The Provider Paradigm, Housing and Housing Finance: An NGO Provider Approach and Its Challenges

Housing projects of a few hundred to a few thousand units are impressive and can capture the imagination as successful housing interventions. Indeed, housing conditions are usually materially improved for the participating households. But what is the relative significance of such interventions in the overall housing environment? A challenge in the field of housing is how to effectively serve statistically significant numbers of households with low incomes. In Housing Without Houses: Participation, flexibility and enablement, Nabeel Hamdi defined two paradigms of housing which he described as being in conflict: The Provider and Supporter Paradigms (see posting from 8th August, 2009). The provider solution to this housing challenge is simply to build houses. Hamdi gives a sampling of some key features of the provider paradigm as follows:

… Providers go on to argue that if production is to be increased, then housing types (tightly defined packages of space and material) should be rationalized to represent a careful balance of technical feasibility, building regulations, and planning controls – the very things that supporters argue have raised costs beyond the reach of those with the lowest income, especially in developing countries.

Providers argue that to mass produce effectively is to mechanize, to mechanize effectively is to standardize, to standardize effectively enables better control of quality, quantity and cost. For providers, to build fast is to build instantly, to solve problems of housing is to build a large number of houses…[1]

Hamdi was largely applying the paradigms to governmental approaches to housing; the provider paradigm being epitomized in public housing projects. The overall concepts behind the paradigms, however, equally apply to many non-governmental organizations (NGOs) and other institutions engaged in housing, particularly as a social service to the poor. Housing NGOs or cooperatives that hold a provider paradigm tend to build houses, have standard house designs and often prescribe required types of materials and technologies (which may be described as “appropriate technologies”). They  frequently define the  housing process for those who would participate in the program as essentially a take it or leave it option. The organization's housing process becomes the clients' housing process. For the purpose of this posting, I will call anything that falls relatively close to these general housing program features the “NGO provider approach,” although it is not to say that all housing NGOs or cooperatives use or are restricted to it. Key stated objectives of the approach typically include ensuring the quality of the house, keeping costs low and affordable and serving a large number of households with the house as a product. The overall results of the NGO provider approach are not always as intended, especially when it enters the realm of housing finance by being building houses as a credit facility.

I was an adherent of the provider paradigm using the NGO provider approach for many years across various countries in Sub-Saharan Africa. Our key metric for performance was the number of houses built. It might have been stated with different euphemisms, but when asked what we did the answer would essentially be: “We build houses.” Although slightly varied for local situations, house designs were roughly the same across the continent. The houses we built were to standard designs, (commonly 24 – 48 square metres by the early 1990s), with two or three bedrooms, a sitting room and sometimes a corridor, storage space and/or veranda. Each house was complete with an outdoor latrine; usually a ventilated improved pit latrine. At various times appropriate technologies were being promoted, such as roofing tiles or stabilized soil blocks, and these would then become mandatory for the clients for as long as the organization was committed to using them. The quantities and types of materials used in construction were specified and the organization did the procurement of materials and services for the client, who did not have much say in the matter.

In most cases the organization provided the skilled labor and/or supervision. Clients (sometimes called beneficiaries, homeowners, participants, families, home partners or other terms, but in this posting I will primarily call them clients based on the feature of a loan component) were told exactly what materials they must provide on their own (usually locally available materials such as stones, sand and bricks) as well as what tasks they must perform. The final costs of services provided by the organization were calculated when the house was complete and converted into a loan to be repaid by the client into a revolving fund to build more houses. To make the house affordable for households with low incomes, the loan was made for periods of 7 to 15 years and the beneficiary was charged primarily for direct cost with most of the operational costs carried by the organization as a subsidy. Specific details vary, but this general approach is not uncommon for housing NGOs or housing cooperatives.

I have been a designer and implementer of programs that have built thousands of low-cost houses across the continent and seen many similar programs run by other institutions. The transition from a traditional, substandard house to a more modern and durable house is powerful and dramatic and can bring many positive changes to households. From very early in my work with the NGO provider approach to low-cost housing in Africa, however, I had a few observations that have remained constant over the years:

1. Quality: The quality of structures was not always what we had expected or intended. One theory behind controlling the process was to ensure quality and improve over the local quality standards that existed prior to the program starting. However, to keep the work affordable to the organization and the clients who were to pay for the services, local labour was inevitably used. Because we used the same builders who were building locally anyway, we usually had roughly the same general level of quality when applied across all of the houses built. There were many well-built houses that were notably better than the traditional and non-durable houses we sought to replace. The overall quality, however, was not significantly greater over houses of similar design that were built by households independently and outside of the program. Construction quality was a consistent problematic area mentioned in program evaluations.

2. Dissatisfaction: An early shock as an idealistic field worker was to come across a significant number of clients who were very vocal in their dissatisfaction with the houses they received through the program. Recurrent complaints included the house being too small, the absence of plastering, dissatisfaction with the types of materials used (e.g. roofing tiles), construction quality complaints (the ubiquitous crack in the wall or floor) and restrictions on adding on to the house before it was paid off. We responded with stock answers to explain why the choices made were in the clients’ best interest and what the policies were, but this never seemed to be wholly satisfactory to the clients themselves. I heard the same discussions recurring almost verbatim across multiple countries, programs and projects and reports of the same from other colleagues over a period of more then 15 years as if it was all being read from the same script. In many ways I suppose it was.

3. Poor Repayment: Repayment was poor almost from the onset and the revolving fund simply did not revolve. Where repayment started well, it could rarely be maintained for longer than two or three years. (Note: This is a particularly bad problem with loan periods of up to ten or fifteen years.) New projects were started on the assumption that some previous implementers did something wrong at the foundation of the poorly performing projects that had resulted in the poor repayment. An often-heard comment was that “The model is good, but the implementation was poor.” Extra effort was usually placed in training and sensitization or strengthening systems, but the results were almost always the same, even with different implementers in different locations over a span of many years.

After years of believing I could fix the system and vigorously trying to do so, I began to realize that there were inherent problems with the model itself. The NGO provider approach may work fine as a hand-out or heavily subsidized housing program that serves relatively small numbers of people compared to the total population that is engaged in a housing process. (The delivery system is still cumbersome and expensive even without a credit component, but that is for another posting.) The biggest problem comes with scaling up and maintaining that scale beyond the duration of a donor’s interest. I think the main challenges could roughly be translated into three major issues that, when combined, make the NGO provider approach to housing and housing finance (the house as a loan product) an almost insurmountable challenge:

1. Charity Messages: NGOs that approach housing from the perspective of a social service often see their role as bringing help and hope to the poor and needy. This is how programs are marketed to donors and it impacts who joins the organization and why, and the institutional cultures and methodologies that translate into the client interactions. Language and symbolic events or rituals often place emphasis on poverty, need and the perceived inabilities of the low income clients, with giving by the affluent as the solution. When messages implying a mission of charity are communicated to clients, whether explicitly or implicitly, it immediately creates a very clear connotation of a “soft” program in the context of Sub-Saharan Africa. Clients will then interact with the program based on this and often will not prioritize payment. Because the charity impulse is woven into the fabric of the organization and its staff, when clients don’t pay (based on their assumption that the institution will not really enforce the terms), the perception of the soft NGO is confirmed and reinforced if and when the NGO fails to take action other than exhortations or repeated threats over a prolonged period. Loan portfolio health quickly declines, usually on a permanent basis.

Charity and credit do not mix well. When they do mix, it will be to the detriment of the credit. I have personally seen many clients years behind in house payments to a housing NGO, while diligently servicing other loans. How we inherently view the people who we serve determines how we interact with them. A program designed around the assumption that people are poor, hopeless and incapable of making good, logical decisions or managing their key life processes will look very different from one that is built around an opposite set of assumptions. Clients can easily read the difference and interact with the institution accordingly to their best advantage. If you have a very low income and can see an opportunity to free some cash flow by not paying a housing loan, not paying might look like a wise strategy in order to service other household needs with the cash that would have otherwise gone to the loan payment. If the lender will let you do it with minimal or no consequences, why not? (Despite regular reminders that a revolving fund exists to help others, the provider’s altruism may not always translate into every low income household’s decision making process in managing their precious and sometimes precarious finances.)

2. Standardized House Designs: The provider paradigm tends to build standardized house designs for efficiency and other reasons. The prescription of the housing package by the institution is a natural choice for an NGO or other institution that has a significant charity influence. Many people and institutions implicitly believe that the poor cannot build houses on their own (although they would rarely state it or may not even have a full self awareness of this underlying assumption). If the poor are doing it by themselves, it must be wrong and they will make bad choices (by the providers’ criteria), which necessitates the provider dictating the “right” way for the client to build and overseeing the process. The emphasis is on the house and ensuring that it meets the provider's criteria, which are usually based on the material standards of the structure.

As I mentioned, it is very powerful to see a family moving from a very poor house by material standards and a better quality house built by a housing institution. All inclusive language and development-oriented language aside, however, when the housing delivery mechanism is in actual and practical operations dictated by the provider, it is disempowering at the very least and could become a mismatch with the client’s livelihood framework at worst. It is not unheard of that clients who benefitted from materially good houses with land security and soft loan terms sell the houses and move back to where they were before the project because of just such a mismatch. This is incomprehensible to the provider who delivered the quality house, but may make perfect sense within the logic of the household in a given set of livelihood circumstances.

Another factor when credit is attached to the standard housing package designs of the NGO provider approach, is Turner’s 3rd Law of Housing (see posting from 23rd July 2009). If the housing process, and particularly the specifics of the house itself, were imposed upon or reluctantly accepted by the client, any problem with the house, construction quality or other, will be felt deeply by the client and can result in resentment towards the lender. The potential impact on loan portfolio performance, particularly if the lending organization is perceived to be soft on recovery, is significant.

3. Weak Housing Finance Products: The charity impulse and provider paradigm of building specific housing designs create some challenges to implementing organizations when it comes to the associated housing finance component of the house-as-the-product. Firstly, stakeholders within the institution may not believe that the people with low incomes should pay back the full cost of services provided. This can lead to internal conflict, but also leaves the door open for a wide acceptance of heavy subsidy, which ultimately hampers the institution’s ability to serve significantly large numbers of households. Sustainability is a key driver of outreach, but true self sustainability may be a hard-sell within some institutions. This may result in a pricing structure that does not recover the costs of delivery and cannot sustain a program of lasting signficance and impact.

Secondly, a possible tendency towards subsidy is aggravated by the challenge of bridging the gap between the actual cost of delivering the provider’s house and affordability to the desired target group. The results are either heavily subsidizing the houses because they are not affordable to the people the institutions wishes to serve, not subsidizing and serving wealthier clients, trying to mimic mortgage lending by using long loan terms to make payments more affordable, or a mixture of all of these. None of these work well for sustainably providing housing services to large numbers households with low incomes in Sub-Saharan Africa. The best option would appear to be longer loan terms, but experience has shown that most low income households do not have the long-term financial horizon or stability to use a specific financial instrument over a sustained period of many years. Sustained repayment becomes a challenge quickly, with somewhere around two to three years perhaps being on the outer margins for many of the poor.

Yet another challenge arises from controlling the construction process and pricing product. (And I am I only referring to the direct construction cost and not even the broader issue of pricing for sustainability.) The cost of materials constantly fluctuates (usually upward), yet the bill of quantities remains constant. This means an ever increasing loan amount in cash terms for the same physical product. If the cost is not calculated until the house is complete in inflationary environments, it is impossible to tell the client how much the loan amount will be before the house is done . This might not be comforting to a borrower who must commit to an uncertain loan amount. "How much will my house cost?" "We will tell you when it is finished."

At the same time, there is a related issue when an approximate house cost is quoted by a staff member prior to construction and the final cost ends up being significantly more. It is a common problem. When a program is introduced to a community or an inquiry is made, people always want to know what the cost of the house will be. The cost they are told sticks in their minds a lot longer than it sticks to the actual cost of construction. Avoid quoting the cost and it is not transparent and can create mistrust. Quote a cost but then change it when construction costs rise and the client feels cheated. Quote the cost and don’t change it when construction costs rise and you either need to have included a hefty margin in the quoted cost or will start facing serious financial losses quickly. It is an inherent challenge of the NGO provider approach when it is rolled out to a community in an unstable economy. When my neighbor receives a house costing 500,000 (in whatever currency) and a few months later I get the exact same house but it costs 750,000, it is hard not to feel cheated, which impacts the relationship with the lender.

Construction quality problems and dissatisfaction with the house as a product, although widely acknowledged in the field, are usually not reportable issues and remain field anecdotes or buried in evaluation reports. Institutions can continue with their assumption that because they are going good, they must also be doing well. Repayment rates, however, are often measured and reported even though they may be poor. I have seen periodic revivals to attempt to improve repayment rates and entire “revitalization” initiatives to increase dismal portfolio performance, but they have almost never been successful in the long term without changing the underlying model and approach.

When NGOs started using the language of sustainability, recovery sometimes became more rigorous. As standards increased and words began to be backed with strong and effective actions to enforce recovery, I have noticed an interesting trend in the programs which I have managed: Demand for new houses dropped dramatically and quickly. What we had experienced was a false demand, rather than an effective demand.

The standard low cost house might not be exactly what a household wants at a given time, but it might be all that is readily available. As long as it is perceived to be on very soft terms, it is advantageous to take the opportunity. It is as if people are thinking: “It is not really the house I want or how I would apply that much money to my housing situation, but what do I have to lose?” Once recovery becomes more rigorous and households come to understand that they actually might have to pay for the provider’s house or face negative consequences, it becomes less attractive and there is less demand for the same product in the area. The house as a product might start to represent a potential risk to their household livelihood framework or it simply no longer looks like such a great opportunity.

Households make informed choices about whether or not to participate in a program based on the perceived opportunities, risks, advantages and disadvantages to their livelihoods. If a household is going to take a loan for housing and really pay it back, control over the application of the loan to the household's own housing needs and desires (as experienced by themselves) is more likely to generate an effective demand than if the same amount of money were used for a design that satisfies the provider’s needs in terms of the type of house it wishes to provide. An underlying challenge for the institution is whether or not it can accept the way a household may choose to apply a housing loan to their situation or whether it even believes that the household is capable of managing the process and making the housing and related livelihood choices in their own best interest.

HMF Hypothesis 5: Standard low-cost housing units provided on credit by organizations that are perceived as charity in Sub Saharan Africa will experience an inverse relationship between demand for the house and rigor in recovery.

My assessment of what I am calling the NGO provider approach to housing is presented as a broad generalization. It is based on repeated experiences and observation over the years, but there are certainly going to be many instances where the generalization does not fit. On a whole, however, I am so far willing to stand by my overall belief that the general NGO provider approach described, as visually powerful and compelling as it may be, is not a feasible solution for serving the masses. To change my position, I would have to see some strong, objectively verifiable evidence of the NGO provider approach successfully implemented.

My criteria for success of a program using the NGO provider approach would be having reached at least OSS (operational self-sustainability), thousands of active clients (in one service area, rather than a global total), continued demand with constant and consistent service delivery, and an accurate PAR (portfolio at risk) of less than 5%. If you are aware of such a program Sub-Saharan Africa using what I have called the NGO provider approach , please write in and let me know about it under the comments section of this post. I am ready to be proven wrong and would be willing to promote and discuss the program and its methodologies in this blog and in my own work if it can meet the above criteria. In the mean time, I will continue to promote housing microfinance as what I currently believe to be a more viable practical approach to housing for the low income population in Africa. (All of my positions are, however, subject to change pending further information and learning, of course.)

20th ANNIVERSARY SPECIAL

This is something like a special anniversary posting for me. It was 20 years ago today that I touched down in Kinshasa, Zaire (now the Democratic Republic of Congo) on my way to start work as a community development worker in a rural housing program in the Equateur Region.

That is me in late 1989 near Gemena, Zaire with my neighbors, Mozukala Dogia and family. I was young and energetic, but had no idea of how much I didn’t know or understand. I am now less young, and perhaps a bit less energetic, but at least I have a more realistic impression of how much there is still to learn 20 years later.

[1]Hamdi, N. (1995). Housing Without Houses: Participation, Flexibility and Enablement. London: Intermediate Technology Publication, 29-30.


(all photos and graphics by the author)