo and Angel, 1993). It recommended strengthening legal property rights, developing mortgage finance and rationalising housing subsidies. It also called for the provision of infrastructure for residential land development, the regulation of land and housing development, and better organisation of the building industry. Finally, it recommended the creation of a national institutional framework for managing the housing sector. The enabling strategy has had an uneven impact. Most Latin American governments shifted from mortgages at below market rates to more transparent and quantifiable direct demand subsidies (DDS) for households. This finance method, consisting of three parts (ABC),²  typically gives households a substantial grant to which families must add a market-rate mortgage and their own down payment to purchase a new developer-built unit. The idea is that the purchasing family receives a voucher, which the household uses to choose a home from a wide variety of houses and units, rather than specific projects sponsored by builders or housing agencies. The success of DDS in Chile has influenced many other governments in the region to attempt this approach. The integrity and capacity of the Chilean bureaucracy and its fiscal commitment to the subsidy has made this administratively complex approach work. Most other Latin American governments, however, lacked these strengths and fiscal commitment. Many took the easy way out and allocated funds to projects sponsored by homebuilders and financial institutions, rather than to households, while adopting the three-part format of DDS programme. In short, the selected developers and mortgage lenders ended up with the ability not only to choose the location and type of homes but also the buyers who would get the finished homes. This supply-side decision making lies at the root of the problematic habitats created by government housing programmes attempting to go down-market in much of Latin America: standardised, miniscule units packed into sprawling developments on distant urban fringes or slums largely neglected by formal-sector institutions. Furthermore, this approach has allowed large homebuilding firms and the mortgage finance industry to benefit from substantial subsidies. It has mainly worked for the upper middle-class, spreading traditional mortgage finance to the elite worldwide.  But it has performed poorly in reaching most households and funding most housing investment, contributing to bipolar habitats in much of Latin America. With few exceptions, subsidised developer-built housing in Latin America has been affordable to only the top 20–40% of the income pyramid in the form of new owner-occupied units (Ruprah, 2010). Institutionally, the enabling housing markets approach has created a boom in secondary housing finance liquidity facilities  to raise and lend money at market rates. Nevertheless, many provident housing funds and Institutes of Housing³ dating from the public housing era continue to exist, but have changed function or clientele, or shrunk in size and importance. Institutes of Housing often went bankrupt because a large share of their loans went uncollected, as households did not take repaying these public sector agencies seriously.

The case of Mexico, 2000-2012

Mexico embraced the enabling housing markets approach at the start of the administration of President Fox in 2000, instituting an array of housing market reforms. In short order, the Mexican Federal Government created a National Housing Commission (CONAFOVI) to coordinate housing-sector efforts and a secondary-market housing finance liquidity facility, the Federal Mortgage Society (SHF). It began a thorough transformation of Mexico's ailing National Workers Housing Fund Institute (INFONAVIT),4  which receives 5% of all private-sector formally-employed workers' salaries and provides them with a range of housing-related mortgage products – making it the largest public mortgage bank.  An industry of 14 niche mortgage lenders, initially called SOFOLES, emerged to channel mortgage finance on market terms raised largely by SHF. Both INFONAVIT and SHF have increasingly adopted the three-part format of DDS but have retained supply-side decision making. These institutions and Mexico's top six commercial developers (Consorcio ARA, Casas GEO, Consorcio Hogar, Homex, SARE, Urbi) ended up controlling Mexican federal housing. In contrast, Mexican households have had little choice but to accept the outcomes. To benefit from their INFONAVIT account and purchase a house, they had to accept the project conditions: the type of house, the location and a burdensome mortgage.

These measures underlay a dramatic increase in the number of mortgage loans from around 350,000 a year in 2000 to 700,000 a year at the end of President Fox's term in 2006, almost equaling the rate of new household formation, a rough proxy for demand in Mexico (Figure). The huge increase in mortgage credits for new developer-built units at first appeared a great success. But, aside from the 40% of Mexico's population that is excluded from INFONAVIT mortgages because it is informally employed, cracks soon appeared in this façade (Memoria, Congreso Nacional de Vivienda, 2013). More than two-thirds of mortgage credits went to fund Mexico's largest developers to build huge sub-divisions consisting of miniscule core expandable units on distant urban fringes, with 31 municipalities each having at least one new housing project of more than 25,000 new units. First, this rapid development created low-density urban sprawl far from jobs and services. And for a variety of reasons, such as the high cost of transport, lack of basic services, or social isolation, many Mexican households – one in five – have left their properties either vacant or temporarily occupied. Developers for their part were left with swathes of housing units they could not sell. Secondly, land oligopolies have increasingly excluded the lowest earners and local government. Mexico's large developers started to stockpile land far in advance of use largely by taking advantage of the country's opaque ejido communal rural land-ownership system, which now includes roughly 50% of developable parcels on the fringe of major urban areas – dramatically reducing access to homeownership by bottom half effectively excluded from formal finance and development. Finally, although this new system of mortgage banking appeared to work well from 2000 to 2007, the financial crisis of 2008 resulted in escalating arrears rates and bankruptcy of many of the SOFOLES. Since 2007, Mexico has begun to recognize these drawbacks and shift from homebuilding on the fringe to urban densification, high-rise building and better regulation of the rental sector.

Shaping the market for the majority: the new housing entrepreneurship

A third paradigm is now emerging for habitat and urban development in much of Latin America (Ferguson et al., 2014). Rising real household incomes have made affordable housing a huge new market opportunity for the private sector. The new urban lower-middle class currently numbers approximately 2 billion people and is projected to grow to 4.9 billion of a total world population of 8 billion by 2030. This group both demands and can afford a market-rate home, as can many low-income families. The affordable housing market, however, consists of many distinct segments, reflecting the great variety of urban conditions, incomes, and life circumstances of the Base of the Pyramid (BoP). The public sector and the private sector must collaborate to create viable products for these many segments. Large modern corporations have started to view affordable housing in dynamic emerging economies as a business worth pursuing. Corporate building-materials manufacturers, retail chains and the new breed of affordable-housing developers offer the most powerful institutional platform with the greatest ability to reach mass markets. Modern management methods, particularly a value-chain approach, appear eminently well-suited to organising, streamlining and squeezing the costs of the incremental housing process used by most of the low-to-middle -income majority (Ferguson, 2008; Schmidt and Budinich, 2006). Mexico has some iconic examples of this new housing entrepreneurship (Box 1).

This first generation of initiatives for the base of the income pyramid provides key insights for expanding market-based affordable-housing finance and development (Samaranayake et al., 2011). They show that market-based affordable housing is a radically different business from traditional mortgage finance of developer-built new units. Seizing this opportunity requires careful market studies followed by the creation of pilot projects before expanding to scale. New companies should be created to carry out the process, as opposed to it being done by departments of existing companies, and conventional business models need to be avoided. Another lesson learnt is that stakeholders must work together to package the products and services suited to the diverse segments of the base of the income pyramid. Typically, the lead organization must partner with others to acquire missing pieces of the product package, including citizen-sector organisations to market the product, screen households and troubleshoot. In this context, the single most important role of government is the provision of serviced urban land. A successful example of this can be seen in Thailand (Box 2). Local governments usually hold many of the key levers to the price and accessibility of urban land. Balancing the interests of peasant landowners on the fringe, urban householders, the homebuilding industry and government itself – by recouping the public investment in urbanisation – requires a strong multi-pronged urban land strategy that is currently missing in most of Latin America. The key to success is providing low-to-moderate-income households with the voice, the choice and the information necessary to shape their habitat. The way forward involves developing a public-private institutional infrastructure capable of producing the diverse housing solutions suited to the many submarkets of the base of the income pyramid.

1 Institutos de Vivienda in Spanish

2 In Spanish, these direct demand subsidies were often called ABC programmes, named after the Spanish initials for household savings (Ahorro), the subsidy (Bono) and a loan (Credito).

3 Second-tier housing finance involves funding, guaranteeing and packaging into securities mortgage loans originated by first- tier lenders directly to households. 

4 As of 2000, INFONAVIT had over a 40% arrears rate on its mortgage loans and failed to collect much of the “salary tax mandated by national law from many formal-sector employees and employers.

References

Ferguson, B., Smets, P. and Mason, D., 2014 (forthcoming). The New Political Economy of Affordable Housing Finance and Urban Development in Bredenoord, J., Van Lindert, P. and Smets, P. (eds) Affordable Housing in the Urban Global South. Seeking Sustainable Solutions. Routledge/Earthscan, London, UK. // Ferguson, B., 2008. A value chain framework for affordable housing in emerging countries, Global Urban Development Magazine, November 2008. // Mayo, S.K., Angel, S., 1993. Housing Enabling Markets to Work. World Bank, Washington, D.C. USA. // Memoria. Congreso Nacional de Vivienda. 2013. Universidad Autonoma de Mexico. Mexico. // Orvañanos, L., Ahumada, G., 2011. Investing in Mexico: The Homebuilder´s Perspective. Home Builder = Nation Builder. Mexican Housing Day. February 10th, 2011-New York. February 14th, 2011-London. // Ruprah, I.J., 2010. Do Social Housing Programs Increase Poverty? An Empirical Analysis of Shelter Induced Poverty in Latin America. Evaluation Department of the Inter-American Development Bank, Washington, D.C. USA. // Samaranayake, S., Budinich, V., Kayser, O., 2011. Access to housing at the base of the Pyramid. Ashoka, Arlington, Virginia, USA. // Schmidt, S., Budinich, V., 2006. Housing the Poor by Engaging the Private and Citizen Sectors: Social Innovations and “Hybrid Value Chains”. Ashoka, Arlington, Virginia, USA.