Mexico City, Buenos Aires, development, rental units, rental housing, Latin America, Bogota, Development Bank, Working Paper, World Bank, households, rental development, urban land, Mexico, density, rental sector, Bouillon, Housing Market, landlord rights, Xavier de Souza, Working Paper Series, Housing Finance Information Network, Bouillon Cesar P., Cesar P. Bouillon, Environmental Modeling and Assessment, Palgrave Macmillan, Cecilia de Mendoza, Statement Gilbert, Nadin Medellin, Journal of Planning Literature, Brookings Institution Press, Distinctiveness of Comparative Social Science, sustainability indicators, Evridiki Tsounta, Marja C. Subsidizing Housing Finance, Oscar Olinto Camacho, public housing, Lincoln Institute of Land Policy, David Dowall, Latin America and the Caribbean, Buenos Aires Mexico, urban density, residential development, Conditioned Urban Development Zones, land regulations, affordable rental housing, rental projects, Rental income, subsidized housing developments, urban areas, income differences, urban planning, income quintiles, poor housing market, rent control, Cynthia Boruchowicz, household income, Mexico City Recife Santiago, informal settlements, International Monetary Fund
Using Evidence-Based Global Housing Indicators for Policy Evaluation of Rental Housing and Vacant Properties Authors: Bernadette Baird-Zars, Alarife Urban Associates Jane Katz, GHI Program Manager and Director of International Affairs
& Programs, Habitat for Humanity International Cйsar P. Bouillon, Lead Research Economist, Research Department, Inter-American Development Bank Ophelie Chevalier, Senior Housing and Urban Development Specialist, Inter-American Development Bank Maria Luisa Alvarado-Zanelli, Advocacy/Partnership Outreach Manager, Habitat for Humanity International, Latin America
n and the Caribbean Region Naveen Jawaid, Independent Researcher Contributor: Joyeta Goswami Abstract: Secure, formal, and affordable rental housing is a crucial component of healthy housing market
s, especially for the urban poor. In Latin America, the rental housing sector has remained underdeveloped, in part due to decades of homeownership-focused regulations and subsidies. Using a comparative policy framework, this paper provides an initial snapshot of the enabling environment for rental in five major Latin American cities profiled in the recently-launched Global Housing Policy Indicators platform. Analysis of the indicators, situated in additional city-specific studies and data, confirms a persistent policy bias towards new homeownership in the subsidy programs, regulations, and institutions impacting housing in the five cities. All five rental markets suffer from persistent gaps of secure and formal rental options, particularly for low-income households and vulnerable populations. As the need for affordable and accessible housing becomes increasingly acknowledged and prioritized by national policy-makers, standardized global policy indicators can provide a critical `wide-lens' approach that, grounded in localized market and case-study findings, can pinpoint key subsidy and regulatory changes to enable sector growth. Further, new tools, such as developing centrally-located vacant or underutilized housing, buildings and land into healthy rental units, could unlock immediate opportunities to decrease housing deficits. Abstract: 195 words Paper: 7530 words 1
Introduction: Utilizing policy indicators to understand comparative housing markets Robust housing indicators can illuminate policy biases and key regional trends that impact urban housing outcomes. Housing, and particularly rental housing, is an under-measured sector of significant public investment, yet is a critical metric for successful cities. Residential space makes up around three-quarters of the urban built environment and is a key determinant of quality of life and access to opportunity for low-income households. Existing institutionalized indicators used to assess the health of the housing sector in Latin America are generally constrained to quantitative measurements of the `housing deficit': infrequent national censuses and city-level ad-hoc surveys provide data on overcrowding, the physical conditions of the dwelling, and access to basic services. While useful, these measurements leave out the systemic context of policy, finance, development, management, and planning needed to enable a healthy and equitable ecosystem of urban housing. This paper utilizes a new housing indicator platform to examine the enabling environment for rental housing in five large cities in Latin America. It provides a preliminary example of how policy indicator framework, when combined with local-level data and case studies, can identify systemically underdeveloped sectors of housing markets, such as rental housing. We demonstrate that, because rental has been so underdeveloped in key Latin American housing markets, policy tools can be better leveraged in order to provide greater opportunity for secure tenure for low-income communities. Growing momentum over the last two decades among policy makers and the development sector for systemic housing indicators at a global level has recently come to fruition with increased institutional engagement.1,2 Habitat for Humanity International (HFHI) led the development of the nearly 200 Global Housing Policy Indicators (GHI) and commissioned surveys for an initial set of 24 cities between 2011 and 2012.3 More recently, over the last two years, a new cross-institutional working group, including members from the Inter-American Development Bank (IADB), Habitat for Humanity, United Nations human settlements
Programme (UN-HABITAT), the U.S. Department of Housing and Urban Development (HUD) and the World Bank, was formed to further coordinate housing indicators and create inter-connected platforms. In Latin America, national policy makers and research institutions formed a GHI-affiliated working group to deepen regional coordination around indicators. The analysis was commissioned by HFHI, and developed in close coordination with the urban and research departments at the IADB as a trial regional policy analysis of the new indicators data. Rental housing was selected as the first topic, as it is an area of current interest among policy makers and the regional working group, with a sample group of the five largest cities surveyed by GHI in Latin America between 2011 and 2012. GHI indicators are complemented with place-specific practices and additional 1 Early prototypes for a broader set of housing policy indicators were discussed at the World Bank and developed for and piloted at the IADB for a set of Latin American countries. A major housing finance indicator platform, the Housing Finance Information Network (HOFINET), a non-profit based out of the Wharton School, launched in 2008, and now collects around 220 indictors for close to 100 countries. 2 The nearly 200 indicators in GHI assessments were developed in a multi-year process with more than 50 field experts, and are sorted into five sections: property rights
, Housing Finance, Housing Subsidies, Residential infrastructure, and the Regulatory Environment. 3 Full data is hosted at www.globalhousingindicators.org. The GHI "aims to enable rapid identification of regional and sub-national trends and policy gaps in housing for national-level policy makers, municipal actors, and international donors." In theory, identifying potential policy bias and trends can catalyze well-targeted intervention across the market. The GHI indicators and their platform were designed to provide structure, but also to evolve and network with other institutions. 2
contextual data from various global data resources including the IADB, the Housing Finance Information Network (HOFINET), the Global Property Guide, and the World Bank. Additional primary resources and specific case study knowledge was contributed by country-level experts who have collaborated on GHI.4 Using a comparative policy framework at the sub-national level, the paper compares city-level data by analyzing indicators with a case-study approach (Snyder 2001).5 The comparative methodology focuses on quantitative and qualitative strategies to compare how policy tools central to creating a more affordable rental housing market interact in each of the five cities of Bogotб, Colombia, Buenos Aires, Argentina, Mexico City, Mexico, Recife, Brazil, and Santiago, Chile (Ragin 1987, Graham and Marvin 2001).6
Methodologies to deploy housing policy indicators for city or region-specific analysis is an additional theme explored throughout this paper. Our analysis of the rental sector illustrates one of many ways in which indicators can be critically combined with other sources of knowledge to build a more holistic and nuanced picture of dynamic and complex metropolitan housing markets. In our framework, policy indicators provide a structure for our findings, which were developed after indicators were placed in a local context with city-specific data and cases. In this way, and similar to the concept of `situational indicators' used for environmental work,7 indicators were a tool that leveraged global expertise to identify potential areas of strength or weakness in local policies impacting housing market function (Gallopin 1996).8 After these areas were identified, metro-area census data, market data, and descriptive cases were needed to confirm local relevance, and later extrapolate to meaningful regional trends.
Rental in the context of housing markets in Latin America
Rental housing has been persistently left out of housing subsidy and regulatory development in favor of
new homeownership. The last three decades of housing policy in Latin America has been shaped by the
view that homeownership is the best pathway to provide low- Percentage of units that are rental income families with a stable and secure base for wealth creation
(Gilbert 2011). While homeownership can do all of those things, Bogotб
there has been an increasing realization among experts, housing Buenos Aires 11%
practitioners, and policy makers that robust housing markets can, Mexico City 22%
and should, also deliver alternative tenure types that are secure Recife
and provide opportunities for asset-building.9 Rental housing has Santiago
been established as critical to a healthy housing market that
Source: GHI 2010-11
offers access to shelter, accessibility to jobs and services, and
4 Particular thanks to Jorge Torres Ramнrez, the executive director of the Center for Construction and Urban and Regional Development Research (CENAC) in Bogota. 5 This paper utilizes city-level GHI Indicators in order to expose a more varied experience and depth of analysis of regional housing trends. By focusing on these five heterogeneous cities, the paper exposes a more nuanced and complex picture of their respective rental housing markets by shadowing a "small-N research design
." 6 Please note that while other sectors of a healthy housing market are critical to the holistic study of a city housing rental market, this paper is structured around a policy framework and will not address other components that may or may not affect the rental housing discussion, such as real estate economics or land regulations. 7 Described well by Gallopin, 1996, in the context of the groundswell in interest for environmental indicators. 8 The role of indicators identifying potential areas became an important distinction for this paper, as even good indicators cannot be used to make value judgments. 9 See Angel 2000 for an overview with Latin American context, Retinsas and Belsky 2002 for a US-based overview, Briggs 2006 for an analysis of mobility benefits on households, and many more describing new innovations in limited equity mechanisms between rental and ownership, including co-ops, (Saegart and Benitez 2005), and land trusts (Abramowicz 1991). 3
good equitable urban development.10 Further, secure rental housing has been acknowledged to more closely fit the preferences of many households, including many of the poorest and most vulnerable. Rental can no longer be cast as a `second-tier' tenure type, but a priority investment for cities that hope to become inclusive hubs of growth.11 The current status of the housing market supports the argument that decades of skewed public sector support for homeownership effectively disincentivized all other tenure types for households and developers.12 Between 2009-11, renters made up a mere 17 percent of households across Latin America, two thirds of the Euro area average and indeed lower than nearly every other region in the world13,14. Rental rates in the five major cities studied range widely, from a low of 11% in Buenos Aires to a high of 41% in Bogota, but remain far lower than those of their counterparts in Africa, Asia and Europe. The resurgent interest in rental is part of a larger acknowledgement of persistent regional `deficits' for all types of housing. All five cities have high percentages of households who live with a qualitative or quantitative housing deficit (Figure 1).15 While large, these number understate the true shortages, as the measures of qualitative and quantitative deficits omit two critical drivers of household demand: connectedness to infrastructure, amenities and jobs, and tenure flexibility. To capture the economic benefits of cities, poor households need access to opportunity, a large component of which are jobs that are increasingly mobile and urban (as opposed to fixed-place manufacturing often with greater job stability).16 While greater Mexico City has a formal deficit of only 34,000 units yearly, there is a city-wide shortage of affordable housing with flexible tenure rental or otherwise, in areas close to employment; large developments of new-build units that are yet to be connected to schools and transit sit vacant on the edges of cities. Income and access to finance further compounds the deficit. Close to half of families in the five cities cannot afford or lack access to credit to buy a formal house 10 See Gilbert 2001, Blanco et al 2012. The forthcoming book by Peppercorn and Taffin will likely also be a valuable contribution. 11 While examined here primarily as a means to increase access to secure, accessible and affordable units for low and middle-income households, a formal stock of affordable rental housing is also correlated with growth of new industry and attracting high-skilled mobile workers. See analyses of mobility and quicker labor market adjustments in Belsky and Drew 2007 12 As many have noted, in a well-functioning market, theoretically both producers and consumers of housing would be ideally indifferent to such a choice with elastic supply (Shlomo 2000). 13 Bouillon et al. 2012. 14 Blanco et al 2012; Jha 2007 15 Quantitative housing deficit is defined as either a temporary housing solution or overcrowding, and qualitative housing deficits as 16 See Oswald 1996, regarding ownership and increased exposure to income shocks and unemployment. 4
(Bouillon et. al, 2012).17 Figure 2 displays the total affordability gap, including lack of affordability due to low income, shelter-driven poverty, house price, interest rate
s, and a range of factors preventing access to a formal mortgage, such as inadequate income documentation. The lack of affordability is expected to grow, compounded by the effects of urbanization, demographic trends toward smaller household sizes and employment shifts toward centrally-located service-sector jobs. These factors will further increase the demand for more affordable and accessible urban housing.18
Undesirable urban outcomes such as informal land and housing development and large quantities of underutilized vacant housing are closely related to poor housing market function. Faced with few alternatives, opting for the informal self-built housing market has been a defining strategy of low-income households and self-employed households. Historically, rates of homeownership rose during Latin America's rapid urbanization, even though the absolute number of tenants rarely fell (Gilbert 2011).
Bogotб Buenos Aires Mexico City Recife Santiago
Number of households living in informal settlements in the greater municipality. 1736519 20000000 275720 102392 4722
Number of households living in informal settlements in the greater municipality who have titles provided by a regularization program. 22000 135983 20000 105957 Source: GHI surveys 2011-2012
Rental housing appears to be a preferred good in all cities except for Bogota, where the bottom four quintiles of households income all have similarly high percentages of renters. Minor variances appear by city. Mexico City has a `u-shaped' distribution where the poor and the wealthy are both more likely to rent than middle-income households. In the two cities of the southern cone, Buenos Aires and Santiago, renting is most common among lower-middle class households and even more so for households in the top two income quintiles. Finally, a direct correlation between income and rental rates exists across all household income categories in Recife.
Given the preference for rental, significant pent-up demand for rental units is especially likely among lower-income urban populations, who either cannot afford or do not qualify for pathways to formal homeownership due to income informality. More
17 As calculated in Bouillon et al., 2012, defining formal housing affordability as the "capacity of a household to buy outright or qualify for a private mortgage loan for a house built using legal building codes on legally subdivided and serviced land." 18 Gilbert 2011; Bouillon 2012. Demographic trends include a growing aging population, increased divorce rates, and growing personal income. 19 69,461 people live in diagnosed settlements or under a legalization development process as of September 2011. The average size of households identified in informal city neighborhoods is 4.0 20 This number does not include the indirect subsidies in Mexico, which cover over 80% of the formal housing market. 5
importantly, key determinants of renter status across the region highlight preferences and characteristics that are on the rise in the urban population, including higher household inter or intra-urban mobility, female headship, shorter job tenure, migrants or emigrant status, elderly or other household members in need of supportive services, small household size, young households with small children and overall preferences for greater accessibility to opportunity, employment and urban amenities.21
Locational preferences are also a strong determinant of the demand for rental housing. Despite the unavailability of formal rental options, some low-income households will opt for informal housing closer to the urban center to have better access to jobs. A recent study in Santiago shows that the prevalence of slums in the city evidences a response to location preferences. Households are "more willing to bear the myriad costs of informality in order to have better access to employment, social services, and transportation."24 While none of the five cities evaluated here have any formal or legal restrictions on mobility, in practice the lack of tenure options often does restrict the ability of a household to relocate for employment or other reasons.25
Private development and finance for formal rental housing remains minimal in the region's cities, usually lacking subsidy assistance and suffering from higher perceived risks from less enabling regulatory and institutional frameworks. As a result, the rental housing supply market is roughly bifurcated. At one end of the market are high-end formal developments that serve the
Rental Yield for 75-120m2 apartments22
Mexico City Recife Santiago
8.8% 5.7%23 7.3% Source: Global Property Guide
mobile wealthy. Beyond these units, much of the current
supply of rental housing is provided by small-scale landlords and informal units (sometimes as simple as
an adjacent bedroom to a landlord's primary residence). These rental units are often rented to the most
vulnerable and sometimes lack basic services, urban amenities or easy access to public transport. Often
these informal units are under the authority of de-facto landlords trying to supplement their own income
(Gilbert et. al 1993). In Bogotб, for example, two-thirds of informal dwellers earn additional income by
renting their homes, and one third of units rent out individual rooms (Blanco et al 2012).
Private financing, either through bank loans or investor equity, is infrequently deployed for rental construction except at the top tier of the market. Rental yield comparable to rates in other countries, and is roughly equivalent to compared to standard rates of formal mortgage, except in Recife, where it is far below.
Mortgage lending available for rental
Yes Source: GHI 2010-2011
21 See studies on tenure choice and housing demand, including Arbelбez et al. 2011; Cristini, Moya, and Bermъdez 2011; Piza, Litchfield, and Balderrama 2011. 22 These rental yields are not based on your average square meter apartment for the majority of low-income urban dwellers; it is included here for comparisons to rates in other countries. 23 Based on national average for Brazil 24 Bouillon et al. 2012. 25 GHI indicator 6
Indicator-based analysis: subsidies, laws, institutions and regulations impacting rental
Creation and rehabilitation of formal, accessible, and serviced rental units is challenged by lack of support from the public sector in all five cities. Drawing on the GHI survey indicators to shed light on the broad trends across the policy landscape, the following sections will examine key policy barriers to rental development: (a) the subsidy skew towards homeownership, with few to none demand-side subsidies to enable households to access rental and no major formal programs to encourage supply-side delivery of rental units; (b) outdated and underdeveloped legal context for rental market that articulates a balance of landlord/tenant rights, including eviction processes, protection against discrimination, and phasing out rent control; and (c) planning and land-use regulations that discourages density and small-unit size essential for the low and medium rental market feasibility, as well as missing tools to capitalize on opportunities such as vacant lands, floor space, and units.
Demand-side subsidies almost exclusively facilitate homeownership. Specific subsidies to support households to acquire a rental unit or assist with payment of rents, leases, or security deposits, are few and far between. All five cities studied here, along with most of their counterparts in Latin America, participate in national direct transfer programs to assist low-income families with home acquisition, and three cities Bogotб, Mexico City, and Buenos Aires supplement this with their own municipal programs
. The GHI surveys for the five cities show that approximately US $6.7 billion is spent annually on direct subsidies for homeownership and no ongoing programs exist for the general low-income population to subsidize rental housing.26 Subsidy quantities here represent the yearly allocation from 2010 or 2011 for national-level subsidies, unless otherwise indicated.
Bogotб/Colombia Buenos Aires Mexico City (2012) Recife/Brazil Santiago & Chile national (2011)
Estimated share of the housing budget for direct and indirect subsidies27 39% 9% .014%29 17% 99%
Direct subsidy for housing ownership 396 million USD 454 million USD 38 million USD 5.5 billion USD 281 million USD
Direct subsidy for rental housing 0 28 5.5 million City level - homeless Some improvement assistance available for rental units, not quantified (program habitat) None mentioned None mentioned Source: GHI 2011-2012
Implicit subsidies on a national scale, such as those channeled through subsidized interest rates, often represent an even greater public investment. In Bogotб, Buenos Aires, and Santiago, lending is available but lines of credit generally exclude rental housing, and it is unclear how many loans are actually issued for the construction of rental units. In Recife, federal government
regulations have not allowed lending to rental developers. The major direct demand-side subsides described for each city in the GHI, are summarized below. Implicit subsidies are not described in this paper, but make up a significant portion of
26 A estimate summing the annual allocations (either 2010, 2010-11 or 2011) to the subsidy programs described in GHI's `subsidy worksheet'. 27 GHI defines this as: "Estimated percentage of housing budget at the national level that was given in rent vouchers, direct demand subsidies (housing allowances, vouchers or subsidies for a down payment), interest-rate subsidies, grants given directly to beneficiaries or slum-upgrading in low income communities, this year (%)" 28 6.2 billion one-time rental payments disaster assistance (national) 29 Excludes the major public-sector provident funds.
the subsidies in some countries, such as Mexico, where public provident funds provide low-cost loans to nearly 80% of the formal housing market.
While targeting and delivery mechanisms have improved significantly over the last decade in Latin America and are now cited as international models demand-side subsidies remain almost exclusively for households who own, or wish to own, their home. For example, Mexico's Esta es tu Casa, Columbia's MetroVivienda, and Brazil's Minha Casa Minha Vida represent new national housing programs with improved targeting and delivery systems. These programs, nevertheless, remain overwhelmingly designed for new homeownership.30 Still, rental is an emerging area of policy priority, illustrated by discussions and pilot programs on new rental subsidy programs and pilots in those three same countries: Colombia, Brazil and Mexico. For these subsidies to leverage private investment, the complementary policy, regulatory and institutional framework needs to be strengthened. This study hopes to be the first step in identifying trends in the policy gaps found across five cities and examining how subsidies, regulation, and land use planning interact with the persistent undersupply of rental housing.
When rental programs are included, they are largely framed as a one-time, temporary fix for vulnerable populations, not a sustainable solution. Some cities have developed temporary rental programs to serve especially vulnerable populations with immediate housing needs, such as the homeless, those displaced by violence, and households affected by natural disasters. However, in the specific case of Colombia, housing for displaced populations is provided through demand subsidies for ownership or fully subsidized public housing transferred in ownership.
Bogotб Buenos Aires Mexico City Recife Santiago
Example of assistance to vulnerable populations: percentage of housing agencies budget for the homeless
0% Not for homeless (offer some assistance to homes requiring special assistance)
3% (housing policy is disaggregated and this is found not in the "Housing Institute" but in the Ministry for
0.2 (not separate from other forms of social assistance)
10% (not specified for only homeless but more for at-risk populations)
0% Not for homeless
Source: GHI 2011-2012
Supply-side disincentives for developers and owners of rental units. Developers of low and mediumincome units in the cities studied face significant challenges to financial viability, especially for locations with basic services, social infrastructure, and accessibility to transport. These difficulties are compounded for potential developers of rental units. Long-term project financing is scarce or too expensive for rental developments. For example, new rental projects in Colombia face real interest rates in the range of 13% over for a 15-20 year term.31 Public subsidies for rental development are non-existent, the regulatory framework has no incentives for low-income rental development, and there is often bias against landlord rights. Planning and land use challenges pose particular barriers to development by continuing to discourage infill and density, two critical drivers of financial and project feasibility for rental developers. Taxes credits or deductions for parties in the rental sector are low to non-existent, and taxes on rental are high in all five cities. Despite the clear need for public-sector incentives for the supply of low and medium-income rental housing, owners of rental units in all of the five cities actually face significantly
30 For example, all direct subsidies are now means tested for income in the five cities 31 Buillion Buendia 2012, p 101. 8
more tax burden on income32 than those faced by property owners. This aligns with the regional trend of low property taxes and no tax on imputed rental income, further favoring homeownership.
Rental income tax
es (%)1 (National)
Can mortgage interest payments be deducted from income tax calculation? (Yes=1/No=0 or 100 if not applicable)
Bogotб, Colombia Buenos Aires Mexico City Recife, Brazil Santiago, Chile
15.47% 21.00% 3.36% 15.00% 17.00%
Source: GHI 2011-2012
Government-built public, or `social' housing in Latin America has largely been produced for transfer to ownership over the last several decades, with very little public housing for rental. In the five cities, none of the government's current public housing programs are directed towards rental. In Buenos Aires, for example, direct-build housing stock is transferred to the occupants, in theory absolving the public sector of any property management or upkeep responsibilities. The `hands-off' approach stems largely from experiences of ineffective maintenance of mainly-rental public housing units from the 1920s through the late 1960s. With public housing agencies attempting to operate as property management, housing stock suffered from poor maintenance, units became dilapidated, and the below-market rents often were captured by middle income households. Unlike other regions that have created partnerships with the private and non-profit sector for property management of government-supported rental, Latin American governments have, in general, shied away from rental in favor of ownership-intended social housing.
Supply-side housing subsidies have often favored large developers, which further prevents interaction with the rental sector: much of rental housing is developed by small and medium-sized firms, and in some cases, individuals. For example, the fact that the subsidy systems in Mexico City and Bogotб tend to work best with large developers has contributed to the popularity of large-scale peri-urban subsidized housing developments.
Tax Deductions for Constructing Affordable housing
Bogotб Buenos Aires Mexico, City Recife Santiago
Yes No. Some tax breaks exist for building in the South zone. No No Yes
(b) Legal and institutional framework
Despite significant impact on housing outcomes, the legal and institutional framework for housing is infrequently quantified systematically, especially for rental. The GHI admittedly focuses most of its regulatory indicators on ownership and, while the range of indicators showcased here vary tremendously in scope, they represent a modest sampling of the many which directly influence rental development. Two generalized assumptions are likely true, however, for all of the five cities. First, the legal framework for rental has received less modernization and attention over the last few decades, given the focus on homeownership. Second, the capacity of these five cities to clarify and enforce the rights of the engaged parties, contracts and disputes for rental housing lags far behind the relative clarity around laws for ownership development and transactions and foreclosure.
32 See, for example, "Rental income taxes are punitively high in Argentina". Global Property Guide, 2012. 9
Rent controls and restrictions are classic examples of the regulatory impact on rental markets. While frequently-cited as disincentives for investment in rental units in some cities in the Middle East and Asia, they make up only a fraction of the total units in the cities studied. Santiago, Chile, is the exception; nearly four out of ten units are rent restricted.
Bogotб Buenos Aires Mexico City Recife Santiago
Percentage of Rental restriction or rent Market rent/rent- The dismantling of rent control or
units under rent control is phased out
rent restriction on new rental units is
addressed in recent policy documents.
7 times N/A (no rent control) N/A N/A No rent control
Yes N/A N/A N/A No Source: GHI 2011-2012
The balance between tenant and landlord rights is crucial to enable effective and fair markets where owning and managing a rental property is a secure investment, as well as a stable and protected home for residents. That said, most rentals are usually informal and therefore have no legal contract. As a result, data on informal tenant and resident relationships continues to be scarce, especially on a regional scale.
Drawing from indicators collected by the Global Property Guide, which focuses on the formal sector only, the regulations of owner-tenant relations in the five countries of focus generally seem to favor tenants' rights over landlords' rights. With the exception of Argentina, the eviction process is rated as tedious, likely increasing costs for landlords and risks for investors.
Colombia (Bogotб) Argentina (Buenos Aires) Mexico (Mexico City) Brazil (Recife) Chile (Santiago)
Pro-Tenant or Pro-Landlord33 Neutral to Pro-Landlord market (unclear if this is focused on rental) Pro-Landlord rental market Pro-Tenant rental market Pro-Landlord rental market Neutral
Eviction for Non-Payment of Rent (Total Days)34 500 45 (need to check if there was a recent change in law (accelerated repossession regulations in recent years) in BA that allowed for landlords to recover property in 2-4 months as mentioned in RD [g. 99] 100 120 240 Source: Global Property Guide, 2013
Evictions are a costly and undesirable option for both parties, but represent a necessary last-resort measure to ensure the security of investment for rental properties. A regularized and respected process for evictions is critical to guarantee basic rights for evicted tenants. A formal and enforced legal framework for evictions is present in only three of the five cities. Insecurity in tenure remains significant across the low-income housing sector in Latin America, further disincentivizing development of formal rental solutions for small low-income landlords. Evictions in the informal sector remain a particular challenge for low-income residents. As briefly discussed above, often, rental solutions for the urban poor are found in the informal housing market. 33 Global Property Guide (accessed Feb. 8, 2013); Based on Analysis at the national level 34 This includes duration of completion of service of process, duration of trial, duration of enforcement.
Historically, and currently, slum upgrading remains a feature of many urban housing policies in Latin America, yet the variable impacts on renters or owners has not yet been clarified. Focus on homeownership-only and post-facto upgrading to date appears to correlate with the cycle of peri-urban, low-density, economically segregated areas. Rather than addressing the cause of slum development, most Latin American cities have programs to assist current households through titling slum upgrading, savings, subsidy and credit systems (ABC) or direct housing provision.
Bogotб Buenos Aires Mexico City Recife Santiago
Established legal guidelines followed for evictions No No
Notification period before eviction.
Involvement of violence in eviction
Less than 1 month No
Residents who were resettled or given compensation roughly equivalent to the value of their homes. (%)
N/A 0 (but some families were helped with a subsidy) 0
Source: GHI 2011-2012
As a policy intervention, slum upgrading has involved improving basic infrastructure in developments, regularizing tenure, or bringing slums key social services. In the long run, infrastructure and regularization have proven to be more expensive than providing upgrades through established land development process.35 These interventions can also lead to gentrification-like effects that can push out the populations that the investments were initially aimed for (Rojas 2010). To improve the longer-term efficiency of the housing market, government subsidies and development strategies that aim to increase supply of affordable urban land and incentivize densification should complement programs aimed at urbanizing existing slums. All five cities studied do have a government slum upgrading program that provide infrastructure improvements in low-income settlements, but many either have no programs or only incipient programs to increase the supply of low income rental solutions.
Bogotб Buenos Aires Mexico City Recife Santiago
Estimated share of the housing agency's budget dedicated to upgrading homes and infrastructure in low-income communities 7 17 .007 3 31
Slum clearance and redevelopment is an established policy of the national or local housing agency. Y N N N Y Source: GHI 2011-2012
While further discussion of urban informality in these five cities is beyond the scope for this paper, it is important to note that in the Latin American region, generally speaking (with the exception of Chile and Costa Rica), the rental housing stock that is available is often of better quality or equal to that in the homeownership market. Similarly, the proportion of urban households suffering from a lack of basic infrastructure (lack of access to piped water, sanitation, and electricity) is higher for homeowners than for
35 Abiko et al. 2007 as discussed in Bouillon et al. 2012. 11
renters (Blanco et. al. 2012). 36
Protections against discrimination. While legal frameworks to prevent discrimination are in place in most of the five cities, they are weakly enforced, and neighborhoods are highly segregated by income. The lower end of the rental market, as discussed above, represents the `bottom of the pyramid' and houses many of the most vulnerable populations in a city with the least protections for tenants, no subsidies for landlords, no knowledge about scale. High percentages of the most vulnerable who rent in informallydeveloped areas far from opportunity and unserviced land. No reliable data exists on the disparities in access to housing for women, minority groups, and vulnerable populations, but GHI indicators note that enforcement of the laws that are in place to protect discrimination remains low in all five cities.
Bogotб Buenos Aires Mexico City Recife Santiago
Are there policies, laws, or regulations prohibit the refusal to rent or sell property to someone based on race, religion, ethnicity, gender, or marital status. No: Yes Yes Yes No
If yes, are the policies, laws, or regulations enforced? (Rate 1 to 5, Strong enforcement=5/No enforcement=1, Not Applicable=100)
Are residential neighborhoods in the city segregated by race, income, or religion? (5=all neighborhoods are mixed,1=highly segregated).
In 2007, it was estimated that 35% of the leasing homes in Bogotб were not leased the unit to potential tenants for several reasons, among them their race or gender37 3, reluctance to rent to single women with Young Children
, some cases against immigrants from neighboring countries or for socioeconomic reasons; insufficient access to justice 3
1, especially with income. 1, income differences in the north and south zones; some concentration of ethnic and immigrant neighborhoods. 4, some spatial segregation
2, Spatial and socioeconomic
concentration; 66.2% of slum
households are black
1, income due to land values)
Source: GHI 2011-2012
(c) Land use regulations and planning Regulations that enforce low density and that place all servicing costs on developers are major deterrents to the creation of low and middle-income housing. These deterrents are especially relevant for potential developers of rental. In all five case studies, land regulations require that new residential land be fully serviced before it is occupied. In most cases, there are no subsidies for low-income developments or access to public serviced land. With rapid rates of urbanization, the region has largely been characterized by a high demand for serviced urban land far exceeding what municipal governments have been able to provide. The inability of municipalities to capture land value appreciation is a fundamental constraint of public-sector capacity to urbanize land prior to development, placing further pressure on urban land markets that often already are
36 Gilbert, 1999 in "A Home is For Ever" argues that often poorer tenants concentrate in areas that are wellestablished and tend to increase as self-help services age. 37 See "Urban leasing market characterization for the low income population segment", carried out by CENAC for the Ministry of the Environment, Housing and Territorial Development, UN- Habitat and First Initiative 12
prohibitively costly for middle income housing development (World Bank 2012a). The scarcity of affordable and serviced land remains a primary constraint in supplying formal housing to low-income communities (Jha 2007).
Bogotб Buenos Aires Mexico City Recife Santiago
Most distant year for which populations projections for the metropolitan area
of the city are available to municipal or metropolitan planners 2020 2015 2050 Not available 2050
Estimated number of years it would take for raw land where residential development is allowed on the periphery of the city to be filled up, given present densities and present annual levels of land consumption. 6 38 No annual data No data No data 7-20.39
City has an urban planning
document designating areas for urban expansion Yes Yes Yes No Yes Source: GHI 2011-2012
Land use planning includes regulations controlling urban density and urban growth, which, in practice, limits urban residential land and housing supply.40 How these controls are enforced varies. However, regardless, they drive costs upward and often disproportionately affect the poor. Most Latin American cities have actually witnessed a decrease in urban density. However, this is most likely due to the land area of cities expanding more quickly than the population is growing.
Rental housing has demonstrated its potential to rectify low-density land use trends with the ability to build or rehab formal units at high densities. This facilitates the development of affordable rental in urban areas
. Latin American cities have some of the lowest densities worldwide, and land supply is a key issue facing each of the five cities. Relatively large minimum lot size requirements, such as 300 square meters in Buenos Aires, further decrease access to land and drive down the feasibility of projects for potential developers of a small rental unit. These restrictions directly impact the city fabric; cities that normally have less strict zoning regulations typically have higher densities, particularly at the urban core.41
Smallest minimum lot size for residential building of any kind in the city (m2) 35
Minimum plot size used for social housing (square meters)42 -
Maximum allowable number of stories for new residential multi-unit buildings in the city. Varies
Buenos Aires 300 Mexico City 120
Determined by land area
Maximum allowable density in new housing developments (persons per hectare). Expected to reach 280 households per hectare 1000 inhabitants per hectare Depends on type, on average 1035 persons per hectare N/A
38 "The District Secretariat of Habitat (1099 of July 2012) prioritized development of 1,185 plots in urban land in Bogota, covering 111.31 hectares, which correspond to vacant plots in the consolidated city, connected with the urban infrastructure, which have not been developed." The GHI does not provide a specific estimate; six years is mentioned elsewhere for Colombia's five largest cities, including Bogota (World Bank, 2012) 39 "In consideration of the requirements established by the Metropolitan Regulatory Plan for the development of the city periphery, based on projects conducted in the Conditioned Urban Development Zones include development and building periods that varies from 7 to 20 years, considering between 15,000 and 70,000 inhabitants since they are developed in stages." 40 Bouillon et al. 2012. 41 Bouillon et al. 2012. 42 IADB, 2012
500 Source: GHI 2011-2012
Timely planning tools can leverage immediate opportunities to generate more affordable units. The significant quantity of vacant units and space located in serviced areas in the five cities studied illustrates an important current opportunity to increase affordable rental housing stock in central and accessible locations. While the cities analyzed struggle to finance and upgrade unserviced settlements on the periphery, many have millions of square meters of unused built space in urbanized and often highly central and accessible parts of town. Whether from areas with vacancy rates above 20% for finished housing units, Vacancy rates by delegation to the north of Mexico City, or the 4 million-plus vacant square meters from mostly central office buildings in Bogota, or empty government buildings in Buenos Aires, this built space represents a significant opportunity for rapid creation of affordable and accessible rental units.43
The phenomenon of underutilized vacant housing that exists alongside informally developed housing illustrates a market failure resulting from underdeveloped rental housing for the low and middle-income sector. Vacant units typically stem from the overproduction of what is not needed low-density new units for homeownership far from services. The oversupply leads to high numbers of vacant units.
Few policy tools exist to convert vacant space to rental,
despite increasing awareness and discussion of the issue.44, 45 The first trend in vacancies are residential units on the
edges of cities; housing policy makers in Chile and, to a
much larger extent, Mexico, face sizable inventories ofdata source
: INEGI 2010
finished homes that received subsidy sitting vacant far
from the urban core.46 On the northern edges of Mexico City, where many large projects were built with
subsidies over the last five years, nearly one out of every homes is empty (Figure 4). While these new
homes are typically provided with all basic infrastructure, accessibility to social services, employment
and urban amenities lagged far behind, and the new owners, who received the homes nearly free through
low-cost mortgages, let the units sit vacant.
Paired with a commitment to improve transport and social services, international experience
s suggest that policies to incentivize formal leasing arrangements and partnerships with property management companies in peri-urban areas can spark renewed investment, prevent depreciation, and provide families
43 CENAC, 2013; the city estimates a further vacant, abandoned, or underutilized 596 hectares in unbuilt plots within the Bogota city limits in 2012 (Mayor's Office of Bogota, 2013). 44 On Brazil: http://www1.folha.uol.com.br/cotidiano/1214946-analise-e-preciso-atrair-diferentes-faixas-de-rendapara-mudar-o-centro.shtml 45 Oxfam presenta proyecto para reducir la escasez de vivienda en Brasil, Oxfam.org 2010 46 The Chilean Ministry of Housing and Urbanism estimated 3,788 vacant social housing units nationwide in 2011, and 150,000 eligible households waiting for a unit. The Ministry responded by tightening regulations against rental of social units. 14
with an option to rent a formal and serviced home, and perhaps even build equity.47 In Bogota, several initiatives towards urban redevelopment mechanisms have advanced over the last several years at the city level, but large projects on the outskirts remains the priority for the new national redevelopment company.48
Implicit policy and regulatory bias toward peri-urban development has also disincentivized effective redevelopment of underutilized office space downtown, much of which could logically host well-located rental units. In all cities, and especially Bogota, Buenos Aires, and Recife, vacant commercial or governmental office buildings sprinkle the downtown districts: central Bogota alone has over 4.5 million square meters of empty offices.49 With no formal policy solutions in sight, informal invasions of these spaces are a logical response. Sao Paulo and Caracas are likely most famous for their informal `vertical invasions' of central office buildings, but cities across Latin America face similar pressure from households in need of shelter that is accessible to work and services.50 A handful of pilot conversions in the region, such as the Habitat for Humanity project in Buenos Aires, could be a good starting point for policy design.
Vacant urbanized lands, lastly, is a fundamental constraint to affordable housing development. Infill
development often carries higher direct costs (such as permitting and title clearance) and perceivedopportunity cost
s (increasing land values triggers speculative holdings) that decrease financial feasibility for developers. If partnered with planning and regulatory
The government has torn down existing low-income neighborhoods and transferred the vacant lands to the private sector for development of new
strategies to incentivize, subsidies for the development of commercial, infrastructure or higher-end residential.
high-density rental units can tip the balance. In Recife, the 77% land in the special ZEIS51 districts, a national policy
Bogotб Buenos Aires
instrument in Brazil, has been used for housing, and had Mexico City
much higher densities, averaging 225 residents per Recife
hectare, compared to 65 per hectare in the rest of the Santiago city.52 In Bogota, the Planning Department recently
Yes Source: GHI 2010-2011
calculated that 100,920 low-income multi-family housing units could be built on lots sitting empty due to speculations, which would decrease the city housing deficit by 20 percent.53 As shown in the table above,
while three of the five cities have undertaken redevelopment, it often comes as the cost of existing low-
47 Haram City, a recent peri-urban formal development on the outskirts of Cairo first experienced high vacancy rates, the developer's offer to rent and manage units has been taken up by nearly all absentee owners, and by most accounts has been a popular experiment. 48 CENAC, 2013; the new national redevelopment company (EMBARCO) was created by a presidential decree in late 2011 specifically to focus on a new hub for government offices in the outskirts of Bogota; however few if no affordable housing is planned. 49 CENAC, 2013. 50 In January of 2013, 31 vacant buildings were invaded in central Sao Paolo, including hotels. January 15, 2013. Folha de Sao Paolo. "Centro tem 31 prйdios invadidos por sem-teto". 51 Zonas Especiais de Interesse Social, or `Special Social Interest Zones' is a Brazilian urban instrument that includes higher-density zoning and reduction of certain regulations in specific areas of the city to encourage redevelopment and development of vacant parcels, with an emphasis on affordable housing. In Recife, nearly four out of ten residents in 2000 lived in a ZEIS area (Donovan 2007). 52 Donovan, Michael Geiger. At the Doors of Legality: Planners, Favelados, and the Titling of Urban Brazil. 2007, housing percentage based on 2005 census data from the IBGE and the Prefecture of Recife; density from the 2000 census. 53 Buillion et al. 2012 15
income neighborhoods. With the right tools, infill conversions of vacant space can provide an immediate and less disruptive opportunity for development and creation of accessible and affordable rental housing. CONCLUSION Our analysis supports case-study level findings that a persistent mismatch between the demand and supply for healthy and secure rental in all five cities persists, particularly among low-income households. This mismatch closely correlates with inadequate regulations, institutions and policies that constrain the development of rental housing supply in Bogotб, Buenos Aires, Mexico City, Recife and Santiago. Across the five cities, subsidies are largely skewed in favor of ownership, with $6.6 billion annually spent in direct, national-level subsidies for ownership, and less than $6 million for rental. There is an underdeveloped legal context for rental market with no clear articulation of the balance of landlord/tenant rights, including eviction processes, and thorough protection against discrimination. All five cities demonstrate that planning and land-use regulations discourage density and small-unit size, essential for low and medium rental market feasibility. Finally, the lack of city-level policy tools to capitalize on opportunities such as vacant lands, floor space, and units through conversions and redevelopment portray that policy interventions could be better suited to manage the inefficient housing markets. More broadly, the GHI platform represents a new resource for city-level analysis of housing policy regimes that incorporate property, regulations, finance, subsidies, and infrastructure sub-indicators. This wide range provided by projects such as the GHI is critical not only to measure housing market outcomes and gaps but also to provide a more nuanced understanding of how policymakers can potentially better intervene to correct market disparities, both for sector-specific policy comparisons like this one and incountry policy makers. A secure long-term platform for multi-national housing indicators is still a missing public good for housing. Going forward, the question remains how to encourage and maintain research from the indicators on GHI and others, and how to best compound new knowledge created through indicator usage, collaboration and critical dialogue with key local and national housing policy-makers. Works Referenced Abromowitz, David M. "Essay on Community Land Trusts: Towards Permanently Affordable Housing, An." Miss. LJ 61 (1991): 663. Arbelaez, M. A., R. Steiner, A. Becerra, and D. Wills. 2011. Housing Tenure and Housing Demand in Colombia. IDB Working Paper Series no. IDB-WP-259. Department of Research and Chief Economist, Inter-American Development Bank, Washington, DC. Arnott, Richard. "Housing vacancies, thin markets, and idiosyncratic tastes." The Journal of Real Estate Finance and Economics 2, no. 1 (1989): 5-30. Bancomer. Situaciуn Inmobilaria. January 15, 2013. Mexico. Belsky, E., and R. Drew. "Taking stock of the nation's rental housing challenges and a half century of public policy responses." Revisiting Rental Housing: A National Policy Summit (2007). Blanco, Andres, Cesar P. Bouillon, Vicente Fretes and Andres Muсoz Miranda. 2012. To Buy or Not to Buy: Expanding the Rental Housing Market. In Cesar P. Bouillon (Editor). Room for Development: Housing Markets in Latin America and the Caribbean. Development in the Americas Series. Palgrave Macmillan, New York. Chapter 4, pp 87-108. 16
Bouillon Cesar P., Viviane Azevedo, Nadin Medellin and Cynthia Boruchowicz. 2012. Two Bedrooms, Two Bathrooms and a Big Yard? Housing Demand in Latin America and the Caribbean. In Cesar P. Bouillon (Editor). Room for Development: Housing Markets in Latin America and the Caribbean. Development in the Americas Series. Palgrave Macmillan, New York. Chapter 3, pp 51-86. Bouillon Cesar P., Cynthia Boruchowicz, David Dowall, and Nadin Medellin 2012. Digging up the Truth about Land Markets. In Cesar P. Bouillon (Editor). Room for Development: Housing Markets in Latin America and the Caribbean. Development in the Americas Series. Palgrave Macmillan, New York. Chapter 5, pp 109-148. Briggs, Xavier de Souza, and Margery Austin Turner. "Assisted housing mobility and the success of lowincome minority families: Lessons for policy, practice, and future research." Nw. JL & Soc. Pol'y 1 (2006): 25. Cristini, M., R. Moya, and G. Bermudez. 2011. Argentina's Housing Market in the 2000s. IDB Working Paper Series no. IDB-WP-262. Department of Research and Chief Economist, Inter-American Development Bank, Washington, DC. Cubeddu, Luis, Camilo E. Tovar, and Evridiki Tsounta. Latin America: Vulnerabilities Under Construction. Working Paper 12/193. International Monetary Fund, 2012. http://www.imf.org/external/pubs/ft/wp/2012/wp12193.pdf (accessed February 12, 2013). Custers, Geert. 2002. "inner city
rental housing in Lima: a portrayal and an explanation". Journal of Planning Literature. 16 (3): 397-477. Fay, Marianne et al. "The Urban Poor in Latin America," World Bank. 2005. Gallopнn, Gilberto C. "Environmental and sustainability indicators and the concept of situational indicators. A systems approach." Environmental Modeling and Assessment 1, no. 3 (1996): 101-117. Gilbert. Alan. Rental Housing in Latin America: A Comparative Statement Gilbert, A. (1999) "A home is for ever? Residential mobility and home ownership in self-help settlements." Environment and Planning A, 31(6): 1073-91. Gilbert, Alan, and Ann Varley. 1991. Landlord and Tenant Housing the Poor in Urban Mexico. London: Routledge. Gilbert, Alan, Renй Coulomb, Andrйs Necochea, and Oscar Olinto Camacho. In search of a home: rental and shared housing in Latin America. London: UCL Press, 1993. Gilbert, Alan. Housing in Latin America. No. 10398. Inter-American Development Bank, 2001. Goytia, Cynthia, Cecilia de Mendoza, and Ricardo Pasquini. Land Regulation in the Urban Agglomerates of Argentina and its relationship with Households' Residential Tenure. Working Paper, Lincoln Institute of Land Policy, June 2010. Graham, Stephen, and Simon Marvin. 2001. Glocal Infrastructure and Splintering of Urban Economics. In Splintering Urbanism: Networked Infrastructures, Technological Motilities and the Urban Condition, pp. 304-377. London: Routledge. Giuliano, Genevieve, Peter Gordon, Qisheng Pan, and JiYoung Park. "Accessibility and residential land values: Some tests with new measures." Urban Studies 47, no. 14 (2010): 3103-3130. 17
Habitat for Humanity: Global Housing Indicators: Evidence for Action. http://www.habitat.org/sites/default/files/gov_global_housing_indicators.pdf (accessed Jan. 31, 2013). Oswald, Andrew J. "A conjecture on the explanation for high unemployment in the industrialized nations: part 1." (1996). "Rental Housing." Housing Finance Information Network (HOFINET) http://hofinet.org/themes/theme.aspx?id=59 (accessed February 2, 2013). IADB Rental Housing: Conceptual Framework, Literature Review, and International Experiences. Discussion Paper. Instituto Nacional de Estadнstica y Geografнa, Resultados de la encuuesta de verificaciуn de vivienda deshabitadas y de uso temporal del Censo de Poblaciуn y Vivienda 2010. Jha, Abhas K. "Low-income housing in Latin America and the Caribbean." World Bank. 2007 Hoek-Smith. Marja C. Subsidizing Housing Finance for the Poor. Hofinet.org Mйnrad, Sйbastien. The Social Housing and Rental Markets in an Equilibrium Rent Search Model. Miraftab, Faranak. "Revisiting Informal-Sector Home Ownership: The Relevance of Household Composition for Housing Options of the Poor." International Journal of Urban and Regional Research 21, no. 2 (2002): 303-322. Moya, Ramiro. The Rental Market in Argentina: An Assessment Study. No. 4823. Inter-American Development Bank, Research Department, 2012. http://idbdocs.iadb.org/wsdocs/getdocument.aspx?docnum=36664041 (accessed February 3, 2013). Navarro, M. "Housing finance policy in Chile: The last 30 years." Land Lines 17, no. 3 (2005). https://www.lincolninst.edu/pubs/dl/1037_July05%20Land%20Lines%20final.pdf (accessed February 3, 2013). Piza, C., and J. Litchfield. 2011. Housing Demand in Brazil: Evidence from the Formal and Informal Sectors. IDB Working Paper Series no. IDB-WP-261. Department of Research and Chief Economist, Inter-American Development Bank, Washington, DC. Rojas, Eduardo. Building cities: neighborhood upgrading and urban quality of life. Inter-American Development Bank, 2010. Ragin, Charles. 1987. The Distinctiveness of Comparative Social Science. In The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies, pp. 1-18. Berkeley: University of California
Press. Peppercorn, Ira G. and Claude Taffin. 2012. Rental Housing Pomeroy, Steve, and Marc Godbout. "Development of the Rental Housing Market in Latin America and the Caribbean." Inter-American Development Bank, IDB-Dp-173 (2011). Retsinas, Nicolas Paul. Low-income homeownership: Examining the unexamined goal. Brookings Institution Press, 2002. 18
Sбnchez, Aida Caldera. Building Blocks for a Better Functioning Housing Market in Chile. No. 943. OECD Publishing, 2012. Saegert, Susan, and Lymari Benitez. "Limited equity housing cooperatives: defining a niche in the lowincome housing market." Journal of Planning Literature 19, no. 4 (2005): 427-439. Angel, Shlomo. Housing policy matters: A global analysis. Oxford University Press, USA, 2000. Snyder, Richard. "Scaling down: The subnational comparative method." Studies in Comparative International Development (SCID) 36, no. 1 (2001): 93-110. Struyk, Raymond J. "Understanding high housing vacancy rates in a developing country: Jordan." The Journal of Developing Areas 22, no. 3 (1988): 373-380. United Nations Human Settlements Programme. Rental housing: An essential option for the urban poor in developing countries. UN HABITAT, 2003. UN Habitat: A Policy Guide to Rental Housing in Developing Countries. Quick Policy Guide Series Volume 1. 2011. http://www.unhabitat.org/pmss/listItemDetails.aspx?publicationID=3107 Wood, Gavin, Judith Yates, and Margaret Reynolds. "Vacancy rates and low-rent housing: a panel data analysis." Journal of Housing and the Built Environment 21, no. 4 (2006): 441-458. World Bank. Poverty and Social Impact Assessment, Colombia: First Programmatic Sustainable And Productive Cities Development Policy Loan (P130972). 2012. 19