ADDRESSING AFFORDABLE HOUSING IN A MODERNIZING CITY: THE RWANDA EXPERIENCE

Transcription

1 ADDRESSING AFFORDABLE HOUSING IN A MODERNIZING CITY: Summary: Major progress has been made in the last decade by the Rwandan government in terms of paving the way for sustainable urban development and city management. Key achievements include: the successful implementation of a nationwide land titling program, and the development of a modern vision for Kigali City set forth in a Conceptual Master Plan. Yet, much remains to be done in the housing sector. The cost of housing is generally too high too much money for too little housing as a result of which more than 80 percent of Kigali s population lives in informal settlements. This may, at least in part, be attributed to the detailed district-level master plans being developed, according to which much of the central business district has been re-zoned, with an emphasis on larger plots and multi-story, multifamily residential buildings new developments as opposed to in-situ upgrading. This is having the effect of 'informalizing' the existing housing stock, some of which is of decent quality, and at the same time, not catering to the needs of lower income groups who do not conform to the overall 'vision' of the city. This paper presents the housing story of Kigali city from the perspectives of the supply and affordability of urban housing and housing finance. Based on the wide range of housing typologies existing within the informal sector and the constraints facing the market, it makes recommendations for a range of practical policy interventions in both housing (self-built and developer-built new affordable housing, rental housing, in-situ upgrading, and social or subsidized housing, as necessary) and finance (mortgages, construction finance for developers and individuals, housing microfinance, and subsidies) - to make Kigali attractive not just for the wealthy few, but also livable and affordable to the vast majority of its residents. Key Words: Kigali, Rwanda, housing, finance, affordability, urban development, master planning

2 ADDRESSING AFFORDABLE HOUSING IN A MODERNIZING CITY: I. INTRODUCTION Rwanda is one of the least urbanized and most densely populated countries in Sub-Saharan Africa (SSA). 1 The current level of urbanization is 19 percent compared to an average of 42 percent for SSA and over 50 percent globally. Based on UN population forecasts, Rwanda s urbanization is expected to reach 43 percent by This has implication for urban policy including the need for future housing construction. According to the 2002 population census, Kigali, the capital and primary urban center, has experienced an average annual growth rate of 9 percent between 1991 and In the next 25 years, over one million people are projected to migrate to cities in Rwanda. According to the Kigali Concept Master Plan, if current trends continue, Kigali is expected to bear the brunt of the urban growth pressure created by this extremely high rate of urban migration. This will mean increased demand on urban and community infrastructure and developable land within the city boundaries. The Rwandan government, in particular, Kigali City has made major progress over the last decade in terms of paving the way for sustained urban development and city management. Yet, much remains to be done in the housing sector. The cost of housing in Kigali is generally too high too much money for too little housing and this applies even to middle income households. However, for now let it suffice to say that as a result of this housing market situation, more than 80 percent of Kigali s population lives in informal settlements. Objective of the Study. The overall objectives of this study were: - To produce a baseline diagnostic report which will enable a better understanding of the housing needs of middle and lower income groups in urban Rwanda. - To help Kigali City and the Government of Rwanda to prepare for rising urbanization and the housing needs that come with it. - To explore options for green housing, and highlight opportunities for expanding the use of domestic building materials in housing construction. This is especially important for a landlocked country very reliant on expensive imported materials. - To complement work being done on the financial sector side with a housing supply-side analysis. 1 This excludes island states such as Mauritius or Comoros. 2

3 This paper provides a qualitative assessment of the overall housing market in Kigali city. It is based on a cross-sectoral approach that presents the housing story from the perspective of the supply and affordability of urban housing and housing finance, and makes recommendations for practical interventions needed to make Kigali attractive not just for the wealthy few, but also livable and affordable to the vast majority of its residents. Content and Research. The work involved intense qualitative field research in Kigali carried out between December 2011 and June 2012, funded by the Multi Donor Trust Fund (MDTF) and implemented by the Housing Finance group at the World Bank. The research included: Detailed household interviews with residents of a range of settlements in Kigali city Focus group discussions with several different groups of people: urban professionals (architects, developers, academics); mid-level staff and young interns (new recruits) in government; and a local community of craftsmen currently living in very poor conditions. Meetings with commercial bankers and a financial cooperative to assess the sorts of loans being offered in the market, and the eligibility criteria and borrowing capacity of the customers. Interviews with developers to understand their market products, the target clientele, and the problems they face in housing development, particularly as they pertain to building housing for middle and lower income groups. Interviews with government officials at the national and Kigali City level to get a general understanding of the priorities and on-going programs in the housing and urban development sector. This qualitative research presented a basic picture of housing in Kigali -- i.e. income levels of local residents, existing housing typologies and associated costs, current trends in housing construction, availability of and access to housing finance, and government policies/ programs affecting the market. This data was used to purposively stratify the market and establish housing affordability in a way that is simple and easily to understand. Structure of the Paper. Section II of the paper provides information on the existing housing stock of Kigali city - the various housing typologies, what they cost, and who is living in them. Section III presents an analysis of the cost and affordability of housing, both owner-occupied and rental, and highlights the demand-supply mismatch characterizing the existing market. Section IV presents a discussion on the constraints in the delivery of housing from the perspective of both supply-side issues (such as costs associated with construction and land, and underlying policies that are indirectly 'freezing' the market), as well as finance (availability of and access to housing finance). Section V briefly provides recommendations to practically and comprehensively address the housing problems in Kigali through interventions in the urban housing as well as housing finance sectors. Finally, Section VI presents some key lessons that Kigali's experience has to offer to other modernizing economies and cities. Limitations. First, given the scope and time/budget limitations, the focus of this analysis has been confined to Kigali city. Second, this paper is not, and does not pretend to be, the outcome of a quantitative or statistical analysis; instead, it is a broad-brush qualitative housing market assessment, a first step that will assist the City of Kigali in carrying out a more detailed demandbased quantitative analysis in the future. Finally, this study was commissioned to inform financial sector work in Rwanda, and accordingly the scope is limited to housing and urban 3

4 issues that are directly relevant to housing finance; as a result, it does not include extensive discussion on urban issues such as solid waste management or building codes that are very critical to Rwanda's urban sector. II. KIGALI'S HOUSING STOCK Formal sector housing in Kigali caters to less than a fifth of the city s population. Informal settlements, on the other hand, occupy 62 percent of the land area of Kigali, and house 83 per cent of the city s population. Although a clear number is not available, gauging from aerial photos, it can be estimated anywhere between 80 and 90 percent of the total stock (in terms of # units) is informal. It is not just the poor who live in informal settlements, but also most middle income earners with a household (HH) income of RF600K-RF1.5M per month (USD per month). 2 Many households in RF2M-RF3M per month (USD per month) income range also live in informal settlements because formal sector housing is either unaffordable or unavailable (see Box 1). Box 1. Upper-middle income HH living in an informal settlement Charles is a self-employed wedding planner/ contractor, with a household income of over RF3M (USD5,000) per month. He is 31 years old, and lives with his wife and son in a 3BR rental in an informal settlement in Kigali. The rent is RF100K (USD 170) per month. When he moved into the house, he paid the landlord 20 months rent upfront. The house has 3 bedrooms, and an indoor toilet and shower. The annex has a store that he uses to store his equipment, and two workers' quarters, one each for the guard and nanny. There is also a pit latrine and a bathing area for the workers, and a small vegetable garden. The house is constructed with a mix of mud brick and stone, and has an iron roof. The house has electricity and running water, both included in the rent. A health center, a gym, and a church are located within walking distance. All members of the family have health insurance. Charles is among the fortunate few who have a relatively decent house, even though it is informal. However, most households in Kigali earn significantly less than he does, and have lesser disposable income for housing than he does. A middle income household defined in this report as one with a monthly HH income of RF600K-RF1.5M (USD1,000-2,500) typically lives in a small 2-3BR unit, with a shared pit latrine that is often ill-maintained, for a rent of RF40K-80K (USD65-140) per month. Source: Interview by WB study team in December Formal housing. Formal sector housing in Kigali is typically either being built by developers in the suburbs, or by individuals closer to the city center who are buying parcels and building large up-scale houses. Such housing is typically priced at RF50M (USD83K) upwards for 2BR singlefamily homes or condominiums. For example, a 3-4BR single-family house can cost anywhere between RF75M (USD125M) and RF120M (USD200K); a 5-8BR villa (estate housing single family) is priced RF150M (USD250K) and upwards. Much of this housing is affordable to only those in the top 2-5 percent of the income spectrum. While there is some indication of private developers interest in venturing into smaller format (2BR) homes in the RF45M-50M (USD75K-83K), this market is practically non-existent at present. Of all the developers interviewed as part of this study, there was only one building 2-3BR houses priced at RF46M at the time of the field work (March 2012). Similarly, rents for formal sector housing are quite high: 2 USD1 (US Dollar) = RF (Rwandan Francs). The USD conversions used in this paper are rounded for ease of reading. 4

5 ranging from USD500 to USD1,000 per month for a 70m 2 2BR unit, to above USD3,000 for a luxurious villa. Formal peri-urban housing. In response to the lack of affordability to buy or build a house in the city, there seems to be a growing trend of low and middle income households selling their homes in the city and moving to the outskirts or peri-urban areas, where building standards are lower and construction more affordable. This is resulting in sprawl, 3 which could be costly to the city in terms of planning for future growth. Informal housing. There is a wide spectrum of housing types in the informal settlements: a large percentage of reasonably good quality houses, and some closely resembling 'slums' in other developing countries. The vast majority of informal housing is located in Nyarugenge district, which is also home to the City s Central Business District (CBD). Compared to cities in other developing countries, a key distinguishing characteristic of informal settlements in Kigali is that the vast majority of the households except those in hazardous locations have land titles. This is an outcome of the national level land titling program. Titles have been awarded to every property owner make sure that no household is unfairly displaced (without due compensation) as the development stipulated in the District Masterplans starts to happen. Yet, despite having titles, households are not making improvements or additions to their homes. In Nyarugenge district, this is reportedly because of the standards and zoning regulations set forth in and enforced per the new District Master plan, which renders much of the existing housing stock informal. No new construction permits are being issued by the City even for renovations or minor expansion work unless the proposed house (after renovation) fully meets all the District Master plan regulations and codes, which the vast majority of the population cannot afford. In the other two districts, Gasabo and Kicukiro, no new construction permits are being issued until their respective District Master plans are finalized. 4 Rental housing. According to the 2010 Third Integrated Household Living Survey (EICV3) (National Institute of Statistics, Rwanda), renters constitute over 40 percent of the city s population. Our field work seems to suggest that unlike landlords in other countries who are typically either very rich (speculative, owning multiple properties) or very poor (typically the case in informal settlements and slums), a large percentage of renters in Kigali fall somewhere in between. They include, in large part, urban professionals, salaried government workers, military personnel, and other formal sector wage employees. Typically, lower income home-owners have one or more rental units on their plots. Many of these poor landlords include vulnerable groups e.g., widows, single mothers, orphans. Rental units on their properties serve as the primary and often the only source of income; some even live in a smaller 1BR annex and rent out the bigger main house. 3 For more on the negatives of sprawl, see: 4 Kigali City s Conceptual Masterplan (2009), approved by the Parliament, sets out the overall development framework for the city. Subsequent to the Conceptual Masterplan, a District Masterplan was completed for Nyarugenge District, which spells out in detail the new zoning regulations. Masterplans for Kigali s other two districts (Gasabo and Kicukiro) are now underway, and expected to be completed by December

6 For the middle and low income households, rents are extremely high relative to income levels and the quality of the unit. This is due to a shortage in supply; for instance, a 1BR unit with a small living room, about 20m 2 in area, rents for anywhere between RF20K (USD35) and RF40K (USD70) per month, depending on the location. A 2BR unit rents for RF60K-80K/m (USD ). These units typically have shared pit latrines in an annex. 5 Larger 3-4BR units with indoor flush toilets have a rent of RF80K-150K (USD ), again depending on location and amenities. Rents are reportedly very volatile. Anecdotal evidence suggests that they can be doubled overnight, based on demand. This is part of the reason why landlords are not in favor of long-term rental agreements or contracts which limit their ability to increase the rent for an extended period of time. In the formal sector, the more commonly contracted period is 6-12 months. Tenants go through agents or listings, 6 and pay a retainer of RF100K (USD170) and an equivalent to half or one month s rent. In the informal sector, rental deals are made through word-of-mouth, or an informal 'commissionaire' who charges a fee of RF100K-200K (USD ). The contracts are typically short, ranging between 2-4 months. The landlord takes the rent for the contracted period upfront; after that, it is month-to-month, and vacating the premises requires 15 days' notice from either party (tenant or landlord). Infrastructure/ Basic Services. Kigali s water distribution system appears to be in place: most homes have a tap, or a convenient access to a shared standpipe. Similarly, electricity is widely available, and distribution systems offer the flexible means to modulate usage through a pay-asyou-go meter system. Sanitation, however, remains highly inadequate. There is no central sewage system in the city. Most houses have pit latrines that are self-maintained - in many cases, poorly maintained or simply not dug deep enough due to constraints with the sloping terrain or the water-table. In some of the more dense and poorer settlements closer to the wetlands, human waste is collected in buckets and disposed in the wetlands at night. Only the more recently houses (with permits) or the formal homes have indoor toilets connected to septic tanks. For new developments, exceeding 50 persons in population, the city has mandated the use of sewage treatment plants to be provided by the developer. Wastewater is typically collected in a pit (separate from the latrine pit), or simply channeled into an open drain. Most city communities do not treat or reuse their grey water; concrete gutters are built to direct this water to the wetlands. This has two negative effects: one, the concrete gutters are also the pathways for pedestrians; constant exposure to water creates slippery and hazardous conditions. And two, the steep slopes of the drains channel the water into the wetlands without allowing for adequate earth-filtration of the grey water, a natural process where water pollutants can be allowed to filtered out slowly before reaching natural wetlands. Public Transportation consists of a combination of city buses, privately run buses, taxi cabs and two-wheelers operated by private individuals. Transportation in general is non-integrated and expensive. For example, a cross-town bus trip from Kimisagara to the airport, about 6 miles by 5 'Annex' is the local term used to refer to a building on the compound, separate from the main house, and smaller. 6 For an example of a web portal/listing (equivalent to Craigslist in the US), see: 6

7 road costs RF250 (USD 0.40) one way. This implies a monthly travel cost of RF10K or USD16 (RF500/day*20 days per month). This is very high relative to income levels. For example, in Washington DC, the cost of commuting cross-town by bus is approximately USD1.25; this translates to a monthly travel cost of USD50, a much smaller proportion of US incomes, even for low income households. Similarly, in Mumbai, India the cost of a similar crosstown bus ride is approximately USD0.60; this translates to a monthly travel cost of USD24. Roads. Major arterial roads are paved and graded to suitable slopes; however, local and collector roads are not. In many cases, these tertiary access roads not even leveled so much so that it is extremely difficult to get into these areas, especially during the rains. In many of the poorer areas, a large percentage of houses are only accessible on foot. This often results in hazardous conditions, especially during periods of heavy rainfall. Large landholders such as RSSB build roads and other common infrastructure (drainage, street-lighting, etc.) for their housing developments. This cost is substantial, adding per cent to the cost of the house, which in turn pushes up its selling price. Housing Typologies. Figure 1 illustrates the typical housing options available for the various income groups in Kigali. It is important to note here that while every effort was made in the survey to get accurate numbers on income and expenditure, these are ballpark / illustrative figures derived from a qualitative survey using a purposive sample of households. Only those in the top income bracket defined here as those earning over RF1.5M (USD2500) per month are able to afford any formal sector housing in the city. For those in the 2 nd and 3 rd income tiers, i.e. earning between RF250K-RF1.5M (USD400-2,500) which is the bulk of the population of Kigali the only affordable formal housing options are in the suburbs or peri-urban areas; within the city, only informal housing is affordable. Tiers 4 and 5 live in relatively poorer conditions, and in more precarious locations (e.g. nearer to the wetlands), with inadequate sanitation. Figure 1. Type of housing, by income group Note: (i) The income tiers in the diagram do not represent income quintiles: the percentage of households in each income tier is unknown. (ii) HH expenditure information gathered from the survey has been used as a proxy for HH income. 7

8 Table 1 presents in summary the various income/ population groups living in the various housing types (broadly categorized as formal, informal, and 'slums'), and the illustrative costs associated with such housing, which varies by location (distance from the city center). 8

9 Table 1. Housing Categories and Characteristics: Formal, Informal, Slums FORMAL INFORMAL SLUMS HH Income /m Typical population groups Location House configuration/ description Typical rents >1.5M Expats, diaspora, wealthy Rwandese City-center Large 4+ BR single family estate housing on plots >1000m 2 >RF500K /Suburbs >1.5M Expats, diaspora, wealthy Rwandese City-center 2-3 BR multi-family multi-story housing (70-100m 2 ) also rental >RF300K RF600K- 1.5M Senior officers-military, government, businessmen (formal sector) Suburb Relatively large 3-4BR house, well-constructed single family homes on plots ranging from m 2 (in accordance with standards) RF20-40K for a 1BR annex RF600K- 1.5M Entrepreneurs-informal sector Peri-urban Single family 3-4 BR units (~100m 2 ) on m 2 plot, with relatively good construction RF20-40K for a 1BR annex RF250K- 600K Mid-level government officials, private sector (white collar), military personnel (junior level); entrepreneurs (informal sector) Suburb Single family 3-4BR house (~120m 2 ) on large plot ( m 2 ). Land purchased from savings and house constructed with loan. RF20-40K for a 1BR annex RF250K- 600K RF250K- 600K (and higher) RF600K- 1.5M (and up) RF250K- 600K RF100K- 250K Junior-mid level government officials, military personnel; entrepreneurs (informal sector) Mid-level officers, military, government, businessmen (formal sector), Retired formal sector employees Mid-level officers, military, government, businessmen (formal sector), Young professionals, mid level government officials, private sector (mid-level) Young professionals, entry/ junior level government officials, private sector (entry level jobs), teachers, nurses, poor landlords <100K Students, unemployed youth, poor landlords, informal sector vendors, laborers, car wash, parking attendants, porters, househelp/ maids <250K (sometim es more) Students, unemployed youth, poor landlords, informal sector vendors, laborers, car wash, parking attendants, porters, househelp/ maids Peri-urban Suburb City-center City-center City-center City-center City-center Definitions: City-center: Within or close to the CBD Suburb: Far from the CBD, but within the urban area Peri-urban: At the periphery of the urban areas, with more rural characteristics Single family 3-4BR house (150m 2 ) large plot (600m 2 ). Land purchased from savings and house constructed with loan. Traditional 3-4BR single-story single-family mud houses of relatively decent quality. Mud construction, with plastered walls and cemented floors. Informalized by new District Masterplan regulations. 3-4 BR units with relatively poorer construction, in dense informal settlements in prime locations. Many households in this income bracket are renters. 2-3BR units with relatively poorer construction. Many households in this income bracket are renters. Traditional mud construction, mostly of poor quality. Often multiple rental units on a single plot. Rental income constitutes a large share of owner s HH income. Traditional mud construction typically of poor quality. Often multiple rental units on a single plot. These are mostly inherited or shared ancestral homes, but rent is the primary income for the landlord. Dense, very poor sanitation, located on steep slopes or near wetlands; high percentage of renters - 2/3rds. RF10-20K for a 1BR annex RF20-40K for a 1BR annex RF20-40K for a 1BR RF20-40K for a 1BR RF60-80K for a 3BR RF40-80K for a 2-3BR RF8-20K for a 1-2BR 9

10 III. THE AFFORDABILITY GAP Cost and Affordability of Ownership. There is not enough new formal sector housing being built to accommodate the city s existing and growing needs. (Several reasons are attributed to this, as discussed in section IV.) This applies even to those in the middle and upper income tiers: there is virtually no formal sector housing supply for households earning RF250K-600K (USD ) per month who could potentially afford to purchase a house costing between RF9M-21M (USD15K-36K), or those earning RF600K-RF1.5M (USD ) per month who could afford housing in the RF21M-54M (USD36K-90K) range (Table 2). Income Tier HH income/m Table 2. Housing price and affordability (ILLUSTRATIVE) Price of cheapest formal sector house available today 1 RF>1.5M >RF70M (suburbs developer built) 2 RF1.0M-1.5M RF15M-30M (suburbs - self-constructed) RF600K-1.0M RF15M-25M (suburbs - self-constructed) THE AFFORDABILITY GAP(*) Price of house that would be affordable but is not available RF54M-70M $90K-116K (in-city) RF36M-54M $60K-90K (in-city) RF21M-36M $36K-60K (in-city) 3 RF250K-600K RF10M -20M RF9M-21M $15K-36K (in-city) (periurban self-constructed) 4 RF100K-250K None in-city; only informal housing RF3M-9M $6K-15K (in-city) 5 <RF100K None in-city; some government subsidized housing in peri urban areas <RF3M <$6K (in-city) Note: (*) Assuming a 15 year interest; LTV 70:30; monthly repayment 40% of HH income. Based on this, a family can afford a house roughly 3 times the annual HH income. Note that this is a rough but fair guide, although the situation will vary based on individual circumstances. Also, the lower income tiers might be able to spend a lesser percentage of HH income (than 40%) on loan repayments for housing than has been assumed here. In other words, there are three available options to aspiring home-owners: One, to buy a formal sector house that costs at minimum RF70M (USD116K): this is unaffordable to the majority. Two, to build a house in the peripheries for RF10M-30M (USD17K-50K), which low and middle income households are beginning to do: but this is environmentally unsustainable because it is causing sprawl. Three, to occupy or retain a traditional ancestral house, which is what most households are currently doing: but this is eventually not sustainable if improvement and expansion of the existing housing stock does not or cannot happen simultaneously. This extremely limited menu of housing options in the city, coupled with a gap of housing in the RF9M-70M (USD15K-116K) price range is resulting in more and more people, including middle-income earners, either moving to the peripheries, or living in housing that is underserviced and overpriced. 10

11 Cost and Affordability of Renting. With respect to rental housing, households in the top income tier spend lower on rent than they can actually afford (<15% of HH income); this is a direct consequence of lack of supply of housing for income Tiers 1 and 2 see Table 3. As a result, households who might otherwise be willing to spend more on housing that is of better quality have no choice but to settle for what is in their affordable price range in the market. Conversely, households in the lowest income tier spend disproportionately more, between percent of total income on rent, which indicates a general shortage of supply. Table 3. Rental housing costs (ILLUSTRATIVE) Income Tier HH income/m Type of unit Rental payments Ideally, rental payments should be: RF per month as % of income RF per month as % of income(*) 1 RF>1.5M 3-4BR RF100K-250K ~<15% RF375K-500K 25% 2 RF600K-1.5M 3-4BR RF50K-150K ~<20% RF120K-300K 20% 3 RF250K-600K 2-3BR RF60K-100K ~20-35% RF50K-120K 20% 4 RF100K-250K 2-3BR RF40K-80K ~20-35% RF15K-50K 15% 5 <RF100K 1BR/ shared RF10K-30K ~30-50% RF10K-15K 10-15% Note: (*) By international standards, 25-35% of HH income is considered a reasonable estimate of rental expense on housing. However, this percentage is reduced for lower income households due to their inherently lower disposable income (basic needs such as food constitute a larger share of HH expenditure). Demand Supply Mismatch. Figure 2 below illustrates how the lack of supply of affordable housing for the middle income tiers is thus exacerbating the lack of affordability for households in the lower income tiers. As a result, (i) only the wealthiest can afford any formal sector housing within the city, (ii) the middle income groups get pushed to the suburbs and the peri-urban areas, and (iii) the poorest have no housing alternative other than informal housing in the city. Figure 2. Lack of formal housing supply: Impact on middle and lower income households 11

12 IV. HOUSING MARKET CONSTRAINTS Several underlying constraints pertaining to the supply of formal sector housing and housing finance have led to Kigali city's current housing situation. Housing Supply: (i) High cost of construction: This is attributed to expensive construction materials most materials are imported and subject to taxation of about 40% (VAT plus other taxes). The building code generally disallows the use of traditional materials (e.g. mud blocks) for construction, which are a fraction of the cost compared to modern materials such as cement. Alternative materials that are cheaper are being explored, but the market is yet to get established. In terms of labor, larger foreign contractors often bring their own pool of specialized labor, which tends to be more expensive, and furthermore, inhibits the development of technical skills in the local labor force. (ii) High cost of land: The existing pool of land owned by most developers was reportedly purchased in the past when its cost was relatively cheaper (2-10% of total project cost) and more easily available. However, today it is difficult to find large land parcels in the city at a reasonable cost for housing projects of a meaningful scale (say, more than houses) that are affordable, especially since much of it is subdivided and/or occupied and must be bought from individuals and consolidated. (iii) High cost of on-site infrastructure: Given the sloping terrain, and the large format single family homes, the cost of on-site infrastructure is often quite significant after factoring in roads, retaining walls, landscaping, street lighting, sewage treatment plant etc. (iv) Limited capacity of developers: Most developers, small or big, lack the skill sets for project planning, especially with respect to housing projects. Lack of credible on-the-ground data on housing makes it difficult to assess or adequately respond to market demand. Financing is also another barrier to entry, particularly for small developers. (v) House design and configuration: In a situation where the base cost of construction relative to income levels is already very high, the factor that exponentially affects the cost is the size of the unit, which in Rwanda is very large compared to other countries. The notion that Rwandans like to live in larger homes is one that is commonly heard in many African countries, but one that needs to change in the interest of efficient land/ resource management and city planning. (vi) New zoning regulations: Much of Nyarugenge district has been re-zoned in the District Masterplan (with larger minimum plot sizes ranging from 250m 2 to 1000m 2, higher Floor- Area-Ratio, etc.), which in effect has rendered much of the city's housing stock, including some very good quality housing, informal (see Boxes 2 and 3). The emphasis is on multi-story multifamily residential buildings as opposed to single family housing, particularly for low and middle income housing, and new developments as opposed to insitu upgrading. Accordingly, no permits are being issued for construction (including home improvements) unless the house plan and underlying structure is in full compliance with the new zoning regulations. 12

13 In the other two districts, Masterplans are currently being developed, and during this interim period, no new construction permits are being issued. This is having the effect of, on one hand, freezing the construction market so much so, that reportedly many of the building materials suppliers across the city are going out of business, and on the other, serving as a push factor out of the city for both individuals and businesses. Box 2. Interpreting the Masterplan: minimum versus maximum standards? According to standard international practice, density and FAR requirements are maximum limits that are put in place to guide development--and revisited/ updated periodically--so that the city infrastructure is able to cope with the projected population over a period of time. In the case of Kigali, it appears that the densities and FAR guidelines set forth in the District Level Masterplans are being interpreted as the minimum standard. According to the Nyarugenge District Masterplan, most existing homes must be re-developed per the regulations listed below (not an all-inclusive list): Parcel/lot sizes minimum for residential: ranging from 250m 2 to 1000m 2 FAR: ranges from 0.8 to 2.4 Ground coverage: ranges from 30% for high rise districts to 60% for a landed house. Setbacks: side and rear ranges from 3m to 5m; front setback ranges from 5m to 7m. Height regulations: 1 story for low rise districts to 16 stories for high rise districts. Permitted uses: Primarily residential; does not allow additions and alterations of non-conforming users if there is any material change in the non-conforming use, unless it is per the zoning code. By these specifications, virtually all existing lots are non-conforming, either due to lot size or setback regulations. Proposed FAR and height regulations mandate a much higher density than most existing parcel sizes can sustain. The mandated minimum plot sizes are expected to be achieved by combining existing plots, as needed. In addition, the streets and other public rights of way required for such densities will have to carved out of existing parcels. This is going to be a challenge since most parcels in neighborhoods such as Kimisagara are currently accessed by narrow pedestrian paths. The most likely outcome will be that existing residents will likely move to suburban or peri-urban locations without the adequate infrastructure (utilities and transportation) for such a migration. 13

14 Box 3. Informalization of old but good quality housing stock Tomas is a 56-year old widower, and an entrepreneur with two businesses: one, he operates a truck and transports construction materials to building sites. And two, he operates a mini-bus passenger taxi service. Tomas has two children: a 15-year old son, who attends school, and a 28-year old daughter, a nurse, now married and living elsewhere. Also living in the house is a maid, and a couple who rent one room and a small shop on the property. Tomas bought the current residence in Kicukiro district as an unfinished house with his savings of RF1M about 15 years ago. He then incrementally improved it to the current state over a period of about 10 years. The unit now has a living room, 3 bedrooms, an annex (with a kitchen and a pit latrine/ bath), and a store in front, which is now being used as a shop (rental). It is constructed with burnt bricks and roofed with iron sheets. It has a plywood ceiling. The floor is cemented. The total area of the house is 101m 2, and the plot 341m 2. The property was valued at RF15.4M (in February 2011) and used as collateral for a business loan. Tomas got title to the land in June 2011 for a 20-year lease. Starting 2012, he will start paying land rent of RF35K per year. Now that he has a title, Tomas says he will demolish the house and put up a storied house -- in accordance with the zoning laws stipulated in the forthcoming Kicukiro masterplan. He plans to start this 5 years from now using a loan from a bank and whatever savings he might have. Left: Tomas s house; Right: Stephan s house Stephan is a 58-year old retired civil servant/professional. He got a degree in Management from Switzerland, then lived in Burundi from 1983 through 1994, when he repatriated to Rwanda. In Rwanda, he worked for the Social Security Fund before retiring. His wife is 38 years old, and working at EWASA (national water/ sanitation utility company) for the last 15 years. They have 8 children: five boys aged 7, 12, 14, 16, 23 years, and three girls aged 15, 18, and 20 years respectively. The oldest 4 children are adopted. All the children attend school/ university. Stephan now works as a consultant. He helps individuals and businesses in a variety of areas, such as; audit, pensions, accident claims, individuals who are operating businesses in informal sector and wish to transition into the formal sector. Income from consulting is very little, he says, and he has not kept track or his earnings. His main income is the pension of RF231K per month he gets from his previous government job at Caisse Sociale. Stephan s wife makes RF150K per month, which makes the total household income about RF381K per month. Stephan bought this land in Nyarugenge district in He designed a 6BR house and started to build it incrementally in The house has a flush toilet and a bathroom. The house has two annexes, a pit toilet and a bathing area for the house help. Over the years, Stephan has done a lot of renovations on the house. The walls are made of mud bricks supported by a wood structure. The roofing is done with iron sheets. Cost of land (1998) Cost of construction (2000) Estimated current value of land Estimated current value of property Estimated value if main road gets paved RF50M RF400K RF5M RF5M RF30M Today, this house is deemed informal, and Stephan cannot make any changes to it unless the new plan conforms to the new zoning laws. Source: Interview conducted by WB study team in December

15 Housing Finance: (i) High cost of construction financing: Interest rates for construction loans range between percent, which is a significant financing cost, especially to small developers. In addition, banks require a 20 percent upfront contribution towards the cost of construction. 7 (ii) Limited access to financing: The current rates of interest make longer terms loans too expensive in the long run, and medium-term loans unaffordable (in terms of monthly payments) for the vast majority. Salaried individuals are able to access loans; similarly, a land-owner formal or informally employed can borrow using the property as collateral, so long as the income is justifiable (see Boxes 4 and 5). However, given the low level of incomes and savings, they cannot qualify for big-ticket loans that correspond to the high cost of housing or housing construction. Renters employed in the informal sector who do not own property have practically no ability to borrow from formal sector financial institutions. 7 This accounts for the developer s equity, since land is not considered valuable enough. 15

16 Box 4. Financing a home: With savings and loans from a Financial SACCO Luke is a Lieutenant in the Army. He lives with his wife, and two young girls: a 2-year old and a 2-month old in a single family house in a suburban area of Gasabo district. Luke earns about RF200K from the military service. In addition, he is a minibus operator; this business fetches him about RF10K-15K/day, i.e. a total of about RF450K per month. Luke s wife has just started a job as a nurse in a private government hospital, and her income is RF260K per month. The total household income is thus about RF910K per month. Luke was previously a renter. He married in 2005, and wanted to buy a plot in this location. He joined a savings association in 2001, where he contributed RF5K per month, and saved additional funds in his bank account. Once he had saved enough, he was able to buy the plot in 2004 for RF400K. The plot allocation was done on a lottery system. The building cooperative of the area undertook the responsibility for the overall site planning. He then approached Zigama for a loan to construct his house. The actual cost of construction was about RF20M. He used RF5M from his savings, and took a construction loan of RF15M from Zigama 10 years, 13.5% interest using the land as collateral. The monthly repayment is RF230K. After the house was constructed, he then took another loan from Zigama to buy a minibus, the proceeds from which are used to pay off the house loan. In the case of this second loan, the collateral was the finished house. Cost of land (2004) RF400K Current estimated value RF8M-10M Construction Cost RF20M Estimated value of property (2011) RF74M (appraised by bank) Left and middle: Luke s recently built house. Right: His neighbor s property. Today this RF400K plot will cost RF8M. Someone recently bought a similar plot nearby for RF16M. A finished house like mine will today cost 40K to build. My house was valued at RF74M by the bank in May With respect to a loan, it is important to understand the implications, and how to manage it. This is particularly the case for those who earn less. For example, even in our case, we first built the annex and lived there. And only once the main house was finished did we move here. It has to be a step by step progression. The main problem in this city is land access: too much speculation, prices too high..the city guidelines require a standard size and service level for the plot, which makes it expensive So there is a problem of inequality. A condominium is an excellent idea. But the cost of such a house is too much. It s not so much of an acceptability issue from the point of view of design ; it is more to do with the fact that people don t have money to finance / buy such a unit. Source: Interview conducted by WB study team in December

17 Box 5. Financing a home: With savings and loans from a bank Timothy is a 30-year old entrepreneur. He imports photography materials and sells locally in Rwanda. His wife, Docas, 25 years old, is young graduate, having recently completed her Master s program in Human Resources Management at a local university. She works as a Customer Service staff at Access Bank. Timothy, on the other hand, dropped out of school after Secondary 3. Docas and Timothy got married about 6 months ago. Also living with them is Docas s 9-year old daughter from a previous marriage. Timothy s income from his business is about RF900K per month. His wife s income is RF245 per month. Prior to moving here a month ago, Timothy was living in Kaciyru in a 2BR rental unit, paying a rent of RF50K per month. It was only when the couple got married in November 2011 that they moved here. In 2008, Timothy met a friend who was selling this plot, which Timothy decided to buy. He paid RF900K in cash from his savings for the land (25m x 30m = 750m 2 ). He built the basic structure in 2011 using his own funds of about RF6M. He needed another RF7M to complete the construction, for which he approached a Bank Populaire where he had a savings account. The house and land served as the collateral, valued by the Bank at RF22M. He was given the loan at 19% annual interest for 5 years. The approval took 2 months to process, and he is currently repaying RF180K per month. Land cost (2008) Construction cost (2011) Property value (2011) RF900K RF13M RF25M This is the annex, says Timothy, We will build a proper house with 5 rooms next year once I have been able to pay off this loan. Better to pay off the current loan before getting another. God has helped me through this whole process. I have savings with which to start construction. And with Jesus, everything will come in time.. Left: The new house, which will eventually become the annex once Timothy builds the main building in the large plot of land in front of the house (middle photo). Right: living/ dining area As of now, no public transport can come here because of the bad roads. We are about a 30-minute drive to the next paved road where we can access public transportation. That makes our commute very long and difficult. According to the Masterplan, the road in from of the house will get paved, and that will make access much easier. Also, per the Masterplan, schools and hospitals are in the pipeline for this area. The permitting process is much easier here than in the city, even though the administration is often painful and inefficient. For the land purchase, I had to do an agreement with the seller, and the local leaders had to sign off. In case of a simple extension, they mostly don t care. But for new construction, the district level administration has to approve the drawings and provide a construction permit in order to start construction. The local level leaders don t bother much, but the district level does. Source: Interview conducted by WB study team in December

18 V. RECOMMENDATIONS The wide range of housing typologies within the informal sector will require a range of policy interventions in both housing (self-built and developer-built new affordable housing, rental housing, in-situ upgrading, and social or subsidized housing, as necessary) and finance (mortgages, construction finance for developers and individuals, housing microfinance) see Table 3 below. Table 3. Potential Housing and Finance interventions by income group (ILLUSTRATIVE) Income HH income/m RENTAL Housing OWNER-OCCUPIED Housing HOUSING FINANCE Tier Rent for housing (RF House price (RF) per month) 1 RF>1.5M RF375K-500K RF54M-70M (developer-built) Traditional mortgages; 2 RF600K-1.5M RF120K-300K RF21M-54M (developer built) Small ticket mortgages 3 RF250K-600K RF50K-120K RF9M-21M (self-built) Developer finance; Construction loans; HMF 4 RF100K-250K RF15K-RF50K In-situ upgrading (housing, HMF; subsidies infrastructure) 5 <RF100K RF10K-15K In-situ upgrading / social housing HMF; subsidies; social housing For Income tiers 4 and 5, the government should facilitate the delivery of housing through a combination of (i) on-site infrastructure upgrading (in areas that are 'upgradable', i.e. not located on hazard zones or environmentally sensitive areas), and (ii) social/ subsidized housing, particularly in case of resettled communities. In case of upgrading, the focus should be on infrastructure (roads, water, sanitation, electricity), and the delivery of housing microfinance (along with technical assistance on building construction), which can enable households to improve their homes over time. Social housing programs need to be designed with extreme caution so as to effectively target the poor (or the displaced, in case of resettlement). Experiences worldwide have demonstrated the failure of mass public housing schemes, where the government has played the role of a developer. The most critical role of the government in the housing sector, however, is about making the market work. The premise here is that with the formal private sector in Rwanda catering to the richest 2-5 percent of the population, and assuming that the government/ donor programs will target the lowest percent, there will still remain a big gap for those in the middle, which includes people who could purchase or build a home if market conditions i.e. access to finance and housing supply which determine affordability and pricing were appropriate. This is the missing middle which constitutes those who are too rich for government subsidy programs, and too poor to be of interest to private developers and builders. 18

19 The inefficiencies associated with this gap are extremely costly: not only do the subsidies to the poor get misaligned and captured by those that they were not intended for, but also the bulk of the population is forced to live in housing that is overpriced or less value-for-money ; inadequate in terms of structural safety, space availability, and basic infrastructure; and in locations that pose high disaster risk. This has social implications not just in terms of health and overall productivity, but also the overall sense of security and stability. Addressing issues that make the 'middle market' work will help reduce the strain on limited government resources in the first place, and at the same time, reduce the inefficiencies associated with mis-targeting and leakage of subsidies. This requires addressing barriers to housing supply as well as housing finance, and is the focus of this study's recommendations. Addressing Supply-side barriers Reducing input costs to increase affordability Housing affordability can be addressed either by increasing disposable income or by reducing the input costs associated with the house. Given the scope of this work, this study presumes that income levels are what they are; so, the focus is on the latter. The two variables that offer the most potential to lowering costs include: - Reducing the cost of construction through alternative materials and technologies There is a critical need to promote and formalize the use of locally available and locally manufactured materials, and build local capacity in the construction industry. Several alternate materials are being explored by various agencies and institutions in Kigali (see Box 6). More research needs to go into testing the feasibility of these materials, followed by building an adequate manufacturing base, and marketing / disseminating their inherent cost and environmental advantages. Box 6. Alternative materials and technologies Agro-stone has the potential to become a major part of the building products palette. It is made of crushed volcanic rock and agricultural waste. Both are widely available in Rwanda. Magnesium Chloride is used as a binder this component is not locally available, and needs to be imported. However, the binder is used in small quantities relative to volcanic rock and agro-waste. The labor skills and plant requirements for the material are modest. KIST has conducted various tests for compressive strengths. The material can be used as infill panels within a concrete frame building. Schnell, a German manufacturer is actively marketing a Styrofoam panel and has built two homes in Kigali to showcase their product. They are also committed to manufacturing the product locally. However, the merits of building homes of Styrofoam both from the cultural and the environmental perspective is debatable. Paper-mache panels use recycled paper as a raw material to make infill wall panels. The material could be limited by the lack of an adequate waste collection and recycling program for paper. However, like the Styrofoam panels, paper mache has limitations from the perpective of cultural acceptability (security, privacy). Apart from these, there are other methods being explored to reduce the usage of cement (which is the most expensive ingredient for construction) - among these, expanding concrete, and the use of rice husk as a binding material. Source: Interviews conducted by WB team in December 2011 with KIST staff. 19