Tenure Security , Land Titles and Access to Formal Finance in Upgraded Informal Settlements : The Case of Dar es Salaam , Tanzania

Pro-poor land policies have recently emerged in reconceptualising access to housing finance in low income settlements in developing countries. In Tanzania, this concept has been adopted to enable dwellers in urban informal settlement to enjoy perceived tenure security. Settlement upgrading has further increased perceived security. Interestingly, titling projects aimed at facilitating access to credit from formal financial institutions have been implemented in already upgraded settlements. Based on empirical investigations in two settlements in Dar es Salaam, the paper explores the take-up of titles and subsequent use to access credit. Findings show that take-up is slow and very few property owners have accessed credit. In contrast to arguments by several scholars, tenure security is not a simple matter of being legal or illegal or formal or informal. In the illustrated cases, loans were neither used to improve houses nor open businesses, but to expand on-going profitable businesses. Loans were repaid and no property has been taken away, presenting a win-win situation. The paper concludes that the use of titles to expand on-going profitable businesses provides new lens with which to look at alternative uses of residential licenses (RLs). Through this lens, property owners were able to use titles to access relatively large microenterprise development loans. It recommends further, that there is a need to promote the acquisition and use of titles. Financial institutions are challenged to design and implement context specific housing microfinance products that make ownership of RLs a necessary condition for accessing loans.


Introduction
The world is increasingly becoming urban.In 2000, the world population was 6.07 billion of which 2.86 (47.1 per cent) was urban (UN- HABITAT, 2005).The world population is estimated to have increased to 6.91 billion in 2010 of which 3.49 billion, lived in urban areas ( UN-HABITAT, 2009).Rapid urbanisation which is taking place in most of the developing countries is associated with, among other things, the fast growth of slums 1 .This partly explains why 862,569 thousand people or 32.7 per cent of the 2.63 million urbanites in the developing world cities today live in slums ( UN-HABITAT, 2012).If the current trend continues and no major policy reforms are made, the slum population is likely to increase to 1.4 billion by 2020 (UN-HABITAT, 2006) and 2.0 billion by 2030 (Palmer, Frisca, & Weherman, 2009).
A slum household is: A group of individuals living under the same roof in an urban area who lack one or more of the following five conditions: access to improved water, access to sanitation, durable housing, sufficient living area and secure tenure (UN- HABITAT, 2006: p. 19;emphasis added).
With regard to secure tenure UN-HABITAT provides two definitions that are relevant for the study: • the degree of confidence that land users will not be arbitrarily deprived of the rights they enjoy over land and the economic benefits that flow from it; and • the rights of individuals and groups to effective government protection against forced eviction (UN-HABITAT, 2008a: p. 5).
As argued by several scholars, tenure security is not a simple matter of being legal or illegal or formal or informal: it is a relative concept which involves legal aspects and peoples' perceptions (Gilbert, 2002).Providing secure tenure to land has positive effects on poverty reduction.According to the World Bank Insecure land tenure prevents large parts of population from realizing the economic and noneconomic benefits, such as greater investment incentives, transferability of land, improved credit market access, more sustainable management of resources and independence from discretionary interference by bureaucrats that are normally associated with secure property rights to land (World Bank, 2003: p. xxv).
Basically, there are three schools of thought on the importance of titling: the first advocates the issuance of legal titles to land as "the solution"; the second contends that titles are important but not the only option because there are other land-related interventions that can lead to perceived tenure and still produce equivalent effects.For instance, in a study aimed at understanding household's perceptions on security of tenure in three informal settlements in Dar es Salaam, Tanzania, Moyo (2006) found that it 1 In Tanzania, the word "slum" is rarely used.The commonly used terms are "informal" or "unplanned" settlements.In this paper, the three terms are used interchangeably.largely depended on duration of residence; building of permanent structures; provision of services by government departments and institutions; sale or purchase of land or house; holding of sale agreement documents; and paying of property tax.Proponents of the third school argue that despite the lack of legal titles, self-help home builders feel secure about their houses.This partly explains why Gilbert (2012) asks "Indeed, is this not obvious, given that most of the housing in African, Asian and Latin American cities, some of it extremely solid and well-built, was constructed informally?" (Gilbert, 2012: p. vii).
De Soto and the World Bank belong to the first school.According to De Soto (2000), property holders in informal settlements own vast assets in the form of self-built houses but the properties are held "extra-legally", that is, under insecure tenure rights.Lack of tenure rights constraints owners from using their properties to access loans from commercial banks.The proposed way forward is to convert the extra-legal arrangements into the legal system through the issuance of land titles.De Soto and the World Bank argue that the acquisition of land titles and hence loans, will not only enable property owners to improve their houses or start businesses but will also contribute positively to higher levels of tenure security, increased investment in their properties and hence contribute to poverty reduction.
In the same spirit, the World Bank notes: Without legal title, property is not secure… People do not invest in homes or communities where they are insecure.They cannot sell them or pass them to their children.Perhaps, worst of all, without property titles, homeowners cannot use their dwellings as collateral to borrow money that would help them escape poverty (World Bank, 2003;as cited in Gilbert, 2012: p. v).
Due to the influence of De Soto and the World Bank, several developing countries including the United Republic of Tanzania (URT) (or simply Tanzania), South Africa, Peru, Colombia, Brazil, Philippines, to mention a few, embarked on large titling programmes of their informal settlements.Interestingly, the first titling projects in Tanzania were implemented in informal settlements that had already been provided with basic infrastructure and social services in the 1970s and 1980s.It can therefore be argued that households in these settlements already enjoyed a high degree of perceived tenure security.
In Tanzania, the National Land Act (1999)  So far, few studies have examined the ability of land titling programmes in urban areas of low-and middle-income countries to reach the expected targets (Payne, Durand-Lasserve, & Rakodi, 2009).The situation is the same in Tanzania: little is known on the extent of take-up and use of land titles to access finance from formal financial institutions.For example, a recent study has revealed that: Although no statistics had been provided by the financial institutions as to how many property owners in the registered "informal" settlements had been successful in accessing loans from the top financial institutions, early indicators from an interview with one of the officers at the Ministry of Lands indicated that some property owners had succeeded in obtaining loans from CRDB Bank and Azania Bancorp (Parsa et al., 2011: p. 701).
Based on an empirical study carried out in two upgraded settlements in Dar es Salaam, Tanzania, the study aims to provide answers to the following questions: • What is the extent of take-up of the residential licenses and Certificates of Right of Occupancy?
• Which institutions are involved in issuing loans?
• Who is accessing loans, for what and under which conditions?
• What are the opportunities and challenges of using loan money?
• What should be done to overcome the challenges?
To better understand the use of titles in accessing loans, it was thought necessary to study three types of property owners: 1) those that have accessed loans from formal financial institutions; 2) those who have collected titles and used them for other purposes and 3) those that have not collected their titles.

Site Selection
The

Data Collection Methods
Both primary and secondary data collection methods were used.The authors reviewed published and unpublished international and national literature including books, journals, consultancy reports, periodicals and documents from the Ministry of Lands, Housing and Human Settlements Development (MLH&HSD), the Kinondoni Municipal Council (KMC) and the WAT-Human Settlements Trust (WAT-HST), a local NGO.After the review, the researchers carried out: • Interviews and discussions with 1) officials from MLH&HSD, KMC and WAT-HST 2) ward leaders and key informants from Manzese and Hanna Nassif and 3) officials from formal financial institutions involved in issuing loans.Interviews with officials in the first group helped to establish the status of take-up of RLs and CROs and identify formal financial institutions involved in issuing loans.Interviews held with members of the second group helped to identify property owners that used their documents to access credit.Interviews with officials from financial institutions helped to collect information on the use of loans and loan terms.
• In-depth interviews with a sample of 24 property owners in Manzese and 10 in Hanna Nassif who actually used their RLs and CROs respectively to access loans.The aim here was to triangulate information on the use of loans; loan terms; and explore opportunities and challenges.
• In-depth interviews with a sample of 10 property owners that have paid for, and collected their RLs but used them for purposes other than accessing credit from formal financial institutions.The interest was to investigate reasons why and/or identify other possible uses of the documents.
• Focus Group Discussions (FGDs) with 12 property owners who have not paid for their licenses and have not yet collected them.The FDGs were held in order to find out why after so much sensitization on the importance of RLs and CROs to access credit, property owners are still hesitant to collect them.
• Non-participant observation.During interviews, the authors were able to observe owners' properties as well as existing livelihood activities in and around the houses and within the settlements.It is important to mention that the process that led to the selection of 24 property holders in Manzese and 10 in Hanna Nassif for the in-depth interviews was more challenging and time consuming than anticipated.Initially, 40 property owners were randomly pre-selected.After the pre-selection, the author accompanied by informants and local leaders visited the 40 owners at their homes to 1) brief them on the purpose of the research 2) seek their participation and 3) fix appointment dates for in-depth interviews where there were no objections.
After the preliminary visits, 32 owners agreed to participate in the research and appointments were fixed.On the agreed dates, however, eight owners excused themselves for one reason or another.With regard to Hanna Nassif, 15 property owners were randomly selected but only 10 were available for the in-depth interviews.Thus, 54 property owners from the two settlements participated in the in-depth interviews and FGDs.

Urban Poverty
Poverty is a multidimensional concept.As such, it has many definitions.In this study, "income" poverty, "housing" poverty and "voicelessness and powerlessness" are discussed.
Income poverty is defined in terms of low income or consumption.According to the World Bank (2001), people whose per capita income is below USD 370 per year or less than 1 USD per day (at 1985 purchasing power parity) are considered poor and those with less than USD 275 per year are extremely poor.
UNCHS defines housing poverty as the individuals and households who lack safe, secure and healthy shelter with basic infrastructure such as piped water and adequate provision for sanitation, drainage and the removal of household wastes (UNCHS, 1996: p. 109).
The definition above has many similarities with the operational definition of slums provided in the Introduction.
At the global level the Millennium Development Goals (MDGs) provide a unique opportunity for reducing, among other things, income and housing poverty.For example, MDG 1 aims at halving the proportion of people whose income is less than USD 1 a day and who suffer from hunger by 2015.MDG 7 aims at halving the proportion of population without sustainable access to safe drinking water and basic sanitation by 2015; and improving significantly the lives of at least 100 million slum dwellers by 2020 (UN, 2011).
Voicelessness and powerlessness within political systems and bureaucratic structures is another definition of poverty (Mitlin & Satterthwaite, 2013).Through participatory settlement upgrading slum dwellers are provided space to air their voices and the process is known to empower them.According to Narayan, empowerment is the: expansion of assets and capabilities of poor people to participate in, negotiate with, influence, control, and hold accountable institutions that affect their lives' (Narayan, 2002: p. xviii).

Access to Land and Tenure Security
Access to land is the basis for housing development.The world is rapidly urbanizing and this has impacts on the demand for land.As land is a finite resource and cannot be made or moved to other places, governments and land markets are required to provide sufficient land in suitable locations to meet the growing demand.If the demand is not met, low-income households are likely to face major constraints in accessing suitable land for shelter development.And as discussed in the Introduction, secure tenure is important for improving the socio-economic conditions of slum households.

Access to Credit
The process of land acquisition and subsequent construction of a decent house in urban areas requires substantial amounts of money that very few people can pay from their own pockets.And this is one area where land titles become important for accessing credit.
Mortgage finance targets individuals who can take relatively large value loans of over USD 10,000 (UN-HABITAT, 2005) which they can repay by making regular monthly payments for long periods of time, ranging from 15 to 30 years.This condition best fits high-income earners including permanently employed staff.In addition, mortgage finance requires households to hold full legal titles to their properties which are held as collateral to the loans.The house is ready-made and adheres to all building standards and regulations.In the event of default in payments, the house is repossessed and sold.
Low/middle income households build their houses incrementally and the process accounts for 50 -90 per cent of residential development in the large cities (Ferguson & Smets, 2010).The house is built with little or no support from commercial banks because the households' economic characteristics and financing needs are incompatible with the operational procedures and lending requirements of the financial institutions (Sheuya, 2007).For example, the majority are self-employed, who according to banks, cannot guarantee regular monthly payments for long periods of time.
When credit is needed, low/middle income households depend on, among others, informal finance and housing microfinance (HMF).The former comprises of a variety of financing sources including savings, informal loans from friends and family, remittances from family members working abroad and the sale of household assets (Stein & Castillo, 2005).Other sources consist of windfalls, self-help fabrication of own building materials, sweat equity, barter arrangements and loans from money-lenders (Ferguson & Smets, 2010).
HMF evolved from the early 1980s when it was realized that loans offered for microenterprise development are diverted to home improvement.According to Mesarina and Stickney a housing microfinance loan: follows the basic principles of microfinance such as small loan amounts, short loan terms, and alternative forms of collateral and is used by the customer solely in a housing project, usually home improvement or progressive building (Mesarina & Stickney, 2007: p. 3).
Housing microfinance shares characteristics of both microenterprise and mortgage finance, but it differs in two ways: • The loans are smaller and short term but relative to microenterprise loans, are somewhat larger and of longer duration; and • Strategies to secure loans include co-signers who have equal or superior repayment capacity as the-would be borrowers; land title and buildings; mortgage/lien on assets; assignment of future income; joint liability and group guarantees and other financial assets (UN-HABITAT, 2008b).

Income Generation through the Livelihood Approach
For the majority of the poor in the developing world, land is the primary means of generating a livelihood and accumulating wealth.Poor households command a series of important assets: human, physical, financial, natural and social (Rakodi & Lloyd-Jones, 2002).Depending on the ability of the households to manage their asset portfolios and the local socio-economic and political contexts in which they live, poor people can move out of poverty (Moser, 1998).Moser further argues that home-ownership is by far the most important productive asset of the urban poor because renting and home-based enterprise are important income generating activities.
On their part, Kellett and Tipple argue that the links between the house and home-based enterprises are symbiotic dwellers are able to consolidate their dwellings through the income earned; many households would not have a dwelling without their home-based enterprises and many enterprises would not exist without the use of a building (Kellett & Tipple, 2000: p. 204).
Before presenting the study findings, a brief introduction on compensation and settlement upgrading in the specific context of Tanzania, is made.

Compensation
As mentioned earlier, the National Land Act 1999 is the principal legislation that governs land matters in the country.According to Section 3 of the Act, all land in Tanzania is public land vested in the President as trustee for, and on behalf of all citizens.Individuals and/or groups can acquire rights to occupy and use land on leaseholds for short terms of 5 years or long terms of 33, 66 or 99 years.
Compensation to affected property owners is spelt out in Section 3 of the National Land Act (1999): pay full, fair and prompt compensation to any person whose right of occupancy or recognized long-standing occupation or customary use of land is revoked or otherwise interfered with to their detriment by the State under this Act or is acquired under the Land Acquisition Act Cap.118 (URT, 1999: p. 14).

Settlement Upgrading and Issuance of Titles
Studies conducted in 1995 show that in Tanzania, 70 per cent of the urban population lives in informal settlements (URT, 2000).At the city level, the population of Dar es Salaam, the largest urban area in the country, has grown from 1.4 million in 1988 to about 4.0 million of which 80 per cent (3.2 million) live in unplanned settlements (UN-HABITAT, 2010a).
During the 1970's and 1980's and with financial support from the World Bank, the country implemented Sites and Services and Squatter Upgrading Projects whereby several unplanned settlements in Dar es Salaam and other major up-country towns were provided with 1) basic infrastructure such as roads, storm-water drainage channels and water supply and 2) social services including schools, health care centres and markets.One of the settlements upgraded during this period was Manzese.The upgrading package, however, did not include issuance of property titles.
Two participatory upgrading projects were implemented in Dar es Salaam city.The first took place from 1993-1999 and the second from 2005-2008.The project used community and labour based upgrading approaches to provide basic roads and storm-water drainage channels in Hanna Nassif, an informal settlement located 4 -5 kms from the City Centre.Similarly, no property titles were issued at that time.In the second project (under the Community Infrastructure Upgrading Project (CIUP), 167,000 residents living in other 16 unplanned settlements in Dar es Salaam were provided with basic infrastructure.It can be argued that these interventions have greatly contributed to perceived tenure security among the residents.
The issuance of Residential Licences began in 2005 in 43 wards3 (out of 72) existing at that time in the three Municipalities of Kinondoni, Ilala and Temeke.To obtain a Residential Licence, the homeowner has to pay a fixed fee of TShs4 5,600 which covers application, preparation, registration of the document and stamp duty costs.In addition, the owner has to pay annual land rent which depends on the use and size of the plot.For example, the required amount for a residential plot covering 407 square metres is TShs 3, 256, thus bringing the total amount to about TShs 9,000.
The residents of Hanna Nassif ward, however, demanded CROs of 99 years (instead of RLs) because they offer the highest degree of tenure rights and security.Community members expressed their willingness to meet the costs of planning and cadastral survey of their settlement and WAT-HST, a local NGO, responded positively to the request.When the survey was completed in 2007, 1423 plots were demarcated (WAT-HST, 2010).Altogether, the NGO spent a total of TShs 30 million and in order to recover costs, each property owner was required to pay TShs 30,000.The amount was equivalent to a month's room rent in the settlement.

Extent of Take-Up of RLs and CROs
The 2006 Progress report of the Unplanned Settlement Project of the Ministry of Lands, Housing and Human Settlement Development shows that at the end of the First Phase in Dar es Salaam, 217,407 residential licences were prepared (Table 1).
Data collected from the MLH & HSD in December 2012 show that a total of 96,630 property owners or 44.45 per cent had paid for and collected their licenses (Table 2).
Regarding Manzese ward, data collected from the Residential License Project office in the Municipality show that 5408 RLs were available as of December 2012, and 2920 (54 per cent) had been collected.With respect to Hanna Nassif, available records5 from WAT-HST show that as of 2010, 295 (20.7 per cent) CROs had been collected.Data extracted from the files at the Lands section of the Kinondoni Municipality in December 2012 show that an additional 230 CROs were collected in 2011 and 2012.Thus, in five years, the number of collected CROs was 525 (36.9 per cent).

Formal Financial Institutions Involved in Issuing Loans
Records from the MLH&HSD show that, as of December 2012, a total of 1948 property owners (2.02 per cent) in the City had used their RLs to access finance from different financial institutions (Table 3).
The majority, (687) live in Kinondoni Municipality.Data collected from the Residential License Project office in the Municipality show that 21 financial institutions were involved in issuing loans (Table 4).
The leading institutions were: Promotion of Rural Initiatives and Development Enterprise (PRIDE), Access Bank (ACB), National Microfinance Bank (NMB), Entrepreneurs Financial Centre (EFC) and Cooperative and Rural Development Bank (CRDB).Two are microfinance institutions and three are banks which also provide microfinance services (Table 5).

Introduction
Five aspects were studied: socio-economic characteristics of property owners who accessed loans; use of loans; loan sizes, repayment periods and interest rates; viewpoints from property owners who have collected titles and used them for other purposes; and of those who have not collected titles.Findings from Manzese ward are presented first followed by Hanna Nassif 6 .The study in Hanna Nassif did not cover property owners from the last two categories because CROs from fully planned and surveyed areas have been in use in Tanzania for many years.

Manzese Ward Socio-economic characteristics of property owners who accessed loans
The sample size comprised of 24 property owners of whom eight were women 16 men.All women used residential licenses as collateral to improve own businesses while three out of 16 men used their licenses to guarantee relatives and close friends (social capital) to access loans.It is important to make this distinction because as we will see later, the two groups encountered different types of opportunities and challenges.
With respect to education, 17 property owners had completed primary school and seven secondary school education.Their education levels fairly represent their employment status: 21 were self-employed and the remaining three were employed as junior cadre staff in privately owned offices in the City.
Use of the loans Loans were issued to property owners who wished to expand on-going profitable businesses only.As narrated by one owner Credit officers were explicit on this.To get a loan, you must have an on-going business and the business must be profitable.Once they are satisfied, they check your residential license and take a photo of your house when you are standing in front of it.Then, they start processing the loan (Interview with IP no. 3, March 2013).This information was relayed by all interviewed property owners that accessed loans.And although they took loans, they complained that during the sensitization meetings on the use of RLs they were told that the documents could be used to 1) start own businesses and 2) improve their housing.When using the loan, however, four owners diverted part of the loan money to other activities: two said that they used half the amount to improve their houses and the remaining two used it for different activities altogether.The stories of the four owners are briefly narrated hereunder.
The first person owns a bar.The house where he lives has a separate wing where four tenants live.Initially, the toilet in the tenants' wing was roofed with temporary building materials and when he received the loan, he roofed it with corrugated iron sheets.In addition, he built floor and walls tiles.Thereafter, he increased the monthly room rent from TShs 30,000 to TShs 40,000 and the collected rent was sufficient to meet monthly loan repayments.
The second person owns a hair dressing salon.Before the loan, clients used to wait for their turns in the front veranda of her house.Using part of the loan money, she extended the salon by enclosing the veranda and decorating it.She now has a spacious salon.The third person has a shop and is also involved in water selling from a deep well dug some five years ago in the courtyard of his house.He narrated that water vending is good business in the settlement because the existing public water supply system is unreliable and at times water flows late at night.The well has large water reserves but he lacked storage facilities.So, when he received the loan, he bought a 10,000 litre storage tank and installed an electrical water pumping machine.He did not tell credit officers that the money was used for the water project because he was not sure if they would approve it.On repayment he said So far, I have no problems with making monthly repayments.This is good business because money obtained from water vending is enough to pay the loan.
However, during the rainy season, I have to top it up from other sources including profit from the shop.I am sure that water vending will remain a good business till when City authorities rehabilitate the whole water supply system.(Interview with IP no.6, March 2013).
The fourth person, like the third, owns a shop and managed to convince credit officers that the loan was for expanding his business.When the loan was issued, however, he used half the amount to buy a motorcycle.He used the motorcycle as a means of public transport and employed a young man to operate it.The mode of transport (popularly known as "boda-boda") is used by many people because the City is currently facing congestion problems during peak hours.Any person coming to Dar es Salaam will agree that motorcycle owners are making good business.After four months, however, the motorcycle driver was involved in a fatal accident and the motorcycle badly damaged.It is out of use and the owner now faces regular monthly repayment problems.To meet his obligations, he borrows money from friends and relatives and repays them later.
Loan sizes, repayment periods and interest rates Ten property owners did not wish to disclose the exact amount of money borrowed for what they called "personal or confidential reasons" (See section on Methodology).
They were, however, willing to talk about their repayment periods and monthly instalments.The period for 23 property owners was 12 months and the monthly instalments ranged from TShs 100,000 to 600,000.It was only person, (IP no.19), who had a monthly instalment of TShs 960,000 and a repayment period of 18 months (Table 6).
On interest rates, most property owners said that the "calculations were complicated"7 .In addition, the data are hardly available in the institutions' branch offices or their websites.To obtain the data, one is required to contact the institutions' headquarters.
Viewpoints from property owners who have collected RLs and used them for other purposes These maintained that possessing a RL was one of their democratic rights because it can be used at a time when the need arises.Besides loans from formal financial institutions, they mentioned two other possibilities: their use in court cases where proof of ownership of immovable property is required, and loans from informal money lenders.In fact, Funes (2013) found that there were 16 such lenders in three sub-wards of Manzese providing loans for a variety of activities including housing improvement, business development and children school fees.
Viewpoints from property owners who have not collected RLs In the FGDs, two major reasons were mentioned 1) it was not a priority need at the moment and 2) owners were not threatened by eviction because the settlement was upgraded long time ago meaning that they already enjoy a high degree of perceived tenure security.They also said that should the government decide to evict them, they are entitled to receive full and fair compensation.

Hanna Nassif Ward
Socio-economic characteristics of sampled property owners The sample size comprised of 10 property owners: four women and six men.Seven had attained primary level education and three secondary.All of them were self-employed.Use of loan money Like in Manzese, loan money was used to expand on-going businesses but two property owners managed to divert substantial amounts of the money to housing improvement.By doing so, they now receive higher rents from improved housing.They also felt that the market value of their properties has increased.
Loan sizes, repayment periods and interest rates The interviewed persons accessed loans from the same formal financial institutions.The repayment period was one year for all of them.Unlike Manzese, monthly repayments and hence loan sizes for the majority of the sampled members were relatively high (Table 6).

Manzese Ward
Twenty one property owners have successfully accessed credit.Except the one who diverted part of the loan money to buy a motorcycle, the rest have expanded own businesses and are satisfied with the outcome of taking loans and encouraged others to do the same.To show their satisfaction they used expressions such as "a title is a useful shield for your life" and "a title is your saviour".
Three property owners have acted as credit guarantors to their close friends and relatives.Two were proud of the stock of social capital rendered to their friends and relatives.Through the "helping hand", they were able to expand own businesses.They narrated that the said relatives managed to pay the required monthly instalments in time and in so doing, established good repayment history and can now access loans without using their uncles' RLs.
Challenges Fears of losing the house One of the often quoted causes of property owners distancing themselves from taking loans from formal financial institutions is the fear of losing their houses (Sekijage, 2006).This did not happen among the 23 owners who accessed loans.Upon probing they said that loan amounts were relatively small, repayment periods long enough and monthly instalments affordable.Therefore, to them the fear was "perceived" rather than "real".As narrated by one I want to admit that soon after receiving the loan, the fear of losing my house started to creep in my mind.But this is precisely what made me repay the loan in time.It is therefore a good fear (Interview with IP no.14, April 2013).
The only person who faced the real fear of losing his house was IP no.19.He owns a 12 room-house built in a large plot of approximately 1,000 square meters.His family occupies four rooms; tenants live in six rooms and the remaining two are shops.Besides the house he owns a three-acre farm planted with mature mango and orange trees in Kiluvya, one of the peri-urban areas of Dar es Salaam.
During the interview he narrated that personally he had no intentions of using his RL to access a loan.However, a close friend and a successful business lady who sells fashionable ladies' and children's wear in two shops in the Central Business District requested him to use his RL to guarantee her a TShs 15 million loan.The lady normally travels to Dubai, (United Arab Emirates) and China for shopping.At that particular time she was in need of large sums of money to triple the volume of her business.She promised to repay the 18-month loan within 12 months and to give him a "handsome reward" in two instalments: the first upon receipt of the loan and the second, ten months later.
When all was set, he used his RL and when the money was made available, the lady honoured her promise.After a short period her shops were fully stocked and the lady diligently paid the required monthly instalments for eight months.Thereafter, the shops were closed for several months.As a result, IP no.19 had to repay the remaining balance.He is painfully doing it to avoid losing his house.Part of his story is contained in Box 1.
Diversion of loan money As explained earlier, four property owners diverted part of loan money to other uses.Those who used it for housing improvements and housing related economic activities have no regrets because they earn enough money to repay the loans.It's only the owner who bought a motorcycle who is facing regular repayment problems.To solve them, he borrows money from friends and relatives confirming the importance of social capital.

Hanna Nassif Ward
Like residential license owners in Manzese, CRO holders in Hanna Nassif highlighted the ability to use titles to access loans from formal financial institutions as a major opportunity.No one faced problems with monthly instalments and none encountered the "real fear" of losing his/her property.They said that such fears were constructive in the sense that the borrower had to be extra careful in using loan money.

Discussion, Conclusion and Recommendations
The study explored five issues: extent of take-up of land titles; financial institutions involved in issuing loans; loan recipients, use of loans and loan conditions; opportunities and challenges; and proposals for the way forward.
From the findings, five major issues emerge: • Slow take-up of land titles.At the City level, 45 per cent of the property owners have collected their RLs while in Manzese ward it was 54 per cent.In Hanna Nassif ward slightly less than 40 per cent have collected CROs; • Very few property owners have accessed credit from formal financial institutions.
Three of the five topmost institutions are commercial banks and two are microfinance institutions.
• Use of loans.Loans are meant to expand on-going profitable businesses and not to start businesses or housing improvement as prescribed by De Soto and the World Bank; • Existence of "perceived" rather than "real" fear of losing property; and • Diversion of loan money.Box 1. "Real" fear of losing a house.
At first I did not believe what had happened.But since my house was at stake, I had to do all I can to rescue it.So, I subdivided two acres of my three-acre farm at Kiluvya and obtained 24 plots.I sold them to people who were looking for vacant plots to build houses.I got enough money and repaid the remaining balances.I must admit that credit officers sympathized with what happened to me and extended the repayment period.All the same, I swear I will never take a loan or help someone else to get a loan.(Interview with IP no. 19, May 2013).

Slow Take-Up
The Second phase of the Residential License project has not yet taken off due to, among other things, the slow take-up of RLs.This might have been caused by the already existing high degrees of perceived tenure security brought by pro-poor land policies coupled with the implementation of settlement upgrading projects since the early 1970s.Titling has added more value to the perceived tenure security and in a way rendered the properties more legal.This partly explains why some participants in the FGDs felt that taking RLs was not a priority need since 1) they are not threatened by eviction and 2) should this happen; they are entitled to receive full and fair compensation.The reasons mentioned are precisely the defining elements of tenure security according to UN-HABITAT (2008a).

Access to Credit
Very few property owners have accessed credit from financial institutions.This finding is shared by re-known scholars in the field.For example, Payne, Durand Lasserve, and Rakodi found that "very few of the households in any of the case study settlements in Senegal or South Africa have taken out loans, including bank loans" (Payne, Durand Lasserve, & Rakodi, 2009: p. 455, emphasis added).In the same manner, Satterthwaite notes Land titling often fails to increase access to credit, and low-income households that obtain titles are often as reluctant to take out loans as banks are to lend to them (Satterthwaite, 2009: p. 305).And Gilbert's findings show that "property titles seem to have little effect in convincing private banks to lend to the poor" (Gilbert, 2012: p. viii).

Use of Loans
As a departure from other studies, the researchers went one step ahead and investigated the use of loan money among the very few property owners.Findings show that loans were neither used to start businesses nor improve housing but rather to expand ongoing profitable businesses.The finding departs from De Soto's assumptions.Put differently, De Soto's prescription, did not meet the underwriting requirements of the studied financial institutions.However, it is important to emphasize that RLs were a pre-condition for accessing microenterprise loans exceeding TShs 2.0 million.For lower amounts, group collateral, similar to the Grameen Bank model, was accepted.
By having on-going and profitable businesses, successful property holders had the required experience; dependable sources of income; and could guarantee regular monthly repayments.This helps to minimize credit risks.But even for on-going businesses and despite holding titles, loan amounts were not very high. 8 In short, focusing on the broad picture, the findings support what other researchers have observed.The details of the findings from property owners who used titles to access loans, however, provide a rich mix of different lessons of experience.In view of the above, a study on the effects of titles on accessing credit is likely to be more balanced if it explores both the 1) broad picture and 2) details of the few who accessed credit.The authors are of the opinion that details provide important hints for the way forward.

"Perceived" Fear of Losing the House
One of the reasons why property owners hesitate to take credit from banks is the fear of losing their properties.Study findings show that the fear was "perceived" rather than "real".It is indeed this fear that helped to ensure regular loan repayments.None lost his/her property and the one who experienced real fear repaid the loan by selling land in the peri-urban area, another important household livelihood asset.This highlights the importance of the aspect of managing the portfolio of household assets (human, social, physical, financial and natural) of the livelihood approach.From the above, one can argue that the type of microenterprise development loans issued, do not seem to threaten loss of property.

Diversion of Loan Money
A small number of property owners have diverted loan money indicating that the study sample is not atypical.Studies from the urban Global South have shown that similar findings have contributed to the establishment of housing microfinance products (UN-HABITAT, 2011).

Conclusion and Recommendations
To conclude, the use of titles to expand on-going profitable businesses provides new lens with which to look at alternative uses of RLs.Through the lens, property owners were able to use titles to access relatively large microenterprise development loans.The majority felt that their businesses had expanded.It was only two people9 who faced problems in making loan repayments.And to solve these, they utilized existing social and natural capital.Financial institutions were satisfied with the approach, making the whole process a win-win situation.Similarly, the few owners who diverted part of loan money to housing improvement were able to repay loans from room rents.As argued by Moser (1998), a house is a major livelihood asset.
Thus, there is a need to promote the acquisition and use of titles.Property holders 8 In the study, it was only IP no.19 who took the highest amount totalling TShs 15 million.Furthermore, Access Bank was the only institution which could offer up to TShs 25.0 million (see Table 5).
who have not yet paid for their RLs should be sensitized to do so: they can use them when the need arises.Those who have not used them should be encouraged to judiciously do so.Given more time and through "learning by doing", the number of property owners using the documents is likely to increase and a rich mix of lessons of experience generated.Property owners that diverted part of loan money to housing improvement and housing related economic activities had no regrets because they earned enough money to repay their loans.This suggests that there is an unmet demand for housing microfinance.Financial institutions in Tanzania are challenged to design and implement context specific housing microfinance products that consider ownership of RLs a necessary condition for accessing loans.
At this point, it is important to examine the relevance of the study findings with respect to poverty reduction in urban settings and hence MDG 1.The findings have shown that titles per se, are not a sufficient condition for reducing income poverty.In the study, formal financial institutions were able to issue loans to property holders who had titles and at the same time owned profitable businesses.It is therefore a combination of the two that can facilitate to reduce income poverty.And as mentioned earlier, the finding defeats De Soto's assumption that titles alone can be used to start businesses and therefore reduce poverty.
Similarly, financial institutions did not issue loans for home improvement as prescribed by De Soto.The few owners who used part of loan money for home improvement did so by violating loan conditions and it is interesting to note that credit officers kept a blind eye.There could be three reasons:

•
It is likely that credit officers knew well the livelihood strategies of their clients including their repayment records.They carefully weighed loan amounts Vs. the value of their asset portfolio.When the asset portfolio outweighed the amount of loan taken, they kept a blind eye; • Probably, credit officers knew that income generated from renting home improvements in the settlements was more than what is required to repay loans amounts; and • Most likely they also knew that property owners value their houses so much that they would never risk losing them.From the foregoing, it will be interesting to explore some of these informal "risk management systems" that facilitate the functioning of formal financial institutions.Participatory settlement upgrading can impact positively on "voicelessness and powerlessness" because it can lead to empowerment (as defined by Narayan, 2002).
As mentioned in the Methodology section, Hanna Nassif is the only upgraded settlement in Dar es Salaam where upon demand, CROs instead of RLs, were issued.The demand could be a result of capacity building initiatives built during the implementation of the Community Managed Settlement Upgrading project in the settlement from 1994 to 2000.
During the six years, the residents and their CBO, the Hanna Nassif Community Development Association (HNCDA) participated fully in a joint upgrading initiative in their settlement with the Dar es Salaam City Council, UN Agencies (ILO, UN-HA-BITAT, UN Volunteers) and Ardhi University.These institutions provided the required technical assistance while UNDP and Ford Foundation made available the financial support (UN-HABITAT, 2010b).
The project was innovative in many aspects including, among other things, partnership between the CBO, City Council and Ardhi University; direct participation and decision making by community members; interagency collaboration among UN Agencies; and capacity building among the key stakeholders (UN-HABITAT, 2010b).
It is likely that the collaborative partnership among the key stakeholders, the duration of the project, and the capacity created during the project (and the resulting empowerment) is what made the CBO and its residents to demand CROs.
which operationalized the Land Policy of 1995 is the principle legislation which enables property owners in informal settlements to own titles.These are of two types: intermediate and full.The former are in the form of Residential Licences (RLs) for settlements that are not located on hazardous lands or lands reserved for public utilities.When issued for the first time in 2005, their duration was two years renewable but since 2009, the period has been revised upwards to five years.Full titles are in the form of Certificates of Right of Occupancy (CRO) and their duration can be 33, 66 or 99 years.
study was carried out in Dar es Salaam city.Administratively, Dar es Salaam City Council is composed of three 2 municipalities-Kinondoni, Temeke and Ilala.The municipalities are divided into 90 wards which are further subdivided into a number of sub-wards.According to the 2012 Population Census, the city region of Dar es Salaam has a population of 4,364, 541.Kinondoni has the highest population of 1,775,049 followed by Temeke and Ilala with 1,368,881 and 1,220,611 respectively (URT, 2013).The empirical enquiry was carried out in two wards, Manzese and Hanna Nassif of Kinondoni Municipality.With a population of 70,507, Manzese consists of six subwards and is the largest upgraded informal settlement in terms of surface area.In addition, the residential licenses project in Dar es Salaam was first implemented in Manzese.Hanna Nassif (with two sub-wards) and a population of 37,115 (URT, 2013) is so far, the only upgraded settlement in Dar es Salaam where, upon demand by its residents, full titles (instead of residential licenses) were issued.

Table 2 .
Property owners who had collected RLs in Dar es Salaam (December, 2012).

Table 3 .
Property owners who had used RLs to access credit in Dar es Salaam(December, 2012).

Table 5 .
Basic information of the leading formal financial institutions.
Source: Official websites and interviews with Branch office staff (April-August 2013).

Table 6 .
Monthly instalments and repayment periods in Manzese and Hanna Nassif wards.