Women’s Voice and Agency Research Series 2014 No.8 92762 Expanding women’s access to land and housing in urban areas Carole Rakodi This paper was commissioned by the World Bank Group to help inform a forthcoming report on women’s voice, agency, and participation. It does not necessarily reflect the views and research of the World Bank Group. Feedback and comments are welcome at: genderandagency@worldbank.org More details about the report are available at: www.worldbank.org/gender/agency 1 Contents Summary 1 Introduction 2 Land, property and wellbeing in urban areas 2.1 What do urban women, especially poor women (and men), want? 2.1.1 Location 2.1.2 Secure tenure 2.1.3 The role of housing in gendered social mobility 2.2 Obtaining access to land and property 2.2.1 Legal frameworks 2.2.1.1 International conventions 2.2.1.2 Land and property law 2.2.1.3 Family and personal law 2.2.2 Family and social relationships 2.2.3 Channels of supply for land and housing 2.2.3.1 Formal private sector land and housing delivery 2.2.3.2 Public programmes for land delivery to the poor 2.2.3.3 Alternatives to public land delivery: multiple channels 2.3.3.4 Women’s access to land and adequate housing: an overview 3 Recent land administration reforms 3.1 Legal reforms 3.2 The recognition, regularisation and upgrading of informal settlements 3.2.1 Recognition 3.2.2 Regularisation 3.2.3 Upgrading 3.3 Privatisation of public sector land and housing 3.4 Public programmes for providing serviced plots and/or housing units 4 Expanding women’s access to land and housing in urban areas: strategies and implications 4.1 Strategies and tactics to influence policy and practice 4.2 Implications for legal reform, policy and practice 4.3 Assessing poor women’s access to land and property: research and data needs References 2 Summary Poor urban women give high priority to affordable access to well-located serviced plots and houses. Women’s locational preferences, motives for investing in housing and priorities for services may differ from men’s. Evidence is mounting that registration of property rights has positive effects for poor people in general and women in particular, especially where households previously had little security. These benefits are manifest in increased investment in housing, a reduced need for activities to protect insecure tenure rights, empowerment of women within their households and benefits to families (including increased investment in human capital, especially of children). This paper makes the case that gender responsive land, housing and urban development policies and practices that address the needs of low income urban residents are as important as individual property rights. Legal tenure for example does not necessarily provide low income residents with access to formal credit for investment in either housing or economic activities because financial institutions are often unwilling to lend to the poor, and low income households are often unwilling to risk their main asset. Evidence on the cost effectiveness, sustainability and longer term impact of land and housing policies and legal reforms is limited and the few quantitative evaluations (and even fewer gendered evaluations) are context or project specific and so their lessons only have limited transferability. In practice, many interventions are multi-sectoral (such as upgrading of informal settlements) and legal/policy changes are often accompanied by other initiatives (e.g. microfinance, entrepreneurial support, service improvements), so attributing outcomes and impacts to registration and titling is difficult. Many countries have now improved the legal framework for land administration and personal law to protect women’s rights, but it is rare for all the ingredients of a progressive legal framework to be in place and often inconsistencies between sources of law and individual pieces of legislation remain. In addition, there are gaps between legal provisions and social norms and practices. Whether or not policies and laws address existing unhealthy and insecure living conditions and redress gender inequality depends on political, social and familial relations and whether poor women and men have a voice in decision making. In practice, many recent land registration and housing programmes, as well as wider land and planning policies, are not gender sensitive. Even when gender considerations are integrated into the design, they have not always been implemented in practice. Experience shows that the legal and policy frameworks for land management and property rights need to explicitly recognise women’s rights, requiring gender-specific measures. However, reforms to property and family law are insufficient – they interact with policies and practices related to land administration, planning for increased housing supply and improvements to informal settlements. In order to benefit poor women, laws, governance and administrative arrangements that improve access to affordable, well-located serviced plots and houses for poor residents in general are required. But unless gender is mainstreamed into their design and implementation, they will not meet women’s needs. Although many of the changes appear to be legal and technical, access to and control over land and property is related to socio-economic characteristics and governed by power relations at the family, community, city and country levels. Any changes, including those that seek to increase gender equality, challenge vested interests. Their effects therefore depend on the outcomes of political processes and power struggles. Gender-specific measures to improve women’s access to land and property include: 3 i. Domestic laws relating to property rights, land administration, marital relationships and inheritance should: a. Acknowledge the right to secure tenure and adequate housing; b. Affirm the equal rights of men and women, regardless of their marital status, making full community of marital property (and its joint administration) the default marital regime; c. Mandate joint tenure for urban men and women in civil, customary and consensual unions; and d. Entitle spouses to a share of the deceased partner’s estate, and sons and daughters to equal shares. ii. Streamline and harmonise inconsistent laws, in particular moving towards consistency between statutory and customary and religious law with respect to gender equality. iii. Support the efforts of civil society organisations, especially women’s and grassroots organisations, to raise women’s awareness of their rights and support them to claim those rights, especially through the statutory, customary and religious legal systems. iv. Make affordable or free legal aid available to enable poor women to claim and defend their rights. Mainstreaming gender equality in pro-poor urban land and housing laws, policies and practices, including through: i. Gender should be mainstreamed in land policies, programmes, procedures and practices, through building the gender awareness of staff and professionals and carrying out assessments of the gender responsiveness of tools and processes. ii. Registration/titling of property is likely to enhance women’s well-being and property rights if it is gender sensitive, systematic and affordable, especially where households previously had little security. A rights registration system that can be implemented quickly, systematically, at scale and at reasonable cost is required. In some situations, full titling may be appropriate. Where the capacity of land administration agencies is limited and/or multiple forms of tenure co-exist, determining the ownership of property is complex, and titling is likely to exacerbate inequality. More appropriate systems for securing property rights should be developed, especially those that permit the registration of a range of rights, especially use or occupancy rights. iii. Gender dimensions should be integrated into urban planning and development programmes, through providing opportunities for women’s and men’s participation in decision-making. iv. Policies should balance the need to upgrade and regularise existing informal settlements with guided development of new subdivisions to ensure an adequate supply of well-located and affordable serviced plots. Both need to be based on basic standards and implemented at scale to reduce the likelihood of gentrification of upgraded settlements and the capture of plots intended for the poor by households with higher incomes. Both need to be gender sensitive. In particular, residents should be protected from forced eviction and when relocation is necessary, internationally-recognised good settlement guidelines should be followed, including the equal entitlement of women to restitution and/or compensation. Further, planning for utilities and services, employment, green space (including for cultivation) and public transport should take into consideration women’s and men’s different needs. v. Related policies, such as the design and use of property taxes and regulation of landlord- tenant relations, should take into account the needs of poor people, including women. 4 vi. Low income women (and men’s) ability to access land and housing can be enhanced by support to home owners’ associations/cooperatives, acceptance of collective tenure, support for appropriate savings mechanisms, and possibly loans and subsidies. Countries’ inheritance laws and practices, policy frameworks and administrative systems are diverse, so it is vital that priorities and the content of laws, policies and programmes are adapted to particular national and local contexts. In addition, because the data is poor and the evidence patchy, improved data and more systematic gendered evaluations of the outcomes of legal changes and other interventions is needed. Priorities for research include:  developing a better understanding of the ability of women with different marital status’ to become land and house owners and retain rights throughout their life course, given the changing scale and characteristics of the channels of land and housing supply available to poor urban households, relating the ‘ownership’ and/or value of assets (‘wealth’) to different marital property and inheritance regimes.  Gendered assessments of the outcomes of registration, both single intervention programmes (tenure registration) and those that combine registration with other interventions, especially infrastructure and service provision, in new or existing areas. The outcomes of innovative instruments and approaches (e.g. joint tenure), as well as short- and long-term effects on women and their families should be assessed. Gender disaggregated data are also required for policy making, planning and day-to-day land administration at the household, land unit, city and regional/national levels. Possible ways of generating such data include improved administrative records, piggy-backing on existing data collection exercises (e.g. censuses, household sample surveys), the use of new technology to develop land information systems and participatory community inventories. 5 1 Introduction Evidence is mounting that secure property rights have positive effects for poor people in general and women in particular. The aim of this report is to review what is known about women’s access to and control over land and real property in urban settings, identifying approaches to strengthening property rights that enhance women’s agency, and sharing key lessons. Its objectives are to: i. Explore how and why access to and control over land and property improves the wellbeing of urban women, especially poorer women. ii. Identify and explain patterns and trends of land ownership among urban women and men across cities, countries and regions. iii. Examine how legal and institutional frameworks and norms determine the extent to which women enjoy access to, ownership of and control over land and property. iv. Identify and assess attempts to address the constraints on access to and control over land and property, especially for poor urban residents, with particular reference to the extent to which the design and implementation of interventions have been gender aware, and their outcomes and impacts for women. v. Identify potentially promising approaches to ensuring that women’s needs are fully taken into account in legislative and policy change and programme design. vi. Suggest areas worthy of further research, as well as possible indicators and types/sources of data for assessing women’s access to, use of and control over land and property. Section 2 synthesises the evidence on urban women’s priorities with respect to land and housing; the factors that influence women’s access to and control over land and secure tenure, including legal and institutional frameworks and social relationships, especially within the family; and what is known about the extent to which women have access to, control over and use of urban land and housing, and through what forms of tenure. In Section 3, recent reform of laws, policies and practices to meet the needs of poor people in urban areas, especially women, will be assessed. Both the legal reforms that underpin and interact with land and housing policies, and the most common contemporary approaches to providing secure tenure and adequate housing to poor urban households will be assessed: the regularisation and upgrading of informal settlements, the privatisation of public sector housing, and public sector programmes to increase the supply of plots and houses. 1 Evidence will be presented on the impacts of these approaches, especially where they have been implemented on a relatively large scale and gendered evaluations are available. Section 4 provides a short discussion of some of the strategies, tactics and alliances that are being adopted to bring about legal reforms and to influence the content, design and implementation of programmes in urban areas. It goes on to identify the implications for policy, legislation and practice, 1 As well as tenure and registration, urban land and housing policy needs to include and integrate policy, law and practice related to planning for urban growth and change, the regulation of development and taxation, but restrictions on time and space preclude a systematic consideration of planning, regulation and taxation in this report (see, for example, Lipman and Rajack, 2011). 6 focusing on how gender can be mainstreamed into land and housing policies, what gender-specific measures can be introduced, the priorities for future research, and how the available data can be improved. The report draws on successive research projects by the author and her colleagues on land and housing markets and policies, urban livelihoods, and urban politics and governance in a variety of contexts, in particular Rakodi (2010). First, much of the material on the social and economic impacts of titling reviewed for Payne et al (2008, 2009a, 2010) was re-reviewed to ensure that any consideration of gender in studies available at that time had been captured. This was followed by a search of the wider academic and policy-related literature.2 Simultaneously, contact was made with a number of organisations, scholars, practitioners and consultants working on land and gender, to identify secondary material. These included the World Bank, UN-HABITAT’s Global Land Tools Net-work (GLTN), the Huairou Commission and Shack/Slum Dwellers International (SDI). 3 Although most of those contacted responded, producing some useful material and many expressions of interest in the work, they also confirmed the paucity of gendered studies and systematic evaluations. Much of the available published and unpublished literature is comprised of general discussions, advocacy material or programme documents. Ten years ago, COHRE and UN-HABITAT commissioned a few global/regional reviews of legislation and practice (e.g. COHRE, 2004, 2006; UN-HABITAT, 2005, 2007a), as well as work on land, law and Islam (Sait and Lim, 2006). They have not updated these reviews, concentrating instead on developing programmes and producing guides for practitioners. However, more recently others concerned primarily with agriculture and facilitating enterprise have produced overviews (e.g. Byamugisha, 2013; Hallward-Driemeier and Hasan, 2013; Namubiru-Mwaura, 2013). The amount of relevant and rigorous research on women’s access to land and property in urban areas is limited and systematic assessments of the outcome and impact of programmes aimed at increasing the access of poor urban people to land and housing are few. Very few of the latter adopt a gendered approach. Moreover, few of the programmes and policies that explicitly sought to take women’s needs into account have been systematically evaluated, either by the implementing agencies or by independent researchers. As a result, although it is now possible to discuss the issues in a less simplistic way than previously, there are enormous gaps in evidence. 2 Websites searched include the University of Birmingham library, the Social Science Research Network e- library, Google scholar, UN-HABITAT, UNRISD, Mokoro Africa land database, USAID, World Bank, UNDP, International Land Coalition, Huairou Commission, Rural Development Institute, GSDRC, FiG, COHRE, International Center for Research on Women, UN Office of the High Commissioner for Human Rights and UN- Women. 3 I would like to acknowledge in particular the assistance received from Professor Sylvia Chant (LSE), Professor Ann Varley (University College London), Professor Diana Mitlin (University of Manchester) and UN- HABITAT’s GLTN (Clarissa Augustinus, Lowie Rosales), also Professor Carmen Diana Deere (University of Florida), Banashree Banerjee (independent consultant, India), Dr Emmanuel Nkurunziza (Rwanda Natural Resources Authority), Dr Rose Musyoka (Land Commission, Kenya), Prof Faustin Kalabamu (University of Botswana), Profs Wilbard Kombe and Lusagga Kironde (Tanzania), Rosalie Kingwill, Lauren Royston, Margot Rubin and Professor Vanessa Watson (S Africa), Mark Napier (Urban LandMark, S Africa), Allan Cain (Development Workshop, Angola), Chiara Selvetti (DFID), Emma Samman (Overseas Development institute) and the Gender and Development team at the World Bank. Apologies if I have omitted any others who made suggestions. This paper represents my own point of view. The draft was also sent to the Huairou Commission (Jan Peterson) and SDI (Sheela Patel). 7 2 Land, property and wellbeing in urban areas What all urban people need for wellbeing is a secure place to live, in a healthy environment and within reach of work opportunities and essential services. With respect to land and housing, this implies both affordability and protection against arbitrary eviction, whether by agents of the state, private owners or people’s own relatives. Gaining access to land and real property involves choices, which are shaped by  the legal framework, including both land and personal law  social norms and expectations, especially concerning marriage and the family  the channels through which land and housing are supplied, the types of tenure on offer and the cost of alternative sources of accommodation To assess the access of poor people, including women, to land and housing requires an understanding of the types of tenure available, the ways in which land and housing are supplied, and what resources are needed to access alternative sources (and are available to the poor), including finance, information, time and skills. It is also necessary to consider what happens if households’ or individuals’ claims to land or residence are contested: the legislative safeguards that are available, and the processes by which they can be enforced. In addition to gender, crosscutting sources of social differentiation and social norms include socioeconomic status, ethnicity, religion, caste, residence (urban v rural) and age. While women on average (or the households they head) are often disadvantaged compared to men, it should not be assumed that they are all (or even disproportionately) poor (Chant, 1997, 2003). Both land/property and family laws are critical. In addition, kinship and marriage relations are central to the way rights and laws are shaped, and women cannot be considered in isolation from their roles as wives, mothers, sisters and daughters (Varley, 2013). Social norms and expectations affecting family and marital relationships include those governing the acceptability of cohabitation versus marriage, household composition, practices of seclusion, treatment of widows, and the obligation to remarry following the dissolution of a marriage. The political and administrative arrangements for land governance and urban management are also important influences on the operation of land and housing delivery systems, in addition to:  the patterns of land ownership and use in and around urban areas;  the topography, which affects the scope for expansion, the need for regulatory constraints, the price of land and the cost of construction; and  the socio-cultural norms that influence how people live and interact in rapidly changing urban areas. All these are gendered, yet laws and regulations, administrative structures and processes, and arrangements for representation and decision-making often do not explicitly consider gender. Even if they do, and particularly if assumptions about gender relationships are implicit, they most often reflect established social norms and arrangements rather than contemporary ideas about gender equality as embodied, for example, in the international conventions to which most governments have signed up. 8 Much of the attention in the growing debate about women’s access to land has focused on rural areas. Gradually, the conceptual frameworks used, the data available and the level of gender awareness in the design of interventions have improved, although recent discussions of the continuing gender asset gap, the gulf between constitutional and legislative change and everyday practice, and the gender blindness of policies and programmes (or implementation processes) show that there is a long way to go (e.g. Byamugisha, 2013; Deere and Doss, 2006; Deere and León, 2001, 2003; Deininger and Feder, 2009; Hallward-Driemeier and Hasan, 2013). Some of the findings and conclusions from rural studies may be relevant for urban areas, since the dividing line between urban and rural is not precise (Brown and Lloyd-Jones, 2002; Satterthwaite and Tacoli, 2002; Varley, 2010). Better understanding of the systems for that would provide land for citizens of all income groups and register and record the full range of legal tenure and land transactions in a way that is accessible and beneficial to all groups have only received limited attention from many national governments, international agencies and researchers to date. Many national governments are wary of the political role played by city populations, so have tried to deter rural-urban migration and limit the power and resources available to urban authorities. As a result, policy attention, programming and research on urban land-related issues has been patchy and sporadic, and even when these have been the focus of interest, gender considerations have not been prominent. In much of the published work on access to land in urban areas, gender remains neglected or is dealt with in an over-simplified way. For example, gender receives only passing references in some key edited collections (Durand-Lasserve and Royston, 2002; Fernandes and Varley, 1998; Home, 2011a, 2011b; Lall et al, 2009; Payne, 2002) and analyses of implementation and impact (e.g. Angel et al, 2006; Fernandes, 2011; Gulyani and Bassett, 2007). This report is therefore limited in scope due to the lack of in-depth gender analysis in previous urban land and property research and analysis. 2.1 What do urban women, especially poor women, want? ‘Land’ per se is of little practical value. Potentially it has use and/or market value, realisable through rental or sale, which is influenced by its location in relation to economic opportunities, its accessibility, and the infrastructure and services that are available to those using it. While undeveloped urban land has exchange value (and can increase its value dramatically over time), obtaining land and keeping it vacant is not a strategy available to the urban poor. For poor urban residents the possibility of owning their own houses cannot be realised without access to land, and land cannot be considered separately from what can be built on it (real property). The living environment the house provides also has value, and this is influenced by the infrastructure and services available. Hence land and houses will be bracketed in this discussion. A broad distinction can be made between tenure (“the means by which land is held”) and property rights (“who can do what on a plot of land”) (Payne and Durand-Lasserve, 2013, p 8). UN-HABITAT defines tenure as “the way land is held or owned by individuals or groups, or the set of relationships legally or customarily defined amongst people with respect to land. In other words, tenure reflects relationships between people and land directly and between individuals and groups of people in their dealings in land” (UN-HABITAT, 2008a, p 5). “The term property right refers to an owner’s right to use a good or asset for consumption and/or income generation (referred to as ‘use rights’). It can also include the right to transfer it to another party in the form of a sale, gift or bequest (referred to as ‘transfer rights’). A property right also typically conveys the right to contract with other parties by renting, pledging, or mortgaging a good or asset, or by allowing other parties to use it...” (Besley and 9 Ghatak, 2010, p 4526). Property rights can be based not only on statute but also on custom or informal practices that enjoy social legitimacy. They “may vary within, as well as between, tenure systems. It is therefore possible to have a high level of tenure security, but restricted rights to use, develop or sell land, or a limited level of tenure security, but a wide range of actual [de jure or de facto] rights” (Payne and Durand-Lasserve, 2013, p 8). Generally, the most desirable bundle of property rights is considered to be ‘ownership’, which is associated with a number of benefits over and above use value (see, for example, Moser, 1998, 2010; Moser and Felton, 2010; Rakodi, 1999). It implies possession of an asset that can:  be put to productive use (for example cultivation, business purposes or the construction of dwellings for rent);  reduce a household’s vulnerability to shocks and stresses through sale, rental or otherwise using the asset to generate additional/new income;  enable an owner to realise a capital sum through sale, rental or using the property as collateral for a loan; and  enable the inter-generational transfer of wealth, to benefit succeeding generations. Ownership is often associated with ‘title’, which provides for the ownership to be both long term (often in perpetuity) and guaranteed by government. In practice, while most (poor) urban residents may wish to own property, their ability to do so is highly constrained, and the value and feasibility of titling should not be taken assumed. The legal framework for tenure and property rights will be discussed further in Section 3. A good starting point for understanding the current distribution of resources within urban areas is to consider the functions poor urban residents, especially women, consider land and property to have in their lives and livelihood strategies, and what characteristics of land and the buildings constructed on it are of value to them in achieving their objectives. Any such analysis needs to distinguish between women in terms of their location within an urban area, their household situation (including their migration experience and status, marital arrangements and life course) and their other social characteristics (e.g. ethnicity, religion). Data on women’s and men’s urban land ownership are scarce. Studies of women’s perceptions of their endowments, economic opportunities, voice and agency, even if they are carried out in urban locations, do not always consider land and housing (for example, Munoz Boudet et al, 2012). Most studies on the role played by land and housing in the livelihood strategies of poor urban people focus on ‘households’ –many of which only distinguish between male and female-headed households, and relatively few focus on the life courses and views of women. The concept and definition of a ‘household’ is problematic. The extent to which there is income (and food) sharing within the household, shared responsibility for arranging and paying for accommodation and services (as well as other household expenses), and joint decision making among those within the household, varies between countries, cities, groups and households within cities. 2.1.1 Location The location of available land and housing in urban areas influences its accessibility, access to income earning opportunities, and the availability of infrastructure and services. For women, the preferred type and location of housing depends not only on their incomes but also on their household responsibilities and life cycle stage. For example, studies in Guadalajara and Mexico City show that 10 women, especially younger female household heads with children, often prefer central locations to peripheral settlements, where there are fewer income earning opportunities and paid work is harder to combine with child care because of long journeys to work. This influences their choice of tenure, because while home ownership is more feasible in peripheral areas, renting a room may be the only option in central locations (Miraftab, 1998; Wigle, 2008). The physical configuration of the residential environment is also important. The private space provided by a dwelling, whether owned or not, is important to household welfare, family and marital relationships, as well as providing opportunities for, or constraints on, income generating activities. Its significance for women is influenced by social norms, including religious norms of seclusion and, more generally, norms that regulate the acceptability of women moving around in public space, particularly for work (Brown and Lloyd-Jones, 2002). Bose (1998), for example, in her study of slum dwellers in Kolkata, describes how “Young unmarried women and women in their child-bearing years are subject to the greatest spatial confinement. With regards to livelihoods, over 20 per cent [of the working women] said that …[home-based work] was the only activity which they were allowed to undertake. Selling in the public arena was seen as a last resort” (quoted in Brown and Lloyd-Jones, 2002, p 192). The importance of location reflects the varying availability of infrastructure and services in different neighbourhoods, including roads, public transport, water, sanitation, electricity, education and health facilities. Location is also relevant to the availability of land for cultivation. Farmers in peri-urban areas are concerned with sustaining their livelihoods, changing farming practices to take advantage of urban markets and/or selling all or part of their holdings for urban uses. Urban agriculture, which may occur on residential plots or in any available spaces throughout urban areas, is an important component of the livelihood strategies of some households (see, for example, Hovorka and Lee-Smith, 2006; Hoornweg et al, 2012; Redwood, 2009). It is often undertaken as part of the domestic responsibilities by women who have few other income earning opportunities. Generally on-plot farming is only available to the owners of sufficiently large plots (often a minority), but is increasingly common as the supply of land for off-plot agriculture diminishes as demand and the density of development increase. Such farming can can provide a useful source of food for poor urban households and sometimes a surplus for sale. Despite the importance of farming in the livelihood strategies of some poor households, it is often considered to be an inappropriate use of urban land by the authorities and is penalised or neglected. Given the potential contribution of urban agriculture to household wellbeing and urban food security, access to land for this purpose should be taken into considered in land use planning and land administration. 2.1.2 Secure tenure The types of land and property rights that provide urban residents with security, and the delivery channels through which they are achieved, varies greatly from place to place.4 “A person or household can be said to have secure tenure when they are protected from involuntary removal from their land or residence, except in exceptional circumstances, and then only by means of a known and agreed legal procedure, which must itself be objective, equally applicable, contestable and independent” (UN - HABITAT, 2011a, p xii). 4 Security includes regular income flows and personal safety, as well as secure tenure, although the discussion here focuses on the latter (Patel, 2013; Rakodi, 2002). 11 Security has perceptual and practical as well as legal dimensions. These dimensions include: Legal tenure confers security if holders know their rights and challenges to these rights can be defended. This requires sound and enforceable law, understanding of their rights, and access to the legal system. Poor people in general and poor women in particular, often lack legal knowledge and the skills to interpret legal documents such as titles or tenancy agreements. In addition, their access to legal redress or channels for dispute resolution is often constrained by the cost, limited capacity in the judicial system and potentially the gender bias in law and/or practice. Legal tenure does not necessarily translate into security (Patel, 2013). For example, residents of informal settlements in Recife, Brazil, think that legal tenure leads to threats to security because of the increased attractiveness of titled plots to external buyers, while it does not block other threats to personal security such as disputes and violent events. For these residents, the provision of infrastructure and services would provide better practical security than tenure security in the form of legal titles (de Souza, 1999, 2001a, 2001b; see also Payne and Durand-Lasserve, 2013). The perceptual dimension of security has both negative and positive drivers – fear of eviction and a sense of belonging and safety. Security from eviction by landlords, private landowners, public agencies or members of ones own family (for example following divorce, separation or widowhood), is a widespread desire. COHRE’s research (2008) showed that women everywhere wish to retain rights to the marital home if they are widowed, and a share of marital property if they divorce or separate, whether the union was a legal marriage or a consensual union. (COHRE, 2008). In addition to the degree of legality conveyed by different forms of tenure, perceived security is inversely related to people’s fear of eviction, which research in Buenos Aires found may exceed the real probability of eviction (van Gelder, 2009). Even without legal tenure, research in Recife and elsewhere shows that if the probability of eviction is perceived to be low, owners make considerable investments in house improvements. In addition, the research shows that owners often believe that, should they be threatened by eviction, their claims will be stronger if they have invested in the housing structure (de Souza, 1999, 2001a, 2001b; Musyoka, 2006; van Gelder, 2009). In addition to freedom from the fear of eviction, the affect-based (emotional) dimension of perceptions of security identified by van Gelder (2009) has positive aspects. These are expressed by respondents as a sense of belonging, peace and achievement, and in order to achieve this they are willing to make considerable sacrifices. For example, urban residents in Guadalajara, Mexico expressed a distaste for renting or living with others rent free (typically young couples living with the man’s parents) because both were felt to make households vulnerable to eviction or lead to conflict. Instead, they wanted a home of their own and were “willing to face the hardships of purchasing a plot of land illegally and building a home little by little.... Despite the hard work of building - mixing cement without a water supply close to hand, for example - and the multiple deprivations of living without services (sometimes for years), people told us that what they value about their home was the chance to rest or to find ‘tranquillity.’ They sometimes expressed this sense of peace in terms of ‘rootedness’” (Varley, 2013). This suggests that security of property rights conveys social status and a sense of belonging. The practical dimension of security refers to the functions played by a dwelling in the lives and livelihood strategies of urban households. For some households, sharing or renting a dwelling is an active and positive choice because of its low cost, flexibility and sometimes locational advantages. 12 However, for many it does not provide the security they desire and so they seek to become owners. Their ability to do so can strongly influence their wellbeing. Research in India and elsewhere suggests that households with improving livelihood trajectories typically have a range of assets (including a house) sufficient to enable them to deal with shocks and stresses and take positive steps to improve their livelihoods. In contrast households that are merely coping, have insufficient assets to provide an adequate cushion against shocks and stresses, while declining households lack assets entirely, making them the most vulnerable (Loughhead and Rakodi, 2002). Migrants face particular challenges. Upon arrival in a new urban area, migrants generally have little choice but to stay with relatives, rent, or perhaps live in accommodation provided by their employer. None of these provide the type of security outlined above, and as such most seek to obtain a property of their own. Their ability to do so depends on their individual characteristics and livelihood strategies, and the opportunities available in their new home. For example, at least a third of the households in Gaborone, Botswana, are headed by women, largely the result of male labour migration to South Africa and female migration to urban areas in search of income earning opportunities. Research has shown that, with the support of gender-neutral government housing programmes, many women acquired a serviced plot and arranged the incremental construction of houses (Larsson, 1988, 1989; Yahya, 2002). Moreover, for many female household heads renting out rooms became an important source of income. Datta (1995) found that not only did many married women play an important role in managing the rental accommodation in which their husbands had invested; half the landlords in low income areas were female heads of household. The decisions many women made to remain unmarried, become home owners and invest in rental housing demonstrate conscious strategies to control their own lives, achieve security, provide a good start for their children and ensure continued support in their old age (Larsson, 1989; Kalabamu, 2006). The important roles played by home ownership in household strategies over time are also demonstrated by Moser’s longitudinal 26-year study of an urban community in Guayaquil, Ecuador. She found that the proportion of female-headed households had increased from virtually none in the early 1970s, when the settlement was established, to a third in 2004 (due to death, desertion or separation). Nevertheless, a higher proportion of female than male-headed families had risen above the poverty line (half compared to a third), due to their smaller families, lower dependency ratios, women’s greater access to stable jobs and the receipt of remittances. Although the adult children of the original settlers were finding it difficult to obtain housing of their own, their parents’ secure possession of a plot and house had helped them to provide their children (especially daughters) with an education and enabled more than half of them to subdivide the family plot to accommodate adult children and their households (Moser, 2010; Moser and Felton, 2010). 2.1.3 The role of housing in gendered social mobility Women’s experience of gender inequality can also be conceptualised in terms of their power and freedom relative to men. The possession of major assets, including rights to property, can contribute to gendered social mobility. For example Boudet et al (2012) attempts to provide not only a general picture of women’s experience of gender inequality, but also an analysis of whether and why they have experienced changes in their position relative to men in both rural and urban areas. Urban women report they “perceive significantly more gains in their power and freedom over the past decade than any other group sampled....The reports by urban women, moreover, starkly contrast with urban men’s sense of loss of power and freedom over the same time period and the challenge that this presents to their compliance with expected models of masculinity” (p 103). 13 The primary factors that empower men and women and explain upward and downward social mobility are identified as occupational and economic opportunities, with men’s downward social mobility being linked primarily to slow economic growth and limited job opportunities. The study found that “urban women perceived more extensive gains in their power when they control major assets, are free (or freer) from domestic violence, acquire greater social capital, and have a supportive local opportunity structure...Urban women also benefited from residing in neighbourhoods where gender norms are more relaxed, markets are stronger, and public services more accessible” (p 102). The research found that upward and downward social mobility are also influenced by attitudes and behaviours, education and training, community roles and responsibilities, and marital and family relationships, as well as (for women) savings and borrowing habits and (for men) access to political and financial resources. Control over assets (including a house with secure tenure), while being recognised as important to urban respondents, was not identified as a primary channel through which respondents said that they could gain more power and freedom in their lives. This lack of recognition may have been due in part to the way the questions were framed. 2.2 Obtaining access to land and property Urban women’s access to land with secure tenure is influenced by the legal framework, social and familial relations and the opportunities available through various channels of land administration. 2.2.1 Legal frameworks The main areas of law that govern the access to and control over urban land and property include international treaties and national constitutions, land and property law, and family and personal laws. Land law not only provides for the allocation, possession and transfer of property rights, it also provides the basis for urban planning and development; the regulation of economic activities, development and construction on land; and property taxation. Family law deals with marital relationships and inheritance, which are critical channels through which individuals obtain access to property. These laws may be derived from statute, customary law (which can be codified or not) and religious law. Legal pluralism and the complex and evolving nature of law frequently results in ambiguities and inconsistencies between these different sources of law. In addition, implementation and enforcement of laws are impacted by political and economic interests and are intricately linked to social norms and relationships, which will be considered further in Section 2.2.2. This paper does not to attempt a comprehensive global review of law affecting women’s access to and control over land and housing – more extensive reviews are available (e.g. Hallward-Driemeier and Hasan, 2013; Knox et al, 2007a; OHCHR and UN-Women, 2013). 2.2.1.1 International conventions International human rights law provides a framework for attaining gender equality (de Silva de Alwis, 2013). The right of all people to adequate housing is also specified as part of the right to an adequate standard of living in the UDHR and the International Covenant on Economic and Social Rights (Article 11(1) 1966). In 1991, in General Comment No. 4, the Committee on Economic, Social and Cultural Rights, specified that the right to adequate housing consists of the following elements (see also Rolnik, 2011; OHCHR and UN-Women, 2013): 14  security of tenure (which guarantees legal protection against forced eviction, harassment and other threats);  the availability of adequate services, especially water and sanitation;  a location that provides access to employment opportunities, education and health services and social facilities;  affordability;  habitability;  accessibility; and  cultural adequacy. In 1997, in General Comment No. 7 on forced evictions, the Committee recognised the particular vulnerability of women during evictions and set out the procedures to be complied with when eviction is necessary. Regional frameworks also set out explicit rights to adequate housing for women. For example, the Maputo Protocol to the African Charter on Human and People’s Rights on the rights of women (1981), adopted in 2003, recognised women’s right to equal access to adequate housing, whatever their marital status, through it’s provision for an “equitable share” in property inherited from a spouse or parents. 2.2.1.2 Land and property law The combination of imported legal systems and customary or religious systems has given rise to considerable difficulties in establishing coherent land law. This is due in part to the different interpretations of concepts of tenure and ownership that exist across the different systems, and the different rights that are conveyed by different forms of landholding, for example family vs individual, and absolute possession in perpetuity vs use rights (Fernandes, 2007a; Hallward-Driemeier and Hasan, 2013). In addition, customary law varies between ethnic groups and is often not codified, making it difficult to determine the exact basis upon which rights exist or decisions will be made. Few constitutions contain explicit legal obstacles to women’s land and property ownership. For example, Hallward-Driemeier and Hasan (2013) half of all countries in Sub-Saharan Africa (mostly those with common law systems) formally recognise customary law, and one-third exempt it from the principle of non-discrimination in family and inheritance laws. Legislation related to property rights and registration may discriminate against poor people in general and women in particular, explicitly or in practice. Most worldwide reviews of the equality of land laws consider the national situation (see, for example, UN Women, 2011, p 39, cited in Chant, 2013, p 17), and only one considers the position of urban women. Based on returns from nearly 150 cities in the developing world, this review found that in 40 percent of cases there were some or considerable legal impediments to women owning land. Overall the review found that women had the most access to land rights in Latin American and Caribbean cities, and the least in African cities, with Asian cities in an intermediate position (UN-HABITAT, 2007b, p 396-8). However, many countries have reviewed their constitutions and laws since this survey was carried out in 1998. The effectiveness of the land administration system also plays an important role in determining the extent to which women enjoy their rights to land. The statutory requirements for proof of ownership and the procedures for registration are typically onerous for both owners and governments. Whether or not the statutory provisions are capable of providing secure tenure to all urban residents depends on the capacity of government agencies to implement the system. Most developing country land agencies 15 do not have the necessary capacity, with the result that many land transfers do not comply with regulatory requirements. Even where women have equal rights to property, therefore, their ability to realise those rights depends on whether the wider land administration system is well designed and effective. 2.2.1.3 Family and personal law Provisions specifying how ownership rights apply to men and women during or upon the dissolution of marriage, and in the case of inheritance may be established in property law and/or in family law. While in many countries, and for certain groups within countries, customary or religious laws may prevail with regards marriage and inheritance practices. Marital property regimes govern how rights to property are allocated between men and women during marriage, and following divorce or the death of one spouse. Marital property regimes that recognise community of property are generally acknowledged to be the most gender equitable, especially for property acquired during marriage, and this is the mandatory default regime in increasing numbers of countries. Community of property implies joint registration and administration of all property during marriage and provides each spouse with protection against the unilateral sale or mortgaging of marital property by their partner. It also provides for equal distribution of marital property upon widowhood, divorce, abandonment or separation.5 However, it is generally considered desirable for real property brought into marriage by a wife to be retained in her own name, and for there to be provision for married women who wish to acquire/register property to be able to do so in their own names. In practice:  Women’s ability to enforce their rights is often hindered by unclear or inconsistent provisions, the cost of judicial redress and a backlog in the courts.  The law may vary for different conjugal relationships – civil marriage, religious marriage, customary marriage (and within this, monogamous and polygamous marriages) and consensual unions. In Sub-Saharan Africa, for example, intestate succession laws exclude customary land from the property a widow can inherit from her husband (Hallward-Driemeier and Hasan, 2013). Often, consensual unions are not recognised in property law at all (statutory, customary or religious), disadvantaging large numbers of women.  Even where joint tenure is specified (or permitted), it is common for relatively few titles to be registered in joint names (or women may not be informed that their name has been included on a registration document) (UN-HABITAT, 2007a).  In some places the law makes the husband the administrator of the community property, potentially enabling him to sell or mortgage it without his wife’s consent (for example, in some Mexican states, although fewer than in the past, Varley, 2007)  Cultural practices that relegate women to a secondary position in relation to property, including bride price and dowry practices, also inhibit the effective implementation of gender equitable statutory laws. Customary and statutory forms of property tenure may be collectively held or held by anindividual. The former regards land as the property of a group (typically an extended family) and generally provides women with only secondary rights to property. The marital property rules governing distribution of collective property during or upon dissolution of marriage generally limit each spouses share to the property they themselves brought into, or acquired during, the marriage, although these 5 See UN-HABITAT, 2005, p 22-33 for a country by country summary of legal provisions for joint tenure. 16 provisions differ between patrilineal and matrilineal systems (see Hallward-Driemeier and Hasan, 2013). Statutory collective arrangements (mainly cooperatives or trusts) do not necessarily include women or treat them equally unless this is a specific objective (UN-HABITAT, 2007a). For example in Kenya, land buying groups were formed to purchase undeveloped land, which is subsequently subdivided for urban residential use. Group members are typically from a single ethnic group and do not explicitly discriminate against women, however to the opportunity to buy shares requires resources which many women do not have access to (Musyoka, 2006). Religious law, especially Shari’a law, is the primary source of law in some countries and recognised as a source of personal law in many others, where it co-exists and overlaps with statutory and customary law (for example, in more than a quarter of Sub-Saharan African countries, Hallward- Driemeier, 2013, p 56). In some countries it is exempt from the non-discriminatory principles embodied in statutory law and it may sometimes be trumped by customary law and practice (COHRE, 2006; Sait and Lim, 2006). Throughout history, whether married or not, Muslim women have enjoyed autonomous legal identities and separate property rights. Under Islamic law, they retain control over their pre-marital property and finances through marriage and beyond, face no restrictions on purchasing property using their own resources, or on gifts received. Shari’a law provides for a husband to make a payment to his wife on marriage (mahr), specifies his duty to support and maintain her and provides for other avenues through which a woman can obtain property (such as gifts, dowry, maintenance and as a beneficiary under waqf). These provisions are used to justify Islamic inheritance rules ( mirath) that provide for a woman’s (and daughter’s) share of inherited property to be generally half that of a man (or son) in a similar situation (Sait and Lim, 2006, p 110; see also COHRE, 2006). In theory, the fixed shares rules support women’s rights to gain, retain and manage their own land and wealth . However, in practice they often act to restrict women’s property ownership as they do not have the means necessary to purchase their own property. Interpretation of Shari’a law also varies (see, for example, Sait’s discussion of Sunni and Shia interpretations and the interactions between statutory, customary and Islamic land laws in different parts of Africa, 2011), and there is no mechanism to ensure that women are compensated for lost inheritance (COHRE, 2006; Sait and Lim, 2006). Except in the few remaining legal systems that consider women to be legal minors (e.g. Swaziland), spouses can, if they wish, choose to arrange their marital relations and rights to property by contract. However, the likelihood of a formal contract being entered into depends awareness of the options available and the ability to fulfil the legal and bureaucratic requirements. In many cases poor women may not be aware of the advantages and disadvantages of different forms of marriage, including their implications for property rights. Even when they are, their ability to fulfil the legal and bureaucratic requirements and willingness to face social disapproval if they depart from accepted social norms is likely to be limited. Laws governing inheritance have a major influence on women’s access to and control over land and property. Dispossession of widows is a serious and widespread problem. It has been exacerbated by the HIV/AIDS pandemic, especially in Africa (COHRE, 2008; Strickland, 2004). Even if widows have the right to inherit or use marital property (especially the family home), their ability to exercise this right may depend on their relations with the deceased husband/partner’s family and compliance with certain practices and rituals that are sometimes onerous and degrading (e.g. levirate marriage) (Chant, 2013, p 18; UN-HABITAT, 2007a). 17 Analysis of DHS surveys since 2003 for 15 Sub-Saharan African countries found that 5 percent of women aged 15-49 had ever been widowed, but fewer than half of these (47 percent) had inherited any assets.6 Only in Rwanda and Senegal was it reported that widows and their children had inherited the majority of assets – in almost all the countries studied, most widows reported that their deceased husband’s property had been inherited by his natal family members (including his other children), rather than themselves and their children. Overall, the spouse’s children/family inherited the majority of assets in 48 percent of cases, the widow and/or her children in 32 percent, and another wife or relative(s) in 8 percent (8 percent of husbands had no property to leave). The data revealed differences both between and within countries, although urban/rural location was not significantly correlated with the likelihood of inheritance, possibly indicating the continued influence of customary law and practice in urban contexts (Peterman, 2012, p 550). A recent study in Malawi found that the vast majority of urban residents had access to an owner-occupied house, but that dispossession was deeper and more extensive for widows than widowers, especially in urban areas (66 percent compared to 26 percent) (Ngwira, n.d.). Deere et al (2013) find that urban women own a larger share of couple wealth than rural women in Karnataka and Ghana, and a similar proportion to men in more urbanised Ecuador – they attribute this to the smaller proportion of agricultural land in inherited wealth. The data is not analysed separately for urban and rural areas, but the report notes that in urban Ecuador it is common to inherit an undeveloped plot, on which the house built by a couple is considered to be jointly owned in nearly two thirds of cases (compared to a fifth of houses in Ghana and only 5 percent in Karnataka). This does not however prevent the woman’s rights from being disputed after divorce or widowhood, unless the title is registered in both names (Deere et al, 2013), but indicates that attitudes to joint ownership may change in urban areas (see also Varley, 2010, on Guadalajara, Mexico).7 In many countries statutory laws are gender neutral, and may contain specific provisions promoting women’s rights and/or gender equality. However, this is not sufficient to ensure that women enjoy equal access to property rights, and the relationships in which they are embedded and the norms that govern social interaction are in many cases as, if not more, important. Even where joint title is encouraged and practised, the property rights and security of those living in male-headed households are determined not only by law governing but also by social custom. While attitudes and behaviour are likely to change more quickly in urban than rural areas, urban populations in rapidly urbanising countries contain a large proportion of first generation migrants, and many urban households maintain ties with their rural areas of origin, so the social mores and practices of (more conservative) rural society may continue to have a strong influence on urban social practices. 2.2.2 Family and social relationships In many parts of the world social disapprobation of women who remain unmarried, combined with economic disadvantage, prevent many women from establishing independent households and, if they do, constrain their ability to afford land or a house.The complexity of women’s and men’s views on their obligations towards their spouses and children, and their uneven ability to exercise agency with respect to property, are illustrated by Varley’s research in Guadalajara. Despite the legal provision 6 Note that this data refers to a minority of all widows, because the DHS focuses on women of childbearing age. 7 The ways in which the intergenerational transmission of property (or wealth) contributes to chronic poverty or lasting improvements in wellbeing has recently received significant research attention. The literature is generally gender sensitive, but most of the studies are rural or do not distinguish between rural and urban areas. 18 that if an owner dies intestate the children will inherit property in equal shares, respondents’ views about who should inherit varied widely. Many believed that their sons should come first, because they are expected to marry and become responsible for their own families, whereas others favoured their daughters because women are considered to be at a socio-economic disadvantage. On the whole, male householders favoured sons over daughters, while women householders were more likely to favour daughters over sons. However, there were indications that attitudes are changing: some believe that whichever of their children look after them in their old age should be favoured, and younger people were more likely to think that children should inherit in equal shares. Overall, “men talked about protecting their wives and children; women about being protected. This suggests women’s relationship with property is a more indirect, passive one, contrasting with the agency assumed by men” (Varley, 2010, p 80). As discussed, although women’s financial and labour contributions are vital to land and house purchase and construction, enduring attitudes and practices towards marital relationships and inheritance mean that women often lose security following partnership breakdown or bereavement, even when statutory law is relatively egalitarian. To date, the evidence concerning the impact of inheritance is inconclusive. While some attribute gendered differences in wealth and wellbeing within families at least in part to inherited assets (for example, Deere et al, 2013), recent research in Senegal found that the inheritance of land or housing did not have statistically significant effects on the consumption levels of women or men in either urban or rural households, once controls for location and level of education were included. There was also no evidence from this study that inheriting land has a significant effect on the likelihood of people undertaking non-farm activities, although inheritance of other (non-land, non-housing) assets correlated with a higher probability that women will be employed in the non-farm sectors, particularly in urban areas (Lambert et al, 2011). There is a tendency to assume that female-headed are poorer than male-headed households, although there is considerable regional, national and local variation. In practice, the dependency ratio in households is the most important determinant of poverty, so young families with children are more likely to be poor, especially if they contain only one adult. Young women household heads with children are therefore more likely to be poor, especially if their natal families do not or cannot provide support. In addition, the numbers of older women in urban areas are increasing more rapidly than those of older men, as evidenced by markedly feminised sex ratios in senior age cohorts across developing regions. This is partly due to cumulative legacies of female-selective migration, especially in Latin America, and partly to women’s greater longevity (see Chant, 2013, p 11). Older women are disproportionately likely to be poor, if they do not have access to income or a pension, their families do not support them, and they do not hold or have use rights to family property (Varley, 2013). Whether older women are supporting or are being supported by family members varies. In some parts of the world widows are expected to remarry, but elsewhere they are not and must continue to earn a living . In India, for example, widow remarriage is socially stigmatised and Vera- Sanso found that lone widows may have to sacrifice opportunities for higher earnings in favour of lower paid work that is more regular, such as domestic work, leading to a downward spiral of ill health and impoverishment (Vera-Sanso, 2010). Land for housing in urban areas may be provided through the allocation of customary land to individual members or households who make up the wider ownership group. In these systems, women’s property rights are linked to their familial relationships, in particular with male relatives (see for example Ikejiofor, 2006, on Enugu in Nigeria; Leduka, 2006, on Maseru, Lesotho). Their use rights and concomitant responsibilities for household maintenance and childcare are secure as long as their marriages are intact, but reliance on secondary rights reduces their bargaining power and makes 19 them vulnerable to dispossession in the event of conjugal dissolution. In many African societies, according to custom widows are supposed to retain use rights while they remain unmarried and care for minor children, but the increasing value of urban property often tempts husbands’ families to claim the property and evict the widow (Rakodi, 2006). Sometimes widows may be entitled to land and property from their natal families, but more often in contemporary urban areas they are forced to move into inferior and more insecure accommodation where, like other women affected by eviction, they are vulnerable to violence and impoverishment, as found in COHRE’s research in Nigeria and India (COHRE, n.d., p 74). Family relationships and attitudes towards marriage and gender roles interact in complex ways with laws to influence men’s and women’s access to land and housing, their control over property, and their entitlement to inherit. Varley notes that the arguments for both preferring sons to daughters as heirs and putting property documents in a husband’s name largely hinge on the belief that it is first and foremost men’s responsibility to work and to house their families (Varley, 2010). Men’s obligation to provide for their family’s legitimizes their assumption of ownership. Research also suggests that “[m]any women do not assert their rights to property for fear of being seen as less than fully committed to the relationship” (Varley, 2010, p 91). While a union l asts, Varley suggests, women can fulfil gendered expectations of their roles in the family without having to assert their rights, and it is not until the relationship ends in separation, divorce, or death that the right becomes relevant (Varley, 2010). In an effort to address the often vulnerability of widow’s, Mexico has established a practice whereby officials “may allocate title to a woman who has been on her own for five years if she has a Certificate of Abandonment from the police and neighbours willing to support her claim” (Varley, 2007, p 1746). 2.2.3 Channels of supply for land and housing In addition to laws and family relationships, the ability of urban residents, including women, to access land and housing depends on the availability of suitable plots and houses, which is influenced by the operation of property markets and policies. Information on the impacts of land distribution processes, in terms of who gets access to assets on what terms, is incomplete and unsystematic. In particular, Deere and Doss (2006) suggest that we know relatively little about the gender asset gap because less information is available on wealth than on income at the individual and household levels. Reasons include the difficulties of assembling the relevant data; the complexity of marital arrangements; and the varying timing and of wealth transfers (e.g. dowry, bride price, bequests). These problems are compounded by the multiple types of claims and rights to land that can exist, and the associated difficulty of measuring ‘security’. While the private sector plays an important role in meeting the demand for urban land and housing, the need for public programmes to meet the needs of poorer households are well recognised. In practice, in urban areas of developing countries the need for affordable plots and dwellings is rarely met through formally regulated land subdivision or house construction by either the public or the formal private sector. As a result a large proportion of urban residents, especially poor households, must seek accommodation through informal channels. 2.2.3.1 Formal private sector land and housing delivery Private markets in residential land and property are, in theory, underpinned and regulated by state law governing tenure, subdivision, transfers, regulation of use and property taxation. 20 In middle income countries the financial and construction sectors may be sufficiently developed to finance speculative development, provide mortgages and produce large volumes of subdivided plots or completed houses/apartments for sale or rent. For example in Brazilian urban areas in 2005, one study suggests that about 70 percent of households owned their own house, and one-fifth rented in the formal sector (Morais and Cruz, 2009, p 266). However, a detailed analysis of housing starts and family income in São Paulo shows that “the formal housing market serves only middle - and high- income families,... [In 2007, it produced] very few units for families with incomes of 5 minimum wages or less, which represent almost 65 percent of the demand” (Haddad and Meyer, 2009, p 300-1). Formal sector supply is even more limited in low income countries. In low- and some middle-income countries, the cumbersome nature of legal processes, the limited management capacity of the agencies responsible and the cost of complying with official standards and legal requirements mean that the volume of officially subdivided residential plots lags far behind demand. In these contexts effective control over private land subdivisions is generally lacking, as a result formal housing often does not fully comply with regulations, and prices exceed the capacity of many urban residents to pay, even if they can access credit. If women can afford to purchase property in such a market, they may be less constrained by discriminatory attitudes and administrative practices than in attempting to access plots or houses through customary channels or from public agencies (UN- HABITAT, 2012, p 28). In general however legal, social and economic restrictions disadvantage women relative to men, whether as household heads or individuals. For example, married women in some countries are still legally or in practice unable to borrow from formal institutions without spousal approval. There may also be personal or social stigma associated with expressing a desire to own property in their own names. As a result, most are unlikely to be in a position to purchase land and property in the formal market or to have a formal tenancy agreement. For example, in Brazil, an analysis of data from the 2005 National Household Sample Survey shows that “wealth is a good predictor of formal home ownership... [but] More vulnerable households, such as the poor... and single women with children under age 14, have a higher probability of living in informal settlements, showing that they have limited tenure choice” (Morais and Cruz, 2009, p 278). 2.2.3.2 Public programmes for land delivery to the poor Public programmes to meet the housing needs of the urban poor have traditionally focused on sub- dividing publicly-owned land into serviced plots for ‘self-help’ house construction or the construction of complete housing units, usually for rent. This is particularly the case in the former planned economies. In practice, efforts to subdivide and service publicly owned land in volumes sufficient to meet demand and at costs affordable to poor residents have met with mixed and generally limited results. In addition, general shortages of affordable residential plots lead to ‘leakage’ up the income distribution of those supplied by public authorities. As a result, in most countries and cities, few poor households have been able to access publicly provided land and houses. Almost without exception, public programmes deal with households. In the earlier programmes, public housing for rent or sale was predominantly allocated in their own names to household heads in wage employment. Most beneficiaries were men owing to their domination of wage labour markets, and joint titles were rare. No long term tracking studies of publicly provided serviced plots seem to be available. 2.2.3.3 Alternatives to formal land and housing supply: multiple channels 21 Because the supply of suitable land and dwellings by formal private or public providers falls far short of demand, many (sometimes most) middle- and low-income households have to obtain land through informal channels. Many poor people seek out unused land in the interstices of urban development within easy reach of economic opportunities (for example in areas unsuitable for residential use or on pavements) or occupy rental accommodation in tenements or employers’ housing. Renting accommodation In many cities, poor people have little choice but to rent a dwelling (in a formal or informal building), although evidence suggest that most would prefer to own a plot and house. Single people or recent migrants, whether men or women, generally stay with relatives or rent a room, especially if they only intend to stay for a short period. For example, women industrial workers in South East Asian countries such as the Philippines may share a rented room, sometimes sleeping in shifts (Chant and McIlwaine, 1995, p 161ff). Some long-term in-migrants intend to return to their area of origin. They may choose to remit their savings to rural relatives or invest them in farming and/or building a rural home rather than investing them in urban construction. Other households become tenants because they are unable to buy a plot or house (in a formal or informal area). Some rent from relatives (perhaps at a reduced rent) and many obtain a room through social contacts. As a result, frequent moves within the same neighbourhood are common, especially because relations with landlords in overcrowded accommodation can be fraught and it is difficult to keep up regular rent payments when incomes are irregular. Because women are often disadvantage in terms of education, income earning capacity, marital position and ability to become home owners, when they head households, they are disproportionately represented among renters (UN-HABITAT, 2003; 2011b, p 16; 2012, p 28). For example, 33 percent of households in urban areas in Ethiopia are headed by women (compared to 17 percent of rural households). Three quarters are poor and the majority live in rented housing (70 percent in a sample of female-headed households in two kebeles in Addis Ababa, 12 percent of whom were single, 52 percent divorced or separated and 36 percent widows) (Eskezia, 2011, p 24). Women landlords may prefer to let to women tenants, but the latter may be discriminated against by male landlords and are particularly vulnerable to bullying, violence and arbitrary eviction (Baruah, 2007; Chant, 2013). Thus in the absence of legal safeguards and effective enforcement, women tenants are typically even less secure than owner-occupiers. Their insecurity is exacerbated when they rent accommodation in informal settlements where landlords themselves have only limited security. Women play a role in the rental housing market not only as tenants but also as landlords. Many low income urban residents let out rooms in the house they occupy, either by building additional accommodation or by vacating rooms in times of economic difficulty or when their households shrink, perhaps as children leave home. Others own accommodation for rental in which they do not live. While there is large scale commercial landlordism in some cities (e.g. Nairobi), many landlords, especially in informal housing areas, have incomes little different from their tenants (Kumar, 2010). Letting accommodation can be a desirable strategy for women, especially older women who are less able to access wage employment or pursue arduous informal sector activities. However, their ability to become landlords is affected by whether:  they can raise the funds for construction – often they are less able to invest in housing for rent than men;  housing construction and management is traditionally seen as a man’s or woman’s role (as it is, for example, in Botswana, Datta, 1995); 22  they can inherit the marital home if their husband dies. For example, Varley (2007) notes that in the inner city areas of Guadalajara and Puebla, Mexico, renting is often referred to as a “widow’s business.” Similarly more women than men are landlords of backyard shacks in South Africa, where many men have died of HIV/AIDS or abandoned their families. However, elsewhere the marital home is claimed by the husband’s family, especially once the children are grown or the widow remarries.  the social acceptability of women being landlords, especially of male tenants. As a result, the proportion of landlords who are women varies, although our comprehensive understanding is limited by lack of data (Kumar, 2010). Home ownership Informal housing supply channels can be considered in terms of: the proportion of total demand they meet; the extent to which they comply with formal regulations; and/or the degree of security they provide to occupants. Today, the proportion of households which can obtain access to free or low cost land through incremental squatting or invasion has declined because areas of undeveloped publicly-owned land have diminished and those holding legal land rights can generally either subdivide and sell their land or use the formal court system to evict squatters and others with informal occupancy arrangements. Although squatting is a cheap way of obtaining land, limited opportunities, the physical hardships and the level of vulnerability rule it out for many women. In Africa in particular, cities are often surrounded by land held under customary tenure. Within the relevant group (extended family, lineage, clan etc.), homestead land is retained and rarely alienated, although plots may be allocated to those with entitlements – generally men upon marriage and rarely women in their own right (for example around Enugu or Maseru, Ikejiofor, 2006; Leduka, 2006). In addition, members of successive generations in a family may have rights of residence and/or ownership in a family house in a central or consolidated area within an urban centre (e.g. in Ghana). In these systems, women’s property rights are linked to their familial relationships, and so are usually secondary rights – in the past and in some countries today it is uncommon for family land to be allocated to or inherited by women, although there are signs of change. In Lesotho, for example, 54 percent of urban households are headed by women, 25% de jure (never married, divorced, widowed) and 29% de facto (husbands working in South Africa). Between one third and one half of plot owners in samples from three informal settlements were female-headed. Many of those allocated plots were de facto household heads at the time they had been allocated plots through customary channels (by the local chief, family head or by inheritance), while few of those who were de jure heads at the time of distribution had been allocated land. Instead, most de jure women heads had purchased their plots independently – these women had educational levels above average for plot owners and were among the least poor in the informal settlements studied (Leduka, 2004, p 167-9). Probably most common today is the informal subdivision of land to which the sub-divider has legal or customary rights, although illegal occupation or unlicensed subdivision of public land continues to occur where there is unoccupied land in public ownership. Such processes operate on the basis of informal social rules, which may mimic and draw upon elements of the formal system (Byamugisha, 2013; Fernandes, 2011; Rakodi, 2006) They provide access to land for low and middle income households, including women who can both act independently and afford to purchase land, but not to the poorest. For example, in Eldoret, Kenya, most of the land available to low and middle income 23 households is in areas legally owned by land buying companies, the shareholders in which are allocated plots for occupation and/or further subdivision. A sample survey showed that 16 percent of plot owners were women, probably 80 percent of whom had been married at the time they acquired their plots, and only 20 percent were never married women who had obtained land in their own right, mostly through purchase. Although women respondents were clear that the land buying companies were willing to accept female shareholders, the social pressure to marry and the disadvantaged financial position of women meant that relatively few women held property in their own names (Musyoka, 2004, p 115-6). The ability of residents in informal settlements to claim their rights as citizens, demand services and defend themselves (especially against eviction) varies, depending on government attitudes, their political connections and the nature and scope of community organisation. Although women generally lack voice in the formal political system, grassroots organisation often depends heavily on their inputs. Typically public services such as water, sanitation and refuse collection are initially absent, but over time, buildings are consolidated and some services are provided by either public or private suppliers – the latter tend to be more expensive. Service improvements often stimulate more house construction. The financial regime (payment for infrastructure and services, property taxation) varies – sometimes users are charged for services and are expected to pay property tax, but often informal settlements do not generate any revenue. After many years of international and national advocacy, governments have increasingly adopted accommodative policies towards informal settlements, varying from neglect through basic service provision to regularisation. These policies and their outcomes will be explored in Section 3. 2.2.3.4 Women’s access to land and adequate housing: an overview In the absence of data on urban land and housing ownerhsip, it is impossible to summarise women’s land and house ownership accurately, nor to make international comparisons. Attempts to make overall estimates are few and adopt different methods. UN-HABITAT has made the most systematic efforts, but in recent years these have focused on the need to monitor achievement of MDG Target 11 (to improve the lives of at least 100 million slum dwellers by 2020) and so have focused on monitoring the number and proportion of urban households living in ‘slums’. Slum households are defined as those lacking one of more of the following five amenities (shelter deprivations):  durable housing (a permanent structure providing protection from extreme weather);  sufficient living area (no more than three people sharing a room);  access to improved water (water that is sufficient, affordable and can be obtained without extreme effort);  access to improved sanitation (a private toilet, or a public one shared with a reasonable number of people); and/or  secure tenure and protection against forced evictions. Because information on secure tenure has not been available, monitoring of progress towards Target 11 has been based on the first four of these indicators. The only gendered data available reveal that between 1997 and 2003 about 20 percent of all households in the 160 cities included in the 2006 Urban Indicators Database were female-headed (although the proportion varies greatly between and within cities). The proportion of households with women heads living in neighbourhoods with three or more shelter deprivations varied from less than 20 percent (15 percent in Tanzania, 17 percent in 24 Indonesia and Armenia, 19 percent in Nepal) to over half (51 percent in Ghana, 57 percent in Haiti). In Kenya and Nicaragua, one-third of female-headed households experienced all four shelter deprivations, suggesting that “in some countries, woman-headed households disproportionately suffer from inadequate housing in poor urban neighbourhoods” (UN-HABITAT, 2008b, p 105). Efforts to define the concept of secure tenure, devise suitable indicators and collect relevant data began in 2002, led by UN-HABITAT, with a view to developing a mechanism that is gender sensitive, measures and defines types of tenure security beyond formal registration of land (conceptualised as a continuum of land rights), and monitors evictions and access of urban residents to land rights (UN-HABITAT, 2011a, p 7; UN-HABITAT and IIRR, 2012). Building on work by other agencies and a series of UN-HABITAT pilot exercises, an operational definition of secure tenure has been agreed and a possible methodology for assessing it at the individual unit, settlement and city levels devised. The pilot in São Paulo made did not include gender considerations (UN- HABITAT, 2011a). In an alternative approach focused on the assets or wealth owned by women, Deere and her colleagues reviewed examples of household sample surveys (LSMS and similar surveys) from around the world, particularly in Latin America. In this urbanised region, they found that housing data are available in most countries and housing appears to be the “main asset whose ownership appears to be ... equitably distributed..., with women’s share of homeowners nationally ranging from a low of 27 per cent in Guatemala (in 2000) to nearly 50 per cent in Nicaragua (in 2005) and Panama (in 2003)..., with urban women constituting a much larger share of total homeowners than rural women in these countries” (Deere, 2010, p 348). This study finds however that the gender analysis of poverty is primarily limited to analysing the differences between male and female household heads (based on per capita indicators). This potentially has value, given the increasing proportion of all urban households that are headed by women (see Chant, 2013, p 12 for Latin America). However, it tells us little about the position of women and men within households, especially women in male-headed households. 3 Recent reforms to land administration Ideally a review of land and housing ownership and acccess should assess the extent to which policies, programmes and legal reforms have achieved relevant outcomes: well-located affordable serviced plots and/or dwellings and security (sustainable livelihoods and personal safety as well as secure tenure). It should assess whether the constraints that prevent poor women and men from satisfying these needs have been successfully alleviated by measures aimed at facilitating the private market supply of appropriate plots/houses, improving living conditions in existing informal settlements and/or the public provision of plots/houses. However, systematic evaluations of the impacts of policies, programmes and legislative change are scarce and gendered assessments even rarer.8 For example, a recent meta-analysis of studies that analysed “the channels through which land titling affects women’s productivity and earnings in agriculture, as well as other indicators of women’s well-being and autonomy” (Rodgers and Menon, 2012) found urban material from only 8 Most of the available material reviews legislative and policy change (e.g. UN-HABITAT, 2007a; Hallward- Driemeier and Hasan, 2013; OHCHR and UN-Women, 2013) or analyses the history, design and characteristics of government programmes, assessing progress with implementation rather than evaluating outcomes and impacts. 25 three countries – India (Datta, 2006), Peru (various studies by Field) and Tanzania (Ayalew Ali et al, 2011) (see below). Relevant data and evidence is lacking to enable a comprehensive review in light of the dearth of data and evidence, this section will consider legislative changes and other programmes designed to improve the access of poor urban residents, especially women, to land and housing, concentrating on countries and cities where these have been implemented on a relatively large scale. The review identfies ways in which gender equality can be mainstreamed in land and housing policy and practice, as well as specific measures to improve women’s access to land and housing. The role legislative reform provides as the basis for other programmes, influencing their content and outcomes, will be addressed first and further referred to during the subsequent analysis of efforts to provide secure tenure and adequate housing through the regularisation of informal settlements, the privatisation of publicly owned land and housing, and large-scale house construction. Particular attention will be given to the outcomes of alternative approaches for women. 3.1 Legal reform Most commonly, legal reforms have focused on property law, specifically titling, while changes to related laws have received less attention. Titling programmes are usually, but not always, national programmes aimed at establishing the legal basis and administrative arrangements for issuing and registering titles for land, as well as the subsequent transfer (rental, sale, inheritance etc.) of undeveloped land and land that has already been developed but for which owners/holders lack legal ownership documents (regularisation, involving adjudication prior to registration). Titling programmes aim to:  provide tenure security;  encourage investment in upgrading;  increase access to formal credit for building and business investment by using the titled plot as collateral;  enhance municipal revenue by integrating plots into the property tax system;  encourage active land and housing markets, to contribute to economic development and poverty reduction. For the most part titling programmes are designed to register full legal rights, but some may register rights other forms of rights, such as use rights based on a certificate of occupancy, license or permit rather than a title or lease (Payne and Durand-Lasserve, 2013). In India, for example, following the adoption of a policy of regularisation by the central government in the mid-1980s, various State governments introduced legislation permitting the grant of pattas (usually non-transferable leasehold rights, typically for 30 years) to squatters on suitable sites after five years of occupation (Ansari, 1998; Banerjee, 2002, 2004; Payne and Durand-Lasserve, 2013). In Tanzania, the 1999 Land Act provides for the issue of short term renewable leases (non-transferable residential licenses conveying the right to occupy land for two years) and longer term titles (certificates of the right of occupancy) in unplanned settlements (Kironde, 2006). Growing experience with the Social Tenure Domain Model demonstrates that it is possible to register different forms of rights in informal settlements (UN- HABITAT, 2012).9 Collective ownership may be adopted if it is too difficult to demarcate individual 9 The Social Tenure Domain Model is a software package that is compatible with standard approaches to Land Information Systems, but capable of being used to record various types of tenure rights in informally settled areas. It uses handheld GPS and community enumerations to record geo-referenced data for the purpose of registering rights and transfers, planning and upgrading (UN-HABITAT, 2012). For an account of its use in Uganda, see Antonio et al (2013). 26 plots in a regularised settlement or to prevent gentrification (see below). Both individual and collective rights are provided for in Namibia’s 2012 Flexible Land Tenure law. The legal framework may also specify what rights title holders receive, in whose names title may be registered (individual, joint, a group/community or a business). How titles are registered and in whose names are influenced by policy as well as the preferences of owners, and thus vary. It has been suggested that, building on the idea of a continuum of land tenure and property rights, a typology can be developed, ranked in terms of the degree of security conveyed by different combinations (Durand-Lasserve and Selod, 2010, p 103; Payne and Durand-Lasserve, 2013, p 21). The nature and level of existing and potential rights are determined by the degree of formality, the regulatory framework, the extent to which transactions are recognised by the public authorities and other local actors, and the contract between parties to land transactions, which generally specifies the period of time for which the rights are transferable, the use and type of development, and whether the agreement is renewable or transferable. These categories can also be refined to show whether or not rights are available to women as well as men. Such a typology is a potentially useful analytical tool for identifying the full range of tenure options available in a particular city, and assessing the need for reform of local tenure arrangements. Many countries have adopted titling and regularisation policies (for example, 17 Latin American and Caribbean countries, according to Angel et al, 2006), but large-scale registration has been achieved in relatively few and implementation in most places has lagged far behind targets. In addition, because registration/titling is rarely, if ever, introduced separately from related policies and administrative changes - such as provisions for land use planning and subdivision regulation, infrastructure installation and service delivery, and housing policies and programmes - it is often difficult to attribute effects to registration of tenure per se. Assessments of titling and registration programmes are often concerned with the implementation and outcomes of national programmes in rural areas, or programmes confined to rural areas alone, and relatively few are available for urban programmes (Payne et al, 2010; Fernandes, 2011). The analysis is also often not gendered; nonethless the discussion below focuses on the few assessments that have considered the effects on women. Reforms to family law have sought to ensure that women can acquire property in their own names, retain ownership of their own property during and after marriage, and inherit property from their natal families and/or husbands. For example, the Indian 1955 Hindu Marriage Act affirms that property owned by a woman on marriage remains her own property. Amendments to the 1956 Hindu Succession Act, which governs intestate succession for 80 percent of India’s population, specify that daughters should have equal rights with sons to inherit equal shares of joint family property. These were initially introduced by various State governments and then at the national level in 2005. Only one of the studies that have attempted to assess whether the amendments have resulted in increased ownership of property by women, or associated benefits (for example, more education and later age of marriage for girls, more autonomy for married women within their households) focuses on urban areas. A survey of three generations of Hindu and Muslim household members (household heads, their spouses, and the siblings, parents and children of both) in four cities in Karnataka, where the Act was amended in 1994, found that the reform had:  increased the share of assets (especially joint property) received by women who were single before the reform by over 20 percent if their fathers passed away after it came into force, but that fathers willed away separate property from daughters who were single before the reform, so they received less separate property in the long run; 27  increased the dowry/gifts received by women who married after the reform by about 15 percent;  reinforced human capital investment in daughters, demonstrated by an increase in the educational attainment of girls enrolled in primary schools since the reform compared to that of boys, leading to improved long term prospects (Deininger et al, 2013). Changes to inheritance law and practice are widely needed, but clearly the scope for and outcomes of such changes are likely to vary. 3.2 The recognition, regularisation and upgrading of informal settlements Given the growth and prevalence of slums and informal settlements much attention has been given to how living conditions in existing substandard housing areas might best be improved. Views differ about the most appropriate way of tackling the problems, especially insecure tenure and inadequate services. Not all informal settlements are suitable for permanent settlement, because the land on which they are located is dangerous or is required for legitimate alternative uses. In addition, well- located sites are attractive to private developers, providing that their occupants are removed. When resettlement is pursued, compliance with laws and guidelines designed to protect occupiers is key, especially for women, are particularly vulnerable to abuses during and after evictions (COHRE, n.d.). 3.2.1 Recognition The quickest way of providing residents with greater security and improved services is to recognise an informal settlement. This is generally done informally, either through a verbal or written promise that a settlement will be permitted to remain on its current site, either temporarily or permanently. Presidential recognition has been an important part of policy and law in the Philippines (UN- HABITAT, 2011c) and some municipalities in India (e.g. Ahmedabad) prefer this approach because most slums are on private land (Baruah, 2010). Government actions such as collecting tax or providing water and sanitation may have the same effect of legitimising the settlement. The evidence shows that recognition can increase perceptions of security sufficiently for residents to invest in their houses and negotiate with service providers (Varley, 2007). Practically however because the promise if informal, it may not survive changes in political control, personnel in the relevant government agency or policy towards informal settlements, or the prospect of a more lucrative use for the site concerned. Informal recognition therefore does not provide legal protection or long term security. 3.2.2 Regularisation Regularisation involves the formal legalisation of an area and registration of ownership, typically through the issue of individual titles (in perpetuity or for a limited period) to owners of buildings. Regularisation is expected to improve tenure security, improve access to formal credit, increase investment in house improvements and/or infrastructure, and improve the economic status of poor households. In most urban areas, claims to land are complex, land is potentially valuable and there are multiple competing interests, so that regularisation is complex and takes years. It is easier on publicly owned land or where the original squatting/invasion had political backing or was tolerated. Depending on who owns the land and other factors, the financial costs vary widely, and residents may or may not be expected to contribute (for example by paying property tax after regularisation). Determining eligibility for the issue of a title can be tricky. Usually, residents are only eligible if they can demonstrate house ownership in the area before a defined date, but such dates are often arbitrary, discriminating against recent arrivals (who may include women evicted from the marital home 28 following marriage breakdown), and obtaining proof can be particularly difficult. In addition, the plots and houses in informally settled areas often have lower standards and higher densities than those permissible under planning and building regulations. While some governments recognise that the official standards are unrealistic and unaffordable for low income households and are prepared to compromise, others are not. If an area is re-planned to ensure that all plots comply with minimum standards, an alternative location must be identified/purchased to accommodate those who lose their plots. Such areas are often less well located than the original settlement, disrupting social ties and work patterns, with implications for women that may not be recognised. In addition, resettlement increases the costs for both the public agency and those forced to relocate. Where registration is demand-driven rather than mandatory, the cumbersome procedures, exacting requirements and cost may deter many occupiers from applying, especially households with low incomes, single adults and less literate heads, all of which are often disproportionately headed by women. Sometimes, when owners consider allocation documents to provide sufficient security, the benefits of completing titling are outweighed by the costs. For example, in Dar es Salaam, Tanzania, in the first ten years after the 1999 Land Act was passed, only 90,000 of the estimated 400,000 owners of informally subdivided plots been issued with licenses, and few of these expressed interest in applying for full title (Kironde, 2006, 2009). Registration of subsequent transfers of land (sales, inheritance etc.) are deterred for the same reasons, with the result that formalisation is soon followed by de-formalisation (see, for example, Galiani and Schargrodsky, 2011 on Buenos Aires, where 29 percent of the titled parcels in the settlement studied had become de-regularised because of the costs of formal divorce and registration; Varley, 2010 on Guadalajara). Although transfers can be expected to have gendered effects, the available analyses are not gendered. Improved tenure security The available evidence suggests regularisation does improve tenure security for those residents who receive registration of their land rights. Registration is valued because it provides a stable base for their livelihoods and families and gives an asset they can bequeath. However, security may not dramatically improve if perceived security was high prior to titling (e.g. in Buenos Aires, see van Gelder, 2013; Guayaquil, Ecuador, see Lanjouw and Levy, 2002; Dar es Salaam, see Kironde 2006). In addition, the benefits may be undermined by the additional costs for which titled owners may be liable. Furthermore, not all the owners of land in informal settlements are poor, and often some are absentee landlords, so regularisation may reinforce existing inequality. Finally, ownership rights are rarely allocated to tenants and regularisation is generally followed by rent increases, so tenants, many of whom may be women-headed households, are often disadvantaged as a result of regularisation processes (see, for example, Baruah, 2007, on upgraded slums in Ahmedabad, India). Assessments have considered whether women household heads have benefited equally to men, and the effects of women owning property in their own names or the issue of joint titles. In Peru for example, rapid and large-scale regularisation through titling of plots in informal subdivisions on the urban periphery since the mid-1990s has been possible because the land is publicly owned and barren, earlier initiatives had prepared the way, and there was Presidential commitment and external funding (Angel et al, 2006; Atuahene, 2004; Calderón, 2004, Field, 2007). Between 1996 and 2006, the specialist agency concerned issued nearly 1.6 million titles, reduced the time taken to obtain a title from 7 years to 45 days, and reduced the cost from US$2,156/title to an average of $64 (Fernandes, 2011). Titles are issued jointly where couples are married (almost two thirds of beneficiaries are 29 married couples) (Consorcio Informet-Input, 2002, p 34, quoted in Varley, 2007, p 1747). Twenty- three percent of both titled and as yet untitled households surveyed in 2000 were female-headed, a similar proportion to the urban population as a whole (Field and Torero, 2004, p 33; Tacoli, 2012, p 16). The law does not however recognise consensual unions and so for such couples titles are issued in the name of the male head of household. Three quarters of respondents in both titled and yet-to-be titled households in 2000 cited tenure security as the main benefit of obtaining title (Field, 2004). A study using data from urban Peru exploring whether joint titling empowers women within the family - using fertility decisions as an indicator - finds “a 22 percent reduction of fertility for squatters who received a property title and ...that females who received joint title experienced twice the reduction in the probability of having a child than those where the title is in one name only” (Field, 2003, quoted in Deininger and Feder, 2009, p 249). Insufficient information is available however to determine whether this effect can be solely attributed to titling. Doss’s review of the literature indicates that in addition to assets, women’s education and incomes are important to their intra-household bargaining power, and that there is as yet insufficient evidence to show how this is affected by different policies (Doss, 2013). In India, the regularisation process varies between States and cities. Pattas vary in form and duration. They are issued in women’s names in some States and joint names in others. There have been no systematic assessments of the outcomes, despite the scale of some of the programmes (Banerjee, personal communication), although a few studies have examined specific aspects. For example, Agarwal and Panda (2007) set out to test whether women’s ownership of property in their own names is as associated with reduced levels of domestic violence. Their panel survey samples of ever-married women in Thiruvananthapuram District in Kerala included both urban and rural households. The results showed that women’s ownership of a house or land significantly reduced their risk of marital physical or psychological violence, because ownership of immovable property “provides a woman economic and physical security, enhances her self-esteem and visibly signals the strength of her fall- back position and tangible exit option” (Agarwal and Panda, 2007, p 359). This has the effect of deterring violence and providing an escape if violence occurs. Seventy one percent of the propertied women who experie ced long-term physical violence between 2000-1 and 2004-5 had left home and few had returned, compared to 19 percent of property-less women, almost all of whom went to live with their parents and half of whom had returned to their husbands. The results were equally true for urban and rural women. The study also demonstrated that women with property of their own had a greater say in household decision making (specifically, taking out loans, using contraceptives and having sex with their husbands). Gupta (2006, quoted in Agarwal and Panda, 2007) found that in Bengal, where women bring little real property to marriage, the same association between property ownership and domestic violence occurred. In practice in India, much of the property women bring to marriages is dowry (although not all dowries include real property), and how much of this remains the property of the bride (stridhan) varies and is contested – much of it is beyond her control, “not women’s wealth, but wealth that goes with women” (Sharma, 2005, p 21). Datta (2006) examines the effects of joint titles, which have been issued in Punjab since 2000, through a study of married couples in three settlements in Chandigarh (one area where pattas had been issued in the name of the husband, one in which they had been issued in joint names and an unregularised settlement). She found that women value joint tenure because of the greater security and bargaining power it conveys. Respondents reported that it legally enables them to prevent their husbands from selling the house without their consent, and leads to them being more involved in household decision making, less vulnerable to abandonment, more able to leave an unsatisfactory marriage, more secure should they become widows and in old age, and more able to give their 30 children a stable future (Datta, 2006). So strong were Datta’s findings that she dubbed joint titling a “win-win policy.” Other research, for example Vera-Sanso’s in-depth analysis of households in low income settlements in Madras/Chennai, suggests that the outcomes of putting titles either solely in women’s names or jointly in husband and wives names are variable and unlikely to straightforwardly empower women (Vera-Sanso, 1997). Large scale registration of titles in Rwanda has been under way since 2010. By March, 2012, 83 percent of parcels countrywide were registered in the joint names of married couples, 11 percent in the name of a woman and 5 percent in the name of a man (Ngoga, 2012, p 5). Daley et al (2010) and Ayalew Ali et al (2011) report similar beneficial effects to those found in other countries for rural households, despite resistance from men and their families, the continuation of discriminatory cultural practices and the exclusion of consensual unions. However, data collected for a sample of households in peri-urban Kigali were not included in Ayalew Ali’s analysis, and there has not been any evaluation of the outcomes of registration in urban areas to date (Nkurunziza, personal communication). In Costa Rica, where 27 percent of all households are headed by women, Chant (2009) found that despite an overall decline in the incidence of extreme poverty (to 5.6% in 2005), the proportion of female-headed households amongst the poor had increased to 33.5% and amongst the extremely poor to 43.5% (p 27). Her qualitative data showed that female headship was rising, because women saw it as having advantages and being more feasible than in the past. The promotion of titling among women, in addition to legal changes that grant them greater entitlements to conjugal assets and protection from domestic violence, had strengthened their ability to survive alone and also made them wary of letting men get too involved in their lives lest they lose control over their assets (Chant, 2009). In contrast, another study reported that although joint title in a housing project in Heredia, Costa Rica, had given women more bargaining power during divorce, it also meant that couples were likely to stay in unhappy or violent relationships after they would otherwise have split up, since neither could afford to buy the other out (Blanco Rothe et al, 2002, quoted in Varley, 2007, p 1747). Despite growing recognition that joint titling can safeguard and empower women during and after marriage, many titles issued during regularisation continue to be issued solely in the name of husbands. For example, a 2003 survey shows that 63 percent of the titles issued during the Ghana Urban II Project (1994-7) were in the names of men, 25 percent in the name of women (some of whom were married) and only 10 percent in joint names (mostly businesses) (World Bank, 2005, p 52). Similarly, in Kinondoni municipality in Dar es Salaam, where most owners were married, 65 percent of the licenses issued between May and September, 2006, were in the name of men and only 5 percentin joint names. Informants attributed this to poor design of the forms, staff attitudes and men’s and women’s preferences (Kironde, 2006). Although the Land Act provides husbands and wives with equal shares in marital property and requires spousal consent for transfers, it appears that in the absence of both names being recorded on titles, these provisions are not respected. In response to the poor uptake of provisions for joint titling, a path-breaking experiment was launched in October 2010 to assess the impact of reducing the cost of obtaining a full title (Certificate of the Right of Occupancy) and making discounts conditional on including a woman on the title application. Randomly selected parcels were issued with discount vouchers, half of which were conditional on including a female household member as joint owner on the application. Results suggest subsidies had a positive impact on the purchase of land titles, households receiving the conditional subsidies were just as likely to purchase as those receiving unconditional subsidies, and the receipt of a conditional subsidy substantially increased the probability that a woman’s name was included. The authors claim that the experiment demonstrates that it would be relatively easy to design and implement incentives 31 to encourage joint titling without reducing demand for titles, but also that the ease with which couples were nudged into co-titling might indicate that men did not take its possible impact on gender relations within the marriage very seriously (Ayalew Ali et al, 2013). i. Improved access to credit Evidence suggests titling does not automatically improve access to formal credit for either housing investment or business activities (Deininger and Feder, 2009; Galiani and Schargrodsky, 2004; Kironde, 2006; Payne et al, 2008; van Gelder, 2013; Varley, 2010). The primary obstacles include that formal financial institutions are unwilling to lend to poor people, including women, or to accept low value property as collateral because of the difficulty and cost of foreclosure, and low income owners are unwilling to risk losing their only significant asset, which they regard as a family rather than individual asset (Varley, 2010). There is, however, some evidence that if subsidised public loans are available for house improvement, owners are more likely to borrow. For example, nearly half of titled households in Peru had applied for loans, almost all from the public sector Materials Bank. The applicants tended to have lower socio-economic status than owners in regularised areas overall, including a greater fraction of female-headed households (Field and Torero, 2006, p 14). ii. Increased investment in housing and/or infrastructure Generally, regularisation is associated with increased house investment despite the lack of access to formal credit. For example in Peru, titled households were “10 percent more likely to have undertaken housing improvements in the two years prior to the survey” than yet to be titled owners (56 percent compared to 46 percent) (Field and Torero, 2004, p 9, 33). Some studies do not control for income, so the relative effects of regularisation and ability to afford investments cannot be determined. For example, Banerjee (2004) noted that investment increased following the issue of pattas to occupiers in Delhi, Bhopal and Visakhapatnam in India, especially in settlements previously scheduled for removal, but this also depended on the ability of households to afford it (see also the studies of Bhopal by Mitra, 1988 and Rizbud, 1988, quoted in Ansari, 1998). Where titled and untitled households have similar pre-treatment socioeconomic characteristics or incomes, as in the natural experiments studied by Galiani and Schargrodsky (2010) and van Gelder (2013) in Buenos Aires, titling has been found to have an independent positive effect on investment, because of the reduced risk of eviction and households’ attachment to their house and neighbourhood following a long struggle to overcome the hardship, insecurity and social stigma associated with residence in an informal subdivision (van Gelder, 2013). Savings are often the main source of capital for this purpose, illustrating the important potential contribution of opportunities for small scale saving. 10 The experience of national slum dwellers’ federations in several countries shows that women in particular value the opportunity to save and are active participants in savings groups. For example, in Malawi, “participating in savings groups is generally viewed as complementary to domestic tasks by powerful figures within the family. The fact that achievements reinforce women’s existing reproductive role ([among other things enabling women to] …address housing improvements) reduces the likelihood of patriarchal challenge …. The altruistic element…is seen as ‘morally legitimate’ ….It is the immediate engagement in local development activities, running meetings, managing finance, constructing homes, surveying their neighbours and 10 A household survey in Bhopal, India, in 2003, showed that almost seven out of ten of the low income households which had moved out of slums had relied on savings as their source of housing finance. While almost a fifth had borrowed informally, fewer than a tenth had obtained a formal loan. The ability to save regularly was the most important determinant of households’ ability to move (Lall et al, 2006). 32 digging sewers that enables women [in Malawi and elsewhere] to discover their own capabilities and future potential” (Satterthwaite and Mitlin, 2013, p 217-8). iii. Improved economic status of poor households Regularisation is also expected to contribute to poverty reduction by improving the economic status of owner households, through increasing the value of their property and enabling them to invest in economic activities and/or human capital. In addition to encouraging investment in house construction and improvement, the value of property increases following regularisation. For example, Galiani and Schargrodsky (2010) found that titled plots in their study area in Buenos Aires were valued at nearly a fifth more than yet-to-be titled plots. Similarly in urban Ecuador, Lanjouw and Levy (2002, p 988) found that “having title is associated [by owners] with a sizable increase in the expected market value of their properties – on average almost 24 percent of untitled property values” and even more in newer, less well-established areas. This increased value also benefited ‘vulnerable’ households, defined as those containing no adult men, who had been less able to defend informal rights. It is sometimes suggested that men may be more inclined to realise the increased value of a house by selling it (potentially without the consent of their partner), while for women retaining ownership to provide a secure family base and a potential location for income generation through home-based enterprise or renting rooms are more important. The sale of the family home trades off short term gain for potential long term disadvantage, if households have no choice but to settle in another slum area, potentially disadvantaging women and perpetuating the vicious cycle of informal settlement, as well as carrying the risk of gentrification, enabling higher income buyers to benefit from subsidised regulation and undermining community cohesion. Various ways of preventing such sales and gentrification have been tried, such as joint titling, issuing titles in the name of the woman (e.g. pattas in India) or collective titling, for example in Community Land Trusts (Lipman with Rajack, 2011, p 25-30), the Philippines Community Mortgage Programme (Berner, 2000; Porio and Crisol, 2004; UN- HABITAT, 2011c) and the Baan Mekong programme in Thailand. The latter, launched in 2003, is based on collective purchase and ownership of land by communities. Although apparently successful and implemented on a large scale (858 projects in 277 cities, involving 91,000 households in 2013 (http://www.codi.or.th/housing/results.html), no gendered evaluation appears to have been carried out (Archer, 2012; Boonyabancha, 2005, 2008; Satterthwaite and Mitlin, 2013, p 150-9). As noted above, titling does not provide low income households with access to formal credit for investment in business activities, as commonly hypothesised, so more recently it has been suggested that they may instead invest in human capital, by devoting more household resources to health and education. Galiani and Schargrodsky (2004, 2010) found modest improvements in the health of children, fewer children (especially in younger families), lower teenage pregnancy rates, higher levels of school attendance, more years of schooling on average, and a higher proportion of children completing secondary school in households with titles compared with those yet to be titled. They and others suggest that allocating solid property rights to the poor might reduce poverty in future generations through investment in physical and human capital. 3.2.3 Upgrading Insecure tenure is not the only problem faced by residents of informal settlements and titling may be insufficient to enable investment in human capital if the addition challenges are not also addressed. In particular, sites may not have been reserved for social facilities such as schools and residents may lack 33 access to infrastructure and services, the opportunities and skills needed to earn an adequate living, and a voice in local decision making. Often, regularisation is impossible or undesirable without tackling these other problems, so it is accompanied by upgrading, especially infrastructure installation and service delivery, and sometimes re-planning (requiring the resettlement of some residents). However, views differ on whether upgrading needs to be preceded or accompanied by regularisation of tenure, as the latter can delay other vital improvements and is not necessarily required to improve owners’ security sufficiently for them to invest in house improvements. In addition, the cost and complexity of regularisation combined with upgrading generally means that it is only accomplished on a smaller scale than titling alone, benefiting selected informal settlements and fuelling the tendency for gentrification of upgraded areas (Fernandes, 2011). Unlike Peru, Brazil has mostly, since the 1980s, adopted a combined regularisation and upgrading approach to improving conditions for the estimated 2.36 million households living in 11,754 favelas in 2000 (Fernandes, 2011, p 14). At US$3,500 – 5,000 per household, upgrading is more costly than titling alone, but municipalities have made more progress with it. Overall, Fernandes estimates that by 2009 regularisation had been initiated for 1.7 million households in 385 municipalities, but only 22 percent of those households had been issued with titles (Fernandes, 2011, p 32). Efforts to make upgrading affordable, avoid resettlement and prevent gentrification include pegging standards below those attractive to middle and upper income households. No gendered evaluations have been identified, although accounts of experience indicate that even organised women residents may struggle to influence decision-making.11 Upgrading has also played an important part in India’s policies for informal settlements since the 1970s, under acts that provided for designating some slums for improvement and provision of basic services, coupled with a guarantee of non-eviction for at least ten years following infrastructure installation. For example, between 1972 and 1996 about 40 million people benefited from the Environmental Improvement of Slums programme (Banerjee, 2002, p 39). Designation improves security and in theory, a guarantee of resettlement if a slum is declared unsuitable for designation. Because regularisation and the issue of pattas is a complex, slow and relatively costly process, most residents prioritise improvements to basic utilities, employment opportunities and incomes, and schools and health facilities over and above regularisation, as demonstrated by a study in Lucknow (Kantor and Nair, 2005). Similarly, Gulyani and Bassett, in their review of experiences of upgrading in African towns and cities since the 1970s, note that whereas the early approaches were based on the assumption that “tenure security is a precursor for housing and infrastructure investment, current programmes tend to use infrastructure investment as a means for enhancing tenure security and encouraging housing investments” (2007, p 487). Overall, therefore, most analysts agree that the formal registration of land ownership to convey de jure security of tenure generally benefits house owners and can potentially benefit women. However, in many places progress has been limited, it does not necessarily benefit all households equally, it may increase insecurity for some, and the potential benefits for women are not always realised. In some contexts, infrastructure investment can increase perceived security and lead to improved living environments without formal titles being issued. Experience indicates that, while both men and women desire security, women often prioritise improved water and sanitation, while men may 11 For example, in Ponte de Maduro informal settlement in Recife, Brazil (Chaves, 2013; OHCHR and UN- Women (2013, p 19). 34 prioritise road connections with employment centres in other parts of the city and titling (especially if they stand to benefit from ownership and inheritance). Effective programme design therefore requires that strategies are tailored to local circumstances and women’s as well as men’s needs are considered. Potential gender differences in priorities and outcomes make it all the more surprising that so few gendered evaluations seem to be available. 3.2 Privatisation of public sector land and housing In the former planned economies in Eastern Europe, most low income urban households continue to live in publicly constructed housing, although new construction has increased in recent years and active, often informal, land markets emerged during the 1990s in urban and peri-urban areas. Land registration therefore has to encompass the privatization of formerly publicly owned dwellings and the regularisation of both subsequent transfers and new informal subdivisions. Although many of the existing dwellings are in urban areas, most of the available discussions concentrate on titling of land, especially agricultural land, rather than the privatisation of publicly provided housing. In most of the countries of Eastern Europe and Central Asia, property and inheritance laws allow women to own and inherit land equally with men, and in theory registration procedures are also gender neutral (Stanley and Di Martino, n.d.; see also Lamb, 2009 and Joireman, 2013 for detailed discussions of the legal framework in Kosovo, although both are concerned with rural rather than urban areas). In addition, to ensure that women are treated equitably, they need knowledge of their rights, staff of land agencies need to be aware of gender inequality issues, and monitoring systems should assess whether women’s rights to land and housing are recognised and registered. In practice, Stanley and di Martino’s review of World Bank-financed land reform and privatisation in the Western Balkans indicates that many of the provisions in project designs are not being implemented. 12 For example in Albania, where a land reform programme was launched in 2007, data on women’s ownership of property is lacking, no public awareness campaign oriented towards educating women about their property rights had been mounted, the project does not monitor women’s participation in the titling process, and although data from a 2010 customer survey were said to be gender specific, no sex-disaggregated data are included (Stanley and di Martino, n.d.). As a result few women are aware of their rights or of the benefits of joint registration: “even well educated women in urban areas are not familiar with their rights to land and property and do not enforce those rights regularly” (p 6) . Furthermore, property is registered in the sole name of husbands because of the continued influence of customary practices, despite the supportive legal framework. Similarly, in Bosnia-Herzegovina neither targeted public awareness campaigns nor gender training for land agency staff had been fully implemented at the time of the review and no gender-specific indicators or data were available, notwithstanding provision in the current Real Estate Registration Project requiring that all data to be collected in the cadaster will be disaggregated by gender and a baseline household study for an impact evaluation that will monitor gender issues is to be carried out (Stanley et al, 2012). Overall, these early reviews concluded that:  women’s access to both information and legal recourse is substantially lower than men’s;  implementation may be affected by local practice, culture and tradition, as well as civil war and social dislocation, which have reinforced women’s lower social status (see also Joireman’s detailed discussion of Kosovo, 2013); 12 The Bank is financing projects in a number of countries but some, for example in Tajikistan and Romania, are only concerned with rural land (Stanley et al, 2012). 35  “women have generally benefited from the privatization of apartments [although] titles … are usually in the name of just one spouse and most often the male” (Stanley and di Martino, n.d., p 3);  gender disaggregated data is lacking in most countries, making it difficult to know the exact pattern of land and property ownership. In contrast, it appears that in the Kosovo Real Estate Cadastre Project (RECAP) activities are under way to mainstream gender issues in all activities (Stanley et al, 2012). The project aims to extend existing semi-systematic registration of apartment ownership to systematic registration, making it possible for awareness raising campaigns and other ways of safeguarding women’s rights to be more effective. At the time the report was written, systematic registration of the ownership of pre-1999 apartments in two cities (Ferizaj and Gjakova) was under way, as part of a plan for seven cities (five World Bank sponsored and two Norwegian government sponsored) (Lamb, 2009). Reform in Vietnam created a land market by initially granting households use rights which can be transferred and mortgaged, followed by large scale titling. By 2007, 11.71 million Land Tenure Certificates had been issued in rural areas, covering 77 percent of the target area, and 2.84 million in urban areas, covering 62 percent of the target area (World Bank, 2008, p 56). In the first ten years, titles were issued to married couples in the name of the man, but since 2003, joint titling has been mandated. Since the mid-1980s, rapid expansion of urban development into peri-urban areas has occurred, including the informal subdivision of farm land for residential use, and by the end of the 1990s it was estimated that between 70 percent and 90 percent of all urban development was illegal, although most subdivision and building had been agreed with local ward officials (Leaf, 1999, 2002). A 2007/8 study in four sites in three provinces finds that both women and men in urban areas were more knowledgeable about land rights and property than in rural areas, especially if couples had joint title. Women respondents expressed appreciation of joint titles because of the protection they provide in cases of marital conflict, transactions and inheritance; the greater equality in marriage and decision- making with which they are associated; and the increased scope they provide for engagement in business. As a result, all female respondents favoured joint title (compared to 85 percent of men) and urban women considered the benefits of converting single into joint titles would outweigh the costs. The study found that women with joint title were more aware of legal issues, more likely to proactively seek a Land Tenure Certificate, had more say in the use and disposition of land and were more likely to earn independent incomes (World Bank, 2008). In other developing countries publicly provided rental housing was privatised, often in the 1990s, although few assessments of impacts are available, with the exception of Zambia, where a large stock of municipal and mine housing was transferred into the hands of ‘sitting tenants’. Although ostensibly gender neutral, the policy had gendered effects. Most of the original tenants of the housing were employees, mostly men, and where tenant households still contained a man, the title was made out in his name alone. Although divorced or widowed women had been able to continue as tenants, when titles were issued, they were vulnerable to eviction by their husband’s relatives on the grounds that under customary law they were not entitled to inherit (Schlyter, 2004a). This led them to position themselves as mothers of their husband’s children and to carefully negotiate customary law and expectations in order to secure continued access to the family house for themselves and their children, often investing in house improvements if they could afford to do so in order to strengthen their claims (Butcher, 2009). Some men tried to prevent their widows (and children) from dispossession by putting the house in the name of one or more of the children, but the effect was to make widows dependent on their children (Schlyter, 2004b). Also tenants had to clear any payment arrears and, if they could not 36 afford to do so, were expected to find another buyer or face eviction. Women were more likely to experience this situation (Butcher, 2009). At the same time, while relatively few women had been tenants of municipal (and especially mine) housing because of their under-representation in wage employment, the privatisation of municipal housing created a market that provided opportunities for some women with means to purchase houses. 3.3 Public programmes for providing serviced plots and/or housing units Ideally, newly formed urban households should be able to obtain a plot and house in a planned and serviced neighbourhood. Scaling-up of planned development is also essential to prevent the further growth of informal settlements. Although the housing policies of some developing countries concentrated on public house building in the 1960s and sites and services schemes in the 1970s and 1980s, few succeeded in delivering large volumes of plots/houses for sale or rent. Programmes to provide serviced plots (and sometimes houses) have continued since the 1990s, but few have been on a large scale, despite the continued growth of urban populations. Most contemporary programmes are officially gender neutral, in the sense that all household heads in the eligible group, whether male or female, may be allocated a plot or house, and there is provision for joint tenure for married couples. Where these principles are implemented in practice, women can benefit. However, as in regularisation and privatisation programmes, if an adult male is present, he is assumed by officials in most cities to be the ‘household head’. Allocation and registration documents (and credit agreements) are almost always issued in the name of the man, unless a household clearly has a female head (typically a widow, divorcée or never married woman), even when joint tenure is possible or mandated. In addition, the realisation of rights is inhibited by the time and money associated with completing the cumbersome procedures required to convert an initial allocation into a title or adding a wife’s name to an already issued title (as in Ahmedabad, India, Baruah, 2007; Unni, 1999). Moreover, widows, divorcées and separated women frequently do not have equal inheritance rights, as discussed above, and fear that joint tenure may indicate a woman’s distrust of her husband (and a marriage in trouble) is a deterrent for many (Rakodi, 2006; see also Kalabamu, 2006, 2009 on Botswana). The standards adopted and cost recovery arrangements can also influence whether women can afford serviced plots, especially if they have unequal access to income generating opportunities. For example, the increased standards and cost recovery introduced in Botswana’s serviced plot programme in 1987 decreased the proportion of female-headed households who could afford to build (Datta, 1996). South Africa has one of the few contemporary large-scale programmes to provide housing to low income households, through provision of a serviced plot with individual title and a one-off capital grant sufficient to pay for a starter unit. These are mostly assembled into packages through which developers build houses for allocation to qualifying beneficiaries, although recently the government has also accepted the need for upgrading of informal settlements. Between 1994 and the end of 2008, serviced plots, subsidies and houses were provided to 2.7 million mostly urban households (Payne et al, 2008). The subsidy is targeted at households with a head who is married or co-habiting or, if single, has dependents, usually children, and an income below a threshold. The aim is that the share of female-headed households receiving subsidies will be the same as their overall share of the population and in some areas this appears to have been achieved. However, the disadvantaged position of women means that some find it more difficult than men to access the subsidy and meet additional costs, even though the very poor are exempt from user charges for utilities (Goebel et al, 2010). In addition, as only those over 21 are eligible, young women with children are excluded (Chenwi and McClean, 2009). 37 A small sample survey in a newly developed housing area in Ekurhuleni, adjacent to Johannesburg, found that nearly three quarters of households acknowledged a woman as head (although some contained a male partner). All owners expressed appreciation of the tenure security provided by official allocation documents, but women reported that they are more likely than men to complete the titling process because of the increased security it provides (Marx and Rubin, 2008; Payne et al, 2008). Sixty percent of both male- and female-headed households had improved their houses, but very few had loans and none had used their house as collateral – the cost was met by drawing on their own savings and contributions from relatives, although often women took longer to carry out improvements because they lacked the skills to do the work themselves. Two thirds indicated that they had made the improvements for the benefit of their children. While owners were aware that selling their houses was prohibited for five years (now eight) and did not envisage selling, in practice informal transfers are relatively common (Marx and Rubin, 2008). In addition, because of the cost of acquiring privately owned land and the peripheral location of much land in public ownership, many areas are remote from economic opportunities. Such areas are said to be more marginalised, disproportionately inhabited by female-headed households and prone to distress sales (Goebel et al, 2010; Venter et al, 2007). In practice, although the subsidy is received by a couple and policy is for the plot to be registered in both names, title is often issued solely in the name of the husband. (Kabane, 2011), Because this leaves women vulnerable to loss of the property or any future subsidy upon dissolution of the marriage, women often feel compelled to remain in relationships that they would otherwise leave, for example because of domestic violence or the fear of HIV infection (Chenwi and McClean, 2009). In addition, it is alleged that selection depends on political affiliation and access to local power brokers, which may disadvantage women whose networks are generally more limited. This may partly explain why women are disproportionately representated in alternative schemes organised by microfinance providers, which provide housing loans for participants in savings groups (the large majority of them women) (Mills, 2010; Ndinda and Uzodike, 2008). 4 Expanding women’s access to land and housing in urban areas: strategies and implications Responsibility for devising and implementing policies and legal reforms designed to improve women’s access to land and housing rests with national and local governments . Unless there is a demand for greater gender equality however, such issues may not be tackled. This section discusses promising approaches and those that have demonstrated to be effective in influencing policy and implementing gender equality provisions (see also de Silva de Alwis, 2013). Awareness of how demand for change can be built can inform external agencies’ engagement with government policy making processes, their choice and design of programmes, and their support for civil society organisations. 4.1 Strategies and tactics to influence policy and practice No one organisation or single approach can address the multiple causes and consequences of women’s insecure property rights. International pressure can help, as governments negotiate and ratify conventions, report to international bodies on their progress and are influenced by inter-national agencies and the experiences of other countries and cities. Domestic actors can in turn hold their governments to account for their failure to comply with international agreements which they have ratified. Although statutory laws relating to planning, land and housing and women’s property rights have become more progressive in many countries, there are inconsistencies between different sources of law (especially between statutory and customary or religious law), as well as gaps between law and 38 practice. Without pressure for change from significant constituencies within countries, governments are unlikely to introduce the necessary legal, policy and administrative changes. Thus without active urban citizens, it is unlikely that coherent urban development, land and housing policies will be devised and implemented or that they will meet the needs of poor people. Similarly, gender responsive legal change and reforms to land administration and judicial systems are unlikely to occur without organised pressure from women. Agarwal (2002) analysed the social forces that led to the 1956 changes to the Hindu Succession Act in India and assessed whether similar legal reforms are likely to occur today. She attributes success in the 1950s to the cohesion and strength of the Indian women’s movement, which helped to overcome opposition from elements of the state and men. Today, however, she concludes that, despite the growth of the women’s movement, women’s organisations: (a) do not, on the whole, prioritise gender equality in property rights, including inheritance law; (b) still lack representation in the highest level decision-making bodies; (c) have failed to win round those opposed to greater gender equality or made allies among civil society organisations (CSOs); and (d) have failed to generate sufficient pressure to achieve change, partly because Indian society has become more divided along communal (religious) lines and also because they have divided their efforts by focusing on changing separate bodies of personal law (such as the efforts of Muslim women ’s organisations to change the nikanahma or marriage contract, see Kirmani, 2011). Killian’s (2011) analysis of Tanzanian women’s efforts to influence the contents of the 1999 Village Land Law and Adamu and Para-Mallam’s (2012) study of the outcomes of efforts to change laws affecting women in Nigeria demonstrate the need for a coherent and representative women’s movement, effective strategies and tactics, and wider alliances. In Nigeria, repeated attempts to integrate CEDAW into Federal law failed, largely because of opposition to certain clauses from male parliamentarians and religious bodies, but also because the women’s organisations concerned failed to anticipate this opposition and develop strategies to difuse it. In contrast, the tactics adopted by women’s groups to build support and an alliance with the locally dominant Roman Catholic church resulted in a law prohibiting the mistreatment of widows (and widowers) being passed by the Anambra State government in 2005. Attitudinal changes and implementation are progressing, with the exception of provisions related to inheritance, which are still opposed by customary leaders and men. These experiences and ICRW’s scan of the strategies and programmes of national and international NGOs relating to women’s property rights (Knox et al, 2007b) reveal that the approaches most likely to be successful are those in which NGOs build on community level experience and work with grassroots organisations, using multiple approaches including:  awareness-raising activities amongst women and their communities, to increase knowledge of rights and change social norms;  capacity building, to equip women to exercise, secure and defend their rights;  assembling information and evidence to support these activities and strengthen attempts to influence formal laws and policies;  providing legal and policy assistance to women to enable them to secure their legal rights and/or training community members as paralegals; and  identifying women’s needs during periods of crisis (such as the HIV/AIDS epidemic), conflicts and disasters and seeking to ensure they are addressed. 39 ICRW’s research showed that many CSOs address women’s property rights as part of broader human rights, poverty reduction and women’s empowerment agendas, although some, such as the Huairou Commission, have a particular focus on land, housing and property rights. In contrast, relatively few of the organisations working on land and housing rights have a gender focus (COHRE is an except- ion). The scan showed that when NGOs working on single issues attempt to address the complex causes and consequences of women’s insecure property rights, they tend to be out of their depth or too isolated to be effective, suggesting that a network or partnership model is more effective. That multiple approaches, rooted in community-based participation and drawing strength from forming a network, are most likely to be effective is confirmed by the experience of urban poor organisations and their federations, initially in India and now in more than 15 countries. These are based on neighbourhood savings groups initiated and managed by women, which form federations at city and then national levels to manage savings and credit, secure land, upgrade houses, and improve infrastructure and services (e.g. community toilets and washing facilities). Most work is carried out in partnership with a local NGO and government agencies, especially municipal authorities, to demonstrate alternative approaches, mobilise communities, and influence policy and practice. A particularly powerful tool they have developed is a methodology for community level enumerations, which help to facilitate priority setting and provide a basis for engagement between communities and government (Patel et al, 2012). Women play a central role in savings groups and the Federations. Savings groups provide a supportive space for women to become active in local organisations and gain leadership experience, enabling many to become settlement leaders. Many enumerations are led by women, and often young women (and men) are enthusiastic data collectors. Depending on the context, specific women’s programmes are developed. In addition, Federation culture attempts to build mutually respectful and equitable social relations between their members. By 2010, the SDI federations claimed to have “secured a total of 108,000 plots, 54,000 houses, services for thousands of additional plots and over 600 communal washing and toilet blocks serving hundreds of thousands of people” (Patel and Mitlin, 2010, p 383). 4.2 Implications for legal reform, policy and practice The main implications of this review for legal reform and land and housing policy and practice are identified below. Experience shows that the legal and policy frameworks for land management and property rights need to explicitly recognise women’s rights and require gender-specific measures to be undertaken. However, reforms to property and family law underpin and interact with other policies and practices related to land administration, planning for increased housing supply and improvements to existing informal settlements. Unless laws, governance and administrative arrangements succeed in improving access to affordable, well-located serviced plots and houses for poor households, they are unlikely to benefit women. The second set of implications outlined below therefore identifies laws, policies and practices required to address the needs of poor urban residents, while emphasising the need for gender to be mainstreamed in them to ensure that they meet the needs of women. Countries’ and cities’ laws, policy frameworks and administrative systems are diverse, so it is vital that priorities and the content of laws, policies and programmes are adapted to particular national and local contexts. In addition, because the data is poor and the evidence patchy, improved data and more systematic evaluation of the outcomes of legal changes and other interventions are needed to enable evidence- based policy making in this area.13 13 This section draws, inter alia, on COHRE (2008), Deininger and Feder (2009), Byamugisha (2013), Giovarelli and Wamalwa (2011), Hallward-Driemeier and Hasan (2013), Knox et al (2007a), Lamb (2009), Lipman with 40 4.2.1 Gender-specific measures to improve women’s access to land and property i. The principle of non-discrimination enshrined in international conventions and national constitutions should be reflected in domestic law relating to property rights, land administration, marital relationships and inheritance. Such laws should: a. acknowledge the right to secure tenure and adequate housing; b. affirm the equal rights of men and women, regardless of their marital status, and whether they are in civil, customary or consensual unions; c. provide for marital regimes of separate, partial and full community of property, with full community of marital property (and its joint administration) as the default regime; d. mandate joint tenure for urban men and women in civil, customary and consensual unions; e. entitle spouses to a share of the deceased partners’ estate, through the law governing intestate succession and restrictions on testamentary freedom, in order to recognise wives’ contribution to the acquisition of marital property and guarantee them continued rights to the marital home; and f. entitle sons and daughters to equal shares in the case of intestate succession. ii. Such legal changes require the streamlining and harmonisation of often inconsistent laws. In particular, customary and religious laws should adhere to the principle of non-discrimination embodied in national constitutions and statutory laws. Given the complex relationships between these sources of law, social practices and gendered conceptions of men’s and women’s roles and responsibilities, the aim should be progressive achievement of consistency between different sources of law and ultimately gender equality. iii. Civil society organisations, especially women’s and grassroots organisations, have important roles to play in making women aware of their rights to land, housing and property, supporting their efforts to claim those rights and exploring whether and how customary or religious law, especially alternative interpretations and innovative practices, can be drawn upon to reduce gender discrimination and inequality. iv. Affordable or free legal aid is needed to enable poor women to claim and defend their rights. 4.2.2 Mainstreaming gender equality in pro-poor urban land and housing laws, policies and practices i. As land administration is modernised, gender should be mainstreamed in policies, programmes, procedures and practices, and the staff of bodies dealing with land allocation, inheritance and dispute settlement, including land professionals, government employees and the judiciary, trained in gender awareness. The GLTN’s Gender Evaluation Criteria provide a framework for assessing the gender responsiveness of land policies, administration and Rajack (2011), OHCHR and UN-Women (2013), Satterthwaite and Mitlin (2013), Payne et al (2008, 2009b), Rakodi (2010), Rolnik (2011), UN-HABITAT (2011a), UN-HABITAT and IIRR (2012) and World Bank (2005). Although Rodgers and Menon’s (2012) meta -analysis includes a few urban studies, the implications they identify concentrate solely on rural women. 41 dispute resolution, as well as a means of engaging and raising the awareness of relevant stakeholders.14 ii. Registration/titling of property is likely to enhance women’s wellbeing and property rights if it is gender sensitive, systematic and affordable, especially where households previously had little security, property-related conflict is common and poor people experience multiple sources of discrimination. Affordability is related not just to the costs and fees associated with registration/titling, but also with associated requirements and provisions, especially improvements to infrastructure and services. iii. To improve the security of poor women and men, a rights registration system that can be implemented quickly, systematically, at scale and at reasonable cost is required. In some situations, full titling may be appropriate. Where the capacity of land administration agencies is limited, multiple forms of tenure co-exist - including customary and hybrid systems, determining the ownership of property is complex and contentious, and titling is likely to exacerbate inequality, more appropriate systems for securing property rights should be developed. In particular, these should: a. permit the registration of a range of rights, especially use/occupancy rights, with provisions for upgrading if appropriate; b. ensure that procedures for adjudication, initial registration, subsequent registration of transfers, dispute resolution, and payments of fees and taxes are accessible and user friendly to the poor in general and women in particular; c. make use of new technologies, especially to computerise registers and make them publicly available on the internet, and develop land information systems; and d. recognise that tenure registration and land administration are not purely technical/administrative matters, but that women’s access to land is related to their socio-economic characteristics and the power relations in which they are embedded at family, community, city and regional/national levels, so that achieving effective changes involves political as well as technical and administrative processes. v. The potential use of financial incentives to encourage compliance with gender equitable behaviour (for example, joint tenure) should be explored. vi. Gender dimensions should be integrated into urban planning and development programmes through providing both women and men with opportunities to influence policy and programme design, and participate in monitoring and evaluation. vii. Policies should balance the need to upgrade and regularise existing informal settlements (including individual tenure registration/titling where appropriate and feasible) with guided development of new subdivisions to ensure an adequate supply of well-located and affordable serviced plots, which can in turn augment the supply of rental dwellings. viii. Both upgrading and guided development policies need to be gender-sensitive and implemented at scale, based on basic standards and incremental improvements, to reduce the likelihood of gentrification in upgraded informal settlements and leakage of serviced plots 14 UN-HABITAT (2012). See Obaikol et al (2013) for an account of its application in ten districts in Uganda, in four of which there are urban centres with populations of 50,000+. 42 intended for low income people up the income distribution. Income earning opportunities, especially for women, can be enhanced by ensuring plots and houses can be used for suitable business enterprises, transport links to employment locations accompany new development and public transport services are affordable, safe and women-friendly. The local provision of services and utilities, and the allocation of sites for social facilities and green space (for cultivation and/or recreation) are vital to enhance women’s wellbeing. ix. A proportion of the increased value of property resulting from public intervention in land development and infrastructure installation should be captured to finance further development and improvements, including, where necessary, subsidies to improve the property rights and living conditions of poor households, especially women. x. Residents should be protected from forced eviction and displacement and, where relocation is necessary, internationally-recognised good resettlement guidelines should be followed, including the equal entitlement of women to restitution and/or compensation. xi. A gender sensitive legal framework for regulating landlord/tenant relationships should be devised, with easy access to affordable enforcement. Because experience is lacking, pilots of alternative approaches will be needed. xii. Low income women (and men’s) ability to access land and housing can be enhanced by support to home owners’ associations/cooperatives, acceptance of collective tenure, support for user friendly and safe savings mechanisms and provision for low interest loans for the purchase of land (and perhaps the purchase, construction and improvement of housing). 4.2.3 Assessing poor women’s access to land and property: research and data needs Understanding of poor urban women’s access to and control over land and property is constrained by the lack of quality sex-disaggregated data. Inadequate data hinders the design of legal reforms, the modernisation of land administration systems, and efforts to ensure that urban development policies and programmes meet the needs of poor women and men. It also partly accounts for the scarcity of systematic assessments of the impacts of gender-sensitive legal reform and land and housing policies. Collection of sex-disaggregated data is essential for assessing women’s access to land and housing and identifying the drivers and constraints that explain national and local variations and the impacts of policies and legal reforms. Such data are essential for improving understanding, introducing legal and policy changes, and day-to-day planning, implementation and administration. Many of the relevant research questions are common to rural and urban areas, suggesting research designs that include both urban and rural areas and permit comparative analysis may be useful.15 However, research specific to urban areas is also needed. Two priorities emerge from this review:  First, developing a better understanding of the changing scale and characteristics of the channels of land and housing supply to which poor households have access, paying particular attention to the factors that influence the ability of women with different marital status (including women household heads) to become land and house owners and retain rights 15 For example, the GLTN and its partners are currently planning a multi-country action research programme to identify the forms of tenure that work for women with varying socio-economic characteristics in different contexts (Lowie Rosales, personal communication). 43 throughout their life course. Studies should examine the ‘ownership’ and/or value of assets (‘wealth’) related to different marital property and inheritance regimes.  Second, gendered assessments of registration programmes, both single intervention programmes (tenure registration) and those that combine registration with other interventions, especially infrastructure and service provision. These may be programmes to supply serviced plots (and sometimes ready built houses) to newly formed or relocating households, or informal settlement upgrading and regularisation. Several recent systematic urban tenure registration programmes are ripe for gendered evaluation, in Africa (Tanzania, Rwanda, Lesotho) and some former socialist countries (Albania, Vietnam). Evaluations of longer standing programmes are needed to assess the long term impact of titling on women and men in urban households (Peru and Mexico would be suitable locations). Such assessments would enable the outcomes of gender-specific measures such as joint titling to be examined. In addition, gendered evaluations of informal settlement upgrading are needed, to assess the outcomes of both well-established approaches (e.g. Brazil, Thailand) and programmes that have been scaled up recently (e.g. South Africa). These research priorities raise methodological challenges. Most of the researchers studying women’s land and property ownership, such as Agarwal and Panda (2007), Datta (2006), Deere et al (2013) and Deininger et al (2013), rely on household sample surveys, often panel surveys. However they stress the need for quantitative data and analysis to be complemented by qualitative data because correlation does not necessarily imply causation. ‘Empowerment’ or ‘bargaining power,’ for example, can only be assessed by the use of proxy variables. Meanwhile ownership arrangements and familial relationships are complex and require nuanced understanding, and perceptions and experiences (e.g. of security, social norms) are also important (see, for example, Doss, 2013). Deere and Doss and their colleagues reviewed the potential for using Living Standards Monitoring Surveys for the purpose of collecting gendered data on the ownership and value of assets in addition to socio-economic data, culminating in guidance on how an asset module conducive to gender analysis might be incorporated in future surveys (Doss et al, 2008). However, limited sample sizes may limit the potential for disaggregated analysis at the city and local levels. UN-HABITAT (2011a) suggests that a similar module could be added to Demographic and Health Surveys, but there is a similar issue of sample size and the DHS are restricted to women of child-bearing age. Doss (2013) concludes her review of methodological approaches for studying intra-household bargaining power by noting that experiments (both randomised controlled trials and natural experiments) enable the most rigorous analysis. However, RCTs are seldom feasible (although see Alayew Ali et al, 2013) and natural experiments rarely occur and do not necessarily deal with the issues of most interest. Instead, she recommends the adoption of a difference-in-difference approach (comparison before and after an intervention, as well as between those affected and not affected). Better gender-disaggregated data are needed not only for research but also for day-to-day administration and the monitoring and evaluation of reforms and programmes. Its collection poses conceptual, methodological, operational and institutional challenges. For example, given the complexity of family and tenure relationships, it is difficult to identify appropriate ways of gender disaggregating data, capturing the full complexity of existing tenure and rights in any city, and monitoring security of tenure. As noted in Section 3.1, a typology of tenure and property rights that specifies whether or not they are available to women (as individuals or household heads) as well as men and ranks them in terms of the 44 degree of security they convey can be used to identify the full range of tenure arrangements operating in a city and assessing the need for recognition or reform of local tenure arrangements (Payne and Durand-Lasserve, 2013). Further, UN-HABITAT has been attempting to develop ways of assessing secure tenure (rather than ownership or wealth). It suggests using the following indicators (UN- HABITAT, 2011a), although like all indicators, these may be over-simplified, reflecting what can be easily assessed and measured rather than what is important: Evictions % of households which have experienced eviction % of households that fear eviction % of women that fear eviction after divorce, separation or widowhood Documentation and Households with secure/insecure documents x type acquisition % of households that have acquired a dwelling and/or land using formal finance, purchased using their own resources, or self-arranged building Duration of residence/use % of households in current dwelling/workplace for 10+ years (can be a of current proxy for adverse possession) dwelling/workplace Rights to % of households which believe they have the right to  restrict  prevent others from entering their property  develop  develop their dwelling  sell  sell  inherit  inherit Data are needed at national, regional, urban and local levels, and for people, households and land units. National and regional/state/provincial data are needed to compare the access to land and property of women in varying contexts, especially women living in rural areas, towns, secondary and large cities. Regional (state/province) data is important where these levels of government have a degree of autonomy, which enables them to adopt distinctive approaches at different points in time. Comparative study of the factors influencing gendered access to land and the outcomes of policies and programmes across different regions will be important. Citywide data relating to households and individuals are needed for gendered assessments of living conditions, prosperity and poverty as well as for devising measures to improve service and support income generating activities. At the same time data on land units are essential for planning land allocations, infrastructure investments and raising revenue through property tax. Citywide data are also needed to assess the outcomes of policies and interventions. Some data can be generated for both administrative and research purposes from improved records, especially tenure registration systems, in which the rights initially allocated/registered and subsequent transfers are recorded. These can be used and maintained locally, be geo-referenced and computerised, and then linked with the records of other agencies to form a land information system with both spatial and administrative dimensions (Byamugisha, 2013). The system needs to record not just land parcels and whether they are registered in the names of male, female and joint household heads, but also the rights and interests of all household members, transfers, and the sources of funding for purchase of and investment in the land/house (especially if the property has been used as collateral). In the absence of citywide data, information collected at the local level can gradually build a wider picture, through the collection of primary data for studies in selected areas and neighbourhood level data for the design and evaluation of specific interventions. The Social Domain Tenure Model software package, as currently being developed by the GLTN and its partners, appears to have considerable potential. In addition, the experience of slum dwellers’ federations has demonstrated the 45 potential value of participatory enumerations (Patel et al, 2012). Based on this experience, the GLTN has produced a handbook that provides guidance on and examples of the use of community-led enumerations in a number of cities. The experience shows that such enumerations have empowered communities to resist eviction, secure appropriate treatment if they are being relocated, gain recognition for informal claims to land and provide a basis for the organisation of savings and credit groups. In addition, they have contributed to adjudication prior to regularisation, the preparation of city-wide upgrading programmes and dealing with post-conflict situations (UN-HABITAT, 2010). The handbook notes the leading role that women have played in many enumerations, which has ensured that relevant gendered data is collected. Although it cautions that relying on women can reinforce the (mis)conception that they have plenty of time, potentially lead to family and gender conflict, and fail to influence programme design and implementation, it also suggests that the skills and knowledge women gain in the process have empowered them at the community level, in negotiations with external agencies, and in monitoring the outcomes of interventions. In addition, if the campaign to include women’s access to land and property in the post -2015 international development goals is successful, indicators suitable for cross-national use will be required for monitoring purposes. Amongst the criteria for selection, the suitability of indicators for international comparison and the feasibility of collecting national data that can be disaggregated for urban and rural areas and by gender are important. Finally, the GLTN and its partners have developed a framework of gender evaluation criteria for use in the design, monitoring and evaluation of policies and practices. The criteria include:  equal participation by men and women in land governance and the design and implementation of tools;  capacity building, organisation and empowerment of women and men to use, access and benefit from tools;  legal and institutional issues relevant to women and men;  social, cultural and economic considerations affecting women’s and men’s access to land; and  coordination arrangements and the potential for scaling up and sustaining outcomes. These are translated into a set of questions, which can be tailored to particular contexts, with indicators suggested for each. The approach uses a score card to summarise stakeholders’ views on the gender responsiveness of local policies and practices and can involve a range of land actors at various scales and in different locations. It can be used to involve and raise the awareness of a range of stakeholders, add to knowledge about women’s access to land, train land professionals (UN- HABITAT (2011d) and provide a benchmark against which progress can be assessed. References Adamu, Fatima L. and Para-Mallam, Oluwafunmilayo J. (2012) The role of religion in women’s campaigns for legal reform in Nigeria, Development in Practice, 22, 5-6, p 803-18. Agarwal, Bina (2002) ‘Bargaining’ and Legal Change: Toward Gender Equality in India’s Inheritance Laws, Brighton: Institute of Development Studies WP 165. Agarwal, Bina and Panda, Pradeep (2007) Toward freedom from domestic violence: the neglected obvious, Journal of Human Development, 8, 3, p 359-88. Angel, S., Brown, E., Dimitrova, D., Ehrenberg, D., Heyes, J., Kusek, P., Marchesi, G., Orozco, V., Smith, L. and Vilchis, E. (2006) Secure Tenure in Latin America and the Caribbean: Regularization of Informal 46 Settlements in Peru, Mexico and Brazil, Princeton NJ: Princeton University, Woodrow Wilson School of Public and International Affairs http://wws.princeton.edu/research/final_reports/f05wws591g.pdf Ansari, Jamal H. (1998) Improving urban land management in India, in Jamal H. Ansari and Nathaniel von Einsiedel (eds) Urban Land Management – Improving Policies and Practices in Developing Countries of Asia, New Delhi: Oxford and IBH Publishing. Antonio, Danilo, Jack Makau and Samuel Mabala (2013) Addressing the information requirements of the urban poor – a government-community partnership in piloting the Social Tenure Domain Model in Uganda, Paper presented to the Annual World Bank Conference on Land and Poverty, 8-11 April. Archer, Diane (2012) Baan Mankong participatory slum upgrading in Bangkok, Thailand: Community perceptions of outcomes and security of tenure, Habitat International, 36, 178-84. Atuahene, Bernadette (2004) Legal title to land as an intervention against urban poverty in developing nations, The George Washington Intl Law Review, 36, 1109-79. Ayalew Ali, Daniel, Klaus Deininger and Markus Goldstein (2011) Environmental and Gender Impacts of Land Tenure Regularization in Africa: Pilot Evidence from Rwanda , Helsinki: UNU-WIDER WP No. 2011/74. Ayalew Ali, Daniel, Matthew Collin, Klaus Deininger, Stefan Dercon, Justin Sandefur and Andrew Zeitlin (2013) The Price of Empowerment: Land Titling and Female Inclusion in Urban Tanzania, Washington DC: Paper presented to the Annual World Bank Conference on Land and Poverty, April 8-11 http://www.conftool.com/landandpoverty2013/sessions.php Banerjee, Banashree (2002) Security of tenure in Indian cities, in Durand-Lasserve, Alain and Lauren Royston (eds) Holding their Ground: Secure Land Tenure for the Urban Poor in Developing Countries, London: Earthscan, p 37-58. Banerjee, Banashree (2004) Maximising the impact of tenure and infrastructure programmes on housing conditions: the case of slums in Indian cities, Paper presented to the International Conference on Adequate and Affordable Housing, Toronto, 24-27 June. Baruah, Bipasha (2007) Gendered realities: exploring property ownership and tenancy agreements in urban India, World Development, 35, 12, 2096-109. Baruah, Bipasha (2010) Women and Property in Urban India. Vancouver: University of British Columbia Press. Berner, E. (2000) Defending a Place in the City: Localities and the Struggle for Urban Land in Metro-Manila, Manila: Ateneo de Manila University Press. Besley, Tim and Ghatak, Maitreesh (2010) Property rights and economic development, in Rodrik, Dani and Rosenzweig, Mark (eds) Handbook of Development Economics, Vol. 5, North-Holland: Elsevier, p 4526-95. Boonyabancha, Somsook (2005) Baan Mankong: going to scale with ''slum'' and squatter upgrading in Thailand, Environment and Urbanization, 17, 1, 21-46 Boonyabancha, Somsook (2008) Land for housing the poor – by the poor: experiences from the Baan Mankong nation-wide slum upgrading program in Thailand. Unpublished paper. Bose, M. (1998) Surveillance, circumscriber of women’s spatial experience: the case of slum dwellers in Calcutta, in Dandekar, H. (ed) City Space and Globalization: An International Perspective, Ann Arbor: University of Michigan, College of Architecture and Urban Planning. 47 Brown, Alison M. and Tony Lloyd-Jones (2002) Spatial planning, access and infrastructure, in Rakodi, Carole with Tony Lloyd-Jones (eds) Urban Livelihoods: A People-centred Approach to Reducing Poverty, London: Earthscan, p 188-204. Butcher, Sian (2009) Lived Experiences of Privatization in Lusaka and Cape Town: women and State Rental Housing, Cape Town: University of Cape Town, Faculty of the Humanities, MA dissertation. Byamugisha, Frank F.K. (2013) Securing Africa's Land for Shared Prosperity, Washington DC: World Bank, Africa Development Forum. Calderón, Julio (2004) The formalisation of property in Peru 2001-2002: the case of Lima, Habitat International, 28, 2, p 289-300. Chant, Sylvia (1987) Domestic labour, decision-making and dwelling construction: the experience of women in Querétaro, Mexico, in Caroline Moser and Linda Peake (eds) Women, Human Settlements and Housing, London: Tavistock, p 33-54. Chant, Sylvia (1997) Women-headed households: poorest of the poor? Perspectives from Mexico, Costa Rica and the Philippines, IDS Bulletin, 28, 3, p 26-48. Chant, Sylvia (2003) Female Household Headship and the Feminisation of Poverty: facts, fictions and forward strategies, London: LSE Gender Institute, New Working Paper Series, Issue 9 http://eprints.lse.ac.uk/574/1/femaleHouseholdHeadship.pdf Chant, Sylvia (2009) The ‘feminisation of poverty’ in Costa Rica: to what extent a conundrum?, Bulletin of Latin American Research, 28,1, p 19-44. Chant, Sylvia (2013) Cities through a ‘gender lens’: a golden ‘urban age’ for women in the global South? Environment and Urbanization, 25, 9, p 9-29. Chant, Sylvia and McIlwaine, Cathy (1995) Women of a Lesser Cost: Female Labour, Foreign Exchange and Philippine Development, London: Pluto. Chaves, Patricia, Regina Pritchett and Ivan Lucá (2013) Building land governance and transparency in land policies: practical lessons from Ponte de Maduro, Paper presented to the Annual World Bank Conference on Land and Poverty, 8-11 April. Chenwi, Lilian and McClean, Kirsty (2009) 'A Woman’s home is her castle?’ – Poor women and housing inadequacy in South Africa, South African Journal on Human Rights, 25, p 517-45. COHRE (2004) Bringing Equality Home: Promoting and Protecting the Inheritance Rights of Women – A Survey of Law and Practice in Sub-Saharan Africa, Geneva: Centre on Housing Rights and Evictions. COHRE (2006) In Search of Equality: A Survey of Law and Practice Related to Women’s Inheritance Rights in the Middle East and North Africa (MENA) Region, Geneva: Centre on Housing Rights and Evictions. COHRE (2008) Women, Slums and Urbanisation: Examining the Causes and the Consequences, Geneva: Centre on Housing Rights and Evictions, Women and Housing Rights Programme. COHRE (n.d.) Violence: The Impact of Forced Evictions on Women in Palestine, India and Nigeria, Geneva: Centre on Housing Rights and Evictions. Daley, Elizabeth, Rachel Dore-Weeks and Claudine Umuhoza (2010) Ahead of the game: land tenure reform in Rwanda and the process of securing women's land rights, Journal of Eastern African Studies, 4, 1, p 131-52. 48 Datta, Kavita (1995) Strategies for urban survival? Women landlords in Gaborone, Botswana, Habitat International, 19, 1, 1-12. Datta, Kavita (1996) The organization and performance of a low income rental market: the case of Gaborone, Botswana, Cities, 13(4), 237-45. Datta, Namita (2006) Joint titling – a win-win policy? Gender and property rights in urban informal settlements in India, Feminist Economics, 12, 1-2, 271-98. De Silva de Alwis, Rangita (2013) Women’s Voice and Agency: Four Central Freedo ms, Washington DC: Background paper for the World Bank Gender and Development Team. De Souza, Flávio A.M. (1999) Land tenure security and housing improvements in Recife, Brazil, Habitat International, 23, 1, 19-33. De Souza, Flávio A.M. (2001a) The future of informal settlements: lessons in the legalization of disputed urban land in Recife, Brazil, Geoforum, 32, p 483-92. De Souza, Flávio A.M. (2001b) Perceived security of land tenure in Recife, Brazil, Habitat International, 25, p 175-90. De Souza, Flávio A.M. (2004) Security of land tenure revised: the case of CRRU in Recife and Porto Alegre, Brazil, Habitat International, 28, 2, 231-44. Deere, Carmen Diana (2010) Assessing gender inequality in asset ownership, in Chant, S. (ed) The International Handbook of Gender and Poverty, Cheltenham: Edward Elgar, p 347-52. Deere, Carmen Diana and Doss, Cheryl R. (2006) The gender asset gap: what do we know and why does it matter? Feminist Economics, 12, 1-2, p 1-50. Deere, Carmen Diana and León, Magdalena (2001) Who owns the land? Gender and land-titling programmes in Latin America, Journal of Agrarian Change, 1, 3, p 440-67. Deere, Carmen Diana and León, Magdalena (2003) The gender asset gap: land in Latin America, World Development, 31, 6, 925-47. Deere, Carmen Diana, Abena D. Oduro, Hema Swaminathan and Cheryl Doss (2013) Property rights and the gender distribution of wealth in Ecuador, Ghana and India, Journal of Economic Inequality, 11, 2, p 249-65. Deininger, Klaus and Feder, Gershon (2009) Land registration, governance, and development: evidence and implications for policy, World Bank Research Observer, 24, 2, p 233-66. Deininger, Klaus, Songqing Jin, Hari K, Nagarajan and Fang Xia (2013) Does inheritance law reform improve women’s access to capital? Evidence fro m urban India. Paper presented to the Annual World Bank conference on Land and Poverty, Washington DC, April 8-11. Doss, Cheryl (2013) Intrahousehold Bargaining and Resource Allocation in Developing Countries, Washington DC: World Bank, Policy Research WP 6337. Durand-Lasserve, Alain and Lauren Royston (eds) (2002) Holding Their Ground: Secure Land Tenure for the Urban Poor in Developing Countries, London: Earthscan. Durand-Lasserve, Alain and Selod, Harris (2010) The formalization of urban informal land tenure, in Lall, S.V., Freire, M., Yuen, B., Rajack, R. and Helluin, J-J. (eds) Urban Land Markets: Improving Land Management for Successful Urbanization, Dordrecht: Springer Verlag, p 101-32. 49 Eskezia, Berhanu (2005) The Incidence of Urban Poverty in the Female-Headed Households in Addis Ababa, Ethiopia, Addis Ababa: http://www.gdn.int/admin/uploads/editor/files/2012Conf_Papers/2_3_Berhanu%20Eskezia_Paper_T2.pdf Fernandes, Edesio (2007a) Constructing the ‘right to the city’ in Brazil, Social and Legal Studies, 16, 2, p 201- 19. Fernandes, Edesio (2007b) Principles, bases and challenges of the National Programme to Support Sustainable Urban Land Regularisation in Brazil, in Freire, Mila, Ricardo Lima, Dean Cima, Bruce Ferguson, Christine Kessides, José Aroudo Mota and Diana Motta (eds) Land and Urban Policies for Poverty Reduction, Vol. 1, Washington DC and Brasilia: The World Bank and ipea (Institute for Applied Economic Research), Proceedings of the Third International Urban Research symposium held in Brasilia, April, 2005, p 191-214. Fernandes, Edesio (2011) Regularisation of Informal Settlements in Latin America , Washington DC: Lincoln Institute for Land Policy. Fernandes, Edesio and Rolnik, Raquel (1998) Law and urban change in Brazil, in Fernandes, Edesio and Varley, Ann (eds) Illegal Cities: Law and Urban Change in Developing Countries, London: Zed, p 140-56. Fernandes, Edesio and Varley, Ann (eds) (1998) Illegal Cities: Law and Urban Change in Developing Countries, London: Zed. Field, Erica (2004) Property rights, community public goods, and household time allocation in urban squatter communities: evidence from Peru, Wm and Mary Law Review, 45, 3, p 837-87. Field, Erica (2007) Entitled to work: urban property rights and labor supply in Peru, The Quarterly Journal of Economics, 122, 4, p 1561-1602. Field, Erica and Torero, Maximo (2004) Do Property Titles Increase Credit Access Among the Urban Poor? Evidence from a Nationwide Titling Program, Harvard University, Department of Economics, mimeo http://www.rwj.harvard.edu/papers/field/Field%20Do%20Property%20Titles%20Increase%20Credit....pdf accessed15th June 2013 Field, Erica and Torero, Maximo (2006) Do Property Titles Increase Credit Access Among the Urban Poor? Evidence from a Nationwide Titling Program, Harvard University, Department of Economics, mimeo. Galiani, Sebastian and Schargrodsky, Ernesto (2004) Effects of land titling on child health, Economics and Human Biology, 2, p 353-72. Galiani, Sebastian and Schargrodsky, Ernesto (2010) Property rights for the poor: effects of land titling, Journal of Public Economics, 94, 9-10, p 700-29. Galiani, Sebastian and Schargrodsky, Ernesto (2011) The Dynamics of Land Titling Regularization and Market Development, Helsinki: UNU-WIDER WP No. 2011/88. Giovarelli, Renee and Beatrice Wamalwa (2011) Land Tenure, Property Rights, and Gender: Challenges and Approaches for Strengthening Women’s Land Tenure and Property Rights, Washington DC: USAID Property Rights and Resource Governance Briefing Paper #7. Goebel, Allison, Belinda Dodson and Trevor Hill (2010) Urban advantage or urban penalty? A case study of female-headed households in a South African city, Health and Place, 16, p 573-80. Gulyani, S. and E.M. Bassett (2007) Retrieving the baby from the bathwater: slum upgrading in Sub-Saharan Africa, Environment and Planning C, 28, p 486-515. 50 Haddad, Emilio and João Fernando Pires Meyer (2009) Housing conditions and income distribution: evidence from São Paulo, in Lall, Somik V., Mila Freire, Belinda Yuen, Robin Rajack and Jean-Jacques Helluin, Urban Land Markets: Improving Land Management for Successful Urbanization, Dordrecht: Springer, p 283-302. Hallward-Driemeier, Mary and Tazeen Hasan (2013) Empowering Women: Legal Rights and Economic Opportunities in Africa, Washington DC: World Bank, Africa Development Forum. Home, Robert (ed) (2011a) Essays in African Land Law, Pretoria: Pretoria University Law Press. Home, Robert (ed) (2011b) Local Case Studies in African Land Law, Pretoria: Pretoria University Law Press. Hoornweg, D., Marielle Dubbeling, Judy Baker and Anjali Acharya (2012) Urban Agriculture: Case Studies of Accra, Bangalore, Lima and Nairobi and a Global Review, Washington DC: World Bank. Hovorka, Alice J. and Diana Lee-Smith (2006) Gendering the urban agriculture agenda, in van Veenhuisen, René (ed) Cities Farming for the Future: Urban Agriculture for Green and Productive Cities , Ottawa and Silang: RUAF Foundation, IDRC and International Institute of Rural Reconstruction, p 125-44. Ikejiofor, Uche Cosmas (2006) Integrative strategies or functional interface? Emerging trends in land administration in contemporary Enugu, Nigeria, International Development Planning Review, 28, 2, p 37-58. Joireman, Sandra F. (2013) Securing property rights for women (and men) in Kosovo, Washington DC: Paper presented to the World Bank Conference on Land and Poverty, April 8-11. Kabane, Noxolo (2011) Landfirst dignity through land, The Transformer, Oct/Nov/Dec http://www.afesis.org.za/sustainable-Settltments-Articles/landfirst-dignity-through-land accessed 1/7/13. Kalabamu, Faustin T. (2006) Changing gender contract in self-help housing construction in Botswana: the case of Lobatse, in Stifel, D.C., Vanessa Watson and H. Acselrad (eds) Dialogues in Urban and Regional Planning: volume 2, London: Routledge. Kalabamu, Faustin T. (2009) Towards egalitarian inheritance rights in Botswana: the case of Tlokweng, Development Southern Africa, 26(2), 209-23. Kantor, Paula and Nair, Padmaja (2005) Vulnerability to Crisis in Lucknow, India: The Role of Assets in Mitigating Risk, New Delhi: Oxfam. Kellett, Peter and Garnham, Ann (1995) The role of culture and gender in mediating the impact of official interventions in informal settlements: a study from Colombia, Habitat International, 19, 1, 53-60. Killian, B. (2011) The Women’s Land Rights Movement, Customary Law and Reli gion in Tanzania, Birmingham: Religions and Development WP 57. Kirmani, Nida (2011) Re-thinking the promotion of women's rights through Islam in India, IDS Bulletin, 42, 1, p 46-56. Kironde, J.M. Lusagga (2006) Issuing of Residential Licences to Landowners in Unplanned Settlements in Dar es Salaam, Tanzania, Dar es Salaam: Consultancy report for UN-HABITAT, Shelter Branch, Land and Tenure Section. Kironde, J.M. Lusagga (2009) Improving land sector governance in Africa: the case of Tanzania, Paper presented to the Workshop on Land Governance in Support of the MDGs: Responding to New Challenges, Washington DC, 9-10 March. Knox, Anna, Nata Duvvury and Noni Milici (2007a) Connecting Rights to Reality: A Progressive Framework of Core Legal Protections for Women’s Property Rights, New York: International Center for Research on Women. 51 Knox, Anna, Aslihan Kes and Noni Milici (2007b) Mending the Gap Between Law and Practice: Organizational Approaches for Women’s Property Rights, New York: International Center for Research on Women. Kumar, Sunil (2010) Gender, livelihoods and rental housing markets in the Global South, in Chant, S. (ed) The International Handbook of Gender and Poverty, Cheltenham: Edward Elgar, p 367-72. Lall, S.V., Freire, M., Yuen, B., Rajack, R. and Helluin, J-J. (eds) 2009. Urban Land Markets: Improving Land Management for Successful Urbanization, Dordrecht: Springer Verlag. Lamb, Tony (2009) Report on Gender and Land Titling in Kosovo in connection with a proposed World Bank- financing for a Real Estate Cadastre and Registration Project, Washington DC: World Bank, unpublished paper. Lambert, Sylvie, Martin Ravallion and Dominique van de Walle (2011) Is it What you Inherited or What you Learnt? Intergenerational Linkage and Interpersonal Inequality in Senegal, Helsinki: UNU-WIDER WP No. 2011/62. Lanjouw, J.O. and Levy, P.I. (2002) Untitled: a study of formal and informal property rights in urban Ecuador, The Economic Journal, 112, p 986-1019. Larsson, Anita (1988) The housing situation of urban female-headed households in Botswana, Habitat International, 12, 3, 17-27. Larsson, Anita (1989) Women Householders and Housing Strategies: the Case of Gaborone, Botswana, Gävle: National Swedish Institute for Building Research, Research Report SB:25. Leaf, Michael (1999) Vietnam’s urban edge: the administration of urban development in Hanoi, Third World Planning Review, 21, 3, p 297-315. Leaf, Michael (2002) A tale of two villages: globalization and peri-urban change in China and Vietnam, Cities, 19, 1, p 23-31. Leduka, R. Clement (2004) Informal Land Delivery Processes and Access to Land for the Poor in Maseru, Lesotho, Birmingham: University of Birmingham, Informal Land Delivery Processes in African Cities, WP 5. Leduka, R. Clement (2006) Chiefs, civil servants and city council: state-society relationships in evolving land delivery processes in Maseru, Lesotho, International Development Planning Review, 28, 2, p 181-208. Lipman, Barbara with Robin Rajack (2011) Memo to the Mayor: Improving Access to Urban Land for All Residents: Fulfilling the Promise, Washington DC: World Bank Urban Development Series Knowledge Paper No. 11. Loughhead, Susan and Rakodi, Carole (2002) Reducing urban poverty in India – lessons from projects supported by DFID, in Rakodi, Carole with Tony Lloyd-Jones (eds) Urban Livelihoods: A People-centred Approach to Reducing Poverty, London: Earthscan, p 225-36. Marx, Colin and Rubin, Margot (2008) The Social and Economic Impact of Land Titling in Selected Settlements in Ekurhuleni Metropolitan Area, Johannesburg: University of the Witwatersrand, Centre for Urban and Built Environment Studies. Mills, Sophie (2010) Renegotiating the household: successfully leveraging women’s access to housing microfinance in South Africa, in Chant, Sylvia (ed) The International Handbook of Gender and Poverty , Cheltenham: Edward Elgar, p 373-8. Miraftab, Faranak (1998) Complexities of the margin: housing decisions by female householders in Mexico, Environment and Planning D: Society and Space, 16, 289-310. 52 Morais, Maria da Piedade and Bruno de Oliveira Cruz (2009) Housing demand, tenure choice, and housing policy in Brazil, in Lall, Somik V., Mila Freire, Belinda Yuen, Robin Rajack and Jean-Jacques Helluin, Urban Land Markets: Improving Land Management for Successful Urbanization, Dordrecht: Springer, p 253-82. Moser, Caroline O.N. (1998) The asset vulnerability framework: reassessing urban poverty reduction strategies, World Development, 26, 1, p 1-19. Moser, Caroline (2010) Moving beyond gender and poverty to asset accumulation: Guayaquil, Ecuador, in Chant, Sylvia (ed) The International Handbook of Gender and Poverty , Cheltenham: Edward Elgar, p 391-8. Moser, Caroline and Felton, Andrew (2010) The Gendered Nature of Asset Accumulation in Urban Contexts: Longitudinal results from Guayaquil, Ecuador, Helsinki: UNU-WIDER WP No. 2010/17. Munoz Boudet, Ana Maria, Patti Petesch and Carolyn Turk with Angelica Thumala (2012) On Norms and Agency: Conversations about Gender Equality with Women and Men in 20 Countries, Washington DC: World Bank. Musyoka, Rose (2004) Informal Land Delivery Processes and Access to Land for the Poor in Eldoret, Kenya, Birmingham: University of Birmingham, Informal Land Delivery Processes in African Cities, WP 4. Musyoka, Rose (2006) Non-compliance and formalisation: mutual accommodation in land subdivision processes in Eldoret, Kenya, International Development Planning Review, 28, 2, p 235-61. Namubiru-Mwaura, Evelyn (2013) Land Tenure and Gender: Approaches and Challenges for Strengthening Rural Women’s Land Rights, Washington DC: Background paper for the World Bank Gender and Development team. Ndinda, Catherine and Uzodike, Ufo Okeke (2008) Accessing housing finance in South Africa: the role of women activism, Wagadu, 6, p 75-91. Ngoga, Thierry Hoza (2012) Empowering women through land tenure reform: the Rwandan experience. Paper presented to the UN Women and UN Office of the High Commissioner for human rights Expert Group meeting on Good Practices in Realizing Women’s Rights to Productive Resources, with a focus on land, Geneva, 25 -27 June. Ngwira, Naomi (n.d.) Women's Property and Inheritance Rights and the Land Reform Process in Malawi, Blantyre: Institute for Policy Research and Analysis for Dialogue for USAID. Obaikol, Esther, A. Jonsson and Lowie Rosales (2013) The gender evaluation criteria – evaluating land tools with a gender lens for transparent land governance: the Uganda experience, Paper to the Annual World Bank Conference on Land and Poverty, April 8-11. OHCHR and UN-Women (2013) Realizing Women’s Rights to Land and Other Productive Resources , New York and Geneva: Office of the High Commissioner for Human Rights and UN-Women. Patel, Kamna (2013) The value of secure tenure: ethnographic accounts of how tenure security is understood and realised by residents of low-income settlements in Durban, South Africa, Urban Forum, 24, p 269-87. Patel, Sheela and Mitlin, Diana (2010) Gender issues and shack/slum dweller federations, in Chant, Sylvia (ed) The International Handbook of Gender and Poverty: Concepts, Research, Policy, Cheltenham: Edward Elgar, p 379-84. Patel, Sheela, Carrie Baptist and Celine D’Cruz (2012) Knowledge is power – informal communities assert their right to the city through SDI and community-led enumerations, Environment and Urbanization, 24, 1, p 13-26. 53 Payne, Geoffrey (ed) (2002) Land, Rights and Innovation: Improving Tenure Security for the Urban Poor, London: ITDG Publishing. Payne, Geoffrey and Durand-Lasserve, Alain (2013) Holding On: Security of Tenure – Types, Policies, Practices and Challenges, Geneva: Office of the High Commissioner for Human Rights, Research paper commissioned by the Special Rapporteur on the right to an adequate standard of living for an Expert Group Meeting on Security of Tenure, 22-23 October 2012 http://direitoamoradia.org/wp- content/uploads/2013/03/Payne-Durand-Lasserve-BackgroundPaper-JAN2013-FINAL-1.pdf Payne, Geoffrey, Durand-Lasserve, Alain and Rakodi, Carole (2008) Social and Economic Impacts of Land Titling Programmes in Urban and Peri-urban Areas: International Experience and Case Studies of Senegal and South Africa, London: Geoffrey Payne Associates. Payne, Geoffrey, Durand-Lasserve, Alain and Rakodi, Carole (2009a) The limits of land titling and home ownership, Environment and Urbanization, 21, 2, p 443-62. Payne, Geoffrey, Durand-Lasserve, Alain and Rakodi, Carole (2009b) Gender Sensitive and Pro-poor Principles when Regularising Informal Land Settlements in Urban and Peri-urban Areas, Oslo: UNDP Oslo Governance Centre, DP 10 http://www.undp.org/oslocentre/overview/land_governance.html Payne, Geoffrey, Durand-Lasserve, Alain and Rakodi, Carole (2010) Social and economic impacts of land titling programs in urban and periurban areas: a short review of the literature, in Lall, S.V., Freire, M., Yuen, B., Rajack, R. and Helluin, J-J. (eds) Urban Land Markets: Improving Land Management for Successful Urbanization, Dordrecht: Springer Verlag, p 155-62. Peterman, Amber (2012) Widowhood and asset inheritance in Sub-Saharan Africa: empirical evidence from 15 countries, Development Policy Review, 30, 5, p 543-71. Porio, Emma and Christine Crisol (2004) Property rights, security of tenure and the urban poor in Metro Manila, Habitat International, 28, 203-19. Rakodi, Carole (1999) A capital assets framework for analysing household livelihood strategies, Development Policy Review, 17, 3, p 315-42. Rakodi, Carole (2002) Conclusions, in Rakodi, Carole with Tony Lloyd-Jones (eds) Urban Livelihoods: A People-centred Approach to Reducing Poverty, London: Earthscan, p 288-98. Rakodi, Carole (2006) Social agency and state authority in land delivery processes in African cities: compliance, conflict and cooperation, International Development Planning Review, 28, 2, 263-85. Rakodi, Carole (2010) Gender, poverty and access to land in cities of the South, in Chant, Sylvia (ed) The International Handbook of Gender and Poverty, Cheltenham: Edward Elgar, p 353-9. Redwood, Michael (ed) (2009) Agriculture in Urban Planning: Generating Livelihoods and Food Security , Ottawa and London: IDRC and Earthscan http://idl-bnc.idrc.ca/dspace/bitstream/10625/36354/1/127655.pdf Rodgers, Yana and Nidhiya Menon (2012) A Meta-Analysis of Land Rights and Women’s Economic Well- being, Nov. 27. Rolnik, Raquel (2011) Report of the Special Rapporteur on adequate housing as a component of the right to an adequate standard of living, and on the right to non-discrimination in this context, New York: UN General Assembly, Human Rights Council, Nineteenth session A/HRC/19/53. Sait, Siraj (2011) Not just another ‘custom’: Islamic influences on African land laws, in Home, Robert (ed) Essays in African Land Law, Pretoria: Pretoria University Law Press, p 91-112. 54 Sait, Siraj and Lim, Hilary (2006) Land, Law and Islam: Property and Human Rights in the Muslim World , Nairobi and London: UN-HABITAT and Zed Books Satterthwaite, David and Mitlin, Diana (2013) Reducing Urban Poverty in the Global South , London and New York: Routledge. Satterthwaite, David and Tacoli, Cecilia (2002) Seeking an understanding of poverty that recognizes rural-urban differences and rural-urban linkages, in Rakodi, Carole with Tony Lloyd-Jones (eds) Urban Livelihoods: A People-centred Approach to Reducing Poverty, London: Earthscan, p 52-70. Schlyter, Ann (2004a) Gender issues in housing privatization: intricacies of property transfer in Zambia, Trialog, 1, p 18-23. Schlyter, Ann (2004b) Privatization of Council Housing in Lusaka, Zambia, London: RICS Foundation, Our Common Estate Paper Series. Sharma, Ursula (2005) Dowry in North India: its consequences for women, in Basu, Srimati (ed) Dowry and inheritance, New Delhi: Women Unlimited, p 15-26. Stanley, Victoria and Di Martino, Samantha (n.d.) Assessing Land Administration Project’s Gender Impacts in the Western Balkans: Country Case Studies: Albania, Bosnia-Herzegovina and Montenegro, Washington DC: World Bank (mimeo) Stanley, Victoria, Lamb, Tony and De Martino, Samantha (2012) Gender and Land Administration: Issues and Responses, Washington DC: World Bank, Europe and Central Asia Knowledge Brief 53 www.worldbank.org/eca Strickland, Richard S. (2004) To Have and to Hold: Women’s Property and Inheritance Rights in the Context of HIV/AIDS in Sub-Saharan Africa, New York: International Center for Research on Women http://www.sarpn.org.za/documents/d0000864/P962-ICRW_ToHaveToHold_June2004.pdf Tacoli, Cecilia (2012) Urbanization, Gender and Urban Poverty: Paid Work and Unpaid Care Work in the City , London: International Institute for Environment and Development and UN Fund for Population Activities, Urbanization and Emerging Population Issues WP 7. UN-HABITAT (2003) Rental Housing: An Essential Option for the Urban Poor in Developing Countries, Nairobi: UN-HABITAT. UN-HABITAT (2005) Shared Tenure Options for Women: A Global Overview, Nairobi: UN-HABITAT. UN-HABITAT (2007a) Policy Makers’ Guide to Women’s Land, Property and Housing Rights across the World, Nairobi: UN-HABITAT. UN-HABITAT (2007b) Global Report on Human Settlements 2007: Enhancing Urban Safety and Security, Nairobi: UN-HABITAT. UN-HABITAT (2008a) Land Rights for All, Nairobi: UN-HABITAT. UN-HABITAT (2008b) State of the World’s Cities 2008/9: Harmonious Cities, London: Earthscan. UN-HABITAT (2010) Count Me In: Surveying for Tenure Security and Urban Land Management , Nairobi: UN-HABITAT. UN-HABITAT (2011a) Monitoring Security of Tenure in Cities: People, Land and Policies, Nairobi: UN- HABITAT. 55 UN-HABITAT (2011b) A Policy Guide to Rental Housing in Developing Countries, Nairobi: UN-HABITAT. UN-HABITAT (2011c) Innovative Urban Tenure in the Philippines: Challenges, Approaches and institutionalisation, Nairobi: UN-HABITAT. UN-HABITAT (2011d) Designing and Evaluating Land Tools with a Gender Perspective: A Training Package for Land Professionals, Nairobi: UN-HABITAT. UN-HABITAT (2012) State of Women in Cities 2012-2013: Gender and the Prosperity of Cities, Nairobi: UN- HABITAT. UN-HABITAT GLTN and IIRR (2012) Handling Land: Innovative Tools for Land Governance and Secure Tenure, Nairobi: UN-HABITAT and International Institute for Rural Reconstruction. Unni, Jeemol (1999) Property rights for women: case for joint titles to agricultural land and urban housing, Economic and Political Weekly, 34, 21, p 1281-6. Van Gelder, Jean-Louis (2009) Legal Tenure Security, Perceived Tenure Security and Housing Improvement in Buenos Aires: An Attempt towards Integration, International Journal of Urban and Regional Research, 33, 1, p 126-46. Van Gelder, Jean-Louis (2013) Then I'll Huff, and I'll Puff, and I'll . . . : A natural experiment on property titling, housing improvement and the psychology of tenure security, International Journal of Urban and Regional Research, 37, 2, p 734-49. Varley, Ann (2007) Gender and property formalisation: conventional and alternative approaches, World Development, 35, 10, p 1739-58. Varley, Ann (2010) Modest expectations: gender and property rights in urban Mexico, Law and Society Review, 44, 1, p 67-99. Varley, Ann (2013) Feminist perspectives on urban poverty: de-essentialising difference, in Peake, Linda and Rieker, Martina (eds) Rethinking Feminist Interventions into the Urban , London: Taylor and Francis. Venter, C., Vokolkovab, V. and Michalekc, J. (2007) Gender, Residential Location, and Household Travel: Empirical Findings from Low-income Urban Settlements in Durban, South Africa, Transport Reviews, 27, 6, p 653-77. Vera-Sanso, Penny (2010) Gender, urban poverty and ageing in India: conceptual and policy issues, in Chant, Sylvia (ed) The International Handbook of Gender and Poverty, Cheltenham: Edward Elgar, p 220-5. Wigle, Jill (2008) Shelter, location and livelihoods: exploring the linkages in Mexico City, International Planning Studies, 13, 3, 197-222. World Bank (2005) Gender Issues and Best Practices in Land Administration Projects: A Synthesis Report, Washington DC: World Bank, Gender and Rural Development Thematic Group and the Land Policy and Administration Thematic Group. World Bank (2008) Analysis of the Impact of Land Tenure Certificates with both the Names of Wife and Husband in Vietnam, Hanoi: World Bank Yahya, Saad A. (2002) The Certificate of Rights story in Botswana, in Payne, Geoffrey (ed) Land, Rights and Innovation: Improving Tenure Security for the Poor, London: ITDG Publishing, p 193-208. 56