Three lessons for policy makers building housing finance systems

Rob McGaffin attended the Wharton Business School International Housing Finance Programme (IHFP) in Philadelphia from the 4 – 15 June 2012.  A similar course that is more tailored to the African context is to be run in Cape Town from the 1 – 6 October 2012.  The course is called the Housing Finance Programme for Sub-Saharan Africa and is being jointly presented by the Wharton Business School, the Centre for Affordable Housing Finance and the University for Cape Town.  More information about the programme can be found at this link.  In this blog, Rob sets out three of the key lessons offered by the course.


Policy makers considering the development of housing finance systems  should consider three especially important lessons:

  1. Housing finance is but one of many key elements that need to be in place for successful housing delivery to happen.
  2. Finance is required at each stage of the housing delivery chain – lack of finance in any one area will undermine the entire chain.
  3. Mortgage-backed finance systems that dominate housing finance thinking fail to address the breadth of housing finance needs.


Lesson 1: Housing finance is but one of many key elements that need to be in place for successful housing delivery to occur. 

To begin with, the entire delivery chain needs to in place or else the financial interventions will have a limited impact and or negative unintended consequences will result.

Delivery Chain:

Land – Development Rights – Infrastructure – Development – Construction – End user Finance – Public Services – Private Services – Management

However a number of key additions, including public/private services and management, need to be added to the chain.  Although the term “housing delivery” is used, most people actually have the objective of delivering viable neighbourhoods made up of both residential and non-residential uses.  Similarly, it is recognised that a house is in fact a multi-dimensional asset containing an array of values.  However, a significant amount of a houses value is created by the context in which it is situated.  Therefore, if one is to deliver houses of value to people, one has to in fact deliver the entire package of viable neighbourhoods.  For this reason, housing finance should not be restricted to only the residential structure itself.

Furthermore, due to the fact that property is characterised as being long-lasting and fixed in space means that it is susceptible to externalities and depreciation.  Numerous life-cycle costing studies have shown that often the operating costs over a life of a property are significantly higher than the initial capital costs to construct it.  Therefore, to protect the value of a residential unit (and the neighbourhood); it is imperative that the unit and area be properly managed.  As a result, finance pertaining to on-going management costs should be part of any housing finance programme and strategy.


Lesson 2: Finance is required at each stage of the housing delivery chain – lack of finance in any one area will undermine the entire chain.

The second lesson follows on from the first: finance is required at each stage of the delivery chain and conceptually therefore housing finance should cover all these stages on the chain.  Furthermore, different types of finance will be applicable to these different stages and in different contexts where the underlying conditions may differ.  For each stage, the following forms of finance will be applicable to varying degrees:

Forms of Finance:

Secured debt finance – Unsecured debt finance – Savings – Direct income – Equity – Grant finance


Lesson 3: Mortgage-backed finance systems that dominate housing finance thinking fail to address the breadth of housing finance needs.

The third lesson is that the mortgage-backed finance systems that dominate most of the thinking around housing finance are flawed in the following ways:

  • In many African countries and cities, a number of the underlying conditions necessary for mortgage finance (legally registered title; consistent, regular, long-term income stream; ability to repossess; etc.) do not exist.
  • Due to the mis-match between the duration of the source of funding and the lending that occurs, derivatives are often used to source funding from the capital markets.  This in itself is not a problem and such markets have an important role to play in enabling such financial systems to operate.  However, where such derivative trading is used as a source to generate profits, the entire systems is threatened.  In order to make profits through derivatives one has to in essence bet against the market, which by definition will only be successful in a limited number of cases.  Doing so is therefore very risky and considering the high levels of gearing of banks, the consequences can be extreme as seen in the sub-prime crisis.
  • In many cases, the easing of underwriting to facilitate housing ownership and address the affordability constraints of households is highly problematic, as it does not address the source of the problem.  The Wall Street Journal (4 June 2012) published an article stating that the real incomes of US households had not increased since 1990.  This, together with the fact that real house prices had increased over that period, resulted in the decline of housing affordability.  This is not a problem that can be overcome by extending greater finance into this market as the affordability issue is not addressed and repayment problems over time will occur.  Over-gearing poorer households is a recipe for disaster.
  • Similarly, the use of standard debt to household income ratios such as 30% are equally problematic as they fail to recognise the different spending patterns and demands on households across the income bands.  Research in South Africa suggests that such ratios should be closer to 15 – 20% rather than 25 – 30%.  In short, the levels of debt finance extended should be limited to what households can sustainably afford.
  • Loan to value ratios fall into a similar trap.  This is because the value of the house is determined at the date of the loan and yet the value of a house is likely to fluctuate over the term of the loan depending inter alia on the nature of the housing cycle.  To make matters worse, the house is by definition over-valued at the top of the cycle and under-valued at the bottom.  This is because prices will revert to the mean over time, which means that by definition, house prices at the top of a cycle must come down over time.

The problem is further exacerbated because lending criteria are generally less stringent over boom periods and higher loan to values are permitted.  Therefore, not only are the houses over-valued but also higher loans are given out against the “over-priced” assets, putting both the household and bank at risk.  Secured lending can be problematic at the lower end of the market because the asset value of the houses is low, liquidity in the market is low, household incomes are low and very importantly, the nature of the household income is uncertain.  As a result, it is often very rational for a household not to take up secured lending opportunities because their precarious economic position results in their houses being vulnerable to repossession.  Shorter term, unsecured lending (micro-finance) is in many ways better suited to this market.  The downfall of this finance is that it is very expensive.  The question remains whether a hybrid solution can be found?  Could a product be designed where a small (R5000 – R50 000) secured loan is made over a shorter period (e.g. 3 – 5 years) on a rolling basis?  If so, the key challenge to be overcome is the cost associated with small loans.

It is clear therefore that trying to address the affordability issue through housing finance alone will not solve the problem and may in fact threaten the financial and other related markets in the process.  Furthermore, the housing finance needs to relate and be structured around the housing delivery process, with the finance being tailored to each stage of the process and the different contexts in which the delivery takes place.


Leave a Reply

Your email address will not be published. Required fields are marked *

6 responses to “Three lessons for policy makers building housing finance systems

  1. I am interested in Housing Finance. African people have irregular incomes and mortgages should be tailor made to reflect that obvious fact. I note Jimmy Carter has been involved with Habitat providing interest free mortgaes, providing 2 million homes to date. I applaud this effort. I believe interest free mortgages should be routinely supplied, but with quid pro quo enforced interest free savings – which is drawn upon in retirement. I believe the house should itself be a unit of currency (It has constant utility value). Annual repayments rise with inflation and fall with deflation, because one is repaying say 20% per annum of the value of the house – which is a constant fraction. The trajectory of each mortgage would be unique, reflecting personal circumstances of unemployment or sickness. No income means no repayment (and no penalties)
    It implies a new discontinuous cashflow macroeconomics pertaining to the household. All money would be electronic. It would wipe out the blackmarket, and money cannot be hidden under a bed to avoid repayment. With moneys pooled in such a manner, there would be a high velocity of circulation of money, thus excessive money supply growth with all the inflation asscociated with it would be histroy. Banks would bid to manage the housing fund and be rewarded with a management fee if successful. They would be small enough to fail, as mortgages would not be on the banks’ balance sheet.

  2. Martin,

    I agree with you, there is definitely a disconnect between irregular incomes and the requirements of mortgage finance.

    I am curious though how you would attract funding to offer interest free mortgages?

    The idea of payments adjusting with the house value is similar in concept to rental turn-over clauses in retail centres. The challenge I guess, is how one would regularly track the changing house values over time.

    1. Hi Rob,

      I did not realise my comments were posted on the website, hence my tardiness in replying.

      I am thinking that the pool of money to finance interest free mortgages coupled with enforced interest free savings is initially financed by a government through quantitive easing. Obviously, the actuarial mathematics to do this is not in place now. However, if the actuarial mathematics can be found, the new method of finance promises to be stable and self sustaining. It presupposes that all houses in a neighbourhood are very similar – if not identical – and all have an identical utility value. Average communal household incomes is the basis to assign a current price to the houses, and mortgage holders would pay a current price (which varies from year to year) to pay off a fraction of a % fraction of the house. mortgage. One starts off with a 100% house mortgage (not denoted in terms of dollars, rands or any other currency). When ownership is complete, with 100% ownership, then the proud householder will be required to save so that he or she owns 120% or 130% ownership of the house. This will be given back as and when needed (unemployment, sickness, retirement). The mathematical integral determining this would be: B.dT = zero integrated over the period of the home ownership

      In the event of no income, the mortgage would flatline. With no interest, the mortgage cannot increase. There would be no bailiffs banging on the door with a court repossession order. Electronic money would be the absolute proof that there is no income.

      Is there others in cyberworld entertaining this line of thinking?

      1. Nigel,

        I like your approach for two reasons:

        Firstly, because your suggestion is not driven by the existing financial model but is rather trying the develop a financial model to fit the context/problem.

        Secondly, I like your income driven valuation as it reduces the possibility for speculation and housing bubbles.

        However, like most things, the devils in the detail and I think the institutional and data requirements for your model are quite onerous. E.g. getting income levels at household or suburb level that are current is going to be a difficult task.

        I do like your fresh way of thinking though…


  3. Rob,
    I have though of this problem for years as a residential landlord in the UK. I have been infuriated by the asset price inflation over the past decade leading up to the incredible subprime mortgage crisis. House price valuations must be based on incomes for houses to be affordable. Good housing is the first line of defence to keep the doctor away, to keep body and soul together, and for a family to grow securely. The burden of stress should not fall on the mortgage holder unduly. My idea does indeed keep speculation and price bubbles at bay.

    Emphasis on high velocity of circulation of money nullifies the need for money supply growth and all that implies. Interest, rent, and mortgage insurance does not drift upwards. The mortgage holder is housed adequately while simultaneously acquiring real wealth in the form of brick and mortar.

    The institutional and data requirements could not be more severe.

    David Cameron is letting 40,000 or 50,000 statisticians leave government employment in the UK (according to press reports 2 years ago). They would certainly be employed if the data generated is to be properly processed.

    I have been thinking about the nature of debt. The subprime mortgage crisis has created massive public debts to bail out the banks. In the UK, the government after 2 years in power has reduced new borrowing from £1 in £4 to £1 in £5. In my opinion, we so addicted to government deficit spending it is analogous to addiction to crack cocaine. The working middle classes are being hollowed out by onerous payments to the banks, and rising taxes to the government.

    My way of thinking has the promise of taking all mortgages of bank balance sheets, and reduce government welfare. Housing benefit must necessarily become history. Balanced household budgets and balanced government budgets must become the new norm in the new financial architecture that needs to be created. As it is, we are collectively moving like lemmings to the edge of the cliff.

    all the best


  4. Hi there, simply became alert to your blog thru Google, and found that it is truly informative. I am going to be careful for brussels. I will be grateful when you continue this in future. A lot of other people will probably be benefited out of your writing. Cheers!