Housing for All: The Challenges of Affordability, Accessibility and Sustainability

Full text

Housing for All: The Challenges of A ordability, Accessibility and
Sustainability
The Experiences and Instruments from the Developing and Developed Worlds
A Synthesis Report
United Nations Human Settlements Programme
Nairobi. 2008

ii
 e Human Settlements Finance and Policies Series
Housing for All:  e Challenges of Aff ordability, Accessibility and Sustainability
First published in 2008 in Nairobi by UN-HABITAT.
Copyright © United Nations Human Settlements Programme (UN-HABITAT) 2008
HS/1012/08
ISBN:978-92-1-132025-1 (series)
ISBN: 978-92-1-131992-7
Acknowledgements
 is report is prepared by Xing Quan Zhang. Guidance and/or comments by Anna Tibaijuka, Inga Klevby, Oyebanji
Oyeyinka, Don Okpala and Fred Neto. Editing by  ierry Naudin and Roman Rollnick. Design and layout by Irene
Juma.
Disclaimer
 e designations employed and the presentation of the material in this publication do not imply the expression of any
opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of any country,
territory, city or area or of its authorities, or concerning the delimitation of its frontiers of boundaries.
Views expressed in this publication do not necessarily refl ect those of the United Nations Human Settlements Programme,
the United Nations, or its Member States.
Excerpts may be reproduced without authorization, on condition that the source is indicated.

iii
Foreword
As we move into the new millennium, one trend overwhelms our concerns: the rapid
urbanization with deepening poverty, environmental degradation and increasing slums,
which poses tremendous challenges for achieving adequate shelter for all.  e challenges
we face in human settlements cannot be met by governments, private sector or civil
society alone. It requires the actions of all sections of the society.
Human settlements are places of organised human activities.  e way in which human
settlements are organised is infl uenced by the pace and breadth of socio-economic
development. Such development cannot take place without linkages and continuous
interactions between physical, institutional, economic and social structures. Human
settlements are the product of deliberate planning or of spontaneous and uncontrolled
economic and social activities.  e problems and issues of human settlements cut across
the conventional socio-economic sectors and are of multi-sector and multi-disciplinary
nature.
 e national resource allocation and fi nance strategies are evolving towards the
identifi cation of national development priorities and challenges.  erefore, a full
understanding of human settlements needs to be looked at in the national policy context,
and its links to fi nance and to policy debate.  is approach is increasingly appreciated
by policy-makers and planners when addressing human settlements problems and policy
options.  e Human Settlements Finance and Policies series aims to explore the
intricacy of fi nance and policy interrelations and to promote better human settlements
fi nance policy and strategies.
 is series addresses the most important issues in improving human settlements. It
draws the intellectual leaders and practitioners from the governments, local authorities,
private sectors and civil society to confront human settlements and fi nance problems
and to exchange views and experiences in tackling human settlements problems and
issues, and to explore and promote innovations in policies, strategies and methods to
address challenges in human settlements. Papers in this series provide opportunities
to move towards a deeper understanding of the broad range of human settlements
and fi nance issues.
Our habitat is shaped by human actions and policies. Policies have profoundly
shaped our cities, towns and villages in the past, they will continue to defi ne the
21st century. Decision-makers face challenges of designing policies that allow their
countries and cities to meet the increasing human settlements challenges. In the new
era I hope that this series will contribute to the policy debate and will enhance the
capacity of member states to design new policies and strategies to address human
settlements challenges. In human settlements policy debate, choices made today will
impact our common future in the habitat of tomorrow.
Dr. Anna K. Tibaijuka
Under-Secretary-General and Executive Director
United Nations Human Settlements Programme

iv
Abbreviations and Acronyms
ACHR – Asian Coalition for Housing Rights
ADB – Asian Development Bank
ASA - Association for Social Advancement
BAAC - Bank for Agriculture and Agricultural Co-
operatives
BAPINDO – Bank Pembangunan
BBD – Bank Bumi Daya
BDN – Bank Dagang Negara
BNI – Bank Negara Indonesia
BRAC - Bangladesh Rural Advancement Committee
BRBD – Bangladesh Rural Development Board
BRI - Bank Rakyat Indonesia
BTN – Bank Tabungan Negara
CBO – Community-based organisation
CDT – Community Development Trust
CEL - Comptes d’Epargne Logement
CHF – Co-operative Housing Foundation
CMHC – Canadian Mortgage and Housing Corporation
CMP – Community Mortgage Programme
CNL – Caisse Nationale du Logement (National Housing
Fund)
CRA - Community Reinvestment Act
CSH – Contractual Savings for Housing
CUDS – Center for Urban Development Studies
DFID – UK Department for International Development
EIU - Economist Intelligence Unit
EXIM – Ekspor Impor Bank Indonesia
EHLP – Expanded Housing Loan Programme
FHA – Federal Housing Administration
FDI – Foreign Direct Investment
FONHAPO – Fondo National de Habitaciones Populares
(National Housing Fund)
FOVI – Fondo de Operación y Financiamiento Bancario a la
Vivienda (Bank of Mexico Trust Fund)
FOVISSSTE - Institute of Social Security and Services
for State Workers (Fondo de la Vivienda del Seguro Social al
Servicio de los Trabajadores del Estado)
FUNHAVI – Foundation for Habitat and Housing
(Mexico)
GBPHBUMC –  e General Board of Pensions and
Health Benefi ts of the United Methodist Church
GDS – Gross Debt Service
GHB – Government Housing Bank in  ailand
GHIF –  e Group for Home & Infrastructure Finance,
Inc.
GLAP – Group Land Acquisition and Development
Programme
GMD – Gender, Media and Governance
GOT – Government of Tanzania
GSIS – Government Service Insurance System
HA – Housing Authority
HDMF – Home Development Mutual Fund
HFC – Home Finance Corporation
HFHT – Habitat for Humanity Tanzania
HIS – Institute for Housing Studies
HIGC – Home Insurance Guaranty Corporation
HRF – Housing Revolving Fund
HUD – US Department of Housing and Urban
Development
IDB – Inter-American Development Bank
IMF – International Monetary Fund
INFONAVIT – National Housing Fund for Workers
(Instituto del Fondo Nacional de la Vivienda para los
Trabajadores)
JPMC – JPMorgan Chase
KUK – Kredit Usaha Kecil (small business loans)
LDU – Direction locale de l’Urbanisme (Local Urban
Planning Directorates)
LGU – Local Government Unit
LEHC - Limited Equity Housing Co-operative
LIHTC – Low-income Housing Tax Credit
LISC – Local Initiative Support Corporation
LTV – Loan to Value
MBS – Mortgage-Backed Securities
MDG – Millennium Development Goal
MFI – Microfi nance institution
MIS – Management Information System
MLGHE – Ministry of Local Government, Housing and
Environment
MMIF – Mutual Mortgage Insurance Fund
MOF – Ministry of Finance
MOH – Ministry of Housing
NACHU - National Co-operative Housing Union
NBFI – Non-bank fi nancial institution
NGO – Non-governmental organisation
NHC – National Housing Corporation
NHDFC – Nepal Housing Development Finance
Company

v
NHMFC – National Home Mortgage Finance
Corporation
NLTB – Native Land Trust Board
NSB – National Statistics Bureau
PEL – Plan d’Epargne-Logement
PELITA – Five-Year Development Plan
PRIDE - Programme Intégré pour le Développement de
l’Entreprise
PRB – Public Rental Board
RDA – Rural Development Administration
SACCO - Savings and Credit Co-operative
SHDLP – Social Housing Development Loan
Programme
SHF – Sociedad Hipotecaria Federal (Federal mortgage
corporation)
SIDA - Swedish International Development Agency
S & L – Savings & Loans institution
SMMS – Secondary Mortgage Market System
SOFOLES – Sociedades Financieras de Objecto Limitado
(Non-banking fi nancial institutions)
SPARC –  e Society for the Promotion of Area Resource
Centers
SRA - Slum Rehabilitation Authority
SSS – Social Security System
SU – Support Unit
SUPF – Solidarity for the Urban Poor Federation/Squatters
and Urban Poor Federation
TAWLAT – Tanzania Women Land Access Trust
THB – Tanzania Housing Bank
UHLP – Unifi ed Home Lending Programme
UNCTAD – United Nations Conference on Trade and
Development
UNDP – United Nations Development Programme
UN-HABITAT – United Nations Human Settlements
Programme
UNDESD – United Nations Department of Economic
and Social Development
UNIDO – United Nations Industrial Development
Organisation
UOD – University of Dar es Salaam
UPDF – Urban Poor Development Fund
USAID – United States Agency for International
Development
USG -  e Urban Sector Group
VA – Veteran Administration

vi
Table of Contents
Foreword ................................................................................................................................................................ iii
Abbreviations and Acronyms .................................................................................................................................. iv
PART I Introduction .......................................................................................................... 1
Chapter 1: Introduction ........................................................................................................................................ 2
PART II  e Challenge of Housing Finance ..................................................................... 2
Chapter 2:  e Challenge of Access to Housing Finance for Low-Income Households ........................................ 2
Why Formal Financial Institutions Do Not Work for the Poor .................................................................................... 2
 e Mystery of Lack of Money ................................................................................................................................... 3
Lack of a Property Market to Facilitate a Housing Financial System ........................................................................... 3
 e Long Process towards Ownership ......................................................................................................................... 4
Lack of Property Rights Restricts Women’s Access to Credit ........................................................................................ 4
Poor Credit Cultures and Information Systems ........................................................................................................... 5
 e Absence of Robust Credit Markets in Developing Countries ................................................................................ 5
Asset-Liability Mismatches and Lack of Long-Term Credit ......................................................................................... 5
High Collateral Requirements ..................................................................................................................................... 6
Poor Management of Financial Assets ......................................................................................................................... 6
High Borrowing Costs ................................................................................................................................................ 7
Ineff ective Legal and Judicial Systems .......................................................................................................................... 7
 e High Costs of Lending to the Poor ....................................................................................................................... 7
Low Individual Participation in the Banking Sector .................................................................................................... 8
PART III Experiences and Instruments for Low and Moderate Housing ......................... 9
Chapter 3:  e Hopes and Challenges of Housing Microfi nance for Low-Income Households ......................... 9
 e Formal System Hinders Aff ordable Housing Finance ......................................................................................... 9
Can Microfi nance Meet the Housing Finance Needs of the Poor? ............................................................................. 9
Poverty and the Microfi nance Revolution ................................................................................................................. 10
 e Emergence of Housing Microfi nance ................................................................................................................. 11
Providers of Housing Microfi nance Services .............................................................................................................. 12
 e Funding Strategies of Housing Microfi nance ...................................................................................................... 15
 e Characteristics of Housing Microfi nance Products ............................................................................................ 17
Limitations of Housing Microfi nance ....................................................................................................................... 23
Chapter 4:  e Co-operative Approach to Low-Income Housing ...................................................................... 26
Housing Demand is a Challenge for the Conventional Formal Sector ....................................................................... 26
Housing Co-operatives ............................................................................................................................................. 26
Self-Build Housing ................................................................................................................................................... 31
Chapter 5:  e Role of Government in Low-Income Housing ............................................................................ 33
 e Role of Government in Low-Income Housing ................................................................................................... 33
Direct Provision of Public Housing: the Case of Hong Kong .................................................................................... 33
Government Mortgage Loan Schemes: the Case of Singapore ................................................................................... 36
Government-Sponsored Mortgage Finance: the Case of Fannie Mae and Freddie Mac in the USA ........................... 36
Government Programmes for Improved Conditions in Slums: the Case of Colombo, Sri Lanka ............................... 40

vii
Chapter 6: Financial Instruments for Low-Income Housing .............................................................................. 45
Financial Instruments for Low-Income Housing ....................................................................................................... 45
Community-Based Savings Schemes for Lowest-Income Housing ............................................................................. 45
Compulsory Savings Schemes ................................................................................................................................... 45
Contractual Savings for Housing .............................................................................................................................. 46
Special Housing Funds .............................................................................................................................................. 48
Housing Bonds ......................................................................................................................................................... 51
Housing Banks .......................................................................................................................................................... 53
Rent-to-Purchase Housing Schemes .......................................................................................................................... 57
Trusts ........................................................................................................................................................................ 57
Chapter 7:  e Multilateral Financial Institutions and Low-Income Housing Projects ...................................... 61
 e Inter-American Development Bank and Low-Income Housing Finance ............................................................. 61
 e Asian Development Bank and Low-Income Housing Finance: the Case of Fiji ................................................... 62
 e World Bank and Low-Income Housing Projects: the Case of Algeria .................................................................. 64
Chapter 8: Revolving Funds for Human Settlements .......................................................................................... 67
Designing Revolving Funds for Housing Programmes .............................................................................................. 67
 e Revolving Fund for House Improvement Loans in the Dominican Republic ..................................................... 67
 e Safe Drinking Water Revolving Loan Fund in Oregon ....................................................................................... 71
Chapter 9: Credit Enhancement ........................................................................................................................ 72
Forms of Credit Enhancement ................................................................................................................................ 72
External Credit Enhancement ................................................................................................................................. 72
US Government Mortgage Insurance ......................................................................................................................... 73
Internal Credit Enhancement ..................................................................................................................................... 78
Chapter 10: Case Studies of Housing Finance Institutions and Systems ............................................................. 81
 e Development of Housing Finance Institutions in Indonesia ................................................................................ 81
 e Development of Housing Finance Institutions in Nepal ...................................................................................... 83
 e Housing Finance System in the Philippines ......................................................................................................... 86
References ...........................................................................................................................................................92

Rapid urbanization in the developing world is the most
unprecedented phenomena of the world’s development
in the past few decades.  e pace of urbanization has
exceeded many developing cities’ capacity to absorb the
needs of a growing population, despite all innovations and
eff orts. One of the most pressing problems is to provide
adequate housing for all, particularly for the poor.
Poor urban housing conditions are a global problem,
but conditions are worst in developing countries. UN-
HABITAT (2003) reports that 1 billion people live in
life-and health-threatening homes.  is represents about
one third of the world’s total urban population, while the
developing world has a substantial proportion of the urban
population living in inadequate housing conditions.
 e threat of mass homelessness is greatest in Africa, Asia,
and Latin America because that is where population is
growing fastest. It seems that old paradigms are unworkable,
many existing formal supply channels often hopelessly
inadequate to low income people, and most conventional
approaches largely irrelevant given the magnitude of the
growth of the sub-standard settlements.
However, the commitment and eff orts for the poor people
are often limited, given the limited resources faced by the
developing world. Many developing countries struggle to
solve their housing problems.  ey often fi nd that there are
lack of adequate knowledge and experiences in housing.
 eir housing poverty is not only linked to economic
poverty but also linked to knowledge poverty and skills
poverty.  is is the reason why member states consistently
call for increased normative activities.
In the past century, the world has witnessed great changes
and improvement of living conditions in many countries
along with their economic growth. Various innovations
and instruments arise to tackle the housing problem.  ere
is much variation in policies, instruments and innovations
in diff erent countries.  ere is a need to fi nd common
ground in search for solutions for housing.  ere is a great
need for sharing experiences and practices across countries.
 ere is a huge demand among countries to learn from
other countries’ experiences and practices.
 is book provides a comprehensive synthesis of various
innovations and instruments in low and moderate
income housing. It focuses on the issues of aff ordability,
accessibility and sustainability, with special reference to
housing fi nance. It illustrates the principal instruments,
innovations and policies in housing provision and housing
fi nance; from specialized housing fi nance institutions to
community based fi nancing initiatives; from mortgage
fi nance to housing microfi nance; from self-build to
cooperative approaches; from market-driven instruments
to government-led initiatives; from revolving funds to
credit enhancements; from experiences of individual
countries to practices of multilateral institutions; from
developing countries to developed countries; from specifi c
instruments and methods to the improvement of systems.
It aims to provide an “all-in-one” synthesis of tools and
experiences and practices in housing provision and housing
fi nance for low and moderate income households.
PART I
Introduction
Chapter 1: Introduction

2Housing for All
Why Formal Financial Institutions
Do Not Work for the Poorw
Rapid urbanisation leads to a crisis of unprecedented
magnitude in urban housing delivery. UN-HABITAT
reports that every year the world’s urban population
expands by some 70 million, most of it in developing
countries where economic capacities cannot cope in terms
of housing and urban services provision. As a result, cities
feature very high proportions of informal dwellings, which
either were constructed to standards that do not conform
to established legislation, or built on land for which the
occupier has no proper title – or both. In some cities, up to
60 per cent of dwellings are in informal settlements.  ese
dwellings are appropriate responses of the poor to their
social and economic circumstances. In many countries,
there is de facto security of tenure as governments
discourage evictions, acknowledging their failure to enable
the land and housing market to supply shelter aff ordable
to the poor, and cognizant of the votes wielded by the
occupants of these areas. Housing in such areas can be built
incrementally according to the budget and circumstances
of the households involved.  e construction of housing
is fi nanced by informal fi nancial mechanisms, mostly
through household savings, loans from relatives, and other
means.
If Millennium Development Goals (MDGs) are to be
achieved, aff ordable housing is goint to be required
on a massive scale and strategies must be developed
for immediate implementation.  e key to providing
housing does not lie solely in the number of humanitarian
programmemes launched by institutions such as
UN-HABITAT, the World Bank, non-governmental
organizations or even governments. It has taken decades
to recognise that marginalized communities have a major
role of their own to play, too.  ey must be given the
technical and planning skills and techniques enabling them
to implement acceptable solutions to provide aff ordable
housing. Governments over the last three decades have
tried to address the problem of aff ordable housing.  e
United Nations Declaration of 1974 was drafted primarily
to encourage developing nations to expand low-cost
housing on a “self-help” basis, through establishment of co-
operatives utilising, as much as possible, local raw materials
and labour. Many governments around the world have
planned to increase housing delivery along these lines on
a sustainable basis. Progress as observed is very good, such
as in China and Eastern Europe, while remaining slow in
many other countries, and the problem of large informal
settlements is not just still there, it is growing, too.
Despite the increasing expectation that the private sector
should play a very important role in fi nancing the down
market, the formal fi nancial sector is reluctant to serve the
low income sector.  e poor are excluded from the formal
fi nancial institutions because they (1) require to have
bank accounts with minimum deposits, discourage small,
regular deposits from poor clients; (2) charge high fees; (3)
fi nancial institutions are located outside the areas where
the poor live and increase the travel cost for the poor to
access such services; (4) the poor do not have formal legal
Chapter 2: The Challenge of Access to Housing Finance
for Low-Income Households
PART II
 e Challenge of Housing Finance

3
The Challenges of A ordability, Accessibility and Sustainability
titles to provide collateral security as required by formal
fi nancial institutions; (5) the poor do not have capacity
to make monthly repayment over a long period; (6) the
poor often do not have formal employment.  erefore
the formal fi nancial institutions fail to provide fi nancial
services to the low-income sector.
The Mystery of Lack of Money
It is a common belief that poor countries lack the resources
to provide fi nance for low-income housing. However,
research shows that poor countries are not as poor as they
are thought to be.  ey have capital.  e poor do save
portions of their earnings.  e value of savings among
the poor is huge – 40 times all the foreign aid developing
countries have collectively received since 1945. In Egypt,
the wealth accumulated by the poor is 55 times foreign
direct investment (FDI). In Haiti, the total assets of the
poor are more than 150 times greater than all the FDI
received since independence in 1804; untitled real estate
assets are together worth some USD 5.2 billion. However,
to use the phrase coined by De Soto, these assets are “dead
capital” which cannot be transferred in the market. Many
poor people live in houses that do not carry any legal titles.
In Egypt, “dead capital” housing accounts for 92 per cent
of urban residents and 83 per cent of urban residents x.
However, this argument may lead to an illusion that poor
people are actually “rich”, having locked money in the
form of unmoveable assets like dwellings.  is optimism
obviously overestimates the value of poor people’s
housing. Most poor people live in substandard housing
and slums which have little exchange value on the market.
 erefore, the housing asset worth of the poor is much
more imaginary than real.
Lack of a Property Market to
Facilitate the Functioning of
a Housing Finance System
Assuming that poor people do have a wealth of capital
locked in the form of housing, it is diffi cult to make use of
it for lack of a property market. Many developing countries
have no clear, defi nite, private property rights, which are the
pre-conditions for any eff ective property market. Informal
settlements often exist outside the legal framework.  ey
lack titles for formally legal transactions. Many informal
settlements are under constant threats of demolition from
government (see Box 1).
In some developing countries, people are required to go
through extremely tedious procedures to legalize their
land plots and housing units, or to obtain legal titles for
their properties. In Peru, it takes six years and 11 months
to obtain authorization to build a house on State-owned
land, and it takes 728 steps to obtain a legal title for that
piece of land. In the Philippines, it takes 13 to 25 years
to purchase land legally, involving 53 public and private
agencies. In Egypt, would-be buyers must go through at
least 77 procedures to acquire and legally register a piece
of State-owned land
y
. Although these might be extreme or
even exaggerated cases, they refl ect the challenging nature
of legal titles and attendant transactions in some developing
countries.  e extreme diffi culties involved in the delivery
and transfer of legal titles make it more diffi cult to use
land and housing property as collateral for access to loans.
Compare what has occurred in Mumbai with evictions and homelessness in Kenya or Mexico City or New Delhi. In
terms of both the sheer numbers of poor people aff ected and the brutality of the demolitions, the evictions in Mumbai
qualify as amongst the worst cases anywhere in the world, with as many as 90,000 shanties torn down between 4
December 2004 and 5 March 2005. Evictions took place on Sundays or religious or festival days. Evictees’ belongings
were then set on fi re. Bulldozers demolished homes when people were still inside.  is violates a range of human
rights, including the right to adequate housing.  e demolitions are in eff ect creating apartheid in the city, with a clear
demarcation of where the rich and the poor live.  e poor are standing in the way of enormous profi ts to be made
through land speculation — among the highest in the world — and if that helps build a so-called world-class city,
well, who cares for the poor? Nobody seems to be looking at the social and psychological impacts of such a clear case of
dispossession. Is anyone looking at the impact on specifi c groups—on women, on children, on Dalits? Mumbai needs
a judicious mix of public housing with lower rent costs, co-operative housing, microfi nance (an option where the poor
can contribute partly even where there is a subsidy) and a socially sensitive use of available legislation.  e solution
has to be based on a human rights approach that meets the needs of the vulnerable fi rst, that respects the views of all
women, men and youth in the city and includes their participation in the planning process.
– Miloon Kothari, United Nations’ Special Rapporteur on Adequate Housing after visiting the scene of the evictions.
Source: Habitat Debate, Volume 11, No. 2, June 2005, Page 15
Box 1 Informal settlements under government threats of demolition

4Housing for All
However, some developing countries have made eff orts to
survey and regularize land parcels (Figure 2.1).
The Long Process Towards Ownership
Research shows that most low-income households aspire
to own houses. Housing is the most expensive asset for
most households.  e substantial investment required, to
buy a fi nished unit is often way beyond the means of low-
income households. Rather, it becomes a socio-economic
process that involves self-build and incremental housing
construction. Families and relatives provide assistance
towards achieving home ownership such as (whenever
possible) in the form of “soft loans” requiring little if any
interest and repayments. In developing countries, the
vast majority of low-income households have built their
houses incrementally over long periods of time z.  e
incremental process of construction refl ects their limited
fi nancial resources. It also creates emotional ties between
families and their houses; as a result, poor families may
be more reluctant to use their homes as mortgages for
loans.  e high rate of home ownership among the poor
in Lima points to the fact that they value their “houses”
very highly.
Lack of Property Rights Restricts
Women’s Access to Credit
Access to credit is more diffi cult for women because they
lack property rights under their own names. In 1995
Kenya, less than one per cent of women owned land. In
2005, this percentage rose to four per cent – still leaving
96 per cent of land in male ownership.  e land tenure
system and property rights are severely skewed towards
men aa.
Kenya is not an isolated case. Women around the world
face various constraints to access to land and property. In
some African countries, patrilineal kinship is the basis that
justifi es men’s superiority over women in practice.  e
prejudice against women starts from the day of birth, as
they are always under the control of men – father, husband
or brother. Land is passed from men to men, for example
from father to son. Lack of access to property rights in
Figure 2.1 Survey of Land for Regularisation in Tanzania
Source: GOT

5
The Challenges of A ordability, Accessibility and Sustainability
rural areas often forces women to migrate to urban areas.
Custom is not the only factor that denies women’s access
to land and property: their economic inability to purchase
these is another factor. As women are not able to provide
collateral, they often fi nd themselves in no position to
secure loans from the formal banking system ab.
Poor Credit Cultures and
Information Systems
Credit culture has a major impact on the fi nancial sector.
Financial institutions have their own internal credit
culture to guide their lending operations. Credit is not
granted unless there is a demonstrated capacity and
willingness for repayment, which goes through a rigorous
analytical scrutiny for every loan.  is scrutiny focuses on
the ability of the borrower to generate constant cash fl ows
to cover the periodic repayment ac. Such scrutiny is often
found humiliating by the poor and scares them away from
the formal fi nancial sector, since many have no regular
employment or stable income sources.
In centrally-planned countries, the fi nancial sector suff ers
from a lack of credit culture. Loans from State-owned banks
are often subject to government directives or plans, though
not to a thorough analysis of the borrower’s repayment
capacity. In the absence of a credit culture, State-owned
banks have been associated with lower growth in income and
productivity.  ey often have high rates of bad loans
ad.
In countries where the credit culture is poor, banks
often do not assess the borrower’s repayment capacity,
and borrowers do not take loan repayment obligations
seriously. In some African societies, some people even
treat loans as grants or income.  is culture is encouraged
by non-enforcement of loan contracts by State-owned
fi nancial institutions.  is poor credit culture results in
signifi cant increases in banking risks. For example in the
year 2000, one Tanzanian State-owned bank had a 44 per
cent bad loan rate ae.
Another issue associated with poor credit culture is lack
of credit information systems.  ese act as information
brokers that improve the transparency of credit markets.
However, in many developing countries credit information
systems are lacking or still in infancy. Financial liberalization
enhances competition between lenders, which increases
over-indebtedness, reduces loan repayment incentives and
causes an accumulation of repayment arrears. Asymmetric
information between borrowers and lenders can lead to
adverse selection and moral hazard.  e high default rate
and non-sustainability of many microfi nance institutions,
which specialize in lending to the poor, point to the
problems caused by an absence of credit information
systems af.
Absence of Robust Credit Markets
in Developing Countries
 e absence of robust credit markets in developing countries
is a signifi cant impediment to solve the housing problems.
To most families, housing is the largest investment in
their lifetime.  ey need loans to fi nance investments
in homes. However, in developing countries, dreams of
decent homes run against most people’s inability to obtain
loans. In contrast, there is widespread access to credit in
most developed countries, including for home purchases.
Research shows that credit is one of the most important
factors that facilitate expansion in real estate markets and
housing in developed countries ag.
Figures 2.2 and 2.3 show a strong correlation between
economic development and credit markets.  e more
developed a nation’s economy, the more developed its
credit markets. High-income countries feature far higher
rates of domestic credit to the private sector in terms of
per centage of GDP.  e United States has the highest
rate.  e least developing countries have the lowest rates
of domestic credit to the private sector, followed by heavily
indebted and low-income countries. Domestic credit from
the banking sector follows the same trends as domestic
credit to the private sector. In 2000, annual domestic
credit provided by the banking sector in high-income
countries averaged 210 per cent of GDP. For the USA,
the fi gure was 258 per cent of GDP.  e average was 42
per cent of GDP for low-income countries,; 27 per cent
for heavily indebted poor countries; 24 per cent for least
developed countries; and only half of that – 12 per cent
– in Tanzania ah.  e disparity in domestic credit markets
between low-income and high-income countries is even
greater than these fi gures indicate, because the range of
available fi nancial instruments is smaller in those with
low incomes For example, high-income countries have
sizeable bond markets and other signifi cant sources of
credit, while banks are the primary sources for credit in
low-income countries ai.
Asset-Liability Mismatch and
Lack of Long-Term Credit
Developing countries have low levels of domestic credit,
but, even when credit is available, loans are often to be
repaid in a very short time period. In many of the least
developed countries, three to fi ve years will often be the
longest available maturities for loans. For example, in
Tanzania, a women’s group had reached an agreement
with a commercial bank on provision of 10-15 year home
loans; but eventually the bank cancelled the agreement and
reduced the maturity to three years, which was the longest
term the bank could off er in 2005.  is contrasts with the
10 to 40 years it will often take for a household on salaried

6Housing for All
income to repay in full.  erefore, short repayment periods
make most families unable to generate enough income to
meet repayment requirements.
On the other hand, the lending policies of the mortgage
fi nance sector in many developing countries are also
aff ected by a number of factors – economic instability,
fl uctuations in infl ation, and foreign exchange rate risk
– which force the banking sector to raise real interest rates
and to reduce maturities in order to curb the high risks aj.
 ese factors further contribute to the lack of long-term
credit.
Figure 2.2 Domestic Credit to the Private Sector (per
cent of GDP)
Source: Based on World Bank data 2005
Figure 2.3 Domestic Credit by the Banking Sector (per
cent of GDP)
Source: Based on World Bank data 2005
High Collateral Requirements
Many banks in low-income developing countries do not
off er home loans to individuals. When they do so, they
will often impose more requirements than lending to
corporations. High collateral requirements for home loans
is another hassle for borrowers. In a recent discussion with
a State-owned bank in Tanzania, it appeared that collateral
requirements for home loans exceeded 180 per cent of
loan amounts, while the maximum ratio of loans to value
was only 65 per cent.  ese harsh conditions preclude
most potential borrowers, and the poor in particular, from
obtaining loans.
Poor Management of Financial Assets
Low credit amounts are not entirely due to a lack of
funds in the banking sector. For example, in Tanzania,
the banking sector has experienced signifi cant increases in
funds in recent years, which contributes to an expansion
in deposits. In 1999, deposits accounted for 53 per cent
of total funding sources, and other liabilities 43 per cent.
 e deposits expanded very quickly and their proportion
in total funding sources rose to 84 per cent in 2003. Other
liabilities accounted for six per cent, share capital for
another six per cent, and other capital for four per cent
(Figure 2.4). Banks increasingly rely on deposits to fund
loans.

7
The Challenges of A ordability, Accessibility and Sustainability
Figure 2.4 Funding Sources of the Banking Sector in
Tanzania (billion Tshs)
Source: Based on BOT data
However, a large por tion of the deposits do not go into loans
but instead remain in the form of liquid assets such as cash
positions, inter-bank loans and government securities.  e
three largest domestic banks in Tanzania (National Bank of
Commerce (NBC), Co-operative and Rural Development
Bank (CRDB) and National Microfi nance Bank (NMB))
together hold over 60 per cent of total deposits, but lend
only about 30 per cent of these, compared with some 60
per cent for those international banks operating in the
country. In the USA, by comparison, banks only keep
six per cent of total deposits in liquid assets and the bulk
of their capital is used for loans. Banking in Tanzania is
characterised by excessive liquidity and extremely low
credit.  is represents a massive failure of the fi nancial
system to allocate capital to the most productive uses ak.
High Borrowing Costs
In Tanzania, one of the main challenges for borrowers is
the high cost of credit. In 1995, the average deposit interest
rate was 25 per cent and the lending rate was as high as
43 per cent. Since then, gradual economic stabilization
and improved competition have brought the rates down
(Figure 2.5). By May 2005, lending rates had dropped
to 14-17 per cent and deposit rates to 3-4 per cent.  e
spread between lending and deposit rates decreased from
some 18 per cent in 1995 to about 11-13 per cent in
May 2005. One of the reasons for maintaining such high
spreads was to compensate for low utilisation of deposits
due to weak market demand and high transaction costs.
 is practice brings additional challenges to individual
customers. It makes credit unaff ordable for the vast
majority of individuals, and next to impossible for those
on low income.
Ineffective Legal and Judicial Systems
Eff ective legal and judicial systems are essential for any
fi nancial system. Whether the legal and judicial system
will enforce contracts and the law eff ectively will aff ect
the willingness of lenders to make mortgage lending
or other loans. In addition, lenders want a system that
can settle disputes promptly.  erefore, availability of
alternate dispute resolution mechanisms is important to
boost bankers’ confi dence in housing fi nance. In Tanzania,
the judicial system is far too slow when it comes to
commercial dispute settlement. An offi cer at the National
Bank of Commerce (NBC) told of diffi culties they faced
when trying to settle default issues in court. One client
used land as security against a loan. When he defaulted,
NBC attempted to repossess the land. But the case had
been pending in court for several years no settlement was
yet in sight.
Figure 2.5 Interest Rates in Tanzania (per cent)
Source: Based on IMF data
The High Costs of Lending to the Poor
To banks, the risk of credit default is higher with the poor
as their circumstances are much more likely to change
over time than for any other segment of the population.
 e amounts they borrow tend to be relatively small,
maturities are short and transaction costs are higher.
 erefore, the formal banking sector is reluctant to
enter this market. Many lenders in this market ‘manage’
customer repayments, rather than relying on customers to
pay on time: agents collect repayments from customers’
homes, which further contributes to high lending costs 97.
0
5
10
15
20
25
30
35
40
45
1994 1996 1998 2000 2002 2004
Deposit interest rate (%)
Lending interest rate (%)

8Housing for All
Low Participation of Individuals
in the Banking Sector
Currently, the banking sector in Tanzania focuses on
corporate banking services.  e participation of individuals
in banking services is very low.  e Government’s household
budget survey showed that only 6.4 per cent of households
had a bank account in 2000. Privatization of national banks
resulted in reduced outreach of individual customers due
to high transaction costs.  e country’s poor credit culture
makes the banking sector reluctant to expand business and
serve the general public, particularly the poor. However,
in recent years, banks began to show their interest in
reaching the general public through microfi nance services.
Some banks have established microfi nance departments
or subsidiaries. Although microfi nance loans are typically
short term, the move refl ects the changing attitudes of
Tanzania’s banking sector towards low-income people. But
there is a long way to go to build a robust credit market
for individual clients.

9
The Challenges of A ordability, Accessibility and Sustainability
PART III
Experiences and Instruments for Low and
Moderate Housing
The Formal System Makes Affordable
Housing Finance Next to Impossible
Formal housing fi nance remains a Utopia to the poor.
 e formal system has proved unable to meet the needs
of poor people around the world am. Any eff ective housing
fi nance system requires the balancing of three elements:
aff ordability for households, profi tability for fi nancial
institutions, and the capacity to scale up the transactions.
Daphnis states that in practice, lenders, governments
and donors understand these three principles as the need
to fi nance a complete housing unit under terms that
are aff ordable to the poor, profi table to the commercial
sector, and on a scale large enough to help solve the
housing problem. However, the same author claims that
this is simply impossible to achieve in poor and very
poor countries, where mortgage lending and secondary
mortgage markets have never materialized as fi nancially
viable options when addressing the housing needs of low-
income households. Poor households lack the capacity to
repay large amounts of money borrowed at real interest
rates to fi nance complete housing units. Housing fi nance
normally requires long repayment periods. However, in
developing countries, most sources of funding are short-
term, which creates a serious asset/liability mismatch
for fi nancial institutions, which signifi cantly raises the
risk of long-term lending. Poor households often have
diffi culties to sustain regular repayments for long periods
of time. Investors do not want to risk their savings in
securities backed by such loans.  erefore, no reliable
secondary mortgage market for housing loans targeting
poor people exists in developing countries.  e donor–
supported housing banks that had been aiming to serve
the poor either went bankrupt or evolved into full-fl edged
commercial banks and shifted loan portfolios away from
poor households an.
Can Microfinance Meet the Housing
Finance Needs of the Poor?
Poor people face two problems when looking to use
fi nancial services: access and cost. Many poor people have
to turn to informal moneylenders for access to credit. But
these fi nancial services come at high costs, with nominal
monthly interest rates typically ranging from about 10 to
more than 100 per cent – a multiple of the monthly interest
rates of formal fi nancial institutions
ao
. Such high interest
rates make it prohibitive for poor people to borrow large
amounts of money for long periods.  e informal fi nance
mechanism is too costly for the poor to fi nance complete
housing units.  erefore, slum clearance and rebuilding
through informal fi nance is often not a feasible option for
the poor.
As for microfi nance institutions (MFIs), they can only
provide small amounts of credit at a time to individual
poor households. Although MFI interest rates are typically
lower than those of informal moneylenders, they are much
higher than those charged by formal fi nancial institutions.
In Mexico, the average size of housing micro-credit
introduced by CHF International and FUNHAVI is USD
1,800 with annual interest rates of 54 per cent. Loans must
Chapter 3: The Hopes and Challenges of Housing
Micro nance for Low-Income Households

10 Housing for All
be repaid with 18 months for fi rst-time borrowers. Table
3.1 shows the typical housing microfi nance schemes in
operation ap. Loans are typically very small and repayment
periods are short with a maximum of three years.  e
basic ideas behind housing microfi nance evolved from the
microfi nance revolution.  erefore, the main principles
of housing microfi nance refl ect those of microfi nance in
general: (1) loan sizes are relatively small and are disbursed
based on borrowers’ capacity for repayment; (2) repayment
periods are relatively short; (3) loan pricing is expected
to cover the real, operational cost and risk premium; (4)
loans are not heavily collateralized; (5) loans do not aim
to fi nance complete housing units at a time, but rather
housing improvements and incremental expansion of
home space; (6) although savings are not a prior condition
for obtaining housing micro-credit, MFIs often link loans
to clients’ savings profi le and habits aq. To put it briefl y,
housing microfi nance enables the partial fi nancing of
housing needs which meet only some of poor people’s
aspirations to proper homes.
Various initiatives have been launched to address the
high cost of credit for the poor around the world.  ese
programmes try to bring down the borrowing cost for
the poor.  ey are often supported by government or
donor-fi nanced nonblank fi nancial institutions. But most
of these institutions are capital constrained and can not
meet the needs of the vast majority of the poor as.  ey are
often not fi nancially self-sustainable.  e survival of these
fi nancial institutions depends on constant capital injection
from governments or donors. In other words, they are
heavily subsidized by governments or donors. However,
subsidies can be on a large scale to cover all the needs of
the poor. When coming to housing, the scale of subsidies
is too high for the government to aff ord.  erefore, the
wide replication of such initiatives is often economically
not viable for the poor countries.
Poverty and the Microfinance
Revolution
Growing Poverty Drives Poor People Away from
Access to Formal Sector Financial Services
About half of the world’s population (three billion) live
in poverty (less than two US dollars per day). Of these,
1.3 billion population live in severe poverty (less than one
US dollar per day), and 800 million lack access to basic
healthcare. Nearly 800 million population are hungry
or malnourished and 2.4 billion lack access to proper
sanitationat. As many as 1.2 billion have no access to safe
drinking water. Some 275 million children never attend or
complete primary school education, while 870 million of
the world’s adults are illiterate. As for health, three million
people die from HIV/AIDS every year (a cumulated 25
million in the last 20 years), and 70 million will die from
the condition by 2020; 40 million people are currently
infected with HIV/AIDS and will die within 10 years. 13
million children have been orphaned by HIV/AIDS since
the epidemic began, and the number is expected to double
to 26 million by 2010. With regard to housing, over 100
million people live in slums. An estimated 25 to 50 per
cent of urban residents in poor, developing countries live
in impoverished slums and squatter settlementsau.
Poor developing countries (host to 80 per cent of the world’s
population) pay the rich developed countries an estimated
nine times as much in debt repayments as they receive in
aid. Africa spends four times as much on repaying debt as
it spends on healthcare. In 1997, the foreign debt of poor
countries stood at over two trillion US dollars and was
still growing.  is amounts to USD 400 of debt per head
in the developing world – where average annual income
in the very poorest countries is less than a dollar a day,
as mentioned aboveav. With its associated vicious cycle of
low productivity due to lack of capital and skills, poverty
makes people unable to meet the standard requirements for
credit. In those areas where poverty is concentrated, many
economic activities are also organized in informal ways.
 ese entail specifi c characteristics that are often diff erent
Table 3.1 Selected Housing Micro nance Schemes
Institution Average
Loan Size
(US$)
Maximum
Repayment
Period
(months)
Security
Collateral
Required
Time with
Programme
Savings
Required
Solution
Type
TA to
Clients
ADEMI 1,800 36 collateralized None No Variable No
FUNHAVI 1,500 20 2 co-signers None No Variable Yes
CHF/Gaza 4,800 36 2 co-signers None No Variable Yes
Source: Daphnis & Ferguson (2004)ar

11
The Challenges of A ordability, Accessibility and Sustainability
from those of formal enterprises, including, for example,
scarcity of capital, family ownership, small-scale operations,
lack of legal status, lack of security of business operations,
labour-intensive production, informal education and
training, low skill levels, products transacted on informal
markets, etc.aw  ese defi ning features restrict the scale
of production and high risks and instability of informal
economies, which in turn deters the formal fi nancial sector
from providing fi nancial services.
The Demand of Microfinance Unmet
in Formal Financial Sector
 e microfi nance revolution is best understood against
the background of developing countries’ population and
income patterns as well as the formal fi nancial sector’s
inability to meet demandax.  e Consultative Group to
Assist the Poorest (CGAP) estimates that over 500 million
poor people around the world need fi nancial servicesay.
Robinson argues that the number is much larger, with
some 360 million households badly needing savings or
credit services from formal fi nancial institutions. If we
assume that the average household size is fi ve persons, then
it is some 1.8 billion people who demand microfi nance
services.
Research shows that in a country like Morocco, most
microfi nance demand from the poor is met by the informal
sector, including friends (38 per cent), family and relatives
(38 per cent), colleagues (21 per cent) and money lenders
(3 per cent).  e poor hardly have any opportunity to
turn to formal credit institutions since these often deny
them access. Among poor people, 20 per cent said that
when they have other alternatives, they will not consider
formal fi nancial institutions; 18 per cent said that they
did not approach formal credit institutions because they
had no collateral available;17 per cent worried about
repayment ability; 12 per cent thought that formal
fi nancial institutions were not pro-poor; and six per cent
were put away by the diffi cult and lengthy procedures of
the formal fi nancial institutions (Table 3.2).
The Rapid Expansion of Microfinance Services
Across the world at the end of 2003, as many as 2,931
microfi nance institutions were in operation serving
80,868,343 clients, 54,785,433 of whom were among the
poorest when they took their fi rst loan. Of these poorest
clients, 82.5 per cent, or 45.2 million, were women.
Assuming an average fi ve persons per household, the 54.8
million poorest clients served in late 2003 spread the
benefi ts of microfi nance loans to some 274 million people.
 is represents a 38 per cent annual growth in the number
of microfi nance clients, compared with its starting point
of 7.6 million poorest families at the end of 1997.  e
overall microfi nance growth of 621 per cent between
1997 and 2003 averaged about 39 per cent per year (Table
3.3).
The Emergence of Housing
Microfinance
Housing fi nance sources in developing countries generally
falls into three categories or tiers.  e fi rst category is
Table 3.2 Reasons for Not Applying for Formal Loans
Reasons First Quartile
per cent
Second Quartile
per cent
 ird Quartile
per cent
Four Quartile
per cent
Total per cent
Other Credit
Sources Available
17 20 21 23 20
Lack of Collateral 21 17 16 18 18
Fear of Inability
to Repay the
Loan
19 18 16 18 17
Lack of Bank
Relations
10 12 12 14 12
Diffi cult
and Lengthy
Procedures
73676
Lack of Financial
Documents
24523
Other 11121
Source: CHF International (2005)ba

12 Housing for All
comprised of private commercial institutions providing
credit for upper-income households at market interest rates
upon the certifi cation of income streams and provision
of collateral.  is category of fi nancial institutions avoid
involvement in provision of housing fi nance for the
poor due to their lack of collateral and steady income,
perceived high default risk, and high transaction costs.
 e second category is the public sector, which usually
provides subsidized funds for middle-income groups and
civil servants by way of specialized or non-specialized
housing fi nance intermediaries. In many developing
countries these public housing programmes have failed
to reach the poor.  eir eligible benefi ciaries normally
operate within the formal economy, can demonstrate
basic home ownership capacities, and have some access
to capital. Public programmes attempting to target lower
income households have been impeded by weak political
will, a paucity of available funds, leakage of funds to
non-eligible groups through corruption, or a failure to
comprehend the socio-economic and political dynamics
of the situation within which the poor operate. Since
they often work in the informal economy, the poor have
(with only a few exceptions) been excluded from access to
capital from formal private or public fi nancial institutions.
 ey have to rely on informal sources, including savings,
informal loans from friends and family, remittances from
family members working abroad, and the sale of whatever
assets they have. Housing microfi nance programmes, as
administered by microfi nance institutions and shelter
advocacy groups, have recently emerged to address the
shelter needs of these groups and to fi ll the fi nancing gaps
not covered by traditional, more formal institutions.  e
target population with unmet demand for housing credit
from formal fi nancial institutions typically account for
the bottom 40 per cent to 70 per cent of national income
distribution bc.
Ferguson points out that in developing countries, the vast
majority of the population (80 to 90 per cent) do not qualify
for mortgage fi nance from formal fi nancial institutions to
purchase the least expensive, economically built housing
unit.  ey are left to build their own housing units without
formal fi nancial sector support, and must rely on piecemeal
or incremental fi nancing support from non-formal fi nancial
institutions. Many developing countries even have nothing
like a viable mortgage fi nance sector
bd
.
Providers of Housing
Microfinance Services
On aggregate, housing loans are still very small in low-
income developing countries – only three per cent of
outstanding credit, compared with 27 per cent in high-
income countries.  e experience of the microfi nance
industry in housing loans has been neither very lengthy
nor extensive. MFIs typically off er working capital loans
to business. Recently, MFIs have become more responsive
to the wide spectrum of poor people’s needs, including
housing fi nance. Some banks have also managed to reach
the bottom tier of the market, which includes those low-
income households who are willing to invest in home
improvement. In Tanzania household income ranges
between USD 50 and 500 per month. Loans for new
(often incremental) house construction are also off ered
by some MFIs, but general-purpose MFIs typically will
not allocate more than 10 per cent of their portfolios to
housing. Loans for new home purchases are normally not
available.  ose for land acquisition and infrastructure
are mostly available from NGOs and CBOsbe. A variety
of institutions are involved in diff erent types of housing
microfi nance.
Indeed, an increasing number of institutions are becoming
involved in housing microfi nance services, including
MFIs, NGOs, community-based organisations (CBOs),
and formal fi nancial institutions (Table 3.4). MFIs use a
wide range of mechanisms and techniques for providing
housing services.  ese are either “stand-alone” or “linked”
Table 3.3 Expansion of Micro nance Services
Year Number of Programmes
Reporting
Total Number of clients
served
Number of “poorest”
clients reported
12/31/97 618 institutions 13,478,797 7,600,000
12/31/98 925 institutions 20,938,899 12,221,918
12/31/99 1,065 institutions 23,555,689 13,779,872
12/31/00 1,567 institutions 30,681,107 19,327,451
12/31/01 2,186 institutions 54,932,235 26,878,332
12/31/02 2,572 institutions 67,606,080 41,594,778
12/31/03 2,931 institutions 80,868,343 54,785,433
Source: Daley-Harris S (2004)bb

13
The Challenges of A ordability, Accessibility and Sustainability
Table 3.4 Providers of Housing Micro nance Services
Types of Institutions Area of Focus Examples
MFIs Large-scale MFIs with over 100,000
clients;
Housing portfolio often borne
out of a disaster situation or as
diversifi cation;
May be a reward for successful
completion of a micro-enterprise
loan.
Medium-sized MFIs with 10,000-
100,000 clients;
Most have already achieved
best practice in microfi nance;
Similar principles are applied to
housing products (short term,
small amounts); Some have taken
government funds for expansion;
Commercial funding usually not
available for these loans, resulting in
funding mismatch
Grameen Bank
CALPIA (El Salvador, specialized
fi nance company), BancoSol (Bolivia,
bank), ADEMI (bank, Dominican
Republic), MiBanco (bank, Peru),
CARD Rural Bank (specialized bank,
Philippines)
NGOs and CBOs Capacity to transfer technologies
across affi liates in various countries;
Limited focus on technical assistance
for housing products; Currently
working on commercial funding
for conventional micro-enterprise
portfolios; Could leverage fi nancing
for housing; Some are direct lenders
and some are wholesale providers of
credit
Accion, CHF International, FINCA,
Homeless International
Co-operatives, Mutuals and
Municipals
Locally owned and often locally
started housing programmes; Good
experience and best practice; Usually
part of networks that enable cross-
experience sharing
Jesus Nazareno (S&L co-op,
Bolivia), Mutual La Primera (housing
co-op, Bolivia), Caja Arequipa
(municipal co-op, Peru)
Government Housing Programmes Some are professionally run; others
are very political and/or not market-
based; Major source of second tier
fi nancing for housing but with
limited outreach; Demonstrated
outreach to low income clients
Ex-FONVIS (Bolivia), FONAVIPO
(El Salvador)
Commercial Banks Some downscaling to housing faster
than to microcredit; Security is
a major issue; Have the capacity
to expand; Could mobilize large
amounts of commercial fi nancing
Banco de Desarrollo (Chile), African
Bank (South Africa)
Source: adapted from Escobar and Merrill (2004)bg

14 Housing for All
housing services. Stand-alone housing microfi nance does
not require clients to demonstrate a prior loan/repayment
history with the provider as a criterion for lending. Loans
are typically off ered to individuals rather than groups. Loan
sizes are small (USD 1,000 to 5,000) with short repayment
periods (1.5 to four years). No mortgage is involved since
loans are not collateralized by any property. Some MFIs
have introduced “linked” housing microfi nance services,
which set additional conditions for qualifying loans such
as participation in savings schemes.  erefore, “linked”
loans are secured to a degree.
Microfi nance institutions have grown rapidly to try to meet
huge demand for credit. Some have developed into ‘giants’,
with over a million clients each, including : Grameen Bank,
the Bangladesh Rural Advancement Committee (BRAC)
and the Association for Social Advancement (ASA) (in
Bangladesh), Bank Rakyat Indonesia (BRI), the Bank
for Agriculture and Agricultural Co-operatives (BAAC)
in  ailand, and the Sri Lanka National Savings Bank.
However, MFI outreach remains dwarfed by demand,
with only fewer than fi ve per cent of poor households
having access to microfi nance services.
Housing Microfinance by Commercial Banks
Accion has identifi ed several models for those commercial
banks intent on microfi nance business of a general nature:
(1) creating an internal unit within the bank; (2) creating a
fi nancial subsidiary; (3) creating a private service company;
(4) creation of new MFIs with bank co-investors . However,
formal fi nancial institutions are not keen on involvement
in housing microfi nance. In Asia, only four commercial
banks have a profi table microfi nance business: BRI in
Indonesia, together with three small private commercial
banks (Bank Dagang Bali in Indonesia, Hatton National
Bank in Sri Lanka, and Krishna Bhima Samruddhi Bank
in India). In addition to these, a 1997 USAID study found
eight commercial banks in Latin America and three banks
in Africa where microfi nance is a small but profi table
business. Altogether, in 1998 only 15 commercial banks
were engaged in profi table microfi nance business in the
world120.  e practice of commercial banks in housing
microfi nance is more limited, with only very few
commercial banks involved such as Banco de Desarrollo
in Chile, CashBank/BoE and African Bank in South
Africa121. Overall, commercial banks account for 78 per
cent of the total number of outstanding microcreditbk.
Housing Microfinance by Co-operatives
Co-operatives play an important role in the provision
of housing microfi nance.  ey enable low-income
households to save and borrow, where formal fi nancial
services are not available.  ey can provide signifi cant
linkages between banks, housing agencies and individual
members of low-income communities.  ey are more
responsive to the needs of low-income members. Terms
of credit are more attractive and lower than most market
rates. Although credit unions or savings and credit co-
operatives (SACCOs) do not aim to provide housing
microfi nance, the amount of loans used for housing has
increased rapidlybl.
One example is the National Co-operative Housing Union
(NACHU) in Kenya. It was established in 1978 to support
housing co-operatives through technical assistance and
training, and off ering housing microfi nance to co-operative
members. Its main eff orts are focused on low-income
housing co-operatives
bm
. On top of funding, NACHU
has also sought to address issues like land availability and
collateral through a combination of savings and lending
programmes with resettlement. While it is pursuing this
approach in several communities, the most advanced
project is to be found at Bellevue, a fi ve-acre community
west of Nairobi. Launched in 1994, the project involved the
resettlement of 184 households. NACHU lends USD 705
per quarter-acre plot, the interest rate is 15 per cent, and
the maximum loan term is four years. Unlike most housing
microfi nance schemes, NACHU retains land titles until all
members have repaid their share.  e NACHU experience
at Bellevue provides valuable insights into the way creative
housing fi nance can overcome obstacles relating to land
availability, access to basic services, and aff ordability
bn
.
NACHU runs three distinct types of housing microfi nance:
(1) housing rehabilitation/improvement loans; (2) new
house loans; and (3) resettlement and infrastructure loans.
Housing rehabilitation/improvement loans are small loans
averaging Kshs 50,000 each targeting the poor who cannot
aff ord to build or buy new housing units, but look for
opportunities for housing improvement.  e maturity is
four years with an annual interest rate of 19 per cent.  e
new house loans are designed for the poor co-op members
who have bought land and started construction of their
housing units in an incremental sort of way.  e loans
help them to expand the units. New house loans range
between Kshs 100,000 and 400,000.  e resettlement
and infrastructure loans target co-operative members who
live in slum areas and want to use their own resources to
relocate to areas with more secure tenurebo.
NACHU grants loans to co-operatives which on-lend the
monies to eligible applicants. Conversely, the co-operatives
pass repayment monies on to NACHU. But variations
may be allowed. For example, borrowers can pay either
at the co-operative offi ce, at the society bank or at the
predetermined location that suits them best.  e mode
of repayment is usually discussed and agreed with the
community in order to reduce the potential for defaultbp.

15
The Challenges of A ordability, Accessibility and Sustainability
Housing Microfinance by NGOs and CBOs
NGOs and CBOs are involved in many housing
microfi nance activities, including for instance Accion,
CHF International, Habitat for Humanity, FINCA and
Homeless International. Habitat for Humanity Tanzania
(HFHT) provides loans to families who live in inadequate
shelter with monthly incomes between Tshs 40,000
(or USD 39.00) and 120,000 (or USD 144.00). Loan
repayment periods range between seven and 10 yearsbq.
In Cambodia, most housing microfi nance schemes have
been launched by NGOs and CBOs, including for
example the Urban Sector Group (USG) and the Solidarity
for the Urban Poor Federation (SUPF). NGOs’ primary
services are savings schemes for land and housing, with
housing micro-credit as an aside. SUPF operates in half
of Phnom Penh’s 500 poor settlements through district-
based “Khan” units and women’s savings groups. On top
of using members’ savings to acquire land and housing,
SUPF has managed to raise awareness of land and housing
issues among one third of all squartter communities.  e
housing microfi nance loans of CBOs and NGOs are
granted to small groups, and most NGOs and CBOs in
Cambodia have resorted to this group guarantee as an
eff ective tool for loan recoverybr.
In 1998 and in co-operation with several other NGOs
such as the Asian Coalition for Housing Rights (ACHR),
Cambodia’s SUPF created another NGO, known as the
Urban Poor Development Fund (UPDF). UPDF aims
to make housing micro-credit aff ordable for poor urban
communities in order to enable them to improve housing
and settlements. It was also intended to act as a tool to
strengthen the Federation of community savings groups
through support of, or ‘topping up’, the fi nancial resources
required to scale up community-driven development
schemesbs.
As for CHF International, it is also a very active provider
of housing microcredit through programmes in various
countries. For example, CHF Romania has been off ering
individual and group home improvement loans since
1998. It makes individual housing loans available to fi rst-
time and repeat borrowers with secure monthly incomes
and good credit histories. Typical lending periods range
between three and 18 months, with 12 per cent monthly
interest rates plus a three per cent up-front commission.
Loans range from USD 50 to 450 for fi rst-time borrowers
and can reach USD 800 for repeat borrowers. Loan
amounts refl ect borrowers’ individual repayment capacity,
which is estimated at 25 per cent of the household monthly
income. Security against loans is secured through recourse
to co-signers, whose mailing addresses must not be the
same as applicants’.bt
CHF Romania also off ers housing microfi nance loans to
groups.  ese group loans allow securitization through co-
signers or through a mortgage instrument for the largest
principal amount available. Maximum loan size is USD
750 per person in a condominium homeowners group and
the maximum lending period is three years.  e monthly
interest rate is 16 per cent and loan commissions range
between two and four per cent, depending on loan sizes.
However, housing microfi nance group loans face four
main operational diffi culties, including: (1) wide ranges of
incomes and repayment capacities within target groups; (2)
the challenge of reaching consensus among condominium
residents; (3) the limited number of collateral options
available to them; and (4) the high overhead costs of
putting together such group loans.bu
Housing Microfinance by MFIs
If housing loans are small with short maturities, MFIs
may be the ideal providers.  e amounts and maturities of
housing microfi nance loans are usually signifi cantly lower
than those of mortgage loans. Interest rates on housing
microfi nance are closer to those on microbusiness loans
than on conventional mortgages. Clients are mostly poor
households, with whom MFIs are used to deal. MFIs
have developed specifi c lending methodologies to reduce
the risk of low repayment rates, and many have achieved
signifi cant results such as BancoSol, FIE, Caja de los Andes
and Crecer in Bolivia; Compartmos in Mexico; Genesis
Empresarial and SIFFE Credit Unions in Guatemala;
Banco Ademi in the Dominican Republic; Financiera
Calpia in El Salvador; and Caja Social in Colombiabv.
FINCOMUN is a Mexican MFI with 17,200 clients in low-
income Mexico City neighbourhoods and an outstanding
loan portfolio of more than three million US dollars.
FINCOMUN also provides housing microfi nance loans
as an extension of existing enterprise loans; loan sizes range
from USD 500 to 1,000, with weekly instalments and a
six per cent monthly interest rate calculated on a declining
balance basis. In addition, FINCOMUN estimates that
10 to 15 per cent of business loans is diverted to housing
improvements related to business operationsbw.
Housing Microfinance by Developers
With proper incentives, some developers off er housing
microfi nance services.
Funding Strategies of
Housing Microfinance
Savings
Traditionally, savings has been a primary source of funds
for fi nancial institutions. In the USA, 97 per cent of

16 Housing for All
total commercial bank liabilities are deposits of various
maturities which fund the banks’ lending operations.  e
fi nancial success of many housing microfi nance providers,
including Bancosol, BRAC, BRI and EBS, depends largely
on their ability to raise savings (Table 3.5). However,
many MFIs are not allowed to take deposits until they can
meet certain minimum capital requirement mandated by
regulatory authorities. Many MFIs are small and unable to
meet these criteria, and those that can often fi nd themselves
falling far short of loan demandbx. Table 3.5 shows the
savings and loans portfolios of selected microfi nance
providers. Average deposits amount to USD 32,395,791
against an average loan portfolio of USD 57,339,910.  e
number of borrowers are more than twice that of deposits.
Most microfi nance lenders cannot rely on client deposits
alone and must rely on additional, alternative sources to
meet the demand for loans.
Donors, Governments and
International Institutions
Donors play a very important role in promoting and
funding microfi nance programmes. For most MFIs, the
principal source of funding remains grants and highly
subsidized loans. Grants and loans are mainly provided
by international donors (e.g. Sweden’s SIDA, the UK’s
DFID, USAID), multilateral banks (e.g. World Bank,
Inter-American Development Bank) foundations (e.g.
Ford Foundation), and apex organizations (e.g. Women’s
World Banking, ACCION, FINCA) . For example, South
African NGO People’s Dialogue is 95 per cent funded by
international donors and the remaining fi ve per cent by
the South African governmentbz. International grants and
loans typically include conditions and requirements, and
they are limited in amounts.
While donors are valuable sources of funds, they are
not the most desirable ones for MFIs as they are neither
suffi cient nor sustainable. Donor pritorities and focuses
can be versatile and their support is often conditional and
temporary, not to mention lack of knowledgeable staff .
From a business perspective, this segment is not growing.
For example, the United States foundation giving is down
as much as 20 per cent since 2000ca.
Access to Commercial Banks through
Loan Guarantee Programmes
 e format developed for the Rajiv Indira Suryodaya
project in Mumbai, India, demonstrates how NGOs
can access funds from commercial banks to leverage
Table 3.5 Selected Micro nance Providers (December 2003)
Member Country Number of
Active Borrowers
Gross Loan
Portfolio (USD)
Number of
Depositors
Deposits (USD)
ASA Bangladesh 2,130,000 166,500,000 2,330,000 16,200,000
Banco del
Desarrollo
Chile 33,500 31,500,000 18,100 8,500,000
BancoSol Bolivia 56,700 91,200,000 53,300 70,100,000
BRAC Bangladesh 3,400,000 190,900,000 4,100,000 104,400,000
BRI Indonesia 3,100,400 1,717,700,000 29,869,200 3,244,900,000
CERUDEB Uganda 44,800 34,900,000 397,800 59,400,000
Citi Savings Ghana 1,100 3,000,000 34,000 5,100,000
EBS Kenya 67,000 21,800,000 252,000 42,200,000
FINAMERICA Colombia 20,700 18,300,000 1,100 11,900,000
Kafo Jiginew Mali 52,700 11,000,000 120,300 12,200,000
K-REP Kenya 45,400 20,500,000 17,300 4,300,000
Los Andes Bolivia 49,700 80,200,000 39,000 48,500,000
Mibanco Peru 116,700 113,900,000 36,600 44,900,000
PRODEM Bolivia 25,300 60,500,000 65,900 48,600,000
UMU Uganda 28,600 6,300,000 35,600 700,000
XAC Bank Mongolia 18,535 9,598,647 25,666 8,936,865
Average 574,446 57,339,910 244,762 32,395,791
Source: based on Microfi nance Network

17
The Challenges of A ordability, Accessibility and Sustainability
the security provided. In 1997 and under the new
Slum Rehabilitation Authority (SRA), SPARC-Nirman
embarked on the fi rst-ever attempt by an NGO in India
to construct apartment buildings for 209 households
living in a slum in Mumbai.  is slum development
scheme was established by the government of the State
of Maharashtra in 1995 to provide incentives for private
builders to construct free tenements for 800,000 poor
households. However, the SRA scheme did not take off ,
and only a fraction of the anticipated tenements have
been built to date. In light of the poor performance of
the SRA, the SPARC-Nirman alternative to commercial
builder-driven housing is an important step. Five
apartment blocks will be built and each tenement will
be a minimum of 225 square feet.  ree buildings will
house community members and the other two will
be sold on the market to make up costs and generate
profi ts
cb
.
Apart from using its own revolving funds, SPARC-
Nirman took a loan for this project from Citibank and
the UK-based NGO Homeless International acted as
a guarantor for 20 per cent of the principal. Nirman
expects to recover its costs once all the buildings are
completed and the fl ats and commercial premises are
sold.  e costs are expected to be met from the sale
of transferable land development rights (34 per cent),
apartments (51 per cent), and commercial spaces (15
per cent). Backed by collateral provided by the Homeless
International Guarantee Fund (established in 1994),
organised groups of poor urban residents can negotiate
more equitably with banks for the fi nancial services
they need.  e fund has had some support from the UK
Department for International Development (DFID)
and from the European Union. However, the bulk of
funding came from non-governmental sources, such
as deposits by the Airways Charitable Trust (Figure
3.1)
cc
.
A variety of funding strategies are available. However,
MFIs fi nd that any single funding source is generally
insuffi cient.  ey often use mixed funding sources
to meet demand for loans. Table 3.6 shows the most
common funding strategies adopted by microfi nance
providers.
The Characteristics of Housing
Microfinance Products
Criteria for Housing Microfinance
Lending criteria for housing microfi nance are mainly
based on repayment capacities.  e home is the most
Figure 3.1 The Homeless International Guarantee Fund
The Guarantee for the Rajiv Indira Suryodaya Project
Source: CGAP (2004)cd

18 Housing for All
important asset for poor people and something to which
they are strongly attached.  erefore, they are prepared
to spend a high per centage of their income toward
building, expanding and maintaining their homes. Most
lending institutions recommend that 20 to 35 per cent
of household income can be spent on housing loans.
For example, the percentage of household income used
for repayment is 25 per cent for CHF in Gaza, and 33
per cent for FUNDHAVI
cf
.
Loan Size
 e size of housing microcredit loans is usually small,
between USD 300 and USD 5,000. Sizes are smaller
in Asia than in other regions, typically a few hundred
US dollars, compared with up to USD 5,000 in South
America. For middle-income countries, loan sizes are
accordingly higher. In South Africa, housing microcredit
loans can be as high as USD 8,000 (Table 3.7).
Loan Terms
Loan maturities vary according to purpose. Home
improvement loans are for two months to two years. Most
land purchase or construction loans range from two to fi ve
years. Some housing microfi nance maturities are almost as
long as those of mortgage fi nance. For example, Banco de
Desarrollo in Chile and People’s Dialogue in South Africa
both off er loan terms of 15 years. Table 3.8 illustrates the
range of terms currently off ered by housing microfi nance
providers.
Interest Rates
Interest rates on housing microfi nance loans are determined
by several factors such as cost of funds, transaction
costs, r
isks, and aff ordability for clients. Interest rates are
normally higher than those charged by commercial banks.
However, they may be lower or higher than for micro-
enterprise loans, or the same. For example, for housing
Table 3.7 Loan Size of Selected Housing Micro nance Providers
Institution Country Average/Maximum Loan Size USD
Diaconia Bolivia 800
FUNHAVI Mexico 1,400
ADEMI Dominican Republic 5,000
BancoSol Ecuador 1,095
Grameen Bangladesh 600
SEWA India 300
CARD Philippines 350
People’s Dialogue South Africa 1,200
CashBank South Africa 8,000
African Bank South Africa 2,500
Source: Escobar & Merrill (2004)cg
Table 3.6 Funding Strategies of Selected Micro nance Providers
Strategies SPARC CARD BRI Grameen Bank People’s
Dialogue
Mandatory Savings Yes Yes No Yes Yes
Bank Funding No Yes Yes Yes No
Donor Funds Yes Yes No Yes Yes
Foundation Funds Yes Yes No Yes No
Public Funds Yes No No No Yes
Credit Enhancement Yes No No No No
International Investors No No No Yes No
Source: Escobar and Merrill (2004)ce

19
The Challenges of A ordability, Accessibility and Sustainability
microfi nance programmes such as Calpia, Grameen and
SEWA, interest rates are lower than for micro-enterprises;
for FUNHAVI, the interest rate is higher than that for
microenterprises; CARD charges the same interest rates for
housing and micro-enterprises (Table 3.9). In most cases,
the interest rates for housing and micro-enterprises range
between 20 and 50 per cent. Grameen charges much lower
and subsidized interest rates (eight per cent) for housing,
which is not common in the microfi nance industry. High
risks and signifi cant transaction costs make low interest
rates commercially less
viable and sustainable from the
microfi nance institutions’ perspective.
 e eff ective rate of credit is often higher than nominal
interest rates because of fees and other charges.  e
eff ective rate captures all fi nancial costs related to the loan.
Borrowers pay not just the nominal rate on the principal
and additional fees, but also forego the interest they would
earn if they were not forced to keep the required balance
on their account. For example, Union Popular Credit
Union in Guatemala requires members to have credit
access-linked share accounts to qualify for loans, and these
accounts carry lower interest rates than passbook deposits.
A borrower will lose 10 per cent on interest earnings for
keeping money in share accounts in order to secure access
to credit .
MFIs widely adopt fi xed interest rates for their housing
microcredit.  e short period of loans and higher
interest rates of housing microcredit will reduce the risk
Table 3.8 Loan Terms of Selected Housing Micro nance Providers
Institution Country Term
ADEMI Dominican Republic 12-36 months
Calpia El Salvador Up to 60 months
MiBanco Peru Up to 120 months
FUNHAVI Mexico 18 months
CHF/Gaza Gaza 36 months
Genesis Guatemala Average of 30 months
BancoSol Bolivia Average of 80 months
CARD Philippines 12 months
Grameen (basic) Bangladesh 120 months
SEWA India 60 months
Banco de Desarrollo Chile 180 months
People’s Dialogue South Africa 180 months
CashBank South Africa 180 months
Source: Daphnis (2004)ch and Escobar & Merrill (2004)ci
Table 3.9 Interest Rates of Housing Micro nance and Micro nance Business Loans
Institutions Housing Microfi nance Microfi nance for Micro-enterprises
ADEMI N.A. 18 per cent - 24 per cent
Calpia 23 per cent 32 per cent
MiBanco 45 per cent (MiCasa) 30 per cent + (Mi Capital)
FUNHAVI 58 per cent N.A.
Genesis 25 per cent 35 per cent
BancoSol 16 per cent - 22 per cent 32 per cent
CARD 20 per cent 20 per cent
Grameen 8 per cent (subsidized) 20 per cent
SEWA 14 per cent 17 per cent
Source: Daphnis (2004)ck

20 Housing for All
of macroeconomic fl uctuations to lenders. Low-income
borrowers are also liken on fi xed interest rates because of
its clarity of periodic payment amountcl.
Security and Collateral Requirements
Many housing micro-loans are unsecured
cm
. However,
housing microfi nance providers increasingly seek
to make security arrangements for housing loans in
order to reduce risk. Housing micro-loans are often
not collateralized. Since many developing countries
have no adequate legal system to support the use of
collateral to secure a loan, it is very diffi cult to liquidate
the collateralized asset to repay the loan balance in
case the borrower defaults.  erefore, many housing
microfi nance providers do not choose to collateralize the
loans. Instead, they resort to co-signers as an alternative
type of security for housing microcredit
cn
.
Most MFIs use co-signers as security for lending. For
example, 71 per cent of loans by Union Popular use co-
signers guarantee. Only fi ve per cent of micro-loans use
a mortgage as security; two per cent use both co-signers
and a mortgage; 16 per cent use savings as securityco. In
Mexico, FUNHAVI off ers housing microcredit ranging
from USD 800 to 2,000 for incremental housing
construction/improvement, such as addition of an extra
room, roof replacement or basic sanitation, with repayment
periods of one to three years. FUNHAVI uses co-signers
as security for loans. Co-signers must feature the same
repayment capacity standards and eligibility requirements
as the borrower, and cannot be members of the borrower’s
immediate familycp. Some housing microfi nance providers
such as SEWA put more emphasis on savings as security;
other institutions, like SPARC, CARD and Genesis, use
group solidarity for housing and infrastructure micro-
loanscq.
Table 3.10 shows the common security measures adopted
by housing microfi nance providers. Most housing
microfi nance providers use co-signers as security. Some
use mortgages. Linked housing microfi nance providers use
both co-signers and savings as security. Since the amount
of loans is small, mortgage is not often used.  e use of co-
signers should be adequate. Co-signers should be credit-
worthy and have resources which can be clearly accessed
by lenders in case clients defaultcr. Since housing micro-
loans are small, short-term and charged with high interest
rates to compensate the possible high risks, the security
for the loans are far less important, compared to long-
term housing fi nance.  erefore, housing microfi nance is
more based on character and willingness to pay than on
collateral.  e strategies housing microfi nance currently
uses to secure loans including: (1) personal guarantee (co-
signers); (2) land title and buildings; (3) mortgage/lien on
assets; (4) assignment of future income (wages); (5) joint
liability and group guarantees (character-based lending);
(6) other fi nancial assets (for example, life insurance
policies and pension funds)cs.
Underwriting
Almost all housing loans are made on an individual basis.
People’s attachment to their homes may make them more
likely to repay housing loans. Many MFIs have developed
a well-functioning individual loan appraisal methodology
that measures borrower repayment capacity before issuing
housing loans. Given the longer loan terms and higher
amounts, some MFIs that off er housing in addition to
other products charge lower interest rates on housing
loans. As a result, some clients may be tempted to apply
for a housing loan to take advantage of a lower cost, and
then apply the funds to another purpose. To avoid the
diversion of housing loans, MFIs can write into the loan
contract that a higher interest rate will be charged if the
loan is not used for housingcu.
Table 3.10 Security Requirements of Selected Housing Micro nance Providers
Institution Land Ownership Requirement Security Requirements
Calpia Yes Mortgage on 59 per cent of housing
loans
FUNHAVI No 2 co-signers
CHF/Gaza Yes 2 co-signers
Genesis Yes Group guarantee and land title
BancoSol Yes Mortgage and ownership title
CARD Yes 5 co-signers; borrower’s savings
Grameen Yes 5 co-signers; borrower’s savings
SEWA No 2 co-signers; borrower’s savings
Source: based on Daphnis (2004)ct

21
The Challenges of A ordability, Accessibility and Sustainability
In housing microfi nance, there is a wide variation in the
ratio of loan to value. Normally it ranges from 40 per cent
to 80 per cent.  e repayment does not exceed 25 per cent
of household income. Some MFIs require previous loan
history with lending institutions, or mandatory savings.
For salaried borrowers, it requires a minimum employment
period and a notarized asset list. Payroll deductions are used
as a payment method for salaried workers, which are common
in South Africa. Savings accounts are another method to
make payments. Mandatory savings are sometimes used
as alternative collateral. Many experienced MFIs require
ownership of land or housing units, even if the loan is not
secured by a mortgage.  ey are even providing technical
assistance to clients in securing titles, for example, BancoSol.
Co-operative Jesus Nazareno keeps the title documents, not
as collateral but as pressure for repayment
cv
.
For micro-entrepreneurs, it requires minimum experience,
proof of ownership and a notarized asset list. MFIs
often provide mandatory training and counseling, and
assistance in securing titles and building technology.  ey
also adopt aggressive serving such as frequent visits and
embarrassment tacticscw.
Delivery Process of Housing Microfinance
Housing microfi nance programmemes attempt to ease
the delivery process. FUNHAVI in Mexico takes 11 main
steps and about an average 10 days from clients’ initial
visit to the actual disbursal of the loan. Most (75 per
cent) of new clients are referrals. FUNHAVI distributes
promotion materials at grocery stores and construction-
material suppliers. Clients wanting a loan FUNAHVI
must go to their offi ce for an initial interview, during
Figure 3.2 Example of Linking Micro nance to a Formal Finance System
Source: based on USAID (2000)db

22 Housing for All
which the potential clients will be briefed about the
programme and eligibility requirements and application
procedures. If applicants meet the criteria for loans, they
need to go to FUNHAVI to deposit a fee for an architect’s
visit to assess the housing improvement needs as well as
assessing the existing housing structure and its quality,
and collects other information about applicants such as
household size, income and housing tenure.  is helps
to determine the loan size and payment duration.  e
architect meets with the applicants at FUNHAVI offi ce
to review and fi nalise housing improvement design,
budget and a list for required construction materials.
 is housing improvement proposal was then referred
to the chief architect for review, who will forward
the application to the credit department.  e credit
department makes credit analysis and further forward
the loan application to the Executive Director for fi nal
review and approval. Upon the Executive Director’s
approval, the credit department meets with the applicant
and the co-signer in which they sign the loan agreement
and promissory note.  e applicants receive instructions
on the loan disbursement process, repayment schedules,
and other technical assistance information
cx
.
FUNHAVI has developed relationships with the
construction industry, particularly the material
suppliers such as Cementos de Chihuahua and local
hardware stores. FUNHAVI’s cooperation with the
construction industry ensures its clients to be able to
access the quality of materials and reduce the costs
to its clients. FUNHAVI also develops a relationship
with the S-MART supermarket group to collect
clients’ monthly loan repayments.  e S-MART
supermarket chain operating 24 hours a day provides
great convenience for clients to make payments.  is
arrangement allows FUNHAVI to streamline the
payment process and reduce operating costs. S-MART
provides the services for free on the consideration that
this arrangement will generate consumer fl ows and
enhance customer loyalty
cy
.
Linking Microfinance to Formal Finance Systems
One promising example of linking microfi nance to a
formal fi nance system is the Guinea Trust. It works through
Ecobank which has identifi d microfi nance lending as one
of the expanding opportunities.  e bank serves as the
fi nancial intermediary for PRIDE microfi nance project,
and allows itself to access one of their targeted markets, but
leaves the intensive groundwork of managing thousands of
microfi nance clients to the experts at PRIDE.  e Guinea
Trust serves as a bridge.  e fi rst tranche of US$2.1
million fi lls an immediate loan capital shortage of PRIDE.
But PRIDE needs to access commercial capital to fi nance
growth in its loan portfolio. A catalyst was needed to
permit PRIDE/Finance access to Guinean capital markets.
 e Loan Guarantee Fund programme with an additional
US$2.6 million in authorized but unallocated funding is
an instrument to provide PRIDE access to commercial
capitalcz.
With the goal of achieving sustainable results, a unique
link was forged with Riggs Bank, N.A. in Washington,
D.C. in order to facilitate PRIDE’s access to private
sources of commercial funding. Riggs agreed to establish
a US$2.6 million trust fund, the “Guinea Trust,” that
would leverage additional loan capital in future funding
for PRIDE by collateralizing standby letters of credit (L/
Cs) in favor of commercial banks such as Ecobank.  e
L/Cs would backstop domestic money market borrowings
by PRIDE from the commercial banks, functioning as a
form of commercial paper stand-by line.  is serves two
purposes: 1) it allows PRIDE market entry for domestic
fi nancing; and 2) it acts as a catalyst for the development
of the Guinean money market. With the loan guarantee
fund in place, PRIDE can negotiate a loan with any local
commercial bank, provided that PRIDE continues to meet
performance-based standards (see Figure 3.2)da.
Linking Housing Microfinance
to Technical Assistance
Housing microfi nance programmes often have components
of technical assistance or develop cooperation with
organizations which provide technical assistance. ITDG
works very hard to bring down the building costs in order
to benefi t a wider range of housing microfi nance users
in terms of their improved housing aff ordability. ITDG
develops and identifi es low cost technology that can be
easily implemented by poor households. For example, in
Kenya, ITDG promotes two low cost building technologies
(i.e. Stabilised Soil Blocks (SSBs) and Ferro-Cement (FC)
construction.  ese approaches reduce the use of high
cost inputs such as cement and stone, and use locally
available cheap or free materials.  e use of these two
low cost technologies leads to the signifi cant reduction of
construction cost and therefore improves the aff ordability
for the poor (Table 3.11).
Variations in Performance of MFIs
Latin American MFIs have achieved greater fi nancial scale
than other regions.  ey have larger average balances per
borrower but reach fewer active borrowers than Asia and
African MFIs.  e average higher loan size is partly due
to the high GDP per capita level in Latin America than
any other region. Latin America MFIs are more successful
in leveraging their equity and access commercial fi nancing
than in other regions.  e average leverage ratio of MFIs
are 2.7 for Latin America, 2 for Africa and 1.6 for Asia.
 e regulatory environments in many Latin American

23
The Challenges of A ordability, Accessibility and Sustainability
countries make MFIs more easier to mobilize savings
in Latin America than in other regions. MFIs in Latin
America pay the highest average cost for fi nancing as per
centage of total assets and are less profi table than in other
regions. In terms of scale, larger MFIs are more effi cient
than smaller ones. Most large and medium-sized MFIs are
fi nancially self-suffi cient. But many small MFIs are not
fi nancially self-suffi cientdd.
Limit of Housing Microfinance
Inadequate Financial Infrastructure
One of the limitations for the development of the housing
microfi nance industry is lack of fi nancial infrastructure
(i.e. legal, information, supervision and regulation). Most
governments have focused on creating institutions or
special programmemes to disburse funds to the poor, but
pay little attention to building the fi nancial infrastructure
that supports, strengthens and ensures the development
and sustainability of the microfi nance industry. Lack
of a legal framework conducive for the emergence and
sustainable growth of small-scale microfi nance institutions
and corresponding supervisory and regulatory systems have
impeded the development of market-based microfi nance
services and limited their access to commercial sources of
fundingde.
High Default Rate
 e most dangerous problem a microcredit programmeme
has to face is the problem of the default of repayment.
If the money invested by the organisations cannot be
recovered, then the whole programmeme might cease to
operate. For this reason, many MFIs (particularly small
ones) are not sustainable. In Bangladesh, a signifi cant
number of organisations could not sustain their operation
due to high default rate of their clients. For example, the
default rates of some NGOs in 1996 were: the Comilla
Proshika Centre for Development (CPCD) - 20.75 per
cent, the Adarsha Samaj Seba Samity (ASS) - 40 per cent,
the Come To Work (CTW) - 45.55 per cent, the Palli
Kallyan Sangstha (PKS) - 50 per cent, the Mukti Shikha
- 62 per cent, the Rural Development Society (RDS) - 82
per cent, the Organisation for Distress People - 84 per
cent, the New Earth Concern (NEC) - 91.67 per cent.
 e increased rate of default, if cannot be checked, can
even jeopardise the whole programmeme. Although the
Association for Social Advances (ASA)’s repayment rate
is satisfactorily high, sometimes it has to face untold
suff erings due to increase in the rate of default in some
regions. In Norshingdi district, ASA’s overdue loan was
370,451 in September 1995 and it rose to 481,464
in February 1996; in Netrokona it was Tk. 70,919 in
September 1995, but rose to 260,489 in February 1996
and in Manikgonj it was 9,772 in September 1995, but
became 663,994 in February 1996. In 1994 the Habigonj
region of ASA was the worst aff ected. Microfi n Caribbean
Holdings Ltd in the Caribbean generates a loss of US$
3.91 million after amortization of pre-operating expenses
of US$ 216,680, depreciation cost of US$ 330,533 and
loan loss provisions of US$ 2.59 milliondf. Although we do
not have data on the default rate of housing microcredit
alone, the above information describes a serious challenge
facing all microcredit programmesdg.
When microfi nance is provided to the poor at a very high
price, it increases clients’ indebtedness and re-enforces a
vicious cycle of poverty and increases the likelihood of
default.
Table 3.11 Impact of Low Cost Construction Technologies on A ordability in Kenya
Type of Construction Construction Cost Monthly Disposable
Income Required to
Save in 5 Years
Monthly Disposable
Income Required to
Repay in 5 Years (20
per cent annual rate)
Without Low Cost Technologies
1 room stone unit, no services US$ 1,090 US$ 17.71 US$ 28.88
Stone basic unit with 2 rooms and
service connections
US$ 2,564 US$ 41.66 US$ 67.93
With Low Cost Technologies
1 room stone unit, no services US$ 545 US$ 8.85 US$ 14.44
Stone basic unit with 2 rooms and
service connections
US$ 1,538 US$ 24.99 US$ 40.75
Source: Cities Alliance (2002)dc

24 Housing for All
High Costs of Operation
 e total cost of providing funding directly to MFI is
usually very high.  is is particularly true when the cost
of feasibility studies, appraisal missions, monitoring,
administration, evaluation, reporting (by both the donor
and the MFI), and so on are included.  is is even more
the case when the funder does not have a permanent offi ce
in the country where the MFI is located, where the donor’s
local offi ce does not have decision-making authority, or
where the local offi ce is located in a capital city far from
the MFI.  e high costs of personnel (head offi ce staff ,
local staff with all their allowances, and consultants) and
the slow-moving nature of aid bureaucracies, which until
recently were not familiar with the unique nature and
needs of microcredit programmes (as compared to more
traditional give-away or relief-type programmes), make for
a costly and slow-moving processdh.
A study by BRBD shows that the microfi nance delivery
cost accounts for 20 per cent to 62 per cent of the loan
amount.  e main factors aff ecting delivery cost are the
cost of operation, cost of fund, and overhead cost in
relation to the total amount of loans disbursed.  e higher
the loan amount disbursed the lower the delivery cost.
 is shows the scale of economy (Table 3.12).  e existing
capacity of RD-9 and RPCP projects is under utilized.
 e RD-9 project also has separate group organizers for
social development and mobilization activities such as
group formation, group management and other social
development activities.  ese are not included in the
estimated staff costdi.
Hard to Reach the Poorest
As a result of the high costs involved in providing funds
directly to MFIs, as well as the high costs incurred by many
MFIs while providing a portion of the funds received as
microcredit, a relatively small amount of these funds are
actually provided as loans to the poorest. It is estimated
that between 10 per cent and 25 per cent of the funds
actually reach the poorest. In the best cases, the funds not
only reach the poorest once, but also reach them again
and again, as they re-circulate through the loan fund every
year, or sometimes as often as every quarterdk.
 ere is an increasing awareness of the failure of the
microfi nance industry to reach the poorest. CGAP noted
that most microfi nance clients today fall in a band around
the poverty line and the extremely poor are rarely reached
by microfi nance. Reaching the poorest is much more
expensive than reaching the poor.  e poorest require
smaller loan amounts and a lender has to make more loans
to achieve the same volume level.  is will require more
staffi ng time and increase the cost.  e risks for lending to
the poorest are higher than for the poor.  e poorest is more
vulnerable in terms of income, health and opportunities.
 erefore the cost of lending to the poorest is much higher
than to the poor, but the lenders can not charge too high
prices to cover their additional costs to lending the poorest
since the aff ordability level of the poorest is lower than
the poordl. Lack of institutional capacity at the retail level
to expand the scope and outreach of the microfi nance
services also limits many microfi nance institutions to cost-
eff ectively reach the poorest of the poordm.
Table 3.12 Delivery Cost of per TK. 100 Disbursed Loans for Selected Projects in Bangladesh
Project Name Cost of Operation Cost of Fund Overhead cost Total
RD-5 10 9 1 20
RD-9 43 10 9 62
RD-12 9 12 1 22
RPCP 43 12 6 61
WP 26 11 2 39
RPAP 20 9 1 30
Source: BRBD (1997)dj

25
The Challenges of A ordability, Accessibility and Sustainability
Small Loan Size
In the case of established MFIs, demand for housing
microfi nance typically involves loan amounts - often in the
$300 to $5,000 range - that are higher than the demand
for low-end microfi nance loans. Unless the provider has
experience with high-value individual loans, fulfi lling a
demand for housing loans may entail altering the MFI’s
worldview on what constitutes an allowable repayment
period. For instance, MFIs that target poor and very poor
clients often develop group-based methods of lending that
feature repayment periods of one year or less.  e short
repayment period is in direct relation to the default risk
the MFI associates with its clients.  e poorer the client,
the more likely the MFI will attempt to manage default
risks by reducing the time over which the client must
repay the loan. As clients successfully complete lending
cycles, loan amounts may increase and repayment periods
may expand. MFIs that use a short repayment period
to minimize credit risk may have to consider signifi cant
increases to the allowable repayment time for housing
loans.
High Interest Rates
People with access to housing microfi nance often do not
need to qualify for the conventional loan requirement. In
addition, the microfi nance has far higher transaction costs
and risks, therefore, the interest rates charged on housing
micro-loans are much higher than conventional loans.
 is means the poor people need to pay a higher price
for access to housing microcredit while their aff ordability
level is lower. Sometimes, the interest rate of housing
microfi nance is as high as a couple of times that charged
for conventional mortgage loans.  is high price makes the
housing aff ordability of the poor further deteriorating.
Short-Term
 e loan term of housing microfi nance is normally very
short. For the same amount of loan with the certain
interest rate, the shorter the term, the higher the monthly
repayment.  erefore the short-term can increase the
signifi cant challenge to the aff ordability issue. For a loan
of US$ 1,000 at the annual interest rate of 15 per cent, if
the term is 15 years, the monthly repayment is US$ 14;
if the term is 1 year, the monthly repayment suddenly
increases to US$ 90.3; if the term reduces to 6 months, the
monthly repayment increases to US$ 174.  e monthly
repayment for a 6 month loan of the same size is 5.3 times
the repayment for a 15 year term.  erefore, we can see
the tremendous challenges the short term puts on the
aff ordability of the poor.

26 Housing for All
The Difficulties of the
Conventional Formal Sector in
Meeting the Housing Demand
 ere are severe housing shortages in many developing
countries.  e rapid urbanization and increasing housing
demand often surplus the pace of housing supply. Housing
supply by the formal sector cannot even meet the new
increased demand each year.  is leads to the increased
housing shortage and the growth of informal housing
sector. Although we do not have updated data on the
housing situation in many developing countries, Table 4.1
shows the typical situations in developing countries where
most countries faced increased housing shortages. Most
housing demand is met by the informal sector. While some
countries like Costa Rica show positive signs of reducing
housing shortage and informal housing settlements.
Housing Co-operatives
Limited Equity Housing Co-operatives
 e Limited Equity Housing Co-operative (LEHC) is
a form of group home ownership that provides benefi ts
to individual home ownership. Individual members of
LEHC owns stock or shares in the co-operative. Each
member has the right to occupy an individual housing
unit.  e deed of the whole project remains with the
LEHC co-operative. Members do not have the deeds for
their individual units. Members enjoy the property tax
exemption for home owners.  ey can also deduct their
share of the co-operative’s mortgage interest and property
taxes on their income tax returns. But the co-operative
has its restrictions on the resale value of its memberships
or shares. Individuals typically pay 5 per cent to 10 per
cent of the market value of the individual units for their
memberships. For a unit of US$ 20,000, the membership
would sell for US$ 1,000 to US$ 2,000.  e co-operative
would borrow the balance by loans that are secured by
the whole project, while members pay monthly mortgage
interest and property taxes for their share dp.
Role of Limited Equity Housing Co-
operatives in Affordable Housing
Limited equity co-operatives are much more aff ordable
over the long run than either market rate rental housing
or condominiums because resale prices are restricted
which result in low carrying charges (the equivalent of
mortgage payments). For example, the median monthly
membership charges in 2003 were just about half of the
Chapter 4: The Co-operative Housing Approaches to
Low Income Housing
Table 4.1 Mismatch of Housing Demand and Supply in Selected Developing Countries
Country Estimated
Accumulated
Housing Shortages
Increase in Housing
Demand (units/year)
Housing by Formal
Sector (units/year)
Estimated Housing
– Informal Sector
Nicaragua 520,000 (1979) 20,000 (1979) 1,100 (1958-1978) 80 per cent
Mexico 8,000,000 (1989) 700,000 (1989) 360,000 (1989) 65 per cent
Guatemala 840,000 (1991) 56,500 (1991) 13,000 (1988) 65 per cent
Cuba 813,000 (1993) 49,000 (1993) 17,300 (1990-1993) 35 per cent
Panama 240,000 (1990) 20,000 (1990) 6,500 (1086-1988) 65 per cent
Costa Rica 265,510 (1993) 25,000 (1992) 34,500 (1993) -
El Salvador 573,676 (1983) 15,000 (1983) 21,800 (1983) 63 per cent
Source: based on Landaeta (1994)do

27
The Challenges of A ordability, Accessibility and Sustainability
central government’s fair market rental rate in the District
of Columbia in USA (Table 4.2). A recent survey of
housing markets in the District of Columbia shows that a
household would need an annual income of US$ 115,000
to be able to aff ord purchase a media-priced 2-bedroom
condominium housing unit costing US$ 352,500 and an
income of US$ 74,000 to aff ord to rent a median priced
2-bedroom apartment at US$1,859 per month. However,
a household with an annual income of only US$ 23,740
could aff ord the current median monthly carrying charge
of US$ 587 for a 2-bedroom limited-equity co-operative
unit (Table 4.3).
Community Land Trust
 e community land trust (CLT) model was developed in
the 1960s by community activists. It is a democratically
structured, community based non-profi t corporation,
designed to achieve a fair balance between individual and
community interests and to meet the strategic requirements
for a new approach to address land and housing problems
to achieve aff ordable housing. What CLT does is to acquire
land and remove it from the speculative, profi t-oriented
market.  e land is made available to individual families,
co-operatives, and/or other non-profi t organizations
through renewable long-term leasehold (typically 99
years). All lessees are members of the CLT, and they are
represented on its Boardds.
Land assigned to individuals can be inherited by the
leaseholders. Most CLTs have limited equity policies
and formulas that restrict the resale price of housing
units in order to maintain the long term aff ordability.
Communities increasingly recognize the versatility and
value of CLTs, they saw the rapid growth of CLTs as one of
aff ordable housing models. In most cases, CLTs have been
formed as a grass-roots response to specifi c local needs for
low income people and communities. Many CLTs have
been established to ensure access to land and housing
for low income people.  ey often focus on increasing
homeownershipdt.
 e Case of Lopez Community Land Trust
 e Lopez Community Land Trust (LCLT) is a non-profi t
aff ordable housing and rural development organization in
San Juan Country, Washington State in USA. 13 per cent
of the population in Lopez lived in poverty in 1998.  e
purpose of LCLT is to act as a resource for low income
households by providing access to aff ordable housing and
land and by cultivating sustainable economic development,
to enhance the entire community in terms of housing
and economic opportunitiesdu. Figure 4.1 depicts the
organizational structure of the Lopez CLT.
In response to the need for aff ordable housing, LCLT
has established three low-income single-family housing
co-operatives.  e housing was built by residents with
additional support from local skilled individuals and
construction interns.  ey were fi nanced with assistance
from Community Block Development Grants, the
Washington Housing Trust Fund, private banks, churches,
foundations and private individuals. LCLT was the overall
contractor and project manager, supervising resident
selection, training, site development, construction and co-
operative development.
 e membership selection process was advertised,
interested individuals made contact with LCLT, received a
handbook and were invited to attend guidance/application
workshops. After application workshops, individuals
submit full applications with a USD 30 fee. Eligible
applicants meet two housing co-operative members and
Table 4.2 Comparative Monthly Costs in District of Columbia USA in 2003
LEC median carrying charge 1 bedroom 2 bedroom 3 bedroom
LEC median carrying charge US$ 504 US$ 587 US$ 761
HUD Fair Market Rent US$ 984 US$ 1,154 US$ 1,573
Source: Coalition for Nonprofi t Housing & Economic Development Study (CNHED) (2004)dr
Table 4.3 Income Required to A ordable Housing in District of Columbia in 2003
Type of Housing 1-bedroom 2-bedroom
Market-Rate Condominium US$ 79,702 US$ 115,566
Market-Rate Rental US$ 51,900 US$ 74,360
Limited-Equity Co-operative US$ 22,000 US$ 23,740
Source: Coalition for Nonprofi t Housing & Economic Development Study (CNHED) (2004)

28 Housing for All
receive further information before proceeding to the fi nal
selection stage. Selected members pay an additional USD
200 fees.  ose who are not selected can start an appeal
processb.
Members do not own the land they occupy. LCLT retains
ownership of the land and subsidies and off ers a 99 year
ground lease with a one-time renewal option.  e ground
lease is the legal document that specifi es conditions and
terms between the co-operatives and the LCLT.  e Trust
raises funds from various sources and acts as owner and
developer for the construction of housing units. Upon
completion, LCLT transfers the housing units to the
co-operative. LCLT and the co-operative sign a legal
document.  e co-operative is responsible for the payment
of any loans and all expenses, including property taxes.  e
monthly rent under the ground lease is determined by the
minimum amount required to pay for land taxes, as well as
a portion of the LCLT offi ce overheads related to LCLT’s
limited responsibilities as lessor.  e co-operative leases the
housing units to its members for a period of 99 years.  e
monthly rent for the housing units is determined by the
annual budget of the co-operative, which includes property
taxes, loan repayments and reserves for maintenance and
repairs. A member can sell his/her membership in the
co-operative and transfer the Occupancy Agreement on
terms they negotiate between the co-operative and eligible
buyers. Such resale is subject to two restrictions: (1) the
potential purchaser/transferee must meet the federral
government’s criteria of a low-income person in the same
Figure 4.1 The Organisational Structure of the LCLT
Source: LCLT

29
The Challenges of A ordability, Accessibility and Sustainability
area; (2) the sale/transfer price cannot exceed the amount
the selling member originally paid for the membership,
increased by fi ve percent (or the percentage rise in the
consumer price index (CPI), if greater than fi ve percent)
simple interest rate per annum and the amount paid by
the member for authorized capital improvement during
the member’s tenancyc.
Housing Co-operatives in Chinac
In the late 1970s, China experienced a most severe housing
shortage on top of substandard housing conditions, which
from 1978 onwards determined the country to embark
on housing reform and to explore new housing solutions.
Housing co-operatives were one of the innovative housing
solutions.  ey fi rst emerged in Wenzhou (Zhejiang) in
1980. Since then, co-operative housing has experienced
rapid growth. By 1992, more than 100 such co-operatives
had been established in China.  e schemes involved
an aggregate 370 million yuan investment raised from
individuals; 1.22 million sq. metres of housing were built
and 20,000 member households moved into co-operative
housing unitse.  e housing co-operatives in China consist
of three major types of organisations: government bodies,
civil society bodies (non-government organisations) and
enterprises (work units).
Government-organised Housing Co-operatives:  is type of
housing co-operative is organised by local governments
or government bodies. Individual membership is open
to all urban citizens within the local administrative
boundaries.  e Kunming housing co-operative was the
fi rst example of this type. It was launched by the Kunming
Municipal Government in November 1987. Households
under housing hardship were given priority to join the
co-operative.  e entrance fee for a member was 50 per
cent of the cost of the housing unit the member applied
for.  e remainder of the cost had to be paid within three
years after the member moved into the housing unit. By
1993, 16 government-organised co-operatives had been
established in Kunming.  ey invested more than 62
million yuan, built 115,000 m2 and provided housing
for 1,572 households (or a 73m2 average per household).
 e funding of co-operative housing was mainly from
individual members, with extra support from work units
and the governmentf.
Social Sponsorship of Housing Co-operatives: Local workers’
unions are the most important type of social body that can
be found behind housing co-operatives in China. In 1980,
the United Workers’ Union in Wenzhou city launched
a co-operative.  e scheme admitted members through
its union’s work unit divisions on a voluntary basis. By
1988, 21.65 million yuan had been raised from individual
members and 2,695 households were provided with new
housing units. Local workers’ unions established similar
co-operatives in other cities, including Beijingg.
Housing Co-operatives based on Work Unit (Employer):
Work units have played a very signifi cant role in this
type of housing co-operative, typically in major areas like
organization, management and partial funding.  ese co-
operatives were based on single work units or single groups
of work units. Individual membership is only open to
employees of the sponsoring work units participating in the
schemes. In 1986, Shanghai No. 6 Toy Factory established
China’s fi rst single work-unit based co-operative, Xin Xin
Housing Co-operative. Individual members were only
required to pay one third of the cost of their housing
units, with the remaining two thirds being funded by the
work unit.  e Government provided support in the form
of tax relief. Housing was allocated and managed by the
co-operativeh. Shengyang Dadong District United Co-
operative was a typical example of multi-work unit-based
co-operatives which involved 230 work unitsi .
Despite their diff erent methods of organisation, housing
co-operatives had several characteristics in common. A
member only needed to pay the full or partial cost of a
housing unit allocated to him/her. He/she did not need
to pay the cost of common facilities, as the commercial
housing purchasers had to do.  e remaining costs were
fi nanced by the sponsoring work units or other co-
operative organising bodies.  e government provided
support for co-operative housing through tax exemption,
low-interest loans, provision of land and low-price
building materials. Co-operative housing was regarded as
a good way of attracting investment from individuals and
of enhancing housing aff ordability for low- or moderate-
income households.
Housing Co-operatives in Kenya
 e development of housing co-operatives in Kenya can be
traced back to 1908 but they offi cially came into existence
in 1931 when the fi rst Co-operative Societies Ordinance
was passed. In 1974, the government regarded housing
co-operatives as an appropriate solution for the national
housing problem. It encouraged the promotion of co-
operative housing programmes in order to mobilize people’s
own resources. At the national level, the Kenya National
Federation of Co-operatives (KNFC) was formed to act
as a mouthpiece for the co-operative movement.  e Co-
operative Bank was established to provide banking services
to co-operatives. By 1983, over 2,400 co-operatives were
in operation with about 1.6 million membersj.
As part of the co-operative movement, housing co-
operatives were launched in 1948 in Kenya and by 1983
over 70 were registered. Housing co-operatives acquire
or build housing units for their members and own them

30 Housing for All
in an initial phase.  ey will subsequently sell the units
to members on a tenant-purchase basis. Members who
have paid the full purchase price own their housing
units, but co-operatives maintain ownership of the land
on which the units are built.  e co-operatives issue
ownership
certifi cates to those members who own
their housing units. However, members cannot sell
their units without permission, since the co-operative
holds the titles to the land. Maisonettes are typically
for rental. Experience shows that the development
of housing co-operatives in Kenya faces three main
obstacles: (1) lack of suitable and aff ordable land; (2)
lack of fi nancial resources; and (3) lack of technical,
managerial and administrative capacities. In response
to these constraints, a National Co-operative Housing
Union (NACHU) has been created to plan and develop
feasible housing programmes for low-income housing
co-operatives, and to help them to secure fi nance and
estate management servicesk. Figure 4.2 summarises
the
institutional and operational organisation of housing co-
operatives in Kenya.
 e Gikomba Housing Co-operative Society in Kenya
 e Gikomba Housing Co-operative was launched by
former slum dwellers in the Gikomba area.  e slum
dwellers qualifi ed for a plot allocation in the Kariobangi
Site and Services Scheme implemented by the Government
Figure 4.2 The Institutional and Operational Model of Housing Co-operatives in Kenya
Source: Adapted from Gatabaki-Kamau (1985)n

31
The Challenges of A ordability, Accessibility and Sustainability
of Kenya and Nairobi City Council. Membership of the
co-operative was open to those people who had been
allocated plots in the Kariobangi scheme. Members paid
a KES 10 entrance fee and were required to buy 20 shares
of KES 50 each over time. Total membership was 45 in
1972 and the co-operative built as many six-room housing
units for them.  e sources of funds for construction of
these housing units were members’ contributions (KES
25,000), a KES 400,000 loan from the National Housing
Corporation and rents from completed housing units (KES
110,000). Housing was initially built through self-help
and subsequently switched to paid labour, both because
of the building skills required and as some members found
permanent jobs elsewherel.
Out of the 45 member households, only 10 (22 per
cent) lived in the housing units built by the co-operative.
Other members rented out their units and lived in rural
areas or other areas in Nairobi. Belonging to a housing
co-operative membership as a homeowner had obvious
economic advantages and improved housing aff ordability.
 e homeowners and the tenants who participated in
the Gikomba Housing Co-operative project had similar
income levels, but homeowers spent about eight per cent
of average monthly incomes on housing, compared with
some 18 per cent for tenantsm.
Self-Build Housingo
Self-build housing has a long history. It has always been
the only form of housing provision in most rural areas.
It was also a dominant form of tenure in urban China
before the Communists came to power. Even after the
1949 Socialist revo
lution, the government continued to
encourage self-build as a supplementary form of public
housing provision to tackle the severe housing shortage.
In the early 1950s, 2.3 million m2 self-build housing
units were completed. Subsequently, left-wingers took
to promoting ‘Yida Ergong’ (nationalised large-scale
construction) which greatly aff ected individual self-
build initiatives, which were even outlawed in many
localities
p
.
Self-build was rediscovered as an approach to
privatisation after 1978. Sine then, growth has been
fast. In 1979, self-build accounted for 1.51 million m2.
It became particularly popular in small cities or towns
– mostly in the latter because of the earlier tradition of
self-build: at one point, for example, 83 per cent of self-
built housing in Jilin and 76 per cent in Helongjiang
provinces were in town areas. In some towns, self-build
has become the largest housing sector since 1979, for
instance accounting for 75 per cent of total housing
construction in Wangkui county and 73.2 per cent in
Shuangcheng county (both in Helongjiang province)
q
.
In July 1980, a government report listed a number
of steps further to encourage self-build.  e report
highlighted the three major advantages of self-build over
other forms of housing provision. (1) Self-build helped
to tackle existing severe housing shortages. (2)  e cost
of self-build was much lower than other forms of housing
provision – by about half compared with public housing.
(3) Self-build is quicker. In view of these benefi ts, the
report proposed subsidies as incentives to accelerate
the development of self-build activities. A few months
later, by the end of 1980, 110 cities had promoted self-
build initiatives. In 1981, the State Council required
local authorities to support and expand self-build.  e
Council reiterated its commitment to self-build in a
1983 set of self-build regulations and stated that any
urban residents or workers with housing hardship could
apply for self-build.  is further encouraged individual
initiative and self-building, which expanded at a rapid,
steady pace. By 1988, self-build housing reached 94.33
million m2, accounting for 18.5 per cent of total housing
completion in urban China that year (Figure 7.3).
However, self-build was aff ected by overall economic
conditions and the political environment. Infl ation
in 1988 and political events the next year forced the
government to put construction under tighter control.
 is led to a decrease in self-build as well. However, in
1992, Deng’s call for fast economic development gave
self-build a fresh boost again.
Self-build in China has developed under a variety of
forms over the last few decades. It can be categorised
into three main types: Full, Subsidised, and Co-
operative Self-Build.
Full Self-Build:  is refers to housing built by households
themselves without the need for any help from public
authorities in terms of fi nance, materials, design and
construction.  is has been the typical formula for
redevelopment of old owner-occupied housing in
China. People who were unemployed or whose work
units were unable to support self-build were likely, or
had, to opt for this form when granted permission for
self-build.
Subsidised Self-Build: Self-build relies entirely, or nearly
so, on individual initiative and labour. Local authorities
or work units have provided only limited fi nancial or
institutional support. In 1983, the Government stated
that public fi nancial support for self-build should be no
more than 20 per cent of housing costs.
Co-operative Self-Build:  is happens when households
interested in self-build come together and provide
mutual (mainly labour) support for construction of
their respective homes. It has been more widespread
in small towns than in cities. China’s small town have

32 Housing for All
Figure 4.3 Self-Build Housing Completion in Urban
China (1979-1992)
Source: Zhang (1998)
been under stronger infl uence from the rural tradition
of co-operative self-build, which was often community-
based; a vivid neighbourhood spirit made it easier to
help each other with self-build housing.

33
The Challenges of A ordability, Accessibility and Sustainability
The Role of Government in
Low-Income Housing
In a perfectly competitive economy, the supply of
goods and services and the set of prices are determined
by the price mechanism in accordance with consumer
preferences and incomes. However, in reality, markets
often operate in circumstances that are at odds with
the assumptions of perfect competitive markets. Left
to its own devices, the market system is unlikely to
be effi cient
s
. Given the potential for market failure,
governments in all countries have perceived the
need to intervene and correct market failures or to
introduce policies or measures to compensate its
eff ects
t
. A vast majority of poor people cannot meet
their housing needs on the open market.  erefore,
even in developed countries, government plays a
strong role in the housing delivery system. As Table
5.1 shows, about one third of the housing stock in
1980 was public housing in Austria, the Netherlands,
Sweden and the United Kingdom.
Government can play four types of role in the
housing market: (1) an allocative role: to intervening
in the allocative function of the market to improve
effi ciency; (2) a distributive role: as market-based
distribution may not achieve equity, government may
step in to align the distribution of outcomes with
equity principles; (3) a stabilization role: government
can stablise the market through various steps and
policies; (4) a regulatory role
v
: the most common
form of intervention in the housing sector is through
regulation, direct provision and subsidies.
Direct Provision of Public Housing:
The Case of Hong Kong
World War II and Civil War in China caused a massive
wave of immigration from mainland China to Hong Kong,
where the population surged from 900,000 in 1945 to 2.3
million in 1949w.  e market system could not cope with
the increase in housing demand on such a massive scale.
Because supply of housing is inelastic in the short run,
any increase will shift the demand curve upward, causing
prices to rocket. Soaring prices became unaff ordable to
increasing numbers, putting the Hong Kong government
under stronger pressure to step in. However, initially the
government maintained a non-interventionist approach
and took no action beyond rent controls.  e government
held the view that in most circumstances it was futile or
damaging to the economy if attempts were made to plan
and control the allocation of resources available to the
private sector and to frustrate the operation of market
forces.  erefore, the government kept well away from any
direct provision of housing in the face of the severe housing
shortage, overcrowding and squats in the late 1940s.
Public housing provision in Hong Kong started as an
emergency response to disaster conditions.  e big fi re of
Shek Kip Mei, on Christmas Eve 1953, reversed government
Chapter 5: The Role of Government in
Low-Income Housing
Table 5.1 Government Intervention in Housing in Selected Developed Countries (1980)
Country Government
Housing
Expenditure as per
cent of GNP
Government
Housing
Expenditure as
per cent of public
expenditure
Public Expenditure
as per cent of
housing owned by
government
Housing
consumption as
per cent of private
consumption
Austria 2.9 11 26 16.8
Finland 0.7 3.3 5.3 13
 e Netherlands 2.7 7 31.7 10.7
Sweden 2.1 3.4 23 18
United Kingdom 2.2 5 32 20
United States 0.2 0.6 1.5 16
Source: Harsman & Quigley (1991)u

34 Housing for All
reluctance at direct intervention.  e fi re made more than
50,000 residents homeless.  e government demonstrated
its eff ectiveness in a crisis situation as it moved to relocate
fi re victims, and at the same time became aware of the
large scale of squatter settlements in Hong Kong. In 1954,
a Resettlement Division was established within the Public
Works Department to provide housing for the victims of the
1953 fi res and to relocate squatters, all at the lowest possible
costs.  e authorities looked to relocate 50,000 individuals
every year in these government-provided housing estates.
 e spatial standard of public housing was 2.2 m
2
per person
with an average room size of 11 sq m
2
. Figure 5.1 shows that
the fi rst public housing estate was built by the government in
Hong Kong in the 1950s. In retrospect, this came as the fi rst
stage in a most ambitious involvement in housing provision.
By the end of the 20
th
century, half of the housing stock in
Hong Kong was government-sponsored, proving that public
provision can be a most eff ective solution to the housing
problem of low-income people.
Government provision focused on resettlement during the
fi rst 10 years of the public housing programme, with the
numbers remaining below 10,000 units a year between
1954/55 and 1959/60. In the early 1960s, the government
began to expand public housing beyond resettlement.
Total annual public supply increased from 9,917 housing
units in 1959/60 to 14,130 units in 1978/79. However,
provision of social housing (public housing plus a very
small number of units produced by housing societies)
has increased tremendously since the 1980s. In 1979/80,
annual completion of social housing increased 115 percent
over the previous year. Average housing production
experienced signifi cant increases in the 1980s and 1990s
(Figure 5.2). Annual production of social housing
increased from 32,975 units in 1996/97 to 89,002 units in
Figure 5.1 First Public Housing Estate Built in Hong Kong in the 1950s
Source: UN-HABITAT/X.Zhang
Figure 5.2 Total Annual Production of Social Housing
in Hong Kong
Source: Based on HK Housing Authority
Annual Reports, various issues

35
The Challenges of A ordability, Accessibility and Sustainability
2000/01.  e government has not only increased its own
production, it has also extended the scope of intervention
since the 1970s, restructuring and reinforcing its housing
function accordingly. Apart from traditional production of
public rental housing, the government began to promote
subsidized homeownership schemes. Figure 5.3 shows the
increase in the production of subsidized owner-occupied
housing units, which refl ects the shift of government
intervention in Hong Kong.
 e government played an extremely important role in
low- and moderate-income housing in Hong Kong. By
2000, 961,200 public rental and 324,700 subsidised units
were available for sale.  e population living in public
housing reached 3,235,200 in 2000/01.
Housing Subsidies in Hong Kong
Subsidies are frequently used by governments to alter the
allocation of resources and the distribution of incomes.
 ere are two types of subsidies in practice: producer
and consumer subsidies. Producer subsidies take the
form of capital grants or below-market interest rates for
housing built by public authorities or non-profi t social
organizationsy. Government can also provide land free
of charge to housing authorities, or at very low prices for
private initiatives in social housing. Increasing involvement
by social organizations (such as the Hong Kong Housing
Society) and by the private sector to achieve certain
housing objectives favours the producer subsidy approach
over direct housing provision by the government.
Under the consumer subsidy approach, public authorities
alter relative housing prices rather than making explicit
housing allowances to the public.  e subsidy takes the
form of relatively below-market prices. Since the prices
of public housing, as those of social housing provided by
the private and community sectors, are controlled by the
government, tenants and purchasers of price-controlled
housing units receive in-kind subsidies.
Source: Based on HK Department of Census and Statistics
Figure 5.3 Public Rental and Subsidised Owner-
occupied Housing in Hong Kong
Source: Based on HK Housing Authority;
Census and Statistics Department
Note: Private housing rent refers to the average rent
of housing units less than 40 sq m. on Hong Kong island.
Rents for large-size housing units are higher.
Figure 5.4 Prices in the Private Sector and Subsidised
Owner-Occupied Housing in Hong Kong
Source: Based on HK Housing Department
Figure 5.5 The Gap between Public and Private
Housing Rents in Hong Kong

36 Housing for All
The Contribution of Public Housing
Initiatives to Affordability in Hong Kong
In Hong Kong, the government plans and builds public
housing and determines public housing rents and prices.
Owner-occupied public housing sell for over 50 per cent
cheaper than private sector housing (Figure 5.4). For rentals,
the gap is even wider, with rents in public housing are on
average one fi fth of those in private housing (Figure 5.5)
.
Government Mortgage Loan
Schemes: The Case of Singapore
 e government plays an extremely important role in
Singapore, as it provides the bulk of housing units: about
90 per cent, which in the year 2000 was host to 88 per
cent of households in the city-State.  e government is
not only the largest housing developer, but also the largest
mortgage provider through the Housing & Development
Board (HDB). Figure 5.6 shows the massive scale of
government mortgage loans: these totalled 60.1 billion
Singapore dollars (SGD) in the year 2000, which was much
higher than the total housing loans of the private sector
(SGD 38.6 billion)z. HDB provides housing loans to all
eligible buyers of public housing. In 2005 the loan ceiling
was 90 percent of the purchase price or market valuation.
Home buyers are assessed for credit risk to determine the
amount of loan that can be granted to them.  is depends
on the buyer’s age, household income, loan ceiling and
the available Compulsory Provident Fund (CPF) balance
in the Ordinary Account.  e maximum loan maturity
is either 65 years minus the applicant’s age, or 30 years
whichever is shorter. Borrowers can use CPF savings to
repay housing loansaa.
However, all buyers must exhaust their CPF funds before
they are eligible for HDB mortgage loans. No loans
will be granted to buyers aged 65 or more. Monthly
loan repayments cannot exceed 40 percent of monthly
household incomeab. From 19 July 2005, the government
introduced a staged scheme for down payments, which
were also reduced from 20 to 10 per cent of the purchase
price of a housing unit in order to ease initial pressure
on buyers.  e 10 per cent down payment is paid in two
stages: fi ve per cent of the price is paid at the time of
signing the Agreement for Lease, and the remaining fi ve
per cent when taking possession of the new fl at, or about
two to three years later.  e interest rate is subsidized.
Since 1986, HDB mortgage loan interest rates have been
pegged at 0.1 per cent above the CPF savings interest rate.
 is amounted to 2.6 per cent per year in 2005. Interest
rates are adjusted every other year.
In Singapore, government policy also considers the
impact of macroeconomic conditions on the aff ordability
of borrowers.  erefore, in response to the economic
hardship brought by the Asian fi nancial crisis, the
mortgage loan policy was adjusted in 1998 to allow public
housing purchasers to include one or more (up to four)
eligible family income earners as joint owners. As a result,
borrowers can use co-owners’ CPF savings to repay the
mortgage and reduce the burden on individual borrowers.
Furthermore, the government has introduced a Reduced
Mortgage Repayment Scheme; under the arrangement,
repayments can be reduced to 75 percent of the normal
amount for the fi rst fi ve years in order to relieve the
pressure during that early period. From the sixth year
onwards, the monthly repayment is re-calculated based on
the balance of the loan as outstanding at the beginning
of that yearac.
Government-Sponsored Enterprises
in Mortgage Finance: Fannie Mae
and Freddie Mac in the USA
History
 e Federal National Mortgage Association (FNMA,
colloquially referred to as Fannie Mae) and the Federal
Home Loan Mortgage Corporation (FHLMC, colloquially
referred to as Freddie Mac) are the two largest housing
fi nance intermediaries in the USA and since 1968 have
operated as government-sponsored enterprises (GSEs).
 e two bodies are privately owned and under shareholder
supervision.  ey are protected and supported by the US
Federal Government. Such support and protections include
access to a credit line with the US Treasury, exemption
from State and local income taxes, and exemption from the
supervision of the Securities and Exchange Commission
(SEC)ad.
Source: based on Government of Singapore data
Figure 5.6 Total Outstanding Government Mortgage
Loans in Singapore
Total Outstanding Government Mortgage Loans in Singapore
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Yea r
S$ Billion

37
The Challenges of A ordability, Accessibility and Sustainability
Fannie Mae was established in 1938 as part of the then
president Franklin Roosevelt’s New Deal favouring
(among other things) government intervention in the
housing market.  e Great Depression of the 1930s
discouraged private lenders from investing in home loans.
Fannie Mae provides banks with federal funding to extend
home mortgage loans in order to spread home ownership
on more aff ordable conditions. A second GSE – FHLMC
or Freddie Mac – was established in 1970. Today, Fannie
Mae and Freddie Mac together control some 90 percent
of the secondary mortgage market in the United States.
Backed by strong government support, both agencies have
experienced unprecedented fi nancial growth.  e current
assets of these two companies are 45 percent larger than
those of the largest bank in the USA.  ey are the only
two Fortune-500 companies that are not held by SEC
public disclosure requirements on their operationsae.
Fannie Mae and Freddie Mac do not lend directly to
home buyers. Instead, they work with credit institutions
and operate exclusively in the secondary mortgage market.
 ey help mortgage credit in two ways. First, they purchase
mortgages from credit institutions and hold them in their
portfolios. Lenders use the proceeds from the sales to grant
more mortgage credit to clients. Second, the two agencies
issue mortgage-backed securities (MBSs) on the fi nancial
markets, and use the proceeds to purchase more mortgage
loans from credit institutionsaf.
 e Privileges of Fannie Mae and Freddie Mac
Fannie Mae and Freddie Mac are no ordinary privately-
owned companies.  ey are government-sponsored
enterprises and enjoy various privileges compared with
other companies.  ey feature the following characteristics
and advantagesag:

 ey were established by US Congress under special
federal charters, while all other companies normally
hold charters granted by a State;

 e US president can appoint fi ve out of the 18 board
members of each company;

Each company has a line of credit with the US Treasury
for up to USD 2.25 billion;

Each company is exempt from State and local income
taxes;

Each company can use the US Federal Reserve system
as their fi scal agent;

 eir debt is eligible for use as collateral for public
deposits, for purchase by the Federal Reserve in
open-market operations, and for unlimited investment
by commercial banks and S & Ls;
Table 5.2 Balance Sheets for Fannie Mae and Freddie Mac on December 31, 2000 (As a percentage of total
assets)
Fannie Mae Freddie Mac
Assets
Mortgage portfolio 90 84
Investments 8 11
Advances n.a. n.a.
Other assets 2 5
Total Assets 100 100
Liabilities and Capital
Debt securities 95 93
Other borrowing 2 4
Equity 3 3
Total Liabilities and Capital 100 100
Total Assets (USD billion) 675 459
Notes: As of December 31, 2000, Fannie Mae and Freddie Mac had contingent liabilities for outstanding mortgage-backed
securities of USD 707 billion and USD 576 billion, respectively.
n.a. = not applicable.
Source: US Congressional Budget Offi ce (2001)ai

38 Housing for All

 eir securities are exempt from the Securities and
Exchange Commission’s registration and reporting
requirements and fees;

 eir securities are explicitly government securities
under the Securities Exchange Act of 1934;

 eir securities are exempt from the provisions of
State investor protection laws.
 ese advantages bring fi nancial benefi ts to both
enterprises, which will be discussed later.
Portfolios and Funding
Fannie Mae and Freddie Mac are heavily invested in
mortgages and depend on debt securities for funding. Up
to 95 per cent of Fannie Mae funding comes from debt
securities, three per cent from equity and two per cent
from other borrowing.  e funding pattern at Freddie
Mac is broadly similar, with 93 percent of debt securities,
three per cent of equity funding and four per cent of other
borrowing (Table 5.2).
By the end of 2003, Fannie Mae held USD 1,010 billion
in assets, compared with Freddie Mac’s USD 803 billion.
 ey are respectively the second and third largest companies
in the USA in terms of assets. At that date, Fannie Mae’s
mortgage portfolio stood at USD 902 billion (with USD
1,300 billion outstanding MBSs) and Freddie Mac’s was
USD 803 billion (USD 769 billion outstanding MBSs).
 e two agencies also stand as the largest and second-
largest MBS insurers (and guarantors) in the USAah.
Guarantees of Mortgage-backed Securities by Fannie Mae
and Freddie Mac
Mortgage-backed securities (MBSs) are fi nancial
instruments that use pools of mortgages as collateral and
are issued to investors on the securities market. MBSs
diff er from traditional debt instruments that promise a
series of predetermined payments to investors. Instead,
MBSs pay a share of the cash fl ows from the underlying
pool of mortgages.  e cash fl ows are often uneven and
unpredictable. For example, when mortgage interest rates
Table 5.3 Outstanding Mortgage-Backed Securities and Debt in Fannie Mae and Freddie Mac, Year-End 1985-
2000 (USD billion)
Fannie Mae Freddie Mac
MBSsa Debt MBSsa Debt
1985 55 94 100 13
1986 96 94 169 15
1987 136 97 213 20
1988 170 105 226 27
1989 217 116 273 26
1990 288 123 316 31
1991 355 134 359 30
1992 424 166 408 30
1993 471 201 439 50
1994 486 257 461 93
1995 513 299 459 120
1996 548 331 473 157
1997 579 370 476 173
1998 637 460 478 287
1999 679 548 538 361
2000 707 643 576 427
Source: US Congressional Budget Offi ce based on data from the Department of
Housing and Urban Development’s Offi ce of Federal Housing Enterprise Oversight.
a. MBSs = mortgage-backed securities; excludes holdings of the enterprise’s own MBSs held in its portfolio.

39
The Challenges of A ordability, Accessibility and Sustainability
fall sharply, borrowers are more likely to prepay loans
through selling or refi nancing their housing units, causing
uncertainty on the pace of repayments or prepayment risks.
A third party’s credit guarantee on an MBS issue provides
assurance against the instability and unpredictability of the
cash fl ows on which investor returns rely. Fannie Mae and
Freddie Mac provide a guarantee of timely payment on
MBSs.  ey assume the credit or default risks in exchange
for a fee, and investors accept the repayment risks in
exchange for a higher rate of return than on a non-callable
debt security. By the end of the year 2000, Fannie Mae and
Freddie Mac had more than USD 1.28 trillion in MBSs.
Guarantees on such large-scale issuance increase both the
risks and the fees for Fannie Mae and Freddie Macaj (Table
5.3)
Government Financial Support to Fannie Mae and Freddie
Mac
As mentioned earlier, Fannie Mae and Freddie Mac are
exempted from State and local income taxes as well as
from SEC registration.  ey also enjoy lower costs of
credit rating services for debt and MBS issues. Such
benefi ts were equivalent to USD 1.2 billion in the year
2000.  e benefi ts increase with the growth of business.
 e value of exemption from taxes accounted for USD
478.6 million for Fannie Mae and USD 282.7 million
for Freddie Mac (Table 5.4).  e largest component
of the total fi nancial benefi ts lies in the reduction in
interest rates for borrowing by Fannie Mae and Freddie
Mac. Subsidies to debt increased from USD 1.7 billion
in 1995 to USD 3.6 billion in 2000 for Fannie Mae,
and from USD 0.8 billion in 1995 to USD 2.4 billion
in 2000 for Freddie Mac (Table 5.5). Both bodies enjoy
an implicit or implied government guarantee.  eir
securities are considered second in safety only to those
of the US Treasury.  e implied government guarantee
is worth six billion dollars to Fannie Mae and Freddie
Mac
ak
.
Table 5.4 Annual Value of Tax and Regulatory Exemptions for Fannie Mae and Freddie Mac, 1995-2000 (USD
million)
1995 1996 1997 1998 1999 2000
Fannie Mae
State and Local Taxes 239.6 312.4 347.0 371.6 435.2 478.6
SEC Registration 55.3 79.4 70.7 139.7 122.2 85.0
Rating Fees 5.3 6.7 8.0 9.3 11.0 12.7
Subtotal 300.2 398.5 425.7 520.6 568.4 576.3
Freddie Mac
State and Local Taxes 126.9 143.8 157.1 188.5 252.9 282.7
SEC Registration 39.9 53.0 44.8 92.7 96.4 66.5
Rating Fees 5.3 6.7 8.0 9.3 11.0 12.7
Subtotal 172.1 203.5 209.9 290.5 360.3 361.9
Source: US Congressional Budget Offi ce.
Note: SEC = Securities and Exchange Commission.
Table 5.5 Subsidies to Fannie Mae and Freddie Mac debt, 1995-2000 (USD billion)
1995 1996 1997 1998 1999 2000
Capitalized Subsidiesa
Fannie Mae 1.7 1.5 1.8 3.2 3.3 3.6
Freddie Mac 0.8 1.1 0.8 3.3 2.4 2.4
Total 2.5 2.6 2.6 6.5 5.7 6.0
Source: US Congressional Budget Offi ce.
a.  e subsidies to GSE debt are present values.

40 Housing for All
Government Support Boosts GSEs’ Rapid Growth and
Monopoly Positions
 e basic purpose of Fannie Mae and Freddie Mac is to
provide more funds to the housing market by linking
prospective homebuyers and investors with the capital
market.  e two GSEs take advantage of their privileged
credit ratings to borrow at low costs on the capital market
and use the funds raised to acquire residential mortgages
from brokers and other mortgage originators.  e GSEs
package the mortgages acquired from originators into “pass
through” mortgage-backed securities that are collateralised
with their residential mortgages. Homeowner principal
and interest payments are then “passed through” to
investors holding the securitiesal.
During much of the period, Fannie Mae did not play a
major role. By 1965, Fannie Mae and other government
credit support accounted for only six percent of the market
in residential mortgages. Savings and loan associations
(S&Ls), savings and commercial banks accounted for the
rest, together with life insurance companies. However,
the GSEs were granted signifi cant privileges, which result
in strong competitive advantages over other participants
in US housing fi nance markets.  ese privileges enable
the GSEs to borrow cheaply and supply low-cost capital
that enables mortgage originators to provide lower-rate
mortgage loans to the public.  is gave Fannie Mae and
Freddie Mac unique opportunities for rapid expansion. By
the 1980s and early 1990s, they were dominant on the
housing fi nance market, and rapid expansion continued
into the 1990s. GSE involvement in outstanding mortgage
credit increased substantially, from USD 1.0 trillion in
1990 to USD 3.4 trillion in 2003am.
 e Limits on Fannie Mae’s and Freddie Mac’s Contributions
to Housing Aff ordability
Fannie Mae and Freddie Mac have traditionally focused
on mortgages for one-family owner-occupied properties.
In the year 2000, of all housing units fi nanced by the
two GSEs, 77 per cent of Fannie Mae’s units and 83 per
cent of Freddie Mac’s were one-family owner-occupied
properties. Fannie Mae purchased USD 235 billion worth
of mortgages which fi nanced about 2.3 million such
properties, and Freddie Mac bought USD 175 billion
mortgages which fi nanced nearly 1.7 millionan.
 e two GSEs focus less on other types of properties. In
the year 2000, Fannie Mae purchased USD 10.1 billion
worth of mortgages on multifamily properties, which
accounted for 13 per cent of the total housing units it
fi nanced that year. Freddie Mac’s purchases of multifamily
mortgages totalled USD 6.8 billion, accounting for 10 per
cent of the total housing units it fi nancedao.
Most one-family owner-occupied properties and
multifamily properties are for well-off households.
 erefore, the vast majority of households benefi ting from
the housing fi nance provided by Fannie Mae and Freddie
Mac are middle-class and upper-class rather than low-
income households.
Most low-income households in the USA cannot aff ord
to buy homes, particularly one-family and multifamily
properties. From 1965 to 1990, residential mortgage
credit had expanded more than tenfold from USD 220.8
billion in 1965 to USD 2.6 trillion in 1990, but the
homeownership rate only rose from 63.3 per cent in 1965
to 63.9 per cent in 1990, which was almost identicalap.
 is means that signifi cant government intervention failed
to turn one third of US families into homeowners. Low-
income households still lived in rental housing, which
hardly benefi ted from any support from the two GSEs.
 e two GSEs were benefi cial to homebuyers because
they provided lower-rate mortgages. Expanded mortgage
credit availability results in higher home pricesaq.  ese
and appreciation of property values are benefi cial to
homeowners. However, higher home prices force low-
income households to give up on home ownership and
improved housing conditions. Fannie Mae and Freddie
Mac make housing more aff ordable for middle-income
and upper class families through lower mortgage interest
rates, but they worsen housing aff ordability for the poor
through a concomitant increase in housing purchase
prices.
Government Programmes for
Improved Conditions in Slums: the
Case of Colombo, Sri Lankaar
Colombo is the capital of Sri Lanka and the country’s
largest city. Offi cial statistics show that 66,022 households
live in 1,506 slum areas without proper sanitary
facilities. About 51 per cent of the total population in
Colombo live in slums . Most of these have no proper
access roads, with only narrow footpaths that are not
wide enough for two people to pass at the same time.
 e paths are full of garbage holes. Houses are fl ooded
in the rainy season. Slums are the most visible testimony
to the housing crisis pervading Colombo. Many families
have migrated from rural areas and are forced to spend
large portions of earnings on rents. In the year 2000,
the average civil servant earned a monthly 5,000 rupees,
(LKR) compared with about LKR 3,000 for a labourer.
However, the monthly rent for a two-bedroom house
with electricity and running water was about LKR
8,000 (or USD 102). As a result, most workers can
only settle for houses without water or power, unless
they rent a single or double room in a house or fl at
as
.

41
The Challenges of A ordability, Accessibility and Sustainability
 ey are in eff ect excluded from the urban mainstream.
 e government of Sri Lanka has introduced legislation
that seeks to improve the living conditions of the poor
through several strategies.
In Sri Lanka, greater emphasis has been placed on the
redistribution of wealth through taxation, land reform
and ceilings on house prices. In addition, a scheme
for worker participation in all economic sectors is
being worked out. In the social sphere, education and
healthcare have been made more accessible to those of
smaller means. Rations of basic foods and other essential
commodities are provided free or at nominal costs, and
school books are distributed free to poor families.
In the housing sector, redistribution of wealth and
improved living conditions for the poor are taking place
through two distinct channels. First, the government
has introduced legislation to control rents, protect
tenant rights and restrict the number of dwellings any
individual may own. Landlords are forced to sell any
excess houses to tenants. Since home ownership refl ects
social status in Sri Lanka, new homeowners stand to
gain in this respect and the gaps between the two groups
are narrowing.
 e second approach takes the form of government
housing programmes to improve the living conditions of
the urban poor in a direct way.  e components of this
approach are:
(a) Slum clearance;
(b) Slum improvement;
(c) Resettlement of shanty dwellers in aided self-help
projects.
Each of these housing programmes is discussed below, but
most attention is given to resettlement.
Slum Clearance
 e traditional approach of slum clearance is applied to
areas where no improvement is possible, i.e., areas that are
overcrowded, with dilapidated buildings beyond repair.
 e policy is to select the areas, plan them to modern
standards, and carry out redevelopment in a phased
programme that reduces disruption to the normal life of
the occupants of the area to a minimum.
Slum Improvement
 ough a policy of slum improvement has existed for many
years in Sri Lanka, implementation has been virtually
impossible until recently.  e main reason is that most
slums were owned by private landlords, and use of public
funds to improve private property was not a practical
proposition.
 is situation changed with the implementation of new
ceilings on housing property, which became law in 1973.
 is law allows an individual, his wife and each of his
children under 18 years of age to own one house each,
and no more. As a result, the State now temporarily owns
most of the slums that have been declared excess property
by the previous owners.
 e government has improved these slums at taxpayers’
cost and a major programme was planned for 1976 under
the auspices of the Common Amenities Board.  is was
regarded as the most humane and democratic way of
helping slum dwellers without breaking up their family
lifestyles, community standards and close-knit integrated
patterns by imposing on them middle-class styles of
apartment living.  e Common Amenities Board did
improve slums through maintenance, repair and the
addition of common facilities such as water taps, latrines
and drains.  is reduced the pressure on,, and over-use
of, the few already existing facilities so that their rapid
deterioration and unhygienic use could be avoided.
Resettlement of Slum Dwellers through Aided Self-help
Housing
 e Ministry of Housing and Construction has launched
a programme of aided self-help housing to provide decent
housing at the lowest possible cost to squatters who either
have no land or cannot aff ord monthly rentals in excess
of 20 rupees. In the self-help scheme, as is the norm
for construction of indigenous housing in Sri Lanka’s
rural areas, land and materials are supplied at the site to
suitable participants; in the next step, they build their own
dwelling units to modern housing standards, under the
supervision of technical offi cers from the Department of
National Housing.
 e following section details this type of aided self-help
programme as carried out in the Colombo suburb of
Hendala.
Project Description
 e project, the fi rst of its kind in Sri Lanka, was launched
at Hekitta in Hendala on 25 November 1972. Forty fi ve
dwelling units were constructed on three acres of land.
 e area allotted to each house varied from seven to 10
square perches, (1 perche = 30¼ sq. yds) and the size of
each dwelling unit, consisting of an open veranda and
two rooms, was 250 square feet. As shown in Figures 5.7
and 5.8, the design provided for future extensions if the
family increased in number and fi nancial circumstances
permitting. Each unit is provided with a separate water-
sealed lavatory. Drinking water is available from two
communal wells.  e canal opposite the suburb was to
be used for bathing, as is customary in the locality.  e

42 Housing for All
houses were allotted to participants only upon completion
of the project, in order to ensure that experienced builders
would apply their skills equally to all construction and not
favour the specifi c houses they expected to own. In this
way, uniform quality was guaranteed. Participants were
required to bring their families, friends and well-wishers
along to assist them, and due credit was given to each
participant for his or her labour input.  e project was
programmed to ensure continuity of work; in order to
meet the target dates for construction, assistance was given
when necessary, with co-operation from local voluntary
organizations.
Selection of Tenants
 e participants were selected by advertisement, and the
following requirements were stipulated:
(a)  e head of household must be legally married;
(b)  e family unit must consist of four persons, including
two children or two dependants living with the head
of household;
(c)  e head of household must be permanently employed
or self-employed with a regular source of income, and
earnings must be such that he can aff ord the monthly
rental;
(d)  e total monthly income of the family unit must be
between LKR 150 and 350;
(e) A participant or his spouse must have had a minimum
10-year period of residence or employment within fi ve
miles of the project;
(f) Participants must be physically able to carry out
construction work and other responsibilities in the
self-help scheme;
(g) Participants must agree to take part in group education
and social work;
Figure 5.7 Floor Plan of Aided Self-help Housing
Source: United Nations (1978)

43
The Challenges of A ordability, Accessibility and Sustainability
(h) Participants must agree to put in a minimum of 20
work-hours per week, even if that means working at
night, on holidays and week-ends;
(i) Participants must be willing to co-operate with each
other and accept the majority decisions of the working
partners;
(j) Applicants must be between 25 and 45 years of age.
 ose applicants complying with the requirements
for selection were interviewed. Carpenters, masons
and unskilled workers were selected in the ratio of
1:3:9.  e 45 participants were advised to form a
society, with offi cers to administer the work during
the construction period and, thereafter, to attend to
their day-to-day maintenance requirements. In later
programmes, where masons and carpenters were
not available within the squatter settlement, they
were either hired or selected to join the programme
without fulfi lling the conditions specifi ed above. A
model house was constructed to serve temporarily as
an offi ce and demonstration house.
Estimation of Labour and Material Inputs
Besides issuing materials and programming and directing
the work, the technical offi cers from the Department
of National Housing maintained a record of the labour
input of each participant, his family and helpers, on the
following basis:
(a) A skilled worker was given 20 points for eight hours
of work;
(b) An unskilled worker was given 10 points for eight
hours of work;
(c) Women and children were given below 10 points
for eight hours of work, depending on their
performance.
Actual output was an important consideration, and any
diffi culty in a proper assessment of poor output was
handled by the technical offi cer with help from the society
established by the participants.  is method saved about
LKR 800 per unit in labour costs and reduced monthly
rentals by fi ve rupees.
Computation of Monthly Instalments for Purchase
Monthly instalments were based on recovery of the basic
cost of material and overhead over a 20-year period.  e
land was leased at a three per cent annual rate.  erefore,
average monthly instalment on the above basis was about
20 rupees.
In order to compute the amount to be paid by each
householder, the total number of points obtained by all
participants was divided by the number of homes built.
 is represented the average number of points per house.
Participants who had put in more points than the average
paid less; those who came under the average had to pay
more. After 20 years the tenants own their homes, while
the government retains ownership of the land.
Benefi ts
Since the fi rst project at Hekitta was such a success, the
Commissioner for National Housing has ascertained by
advertisement the demand for this type of construction in
other areas. Where demand was identifi ed, the Commissioner
has proceeded to acquire land and launch similar schemes.
Construction work on 60 such projects is under way at
present and further projects are being investigated.
 e benefi ts derived by the participants from aided self-
help housing programmes in Sri Lanka are as follows:
(a)  ey acquire a house of about 250 square feet, for
a monthly repayment of only 20 rupees, paid for a
period of 20 years;
(b)  ey obtain a building site without any initial capital
expenditure;
(c) Building materials are made available at the site;
(d) Participants buy their building materials – which
have been purchased at minimum, competitive rates
that take advantage of bulk discounts and with easy,
deferred payments;
(e) Participants receive plans and technical advice at the
site during all stages of construction, so that their
houses comply with the requirements of the Housing
and Town Improvement Ordinance, ensuring
minimum standards;
Figure 5.8 Aided Self-help Housing
Source: United Nations (1978)

44 Housing for All
Table 5.6 Urban Poor Participating in Aided Self-help Programmes in and around Colombo, 1971-1974
Category Number of Persons Percentage of Total Urban Poor
Population
Persons relocated to fl ats and houses
due to slum clearance
13 000 4.0
Persons relocated to aided self-help
housing due to slum clearance
5 340 1.6
Persons benefi ting from slum
improvement through the Common
Amenities Board
8 300 2.6
Persons currently at work on
Common Amenities Board
programmes
11 000 3.4
Total 37 640 11.60
Source: United Nations (1978)
(f)  ey receive the services of other participants without
any apparent hiring costs;
(g) Participants may take advantage of training facilities
and opportunities to acquire contruction skills;
(h)  ey make profi table use of the free time and talent
available in squatter areas:
(i)  ey develop a spirit of co-operation and good-will
towards their neighbours and fellow participants.
 ose who learn a skill in an aided self-help scheme are
later employed by the government in the construction
industry. Over time, residents expand their homes and
variety is added to the initial standard design, refl ecting
resident preferences and individualities.
Unfortunately, because aided self-help houses must comply
with relatively high housing standards, they are beyond the
means of 50 per cent of slum dwellers in Sri Lanka. To
overcome this problem, the government is now supplementing
aided self-help with sites-and-services programmes. At the
time of writing such programmes were still in the planning
stages and, no construction had yet begun.
Table 5.6 shows the number of urban poor who have
benefi ted from aided self-build schemes. Although only
11.6 per cent had benefi ted during 1971-1974, this fi gure
has since been on the increase.

45
The Challenges of A ordability, Accessibility and Sustainability
Financial Instruments for
Low-Income Housing
Community-based Savings Schemes
for Lowest-Income Housing
Formal housing fi nance systems do not work for the
poorest households.  erefore, the poorest have developed
their own survival strategies through informal fi nancing.
Informal fi nance has funded the vast majority of lowest
income housing in the world and is to be found behind
the creation of maintenance of the slums, shanties,
squatter camps and pavement dwellings that provide
shelter for about more than half of the urban population
in developing countries. However, attempts to link this
ingenuity to the fi nancial resources managed by the
formal fi nance system have proved relatively unsuccessful.
NGOs have sought to create fi nancing structures outside
the formal fi nance system, which include Revolving Loan
Funds, providing loans to organizations and groups of the
urban poor specifi cally for housing and infrastructure.
Community-based savings schemes are an essential part
of such community-led organizations. With small savings
deposits pooled together over time, a pool of resources
builds up from which members can take low interest loans
for housing improvements.  ese schemes provide the
poorest the access to credit they are otherwise denied by
the formal fi nance systemat.
Compulsory Savings Schemes
Funds accumulated in compulsory savings schemes (such
as provident or pension funds and other social-security
systems) are the most important form of household
savings; they are also particularly suitable for housing
fi nance since they are held for long maturities and build
up into large amounts. One of the most successful
examples of compulsory savings schemes is Singapore’s
Central Provident Fund (CPF). Contributions to CPF
accrue from two related sources: payroll deduction of a
certain percentage of an employee’s salary, and employer
contributions equivalent to a specifi c percentage of the
employee’s salary.  e percentages are adjusted from time
to time to refl ect economic performance and market
conditionsau. In the mid-1980s, the combined contribution
rate was as high as 50 per cent of an employee’s salary. After
the Asian fi nancial crisis of the late 1990s, the rate was
reduced signifi cantly to alleviate the burden on employers
and to allow employees to have more cash to bring home
for consumption.  e CPF contribution rate also varies
according to age.  e older an employee, the lower rate
s/he contributes to CPF.  e contribution rate is 33 per
cent of monthly salary for employees below 50 years; 30
per cent for those aged 50-55; 18.5 per cent for those aged
56-60; 11 per cent for those aged 61-65; and eight per
cent for those over 66 (Table 6.1).
Virtually all working Singaporeans and permanent residents
are CPF members. CPF has 3.04 million members with
a total balance of 116.96 billion Singapore dollars (SGD).
 e CPF savings interest rate is reviewed quarterly. It is
derived from the major local banks’ three-month interest
rates. Between 1 May 2005 and 31 July 2005, this worked
out as an annual 0.59 per cent. However, the CPF Board
will be paying the higher rate of 2.50 per cent per annum
as the CPF Act provides for a minimum CPF interest rate
of 2.50 per cent per annum.  is minimum rate is higher
than the 12-month fi xed deposit and savings rates of the
major local banksav.
During the second quarter of 2005 alone, SGD 1.16 billion
CPF funds were spent on public housing and another
SGD 633.3 million on private housing. A total 629,174
members used the CPF fund to repay mortgages on public
housing and another 125,124 on private housing. By the
end of June 2005, a total 1,255,358 members had used
CPF funds to pay for their public and 211,674 members
for their private housing unitsaw.  is refl ects the signifi cant
impact of CPF funds on aff ordable housing solutions for
society as a whole.  anks to the contribution of CPF
funds, today about 84 per cent of Singapore’s population
live in public housing and most of the remaining 16 per
cent use CPF funds for mortgage payments for private
sector housing.
A number of other countries have also introduced
compulsory savings schemes in the form of social security
systems or provident funds. Typically, contributions in
the form of payroll deductions as a percentage of monthly
salaries are shared by employees and employers, with the
ratios varying from time to time.  e fund, usually managed
as a government or parastatal body, allows members to
withdraw or borrow against their balances for certain uses,
such as the purchase of housing units. However, members
are typically formal sector employees. Such schemes often do
not involve rural populations or informal sector workers
ax
.
Chapter 6: Financial Instruments for
Low-Income Housing

46 Housing for All
 is contrasts with the fact that in many developing
countries, a considerable proportion of the population lives
in rural areas or is employed in the informal sector, and
therefore cannot benefi t from such schemes.
 e savings held in such institutional funds often look
like an ideal source of funding for human settlements
improvements for three main reasons: (1) funds are
available as medium- to long-term capital; (2) they are
held as fi nancial assets and provide steady infl ows of
capital over time; (3) investment by these funds is closely
controlled by government to ensure safety. Similarly,
these fi nancial resources can be channelled into human
settlements in three ways:. (1) public sector institutions
can issue debentures to fulfi l security requirements; (2)
development of mortgage-backed securities market into
which provident and social security funds can invest;
(3) direct investment in the human settlements sector
by institutional funds
ay
.
Contractual Savings for Housing
Contractual savings can be brought to bear on housing
through dedicated schemes (known as contractual
savings for housing or CSH). Contractual schemes
normally off er below-market interest rates on savings,
while giving savers the right to take out a loan at a
low fi xed interest rate when the savings contract is
fulfi lled.  e major benefi t is that housing loans are
insulated from market interest fl uctuations
az
. CSH
links the accumulation of savings to the extension
of a loan in the future.  e promise of a loan at a
predetermined, below-market fi xed rate of interest
is particularly attractive to households in a volatile
fi nancial market environment, making home
purchasing more aff ordable. Financial institutions can
also raise long-term funds through CSH schemes.  e
regular deposits built into the savings schemes for a
number of years provide fi nancial institutions with the
liquidity and long-term funding required to fi nance
long-duration housing loans
ba
.
 e Characteristics of Contractual Savings for Housing
 ere are two dominant forms of CSH: the “closed”
German Bauspar system and the “open” French épargne-
logement. In a “closed” CSH system, access to housing loans
often is through a queue of savers waiting for their turn for
a loan from a specialist institution. In an “open” system,
the saver can legally demand extension of a loan at contract
maturity, irrespective of liquidity conditions of the relevant
fi nancial institution.  e two models diff er substantially for
structure and options, as shown in Table 6.2.
 e “Closed” System of Contractual Savings for Housing:  e
Case of Germany
 e German CSH system follows the principle of mutuality.
 e funds are only available to members/participants of the
CSH scheme, which is managed by a specialized institution,
the Bausparkasse. When funds are not enough to meet
CSH loan demand, members are served according to pre-
defi ned queuing rules.  e CSH system is insulated from
capital markets. It was established in the 1920s based on
social, economic and fi nancial grounds. At the beginning,
the Bausparkassen (‘saving for housing banks’) attempted to
provide all the monies required by homebuyers. However,
this proved impossible and therefore the Bausparkassen
concentrated on providing second mortgages, backed by
a network of savings banks.  e fi rst mortgage funding
was provided by a network of mortgage banks which
raised funds from the capital markets through mortgage
bonds.  e 1952 Dwelling House Construction Premium
Table 6.1 Contribution Rates to Singapore’s CPF (2005)
Employee Age
(years)
Contribution
By Employer
(per cent of
salary)
Contribution
By Employee
(per cent of
salary)
Total
Contribution
(per cent of
salary)
CPF distribution among accounts
Ordinary
Account
per cent
Special
Account
per cent
Medisave
Account
per cent
35 & below 13 20 33 22 5 6
36 - 45 13 20 33 20 6 7
46 - 50 13 20 33 18 7 8
51 - 55 11 19 30 15 7 8
56 - 60 6 12.5 18.5 10.5 0 8
61 - 65 3.5 7.5 11 2.5 0 8.5
Above 65 3.5 5 8.5 0 0 8.5
Source: Adapted from Singapore CPF Board

47
The Challenges of A ordability, Accessibility and Sustainability
Table 6.2 Characteristics of Contractual Savings for Housing in France and Germany
Features Epargne-Logement (PEL)
France
Bauspar
Germany
Provider Commercial and Savings Banks Specialised Institutions
1. SAVINGS PHASE
Initial Savings Deposit Small minimum No minimum
Minimum Annual Savings Yes
Early deposits are rewarded
No, but preference will be given to
regular savers
Minimum Total Savings Initial + annual + interest Pre-set by saver in contract
Maximum Savings Amount Yes No
Savings Interest Rate Competitive after-tax yield Below-market after-tax yield
Savings Liquidity Yes; contract transferable to relative No, therefore “interim” loans are
extended
Government Incentives:
i. Tax-free yield i. Full interest tax-free i. Full interest tax-free
ii. Interest subsidy ii. State interest subsidy (“interest premium)
based on interest paid by the bank. Ceiling:
FRF10,000 (until 1999)
ii. Income-targeted interest subsidy,
but only for housing loan
2. MINIMUM WAITING
PERIOD
4 years (except 1981-1992: 5 years)
Some contract benefi ts extended on request.
10-year-limit since 1992
2 years
3. LOAN PHASE
Date of Loan Availability Right to loan immediately at the end of
the savings phase, with option to call loan
extensible up to 10 years
After waiting period of uncertain
duration
Maximum Loan Amount Loan such that interest paid on loan equals
2.5 times interest earned in savings. Ceiling
of FRF400,000
Multiple of contracted savings amount
(1-1.5 times)
Loan Term 2 to 15 years at borrower’s options, but
constrained by interest rules.
6-15 years at borrower’s option; rules
give preference to shorter term loans
Loan Rate Contract deposit rate paid by bank plus
regulated servicing fee.
Minimum spread of 2 per cent over
savings rate
Loan Payment Level Level
Loan Servicing Fee 1.7 per cent of outstanding principal 2 per cent spread
4.CONTRACT OPTIONS
State Interest Subsidy State interest subsidy have ranged from 4/9 to
2/7 of bank interest on savings available after
contract maturity with or without loan
State subsidy is part of the contract
Transfer of Rights Mature contract benefi ts can be transferred
to relative to improve subsidy + maturity of
a loan
Limited transfer
Uses of Loan Purchase of new unit, existing unit, new
secondary residence, rehabilitation, energy
retrofi t, housing REIT (since 1993)
Purchase, construction, rehabilitation
Timing of Loan Once contract matures, up to saver
responding to market conditions.
Right to a loan can be extended to 10 years
Up to lender, but actually according to
pre-specifi ed queuing rules
Tax Free Interest Bank paid contract interest remain tax free
beyond the 10 year limit
Not Applicable
Source: Lea & Renaud (1995)bb

48 Housing for All
Act boosted the development of the CSH scheme, which
encouraged savings for owner-occupied housing. Savers
who fulfi lled the savings contract received a government
interest premium equivalent to 25 to 35 percent of their
total savings.  is incentive stimulated the expansion of
the CSH schemebc.
 e “Open” System of Contractual Savings for Housing:  e
Case of France
France established its fi rst contractual savings for
housing scheme – the Comptes d’Epargne Logement
(CEL) (housing savings accounts) – in 1965. It is very
similar to the German Bauspar model. However, this
scheme has not been very successful due to lack of
funds for loans and despite the eff orts made to create a
mortgage bond market. In 1970, France introduced a
new CSH scheme – the Plans d’Epargne Logement (PEL)
(housing savings plans). Compared to CEL contract,
the PEL off ers a higher loan multiplier (2.5 times
for PEL, instead of 1.5 times for CEL).  e savings
period is longer (four years for PEL, a minimum 1.5
years for CEL). CEL allows withdrawals while PEL
blocks them for fi ve years.  e PEL featured attractive
deposit rates (eight per cent) which was competitive
with after-tax yields on long-term Treasury bonds.
 e deposit rate had two components: a basic deposit
rate of four per cent provided by the deposit bank and
an interest rate premium of four per cent from the
government. While the lending rate is 5.5 per cent,
the spread is 1.5 per cent (now 1.7 per cent) over the
banks’ deposit rate
bd
.  is makes it possible for PEL
savers to borrow from the system at a lower interest
rate than the one on their savings.  is acts as a huge
incentive for households to participate in the CSH
scheme, which as a result is able to mobilize a wide
base of savings, including non-borrowing savers.
CSH accounts holders in France and Germany
represent more than one third of the population in
their respective countries.  e schemes have also
been running in a number of other industrialized
countries, particularly in Western Europe, but
have rarely been emulated in the developing world.
Savings for housing schemes can be an opportunity
to deepen domestic fi nancial systems, to reach out to
low-income households, and to mobilize untapped
household savings; still, they run into major obstacles
in developing countries: (1) the savings period of
three to fi ve years required before loans are granted
exceeds the planning horizon of most low-income
households; (2) most low-income households lack
steady income fl ows owing to predominant informal
sector employment, making sustainable savings
diffi cult over a number of years, and therefore are not
eligible for loans; (3) macro-economic instability and
the fear of hyper-infl ation discourage people from
participating in low-interest earning savings; (4) due
to inadequate supply of suitable housing, savings do
not necessarily secure homes; (5) loans are related to
unreasonably restrictive housing standards, further
restricting people’s choices
be
.
Special Housing Funds
 e Mass Housing Fund in Turkey
In March 1984, the government of Turkey established the
Mass Housing Fund (MHF) to counter the negative cyclical
impact of market fl uctuations on housing aff ordability. At
the time, more than 200,000 housing units had been left
unfi nished due to increases in interest rates and reductions
in household incomes. Having realized that MHF would
run short of funds if it relied solely on allocations from
the national budget, the government introduced special
taxes on a number of imports and other items to secure
additional resources. During 1984-1987, the main
funding sources of MHF were taxes on certain imports (28
percent, on tobacco and spirits) as well as on consumption
of petroleum products (27 percent)bf.
Since then, MHF has provided a substantial, steady
infl ow of funds for fi nancing housing and infrastructural
investments. Being funded through taxes it does not
mobilize savings from households, and therefore can be
regarded as a fi scal instrument. Benefi ting from a steady
resource infl ow, MHF does not need to be concerned
with returns on lending capital and can make loans at
subsidized, low interest rates. MHF came in response to the
ineff ectiveness of the conventional housing fi nance system
in the face of accelerating infl ation and increases in market
interest rates of 40 to 50 per cent, which had combined
to make long-term housing fi nance too expensive for low-
and moderate-income Turkish householdsbg. MHF off ers
fi nancing for the construction of up to 100,000 housing
units a year. Funds can be used to improve existing slums
(known as gecekondus) or for new construction, and
can be lent to homeowners, co-operative associations,
or contractors. Although MHF can not solve all slum-
related problems, it helps to contain the proliferation of
gecekondusbh.
 e Housing Provident Fund in China
 e Housing Provident Fund (HPF) scheme was fi rst
established in Shanghai in 1991.  e rationale was to
raise funds from individuals and work units on the
widest possible scale. Individuals and work units were
each required to pay fi ve per cent of individual salaries
into their HPF accounts.  e money could be used for
housing-related purposes such as home purchase, repairs

49
The Challenges of A ordability, Accessibility and Sustainability
and self-build. HPF opened up a source of funding both
for housing construction and for home purchases. By the
end of 1996, HPF had raised 11.4 billion yuan (CNY)
in Shanghai; CNY 8.4 billion were lent to work units for
housing construction and another CNY 2.0 billion in
mortgage loans to 46,000 households for home purchases.
One fourth of total social housing construction in Shanghai
was fi nanced by HPF in 1996bi.
Encouraged by the Shanghai HPF experience, in a 1994
housing reform resolution the government required the
establishment of HPFs in cities across the country. Like the
Singapore model, since 1994 HPF has been compulsory
for all work units and their employees in cities. However,
practice shows otherwise. Participation rates vary across
cities. It is as high as about 90 per cent in coastal cities
in both Zhejiang and Jiangsu provinces and 98 per cent
in Shanghai. But in most cities, the participation rate is
below 50 per cent, and even under 20 per cent in Beijing.
Contribution rates to HPF also vary across individual
accounts, while most cities require rates under fi ve per
cent.  e changes are a function of economic conditions
in individual citiesbj.
Special Housing Funds in Mexico
Mexican authorities established INFONAVIT (National
Housing Fund for Workers) and FOVISSSTE (Housing
Fund of the Institute of Social Security and Services for
State Workers) in 1972. INFONAVIT provides low-cost
housing and fi nancing to urban workers in the private
sector. It is funded through employer contributions (fi ve per
cent of employee salaries) in the formal sector. Initially, the
fund operated primarily as a housing developer. A reform
in 1992 turned it into a mortgage fi nancing institution.
 e fund provides mortgage loans to eligible borrowers
within the target income range (minimum salaries).  e
interest rate is capped at four to nine per cent regressively
according to the income level within the target range. In
2004, the fund had a portfolio equivalent to USD 36
billion, or about two thirds of outstanding mortgages in
Mexico. In 1995, INFONAVIT and FOVISSSTE had a
combined market share of 63 per cent of mortgage loan
origination (Figure 6.1). INFONAVIT grows very rapidly.
In 2002, it had a 57 per cent market share in terms of
mortgages originated and 72 per cent in terms of loan
value, compared with two per cent of originations and
eight per cent of mortgage value for banksbk. FOVISSSTE
provides low-cost housing and fi nancing to urban workers
in the public sectorbl. It had a two per cent market share in
terms of mortgage originations in 2002.
FOVI (Fondo de Operación y Financiamiento Bancario a la
Vivienda) is a Bank of Mexico trust fund. It has no retail
operations as it on-lends funds to low-income borrowers
through commercial banks and specialized non-banking
fi nancial institutions – known as SOFOLES (Sociedades
Financienras de Objecto Limitado) – on a risk-sharing basis.
Since creation in 1994, SOFOLES have been originating
and servicing loans.  ey off er mortgages for purchase of
new or existing housing units by individuals for owner-
occupation or rental, land development, commercial
infrastructure and home equity. Loans are off ered at
fi xed rates or indexed to the minimum wage. SOFOLES
also off er loans to workers in the informal sector and to
the self-employed.  ey make credit more accessible
Figure 6.1 Number of Mortgage Loans Originated in
Mexico in 1995
Source: Based on Babatz (2006)bm
Figure 6.AD FONHAPO Lending Categories in Mexico (2002)
Loan Type Number of Loans Value of Loans (000s of
pesos)
Average Loan Size (pesos)
Home Improvement 117,792 290,668 2,468
Progressive Housing 25,047 705,423 28,164
Finished Housing 85 12,723 149,685
Total 142,924 1,008,814 7,058
Source Joint Center for Housing Studies

50 Housing for All
to informal and low-income households in two distinct
ways. First, the lending criteria they have developed are
appropriate for the informal sector. Households can pay
a monthly sum equal to their desired mortgage payment
into an account and for a period of time in order to
demonstrate their ability to pay and to accumulate funds
for a down payment. Secondly, SOFOLES provide in-
person physical delivery of statements, collect payments at
on-site locations and outside of business hours, which on
top of convenience makes for a congenial atmospherebn.
SOFOLES performance has been very good and their
delinquency rates are very low – below 2.5 percent.  ey
have a 20 per cent market share and have been growing at
a quick pace since 2001 (Figure 6.2).
FONHAPO (Fondo Nacional de Habitaciones Populares) is
a housing fund and since creation in the early 1980s has
been the primary funding source for low-income housing
in Mexico.  e fund targets the lowest income segment
of the population with incomes under 2.5 times the
minimum wage. FONHAPO is funded by the Mexican
government and international donors. It provided more
than 140,000 loans in 2002, which accounted for about
30 per cent of total national housing loan origination,
but only about one per cent of mortgage lending in terms
of value since average loan size is very smallbo. Figure 9.3
shows the distribution of lending across housing categories,
showing that FONHAPO heavily concentrates on home
improvement and progressive housing.
Table 6.7 shows the conditions for mortgage loans provided
by FOVISSSTE, INFONAVIT, SOFOLES and banks.
FOVISSSTE and INFONAVIT do not need any down
payment, while SOFOLES and banks require a minimum
10 per cent down payment. FOVISSSTE and INFONAVIT
off er maturities of up to 30 years, or 10 years longer than
those available from banks and SOFOLES. As mentioned
Figure 6.2 The Growth of National Housing Funds and
Private Intermediaries in Mexico
Source: Babatz (2006)bp
Table 6.7 Conditions for Mortgage Loans in Mexico (2006)
FOVISSSTE INFONAVIT BANKS (pesos) SOFOLES (pesos)
Minimum Down
payment
0 per cent 0 per cent 10 per cent 10 per cent
Maximum Maturity 30 years 30 years 20 years 20 years
Currency /Index Minimum Wages Minimum Wages Pesos Pesos
Interest Rate (A) Δ per centM.W. + 6
per cent
Δ per centM.W. + 9
per cent
N.A. 15.79 per cent
Interest Rate (B) Δ per centM.W. + 6
per cent
Δ per centM.W. + 9
per cent
12.28 per cent 15.74 per cent
Initial payment
per USD1000 of
outstanding balance (A)
5.99 8.94 * N.A. 13.76
Initial payment
per USD1000 of
outstanding balance (B)
5.99 12.22 * 12.36 13.72
Loan Limit in USD (A) 43,000 25,210 N.A. 23,500
Loan Limit in USD (B) 48,050 25,210 59,500 50,510
Note: For an individual of 35 years old or above. M.W. stands for minimum wage.
A) For an individual with a monthly income of USD 980.
B) For an individual with a monthly income of USD 2,100.
Source: SHF

51
The Challenges of A ordability, Accessibility and Sustainability
earlier, FOVISSSTE and INFONAVIT interest rates are
indexed to minimum wages. FOVISSSTE rates are based on
changes in minimum wages plus six per cent. INFONAVIT
interest rates are based on changes in minimum wages plus
nine pe rcent. Maximum loan amounts are USD 48,050
for FOVISSSTE, USD 25,210 for INFONAVIT, USD
59,500 for bank loans, and USD 50,510 for SOFOLES.
Mexico’s housing fi nance system is dominated by a network
of housing funds in the form of public institutions that
serve the various segments of the population (Figure 6.4).
 e richest four per cent are served by commercial banks.
 e income group between the 95th and 85th percentiles
resorts to FOVI. INFONAVIT and FOVISSSTE cater to
households between the 90th and 60th percentiles while
SOFOLES takes care of those between the 100th 50th
percentiles. Finally, FONHAPO serves the people at the
lowest tier of the formal sector.
Housing Bonds
Bond issuance is one of the traditional instruments
mortgage banks use to mobilize fi nancial resources.
Institutional investors, such as insurance companies and
pension funds, are the main purchasers of such bonds.  ese
are registered and traded on stock exchanges. Investors
can convert the bonds into cash while mortgage lenders
receive long-term funds to match their long-term housing
fi nance needs. Making long-term housing bonds attractive
is important to mobilize funds for housing fi nance. In
this respect, exempting earned interest from income tax
is common practice.  is enables mortgage banks to
issue bonds at below-market interest rates, lowering their
cost of capital, while investors can earn higher net after-
tax returns on the bonds compared with other fi nancial
instruments.  erefore, the bonds are remaining attractive
to investors. Income tax exemption acts a government
subsidybq. Housing bonds is an area where the United
States has accumulated a wealth of experience.
Housing Bonds –  e Case of the USA
Tax-exempt housing bonds were fi rst issued after World
War I but have only been widely used since the early
1970s.  is is when many State housing agencies started
to issue tax- exempt bonds for mortgages on apartment
buildings and on owner-occupied houses. States and local
governments issue bonds at relatively low, tax-exempt
interest rates and on-lend the proceeds as mortgage loans at
slightly higher interest rates. For owner-occupied housing,
the State agency issuing the bonds funds the private
lending institutions where individuals apply for mortgage
loans. Applicants are reviewed for creditworthiness and
any requirements imposed by federal and State laws and by
the issuer. Tax-exempt bond issuance for owner-occupied
housing increased dramatically in the late 1970s, with the
proceeds of bond issues soaring from USD 1.3 billion in
1976 to USD 12 billion in 1980br.
Figure 6.4 Government Housing Funds and the Housing Finance System in Mexico
Source: Joint Center for Housing Studies (2004)

52 Housing for All
Bonds for Owner-Occupied Housing
Tax-exempt bonds for owner-occupied housing were fi rst
issued by California after World War I and by Oregon
shortly after World War II.  e rationale was to provide
veterans mortgage loans at below-market-rates. In the early
1970s, State housing agencies began to issue tax-exempt
bonds to fi nance mortgages on single-family housing
for all residents with low or moderate incomes. In 1978
cities and counties started issuing bonds, too; at the same
time, State agencies shifted their housing delivery eff orts
from rental housing to owner-occupied housing that was
often targeted at middle-income families and located in
suburban areasbs.
 e operation and management of each bond programme
is slightly diff erent. Some State and local housing agencies
have adequate capacities and play an active role in the
routine management of their programmes, while others
lack appropriate personnel and consist of boards of local
citizens who only meet to approve bond issues. However,
the basic principles of the issues are all the same. Bond
proceeds are on-lent to private fi nancial institutions
for mortgage lending according to rules laid out by the
issuer.  e private lenders process loan applications,
automatically accepting those that meet the following
criteria: (1) the issuer’s eligibility requirements; (2) any
restrictions imposed by federal and State laws; and (3) the
borrower’s creditworthiness.  ose homeowers who have
been granted mortgages make monthly repayments to
the lenders, who forward the money to another fi nancial
institution which pays the bondholders.  e bonds are
generally not secured by the issuer’s credit; bondholders
and mortgage issuers assume the risks of bad mortgages.
Federal government subsidies are shared by bondholders
and homebuyers, with another portion also going to the
various intermediaries in the processbt. Table 6.3 shows
some of the bonds issued in 1981 for mortgages on owner-
occupied housing backed by State housing agencies.
Many State and local governments have consistently
imposed low-income limits on homebuyers whose
mortgages are fi nanced by tax-exempt bonds. In particular,
benefi ciaries must be fi rst-time buyers.  ere are also
Table 6.3 Selected Bonds Issued for Mortgages on Owner-Occupied Housing by State Housing Agencies (1981)
Insurer Date of Issue
Bond Issue
Amount
(USD
million)
Bond Net
Interest Cost
( per cent)
Mortgage
Interest Rate
( per cent)
Bond Rating
Moody’s S & P
Alabama HFA 12/1/81 100.00 13.47 13.85 A1 AA-
Alaska HFC 11/1/81 100.00 12.50 10.00a AA AA-
Alaska HFC 12/1/81 100.00 11.54 10.00a AA AA-
Connecticut HFA 12/15/81 200.00 12.89 13.50 A1 AA
Hawaii HA 12/1/81 20.00 12.81 12.87 A A
Idaho HA 12/1/81 30.07 12.79 13.00a - A
Kentucky HC 12/15/81 36.00 13.22 - Aa AA
Louisiana HFA 12/1/81 150.00 11.81 13.50 Aa AA-
Michigan SHDA 12/1/81 25.00 13.79 - A AA-
New York SMA 11/1/81 104.75 10.97 14.00 Aa AA-
North Carolina HFA 11/1/81 30.00 12.80 13.30 A1 A1
Oklahoma HFA 12/1/81 100.00 13.72 13.90 A A
Rhode Island HMFC 12/1/81 40.00 12.95 13.75 A1 A+
Rhode Island HMFC 12/15/81 25.00 13.92 14.60 A1 A+
Tennessee HAD 12/1/81 50.00 13.96 12.00 A1 A+
Virginia HDA 12/1/81 100.00 13.28 13.70 A1 AA
Wisconsin 8/1/81 10.05 11.50 12.31 Aa AA-
Wyoming CDA 12/1/81 75.00 13.46 13.00 Aa AA-
Source: US Congressional Budget Offi ce (1982)
bu

53
The Challenges of A ordability, Accessibility and Sustainability
restrictions on the prices of housing units that can be bought
with mortgages fi nanced by tax-exempt bonds.  e limit
for new houses located outside targeted areas is 90 percent
of the area median price of new houses. Limits in targeted
areas are 110 percent of the area median price of new and
existing houses. Data is vary scarce on whether mortgages
fi nanced with tax-exempt bonds go to newly constructed
as opposed to existing housing units.  e proportion varies
across States
bv
. Table 6.4 shows that the percentage ranged
from 10 percent newly constructed in Rhode Island to 95
percent in Montgomery County, Maryland.
Bonds for Rental Housing
 e fi rst tax-exempt bonds for rental housing were issued
by New York State in 1955. In the early 1970s, many
other State housing agencies also began to issue bonds for
rental housing. In the mid-1980s, these agencies became
heavily
involved in the ‘Section 8’ housing project
programme, under which the federal government
pays private project owners a large portion of rent on
behalf of low-income tenants. Housing agencies issue
tax-exempt bonds to fi nance construction of privately-
owned ‘Section 8’ apartment buildings.  e Housing
Act of 1937 habilitated local housing agencies, non-
profi t organizations and individuals designated as public
instrumentalities to issue tax-exempt bonds. In the late
1970s and early 1980s, most tax-exempt bonds issued by
State housing agencies were used to fi nance ‘Section 8’
projects, while some others supported market-rate rental
housing.  e mortgages on market-rate housing projects
are issued by the Federal Housing Administration
(FHA).  e Mortgage Subsidy Bond Tax Act of 1980
subjects tax-exempt bonds for rental housing to two new
requirements: (1) all bonds issued after 1 January 1992
must be issued in registered form, which means that the
trustee or some other party must have a current record
of the names of all bondholders.  e maintenance of a
record of bond bearers will make it easier to collect estate
and gift tax
bx
.
Bonds for Home Improvement Loans
Tax-exempt bonds can be issued for home improvement
and rehabilitation loans as well as for home mortgages.
Loans fi nanced with bonds are insured by the Federal
Housing Administration (FHA) and payable over a 15-
year period. Home improvement projects fi nanced with
these loans must improve the basic conditions or energy
effi ciency of a housing unit. Authorised improvements
under such loans include plumbing and electric system
renovation, kitchen remodelling, and additions to living
space. Home improvement bonds are subject to all the
restrictions imposed on owner-occupied housing, except
the purchase price and fi rst-time homebuyers rules, and
are limited in sizes
by
.
Housing Banks
 e Government Housing Bank in  ailand
 e Government Housing Bank (GHB) was established
under the  ai Ministry of Finance by the Government
Housing Bank Act and was offi cially open for business
in 1953.  e bank had an initial capital of 20 million
Table 6.4 Percentage of Mortgage Funds Expected to be Used for Newly Constructed Housing (by the 10 Largest
Bond Issuers, 1998)
Issuer Amount of Issue
(USD million)
Value of
Mortgages (USD
million)
Value of Morgages
on New Housing
(USD million )
Percentage of
Mortgage Funds
for New Housing
Connecticut HFA 200.0 170.6 30.0 18
Louisiana HFA 150.0 134.1 63.7 48
New York SMA 104.8 89.0 26.7 30
Alabama HFA 100.0 87.6 61.3 70
Alaska HFC 200.0 235.5 82.4 35
Oklahoma HFA 100.0 97.5 N.A. N.A.
Virginia HAD 100.0 85.3 59.7 70
Wyoming CDA 75.0 64.9 58.4 90
Montgomery County,
Maryland HOC
75.0 64.9 58.4 90
Rhode Island HMFC 65.0 63.2 6.3 10
Source: US Congressional Budget Offi ce (1982)
bw

54 Housing for All
baht (THB)  e primary objective of GHB is to provide
housing mortgage loans for low- and medium-income
households.  e bank off ers residential mortgage loans,
house building and purchase loans or for existing housing
renovation. GHB currently operates through 107 main
branches and 38 sub-branches throughout  ailand
bz
.
One of the GHB’s main functions is to mobilize funds
for on-lending to home buyers in the form of aff ordable
mortgages. On top of this, GHB also undertook land
development, housing construction, property rentals,
and hire-purchase, and acted as a developer of residential
real estate. In 1972, several of these roles were taken over
by the newly created National Housing Authority.  is
allowed GHB to concentrate on mortgage loans and other
related fi nancial services. GHB provides fi nancial support
to the National Housing Authority and private housing
developers in the form of fi nancing and guaranteesca.
 e GHB grows at a brisk pace, has a very signifi cant
market share and generates substantial net profi ts.
It is the leading housing fi nance lender in  ailand,
providing about 30 per cent of all new residential
mortgages. As of 30 September 2004, GHB had
outstanding loans worth THB 369.16 billion which
accounted for about 38.5 per cent of the whole  ai
market (THB 962.27 billion)
cb
. Total GHB assets
increased from THB 14.6 billion in 1987 to 448.44
billion in 2004 when liabilities stood at THB 425.49
billion. Before 1984, GHB funding came mainly from
off -shore sources and local intermediaries. Nowadays,
deposits from the general public are the major source.
Other sources of funding are GHB-issued bonds and
other domestic borrowing
cc
(Table 6.5).  e bank’s non-
performing loans are at 11.17 per cent of the amount
outstanding, or THB 41.13 billion.
 e GHB branch network is not wide-ranging, keeping
overheads low.  e bank has started to off er higher
deposit rates and lower lending rates without any
subsidy from the government. In 2006, fi xed deposit
interest rates were 3.75 per cent for one year, four per
cent for two years, 4.25 per cent for three years, and
4.50 per cent for fi ve years
cd
. GHB off ers a range of
mortgage loans with lower interest rates: (1) loans to
purchase land with house; (2) loans for home repairs
and expansion; (3) loans to buy condominium units;
(4) refi nancing loans; (5) additional loans on existing
accounts. A borrower can borrow up to 80 per cent of the
property value, and can choose a fi xed or an adjustable
interest rate. Maximum loan maturity is 25 years. For
the savings-and-loans scheme, borrowers are required
to make 24 monthly deposits to qualify for a loan that
is 75 times the instalment saving amount.  e scheme
is specially designed to assist those borrowers who work
in the informal sector, or the self-employed who have
diffi culties to prove their income or credit-worthiness
through formal means. If employers maintain deposits
equal to or higher than the total loan amount, interest
rates (which are pegged at one to two per cent above the
deposit rate) can be further lowered,
ce
.
In 2006, GHB interest rates were as follows: one year:
3.5 per cent; two-year fi xed step-up: 4.75 per cent
for the fi rst year and 5.0 per cent for the second year.
Five-year fi xed step-up rate: 5.25 per cent for the fi rst
year, 5.75 per cent for the second year, 6.25 per cent
for the third year, 6.75 per cent for the fourth year, and
7.25 per cent for the fi fth year.  e 10-year fi xed step-
Table 6.5 GHB Financial Performance (THB million)
2004 2003 2002
Total Assets 448,437 377,004 333,005
Total Liabilities 425,584 357,428 316,236
Government Capital 17,320 17,320 17,320
Equity 22,853 19,576 16,769
Interest Income 15,256 14,968 15,489
Other Income 1,043 1,127 778
Total Income 16,299 16,095 16,267
Net Profi t 4,644 3,607 2,112
Net Profi t/Total Income 28,49 22.41 12.98
Return on Equity 20.32 18.43 12.59
Return on Assets 1.04 0.96 0.63
Source: GHB Annual Report 2004

55
The Challenges of A ordability, Accessibility and Sustainability
Table 6.6 Interest Rate on GHB Loans O ered to Individuals in 2006
Loan Maturity Annual Interest Rate ( per cent)
1 year fi xed 3.50
2 years fi xed step-up rate
1 year 4.75
2 years 5.00
3 years fi xed step-up rate
1 year 4.75
2 years 5.75
3 years 6.75
5 years fi xed step-up rate)
1 year 5.25
2 years 5.75
3 years 6.25
4 years 6.75
5 years 7.25
10 years fi xed step-up rate
1 year 5.25
2 years 6.25
3 -10 years 7.00
Fixed interest rates for borrowers opting out of fl oating rates
3 years fi xed step-up rate
1 year 5.25
2 years 6.25
3 years 7.25
5 years fi xed rate
1 -5 years 6.25
5 years fi xed step-up rate)
1 year 5.25
2 years 5.75
3 years 6.25
4 years 6.75
5 years 7.25
10 years fi xed step-up rate
1 year 5.25
2 years 6.25
3 -10 years 7.00
Note: Data as of 2 February 2006

56 Housing for All
up rate was 5.25 per cent for the fi rst year, 6.25 per
cent for the second year and 7.0 percent for the period
from the third to the 10th year (Table 6.6). GHB also
off ers re-fi nancing to holders of mortgage loans with
higher interest rates. Taking their cue from the GHB,
commercial banks have alsotaken to lower their own
rates in order to keep customers. Such competition in the
home loan industry generally brings down lending rates,
making homes more aff ordable to borrowers. In 1980,
the cheapest new private-sector housing unit on the
market was aff ordable to only 15 per cent of households
in Bangkok; today, the proportion has soared to some
80 per cent under GHB housing loan conditions.
About 40 per cent of the borrowers are women. GHB
also runs a dedicated home-loan programme for rural
co-operatives.  e bank provides funds at wholesale
interest rates to co-operatives for them to on-lend to
members for housing-related purposescf.
 ailand’s GHB plays a signifi cant role in achieving
government policy initiatives. In addition to developing
innovative fi nancing options, the Government Housing
Bank helps to develop an appropriate legal infrastructure
and the mechanisms for converting slum-dwellers’
and squatters’ informal rights into legal rights. GHB
enables low-income individuals with no prior credit
histories to purchase homes.  e scheme is based on
the hire-purchase contracts from the National Housing
Authority, which holds the titles on these properties
for three to fi ve years while purchasers repay loans.
Once a borrower demonstrates their ability to pay with
these hire-purchase contracts, title is transferred and
the Government Housing Bank grants the purchaser/
borrower a loan mortgaged on their homecg.
The Home Finance Company Limited in Ghana
T
he government of Ghana established the Bank for Housing
and Construction (BHC) in 1973 to boost the housing fi nance
industry. However, BHC was unable to focus on housing
fi nance and shifted to commercial banking. Mobilising long-
term capital was too much of a challenge due to the poor
savings associated with the harsh macro-economic conditions
of the mid-1970s and the 1980s, which combined high
interest rates, substantial non-performing bank assets and an
over-valued currency. BHC was eventually liquidated in the
year 2000.
 e Home Finance Company Limited (HFC) was another
specialized institution which the World Bank sponsored in
1990 as a special vehicle to promote housing fi nance. HFC
was owned by the government of Ghana, the National
Pension Fund (SSNIT) and Merchant Bank (Ghana) Ltd.
Initial capital was USD 24.4 million, including a USD 7.2
million, 30-year loan from the World Bank, a USD 16.2
million 20-year loan from SSNIT and a USD 1.0 million
technical assistance grant from the World Bank
ch
.
HFC acts as a fund raiser for primary mortgage lenders –
mainly commercial banks. HFC raises funds from the World
Bank and SSNIT and through bond issuance, and on-lends
the proceeds to primary mortgage lenders. HFC has already
raised USD 18 million on the Ghana Stock Exchange through
a number of fi ve-year bond issues – the only listed corporate
bonds on the Ghanaian capital market. Table 6.7 details the
capital raised through bond issuance
ci
.
Specialised lenders provide mortgage loans to moderate-
income earners.  e loan-to-value ratio is 80 per cent or
under. Interest rates are determined by the three-month
average change in the consumer price index, plus 3.5 per cent.
Bondholders earn a yield equivalent to the rate of infl ation
plus one per cent. Originators and servicing institutions
earn a 1.5 per cent fee on the mortgage portfolios under
their management. HFC earns a one per cent fee to cover
management expenses. However, commercial banks consider
this margin as too low for them to commit resources to
originate and manage loan portfolios
cj
.
HFC also raises funds on the capital market through public
share off erings. HFC became a public company listed on the
Table 6.7 Capital Raised through Bond Issuance, 2000-2005 (cedi million)
Bond Holders 2000 2001 2002 2003 2004 2005
Ghana Government 30,145 37,870 44,034 49,362 57,401 64,523
SSNIT (National
Pension Fund)
60,221
87,609 103,246 115,738 133,448 151,286
HFC Unit Trust 507 648 748 535 - -
HFC $ Housbond 71,841 69,076 78,956 63,945 59,634 50,253
HFC £ Housbond 12,906 16,068 18,818 20,877 18,829
Total 162,714 208,109 243,052 248,398 271,360 284,891
Source: Akuff o (2006)

57
The Challenges of A ordability, Accessibility and Sustainability
Ghana Stock Exchange in 1994. Today it has more than 1,000
shareholders, but 95 per cent of the shares are held by only
eight institutions. HFC obtained a deposit-taking license in
2001 and a universal banking license in 2003 in order to off er
commercial, mortgage and investment banking services. In
1991, the bank established an HFC Unit Trust to pioneer
a collective investment scheme. Today, HFC manages three
funds that are valued at USD 23 million
ck
.
So far, HFC has granted more than 3,000 mortgage loans,
some 90 per cent of which were for new housing units.
However, for all the eff or ts to diversify funding sources through
public bond and share issues, the number of mortgage loans
has been decreasing rapidly since 2001.  is has prompted
HFC to re-think its role as a specialized mortgage lender and
to diversify both functions and funding sources (Figure 6.5).
HFC realized that capital market funding was more expensive
than other sources.
Accordingly, its strategy for the future is
to attract lower-cost deposits, which caused HFC to turn
into a universal banking institution in 2003cl.
Figure 6.5 Number of Mortgage loans granted
Source: Akuff o (2006)
Rent-to-Purchase Housing Schemes
 e Rent-to-Mortgage Scheme in the UKcm
 e UK’s Rent-to-Mortgage scheme is an upshot of Right-
to-Buy. Under the Right-to-Buy scheme, public (i.e.,
council) housing tenants and some housing association
tenants can buy their homes at a discount.  e rate of the
discount depends on the length of time as a public housing
tenant, up to a limit.  e Rent-to-Mortgage scheme makes
home buying more aff ordable for public housing tenants,
as it allows them to become homeowners through payment
of only a portion of the Right-to-Buy price.  e buyer
can obtain a mortgage from a bank or building society. A
public housing tenant buys a housing unit with an initial
(partial) payment. Since the initial payment is smaller than
the full Right-to-Buy price, the landlord retains a share in
the value of the property.  e share owned by the buyer is
a function if his/her initial payment. If the buyer pays 70
percent of the Right-to-Buy price, the landlord’s share in
the property value is 30 percent.
 e buyer cannot pay less than a minimum, or more
than a maximum, initial payment.  e minimum initial
payment should refl ect the amount which banks or
building societies may be willing to lend the buyer on a
standard 25-year repayment mortgage.  is is based on
the premise that the minimum initial payment should
refl ect what the buyer can aff ord for a mortgage on which
the monthly repayment is equivalent to current rent the
buyer is paying. While the maximum initial payment is
80 percent of the Right-to-Buy price, those buyers willing
to pay more can purchase the housing unit outright under
the Right-to-Buy scheme.
A buyer receives a discount on the initial payment. If the
property is a house, the discount rate starts at 32 per cent
and increases by one per cent for each additional year to a
limit of 60 per cent. If the property is a fl at, the discount
rate starts at 44 per cent and increases by two per cent a
year to a limit of 70 per cent. Once s/he makes the initial
payment, the buyer will own the property on freehold
(although the formal landlord will have a share). If the
property is a fl at, the buyer will be given a long lease,
normally 125 years.
Trusts
 e Community Development Trust in the USAcn
 e Community Development Trust (CDT) is the only
private real estate investment trust with a public purpose.
CDT makes long-term debt and equity investments
in aff ordable projects. CDT was established in 1998 by
the Local Initiative Support Corporation (LISC) and
a number of socially motivated institutional investors.
CDT operates like a mutual fund, mobilising capital from
institutional investors to acquire or provide fi nancing for
aff ordable housing. All CDT investments must meet the
requirements laid out in the Community Reinvestment
Act (CRA). CDT invests in long-term debt capital through
purchases of smaller, fi xed-rate multifamily mortgages
from community lenders.  e trust also invests in equity
capital either in cash or by providing a tax-advantaged
transition for existing properties to a new set of owners
committed to long-term aff ordability.
CDT operates a debt programme that creates a secondary
market for smaller Low-Income Housing Tax Credit
(LIHTC) loans. CDT leverages its own, limited capital
as it mobilizes institutional investors to purchase a 90
per cent senior interest in each loan, with CDT holding
a 10 per cent subordinate interest. In this way, for every
million dollars worth of capital, CDT leverages USD 10
million in loans. For example, CDT signed an agreement
NO OF MORTGAGES
824
417 411 401 351
461
379
160 106 85 82 83
0
100
200
300
400
500
600
700
800
900
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
YEARS

58 Housing for All
with a socially motivated pension fund – the General
Board of Pensions and Health Benefi ts of the United
Methodist Church (GBPHBUMC).  e pension fund
agreed to purchase a senior interest in each mortgage loan
subject to certain underwriting criteria, and CDT retains
the subordinate interest. GBPHUMC initially provided
USD 30 million in capital, raising it to USD 100 million
by 2004. CDT proposed to repurchase a portion of
GBPHUMC senior interests and then combine them with
CDT-retained subordinate interests.  e next step for
CDT was to securitise the whole loan portfolio for sale in
the open market.  e pool totaled USD 44.9 million and
consisted of 31 aff ordable multifamily housing mortgages
and more than 2,000 aff ordable housing units. CDT
swapped the mortgages for an equal amount of Fannie
Mae mortgage-backed securities (MBSs).  e MBSs were
then sold to JPMorgan Chase (JPMC).
 e loans had an average balance of USD 1.4 million and
were secured by properties; the majority were on properties
featuring LIHTCs. Repurchasing the senior interests
CDT had previously sold to GBPHUMC enabled CDT
to expand its programme capacity while demonstrating its
ability to securitise the previously illiquid senior interests.
 e transaction provides GBPHUMC and CDT four
major benefi ts. First, GBPHUMC’s ability to sell its senior
interest at market prices validates the underlying value of
senior interests. Secondly, securitization provided liquidity
to CDT, freeing up the capital committed to its existing
subordinated holdings.  irdly, securitization eff ectively
reduced CDT’s credit enhancement to a level below the
level provided to GBPHUMC. Finally, the repurchase
of the seasoned senior interests enabled CDT to deliver
new senior interests on a dollar-for-dollar basis, thereby
increasing its capacity.
The Broken Dream: Low Income Housing
Finance - Subprime Lendingco
It is ironic that the innovative housing fi nance instruments
for low income people once help millions of people to
realize their dreams to have decent homes and now also
break dreams of millions. During the last couples of years,
millions of people lost their homes and about 50 providers
of housing credit for low come people through subprime
lending have gone out of business. It results in nationwide
subprime lending crisis in USA and aff ects the global
housing and fi nancial market in 2007.
Homeowership Dream through Promotion of Subprime
Lending
 e subprime mortgage market was formed mainly because
of the political will of the Bill Clinton administration to
push homeownership to record levels through promoting
low income homeownership. Low income people fi nd it
diffi cult to access credit in the formal housing fi nance
system. In the US fi nance parlance, ‘prime’ borrowers
are those to whom their good credit histories and ability
to pay. Subprime lending refers to B-paper, near-prime,
or second chance lending, in other words to make loans
to borrowers who do not qualify for best market interest
rates because of their defi cient credit history. Subprime
lending is risky for both lenders and borrowers due to the
combination of high interest rates, poor credit history,
and unstable fi nancial situations often associated with
subprime borrowers. Because of the high risks associated
with subprime lending, lenders use a variety of methods
to off set these risks such as charging a higher interest
rate. Subprime mortgage loans have a much higher rate
of default than prime mortgage loans.
Because low income people are often denied access to
housing fi nance on the conventional housing mortgage
market, they can regard subprime lending is an
opportunity window for access home loans to purchase
homes. About 21 percent of all mortgage originations
between 2004 and 2006 were subprime, and totalled
US$ 600 billion in 2006.
Subprime loans have been typically distributed by so-
called local Community development banks (CDBs)
which serve low-income areas. CDBs can apply for formal
certifi cation as a Community Development Financial
Institution (CDFI) from the Community Development
Financial Institutions Fund at the US Treasury Dept.  e
CDFI Fund promotes economic revitalization in distressed
communities throughout the USA by providing fi nancial
assistance and information to community development
fi nancial institutions. CDBs/CDFIs can also apply for a
State banking charter (from State banking authorities) or
for a Federal banking charter (from the Comptroller of the
Currency at the US Treasury Dept.).
 e benefi ts of US Federal mortgage insurance through
Fannie Mae and Freddie Mac had so far been restricted to
the US middle class and did not extend to lower-income
segments.  ings have changed: by 2002-2003, the two
institutions decided that the middle-class market was
mature and would not grow as it had over the past few
decades; consequently, they began to insure the higher
tier of ‘subprime’ mortgages (which had emerged in the
1990s), which they saw as the next source of growth for
their business.  ey issued securities backed by these
mortgages (MBSs); the interest on the securities was paid
by individual borrowers’ mortgage repayments; Wall
Street fi rms bought the securities and distributed them
across their range of investment funds; when individual
borrowers began to default on mortgage repayments,
it began diffi cult to pay interest on the securities; the
fi rms/the funds found themselves stuck with bad-quality

59
The Challenges of A ordability, Accessibility and Sustainability
securities they could not sell to anyone; instead they had to
sell good quality securities on the stock exchange to make
up for their ‘subprime’-related losses.
 e Broken Dream: the Subprime Lending Crisis
 e subprime mortgage fi nancial crisis refers to the sharp rise
in foreclosures in the subprime mortgage market that began
in USA in 2006 and became a global fi nance crisis in July and
August 2007. It caused some subprime mortgage lenders to
fail or fi le for bankruptcy, such as the USA’s second largest
subprime lender – New Century Financial Corporation.  e
failure of these lending companies caused the prices of the
mortgage-backed securities to collapse. By January 2007,
there was 810 billion USD subprime mortgage-back securities
which are large enough to aff ect the entire housing market
and the broad economy and spread the impact on the global
fi nancial market.  e Hong Kong Hengshen Index dropped
924 points on 17 August 2007 which was even more than the
drop on 11 September 2001.
Factors Driving the Subprime Lending Crisis
What triggered the crisis in ‘subprime’ loans? Developers
had a double role in the subprime crisis.  eir vested interest
was in building and building and building for profi t,
which they did, causing a fall in house prices in the USA.
 ey managed to do this partly by introducing would-be
buyers to mortgage lenders: they bloated applicant income
and other data and pushed adjustable-rate, interest-only,
and other risky loans.  is was how in some cases they
deliberately attracted a fi nancially marginal clientele who
could not aff ord conventional mortgages – abetting some
of the reckless mortgage lending that exposed borrowers to
higher risks than they could bear.
 e main cause behind the crisis remains reckless mortgage
lending.  e loans were granted by credit unions and so-
called ‘community development banks’. CDBs are designed
to serve residents and spur economic development in
low- to moderate-income (LMI) geographical areas.  ey
provide retail banking services (including mortgage loans)
and usually target “fi nancially underserved” customers.
Subprime borrowers frequently pay higher points and
fees and are saddled with more unfavourable terms and
conditions. Some CDBs were predatory (including those
who granted subprime loans to borrowers who would
have qualifi ed for ‘prime’ terms and conditions), others
(including faith-based, often Christian fundamentalists)
were simply reckless in their lending practice.
Countrywide Financial, the largest US mortgage lender,
boasts that it will grant loans to four out of fi ve borrowers
who have what it calls a ‘less than perfect credit rating’.
 ese subprime borrowers are colloquially known as
‘ninjas’ (for ‘no income, no job, no assets’).  e problem
is that they did not understand the mechanics of their
loans in the fi rst place.  ey did not see that initial terms
and conditions were very undemanding, only to escalate
sharply after a couple of years or so. Many borrowers were
caught between a fall in house prices and a sudden rise in
interest rates on their loans (plus an upward trend in the
cost of credit in general in the USA), and the attendant
surge in monthly repayments proved unsustainable.
Strings of delinquencies and bad loans ensued, hitting
MBS issues. With the fall in house prices, lenders could
not recover their full losses through repossession of failed
borrowers’ homes. For these reasons a good many lenders
and borrowers ended up bankrupt – with borrowers losing
their lifetime savings and ending up in the rental market
where they had started in the fi rst place.
Lessons
Subprime enabled some borrowers access to next-to-prime
loan conditions – a less biased credit-scoring system might
have admitted them to ‘A’ ratings.
 e current shakeout is caused by a re-assessment of risk.
For borrowers who have employment, reasonable credit
histories, and within limits of debt-to-income ratios, not
much will change. Fully documented loans will still get
the best pricing and terms considered by lenders as lower
risk.

Behind the subprime bubble was poor scrutiny and
disregard for one of the cardinal rules in banking
– ‘know your client’ - is taken very seriously in some
developing countries.

An obvious need for better supervision/regulation of
community banks/subprime lenders: it is believed that
thousands of fi nancial institutions serving the needs of
low-income people or communities in the USA either
have not applied for CDFI status, or have otherwise
not been able to fulfi ll all of the requirements for
formal CDFI certifi cation, and therefore have not
benefi ted from CDFI Fund expertise and fi nancial
support.

Existing community banking-related programmes
should be more eff ectively implemented. Lenders may
be required to condition loans on an understanding of
credit and family budget management issues (as some
low-income lending schemes are already doing in
developing countries). Many State governments run
special schemes that can help fi rst-time homebuyers
in selected price ranges to access aff ordable housing
fi nance.  e schemes typically involve courses in
family budgeting and home care/maintenance; they
also include ‘Home Buyers Clubs’ to help would-be
borrowers put themselves in positions that will qualify
them for housing loans.  ose borrowers who have

60 Housing for All
been through the programmes and have eventually
qualifi ed for special mortgage loans have been found
to have lower rates of foreclosure.

As suggested by US Federal Reserve Board chairman
in March 2007, mortgage guarantors Fannie Mae and
Freddie Mac may be required by Congress to limit
their massive holdings to guard against any danger
their debt poses to the overall economy. “Legislation
to strengthen the regulation and supervision of
GSEs (government-sponsored enterprises) is highly
desirable, both to ensure that these companies pose
fewer risks to the fi nancial system and to direct
them toward activities that provide important social
benefi ts,”

Emphasis at Fannie Mae/Freddie Mac would be on
social programmes that boost fi rst-time home buyers
and at the same time try to make said programmes
more aff ordable. More eff ective implementation of its
CDFI-targeted ‘My Community’ programme may be
required of Fannie Mae.

Low-income borrowers in the USA would not end
up as complete losers when the dust settles on the
subprime crisis.

61
The Challenges of A ordability, Accessibility and Sustainability
The Inter-American
Development Bank and Low-
Income Housing Finance
Financing the Construction of Housing Units
During the 1960s, the Inter-American Development Bank
(IDB) on average approved fi ve housing loan projects
per year.  e projects adhered to the then predominant
approach, namely, building completed housing units for low-
income households. Projects were implemented by central
government agencies, and housing units were assigned to
benefi ciaries with soft repayment conditions.  is approach
had several drawbacks. High housing standards put these
projects out of reach for lower-income households, despite
the direct and indirect subsidies included in the fi nancial
terms.  e benefi ts of subsidies fi nally went to middle-income
households. On the other hand, the high standards made the
housing attractive and boosted their market value, inducing
the original benefi ciaries to sell the units to higher income
households
cp
.
 e cost recovery of the IDB housing projects was poor.
Most benefi ciaries had irregular income patterns, which
prevented them from making regular repayments under
the transfer terms imposed by the executing agencies.
Loan defaults were widespread, since public agencies
found it diffi cult to enforce repayment requirements.
Moreover, subsidised interest rates below the cost of
funds made the executing agencies unable to recover
the cost of capital.  e seed capital provided through
IDB loans was exhausted rapidly, making it diffi cult to
replicate and upscale the projects, which were fi nancially
unsustainable.  e fi nancial institutions involved
required regular replenishment of funds from the
government budget
cq
.
The Sites-and-Services Approach
to Low-Income Housing
Due to the failure of these fi nancing methods, the IDB
signifi cantly reduced its involvement in the 1970s,
sponsoring only three projects during that period. Two of
the projects were for integrated urban development with
a low-income housing component.  e Bank turned away
from fi nancing completed houses and shifted fi nancing
to the sites-and-services formula. In this respect, IDB
took to fi nancing services for minimum-size housing lots.
 e services included potable water, sanitary disposal of
waste water, roads, drainage, electricity and individual
connections to the services. Benefi ciaries built their
own dwellings on the serviced lots through self-help,
community/co-operatives or other mechanisms. Slum
upgrading projects broadly fell in the sites-and-services
category. Although slum upgrading eff orts focused on
rationalization of land uses and provision of secure land
tenure, they also included other services similar to those
in sites- and-services projects, which could reach wider
segments of the low-income population and reduce the
need for subsidiescr.
Sites-and-Services Projects - In 1984, Costa Rica fi nanced
a sites-and-services programme through an IDB loan.
 e programme was for provision of 2,900 serviced plots
for households earning less than 2.5 times the minimum
wage. Each serviced housing plot was about 100 m2, with
access roads, potable water, sewerage and electricity. Also
included was the construction and equipment of a sanitary
unit connected to the public utilities.  e programme
also provided loans enabling benefi ciaries to purchase
construction materials and build core units of 14 m2 either
by self-help or community help.  e loans met 72 per
cent of the construction costs of a basic housing unit, with
subsidies providing the balance. In 1994, after a three-
year delay, 2,791 serviced plots were fi nally completed.
However, the cost was 40 to 80 per cent higher than
scheduled. Two and a half years after completion, all the
serviced plots had been allotted to benefi ciaries and 73 per
cent were occupied. Within two years of occupancy, over
70 per cent of households had built at least a one-bedroom
house on their serviced plots and 50 per cent enjoyed the
benefi t of two bedrooms. House sizes ranged between 44
and 66 m2. Most benefi ciaries completed their houses
through self-help rather than community helpcs.
Low-Income Housing as A Component of Integrated
Urban Development Projects - One such project was
at Buenaventura, Colombia in 1977. IDB fi nanced
the installation of serviced lots to house the displaced
population who previously resided in low-lying fl ooded
areas close to the sea. In the course of implementation, the
project underwent signifi cant alterations to the original
designs, causing delays and cost overruns.  e location
and low standards of the site-and-services project did not
attract the target families. Eventually, the serviced plots
were used by newly arrived low-income migrants rather
than the originally intended groups.  e project ended
Chapter 7: The Multilateral Financial Institutions and
Low-Income Housing Projects

62 Housing for All
up providing higher subsidies to benefi ciaries and the
implementation agencies absorbed the cost, which made
it even more diffi cult to sustain the eff ortsct.
The Market-oriented Sector Approach
to Low-Income Housing
 e 1980s saw the rise of privatization around the world.
 e IDB was aff ected by this global shift in perspectives. As
a result, the Bank has taken to place more emphasis on the
private sector in housing supply and provides long-term
mortgage loans for households, while the government acts
as market regulator and facilitatorcu.
Isolated projects have proved incapable of solving the
housing problems of low-income households. IDB-
fi nanced housing has often been occupied by higher-
income groups and proved unsustainable. To ensure that
projects reach the targeted low-income groups, one must
examine the operations of the housing submarkets and take
a sector approach to project design and implementation.
Any projects for low-income households should be based
on a review of the housing markets in which they operate,
particularly with regard to the constraints aff ecting demand
and supply of low-income housing. Projects should dovetail
with the local housing markets in order to serve a specifi c
target population as part of a coherent systemcv.
The Asian Development Bank
(ADB) and Low-Income Housing
Finance: The Case of Fiji cw
Project Background and Objectives
High land and building costs make it more diffi cult for
many low-income households to aff ord adequate housing.
 is is why this segment of the population takes to
squatting on land they do not own in order to establish
informal housing structures.  e private sector is mainly
involved in developing and fi nancing housing for upper-
and middle-income households. It is not interested in low-
income housing projects since profi t margins (if any) are
low in that market segment. However, the public sector
lacks adequate resources to meet the challenges of housing
for low-income households.  e ADB’s project in Fiji came
as part of a wider joint initiative also involving the United
Nations and other donors.  e aim was to address the
challenges and improve the housing sector’s capacity to meet
the needs of low-income households in the Pacifi c island-
State.  e project combined normative and operational
objectives. Sponsors aimed at four main objectives: (1)
establishing appropriate housing policies and standards to
support low-income households; (2) strengthening public
sector institutions to cope with the challenges of low-
income housing provision; (3) improving sector effi ciency;
and (4) increasing the supply of aff ordable housing and
mortgage fi nance for low-income households.
Project Components and Activities
 e ADB project in Fiji had two components: (1)
capacity building; (2) operational activities.  e capacity
building component included six items: (1) improvement
of the operational and fi nancial management of the Fiji
Housing Authority (HA); (2) improvement of HA’s
management information system; (3) improvement in
the operations of the newly created Public Rental Board
(PRB); (4) development of housing and urbanization
policies in Fiji; (5) development of operational guidelines
for the new policies under the Ministry of Local
Government, Housing and Environment (MLGHE);
and (6) improvement of land use planning under the
Native Land Trust Board (NLTB).  e operational
activities component involved the development of 3,320
fully serviced housing lots, provision of mortgage fi nance
for the construction of 4,490 houses, and rehabilitation
of 764 rental units for sale to existing tenants and other
purchasers.  e lots and houses to be fi nanced were
targeted for low-income households.  e ADB fi nanced
improvements in the Fiji HA’s management information
system; improvement of 760 lots; mortgage fi nance for
1,570 houses; and rehabilitation of 404 rental units.
 e total cost of the project was USD 51.3 million,
including 20.6 million in ADB funding. In 1989, the
ADB granted Fiji a USD 9.6 million loan.  e executing
agencies were the HA, PRB, MLGHE and NLTB.  e
ADB also extended three technical assistance grants for
the housing sector to support the loan.  e project was
completed in 1995 and the loan was closed in 1996.
Project Design
 e focus of the project design was on housing
aff ordability for low-income households in Fiji.  e
rationale was to lower the housing standards so as to bring
costs down to meet the fi nancial capacity of low-income
households. In addition, the project also advocated an
incremental approach to housing construction. Under
this approach, households could build a very basic core
house unit meeting the minimum standards of sanitation
and structural strength, and extend the core housing unit
over time according to their fi nancial capacity. However,
lower housing standards and incremental building were
not widely adopted due to implementation problems; as
a result, only the relatively better-off households could
aff ord the lots and houses made available under the
project.
 e capacity-building component required to establish
adequate land planning functions was not completed
because of the personal problems of the consultant

63
The Challenges of A ordability, Accessibility and Sustainability
in charge, who was not replaced on departure as the
government was reluctant to use loan funds for this kind
of capacity building.
One shortcoming of this ADB project in Fiji might
have been lack of a better understanding of low-income
groups.  e target group of low-income households was
broadly defi ned as those between the 13th and 70th income
percentiles, which included 57 per cent of Fiji’s entire
population.  e project failed to reach the genuine low-
income households. Another issue was that the project
design sometimes seemed to be infl uenced by ideological
considerations, rather than what could work for the poor
on the ground. For example, PRB had been put in charge
of providing homes for households at the bottom tier of
the low-income group, while at the same time the project
required PRB to reduce the scale of its operations in favour
of the lowest income group and to pursue privatization
through sales of its housing stock.  is externally-imposed
directive ran against existing PRB objectives and practice,
causing the body to withdraw from the implementation of
the ADB project altogether.
Project Outcomes
Low-Income Housing Provision – ADB provided 1,151
lots, which was more than the expected 760 (Table 7.1).
However, the project did not do so well when it came to
improving the housing sector’s capacity to cope with the
challenges of providing for the low-income segment. In
Fiji, the HA is the only signifi cant provider of housing
lots for low- and middle-income households in the formal
sector.  e PRB and the Housing Assistance and Relief
Trust off er small amounts of rental accommodation
and have not increased their stock much over a decade.
 e volume of housing produced by NGOs is relatively
negligible, and the private sector focuses on higher income
housing development. Between 1990 and 1998, the HA
produced 6,070 housing lots and sold 6,165, of which
3,189 were provided under the ADB-sponsored project.
 e number of urban households increased by 16,000, of
which more than 9,000 were on low incomes.  erefore,
the delivery capacity of the housing sector was way behind
the expansion in new households. As the housing shortage
kept increasing, so did informal housing units – by 3,400,
which exceeded the project’s output of housing lots.
Supply of Mortgage Finance –  e ADB project in Fiji
provided 1,901 mortgage loans, way under the 4,490 that
had been originally planned. ADB provided only 707 loans
out of the 1,570 it was expected to grant.  e bulk of the
loans went to the upper segment of the low-income group
or to middle- and upper-income households, since very
low-income households could not aff ord the high costs of
the housing lots. For example, the cheapest housing lots
were sold HA for 3,606 Fiji dollars (FJD), and the cheapest
HA off er was terraced houses for some FJD 24,000.  e
average mortgage loan was FJD 19,900, compared with an
expected FJD 4,200average.
Lessons from the ADB Project in Fiji
Access to aff ordable land with title is essential to any
improvement in the housing conditions of low-income
households. Unless aff ordable housing alternatives are
available, low-income households will continue to rely
on informal housing.
Projects should defi ne target groups more precisely,
avoiding any excessively broad coverage. Where the
intended group of benefi ciaries is too broadly defi ned,
some segments may in eff ect end up being left out
although they may precisely be those with greater needs.
 e ADB proposes a number of follow-up actions to
further technical assistance with regard to institutional
and policy issues. Of particular concern are the following
points: (1) the need to strike a balance between housing
standards and the fi nancial capacity of low-income
households; (2) the redevelopment of existing informal
housing areas; (3) land availability for urban development
and land use planning; (4) sorting out any overlaps and
confl icts with regard to land subdivisions and building
regulations; (5) defi ning a role for the PRB, and (6)
strengthening the institutional framework for the housing
sector as a whole.
Table 7.1 ADB-funded Housing Lot Development and Sales Data under the Project in Fiji
Development
Period
Development
cost
(FJD ’000)
Lots w/o
house
Lots w/ house Total Lots Average Lot
Cost (FJD)
Tavakubu II 1991-1993 3,636 351 255 606 6,000
Tavakubu VI 1994-1996 1,530 156 125 281 5,443
Mnikovo 1994-1996 1,618 164 100 264 6,129
Total 6,784 671 480 1,151 5,894

64 Housing for All
The World Bank and Low-Income
Housing Projects: The Case of Algeriaa
Project Background and Objectives
Severe Housing Shortages – In Algeria, the average increase
in housing provision did not exceed two per cent between
1962 and 1995, while the country’s population had been
growing 2.7 percent a year (5.4 per cent in urban areas).
 e resulting severe housing shortage led to overcrowded
housing conditions and the expansion of illegal settlements
and slums. In 1995, more than 400,000 housing units
were identifi ed as unfi t for habitation and 120,000 slum
dwellings were in urgent need of redevelopment.  e
project sponsored by the World Bank aimed to encourage
the government to shift public housing towards provision
of sites and services and granting slum dwellers full
ownership of core housing units.
Major Sector Issues -  e World Bank project identifi ed
three broad sector issues, as follows:
Project Objectives -  e World Bank housing project in
Algeria aimed to address these issues in the following
ways:
(1)  e project proposed to distribute housing subsidies
to the developers in charge of the construction works
through the National Housing Fund.  e government
sells public land to developers at a cost equivalent to
20 per cent of the estimated market price. In the next
step, the developer on-sells the land to benefi ciaries.
Infrastructure costs are paid through the government
budget;
(2) Sector Financing:  e project sought cost recovery
through higher direct contributions from the
benefi ciaries; resource transfers to the National
Housing Fund from the sale of public land were also
to be improved;
(3) Land Issues:  e project aimed to encourage the
emergence of a proper land market and to facilitate
the sale of public land at market prices, apart from
Major Housing Sector Issues Strategies/Alternatives
Ineffi cient Low- income Housing Programmes
 e government was providing public rental housing
units to high standards and costs, and had to subsidise
maintenance of the stock. In 1997, the government
fi nanced 20,000 public rental housing units for the poor.
 e Ministry of Housing (MOH) had experimented
with the construction of core units with full ownership
- minimum size: 35-50 m2 - at a unit price 2.5 times
as low as public rental housing units, and which
was completed in less than one year. Private sector
involvement signifi cantly improved the construction
process.
A clear privatization-induced approach was in sight,
increasing the rent-to-income ratio for public housing
tenants.
Land Issues
Although a new law liberalized the land market in 1990,
the bulk of public land was not made available to the
private sector, but was allocated to public entities at 80
per cent below the market price.
Eff orts are in order if the 1990 law is to be implemented
and the land market liberalised.  e government planned
to make public land available to the private sector
through auctions, starting in 1998.
Housing Sector Financing and Cost Recovery
 e government’s objective was begin slum eradication
in the medium term within a fi xed budget allocation.
On the other hand, cost recovery from low-income
households was almost non existent.
In 1997, the government spent fi ve billion dinars (DZD)
for low-income housing, supplying the equivalent
of 25,000 housing units to the poor. in 1998 the
government planned to spend DZD 12 billion a year and
was committed to allocate the budget funds needed to
upgrade all slums in the medium term.
A full-cost recovery policy was set up for private
land development, which was entirely fi nanced by
benefi ciaries. Direct cost recovery from benefi ciary
participation was estimated at 20 per cent. Indirect
cost recovery was to be be provided by local taxes and
connection fees.  e total direct and indirect cost
recovery for poor households is currently around 50 per
cent.

65
The Challenges of A ordability, Accessibility and Sustainability
adapting well-adjusted, aff ordable technical standards
and granting benefi ciaries secure titles on land;
(4) Improving the effi ciency of slum upgrading and
infrastructure programmes in two distinct ways:
adapting urban standards to targeted benefi ciairies’
needs and fi nancial capacities; and increasing private
sector participation in such programmes;
(5) Provision of sites and services for low-income
households;
(6) Strengthening institutional and management
capacities.
Project Activities
 e project came in two parts. One part was institutional
capacity building within (1) the National Housing
Agency; and (2) the Ministry of Housing at local level
through training, project preparation and management.
 e second part of the project was to fi nance the
rehabilitation of existing sites and the development of
new ones through the provision of water, sewerage, roads
and power, constructing core housing units and granting
secure land titles to low-income households (Table 7.2).
 e project provided basic services and core housing
units to targeted benefi ciaries. Basic infrastructure was
provided in steps: (1) design and implementation of
land servicing; (2) provision of housing core units to
households living in inadequate conditions; (3) assigning
land titles to those households in non-legal situations.
 e average infrastructure cost was estimated at DZD
1,600 per m
2
.  e project serviced about 50,000 lots.
Land acquisition cost DZD 100,000 per lot. Developers
paid DZD 20,000 per lot, the balance being absorbed by
the government.
 e physical improvement component of the project was
funded by the government through the National Housing
Fund, and the funding was managed by Caisse nationale du
logement (CNL, or National housing fund). For each site,
funds were allocated to developers under development
agreements which defi ned the duties and responsibilities
of each party (CNL, local town planning directorates
(DLU/LDUs) and developers) as well as detailed fi nancing
plans.
 e project provided core housing units of 30-40 m2 free
of charge to slum-dwelling households. Housing units
included a room and adequate sanitary conditions, at a
cost of about DZD 10,000 per m2. A total 13,000 or so
core units were fi nanced under the project. Most were built
by small and medium-sized enterprises (SMEs) through
competitive bidding.
Implementation Agency
 e Algerian Ministry of Housing appointed a Steering
Committee to carry out its responsibilities in project
implementation, such as site selection and approval of
site development contracts.  e committee also included
representatives from the Ministry of Finance (MOF),
MOH and CNL. A central project implementation unit,
the Support Unit (SU) was to be be established in the MOH
Urban Planning Directorate for project coordination. SU
staff brings together technical, economical, fi nancial and
environmental expertise.
CNL was in charge of managing and monitoring all
rehabilitation and development works. It was also to
supervise the fi nancial assessment of developers, approve
expenditures for the works and make disbursements to
Table 7.2 World Bank Project Components in Algeria
Component Category Cost (USD
million)
% of Total Cost World Bank
fi nancing (USD
million)
% of World Bank
Financing
Site Development
for Low- Income
Households
Physical 213.5 98 145.8 68
Training,
Research and
Equipment
to strengthen
MOH and
CNL (National
Housing Agency)
management
and project
management
Technical
Assistance and
institution
building
4.8 2 4.2 88

66 Housing for All
developers’ subcontractors. CNL was also to build and
manage a database on benefi ciaries. CNL was established
under the MOH in 1991. Its main functions are to manage
public funds as part of the country’s housing subsidisation
scheme and to monitor and control allocation of subsidies
to households.
Site Selection
 e Local Directorates of Urban Planning (LDUs)
prepared the site concept document, which included: (1)
site data, complete with perimeter, housing construction
quality, infrastructure and environmental conditions; (2) a
local population census, identifying the benefi ciaries and
local associations; (3) the social objectives of subprojects;
(4) tentative estimates for the fi nancial plan, including
government subsidies and local council endorsement
of site selection.  e MOH selected the sites based on
physical features and development objectives.
Project Costs
 e largest expenditure item in the project was
infrastructure, which accounted for USD 140.5 million,
followed by construction of the core housing units on
the serviced lots (USD 55.7 million). Another USD 12.2
million went to land plot purchases (Table 7.3).
Table 7.3 Project Costs (USD million)
1999 2000 2001 2002 Total
Land Acquisition 6.8 5.4 0.1 N.A. 12.2
Infrastructure 18.1 47.0 58.9 16.5 140.5
Housing 5.6 19.4 24.2 6.5 55.7
Design and
Supervision
0.6 1.7 2.1 0.6 5.0
Consultant
Services
2.5 0.4 0.4 0.4 3.7
Equipment 0.8 0.1 0.1 0.1 1.1
Total Project
Costs
34.5 73.9 85.8 24.1 218.3

67
The Challenges of A ordability, Accessibility and Sustainability
With housing revolving funds (HRF), loan repayment
monies “revolve”, i.e., are recycled and made available
again for fresh loans. A revolving fund is either funded
completely by users, or partly by users and partly by
subsidization.  e rationale behind revolving funds is
to provide a funding mechanism with a business-like
modus operandi on a commercial or quasi-commercial
basis. Operational defi cits must be covered off by draw-
downs that incur interest charges.  ere is a fundamental
diff erence between revolving fund activities and their
equivalents in the private sector. A revolving fund generally
has a mandate to recover full costs and to maintain the
draw-down levels that correspond to the amounts shown
in the business plan; by contrast, the objective of a private
sector entity is to maximize profi tsb.
However, slum formation and poverty have multiple
causes, and lack of fi nancial resources is not necessarily
the main one. Access to credit cannot bring eff ective
solutions if the basic problem is individual lack of skills,
unsuitability for work and/or poor repayment capacity.
 erefore, before proposing a revolving fund, all the
conditions that will make the fund eff ective should be
met.  ere is a tendency to assume that the main challenge
in improving housing conditions for slum dwellers is their
exclusion from institutional credit. As a matter of fact,
access to fi nance does not only vary from one target group
to another, but even within a single target group, there
can be very signifi cant diff erences in asset holdings, cash
incomes and creditworthiness.  e complexity and variety
of slum-related issues call for proper appreciation of how
a revolving fund would further the objectives of a project,
and what makes a revolving fund an appropriate solution
to the underlying problems. What are the obstacles to slum
upgrading which the targeted group members are facing?
In addition to fi nance, what steps can be taken to remove
these obstaclesc?
Designing Revolving Funds for
Housing Programmes
Assessing Financial Needs
 e fi rst step in designing a housing revolving fund is to
assess the target group’s needs for credit – including the
needs as they perceive them.  e next question is: Do the
existing channels of housing fi nance fail to provide the
needed resources? Can a HRF be structured to address
these needs? Housing revolving funds are often designed
to assist low-income rather than median- or high-income
households. Unlike well-off households, low-income people
have no access to attractively-priced long-term fi nancing.
Financial mechanisms such as HRFs are typically designed
to bridge this fi nancing gapd.
Maximising the Benefi ts of HRFs
When designing a HRF, the major social, economic and
housing characteristics of the targeted group/community
should be evaluated.  is includes household size and
composition, employment opportunities, income
generation activities, income levels, housing types, housing
construction modes and costs, living conditions, housing
improvement needs, available fi nancial resources, fi nancing
gaps, and the local community’s degree of interest in slum
upgrading and housing improvement.  e targeted group/
community should consider all the variables involved in
a HRF : eligibility criteria, fi nancing methods, leverage,
uses of proceeds, terms and conditions, collateral and
general administration. Housing revolving funds can make
signifi cant contributions to slum upgrading and housing
improvement for low-income households, as they can
provide the aff ordable funding these programmes requiree.
However, individual HRFs typically do not operate on
any large scale, due to limited resources; therefore, they
will not be in a position to meet all the credit needs of
a targeted group/community. Consequently, a HRF
will fi t particularly well with any activities that link
slum upgrading with local or community economic
development opportunities which in turn, together with
economic empowerment, will help maximize the benefi ts
of the HRF.
The Revolving Fund for House Improvement
Loans in the Dominican Republic
Objectives and Loan Policy
 e Revolving Fund for House Improvement Loans
(RFHIL) is one of the components of the Dominican
Republic’s Special Fund for the Promotion of Self-help
Initiatives. Its overall objective is to improve living and
housing conditions in slum areas.  e RFHIL grants loans
for repairs, improvements, extension and replacement.
Loan sizes depend on individual repayment capacities.
Loans enable poorer households to improve their shelter
step by step.  ey can even take the form of building
materials rather than cash. Loans are disbursed in stages
as a function of the construction process and based on
available budgets. Borrowers must contribute their own
or paid labour whenever able to do so.  ey are required
Chapter 8: Revolving Funds for Human Settlements

68 Housing for All
either to mortgage the property, or to sign a notarized
promissory notef.
 e Structure and Operation of the Revolving Fund
 e initial funding of the Dominican revolving fund
was provided by the German Society for Technical
Cooperation (GTZ).  e amount at the start was
equivalent to USD 190,000, or some 1,130,000
Dominican pesos (DOP), which at an average loan size of
DOP 3,000, could improve the shelter conditions of 376
households during the fi rst project phase and before any
loan recovery. GTZ provided funds for the acquisition
of building materials and the hiring of labour.  e
Dominican National Housing Institute (INVI) acted
as the implementing agency, providing personnel, offi ce
space as well as logistic and administrative services
g
.
For the implementation of the revolving fund, two
agreements were signed: a bilateral agreement between
the Dominican and German governments, and an
implementation agreement between the Dominican
Planning Ministry, INVI and GTZ. Under this
agreement, INVI was responsible for managing the
fund, and GTZ was granted a right of control and
supervision through its local project offi ce, including
periodic accounting and operational audits as well as
review of quarterly fi nancial and progress reports
h
.
INVI established a Project Unit within its regional
offi ce in Santiago.  e Unit also had a fi eld offi ce in
Pekin where the targeted group was located.  e Project
Team consisted of a director, technicians, social workers,
store personnel, an accountant, a lawyer and a secretary.
 e Project Team was responsible for the following
functions:

planning the loan schemes for house improvements;

coordination, supervision and control of the revolving
fund operations;

marketing the project to residents;

evaluating loan applications;

designing and budgeting for individual house
improvements;

determining the creditworthiness of applicants;

recommending endorsement or otherwise of
applications;

formalising loan contracts;

supervision and follow-up of construction works;

coordination and follow-up of loan recovery;

preparing monthly and quarterly reports.
In the course of implementation, the Project Team held
weekly and monthly meetings to assess performance and
any problems, as well as to plan activities and prepare work
schedules.  e Project Director was directly responsible
to the Director General of INVI, instead of the Division
Level Director as in most schemes.  ere were three
reasons for this. First, INVI is highly centralized and
almost all decisions must be approved by the Director
General. Secondly, as the revolving fund is fi nanced
by an external agency, the Director General prefers to
supervise in person both project implementation and
compliance with the agreement with GTZ.  irdly, the
project was regarded as politically sensitive, since it dealt
with slum issues
i
.
Loan Conditions
 e main objective of RFHIL was to provide low-income
households with access to home improvement loans
with more fl exible conditions, options and alternatives
with regard to repayment capacity. Nevertheless, about
Table 8.1 Revolving Fund for House Improvement Loans – Loan Conditions
Type of
Improvement
Maximum
Loan Value
(DOP)
Maximum
Loan Period
(Years)
Annual
Interest Rate
(%)
Percent. of
Loans Per
Type
Minimum
Income
Required
(DOP)
Monthly
Income
(DOP)
Minimal 500-1000
(USD 79-
158)
1.5 14 10 150-300
(USD 5-10)
31-62 (USD
24-48)
Basic 1100-3000
(USD 173-
472)
2.5 14 25 218-600
(USD 34-
94.5)
43-120 (USD
7-19)
Extension 3100 – 5000 3 14 50 530 – 850 106 -171
Reconstruction 5100 – 8000 6 14 15 515 – 825 105 -164
Source: UN-HABITAT (1991)k

69
The Challenges of A ordability, Accessibility and Sustainability
30 percent of the poorest could not aff ord housing
loans, and they did not consider access to credit as a
priority. Dominican authorities also realized that they
had to subsidise housing loans if these were to benefi t
the targeted group – lower-income households. RFHIL
sought to reduce and rationalize the subsidies (which
applied to interest rates only), off ering various types of
loans with diff erent repayment periods. Loan size was
determined by an applicant’s monthly income and the
share of that income available for housing expenditure
(Table 8.1)
j
.
Processing Loan Applications
 e Project Team was trained in the practicalities of
implementation. Loan applications were processed as
shown in Figure 8.1. At any project fi eld offi ce, a social
worker would give prospective applicants an information
sheet about the scheme and loan conditions.  e social
worker would fi ll in the loan application form to avoid
mistakes and misunderstandings. Applicants were
asked to provide supporting documents, including an
employment certifi cate (or income declaration) and a
land-lease contract. In the next step, the project accountant
would review the economic or fi nancial background of
the application. S/he would next determine maximum
loan size and maturity based on monthly income, the
amount of credit required and repayment capacity.
 is socio-economic evaluation would verify the data
provided by each applicant. Based on these elements,
the project technician would visit the applicant’s house
to measure the plot and house, and to prepare a sketch
of the existing shelter. Having reviewed the structure,
materials and space organisation of the house/plot, the
technician would meet with the applicant, discuss the
Figure 8.1 Flow Chart of RFHILP Loan Application Processing

70 Housing for All
requested or proposed improvements, and estimate the
costs.  e technician would also conduct a loan analysis
and advise on economic feasibility and aff ordability. Once
feasible improvements were agreed upon, the technician
would draw a rough sketch of the planned construction
work, identifying the quantities and proceeding with
the preparation of a budget based on the construction
being planned and the disbursement schedule. Once the
evaluation and budget were completed, the applicant’s
fi le was handed to the Approval Committee, which
consisted of the project director, the technical supervisor,
the project lawyer and the project accountant. After
endorsement by the Committee, ,the fi le would go to the
Head Offi ce for fi nal approval by the Director General
or the Sub-Director for Administration, depending on
the type of loan. Successful applicants were invited to the
project’s fi eld offi ce to sign the loan contracts
l
.
Administration of Loan Recovery
After the fi nal disbursement of building materials, the
project accountant would open a loan repayment control
card and inform the head of the collection section of INVI
in Santiago by memorandum.  e project accountant
would prepare a list of borrowers, complete with payment
amounts, schedules and addresses, and every month
would hand the list over to the INVI debt collector.  e
collector would issue repayment receipts as needed (the
original receipt going to the borrower with a copy for the
daily list of recoveries, another for the project accountant
and another one to the collection section).  e Project
accountant would prepare monthly reports on debt
Figure 8.2 Flow Chart of RFHILP Loan Recovery Process

71
The Challenges of A ordability, Accessibility and Sustainability
recovery, specifying debit and credit items.  e report
went to the Director General and GTZ, together with the
project progress report. Figure 8.2 illustrates the process of
loan recovery.
The Safe Drinking Water Revolving
Loan Fund in Oregon (USA)
 e Safe Drinking Water Revolving Loan Fund (SDWRLF)
off ers a long-term, self-sustainable source of fi nance
to build and upgrade drinking water systems in the US
State of Oregon.  e fund can support diverse activities,
from project planning to acquiring land and equipment
to building facilities for improved water supply, fi ltration,
storage, distribution, etc.  e fund provides direct loans
(up to four million dollars per project) to eligible applicants
based on repayment ability. Loan maturities only extend as
long as a project’s useful life, and the maximum maturity
is 20 years. Interest rates are 80 per cent of those on
State/local bonds. Each applicant must demonstrate their
fi nancial, managerial and technical capacities, and projects
must undergo thorough analysis to ensure creditworthiness
and the safe nature of the loanm.
Funding the Revolving Loan Fund
 e US Congress created a Drinking Water State Revolving
Fund in 1996, to be run by the Federal Government.  e
facility enables individual States to make loans to improve
local public water systems in order to meet current or future
drinking water standards.  e State of Oregon receives
an annual grant from the US Environmental Protection
Agency (EPA) and channels the bulk of the monies to
the local Safe Drinking Water Revolving Loan Fund.  e
State is required to match 20 per cent of federal funding
and does so through bond issues to be repaid with Oregon
State Lottery proceedsn.
 e Application and Funding Process
SDWRLF funds three main types of activities: (1) planning
and preliminary engineering work; (2) fi nal design and
specifi cation; (3) construction. Applications are reviewed
by the Oregon Economic and Community Development
Department (OECDD). OECDD must be satisfi ed that
the amount of fi nancial assistance sought is essential to
project completion. Proposed projects must be feasible,
cost-eff ective and of sound design. OECDD may visit
project sites and consult with other appropriate State and
federal agencies to check that all legal requirements are
met, and to make the best fi nancing optionso.
Loan awards are contingent on certain requirements, such
as the receipt of (1) suffi cient documentation that the
debt and the security pledged are valid and binding on the
recipient; (2) a letter of fi nancial commitment regarding
any other funds needed to undertake and complete the
project. Loan contracts include six major conditions,
as follows: (1) the project should be completed within
two years from the date of execution of the contract;
(2) loan repayments are made promptly, when due,
subject to any remedies for non-payment set out in the
contract; (3) a repayment plan is secured, as evidenced by
pledged resources; (4) thorough maintenance of accounts
and records for all matters associated with the project,
which OECDD shall be entitled to monitor, to ensure
compliance with contract terms; (5) certifi cation that a
professional engineer, registered in Oregon, is hired and
will be responsible for design and construction of the
project; and (6) the presence and upkeep of respective
source and service meters on all connections throughout
the water system, as well as a comprehensive operations
programme for all regular reading and maintenance of all
metersp.

72 Housing for All
One of the common features of risk management is the
use of credit enhancement, that is, a “cushion” to protect
lenders or investors against potential losses.  e degree
of credit enhancement is sized to refl ect a potential loss
determined under a series of adverse scenarios that could
aff ect the asset pool during its life. Credit enhancement for
a specifi c deal typically combines several techniques and is
a refl ection of the specifi cs of the asset in questionq.
Forms of Credit Enhancement
Mortgage credit risks encompass short-term delinquencies
as well as unrecoverable losses due to borrower default.
Government agencies can help to eliminate mortgage
default risks by guaranteeing timely payment of principal
and interest.  e agencies can rely on a variety of credit
enhancement forms, which are categorized as internal or
external according to the provider/sources. Internal credit
enhancement is provided by the borrower and relies on
re-allocation of cash fl ows within the structure. External
credit enhancement is provided by an outside partyr. Table
12.1 shows the main credit enhancement instruments.
External Credit Enhancement
Mortgage Insurance
Mortgage insurance insures a lender against default.
If a borrower defaults on repayment of a loan and the
property is foreclosed, the mortgage insurance provider
must compensate the lender for the whole or a portion
of the loss. In the USA, four types of mortgage insurance
are available, namely: (1) Federal Housing Administration
(FHA) mortgage insurance (FHA-insured mortgage loans
are granted by private lenders); (2) the US Department of
Veteran Aff airs (VA) loan guaranty; (3) mortgage insurance
provided by the Rural Housing Services administration;
and (4) private mortgage insurance (PMI)t. Another type
of mortgage insurance is hazard insurance, which may be
required if the property is located in areas where natural
disasters may occur, such as fl oods and earthquakesu.
Virtually all lenders require a borrower to provide standard
hazard insurance. Conventional loans are insured by private
mortgage insurance companies. Most lenders require
mortgage insurance if the Loan to Value (LTV) ratio is
above 80 per cent. Mortgage insurance typically covers 20
to 25 percent of the original loan balance. Where LTV is
below 80 per cent, PMI is not requiredv. Mortgages with
low-down payments make home purchase more aff ordable
to more people. For this type of loan, mortgage insurance
cushions lenders against the additional risks they take,
while on the other side, the insurance premium is an
additional expense for the borrower who cannot aff ord a
20 percent down paymentw.
Private Mortgage Insurance
In the USA, mortgage insurers are required by State
regulators to maintain capital at or above four per cent
of outstanding risk, while Fannie Mae and Freddie Mac
impose on mortgage insurers a capital ratio equivalent to
a minimum fi ve per cent of outstanding risk .  is capital
reserve represents liquid assets that can be sold to raise cash
to pay for claim costsx.
Chapter 9: Credit Enhancement
Table 9.1 Forms of Credit Enhancement
External Provided by an outside party
Bank Letter of Credit
Insurance
Guarantee
Subordinated Loans from third party
Internal Provided by originator or within the deal structure
Reserve account/refunded or build-up from excess spread
Originator’s guarantee
Senior subordinated structure
Excess spread
Over-collateralisation
Minimum required debt service coverage ratio
Source: Adapted from Fabozzi & Choudhry (2004)s

73
The Challenges of A ordability, Accessibility and Sustainability
Private Mortgage Insurance Premiums
Loans that are not insured or guaranteed by the US
government are known as “conventional” loans. When
a conventional loan is greater than 80 percent of the
property price or value, the borrower must secure private
mortgage insurance (PMI). PMI premiums are expressed
in points.  e premium is infl uenced by several factors
such as the amount of risk covered, LTV, loan maturity
and the physical location of the propertyy.  ere are
normally two ways to pay for private mortgage insurance:
(1) adding a small premium amount to each monthly
payment; (2) paying a larger, single premium. If the
borrower is planning to borrow on mortgage for only a
few years, pay-by-month is probably the best option. If the
borrower is borrowing on mortgage for a long period, s/he
should consider the one-time premium approach. Single-
premium plans come in two categories: refundable and
non-refundable premiums. Table 9.2 shows that the total
monthly payment is somewhat smaller when premiums
are fi nanced rather than paid monthly. If a borrower has
a conventional loan with private (renewable or monthly)
mortgage insurance, or an FHA loan with monthly PMI,
mortgage insurance payments will stop automatically
when the LTV of the outstanding balance declines to a
certain level, for example, 78 per centz.
Cancelling Mortgage Insurance
Depending on the type of loan agreement, a borrower may
be able to cancel the insurance once s/he reaches a certain
amount of equity. If s/he has 20 percent equity (80 percent
LTV), s/he may be able to stop paying insurance.  e LTV
ratio can change if the borrower invests more capital in
home improvement, or if the property appreciates quickly,
or alternatively if the borrower has paid the mortgage for
quite a long time and the outstanding loan balance-to-
value decreasesab.
 e Homeowners Protection Act 1999 sets specifi c
guidelines on borrower cancellation or automatic
termination of private mortgage insurance. PMI will
be automatically cancelled when the borrower reaches
22 percent equity in the original value of the property.
However, these provisions do not apply if the borrower
has not maintained current status on loan payments in the
prior year, or if the loan is a high-risk one, or if there are
additional liens against the borrower’s propertyac.
US Government Mortgage Insurance
In the USA, the government shifted from direct housing
production to enabling through promotion of mortgage
insurance.  e government increased the amount insured
under the FHA home mortgage programme to six billion
dollars in 1949; it increased FHA mortgage insurance
authorization by USD 2.25 billion in 1950 and by an
annual USD 1.5 billion in 1951, 1953 and 1954. At the
same time, the US government reduced production of low-
rent housing to 75,000 units in 1950, 50,000 in 1951,
35,000 in 1953, and down to 20,000 units in 1954ad.
FHA Mortgage Insurance
FHA mortgage insurance is fi nanced by the Mutual
Mortgage Insurance Fund (MMIF), which is required to
be self-supporting. FHA-insured loans come with high
underwriting requirements. MMIF focuses on single-
family mortgage insurance. By 1980, the Fund had built
a comfortable USD 3.4 billion capital reserve and had an
estimated value of 5.3 per cent of insurance outstanding
(eight billion dollars). In the year 2000, MMIF was
worth USD 16.6 billion with 4.01 per cent of insurance
outstandingae. MMIF value was expected to grow to USD
34 billion in 2006. In 1999, FHA insured about 1.3 million
mortgage loans with a combined value of approximately
USD 125 billion. FHA-insured mortgage loans benefi t
homebuyers in a number of ways, as followsaf:
Table 9.2 Monthly Payments for Various Mortgage Insurance Plans
Renewable Mortgage
Insurance
Refundable Single
Premium
Non-refundable Single
Premium
Loan Amount USD 100,000 USD 102,650 USD 102,000
Principal and Interest
Payment (8%)
USD 733.37 USD 753.21 USD 748.44
Mortgage Insurance
Premium
USD 29.17 N/A N/A
Total – Principal, Interest,
and Mortgage Insurance
USD 762.54 USD 753.21 USD 748.44
Source: Steinmetz (2002)aa

74 Housing for All

FHA down payments (three per cent) are lower than
the minimum required by many lenders for non-FHA
mortgages.

FHA requirements for homebuyer credit ratings are
more fl exible than those set by many lenders for non-
FHA borrowers.

FHA allows homebuyers to use contributions from
family members and non-profi t groups to make their
whole down payments, while conventional loans
generally require homebuyers to pay at least a portion
of the down payment from their own funds.

FHA allows borrowers to carry more debt than a
private mortgage insurer normally permits.
FHA mortgage insurance allows a homebuyer to make a
relatively small down payment and obtain a mortgage for
the balance of the purchase price. FHA-insured mortgage
loans can be granted by banks, savings and loans (S&L)
associations, mortgage companies, credit unions or other
FHA-approved lenders. FHA compensates the lender if a
borrower defaults on an FHA-insured mortgageag.
FHA Mortgage Insurance Premiums
For mortgage insurance and as of 2001, FHA charged an
up-front premium of 1.75 points (a point is one per cent of
the loan amount) and an annual premium of 0.5 point for
the whole loan period, unless the borrower made a down
payment of more than 10 per cent of the purchase price.
 e up-front premium can either be paid in cash at the
closing or fi nanced as part of the loan.  e annual premium
is paid monthly.  e following example illustrates how to
calculate the mortgage insurance premiums required for
an FHA-insured mortgage. If the loan amount is USD
50,000, the up-front premium amounts to USD 875
(USD 50,000 x 0.0175), the annual premium to USD 250
(USD 50,000 x 0.005) and the monthly premium toUSD
20.83 (US$ 250 ÷ 12 months); therefore a total USD
20.83 is added to the monthly mortgage repaymentah.
VA Funding Fees
VA-guaranteed loans are underwritten by the US
Department of Veterans Aff airs (VA). Immediately after
World War II, the then Veterans Administration (VA)
began to guarantee old soldiers’ mortgage loans in order to
facilitate home purchases with little if any down payment
and with low interest rates.  e VA guarantees payment
of a mortgage loan made by a private lender to a qualifi ed
veteran in case the borrower defaults. VA-guaranteed loans
do not require the borrower to pay any mortgage insurance
premiumai, but instead impose a funding fee.  is fee is
a function of the amount of down payment made by a
veteran. If there is no down payment, the fee is two points
on the loan; if the down payment is fi ve percent, the fee is
1.5 points; if the down payment is 10 percent of more, the
fee is 1.25 pointsaj.
Table 9.3 Mortgage Insurance in Homeowner Properties in the USA in 2001 (unit: ,000)
First Mortgage Insurance
Status
All
properties
(owned and
rented)
Total
(owned)
1 housing
unit
2 to 4
housing
unit
Condo
minimums
Mobile
homes
FHA-insured fi rst mortgage 4989 4439 3985 72 326 56
VA-guaranteed fi rst mortgage 1264 1163 1055 13 84 12
Rural Housing Service/Rural
Development-guaranteed loan
984 845 786 5 32 22
Insured by State agency 1166 944 842 11 43 48
Insured by State agency
with FHA insurance, VA or
Rural Housing Service/Rural
Development guarantee
41 37 27 0 7 3
Insured conventional fi rst
mortgage
5606 4935 4426 70 313 125
Uninsured conventional fi rst
mortgage
32856 27961 25289 446 1575 651
Not available 3663 3313 1558 26 57 1672
Source: US Census Bureau

75
The Challenges of A ordability, Accessibility and Sustainability
Example: Assume a house has a purchase price of USD
100,000
No down payment:
USD 100,000 x 0.02 = USD 2,000 funding fee
5 % down payment:
USD 100,000 x 0.05 = USD 5,000 down payment
USD 100,000 – USD 5,000 = USD 95,000 loan
USD 95,000 x 0.015 = USD 1,425 funding fee
10 % down payment
USD 100,000 x 0.10 =USD 10,000 down payment
USD 100,000 – USD 10,000 = USD 90,000 loan
USD 90,000 x 0.0125 = USD 1,125 funding fee
First Mortgage Insurance Status
Table 9.3 shows that in 2001, the bulk of conventional fi rst
mortgages (32.86 million) were not insured, accounting for
85 per cent of total fi rst mortgages. Insured conventional fi rst
mortgage loans (5.6 million) accounted for 15 per cent of
total fi rst mortgages. Among insured fi rst mortgages, 39 per
cent of mortgage insurance was provided by FHA, 22.5 per
cent by VA, 17.5 per cent by the Rural Housing Service and
Table 9.4 Requirements for CMHC Mortgage Insurance
Number of Units 1 – 4 units, one of which must be occupied by the owner
Maximum Loan-to-Value
Ratio
1 unit: 95%
2 units (e.g. a duplex): 92.50&%
3 or 4 units: 90%
Qualifying Interest Rate  e interest rate used to assess borrower eligibility is determined as follows:

loan maturity under 3 years – the greater of the lender’s 3 year posted rate or the
contract interest rate (or VRM cap).

Loan maturity 3 years or more – the contract interest (or VRM cap).

Standard Variable Rate Mortgage (VRM), regardless of loan term: the lender’s
3-year posted rate.
Minimum Loan Maturity 6 months
Minimum Equity

Minimum equity of 5% from the borrower’s own resources is required.

Gift down payments from an immediate relative are acceptable and need to be
confi rmed in a letter from the donor. Gift money needs to be in the borrower’s
possession 15 days prior to the closing date.

An Approved Lender needs to verify the borrower’s ability to cover closing costs
of at least 1.5% of purchase price. If closing costs are to be borrowed, the loan
repayment is to be included in the Total Debt Service calculation.
Debt Servicing Ratios

Gross Debt Service (GDS) ratio: the borrower can commit up to 32% of gross
household income toward the payment of principal + interest + property taxes +
heat.

Total Debt Service (TDS) ratio: the borrower can commit up to 40% of gross
household income toward housing obligations and all other debts.

when the property has one to three rental units in addition to the owner-
occupied unit, TDS formula is:
(total principal and interest payments + payments on all other debts) x 100
gross household income + up to 50% of confi rmed gross rental income
Maximum Housing Price Maximum housing prices apply when the LTVs are greater than 90%
Premium on Mortgages for
Home Purchases
Loan to Value Ratio
Up to and including 65%
Up to and including 75%
Up to and including 80%
Up to and including 85%
Up to and including 90%
Up to and including 95%
Premium on Total Loan
0.50%
0.65%
1.00%
1.75%
2.00%
3.25%
Source: Based on CMHC

76 Housing for All
Rural Development-guaranteed Loan Programme, and 21
per cent by State agencies.
Government Mortgage Insurance in Canada
Mortgage insurance was introduced in Canada by the
National Housing Act in 1954. Under this law, mortgage
insurers would compensate lenders’ losses in the event of a
borrower’s defaults.
Government mortgage insurance is provided by the
Canada Mortgage and Housing Corporation (CMHC).
 e agency supports the purchase, improvement and
refi nancing of new and existing owner-occupied housing
units as well as the construction, purchase and refi nancing
of rental properties by investors. Clients can use CMHC-
insured fi nancing not only to purchase homes, but also
for a variety of other purposes; those intent on renovating
their housing units can use insured mortgage loan services
known as “Purchase Plus Improvements”.
CMHC Requirements for Mortgage Insurance
 e terms and conditions for CMHC mortgage insurance
are shown in Table 9.4
Mortgage Insurance for Energy-Effi cient Properties
CMHC recently developed a number of fi nancial
incentives towards the development of energy-effi cient
housing. When a borrower uses his/her mortgage to buy
an energy-effi cient housing unit, or to renovate a home to
improve its energy effi ciency, s/he may be eligible for a 10
per cent refund on their CMHC mortgage loan insurance
premiumak.
Enhancing the Purchase/Refi nance Plus Improvements
In order to facilitate smaller-scale home improvements,
CHMC waives the current 10,000 Canadian dollar
(CAD) maximum improvement limit under its Purchase/
Refi nance Plus Improvements facility, only setting the
limit based on the value of the property (10 per cent of the
value).  e agency simplifi es and streamlines the fi nancing
process for an easier, quicker provision of insured mortgage
fundingal.
Pool Insurance
Pool insurance normally compensates for losses caused
by a borrower’s economic circumstances, but specifi cally
excludes losses that result from bankruptcy, origination
fraud, and hazardsam. Since pool insurance eff ectively covers
only those losses resulting from defaults and foreclosures,
additional insurance is required against losses resulting
from other factors such as bankruptcy, etc.an
Pool insurance is a common credit enhancement technique
among mortgage-backed securities (MBS) issuers. Risk is
reduced to a predictable level as it is spread across a large
number of policy holders. Pool insurance is provided by
composite insurers.  erefore, losses in one business can
be off set with profi ts in another.  e rationale behind pool
insurance is to ensure that the interest and principal due to
creditors/investors are eventually paidao.
Letters of Credit
 e issuer of a whole-loan security or a third party
with a relatively low credit rating can provide a Letter
of Credit (LOC) in the amount required by the rating
agency to enhance the entire deal, or for a lesser amount
to complement other forms of credit enhancementap.  e
LOC is normally provided by a commercial bank with an
AAA credit rating.
Corporate Guarantees
Corporate guarantees cover all types of losses and can be
used as stand-alone credit support or in conjunction with
other forms of credit enhancement. Where primary credit
enhancement is provided by a credit-rated entity, the
creditworthiness of the borrower is subject to re-evaluation
and fl uctuations if that entity’s credit is downgradedaq, or
in case of any adverse news or volatility in the quality and
performance of that entityar.
Government Loan Guarantees
 e US VA Loan Guarantee Programme
A typical government loan guarantee programme is run
by the already mentioned US Department of Veterans
Table 9.5 Schedules of VA Guarantees (2002)
Loan Amount Guarantee Amount
Up to USD 45,000 50%
USD 45,001-USD 56,250 USD 22,500
USD 56,251 – USD 240,000  e lesser of USD 36,000 or 40% of the loan
USD 240,001 or more  e lesser of USD 60,000 or 25% of the loan
Source: Lush & Sirota (2003)au

77
The Challenges of A ordability, Accessibility and Sustainability
Aff airs.  e operations are managed by 55 regional offi ces
located throughout the country. Loan guarantees are
granted to eligible veterans.  e programme focuses on
guaranteeing loans made by private lending institutions
such as commercial banks, S&Ls, mortgage banks and
insurance companies. If a borrower defaults on a VA-
guaranteed loan, the VA compensates the lender for any
losses incurred in the foreclosure and subsequent sale of
the property, up to the maximum amount guaranteedas.
 e Characteristics of VA Loan Guarantees
VA loan guarantees are designed to provide veterans with
up to 100 percent fi nancing of their home purchases.
However, unlike the FHA, which insures the total
amount of the balance of a loan, VA loan guarantees
only cover the top portion of the losses, which is where
the greatest risk lies when a borrower defaults.  erefore,
VA loan guarantees off ers the type of protection that
lenders most want. Table 9.5 shows the schedule of VA
guarantees. An eligible veteran can use a VA guarantee
as down payment and obtain a 100 per cent loan that
is four times the remaining eligible amount (currently
USD 240,000, i.e.,USD 60,000 x 0.25)
at
.
Partial Entitlement
Since VA-guaranteed entitled amounts change over
time (see Table 9.6), a veteran who has used his/her
benefi ts in the past may still be eligible for another VA
loan guarantee if s/he has any remaining entitlement.
To determine the remaining entitlement, simply
subtract the amount used previously from the amount
currently outstanding. To calculate the maximum loan
allowed with this partial entitlement, use 75 percent of
the sales price or appraised value, whichever is the lesser,
and add the remaining entitlement to this amount. For
example, a veteran bought a house in 1988 using USD
36,000 of his entitlement. In 2002, the veteran decided
to purchase a larger house but still kept the original
residence.  e new house was priced at USD 180,000.
 e remaining VA entitlement was USD 24,000 (USD
60,000 – USD 36,000).  erefore, the maximum loan
available is USD 159,000 (USD 135,000 + USD
24,000)
av
.
Loan Guarantees by Foreign Agencies
Loan Guarantees by the Asian Development Bankax
 e Asian Development Bank (ADB) aims to promote
prudent, long-term debt fi nancing and the expansion of
loan options to borrowers in the Asia-Pacifi c region. ADB
guarantees a portion of the risks to improve the credit of a
borrower to a level acceptable to lenders.
Types of Guarantees – ADB provides two types of
guarantees: (1) a partial credit guarantee; (2) a political
risk guarantee.
Coverage – ADB extends guarantees either as primary
or secondary obligor. ADB guarantees can cover a wide
variety of debt instruments, including syndicated loans
from commercial banks, capital market instruments such
as bonds and fl oating rate notes, letters of credit, as well
as debt instruments where the government is the primary
obligor. ADB may guarantee debt instruments with
attached options and warrants to purchase equity and
convertible bonds (provided that the entire ADB guarantee
terminates upon conversion into equity of all or part of the
debt instrument or convertible bond).
Table 9.6 Schedule of VA Loan Guarantee Periods
Loan Amount Guarantee Amount (USD)
December 1945 4,000
July 1950 7,500
May 1968 12,500
December 1974 17,500
October 1978 25,000
October 1980 27,500
March 1988 36,000
December 1989 46,000
October 1995 50,750
January 2002 60,000
Source: Lush & Sirota (2003)aw

78 Housing for All
Risk Sharing – ADB guarantees are guided by the principles
of maximizing risk sharing with other co-fi nanciers
while keeping the credit enhancement provided by ADB
at a minimum. In practice, ADB only provides partial
guarantees, covering part of a loan or specifi c risk events.
 e structure and degree of the coverage must be set at
the lowest level required to mobilize the fi nancing needed.
Commercial co-fi nanciers must bear some or all of the risk in
each repayment period. ADB guarantees are only provided
to those projects that cannot secure loans from commercial
fi nancial institutions without ADB guarantees.
Terms and Conditions – ADB guarantees are subject to
the following terms and conditions: (1) ADB will make
payment under a partial credit guarantee due to a payment
default covered by the guarantee. (2) ADB reserves the right
to terminate the liability by purchasing its entire guaranteed
obligation of principal together with accrued interest. It
will not allow benefi ciary lenders to accelerate payment
under the guarantee. (3) ADB requires a counter-guarantee
for guaranteed loans to a member country or government
executing agency. Under the counter-guarantee, the
government guarantees payment of all guarantee fees and
charges payable by the borrower to ADB. (4) ADB requires
that the proceeds of guaranteed loans and bond issues be
used for the specifi c purposes of the project and applied to
ADB procurement requirements.
Loan Guarantees by the US Development Credit Authorityay
 e Development Credit Authority (DCA) is the US
legislative authority that permits USAID to issue partial
loan guarantees to private lenders. DCA guarantees
require risk-sharing where the USAID share of a lender’s
risk does not exceed 50 per cent (except for those specially
approved by the Credit Review Board). DCA seeks high
leverage ratios for their loan guarantees. For example, on
average, every dollar of USAID contingent liability under
DCA guarantees is expected to cost two to seven US cents,
which is much lower than making direct loans.
Conditions for Loan Guarantee by DCA – Under the DCA
scheme, borrowers select their own lenders. USAID reserves
the right to approve the lender selection procedure, the actual
lender(s) and the terms of loans guaranteed such as the interest
rate.  e agency also reserves the right to refuse a guarantee
if it believes that the terms are not suffi ciently advantageous
to the borrower. Although USAID may guarantee payment
or collection depending on specifi c needs, the agency prefers
guarantees of collection as a way of providing incentives
for the guaranteed party better to manage the collection
process. DCA guarantee can be on either dollar- or non-
dollar denominated debt. However, in case of non-dollar
denominated debt, the total USAID contingent liability
must be capped or limited in dollar terms.
Internal Credit Enhancement
Senior/Subordinated Structure
 e most common form of internal credit enhancement
is the senior/subordinated structure, which has become
popular particularly in securitization programmes.
Subordination means that the cash fl ows generated by
the borrower are allocated with diff erent priorities to the
various classes of debt holders.  e cash fl ows generated by
the borrower and his/her assets are fi rst used to repay the
principal and interest, while the subordinated piece (also
called “equity” piece) receives cash fl ows in subordinate
order but absorbs the losses fi rstaz.  e results are that
repayment for the senior class of debt is unaff ected by the
absorption of all losses by junior classes of debt. Junior debt
typically trades at high yields to compensate investors for
the risks attachedba. For example, take a USD 200 million
securitization programme with USD 140 million of class A
securities, USD 40 million of class B certifi cates and USD
20 million of class C certifi cates. In order for any class to
obtain an AAA credit rating, its loss coverage must be equal
to or greater than 10 per cent. Subordinating the USD 20
million worth of class C certifi cates to classes A and B,
both class A and class B securities can obtain AAA ratings.
Class A may become ‘super senior’ due to payment priority
over both class B and class C certifi cates if it features a 30
percent loss coverage ratio.  is additional loss protection
entails an AAA rating for class A securities which, being
‘super senior’, sell at a premium to investorsbb.
Over-Collateralisation
Another common type of internal credit enhancement
is over-collateralisation, where a Special Purpose Vehicle
(SPV) holds assets of greater value than subsequent debt
issuance. Under collateral structures, the cash fl ows on the
underlying assets should be large enough to cover outgoing
paymentsbc. For example, a trust has a market value of
collateral trust assets of USD 150 million.  e trust issues
two tranches. Tranche A is the senior tranche consisting of
USD 125 million worth of securities. Tranche B consists
of USD 25 million worth of subordinated securities and is
repaid after the senior tranche is repaid in full.  erefore,
the senior tranche is over-collateralized by 120 per cent.
 e funds used to purchase the over-collateral for the
senior tranche are from the subordinated tranche within
the trust structure.  is is why over- collateralization is a
form of internal credit enhancementbd.
Excess Spreads
Another type of credit enhancement is known as excess
spread.  is means that there is a large diff erence between
the amount of cash fl ow coming into the special-purpose
vehicle and the amount of debt servicingbe. In practice,

79
The Challenges of A ordability, Accessibility and Sustainability
this can be the excess spread of the loans held in the SPV
and the interest due on SPV-issued securities.  e excess
spread may increase because SPV assets are of lower credit
quality than SPV-issued securities, and therefore yield
higher interest rates. A higher yield on SPV assets may also
result from other factors such as diff erent maturitiesbf.
Cash Collateral
Cash collateral includes cash, negotiable instruments,
accounts receivable, deposit accounts, reserve accounts
or other cash equivalents subject to lender preferences.
Sophisticated lenders may agree to enter into cash
collateral agreements because immediate liquidation of
the borrower’s assets can reduce the amounts realized
and maximize the potential for having an unsecured
claim. Lenders may also consider entering into cash
collateral agreements for limited periods in order to give
borrowers time to explore alternatives that are in the
interests of both lenders and borrowers
bg
.
Excess cash consists of highly rated, liquid instruments
such as Treasury securities or high-grade commercial
paper (CP) that provide security to a creditor. Cash
proceeds received by an SPV from the sale of such
securities are used to purchase the underlying collateral
and the reserve account. However, cash reserves often
earn lower rates of return than those required to fund
SPV securities, and therefore are not an effi cient form of
credit enhancement
bh
.
Mortgage
For large housing loans, a lender will ask for security over
the property that a borrower wishes to buy, or some other
property that the borrower owns.  e requirement for
security will be stated in the loan approval documents.
Mortgage is the most common form of security. A lender
can take a mortgage over most forms of property, including
houses, land, motor vehicles, boats, household goods and
appliances, shares and life insurance policies. In the case of
housing loans, the mortgage is normally housing or estate
property. If a borrower defaults on his/her loan, the lender
can take possession of the mortgaged property and sell it
to repay the outstanding balance on the loan. Normally,
the lender will register the mortgage, which will protect
him/her against claims made by others in respect of the
mortgaged propertybi.
Mortgage loans can be divided into two types –conventional
and non-conventional. A non-conventional loan is backed
by the full faith and guarantee of the government. It is
provided (in the USA) by central government agencies
such as the Federal Housing Administration (FHA), the
Veterans Administration (VA) and the Rural Development
Administration (RDA). Conventional loans do not come
under any formal government guaranteebj.
 e USA featured 67.67 million homeowner properties in
2001, of which 43.64 million were mortgaged, accounting
for 64.5 per cent of the total. Split by categories, there were
Table 9.7 Mortgages for Homeowner Properties in the USA in 2001 (unit:,000)
Mortgage Status
All
properties
(owned
and
rented) Total
1 housing
unit
2 to 4
housing
units
Condo-
minimums
Mobile
homes
All properties 83465 67671 56960 1087 3883 5741
Mortgaged* 50570 43636 37968 643 2437 2588
Non-mortgaged 32896 24035 18992 444 1446 3153
MORTGAGED PROPERTIES
Type of First Mortgage Instrument
Fixed-rate, level-payment mortgage
37541 32900 29850 478 1845 727
Short-term with balloon payment
mortgage
2333 1869 1695 17 101 55
Reverse mortgage 11 11 11 0 0 0
Adjustable rate mortgage (ARM) 6474 5119 4489 109 400 121
Other 4211 3737 1923 38 91 1685
* Includes properties with home equity and/or instalment loans
Source: US Census Bureau

80 Housing for All
mortgages on 66.7 per cent of one-housing unit properties,
59 per cent of two- to four- housing unit properties, 62.7
per cent of condominimums and 45 per cent of mobile
homes. Of the combined 83.47 million rental and owner-
occupied properties, 50.57 million were mortgaged.
Most fi rst mortgages (37.54 million) were fi xed-rate and
only 6.47 million were adjustable-rate mortgages. Only
4.6 percent were short-term balloon payment mortgages
(Table 9.7).

81
The Challenges of A ordability, Accessibility and Sustainability
The Development of Housing
Finance Institutions in Indonesiabk
Introduction
Indonesia’s long-term development is based on several fi ve-
year plans (‘PELITA’) com mencing in 1969. From the
fi rst development plan (PELITA I, 1969/70-1973/74)
onward, housing became a major component and policies
were laid out emphasiz ing research and development
on the technical aspects of housing. During PELITA
II (1974/75-1978/79), three important institutions were
established: the National Housing Board, the National
Housing Corporation (PERUMNAS), and the National
Savings Bank (BTN), a government-owned institution
that was to focus on housing mortgage fi nanc e for low-
income people. PERUMNAS was assigned to build 73,000
houses during PELITA II.
PELITA III (1979/80-1984/85) saw the construc tion
of 150,000 houses and the establishment of PT Papan
Sejatera (PAPAN), a government-sponsored financial
institution engaged solely in housing mortgage loans for
the middle-income segment of the Indonesian population.
Fur thermore, 300,000 new houses were planned under
PELITA IV (1984/85-1988/89), while PELITA V (1989-
1994) was expected to add another 450,000, of which
330,000 units were for the private sector to build while
the balance were to be built by PERUMNAS.
 e fi rst year (1989) of PELITA V saw the construction of
about 100,000 units, but prospects for the rest of the period
looked rather bleak. Financing for construction of houses
as well as for mortgages had become scarce and made the
450,000 target unattainable. In any case, 85 per cent of
housing development in Indonesia is in the hands of
informal-sector developers/builders. Indonesia’s housing
fi nance system is illustrated in Figure 13.1.
Indonesia’s Financial Market
 e fi nancial market in Indonesia has been dominated by
fi ve government-owned commercial banks: Bank Bumi Daya
(BBD), Bank Dagang Negara (BDN), Bank Ekspor Impor
Indonesia (Bank EXIM), Bank Negara Indonesia 1946
(Bank BNI), and Bank Rakyat In donesia (BRI). On top of
these there is one development bank, Bank Pembangunan
(BAPINDO), and a savings bank, Bank Tabungan Negara
(BTN) mentioned ear lier. More than 100 private banks
also operate in the country, along with about a dozen foreign
bank branches and some 20 regional development banks. In
addition to banks, the fi nancial system also includes 12
non-bank fi nancial institutions (NBFIs), which it had been
proposed to turn into commercial banks in 1992, based on
the draft of the the new Banking Act.  e operations of
these banks and NBFIs are controlled by Bank Indonesia,
the country’s central bank.
Over the past two decades or so, a number of fi nancial
regulations have been introduced to improve the operation
of Indonesia’s fi nancial markets. In October 1988, a
regulation looked to improve mobilization of funds,
increase the operating effi ciency of fi nancial institu tions,
improve the eff ectiveness of monetary policy, and accelerate
the development of the capital market. Under this policy,
domestic banks were allowed to open branch offi ces
throughout Indonesia. NBFIs, foreign banks and joint-
venture banks were al lowed to open offi ces in major cities,
namely, Jakarta, Bandung, Semarang, Surabaya, Medan,
Ujung, Pan-dang and Denpasar.  e formation of new
private national banks, rural credit banks and joint-venture
banks was made easier than before.
As a result, the number of commercial banks increased from
122 to more than 150 at the end of 1990. Bank branches also
expanded substantially from about 1,900 at the end of 1988
to more than 3,300 at the end of 1990.  is helped to generate
funds for the banking sector. Additionally, private fi nancial
institu tions were allowed to take deposits from government
enterprises, which no longer had to be held exclusively in State-
owned banks. In other words, the mobilization of funds was
deregulated through elimination of some of the restrictions
im posed on private fi nancial institutions.
In 1990, Bank Indonesia introduced several new measures to
improve the credit system, focusing on spreading bank credit to
a wider spectrum of small businesses (PAKJAN 1990). Each
bank (except for foreign banks and joint-venture banks with
export-credit obligations) was required to grant at least 20 per
cent of total loans either to small businesses (‘Kredit Usaha
Kecil’ or KUK), or to would-be homeowners for the purchase
of small- to medium-sized houses. Banks were required to
comply with this regulation within one year of enactment.
In order to meet the new KUK requirement, a number of
private and State-owned banks launched – or expanded –
residential mortgage loan
programmes.  e
upshot was that
in the fi rst quarter of 1990, and on top of the two fi nancial
institutions already in the business (BTN and PAPAN), as
many as 18 banks, including a foreign one, were involved
Chapter 10: Case Studies of Housing Finance
Institutions and Systems

82 Housing for All
Figure 10.1 The Institutional Structure of Housing Finance in Indonesia

83
The Challenges of A ordability, Accessibility and Sustainability
in residential mortgage fi nancing in Indonesia.  ese banks
could originate loans directly or lend indirectly through
rural credit banks (Bank Perkreditan Rakyat-BPR) or via
other fi nancial institu tions which originated the loans and
acted as channelling agents.  e mortgages were then held
in the bank’s portfolio.  is situation lasted only up to the
beginning of 1991, when the government imposed a tight
monetary policy (PAKTRI 1991).
Sources of Funds
During the early 1980s, funds for home mortgage fi nance
depended heavily on subsidized loans from the Indonesian
government and, to a lesser extent, on fi ve-year bond issues.
It was unfor tunate that banks that were holding savings as
well as time-deposit funds became involved in residen tial
mortgage lending only when the KUK requirement forced
them to do so. However, evidence showed that only a small
portion of the banks’ portfolios consisted of mortgage loans.
 is kind of lending bears two types of signifi cant risks for
lenders: maturity mismatch risk and interest-rate risk.  e
fi rst risk occurs because the majority of deposits with the
banks are short-term. Borrowing short and lending long
entails con siderable risk and causes liquidity problems.
Interest-rate risk arises because fi xed-rate mortgage loans
leave the lender exposed to fl uctuations in the cost of
funding.  is is why, com pared with the other assets of these
banks, mortgage lending is small, so that the risk to each
bank remains manage able.  e government-imposed tight
monetary policy in 1991 further discouraged banks from
residential mortgage lending, which at the present moment
has practically vanished.  is leaves only BTN and PAPAN
as the two fi nancial institutions still extending mortgage
loans. Mortgage interest rates vary with the various product
designs, for instance ranging between 12 and 27 per cent in
the early 1990s.
 e critical issue for housing fi nancing is related to the
absence of a long-term funding market for mortgage
lending in Indonesia.  ere seems to be a compelling case
for the government to put in place a system that provides
a steady and continuous fl ow of funds into housing
fi nance. One solution would be to launch a secondary
mortgage market as a sub-system within the national
financial framework.  is would call for the creation of
secondary mortgage institutions to pur chase mortgages
from originators, encouraging a greater number of these
to par ticipate in residential mortgage fi nancing. Secon-
dary institutions would require government sponsorship
and regulation. Funding could emanate from banks,
pension funds, insurance companies and other institutions
generating steady fl ows of funds from their operations.  is
could only happen under strong regulatory measures.
Conclusions
Indonesia’s residential mortgage financing deserves
serious consideration. Prior to PAKTO 1988 and PAKJAN
1990, the country’s fi nancial system did not provide suffi cient,
if any, infrastructure for chan nelling funds into housing
mortgage fi nance. Any long-term funding that could be
raised was only through fi ve-year bond issues.  ese were not
substantial, though, and competition from non-mortgage
banks did little to help mortgage lending institu tions, like
PAPAN, to raise long-term funds. Another crucial factor
was that interest rates on short-term deposits were much
more favourable than those on long-term instruments.
As a result of PAKTO 1988 and PAKJAN 1990, the number
of fi nancial institutions competing for short-term funds grew
substantially. Long-term funds, which were already scarce,
literally disappeared.  e situation has become worse for
institutions in need of long-term money since 1991, when a
tight monetary policy was imposed (PAKTRI 1991).
In 1990, PAPAN successfully sold 27.7 billion rupiahs
(IDR) worth of mortgages to a govern ment-owned bank.
Although this was a direct sale to an investor, it represented
the start of a secondary mortgage market.  e rationale
behind the bank’s purchase of the mortgages was to fulfi l
the 20 per cent KUK requirement. A secondary mortgage
market, which should be part of Indonesia’s fi nancial
system, is still remote, though. Aside from monetary
issues, some legal aspects of the mortgage instruments must
be ironed out fi rst.  e use of legislative and regulatory
power is man datory to create a sub-system that will operate
within the national fi nance system to provide a steady and
continuous fl ow of funds into housing fi nance. One way of
doing this is to create a secondary mortgage market with or
without secondary market institutions serving as mortgage
market intermediaries. It would be useful if Indonesia’s
forthcoming banking act mentioned earlier required banks
to invest a certain portion of loan portfolios in the form of
mortgage securities.
The Development of Housing
Finance Institutions in Nepal
Absence of Formal Housing Finance
In Nepal, one of the major constraints to the supply of new
houses and maintenance of the existing stock is an absence of
organized sources of housing fi nance, and the development
of formal hous ing finance institutions in the country
is still at very embryonic stages. So far, housing has
largely been an individual rather than a collective concern.
Consequently, the huge investment required to fi nance
housing is almost all undertaken by private individuals
in informal arrangements.  e role of formal fi nance
in overall housing development is very marginal if not

84 Housing for All
negligible, both in terms of investment and eff ectiveness. For
instance, based on construction-cost data for 1981 and fl oor-
area approximations, it has been estimated that informal-
sector investment in housing in the Kathman du and the
Lalitpur Town panchayat (now municipalities) was around
530 million and 200 million rupees (NPR), respectively.
 ese estimates are for building construction only and do
not include land.
Compared with this huge investment in housing by private
individuals, total government expenditure in a scheme like
the Kuleswar Housing Project was just NPR.35 million,
and over a period of more than 10 years.
Similarly, formal housing credit from fi nancial institutions
is equally limited in both scope and amounts. In Nepal, the
Provident Fund and commercial banks do provide loans to
subscribers and employees, respectively. In 1981, the loans
granted by these institutions accounted for about four per
cent of the estimated funding required to build all the housing
for which permits had been issued. Average loan sizes were
quite small: NPR.59,000 for Nepal Bank; NPR.11,815 for
the Provident Fund; and NPR.40,500 for commercial banks.
Loans this size would only fi nance homes ranging between
seven and 50 square metres; in other words, households
would require additional sources to secure adequate fl oor areas
for their homes.
As a result informal mechanisms have, to
this day, been the major source of housing fi nance in Nepal.
Accurate data on the various infor mal sources of housing
fi nance are not available, but conversion of existing family
assets such as land or gold is known to be a major one. Other
sources include loans from friends or relatives, money-lenders
and others. How effi ciently, costly and equitably the infor mal
fi nance sector operates is unknown and calls for research.  e
vacant land within the built-up areas in Greater Kathmandu
can partly be accounted for by the absence of housing fi nance
for construction in the country. More households would be
in a position to develop any vacant land they owned if more
housing fi nance was available to them.
Establishing the Nepal Housing
Development Finance Companybl
As proposed in the Seventh Plan, the Ministry of Housing
and Physical Planning established the Nepal Housing
Development Finance Company (NHDFC) in March 1990
under the Finance Company Act1985.  e main objectives
were improving the existing housing-delivery system
and launching new housing schemes through dedicated
loan facilities and other re lated services, in order to meet
growing demand.  e authorized capital of the Com pany
is NPR.100 million, 50 million of which has been issued
and another 30 million paid-up.  e structure of the share
capital is showed in Table 10.1.
 e Board of Directors is comprised of the General
Manager, the chairman of the National Insurance
Company and representatives from the Ministry of
Housing and Physical Planning, Nepal Bank Ltd., Na tional
Commercial Bank, Agricultural Development Bank, and
Nepal Arab Bank Ltd.
NHDFC focuses on the housing sector and grants loans
for four major purposes, as follows: improvement of
existing housing conditions, launching new housing
schemes, supplying housing loan facilities, and providing
other housing-related services in order to meet growing
demand for housing in Nepal. NHDFC’s eight objectives
are as follows:
a) Mobilising fi nancial resources (with approval from
Nepal Rastra Bank) to launch housing schemes and
provide related physical facilities in accordance with
government housing policies.
b) Extending mid- and long-term loans to civil servants,
corporate bodies or any company or individual,
developing plots of lands for residential purposes,
building new homes or purchasing or improving
existing houses; or acting as an intermediary or
guarantee for these purposes.
Table 10.1 Composition of Nepal’s NHDFC Share Capital ( 2006)
Government 10% 50,000 Shares
Rastriya Bima Sansthan 15% 75,000 Shares
Rastiya Banijya Bank 10% 50,000 Shares
Nepal Bank Limited 10% 50,000 Shares
Agriculture Development Bank 10% 50,000 Shares
Nabil Bank Limited 5% 25,000 Shares
General Public 40% 200,000 Shares
Source: NHDFC

85
The Challenges of A ordability, Accessibility and Sustainability
c) Making available housing schemes and physical
facilities for individuals, companies or institutions,
and supply on a lease or hire-purchase basis the
machinery, tools, equipment and other goods required
for housing schemes.
d) Encouraging the establishment of any commercial or
industrial company or institution whose objective is
to the build physical infrastructures required for the
management and development of housing plots and
the construction of housing facilities; and helping
in the unifi cation, consolidation, reorganization,
expansion and increase in the capital of such companies
or institutions.
e) Operating, or assisting in the running of, schemes,
training or other programmes committed to the
development of residential areas.
g) Acting as a fi nancial intermediary, i.e., supplying
capital to borrowers pursuing similar objectives and
mobilizing fi nancial resources through deposits, loans
debentures, bonds, etc.
h) Performing any other function that may be relevant
and contingent to the functions of the company.
Loan Facilities and Performance
NHDFC provides loans for up to 15 years, a type of long
maturity that can alleviate monthly repayments and be
more aff ordable to low-income people. Table 10.2 shows
the types of loans provided by NHDFC. Table 13.3 shows
NHDFC’s fi nancial performance.
Resource Mobilisation
If a housing-fi nance institution is to be effi cient and
eff ective, it needs a continuous fl ow of resources and proper
use thereof. Consequently, resource mobilization forms an
integral part of housing finance. The following avenues
for resource mobilization may, be worth exploring and
examining in the Nepalese context.
As mentioned above, investment in housing by private
individuals has been quite signifi cant up until now, suggesting
a tremendous potential for internal resource mobilization
in Nepal. Such resources could be partly tapped through
some kinds of “savings and loan schemes” to attract and
retain households and individuals.  e schemes could
benefi t from the lengthy experiences of  ailand and other
countries in this regard.
Commercial banks in Nepal may also be encouraged to
channel portions of their incremental deposits in the housing
sector.  eir longer experience in banking, together with their
trained staff , could be put to good use in housing fi nance
as well. Moreover, they can easily spread their businesses
(even in rural areas) without any additional overhead costs
as they already have branches scattered all over the country.
In fact, Nepal Bank (a commercial bank) had, in the past,
Table 10.2 NHDFC Loan Facilities (2006)
Loan Amount Loan Maturity/Monthly Repayment Amount
5 Years 10 Years 15 Years
1,00,000 2,275.31 1,493,11 1,265.24
2,00,000 4,550.61 2,986.21 2,530.48
3,00,000 6,825.92 4,479.32 3,795.73
4,00,000 9,101.23 5,972.43 5,060.97
5,00,000 11,376.54 7,465.54 6,326.21
6,00,000 13,651.84 8,958.64 7,591.45
7,00,000 15,927.15 10,451.75 8,856.70
8,00,000 18,202.46 11,944.86 10,121.94
9,00,000 20,477.77 13,437.97 11,387.18
10,00,000 22,753.03 14,931.07 12,652.42
11,00,000 25,028.38 16,424.18 13,917.66
12,00,000 27,303.69 17,917.29 15,182.91
13,00,000 29,578.99 19,410.40 16,448.15
14,00,000 31,854.30 20,903.50 17,713.39
15,00,000 34,129.61 22,396.61 18,978.63
Source: NHDFC

86 Housing for All
tried to provide housing loans on the basis of collateral,
usually land, with a higher interest rate and shorter maturity.
 is scheme could not last long, for three main reasons:
the gpvernment’s absence of long-term fi nancial policy
and inability to comprehend the full import of housing
in the national economy, as well as the bank’s own bitter
ex periences in the loan repayment collection.
The Housing Finance System
in the Philippinesbm
Introduction
 e Philippines faces serious problems with regard to the
extent and quality of its housing stock. Demand for
affordable housing units continues to grow in response
to demographic expansion and larger household sizes, in
both urban and rural areas. However, aff ordability poses
a challenge in view of low incomes, inadequate supply
of suitable homes and limited access to home fi nancing
facilities.  e following major factors lie behind the
Philippines’ shelter problem: housing need, housing
aff ordability, home-ownership and land tenure, provision
of basic services to urban households, type of housing
materials, and geographic mismatches between housing
needs and stocks. Housing needs for the period 1987-1992
were estimated at 3.4 million units – 1.8 million units for
rural and 1.6 million for urban areas.
Housing aff ordability remains a problem due to relatively
low household incomes, with a national monthly average
of 2,960 pesos (PHP) (or USD 110). In 1988, the
average monthly family income was PHP 4,420 (USD
164) for urban and PHP 2,072 (USD 77) for rural areas.
 e amount avail able for housing was estimated on the
basis of the current expenditure for housing, rent or its
equivalent, plus savings, and stood at an average PHP 740
(USD 27) monthly, or 23 per cent of monthly income.
 is is aff ordable only to households belong ing to the top
50 per cent of families on the income ladder.  erefore,
housing remains beyond the reach of a sizeable portion of
the Filipino population.
Home-ownership and security of tenure remain a problem
in urban areas, where only 64 per cent of households own
their homes; in Metro Manila, the proportion is 58 per cent.
Moreover, 22 per cent of Metro Manila households have no
tenure over the land on which their houses are built.
An Integrated Housing Delivery System
Between 1981 and 1985, the Filipino govern ment launched
a ‘total systems approach’ to the fi nancing, production and
regulation of housing. An inter acting network of housing
agencies was established and maintained to fulfi l six main specifi c
func tions: funds generation, mortgage purchases, mortgage
guarantees, regulations, and social hous ing,.  is sharper
government focus on housing sector development resulted in
a sig nifi cant increase in overall housing production.
 e housing-fi nance system put in place in the Philippines
integrated savings, secondary mortgage trading and credit
insurance.  is crucial component of housing delivery
was undertaken jointly by three government agencies: the
Table 10.3 NHDFC Financial Performance (NPR)
Fiscal Year
1999/2000
Fiscal Year
2000/01
Fiscal Year
2001/02
Fiscal Year
2002/03
Fiscal Year
2003/04
Paid-up Capital 3,94,92,500/- 3,94,92,500/- 4,66,40,000/- 4,79,18,000/- 4,80,94,000/-
Gross Deposit 25,41,32,260/56 26,55,23,603/68 30,86,22,846/46 32,90,62,124/07 35,97,48,770/18
Gross Loan 22,61,53,446/77 27,85,53,113/80 31,04,63,095/10 31,59,10,100/95 35,07,57,405/23
Gross Income 4,46,37,766/59 5,25,77,899/71 5,70,75,149/13 5,98,17,228/89 6,47,24,833/05
Gross Expenditure 3,78,85,250/78 4,12,23,609/29 4,38,98,784/12 4,63,79,887/98 4,26,05,470/71
Net Profi t (Before
Income Tax)
67,52,515/81 1,13,54,290/42 1,13,76,365/01 1,34,38,140/91 2,21,19,362/34
Paid Income Tax 41,19,672/26 28,61,807/06 44,57,840/90 39,44,371/02 68,14,684/60
Net Profi t (After
Income Tax)
53,32,843/55 84,92,483/36 87,18,524/11 94,93,769/89 1,53,04,677/74
Per Share Income 13.50 21.50 18.69 19.81 31.82
Dividend
Distribution (In
Percent)
10 13 14 15 18
Source: NHDFC

87
The Challenges of A ordability, Accessibility and Sustainability
Home Development Mutual Fund (HDMF), the National
Home Mortgage Finance Corporation (NHMFC), and the
Home Financing Cor poration (HFC).  e scheme made
home loans accessible to low- and middle-income groups,
channelling savings into PHP 6.2 billion (USD 229.6
million) worth of long-term mortgage funds between 1981
and 1985. In addition, the system generated PHP 1.2
billion (USD 44.4 million) in construction loans and PHP
2.0 billion (USD 74.1 million) in long-term mortgages
from the private banking system through HFC and during
the same period.
The Housing Finance System
In keeping with the objectives of the 1986 Revised National
Shelter Programme, the Filipino housing fi nance system
focuses on lower income groups and a more equitable
distribution of loans.
Housing fi nance is a joint under taking by three agencies in
the Philippines: NHMFC, the Home In surance Guaranty
Corporation (HIGC) and HDMF.
NHMFC’s role is mobilise available long-term funds (the
bulk of which emanates from the Social Security System
(SSS), the Government Service Insurance System (GSIS)
and HDMF) to purchase mortgages originated by both
private and public institutions under the Unifi ed Home-
lending Programme (UHLP).
NHMFC administers UHLP, which integrates the
respective housing loan programmes of SSS, GSIS and
HDMF. UHLP is funded entirely from borrowings from
these three institutions. It was designed to operate on a
self-sustaining basis, with a cross-sub sidy mechanism where
the highest income class pays three per cent more than the
average, enabling the lowest income class to pay three per
cent less. UHLP interest rates are considered to be more
socially equitable.  e three institutions behind UHLP
funding in turn get a guaranteed 10.5 per cent yield, which is
suffi ciently attractive for them to provide abundant funding
for housing purposes. As adjusted in 1991, UHLP loan
packages and inter est rates were as follows: loans up to PHP
150,000 (USD 5,556): nine per cent interest rate; over
PHP 150,000 to P225,000 (USD 8,333): 12 per cent; and
over PHP 225,000 to HP P375,000 (USD 13,889): 16
per cent.
A facility for those employed/income-earning individuals
who are not members of SSS, GSIS or HDMF has also been
launched and made an integral part of UHLP.  rough this
so-called “Social Mortgage Window”, the benefi ts of home-
ownership are made available to the widest possible spectrum
of citizen s. Non-members of the three funding institutions
may avail themselves of loans up to PHP120,000 (USD
4,444) for house and lot, and up to PHP 45,000 (USD
1,667) for home lots.
Through the Home Insurance Guaranty Corporation
(HIGC), the Filipino housing finance system also
provides two types of guarantee: (a) developmental
guarantees extend to developmen tal/construction loans by
banks to private developers undertaking housing projects;
and (b) retail guaran tees on individual home-buyers’
loans.
 ese guarantees make commercial funds avail able to
developers and home-buyers thanks to the tax incentives
granted to the lender and the substitu tion of government
credit for the borrower’s own.  is is how HIGC-g uaranteed
mortgages are turned into government debt and are,
therefore, classifi ed as risk-free.  is makes it easier for
banks to roll over their funds, thereby generating a more
active market.
The Home Development Mutual Fund
 e Filipino Home Development Mutual Fund (HDMF)
was developed and launched as a provident savings
system for housing fi nance. It was created to meet the
need for a fi nancial institution to mobilise savings with
the government propelling the system through decreed
employer counterparts. At the same time, the Fund
addresses the urgent need for aff or dable fi nancing to help
solve the housing problem.
HDMF is mandated to administer Provident Fund
contributions of member-employees and employers;
to channel into housing loans for members any funds that
are surplus to provident benefi ts; and to develop savings
schemes for home acquisition for private and government
employees.
In its eff orts to address the country’s two interrelated
pressing needs – savings and housing – HDMF has
eff ectively mobilised domestic savings for long-term
housing fi nance. On the provi dent aspect, HDMF grants
short-term provident loans under a multipurpose loan
programme that is tied with members’ total accumulated
Fund contributions.
Since inception, HDMF has remained in the primary
business of housing, which contributes the bulk of its
investment portfolio. As one of the key agencies in the
Filipino housing fi nance system, HDMF’s tradi tional role
was that of providing end-user fi nancing.  e Fund opted
to transcend this role in order to address more critical
housing problems, which are such that no one agency can
be expected to provide all the necessary solutions.

88 Housing for All
In order to complement UHLP, HDMF has launched the
so-called Expanded Housing Loan Programme (EHLP),
which operates under more liberal guidelines for housing
loans.  e loan pur poses covered by the Programme
include house construction and/or acquisition, lot
purchase, home improvement, refi nancing of an existing
loan, and redemption of a foreclosed property.
 e critical need for increased social housing for the lowest
30 per cent segment of the Filipino population has pushed
HDMF to become involved in direct housing production.
It has been doing so through the Social Housing
Developmental Loan Programme (SHDLP) and the Group
Land Acquisition and Development Pro gramme (GLAD),
which provide fi nancial assistance for land acquisition and/
or site development for social housing projects.
HDMF makes loans at 12 per cent interest for development
of low-income houses priced between PHP 120,000 (USD
4,444) and 150,000 (USD 5,556). For developments
above this range, the HDMF lends to banks at 13 per cent
and they on-lend the funds at 18 per cent as they bear the
credit risk.
Regular Mortgages
Regular mortgages consist mainly of housing loans under
UHLP; the complementary individual programmes of the
three funding institutions, such as EHLP of HDMF, are
also covered through the Social Mortgage Window of the
NHMFC.
Regular mortgage take-outs totalled 106,468 between 1987
and 1990, with an average annual growth rate of 33.87 per
cent. Total fi nancial assis tance aggregated to PHP 15,259
million (USD 565 million) over the four-year period.
In line with the Filipino government’s policy of im-
plementing a continuing programme for social hous ing,
the Abot-Kaya Pabahay Fund was established in 1990
under Republic Act N
o
6846.  is Fund amounts to PHP
2.5 billion (USD 92.3 million) and was to be built up
over a fi ve-year period through annual appropriations of
PHP 500 million (USD 71.4 million) from the national
government.  e rationale is to provide amortisation
support, expedite land development for social housing
through develop ment fi nancing for developers of low-
cost housing projects, and to establish a strong guarantee
system for the funding agencies involved in housing.
Development Loans
 e Social Housing Developmental Loan Pro gramme
seeks to address the dearth in the production of low-income
housing despite availability of funds for low-income
mortgages.  e Programme ex tends low-interest loans to
public and private developers and landowners, in a bid to
encourage them to produce aff ordable housing for low-
income families.
From 1988 to 1990, a total 31,665 housing units were
completed by both public and private developers with the
assistance of NHMFC and HDMF.  is amounted to PHP
2.1 million (USD 77,800) in developmental fi nancing.
 e Filipino government also encouraged private fi nan-
cial institutions to extend construction loans through the
HIGC Development Guarantee Programme, which covered
16,866 units from 1988 to 1990.
Community Mortgage Programmes
 e Filipino Community Mortgage Programme (CMP) was
set up to enable urban poor communities to acquire land,
develop infrastructures and build or improve their homes.
CMP loans enable community-based organizations to acquire
and develop land on behalf of members, issue individual land
titles and provide individual loans for housing construction
or improvement – all under the concept of community
ownership.  e primary objective of the programme is to
help residents in blighted areas to gain ownership of the lots
they occupy, or to relocate to another area, and eventually
improve their neighbourhood and homes to the extent of
their fi nancial abilities. Under the scheme, residents can
only obtain a loan if they are supported by an NGO or a
government agency
bn
.  e CMP was launched in 1988 and is
ad ministered by NHMFC. Community-based organizations
or co-operatives apply for CMP loans through the National
Housing Mortgage and Finance Corporation (NHMFC).
CMP addresses the problem of security of land tenure for the
landless urban poor in three distinct stages.  e fi rst stage
allows for land acquisition by the community; the second stage
provides fi nanc ing for the ‘horizontal’ development of the
acquired property and the individual titling of lands; and the
fi nal stage involves home improvement or house construc tion.
CMP is totally funded by budget appropriations from
the Filipino Treasury. However, Treasury funds are not
always released on time. In 2002, CMP was still overdue
the remaining balance of its 1999, 2000 and 2001 budget
allocations, and the one for 2002 was reduced to PHP 300
million (USD 60 million). In eff ect, CMP allows slum
dwellers to legalise their status without requiring them
to provide collateral. By 2001, CMP had helped 114,911
poor households in 854 communities to secure rights to
housing and tenure. On average, it has assisted about 9,000
households a year
bo
. CMP has achieved the highest collection
effi ciency rate of all government housing loan programmes.
 e average loan amount was USD 665 per household
bp
.
 anks to its inherent advantages CMP is widely recognised,
capturing the interest not just of government spheres but also

89
The Challenges of A ordability, Accessibility and Sustainability
of a number of local authorities as well as community-based
and non-governmental organizations (NGOs). However,
its potential as a major government programme remains
untapped. To date, a substantial volume of community
projects are still under process. NHMFC has opted to exercise
prudence over loan volumes in view of various factors, which
suggests that institutional strengthen ing is in order if the
quality of the CMP portfolio is to be maintained. Although
CMP provides loans to an average 9,000 households a year,
it is far from solving the low-income housing problem on its
own, due to the fast-growing numbers of poor households in
need of homes. Furthermore, the lowest income households
cannot access CMP.  e poorest 20 percent segment can only
access informal fi nance.  e low-income groups between the
20th and 50th income percentiles can aff ord CMP loans,
and median- and high-income segments can resort to formal
housing fi nance (see Table 10.4).
Strengthening the Filipino Housing
Finance System through Integration
Despite the presence of well-developed and spe cialized
institutions in the country, housing fi nance in the
Philippines has not yet been fully in tegrated into the
overall fi nancial system.
Table 10.4 Income Groups and Housing Financing Schemesbq
Income Group Cumulative %
Of Population
Ave. Annual
Income
Ave. Annual
Housing Expend.
Housing
Delivery
Mechanism
Financing
Scheme
<10,000 0.5 7,902 799 Informal Market Informal credit/
Self-fi nancing
10,000- 19,000 3.9 16,107 1,536 Informal Market Informal credit/
Self-fi nancing
20,000-29,000 11.9 25,330 2,091 Informal Market Informal credit/
Self-fi nancing
30,000- 39,000 22.3 35,063 2,864 Informal/ formal
market
Informal credit/
self-fi n/CMP
40,000- 49,000 32.3 44,881 3,863 Informal/ formal
market
Informal credit/
self-fi n/CMP
50,000- 59,000 40.5 54,854 4,971 Informal/ formal
market
Informal credit/
self-fi n/CMP
60,000- 79,000 53.4 69,492 6,822 Formal market Private formal
lenders/CMP*/
/SSS/GSIS/
HDMF
80,000- 99,000 62.3 89,429 9,435 Formal market Private formal
lenders/CMP*/
SSS/GSIS/
HDMF
100,000-149,999 76.9 122,409 15,194 Formal market Private formal
lenders/SSS/
GSIS/HDMF
150,000-249,000 90.0 191,141 22,208 Formal market Private formal
lenders/SSS/
GSIS/HDMF
250,000-499,000 97.5 330,041 51,937 Formal market Private formal
lenders/SSS/
GSIS/HDMF
500,000 and over 100.0 996,047 134,273 Formal market Private formal
lenders/ SSS/
GSIS/HDMF

90 Housing for All
The preceding discussion focused on the government’s
housing programmes and housing-fi nance system, as this
is treated separately from the types of fi nancing provided by
private commercial banks.  ese private institutions off er
fi nancial assistance under more stringent terms and higher
interest rates, as determined by open-market com
petition.
Consequently, these institutions respond to the fi nancial
needs of those in the middle- and upper-income brack ets with
higher aff ordability levels for housing.
In the Philippines, there is absolutely no competition
between government housing-fi nance institutions and
private commercial banks, all of which off er their own housing
and real estate loan programmes.  e two categories ser vice
two totally diff erent markets.  e government provides
much-needed support for the housing needs of the less
privileged groups through increased budget allocations, loan
amortization support and in terest-rate subsidy schemes.
However, eff orts to integrate housing fi nance schemes into
the national fi nancial system are gradually being undertaken
by way of enhancing the acceptability of mortgage-backed
instruments in the Filipino capital market.
On its own, it may be said that the government’s housing
fi nance system has achieved some degree of integration.
UHLP itself integrates the housing programmes of
various housing fi nance institu tions and establishes vital
links with private developers and originating banks.
As one of the housing fi nance institutions, HDMF has been
continuously working towards integrating the government
machinery into the national fi nancial sys tem. Given its
unique and distinct capacity in mobilizing household
savings for long-term housing fi nance, the Fund brings
together both government and private-sector resources to
meet the needs of its members, who represent an entire cross-
section of the population. HDMF has expanded its reach
nation wide with fully decentralized operations.  rough
its various provident and housing-loan programmes, the
Fund brings together members, employers, originating
banks, private developers, non -government organizations
(NGOs) and even local government units (LGUs).
HDMF has, likewise, invested in mortgage-backed
instruments through aggregate purchases of some PHP
120 million (USD 4.44 million) worth of contract-to-sell
receivables from private developers at discounted rates.
 e ultimate objective here is for the Fund to on-sell these
mortgage-backed securities to other private fi nancial
institutions at reasonable rates of return; these eff orts
should, ultimately, result in active trading of mortgage-
backed instruments on the open market, thereby further
increasing the degree of integration of housing fi nance into
the national capital market.
However, the market for mortgage-backed investment in-
struments is still in its infancy in the Philippines.  ese
instruments have not yet gained the full confi dence of either
institutional or individual investors due to several factors. One
perceived reason is the failure of NHMF’s secondary mortgage
market system (SMMS), which ad verse economic and political
developments back in 1983 have not allowed to develop.
Although a revised NSP has reactivated the SMMS through
UHLP, transactions are limited to the home mortgages of the
funding agencies.  is is why investors remain wary of these
mortgage instru ments; another reason could be that the rates
of return on mortgage-backed securities pale in com parison
with other risks and higher yields.
Once acceptability of mortgage-backed securities as
viable investment instruments is estab lished, the Filipino
housing fi nance system can be assured of readily available
funding. More institutions may be lured into purchasing
mortgages once the mechanism that makes this market liquid
proves to be sound and profi table.  e free fl ow of funds
recycled into the system can only mean more fi nancing for
housing production.
Conclusions
Integration of the housing-fi nance system into the Filipino
national fi nancial system will undoubtedly strengthen its
capability to generate more funds for housing production.
However, full integration can only be achieved if a country’s
economic and business climate remains favourable and
interest rates are stable. Un fortunately, this is not the case in
the Philippines where consumer prices continue to fl uctuate
and the infl ation rate, now averaging 17 per cent, is very
much off -course by comparison with the other member
countries of the Association of South-East Asian Nations
(ASEAN).
Given the objectives of the Filipino government’s
housing programme and its focus on the low-income
segment of the population, it may be diffi cult to achieve
full integration. Continued government sup port by way of
legislation, budget appropriations and interest subsidies
for housing dictates the current interplay of forces in the
housing-fi nance system. If the government’s housing
finance programme is to be fully integrated, its interest
rates should refl ect those opn the open market. However,
market rates for housing fi nance are beyond the reach of the
poorest 30 per cent of the Filipino population.
Against this background, only partial integration can be
sought at the moment.  is is what HDMF has been
doing, in line with its mandate and the policy objectives
of the National Shelter Pro gramme.  e Fund recognises
the signifi cance of estab lishing a viable market mechanism
which would bring liquidity to secondary home mortgages

91
The Challenges of A ordability, Accessibility and Sustainability
and recycle them into the general system in order to generate
increased fi nancing for home ac quisition.
 erefore, while HDMF continues to implement its
provident benefi t and housing loan programmes for
members, it is taking gradual steps to integrate hous ing
fi nance into the overall fi nancial system through invest-
ment in mortgage-backed securities. Eventually, the Fund
hopes to establish a viable market mechanism which would
stimulate open market trading of mortgage instruments,
thereby paving the way for its fuller in tegration into the
Filipino capital market.

92 Housing for All
w UN-HABITAT (2005), Towards Poverty Eradication
Goal: Structure and Infrastructure of Microfi nance and
Microcredit Industry in Eastern Africa, Nairobi: UN-
HABITAT
x De Soto H. (2000)  e Mystery of Capital, New York:
Basic Books
y De Soto H. (2000)  e Mystery of Capital, New York:
Basic Books
z Institute for Housing Studies (IHS) (1998) Renting – Its
Place in the Housing Market of Developing Countries,
Rotterdam: IHS
aa GMG (2005) Lack of Property Rights for Women Hampers
MDGs, Gender, Media and Governance, e-newsletter, No.
003, June 7, 2005
ab Qvist E (1995) Women’s Access, Control and Tenure of
Land, Property and Settlement, Paper presented at the
International Workshop on Women’s Access, Control and
Tenure of Land, Property and Settlements, Gavle, Sweden
ac Boumphrey R., Dickie P., and Tukuafu S. (2005) Instill
Credit Culture in State-owned Banks – Experience from
Lao, ADB
ad Boumphrey R., Dickie P., and Tukuafu S. (2005) Instill
Credit Culture in State-owned Banks – Experience from
Lao, ADB
ae GoT & UN-HABITAT (2003) Re-establishing Eff ective
Housing Finance Mechanisms in Tanzania:  e Potentials
and the Bottlenecks, UN-HABITAT
af Luoto J., McIntosh C. & Wydick B (2004) Credit
Information Systems in Less-Developed Countries, paper
presented at a seminar at Princeton University.
ag Freedman P L (2004) Designing Loan Guarantees to Spur
Growth in Developing Countries, USAID
ah Based on data from World Bank database 2005
ai Freedman P L (2004) Designing Loan Guarantees to Spur
Growth in Developing Countries, USAID
aj Ferguson B (2004)  e Key Importance of Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
ak Freedman P L (2004) Designing Loan Guarantees to Spur
Growth in Developing Countries, USAID
al Collard S. & Kempson E (2005) Aff ordable Credit, the
Policy Press
am Daphnis F (2004) “Housing Microfi nance: Towards a
Defi nition”, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
an Ibid.
ao Robinson M S (2001)  e Microfi nance Revolution,
Washington DC: World Bank
ap Daphnis F (2004) “Housing Microfi nance: Towards a
Defi nition”, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
aq Ibid.
ar Ibid.
as Robinson M S (2001)  e Microfi nance Revolution,
Washington DC: World Bank
at Healey J (ed.) (2000) World Aid and Foreign Debt, Issues
in Society, vol. 134, pp1-38
au World Revolution (2005)  e State of the World, http://
www.worldrevolution.org/projects/globalissuesoverview/
overview2/briefdevelopment.htm
(accessed 15 August 2005)
av Ibid.
aw Robinson M S (2001)  e Microfi nance Revolution,
Washington DC: World Bank
ax Ibid.
ay CGAP (2001) Helping to Build a Microfi nance Industry,
http://www.cgap.org/docs/Brochure_helpingtobuild.pdf
(accessed 15 August 2005)
az Robinson M S (2001)  e Microfi nance Revolution,
Washington DC: World Bank
ba CHF International (2005) Practical Guide for Housing
Microfi nance in Morocco, Silver Spring, MD: CHF
International
bb Daley-Harris S (2004) State of Microcredit Summit
Campaign Report 2004, Washington DC:  e Microcredit
Summit Campaign
bc CUDS (2000) Housing Microfi nance Initiatives, Harvard
University
bd Ferguson B (2004)  e Key Importance of Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
be Klinkhamer M (2000) Microfi nance Housing Products and
Experience with Land Title as Collateral, GHIF
bf Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
bg Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
bh ACCION (undated) Assisting Commercial Banks to Start
Microfi nance Programme, Boston: ACCION International
bi Goodwin-Groen R (1998)  e Role of Commercial Banks
in Microfi nance: Asia and Pacifi c Region, Brisbane: BWTP
bj Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
bk Dichter T (1999) NGOs in Microfi nance, University of
British Columbia
bl UN-HABITAT (2004) Community-based Housing Credit
Arrangements in Low Income Housing: Assessment of
Potentials and Impacts, Nairobi: UN-HABITAT
bm Ibid.
bn Cities Alliance (2003)  e Enabling Environment for
Housing Finance in Kenya, CIVIS, No. 4
bo UN-HABITAT (2004) Community-based Housing Credit
Arrangements in Low Income Housing: Assessment of
Potentials and Impacts, Nairobi: UN-HABITAT
bp Ibid.
bq Ibid.
References

93
The Challenges of A ordability, Accessibility and Sustainability
br Domingo M (1998) Formal and Informal Financial Services
Available to the Urban Poor of Phnom Penh, in UN-
HABITAT and Community Development Programmeme
for Asia (eds) Urban Community-based Financial Services
in Asia, Nairobi: UN-HABITAT
bs Boonyabancha S (2000) Cambodia:  e Urban Poor
Development Fund, Conventry: Homeless International
bt Housing Microfi nance Case Study: HILP and
Condominium Loans in Romania, http://www.coop.org/
housing/activities/housing-practitioners/part-2-chapt1-
romania.pdf#search=%22CHF%20housing%20microfi na
nce%22, accessed 25 August 2006
bu Ibid.
bv Goldberg M & Motta M (2003) Microfi nance for Housing:
the Mexico Case, Journal of Microfi nance, 5 (1), 51-76
bw Ibid.
bx Fehr D & Hishigsuren G (2004) Raising Capital for
Microfi nance: Sources of Funding and Opportunities
for Equity Financing, Working Paper, Southern New
Hampshire University
by Ibid.
bz Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
ca Tulchin D (2004) Positioning Microfi nance Institutions for
the Capital Markets, Social Enterprise Associates
cb Ruth Mcleod (2001) “Experiences of Linking Community-
Based Housing Finance to Formal Finance Mechanisms,”
presentation to Homeless International, March 28, 2001
cc Ruth Mcleod (2001) “Experiences of Linking Community-
Based Housing Finance to Formal Finance Mechanisms,”
presentation to Homeless International, March 28, 2001
cd CGAP (2004) Housing Microfi nance, Donor Information
Resource Centre, www.cgapdirect.org (accessed 19 August
2005)
ce Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
cf Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
cg Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
ch Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
ci Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
cj Almeyda G & Branch B A (1998) Microfi nance in
Guatemala: the Case of Creidt Unions, Madison: World
Council Information Center
ck Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
cl Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
cm CGAP (2004) Housing Microfi nance, Donor Information
Resource Centre, www.cgapdirect.org (accessed 19 August
2005)
cn Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
co Almeyda G & Branch B A (1998) Microfi nance in
Guatemala: the Case of Creidt Unions, Madison: World
Council Information Center
cp Vance I (2004) Land and Collateral Issues, in Daphnis F
& Ferguson B (eds.) Housing Microfi nance: A Guide to
Practice, Bloomfi eld: Kumarian Press
cq Ibid.
cr Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
cs Vance I (2004) Land and Collateral Issues, in Daphnis F
& Ferguson B (eds.) Housing Microfi nance: A Guide to
Practice, Bloomfi eld: Kumarian Press
ct Daphnis F (2004) Elements of Product Design for Housing
Microfi nance, in Daphnis F & Ferguson B (eds.) Housing
Microfi nance: A Guide to Practice, Bloomfi eld: Kumarian
Press
cu CHF International (2005) Practical Guide for Housing
Microfi nance in Morocco, Sliver Spring USA: CHF
International
cv Escobar A & Merrill S R (2004) Housing Microfi nance:
 e State of the Practice, in Daphnis F & Ferguson B (eds.)
Housing Microfi nance: A Guide to Practice, Bloomfi eld:
Kumarian Press
cw Merrill S (2004) Housing Microfi nance: State of the
Practice, the Urban Institute
cx Cities Alliance (2005) FUNHAVI’s Housing Microfi nance
Programme in Mexico, Washington DC: Cities Alliance
cy Ibid.
cz USAID (2004) PRIDE/Finance: Linking an African
Microfi nance Institution to a Formal US Banking
Structure, News Stories, http://www.usaid.gov/gn/nrm/
news/microfi nance/article_long.htm
da Ibid.
db Ibid.
dc Cities Alliance (2002)  e Enabling Environment for
Housing Microfi nance in Kenya, Washington DC: Cities
Alliance
dd Miller J (2003) Benchmarking Latin American
Microfi nance, Washington DC: Microfi nance Information
Exchange
de Sharma A (2005) Developing Sustainable Microfi nance
Systems, ADB, www.unescap.org/drpad/publication/fi n_
2206/part5.pdf (accessed on 2 September 2005)
df Operating and Financial Reviews, http://www.dfl caribbean.
com/OpFinReview.pdf (accessed on 31 August 2005)
dg Okpala D & Zhang X Q (2005)  e Role of Microcredit
in Aff ordable Housing for Low Income Households, in
UN-HABITAT (ed.) Towards the Poverty Eradication
Goal:  e Structure and Infrastructure of the Microfi nance

94 Housing for All
and Microcredit Industry in Eastern Africa, Nairobi: UN-
HABITAT
dh Ibid.
di BRBD (1997) Comparative Study on Credit Delivery Cost
for Rural Poor Projects, http://www.alternative-fi nance.org.
uk/cgi-bin/summary.pl?id=18&sourcelang=E&view=html
(accessed on 31 August 2005)
dj Ibid.
dk Okpala D & Zhang X Q (2005)  e Role of Microcredit
in Aff ordable Housing for Low Income Households, in
UN-HABITAT (ed.) Towards the Poverty Eradication
Goal:  e Structure and Infrastructure of the Microfi nance
and Microcredit Industry in Eastern Africa, Nairobi: UN-
HABITAT
dl Fernando N A (2004) Microfi nance Outreach to the
Poorest, ADB Finance for the Poor, Vol. 5 No. 1
dm Sharma A (2005) Developing Sustainable Microfi nance
Systems, ADB, www.unescap.org/drpad/publication/fi n_
2206/part5.pdf (accessed on 2 September 2005)
dn Okpala D & Zhang X Q (2005)  e Role of Microcredit
in Aff ordable Housing for Low Income Households, in
UN-HABITAT (ed.) Towards the Poverty Eradication
Goal:  e Structure and Infrastructure of the Microfi nance
and Microcredit Industry in Eastern Africa, Nairobi: UN-
HABITAT
do Landaeta G (1994) Strategies for Low-income Housing,
Sweden: Lund University
dp San Luis Obispo County Housing Trust Fund, Permanently
Aff ordable Housing, http://www.slochtf.org/housing-
models.htm (accessed on 10 September 2005)
dq CNHED (2004) A Study of Limited Equity Co-
operatives in the District of Columbia, <http://www.
cnhed.org/image/123800_c_sU127242_s_i189945/
Coop%20Study%20PDF.pdf>
(accessed on 12 September 2005)
dr Ibid.
ds Lopez Community Land Trust, LCLT Aff ordable Housing
Programme, http://www.lopezclt.org/aff ordable_housing/
model.html (accessed on 12 September 2005)
dt Peterson T (1996) Community Land Trusts: An
Introduction, http://www.plannersweb.com/articles/
pet112.html (accessed on 11 September 2005)
du Lopez Community Land Trust, LCLT Aff ordable Housing
Programme, http://www.lopezclt.org/aff ordable_housing/
model.html (accessed on 12 September 2005)
a Ibid.
b Ibid.
c Ibid.
d  is section is based on Zhang, X Q (1998) Privatisation: A
Study of Housing Policy in Urban China, New York: Nova
Science Publishers
e Editorial committee (ed.), China Real Estate (Beijing: New
Times Press, 1992) (in Chinese)
f Data from Kunming Housing Co-operative Headquarters
g Editorial Committee (ed.), China Reform : Real Estate
(Dalian: Dalian Press, 1992), 80 (in Chinese)
h Editorial Department (ed.), Shanghai Housing, Shanghai:
Shanghai Science Popularisation Press, 1993 (in Chinese);
Editorial Board, China Reform : Real Estate (Dalian: Dalian
Press, 1992) (in Chinese)
i Editorial Committee (ed.), China Real Estate passim
j Gatabaki-Kamau R (1985) Co-operative Housing
Development in Kenya, Nairobi: University of Nairobi
k Ibid.
l Ibid.
m Ibid.
n Ibid.
o  is section is based on Zhang, X Q (1998) Privatisation: A
Study of Housing Policy in Urban China, New York: Nova
Science Publishers
p Editorial Committee (ed.), Modern China Urban
Construction, Beijing: China Social Science Press, 1990 (in
Chinese)
q Z. F. Cheng, State Urban Construction General Bureau’s
Vice Director Cheng Zhong Fong’s Speech in National
Urban Employees and Residents’ Housing Building
Experience Conference, 1/12/1980 (in Chinese)
r China’s United Workers Union & State Urban Construction
General Bureau, Report of ‘Public Building with Citizen’s
Financial Contribution’ and Citizen’s Building with Public
Support’, D214:1980 (in Chinese)
s Brown C V & P M Jackson (1996) Public Sector Economics,
4th edition, Oxford: Blackwell
t Dunkerley H B (ed.) (1983) Urban Land Policy: Issues and
Opportunities, Oxford: Oxford University Press
u Harsman B & Quigley J M (eds.) (1991) Housing Markets
and Housing Institutions: An International Comparison,
Boston: Kluwer Academic Publishers
v Brown C V & P M Jackson (1996) passim
w Zhang X Q (2000) High Rise and High Density Compact
Urban Form – the Development of Hong Kong, in Jenks
M & Burgess R (eds.) Compact Cities: Sustainable Urban
Forms for Developing Countries, London: Spon Press
x Castells M, Goh L, Kwok R Y W (1990)  e Shek Kip Mei
Syndrome, London: Pion
y Howenstine E J (1986)  e Consumer housing subsidy
approach versus producer housing subsidies, Cities, 24-40
z Monetary Authority of Singapore (2005) Public Finance
and Monetary Statistics, http://www.singstat.gov.sg/
keystats/economy.html#pubfi n (accessed 20 September
2005)
aa Housing & Development Board (2005) Housing Financing,
http://www.hdb.gov.sg/hdbvsf/eampu100.nsf/0/Home_
Financing.html?open&doc=Financial_Scheme_Mortgage_
Loan.htm&topmenu=0 (accessed on 16 September 2005)
ab Tan S Y (1998) Private Ownership of Public Housing in
Singapore, Singapore: Times Academic Press
ac Wong G K M & Zhang X Q (2003) Government
Intervention and Impact on the Housing Market in
Singapore, Journal of Comparative Asian Development, 2
(1), 115-132
ad Alford R (2003) What Are the Origins of Freddie Mac and
Fannie Mae? Http://hnn.us/articles/1849.html (accessed 23
January 2006)
ae Ibid
af AFQs on Fannie Mae and Freddie Mac, http://www.annaly.
com/ie/ff mfaq.html/ (accessed 23 January 2006)
ag White L J (2004) Fannie Mae, Freddie Mac, and Housing
Finance, Policy Analysis, No. 528, October 2004
ah Ibid
ai US Congressional Budget Offi ce (2001) Federal Subsidies
and the Housing GSEs, http://www.cbo.gov/showdoc.cfm?
index=28418sequence=2/ (accessed 23 January 2006)
aj Ibid

95
The Challenges of A ordability, Accessibility and Sustainability
ak Assisted Housing: Fannie Mae, Freddie Mac and the
Aff ordable Housing Goals, http://www.nlihc.org/
news/5199.htm (accessed 24 January 2006)
al Utt R D (2005) Time to Reform Fannie Mae and Freddie
Mac, Backgrounder, No. 1861, June 20
am Ibid
an Manchester P B (2002) Goal Performance and
Characteristics of Mortgages Purchased by Fannie Mae and
Freddie Mac 1998-2000, working paper No. HF-015, USA
Department of Housing and Urban Development
ao Ibid
ap Utt R D (2005) Time to Reform Fannie Mae and Freddie
Mac, Backgrounder, No. 1861, June 20
aq Ibid
ar  is case is based on Piyatissa S W Bulankulame, E Dias
& M E Joachim (1978) City of Colombo, Sri Lanka:  e
Integration of Squaters into the Mainstream of Urban Life,
in United Nations,  e Role of Housing in Promoting
Social Integration, New York: United Nations
as Conditions in Colombo’s Shanties Highlight Sri Lanka’s
Housing Crisis, http://www.wsws.org/articles/2000/
sep2000/sri-s21.shtml (accessed 6 February 2006)
at Homeless International (2005) Community Savings
Financing Human Settlements, http://www.homeless-
international.org/standard_1.asp?category=3&id=2028&id
=2007&id=263
(accessed on 16 September 2005)
au UN-HABITAT (1989) Mobilisation of Financial Resources
for Low-Income Groups, Nairobi: UN-HABITAT
av CPF Board (2005) CPF Interest Rate, http://www.cpf.gov.
sg/cpf_info/goto.asp?page=News/PressRel/N_15Aug2005.
asp (accessed on 20 September 2005)
aw CPF Board (2005) CPF Statistics, http://www.cpf.gov.
sg/cpf_info/News/Statistics/2005-q2n.asp (accessed on 20
September 2005)
ax UN-HABITAT (1989) passim
ay UN-HABITAT (1987) Global Report on Human
Settlements, Oxford: Oxford University Press
az Ibid
ba Lea, M J & Renaud B (1995) Contractual Savings
for Housing: How Suitable Are  ey for Transitional
Economies?, Policy Research Working Paper, Washington
DC: World Bank
bb Ibid
bc Ibid
bd Ibid
be UN-HABITAT (1987) passim; UN-HABITAT (1989)
passim
bf UN-HABITAT (1989) passim
bg Ibid
bh Current Economy Profi le of Turkey, <http://www.photius.
com/countries/turkey/economy/index.html> (accessed on
20 September 2005)
bi Zhang X Q (2000)  e Restructuring of the Housing
Finance System in Urban China, Cities, 17 (5), pp.339-
348
bj Ibid
bk Joint Center for Housing Studies of Harvard University
(2004)  e State of Mexico’s Housing – 2004
bl Murai T (2004)  e Foundation of the Mexicon Welfare
State and Social Security Reform in the 1990s,  e
Developing Economics, XLII, pp. 262-87
bm Babatz G (2006) Mexico, paper presented at the Conference
on Housing Finance in Emerging Markets, Washington
DC, 15-17 March 2006
bn Joint Center for Housing Studies of Harvard University
(2004) passim
bo Ibid
bp Babatz G (2006) passim
bq UN-HABITAT (1989) passim
br US Congressional Budget Offi ce (1982)  e Mortgage
Subsidy Bond Tax Act of 1980: Experience under the
Permanent Rules, Staff Working Paper
bs Ibid.
bt Ibid
bu Ibid
bv Ibid
bw Ibid
bx Ibid
by Ibid
bz Government Housing Bank Annual Report 2004
ca Ibid
cb Government Housing Bank of  ailand, http://www.reic.
or.th/about/eng/about_ghb.asp (accessed 3 February 2006)
cc Government Housing Bank of  ailand, http://www.nird.
org.in/clic/govt_house.html (accessed 3 February 2006)
cd Government Housing Bank in  ailand, http://www.ghb.
co.th/eng/rate/rate_Deposit_eng.htm (accessed 2 February
2006)
ce Government Housing Bank of  ailand, http://www.nird.
org.in/clic/govt_house.html (accessed 5 February 2006)
cf Ibid.
cg Prachuabmoh K (2006) Experience of the Government
Housing Bank of  ailand, paper presented at the
Conference on Housing Finance in Emerging Markets,
Washington DC, 15-17 March 2006
ch Akuff o A (2006) HFC’s Pioneeing Role and Future
Mortgage Market Expansion in Ghana, paper presented at
the Conference on Housing Finance in Emerging Markets,
Washington DC, 15-17 March 2006
ci Ibid.
cj Ibid.
ck Ibid.
cl Ibid.
cm  is is based on Offi ce of the Deputy Prime Minister
(updated) Rent to Mortgage Housing, London: Offi ce of
the Deputy Prime Minister
cn  is is based on Levy J S & Purnell K (undated), Case
Study: the Community Development Trust, Community
Development Investment Review
co Most of this section is contributed by  ierry Naudin
cp Rojas E (1995)  e IDB in Low-cost Housing:  e First
 ree Decades, IDB
cq Ibid.
cr Ibid.
cs Ibid.
ct Ibid.
cu Ibid.
cv Ibid.
cw  is case is based on ADB (1999) Project Performance
Audit Report on the Low Income Housing Development
Project in the Republic of the Fiji Islands, ADB

96 Housing for All
a World Bank (1998) Project Appraisal Document on Low
Income Housing Project in Algeria, Washington DC: World
Bank
b Treasury Board & Public Works and Government Services
Canada (1997) Guide on Revolving Funds, http://www.tbs-
sct.gc.ca/ accessed 3 February 2005
c ILO (1998) Revolving Loan and Guarantee Funds, <http://
www.ilo.org/> accessed 3 February 2005
d National Development Council (1998), Revolving Loan
Fund Handbook, State of California Department of
Housing and Community Development
e Ibid.
f UN-HABITAT (1991),  e Management of Revolving
Funds for House Improvement Loans, Nairobi: UN-
HABITAT
g Ibid.
h Ibid.
i Ibid.
j Ibid.
k Ibid.
l Ibid.
m OECDD (2004) Safe Drinking Water in Oregon, February
2004
n Ibid.
o Ibid.
p Ibid.
q Fabozzi, F J & Choudhry M (eds.) (2004)  e Handbook
of European Structured Financial Products, Hoboken, New
Jersey: John Wiley & Sons
r Fabozzi F J, Ramsey C & Marz M (eds.) (2000),  e
Handbook of Non-agency Mortgage-Backed Securities,
New Hope, Pennsylvania: Frank J Fabozzi Associates
s Fabozzi, F J & Choudhry M (eds.) (2004) passim
t Steinmetz T C (2002)  e Mortgage Kit, Chicago:
Dearborn Trade Publishing
u Fabozzi F J (ed.) (2001)  e Handbook of Fixed Income
Securities, New York: McGraw-Hill Professional
v Fabozzi F J, Ramsey C & Marz M (eds.) (2000) passim
w McCrea B (2005) Home Buyer’s Question and Answer
Book, New York: American Management Association
x Capone C A (2000) Credit Risk, Capital, and Federal
Housing Administration Mortgage Insurance, Journal of
Housing Research, Vol. 11, No. 2
y Ventolo W L & Tamper R (2002) Mastering Real Estate
Maths, Chicago: Dearborn Real Estate Education
z Steinmetz T C (2002) passim
aa Ibid.
ab O’Hara S & Lewis N D (2003)  e Complete Idiot’s Guide
to Buying and Selling a Home, Indianapolis: Alpha Books
ac Weiss M B (2003) Condos, Co-Ops & Townhomes,
Chicago: Dearborn Trade Publishing
ad Pacione M (ed.) (2002) City: Critical Concepts in the
Social Science, London: Routledge
ae Capone C A (2000) passim
af HUD – FHA Will Cut the Cost of Mortgage Insurance,
news released by HUD Secretary Andrew Cuomo, October
30, 2000, Mortgage Bankers Association
ag Guide to Single-family Home Mortgage Insurance, USA
Department of Housing and Urban Development
ah Ventolo W L & Tamper R (2002) passim
ai Floyd C & Allen M T (2002) Real Estate Principles,
Chicago: Dearborn Real Estate Education
aj Ventolo W L & Tamper R (2002) passim
ak Mortgage Insurance & Securitization, http://www.cmhc-
schl.gc.ca/en/moinin/moinle/moinle_012.cfm (accessed 25
January 2006)
al Ibid.)
am Fabozzi F J, Ramsey C & Marz M (eds.) (2000) passim
an Fabozzi F J & Ramsey C (1999) Collateralized Mortgage
Obligations: Structure and Analysis, New Hope,
Pennsylvania: Frank J Fabozzi
ao Stone, C A & Zissu A (2005)  e Securitization Markets
Handbook: Issuing and Investing in Mortgage and Asset-
backed Securities, Princeton NJ: Bloomberg Press
ap Fabozzi F J, Ramsey C & Marz M (eds.) (2000) passim
aq Ibid.
ar Fabozzi, F J & Choudhry M (eds.) (2004) passim
as Lush M & Sirota D (2003) California Real Estate Finance,
Chicago: Dearborn Real Estate Education
at Ibid.
au Ibid.
av Ibid.
aw Ibid.
ax  e case is based on ADB (2003) Operations Manual,
Manila: ADB
ay DCA (2002) Programming Policy, Washington DC: DCA
az Fabozzi, F J & Choudhry M (eds.) (2004) passim
ba Kimber A (2004) Credit Risk: From Transaction to Portfolio
Management, Oxford: Elsevier
bb Kravitt J H P (ed.) (2004) Securitization of Financial Assets,
New York: Aspen Publishers
bc Kimber A (2004) passim
bd Ibid.
be Ibid.
bf Ibid.
bg Garner D R, Owen R R, Conway R P (1994)  e Ernst
& Young Guide to Financing for Growth, Hoboken, New
Jersey: John Wiley & Sons
bh Kimber A (2004) passim
bi English J W, Hicks B, Hrasky S & Gyles N (2003) Personal
Financial Management, Crows Nest, Australia: Allen &
Unwin
bj Fabozzi F J (ed.) (2001) passim
bk  is section is based on UN-HABITAT publication
bl  is section is based on UN-HABITAT publication and
http://www.nhdfcl.com/Current%20company's%20status.
htm
bm Ibid.
bn Community Mortgage Programme, http://www.unescap.
org/rural/bestprac/mortgage.htm (accessed 17 April 2006)
bo Yu, S. O. (2005) Documentation of the Experience of
the Homeless People’s Federation of Philippines, UN-
HABITAT
bp Porio E (2005) Aff ordable Housing in the Philippines,
Slums and Housing, http://www.id21.org/publications/
infrastructure%20highlights.pdf (accessed 17 April 2006)
bq Yu, S. O. (2005) passim