Understanding shelter from a gender perspective: the case of Hawassa,Ethiopia

I. Introduction

This paper draws out key findings on gender and housing from a collaborative study investigating shelter provision in Hawassa, part of a wider research study on inclusive cities in East Africa. The methodology underpinning the Hawassa study recognizes that gender, poverty, displacement and ethnicity can act as major axes of discrimination that impede access to land, shelter and services in East African cities. The study objectives were to: investigate the systems through which people access shelter; establish how low-income and vulnerable groups (including women and girls, refugees and internally displaced people, and people with disabilities) are positioned within these systems; and identify targeted interventions that can allow low-income and vulnerable groups to improve their access to shelter.

Our Hawassa research included a desk review, as well as qualitative and quantitative fieldwork undertaken from April 2018 to May 2019, during which time major political reforms were unfolding in Ethiopia (with implications for the continued validity of some data). The team conducted a total of 146 interviews with residents and key informants, as well as 11 focus group discussions (FGDs) with different groups. Participants were identified with support from our local research team and/or from non-governmental organizations (NGOs) working with women and various hard-to-reach groups.

Findings from this research shed light on how gender affects shelter access in the city and point to recommendations for how to enhance inclusive responses by the Hawassa city administration, the federal government and development partners.

Section II provides background on Hawassa, the state of housing supply and demand, and government efforts on shelter provision for women and vulnerable groups. Section III provides a brief literature review of gender issues related to shelter in Hawassa. Section IV outlines the research methodology. Section V presents the findings from our research on specific constraints faced by women accessing shelter in Hawassa (particularly vulnerable women). These include specific issues around access to services and infrastructure; how gender norms and roles continue to constrain equal property rights for divorced, separated and widowed women; the role of informality and impact of regularization and resettlement on housing provision for low-income groups; and safety and security concerns associated with women living in peripheral locations. Lastly, Section VI summarizes conclusions and policy implications.

References have been inserted throughout the text to relevant sections of the original East Africa Research Fund (EARF) study this paper is based on,⁽¹⁾ where more information is available.

II. Background

Founded in 1959 as one of a handful of planned cities in the country, Hawassa is a medium-sized city located 285 kilometres south of Ethiopia’s capital, Addis Ababa. Today it is a thriving and attractive city, home to one of the country’s biggest industrial developments, the Hawassa Industrial Park (HIP), and capital of the most diverse region in the country, the Southern Nations, Nationalities and Peoples’ Regional State (SNNPR).⁽²⁾ At the time of writing, the city held the status of a “special zone” within the region⁽³⁾ and is a self-governing city administration within the SNNPR. A referendum was held in November 2019 on the creation of a “Sidama Region” – a new regional state, of which Hawassa is expected to become the new capital.

Hawassa’s most significant economic development in recent times is the Hawassa Industrial Park – a key initiative within Ethiopia’s industrialization drive and the Growth and Transformation Plan II. Designed mainly to produce high-end apparel and textile products, HIP was officially opened in 2016. It is expected to generate about 60,000 jobs and increase Ethiopia’s export revenue by US$ 1 billion per year. It has attracted thousands of young women to the city, who account for approximately 90 per cent of the HIP workers [key informant interview (KII) with McKinsey & Company, 2018].

III. Literature Review – Gender and Access to Shelter in Hawassa

There are several studies on women’s access to and use of land across Ethiopia (particularly on the impact of the national land certification process), but the focus is typically on rural areas; the regional state of Tigray; or the capital city, Addis Ababa. Overall, there are very scarce reliable data on women’s access to and experiences of shelter in Hawassa, making this a priority area for our research. We have identified only three relevant sources to date,⁽²⁴⁾ discussing experiences with condominium housing, land certification, and street women and children respectively.

In line with findings from the global literature, data from Hawassa suggest that women with disabilities and divorced, separated and widowed women all face acute shelter vulnerabilities. In Nathan’s small-scale study of women and children sleeping rough across the city,⁽²⁵⁾ 81 per cent of the women begging and sleeping on the streets were found to be separated, widowed or never married, or had disabled husbands (16 per cent). Most of these women had small children; almost one-third were disabled. A 2015 World Bank poverty assessment similarly found that nationally, elderly, disabled and female-headed households in Ethiopia’s urban areas were much poorer than those in other social categories.⁽²⁶⁾ This aligns with global evidence that divorced, separated and widowed women are at particular risk of dispossession and constitute the most vulnerable subsection of urban women.⁽²⁷⁾ It is important to note that the economic vulnerability of women in these categories is understood to be a product of unequal access to assets, finance and property, as well as traditional gender roles vis-à-vis employment and caring responsibilities.⁽²⁸⁾

A World Bank study found that female-headed households represent 33 per cent of Ethiopia’s total urban population, one of the highest in Africa, as compared to 17 per cent in rural areas.⁽²⁹⁾ Of this number, widow-led families represent a remarkable 15 per cent of Ethiopia’s urban population – many of them having migrated from rural areas to improve their livelihood prospects.⁽³⁰⁾ Multiple studies⁽³¹⁾ highlight the trend of female-headed households migrating to urban areas due to social stigma and the lack of job opportunities in rural areas. Tiumelissan and Pankhurst⁽³²⁾ found a high prevalence in Hawassa’s low-income areas of households headed by women, many of whom were migrants from rural areas, particularly Oromia and Amhara. An older study from Addis Ababa⁽³³⁾ found that most of the women heading households in informal urban settlements in the study sample were widowed or divorced.

More recent data suggest that condominiums are less popular in Hawassa than in Addis Ababa, and that women working from home in Hawassa (mainly preparing and selling food and beverages) are concerned that living in condominium housing could entail limits on their business activities.⁽³⁴⁾ But girls were more positive than boys about condominiums, citing access to modern kitchens, private toilets, and a greater sense of security as particular benefits.⁽³⁵⁾ In both Hawassa and Addis, girls highlighted proper toilets as a key advantage of condominiums, likely because of menstruation and their desire for safety and privacy.⁽³⁶⁾

While 90 per cent of Hawassa Industrial Park workers are women, recent research has found that housing is generally unaffordable for female HIP workers, who face difficulty in accessing decent housing and can be forced to live in rooms shared with up to four or five others.⁽³⁷⁾ There are also significant safety concerns as a lack of affordability, given their low wages, is forcing women to live in cheaper, more peripheral locations and to walk home at night, facing the threat of sexual harassment and assault.⁽³⁸⁾

Ethiopia has positive legal frameworks around equal property rights for divorced, separated and widowed women.⁽³⁹⁾ Article 35, Sub-article 7 of the Constitution of the Federal Democratic Republic of Ethiopia⁽⁴⁰⁾ states: “Women have the right to acquire, administer, control, use and transfer property. In particular, they have equal rights with men with respect to use, transfer, administration and control of land. They shall also enjoy equal treatment in the inheritance of property.”

Several steps have been taken by the Ethiopian government in recent years to support female-headed households to access shelter. A revised Family Code⁽⁴¹⁾ came into effect in July 2000, updating the 1960 Family Code. The law covers most areas of family relations, such as marriage, divorce, inheritance, paternity, adoption and child welfare. Legally, the revised Family Code stipulates that pre-owned or later purchased assets are co-owned by spouses after marriage.

Among the key changes pertaining to shelter in the revised law (compared to the previous 1960 Family Code) are that women are given the authority to administer common marital property; and more authority has been given to courts in settling disputes arising in cases of divorce and inheritance (and correspondingly less to traditional arbitrators). This allows greater agency for women.⁽⁴²⁾

Several studies discuss the positive impact of Ethiopia’s joint land titling (with both the husband and wife named on the certificate) upon women’s rights and economic empowerment. In the early 1990s the government led by the Ethiopian People’s Revolutionary Democratic Front (EPRDF) formally proclaimed that land rights were to be granted to men and women, including the right to lease out land.⁽⁴³⁾ A national land certification programme was introduced in the late 1990s, although not in SNNPR until 2004.⁽⁴⁴⁾ It sought to address gender bias concerns in the land-tenure system by issuing a “non-alienable joint certificate to both spouses that confers equity and joint land ownership”.⁽⁴⁵⁾

Whilst held up as a success story,⁽⁴⁶⁾ Ethiopia’s certification process has been implemented quite differently in each region. In SNNPR, regional land laws mandate the inclusion of both the husband and wife on land certificates. In Holden and Tefera’s study,⁽⁴⁷⁾ which sampled half from peri-urban and half from rural areas, around 80 per cent of all distributed certificates were found to include women’s names (wives or female heads), which is a considerable success.⁽⁴⁸⁾ In terms of perceptions, 75 per cent of households “believed that land certification contributed to increased tenure security of women” and that this would increase the probability that women would keep the land after divorce or death of their husbands.⁽⁴⁹⁾ Polygamy is widely practised within SNNPR and the legislation allows multiple wives to be listed on certificates, thereby increasing the sense of tenure security among polygamous wives, particularly later wives.⁽⁵⁰⁾ The legislation also strengthened inheritance rights, conferring equal inheritance rights to sons and daughters,⁽⁵¹⁾ but evidence suggests that widows and divorcees still remain vulnerable to dispossession.⁽⁵²⁾

Several studies⁽⁵³⁾ highlight interesting dynamics around how the certification process has played out in peri-urban areas or other areas affected by rapid urban expansion. As Deininger notes,

“Households in areas where urban expansion is imminent are reported to be particularly eager to get certificates that could help them substantiate their claims for compensation once their land is subject to urban expansion. In fact, observers link the ability to use certificates to insist on compensation to the emergence of innovative in-kind compensation arrangements in a number of peri-urban areas.”⁽⁵⁴⁾

Likewise, Bezu and Holden note a “sharp increase in land values from 2007 to 2012, particularly in Shashemene district, which is experiencing rapid urbanization”.⁽⁵⁵⁾ Early-stage fieldwork indicates that this phenomenon is likely also happening on the outskirts of Hawassa, although the gendered effects of this are not yet understood.

IV. Methodology

While the data collected for this study are primarily qualitative, we have drawn upon quantitative data from secondary sources where possible, and have also made use of a quantitative survey in order to understand household income, rental and utility costs, household characteristics, type of housing and sources of finance for construction.

This review used both qualitative (checklist based) and quantitative (questionnaire based) questions to gather data. This included 86 qualitative, semi-structured interviews; 11 focus group discussions; and 60 household surveys. Interviews and focus group discussions included people with disabilities, female heads of household living in various shelter types (informal settlements, low-cost government housing, condominiums, own housing), and select government and civil society representatives working with vulnerable women, children and people with disabilities in the city.

The 60 household surveys were quantitative (questionnaire based), and were conducted in one formal and one informal settlement (31 surveys in Philadelphia and 29 surveys in Dato respectively), in order to understand household income, rental and utility costs, household characteristics, type of housing and sources of finance for construction. The Philadelphia neighbourhood was selected as an example of a formal settlement and the Dato neighbourhood was selected as a contrasting case study site — a fast-growing informal settlement towards the outskirts of the city.

While our survey sample was not randomized, and hence has the limitation of not being formally representative,⁽⁵⁶⁾ we are confident that the findings provide an indicative and useful snapshot of affordability and perceptions in these two different settlement types.

Qualitative primary data were analysed using Excel, and by using a coding system that reflected the overall research programme’s objectives and core areas of focus. Where possible, during analysis, primary data were triangulated with secondary sources. However, the lack of reliable, up-to-date, published statistical data on shelter in Hawassa rendered this very challenging.

The principal challenge in the research was the fast-changing political landscape, with the result that data collected early in the project needed to be interpreted with caution. Political sensitivity, and the ongoing social unrest, was also a constraint in terms of what questions could be asked, given the imperative to do no harm.

Finally, a limitation of the study was that the scope did not extend to reliable and representative quantitative data that would permit a more detailed analysis of market segmentation. Our research thus paints a broad picture of how particular challenges affect certain population groups, but further research is required to quantify these inequalities to the level of accuracy required for setting budgetary targets or similar aims.

V. Findings

c. Access to services and infrastructure

Beyond affordability, our research found that access to both services and adequate infrastructure was a critical concern for Hawassa residents. Table 1 compares infrastructure provision in two settlements (one formal and one informal), where 60 residents were interviewed as part of our housing survey. The Philadelphia kebele is located within the Addis Ketema sub-city and is one of the oldest neighbourhoods within Hawassa. The Dato neighbourhood, a fast-growing informal settlement, was selected as a contrasting case-study site and a neighbourhood typically hosting a number of HIP workers. It is located within the Hawela Tulla sub-city at the periphery of the city and alongside Hawassa’s rural kebeles that are not within the plan boundary. It is one of the fastest-growing informal settlement sites hosting the city’s low-income population. Residents were asked about a range of issues including rental costs, services, shelter size and other shelter details.

OPEN IN VIEWER

Table 1

Comparing infrastructure provision in in the Philadelphia and Dato settlements

	Philadelphia	Dato
Electricity	• Variable electrical supply, with disruptions reported up to two–three times daily. Cost not reported as a specific issue. • Disruptions to electricity supply hamper use of home appliances. • Some residents have opted for private electricity connections, which offer more reliable supply. • 36 per cent of respondents felt that electricity was not affordable.	• Frequent power outages, hazardous installation of wiring. • Most residents are sharing electricity with neighbours and paying higher-than-tariff fees for this electricity. • Some residents have no access to electricity. • Limited community organizing, e.g. petitioning local authorities to provide electrical supply, which was not responded to. • 60 per cent of respondents felt that electricity was not affordable.
Water and sanitation	• Most residents have water from private taps. • Many respondents did not feel the cost of water was high, but reported that the quality of water can be variable or dirty. • Unsafe communal toilet facilities have a disproportionate impact on women and girls. • Poor design in condominium construction has led to ineffective disposal of liquid waste and problems with water supply on upper floors. • Respondents noted benefits of cleaners and street sweepers employed and provided by the municipality. • The community has established an association to tackle waste and clean up the area. • 25 per cent of respondents felt that water was not affordable.	• Some have access to water through pipes laid down previously, but many have to pay for water from private sellers/kiosks or a communal pump. • Water supply was reported to be reliable where pipes had been installed before settlement, but with limited access and irregular supply if depending on communal taps. • Acute sanitation issues, particularly with a lack of communal facilities and lack of rubbish-collection services. • Residents are paying extra for rubbish collection or sanitation due to the lack of municipal services. • 62.5 per cent of respondents felt that water was not affordable.
Transport	• Acute transport shortage. Buses and minibuses are overcrowded and infrequent. Bajaj (motorized rickshaws) and taxis are increasingly costly. • The kebele has made improvements to road safety.	• Acute transport shortage. Many residents use motorbikes, while many others walk as transport is unaffordable. • At night, taxis increase their prices by over double – which has posed a hazard to women as they then have to walk through the city after dark.
Safety	• There are many safety concerns after dark, particularly because of a lack of lighting. • Night patrols organized by residents have helped promote safety, particularly for women.	• Most residents do not feel safe, noting the lack of street lighting and extensive theft and robbery. • Lack of lighting was reported as a particular concern for women travelling after dark.

Illustrative of broader infrastructure-related issues found in our research, these case studies highlight the following issues:

Unequal costs: Residents of informal settlements in Hawassa often pay more for basic services than residents of formal settlements, who have access to utilities and services provided by the municipality. This places an additional financial burden on low-income residents that could prevent them from investing in other assets such as shelter.

City-wide issues: A number of issues, such as safety concerns, a shortage of public transport, and regular disruptions to water and electricity supplies, were found to be common to the two settlements. These are more likely to be city-wide issues.

Disproportionate impact on women and girls as a result of inadequate infrastructure: The impact on women and girls of the absence of key services can be seen with regard to unsafe communal toilet facilities, the lack of adequate public transport making it unsafe or costly for women to travel at night, and a lack of adequate street lighting.

Community organizing has been limited in both settlements and focused around specific issues, although in Philadelphia the community has established a community night patrol that has improved safety for women.

For condominium residents specifically, despite the generally better conditions there, both the peripheral location and poor infrastructure were critical concerns, especially regarding the provision of sanitation and toilets. Poor construction of the waste system appears to be compounded by a lack of clarity regarding ownership or management. One female respondent complained:

“I think the construction was not studied by professionals. It was carried out by those who were organized under small and micro business enterprises [. . .] It is full of problems. The toilet [. . .] the septic tank does not have an owner. Who owns the septic tank and to whom to appeal is not known [. . .] The biggest and major problem is that the pipeline was not properly designed and implemented. The pipelines from the top leak to the lower floors. This creates problems among dwellers; it is a common cause of complaints [. . .] When the toilet is full, it flows in the house.” (female condominium resident, FGD)

Women in our study were particularly concerned with such infrastructure problems in condominiums since they spend more time in and around the home than their male counterparts. They are also more likely to run businesses from home, which require good water and sanitation – for example, making and selling coffee or injera (flatbreads), or preparing animal products for sale/consumption. Inadequate toilets are also a particular concern for women, particularly in informal settlements, due to menstruation and concerns over privacy/safety (FGD with female household heads, Monopol area).

VI. Conclusions

Findings from Hawassa on the intersection between economic vulnerability and gender roles and norms validates the literature and global recommendations, that, as noted by Baruah in 2007, “without livelihoods, incomes and access to basic services, access to land and property will not take women very far [. . .] pro-urban land policy must be located within a broader urban development framework” that includes gender-sensitive support to raising incomes, livelihood opportunities, education, housing and infrastructure (KII, AWSAD representative).⁽⁶⁶⁾

Crucially, our research demonstrates how multiple intersecting forms of vulnerability – social, economic and spatial – act as barriers to women in accessing shelter in Hawassa. The peripheral location of condominium units, their unaffordability for poor (among whom women are disproportionately represented) and even middle-class urban dwellers, and corruption in kebele housing allocation all undermine the effective implementation of policies designed to support women.

Female-headed households; unmarried, divorced, separated and widowed women; elderly women; and women with disabilities all face severe/acute shelter vulnerabilities and are among the most vulnerable urban residents in Hawassa. These are the groups that are most likely to struggle to access both formal and informal shelter, because of their below-average income levels but also because of other forms of bias and discrimination. Despite a range of legal and policy stipulations that provide for affirmative action measures to prioritize women’s access to shelter, gender norms and roles continue to constrain equal property rights for divorced, separated and widowed women – who are also the most economically vulnerable.

In light of these findings,⁽⁶⁷⁾ the Hawassa city administration, the federal government and development partners should ensure that shelter solutions are genuinely and concretely inclusive, responding to the needs of vulnerable households and individuals, including migrant HIP workers, female-headed households, women and children facing risks of sexual violence in informal settlements, extremely poor households, resettled households and people with disabilities. It is important that inclusion goes beyond policy design to include implementation, which requires close monitoring and beneficiary feedback loops.

The city administration should actively support the formation of housing cooperatives to increase the supply of affordable housing. They should mobilize relevant bureaux and sub-city offices to explore multi-sector options for preventing violence against women and girls, including the provision of safe transport, improvements in water and sanitation facilities, and consideration of reliable daycare facilities for children whose parents work away from home – as well as data collection that allows for adequate assessment of the needs, and monitoring of responses where they are provided.

The Hawassa city government should provide more serviced land for housing at affordable lease prices. This might entail an additional mechanism that permits allocation of land to low-income households, outside of the competitive bidding system that is beyond the reach of these households.

To ease the acute demand for housing, the municipal government should also carefully consider regularization of eligible informal settlements according to the existing legal framework. To mitigate the risk of gentrification, this should ideally be done at scale, using alternative standards, with community involvement/ownership, and with consideration of collective landownership options. The authorities should ensure that any evictions of informal settlement residents are managed with maximum diligence and consultation. The municipality should also engage with residents of informal settlements to formulate responsive policies for the sustainable provision of shelter. If involuntary resettlement is unavoidable, best practices can be drawn from international standards such as the International Finance Corporation (IFC) performance standards on environmental and social sustainability, specifically Performance Standard 5 on Land Acquisition and Involuntary Resettlement, to safeguard vulnerable residents of informal settlements.

Recent reforms offer a major opportunity to involve communities and civil society organizations (CSOs) in organizing, mobilizing and advocating around issues of shelter affecting low-income groups, especially to address the acute challenges affecting women and others facing particular risks.

Companies operating in the HIP should provide proactive and holistic support for the welfare of their staff, including minimizing their exposure to violence on the journey to and from work. It is in these companies’ interest to improve shelter for workers, in order to boost worker retention and the park’s success.