Download The Household Economy Approach

The Household
Economy Approach
A resource manual for practitioners
John Seaman
Paul Clarke
Tanya Boudreau
Julius Holt
Save the Children is the UK’s leading international children’s charity, working in
70 countries, helping children in the world’s most impoverished communities.
Emergency relief runs alongside long-term development and prevention work to
help children, their families and communities to be self-sufficient.
We put the reality of children’s lives at the heart of everything we do.
We learn from the reality of children’s lives and campaign for solutions to the
problems they face. We gain expertise through our projects around the world
and use that knowledge to educate and advise others.
All references to Save the Children in this manual are to Save the Children UK,
unless otherwise stated.
Save the Children Development Manual No. 6
Published by
Save the Children
17 Grove Lane
London SE5 8RD
UK
First published 2000
© The Save the Children Fund 2000
Registered Charity No. 213890
ISBN 1 84187 029 3
All rights reserved. This publication is copyright, but may be reproduced by any method without fee or
prior permission for teaching purposes, but not for resale. For copying in other circumstances, prior
written permission must be obtained from the publisher and a fee may be payable.
Designed by Annie Wasdell
Typeset by Neil Adams
Figures by Carlton Hutchinson
Editorial: Alan Dingle and Paula McDiarmid
ii
About the authors
The authors worked for Save the Children during the development of the
household economy approach. Currently, John Seaman is Development Director
of the Food Security Unit, Save the Children. Paul Clarke works for the World
Food Programme in Rome. Tanya Boudreau and Julius Holt are with the Food
Economy Group.
iii
● The Household Economy Approach
Acknowledgements
The authors wish to express their thanks to Leslie Adams, Christina Archer,
Cassandra Chapman, Haile Kiros, Ellen Mathys, Gary Sawdon, Buzz Sharpe,
Anna Taylor, Richard Mawer and Sonya LeJeune of Save the Children for their
comments and support with the production of this manual. Particular thanks are
due to Annalies Borrel, Jason Matus (World Food Programme) and Lola
Gostelow (Save the Children) for acting as referees. The authors would also like
to thank Mark Lawrence and Alex King (Food Economy Group), and the
headquarters and overseas staff of the many Save the Children programmes
whose work has made the development of the methodology possible. Thanks are
also due to the European Union for supporting the early development of this
work.
Save the Children would like to thank The Stationery Office for permission
to reproduce extracts from Tables of Representative Values of Foods Commonly Used
in Tropical Countries (see Platt, Annexe 4).
iv
Contents ●
Contents
Abbreviations
viii
Introduction
1
1 An introduction to household economy
7
1
Some definitions
7
2
The household economy
9
3
Household management
16
4
Markets and exchange
19
5
Household expenditure and consumption needs
24
6
Individual and household food requirements
25
2 An introduction to household vulnerability
29
1
The effect of shocks on households
30
2
Compensating for loss of income
33
v
● The Household Economy Approach
3 An overview of the household economy
approach
37
1
Defining food economy areas and populations
38
2
Preparing a wealth breakdown
41
3
Describing households within wealth categories
42
4
The economic context
46
5
Analysis
46
4 Framework and definitions
48
1
Classifying household income
48
2
Household food production
50
3
Gifts and relief
52
4
Income from the exchange of household production and labour
54
5
Savings, assets and reserves
56
6
Distinguishing between wealth categories
57
7
The wider economic context of households
59
8
Units
61
5 Finding the information
vi
62
1
Documentary sources
64
2
Key informants
68
3
Information from the community
71
4
Observation
72
5
Measurement
72
6
How many interviews are needed?
73
6 Methods for getting information
75
1
Tools
75
2
Conducting interviews
81
3
Specific areas of enquiry
4
Seasonal calendars
84
103
7 Checking the accuracy of information
106
1
The limits to accuracy
107
2
Checking for the internal consistency of information
107
3
Avoiding pitfalls
114
8 Putting the information to work
119
1
How complex?
119
2
Using food economy information
120
3
Converting primary data to common units
121
4
Developing a scenario
125
5
The analysis
130
6
Calculations using estimates
143
9 Presenting the results
145
Annexes
1
Computer software
151
2
Energy value of foods
153
3
Data collection forms
157
4
Other sources of information
165
vii
● The Household Economy Approach
Abbreviations
FAO
GIEWS
HEA
NGO
PRA
RRA
UN
UNDP
UNEP
UNHCR
WFP
viii
United Nations Food & Agricultural Organization
Global Information and Early Warning System
Household Economy Approach
Non-governmental organisation
Participatory rural appraisal
Rapid rural appraisal
United Nations
United Nations Development Programme
United Nations Environment Programme
United Nations High Commissioner for Refugees
World Food Programme
Introduction
It is now widely agreed that hunger and famine are best understood not only in
terms of the supply of food, but in terms of people’s ability to gain access to
enough of that food. Even in rich countries where food is in abundant supply,
people can go hungry if they do not have enough money to buy it.
This criterion of access to food has been very useful in explaining why hunger
and famine occur in the way they do. For instance, it explains why, in poor
countries where even in normal times some people do not get enough to eat, the
failure of crops and other production is only rarely followed by starvation. Even
in the face of severe production failure, people can often acquire enough food to
survive by using food reserves or by exchanging savings, livestock and other assets
for food. Even the poorest people, those who have neither assets nor reserves,
may be able to survive by finding additional employment or other sources of
income.
However, the concept of access has proved less useful in practice. Although
the monitoring of food production and exchange (for example, by remote
surveillance or price monitoring) has become routine, it is more difficult to assess
people’s ability to access food. A crop failure, an increase in the price of staple
foods or some other shock may be easy to observe, but it is much harder to
explain how they might affect people’s ability to obtain enough food. We can
only understand this if we understand people’s normal economy: how they
usually make a living; their savings, reserves and assets; and how household
production and labour are exchanged for other goods. For example, a
family heavily dependent on agriculture would be more affected by crop
failure than one that relied more on livestock or wage income. A family with
substantial reserves might easily survive a production shock, whereas a family
without reserves or alternative sources of income might be quickly reduced to
starvation.
Acquiring an understanding of food access poses major technical problems.
1
● The Household Economy Approach
Information is often needed about large populations with diverse economies
often situated in remote areas, and it is needed quickly and at reasonable cost. To
be useful, this information must be sufficiently trustworthy to inspire action, and
it should be capable not just of indicating that people are failing to obtain
enough food, but also of quantifying the problem and suggesting possible
approaches to intervention.
The household economy approach (HEA) was developed between 1992 and
1997 by Save the Children in collaboration with the Food and Agricultural
Organisation (FAO) Global Information and Early Warning System (GIEWS) to
meet this need. The aim was to find a method that could indicate the likely effect
of crop failure or other shocks on future food supply. FAO GIEWS was
interested in a system that could be used on a national scale. In the first phase of
the project, several national databases and a computer program (RiskMap) for
analysing large data sets were developed.*
The HEA has two main parts:
1. A quantitative description of the economy of a defined population, including
all the main factors determining current household income and potential
household income under changed conditions, and how these vary between
households. A standardised set of information is collected which includes an
estimate of how households normally obtain their food and other income;
their expenditure on food and non-food items; their savings, livestock and
other assets; the availability of wild foods; and their access to, and use of,
markets. This information is collected using rapid field methods.
2. A system to analyse the relationship between a shock – for example, crop
failure from drought or a rise in the price of a staple food – and the ability of
households to maintain their food and non-food consumption. In practice,
these relationships can be complex. For instance, a drought may affect
household income directly (for example, by reducing crop income) or
indirectly. As the marketed food surplus falls, food prices may rise, and with
reduced production the demand for food may increase; in addition, asset
prices may fall as households sell their assets to buy food. Qualitative issues
can also be important: the outcome may be affected if the asset being sold
* Supported by the European Union
2
Introduction ●
by a household is a cow, firewood or its own labour. The approach used in
HEA is to model the most likely chain of events linking a shock and the
outcome.*
We cannot, of course, foresee the future with complete accuracy. Models of
complex systems have intrinsic limitations – not least because we can never be
sure that the predicted outcome is what will actually happen. However, the aim
of the HEA is not simply to forecast one particular outcome, but to develop
systematic, well-informed arguments about the most likely range of outcomes in
a given situation.
This approach has several strengths:
●
By developing a systematic argument, it creates a logical and transparent link
between an event, its outcome and the possible responses, which is accessible
to non-technicians and which can serve as the basis of discussion with a wide
variety of people.
●
The need for a minimum “complete” set of information, including all
the relevant variables, ensures that the collection of information is
comprehensive. If for example we do not know about some critical variable
(such as the availability of wild foods), then we must find out.
●
It allows us to make use of information normally available only as estimates.
For instance, crop statistics are often only a rough estimate of actual
production; information on rangeland production, where it is collected at all,
is often little better than anecdotal; and estimates of human population
sometimes still span a wide range. HEA allows scenarios to be developed
using different estimates for each variable.
●
It enables us to analyse complex changes in the economic context. Economic
shocks often involve a variety of factors – such as production changes that
affect different crops and livestock differently, changes in price, or restrictions
on access to markets – that sometimes interact over a period of several years.
For example, famine in Ethiopia in the early 1980s followed several years
of production failure and was exacerbated by war; the 1993 famine in
* The idea that famine can be understood only in terms of a model which includes supply and demand factors follows from
Amartya Sen’s theory of exchange entitlements. HEA also draws on participatory and other existing rapid field techniques
(see Annexe 4).
3
● The Household Economy Approach
●
●
●
Somalia followed widespread conflict and economic disruption; and the 1998
famine in Bahr-el-Ghazal, Sudan, followed a long-standing economic
blockade and was triggered by drought and the displacement of a population
by war.
It provides information in a form that can be used to monitor a situation as
it evolves. If, for example, the analysis suggests that people will have to sell
livestock to maintain their income, and that the price of livestock will
therefore fall, this can be easily observed.
It allows an analysis to focus on the needs of defined groups within
populations not just an aggregate measure which reflects an average reality.
It tells us not only that a problem exists, but also the scale of the problem and
the relative contribution of different factors to this, and so suggests possible
responses.
HEA models are relatively simple to construct, although this is most easily done
using a spreadsheet or other computer simulation.
Other uses of HEA
Famine (apart from the malnutrition of refugee and other populations) is now
quite rare. In the past twenty years acute large-scale human destitution and
starvation have scarcely occurred outside Africa, and within that continent
chiefly in three countries: Ethiopia, Sudan and Somalia.
Starvation is, however, only one measure of the effect of shocks on the rural
economy. Although even in poor countries most people survive most economic
shocks, they do this at a cost. These costs can include temporary hunger as
people eke out available food, the risks of migration to find work, and the
need to forego buying clothing, fuel, soap and other basic household items, and
health care and education. For many of the poorest people, survival is bought
only at the cost of longer-term impoverishment as assets are sold and savings
depleted.
Development agencies increasingly recognise that economic shocks do not
have to be severe or widespread to impede development significantly. Poor
households often devote much of their effort and income to reducing their
4
Introduction ●
vulnerability to economic fluctuations, and therefore are often unwilling to risk
economic change. If shocks occur, development gains may be lost.
As the risk of starvation recedes in most regions, so information needs have
changed. Although it is still important to be able to anticipate the likelihood of
starvation, in many locations the need now is for information that will enable
much finer distinctions to be made between different types of economic effect
on different types of household, which will allow more considered choices about
intervention to be made.
The HEA goes some way towards providing this kind of information. The
effect of a shock on a household’s food supply cannot be understood without an
awareness of the household’s ability to acquire soap, fuel, clothing, education and
other non-food goods. Central to the HEA is the need not just to estimate
changes to a household’s food supply, but also the need to estimate the
household’s need to sell assets or to forego non-food expenditures. The approach
enables a range of possible interventions to be identified, including market
intervention, income support and a reduction of household costs (for example,
by suspending taxation and charges for certain items).*
In more general terms, therefore, the HEA allows a response to be made to
questions of this kind: “How will a change in the economic context of a
household or group of households affect their ability to meet their food and nonfood needs?”
HEA also generates detailed budgets for defined categories of household and
detailed information about: the size of household income; its quality (how much
of the household’s food is from cereals, animal products, etc); how this varies
quantitatively, qualitatively and seasonally within and between locations; and
about non-food expenditures (cloth, soap, taxes, education and health). This
may be useful as background information for many purposes and as the basis for
livelihood analysis, the targeting of the poor and as a contribution to project
design.
The HEA has been widely used in Africa (and to a lesser extent in Asia) in a
* The approach, reflecting its original objective of famine early warning, was originally termed the household food economy
approach. As the method has been increasingly used to understand household non-food expenditure, the term household
economy approach (HEA) has been used in this manual. The emphasis remains on food as, for the poor, this makes up the
greater part of their economic activity.
5
● The Household Economy Approach
range of situations, and we now have sufficient experience to be confident that
the method is a useful addition to crop surveillance and other established
techniques.*
The scope of this manual
In developing the HEA, a primary aim was to find an approach that would
enable information on economies to be routinely included in assessments of
crises, and which could therefore be used by non-specialists. This aim has been
at least partly fulfilled and a large group of people from a range of backgrounds
has now been trained. Nevertheless, the subject has its complexities. In most
situations a degree of judgement is required, and there is no doubt that more
experienced users are likely to produce better results.
The aim of this manual is to describe the components of the HEA and the
practical techniques needed to use them. It is primarily intended as a reference
for people who have already had, or are engaged in, practical training in HEA.
We suggest that readers of the manual who wish to use the approach to make
significant operational decisions (for example, to supply food aid) should also
obtain training. Save the Children is now developing training in field techniques
and information use.
The more specialised use of the HEA over larger geographical areas and for
other purposes, and the use of computer techniques for analysis, are briefly
described in the Annexes.
The emphasis of this book is on the rural economy, with particular reference
to Africa, as it is in this setting that HEA has been most widely used, and on the
more typical situations which are encountered. Experience of the approach
outside Africa and in urban areas is more limited, although it has been used in
Nepal, Afghanistan and Pakistan and in urban areas in Mozambique and
Afghanistan.
* HEA has been widely used for local assessments of drought and other shocks (Kenya, Tanzania, Ethiopia, Democratic
Republic of Congo); as part of the early warning/information systems of Operation Lifeline Sudan and the Somalia Food
Security Assessment Unit, and by the World Food Programme (WFP) Vulnerability Analysis and Mapping System (VAM)
in Afghanistan; for the assessment of the economy and food needs of refugees in many locations in East Africa, Nepal and
Pakistan; for the evaluation of the impact of food aid in Ethiopia and for teaching the elements of rural economy and food
security. Maps of rural economy are being developed by the governments of Mozambique and Swaziland.
6
1
An introduction to
household economy
This chapter introduces some basic concepts concerning household economy
and food supply, as a basis for themes developed later in the manual.
1. Some definitions
The household is the smallest coherent economic unit. It is defined here as a
group of people, each with different abilities and needs, who contribute to a
common economy and share the food and other income from this. According to
location, the size and nature of households may vary widely, from “nuclear”
households (a man, a woman and their dependent children) to units of a
hundred or more people. Intra-household management of production and
income – in other words, how households choose between the different options
for production and income, and how they allocate resources within the
household – is not discussed in this book. A household is most often, but not
necessarily, a family.
Household economy is used in this book to mean the sum of the ways in
which the household gets its income, its savings and asset holdings, and its
consumption of food and non-food items.
Food access (rather than food supply). It is now regarded as more useful to
analyse food security in terms of people’s access to food than in terms of food
7
● The Household Economy Approach
production or supply. This enables us to
understand why, irrespective of the
The World Bank defines food
supply of food, some people get enough
security as “Access at all times
to eat and others do not. The world
to enough food of a sufficient
currently produces enough food to feed
quality to ensure an active
everyone, were this equally distributed.
However, even in a rich country where
healthy life”.
food is plentiful people may go hungry
if they are unemployed and do not have
enough money to buy food. People in a
poor country who have insufficient land to grow food may go hungry if they
cannot find employment that will provide them with money to buy food from
the market.
The concept of access is useful even in situations where the food supply is
insufficient to meet everyone’s needs: in a camp, for example, the people who
have more resources or better social connections will often have access to a
greater proportion of the available food. Looking at access helps to explain how
food is distributed between households. However, we cannot understand access
without also understanding supply.
Poverty and wealth. Poverty can be defined in several ways. A household
experiences absolute poverty when it does not have enough income to obtain the
basic material requirements for a healthy and dignified life; some countries have
designated “poverty thresholds”, above which a household can adequately feed,
house and clothe itself.
Relative poverty describes the
poverty of a household relative to other
Starvation is the characteristic
households in the same community, or
of some people not having
in the same or another country. A
enough food to eat. It is not
household may be relatively poor but
the characteristic of there not
not absolutely poor.
being enough food.
Poverty may also be defined in
Amartya Sen, 1981
qualitative terms. At a given level of
income, a household may have greater
or lesser access to food of adequate
8
An introduction to household economy ●
quality, to health care or to education:
for example, in many rural households
HEA is concerned with
access to animal milk (rather than, for
measures of absolute, relative
example, more cereals) is a major
and qualitative poverty, and
determinant of child growth – two
with wealth.
households with otherwise identical
incomes, one of which has milk and the
other not, are not, in terms of the
welfare of their children, equally poor.
Even a well-off household may lack access to health care if this is not provided
by the state or is too far away. Whatever the level of household income, poverty
may also be defined in terms of the outcome: for example, the nutritional status
or risk of illness or death that has resulted.
In the household economy approach (HEA) the wealth of a household is
defined by its holdings of assets that have a market value: that is, can be
exchanged for money or other goods. In rural areas these typically include cash
savings, gold, livestock and farm implements. As the assets of poor households
are usually accumulated by savings from current income (often over several
generations), there is a trade-off between poverty and wealth: for many
households, wealth can only be accumulated if consumption is minimised.
2.The household economy
Especially in rural areas, the concept of household access to food is complicated
by the fact that households often produce both food and non-food items for their
own use and also obtain food and non-food items by exchanging household
production and labour. A household may also have several sources of income. It
is useful to keep three headings in mind:
9
● The Household Economy Approach
1. Household production, including household labour
2. The exchange of production (and labour) for other goods
3. Consumption.
Each of these may involve both food and non-food items.
Production
Exchange
Consumption
Food and non-food
Figure 1
Households in different locations obtain their incomes in an infinite
combination of ways. For example, “livestock production” may involve cattle,
goats, sheep, camels, pigs, poultry, rabbits, guinea pigs, etc, or any combination
of these. These animals may supply food directly (milk, meat, blood or eggs) or
income via the sale of livestock products (wool, skins, meat, milk and eggs) or
livestock. Each type of livestock will have a different value according to its age,
sex and condition, and in some locations its colour, horn length or shape.
Similarly, paid employment may involve seasonal day labour, seasonal migrant
work within the country or in another country, or some combination of these.
However, if the three main headings of production, consumption and exchange
are kept in mind, the specific features of any economy can be classified whatever
the details of the case.
10
An introduction to household economy ●
Sources of household income: production and exchange
Household production and other income that is directly consumed by the
household may be:
●
from agricultural production. This can be in the form of food crops of various
types or animal products (milk and milk products, meat, eggs and blood)
●
wild foods that are gathered or hunted. These include fruits, berries and roots,
insects, fish, and many types of large and small game
●
non-food products, such as firewood, charcoal, fodder and agricultural
implements.
Household production and labour that are exchanged in the market for other
goods may include:
●
Food and non-food agricultural products. A household may produce more
cereals or pulses than it needs for consumption and exchange the surplus, or
it may exchange production such as livestock or cotton for other foodstuffs
and goods.
●
Non-agricultural products, such as firewood, charcoal, fodder or handicrafts.
●
Household labour, either as paid employment or self-employment.
Opportunities for paid employment vary widely according to the area, from
local agricultural work to overseas employment (for example, in Europe, the
USA or the Gulf States) where some of the income is remitted to the
household. There are many forms of self-employment including petty trade,
transporting goods and ploughing for other households.
Non-market exchange may also occur. Poorer households may receive income
through transfers of food, cash or livestock from better-off households. Gifts are
often but not exclusively between kin, although they may also take the form of
food and other goods distributed as welfare or relief by religious institutions,
government and non-governmental agencies. All societies have mechanisms for
redistributing resources from better-off people to poorer people.
As an example, Figures 2 and 3 overleaf show the sources of household food
and cash income, and their relative importance, in the Rift Valley of Ethiopia.
11
● The Household Economy Approach
Wild foods
5%
Milk/meat
10%
Gift
15%
Food purchase
25%
Figure 2:
Sources of food,
Food crops
45%
poor households,
Rift Valley, Ethiopia
Firewood sales
25%
Petty trade
20%
Figure 3:
Livestock sales
30%
Employment
25%
Sources of cash income,
poor households,
Rift Valley, Ethiopia
Seasonality
Household income usually varies according to the season. During any given year,
different crops will be harvested at different times, and milk production will vary
according to rainfall and availability of grazing; the same is true of gathering wild
foods and fishing. Opportunities for paid employment are frequently seasonal;
people often find work in an area, some distance from the household, that has a
different seasonal pattern (see Figure 4 opposite).
The seasonality of income means that:
●
The sources and types of foods available to the household will differ
according to season. For example, the main crop produced by the household
may not be sufficient to meet its needs throughout the year, and there may
be a period when the household has to reduce its food intake or purchase food
to tide it over until the next harvest.
12
An introduction to household economy ●
●
The level of income will vary according to season, and there may be periods
when households are unable to get enough food to meet their needs.
Jan
Feb March April May June
July
Aug
Sept
Oct
Nov
Dec
Rains
Agriculture
Land clearing
Land preparation
Main planting (maize, beans,
cowpeas, millet, sorghum)
Weeding
Planting (pigeon peas)
Bird scaring
Manure spreading
Harvest
Milk production
Other production
Charcoal/firewood sales
Labour migration
Honey gathering
Hunting (dik dik)
Wild crops
Brewing
Construction
Water sales
Fencing
Brick making
Digging terraces
Figure 4:The seasonal pattern of agricultural and other productive
activities in Lower Kitui, Kenya, 2000 (Save the Children)
13
● The Household Economy Approach
Differences in income levels and the relative importance
of different sources of income
In many countries a large proportion of households are, by any definition, poor,
and many of these live in absolute poverty, being unable – at least seasonally – to
meet their food needs. Yet there is a wide variation in the income levels of
households in the same location or between locations.
Within groups of households where each household has the same economic
opportunities, there will sometimes be large variations in both the size of income
and in the relative importance of different sources of income.
Figure 5 gives an example from Tanzania, in 1999. Here the cash income of
the better-off households, after food needs have been met, is approximately seven
times that of the poorer households.
400,000
Tanzania shillings
350,000
300,000
250,000
Trade
200,000
Agricultural labour
150,000
Firewood
100,000
Brewing
Livestock
50,000
Crop sales
0
Very
Poor
Poor
Lower Middle
Middle
Rich
Figure 5: Relative income of different wealth groups in a normal year,
Singida Central Sandy Plains,Tanzania, 1999 (Save the Children/Ellen Mathys)
Figure 6 compares the income of two groups from different areas of Western
Sudan. By the standards of their own areas, both groups are “poor”, yet one has
an income almost three times larger than the other’s.
For the poorest households, acquiring food may account for as much
as 80 per cent of economic activity. Very poor households may have to make
trade-offs between obtaining sufficient food (or food of adequate quality) and
obtaining clothing and other essential non-food items, such as cloth and soap.
14
An introduction to household economy ●
Figure 6:
Relative income of the
same wealth group for
two different locations in
North Darfur, Sudan,
Sudanese pounds/year
1000
800
Livestock sales
600
Agricultural labour
400
Collection
Remittance
200
1999 (Save the Children/
Day labour
Trade
0
Jebel Si
Pippa Coutts,Youssif
Mellit/Sayah
Abaker)
At least seasonally, many poor households may get less food than they need to
maintain good health.
The relative importance of different sources of income also varies within an
area (Figure 7), according to the access households have to the most productive
assets. For example:
●
In an area where income is most easily generated from agriculture, some
people may have better access to land and hence a larger income from crops.
●
In a rangeland area, some people will have more livestock and therefore more
access to the animal products and cereals that can be obtained by trading
livestock.
●
Where the chief income-generating opportunity is employment, the
households with more people (especially more skilled people) available to
work are the more prosperous.
100
80
Figure 7:
Relative importance of
different sources of
Livestock
60
%
Agricultural labour
40
Purchase
income by wealth group,
Singida Central Sandy
Plains,Tanzania, 1999
(Save the Children/Ellen
20
0
Own crops
Gifts
Very Poor Lower Middle Rich
Poor
Middle
Mathys)
15
● The Household Economy Approach
In general, poorer households will, relative to the better-off:
●
demonstrate a greater diversity in their sources of income, as they can survive
only by exploiting a much wider range of opportunities, however unattractive
these may be (for example, migrant field labour)
●
be much more dependent on paid employment for income than the betteroff households
●
rely much more heavily on purchased food to meet their consumption needs
●
spend a much larger proportion of their income on food
●
have fewer savings and fewer assets than the better off.
3. Household management
Households face two main problems in managing their income. One is income
fluctuations that are broadly predictable, the other is income fluctuations that are
unpredictable.
Predictable income fluctuations
Within a given year, household income will not remain consistent, but will
fluctuate according to the seasonality of different income sources (see above).
The household must therefore manage its income to maintain a reasonably
constant level of food and non-food consumption and to provide for periods of
increased need: for example, the additional food required during spells of heavy
physical labour such as land preparation. For the poorest households there may
be a hungry season when household food needs cannot be met. Adults may
minimise their food intake and restrict their activities or give preference to
children. In some economies people can manage the flow of income to some
extent: for example, pastoral people may time livestock conceptions to maximise
the period during which animals are lactating and milk is available. In general,
however, a household’s ability to manage income will depend on its level of
savings and reserves, which can be used to bridge income gaps. Poorer
16
An introduction to household economy ●
households may also have to take loans for this purpose. These range from shortrun credit (where, for example, a richer person will provide seed and other inputs
to enable a poorer person to cultivate crops and will receive part of the crop in
return), through cash loans to be repaid at interest, to systems of hereditary debt,
where the recipient may be bound by a debt that realistically cannot ever be
repaid.
In some semi-arid areas, rainfall and crop production may fluctuate wildly
from year to year. A household can anticipate this and manage its income to
ensure a regular food supply. Crops are stored in years of surplus production to
cover years when the household is in deficit.
Unpredictable income fluctuations
Poor households live in an environment of uncertainty, where there are constant
threats to their income. These may come from within the household – for
example, the illness or death of a family member may result in loss of earnings
or the need to pay funeral costs – or from some change in the wider situation,
such as drought, war or price changes.
In most poor countries there is no government welfare provision – other than
famine relief – and income failure may lead to destitution. The security of a
household will therefore depend primarily upon its own management, and to a
variable extent on the economic support it can expect from other households.
Household risk management depends upon three principal strategies:
Minimising risk
Households tend to choose income options that minimise the risk to their
income and food supply, even if this sometimes means a lower return; in other
words, they are “risk averse”. For example, when offered the choice between paid
employment where the return can be reliably predicted and a cash crop where the
return is potentially greater but uncertain, a household may choose employment.
Many strategies are used to reduce the threat to agricultural production: in parts
of Rwanda, for example, several types of bean seed, each of which flourishes
under different rainfall conditions, are often planted together, thus increasing the
chance of a crop, though at a higher initial investment.
17
● The Household Economy Approach
Maximising savings and assets
Households will often save and invest as much of their surplus income as they
can, even to the extent of minimising consumption. Ideally, savings should be in
a form that is secure, holds its value over time and offers a return, either interest
or some other benefit. In practice, however, the opportunities for saving open to
a household are limited. Cash is seldom an attractive option, as banks are often
inaccessible, particularly to the poorest people, inflation is often high and there
is the constant danger of theft. Households typically diversify their savings, eg:
●
as gold (often jewellery)
●
as livestock (which gives a return in the form of milk and meat, and interest
in the form of offspring. Moreover, cattle, camels, sheep and goats are mobile
and can be moved in the event of drought)
●
as items such as corrugated roofing that are not only useful but also retain
their value and can, if necessary, be sold.
Cash savings may be kept in the form of hard currency remitted from other
countries.
Sharing risk
Within larger kin groups and between households, risk can be shared by the
transfer of goods between households in time of need. This strategy covers a vast
range of situations and methods of transfer, but there are three basic types:
●
gift, where food or some other item is transferred freely and without
obligation from one household to another
●
reciprocity, where the transfer imposes an obligation on the recipient to
return the goods or some other service at a later time
●
obligation, where the giver is obliged to relinquish some item under specified
circumstances, eg, the transfer of a cow in time of drought.
In fact, various intermediate forms are found: for example, where a gift is given
on the understanding that the recipient “pays” by providing some service such as
nominal agricultural work. Transfers may also occur through the transfer of
people between households, eg, by adoption or fostering. The important feature
of all three types of transfer is that they occur on non-market terms: all other
things being equal, the gift is independent of its market value.
18
An introduction to household economy ●
4. Markets and exchange
Household economy does
The purely local subsistence economy,
not exist in isolation but in
where a household produces most or all of
a market context. The
its food needs, has become a rarity. There
household economy can be
are now very few people, even in the
understood only if we
remoter rural areas of poorer countries,
understand this context.
who do not depend in some way on the
market to obtain food. To the extent that a
household relies upon exchanging crops,
livestock and labour for food and other goods, its income will depend on the
relative value of the items it buys and sells.
Trade confers major benefits: it gives access to manufactured goods that
improve the quality of life, and to income that can be used to obtain health care
and education. But involvement with the market is not without its risks,
particularly for poor families.
Definitions
Market is used here and in HEA to mean a place where exchange occurs. The
nature of markets varies widely: trade may occur through local arrangements
between surplus and deficit households, through itinerant traders, through
village shops or through weekly and permanent market centres. Labour markets
may sometimes operate throughout an area – for example, a commercial
agricultural district – rather than in a specific named place. A hierarchy of
markets is often found, with larger markets in district centres, perhaps offering a
wider range of commodities, and smaller markets at a more local level.
Value. In most locations, trade is now conducted through the medium of
cash (usually the national currency, although in border areas and in countries
where the national currency is unreliable, multiple currencies, including hard
currencies, may be used). Barter – the direct exchange of one commodity for
another – is also encountered. Less often, specific commodities (such as tobacco
in parts of South Sudan) is used as a currency, although the range of transactions
that can be carried out with such “cash” may be restricted. But whatever the
19
● The Household Economy Approach
medium of exchange, market exchange (including barter) is conducted on the
basis of value, which is liable to change under different conditions. This should
be distinguished from “non-market” exchange, where food, cash or assets are
transferred, for example by gift or on a reciprocal basis, without involving the
concept of market value.
Market use
Households resort to markets for a variety of purposes:
●
to use any income above their consumption requirements to obtain
manufactured goods (such as soap and cloth), education and health care
●
to increase household income through specialised production and “adding
value”. A household may choose to grow a crop that is most easily and reliably
produced in its area, and to exchange all or part of it for other crops produced
in other areas. These may be food crops – in central Afghanistan, for example,
high-quality potatoes are grown and exchanged for wheat grown in other
areas, with the producer making a profit (in terms of dietary energy) on the
exchange – or they may be cotton, coffee and a wide range of other cash
crops, or products such as handicrafts and firewood. Pastoral people often do
not have enough animals to enable them to subsist directly from animal
products. To get enough food, they depend on the exchange of livestock for
cereals
●
to obtain paid work on larger land holdings, cash-crop plantations, urban
areas or in other countries.
In practice, the way households use markets often depends upon the ease of
physical access. Distances to the markets may often be long and the travelling
time substantial. Some pastoral people may only be within reasonable striking
distance of a market at a few periods during the year. In some areas, access to
markets is insecure and there is a risk that goods will be stolen. It may also be
difficult for people to transport heavy or bulky goods over long distances.
20
An introduction to household economy ●
Price
Prices at any market will vary according to:
●
the extent of competition between traders. A free market in which several
traders compete for custom will tend to have lower selling prices and higher
buying prices than one dominated by a single trader or a cartel (a group of
traders working together), who can, within limits, set the price to their own
advantage. This latter situation is still encountered frequently in poorer
countries
●
the costs incurred by buyers or sellers. Many markets charge sellers for the use
of the market, and in some countries transactions of specific commodities,
such as livestock, may be taxed. There may also be transportation costs.
Governments may also intervene to stabilise and manage food prices
●
the balance between supply and demand at the market. A larger supply to a
market (see Figure 8) relative to demand will tend to produce lower selling
prices, and vice versa.
Price
Figure 8:
Relationship between
supply/demand and price and
Price below
which people
will not sell
Supply
Demand
floor price.
Supply and demand
Patterns of supply and demand vary. For example:
●
The supply of a commodity at a market may be entirely from local producers
or from a much wider geographical area, including international imports.
●
Demand may also be entirely local, or may be from traders who are supplying
a distant urban or export market.
In a given area, the normal supply of (for example) cereals to a market may be
from local producers and the demand from local people who wish to acquire
21
● The Household Economy Approach
grain. Under these conditions, any small
change in the supply to the market,
that supply and demand will
perhaps because of a poor local crop, can
be different for each type
lead to a sharp change in cereal prices. The
number of people wishing to buy cereals
of commodity. Different
may remain the same, or even increase, and
considerations may also apply
prices may rise to high levels. However, in
to different types of one
a market connected to larger cerealcommodity, eg, sorghum,
producing areas (for example, through a
barley, wheat. Some of these
government distribution system), a similar
may be produced in the area,
failure of local supply might have no effect
others traded from a
on prices, or at most cause only a small
distance.
increase, as any local shortfall would be
made up by supplies from other areas.
Levels of supply and demand will also vary with price: in general, a high price
for a commodity will tend to increase supply and reduce demand, and a low price
will have the reverse effect. Demand for a commodity may move between
markets: if prices are high at one market, people may be able to travel to take
advantage of lower prices elsewhere.
Market elasticity is a measure of how responsive one variable (such as price)
is to changes in another variable (such as demand): for example, in Figure 9 when
demand increases, prices tend to rise – the rate at which demand will rise is a
measure of elasticity.
It is important to remember
More elastic
Price
Figure 9:
Less elastic
Demand
Price changes with changes in
demand: two markets with
different elasticity
22
An introduction to household economy ●
Floor price. For most commodities there is a price below which people will
not sell. The rate of pay for day labour, for example, is often very low, as the
supply of labour may exceed the number of jobs available. But there will be a
limit to how far prices will fall as, below a certain level, the rate is too low to
allow subsistence and people will refuse to work. Livestock may be sold for very
low prices when people are desperate to sell: for example, when it is clear that
animals will perish because of a lack of grazing. Under such conditions the floor
price may be any sum that is offered – sometimes merely the value of the animal’s
skin.
Fluctuations in price
major food crops, Gidan
Woreda, North Wollo,
Ethiopia, 1999 (Save the
300
250
200
Wheat
150
Sorghum
Teff
100
Barley
50
Beans
Peas
0
Se
pt
O
c
No t
v
De
c
Jan
Fe
b
Ma
r
Ap
r
Ma
y
Jun
e
Jul
y
Au
Se g
pt
O
ct
Figure 10: Market prices of
Price (Birr) per Quintal (100kg)
A major problem for household management is that the price of many
commodities fluctuates between seasons and between years, in some market
systems in an unpredictable way.
The seasonal price fluctuations of locally produced commodities at least are
usually broadly predictable. For example, crop prices (Figure 10) in producer
areas are generally lowest immediately after the harvest. If farmers have to sell
produce immediately after a harvest in order to obtain cash to meet their
consumption needs, they will obtain a much lower return than if they sell at a
later period. Richer households with savings and reserves may be able to wait and
September 1998 – October 1999
Children/Ellen Mathys)
23
● The Household Economy Approach
sell their products at a later time to get a higher price. Poorer households, which
depend on exchange of their labour, livestock or cash crops for survival, may have
to sell even if prices are against them. Food and other goods produced outside
the area and imported for sale will tend to show the seasonal price pattern of the
producer area.
Larger and less predictable price fluctuations may follow surges in supply or
demand. In some cases these may drastically alter the value of the household’s
assets: for example, a household with a hundred goats each worth $10 may be
considered wealthy, but a sudden fall in livestock prices may reduce the
household to poverty (see Chapter 2).
The unpredictability of commodity prices is one reason why people may
choose to keep their savings in edible form, as food stocks and livestock, rather
than converting them into cash or other goods.
5. Household expenditure and consumption
needs
All households have needs over and above an adequate food supply. A reasonable
standard of living is difficult to define, but even the most basic level of existence
requires clothing, shelter, salt, fuel and soap. Many people would also include
kerosene or some other source of light; some would include alcohol and tobacco,
almost the only luxuries available to the rural poor. Also, expenditure on
weddings, funerals and other ceremonies are essential for a proper and dignified
social existence. Access to basic health care and education, in areas where they are
available at all, now usually involves some costs, in direct charges and incidental
costs such as books.
In practice, many households in the poorest areas have difficulty in meeting
even their basic food needs, and they often have to make trade-offs between
acquiring enough food, acquiring food of adequate quality and meeting nonfood costs.
24
An introduction to household economy ●
Household expenditure can be divided into two broad categories, although
the distinction is not always precise:
●
Obligatory expenditure. That which the household cannot avoid: for
example, food, clothing, salt, the minimum costs associated with births,
deaths and weddings, emergency medical costs, debt repayments, etc
●
Discretionary expenditure. That over which the household has greater or
lesser control: for example, school fees, most incidental medical expenses and
in some locations, meat, fats, oils and other more costly foodstuffs.
The division between the two will depend on the absolute level of household
income. Poorer households may have to make trade-offs between different costs:
for example between expenditure on food and the costs of sending a child to
school.
6. Individual and household food
requirements
HEA is chiefly concerned with the adequacy of household access to food energy.
Protein and other specific nutrients which are necessary for an adequate diet are
not discussed in this manual.
Human beings can consume a wide variety of diets and still meet their food
and nutrient requirements.
Food energy is usually measured in terms of kilocalories (kcal). A kcal is
defined as the quantity of heat required to raise the temperature of a litre of water
from 15°C to 16°C.
The energy requirement of individuals vary according to their age, sex and
level of physical activity and the environment to which they are exposed. In
general, growing children have a greater energy requirement relative to their
weight than adults. People require more food energy to maintain body weight in
cold climates or if they are doing hard physical work.
25
● The Household Economy Approach
Table 1 shows the energy requirement to maintain body weight for
individuals of different ages in a population with the demographic profile of a
developing country. The average requirement across the whole population is
2,070 kcal per person per day.
Table 1:
Energy requirements for a population with a developing country
demographic profile assuming a light activity level and an ambient temperature
of 20°C, a body weight for men of 60 kg and 1.55 BMR, and a body weight for
women of 52 kg and 1.56 BMR. From:World Food Programme/United Nations High
Commissioner for Refugees, Guidelines for estimating food and nutritional needs in
emergencies. December 1997
Age group (years)
% of
Male and female
total
Energy requirement
per person
combined
population
kcal per day
0
2.59
820
1
2.46
1,220
2
2.45
1,380
3
2.44
1,500
2.43
1,620
0-4
4
12.37
1,290
5-9
11.69
1,860
10-14
10.53
2,210
15-19
9.54
2,420
20-59
48.63
2,230
7.24
1,890
60+
Pregnant
2.4
+285 kcal/day
Lactating
2.6
+500 kcal/day
Whole population
2,070
In practice there is much variation in actual energy intake within and between
populations, and this may be less than or more than 2,070 kcal per day. Food
intake is very difficult to measure and exact measures of actual human intakes are
26
An introduction to household economy ●
rarely available. However, rough estimates of household consumption can be
obtained from household income and consumption data, and there is indirect
evidence from anthropometric surveys. Body weight varies widely between
populations and some populations will need more energy to meet environmental
costs: eg, at higher altitudes where ambient temperature is lower, or where people
cannot afford adequate shelter, fuel and clothing even when it is relatively warm.
Poorer groups often have access to less food energy than the better-off and, as
already noted, there may be seasonal swings in food availability and
consumption.
Household energy requirement will vary according to the composition of the
household. For example, using the data in Table 1, a household with six people,
two adults under 59 years of age, and four children with ages 1, 3, 7 and 10 years
would have an average requirement of 1,875 kcal per person per day.
A household with two adults under 59 years and four older children (5, 8, 10,
and 13 years) would have an average requirement of 2,100 kcal per person
per day.
In any case, it is only possible to estimate actual consumption and to make
rough estimates of actual variation in requirements. As a rule of thumb:
●
moderate levels of activity will add about 140 kcal per person, and heavy
work (eg, heavy land preparation) will add about 350 kcal per day when
averaged across a whole population
●
each 5°C below an average ambient temperature of 20°C will add 100 kcal to
average requirement.
27
● The Household Economy Approach
Summary
●
For HEA, food security is defined in terms of household access to food,
not in terms of the supply of food.
●
Poverty may be defined in absolute, relative and qualitative terms and in
terms of the achieved outcome.
●
Household income varies in size and quality within households and
between households in different locations and across seasons.
●
In general, households manage their affairs so as to minimise the risk of
income failure and to secure their future food supply and livelihood,
rather than to maximise income.
To the extent that a household depends upon market exchange, its
income will depend on the relative value of the commodities sold
(including labour) and the food and non-food items purchased.
28
2
An introduction to
household vulnerability
This chapter discusses how households are affected by, and respond to, economic
shocks, whether those shocks are more or less severe, on a larger or smaller
geographical scale, or the ways in which they occur in combination. For example,
a shock might take the form of a fall in one or more sources of household income
caused by the reduced production following drought, a change in the price of a
commodity, a decline in employment opportunities, or conflict which reduces
people’s access to markets, or a combination of two or more of these factors. Our
interest here is not only in the effect that shocks have on household food
supplies, but also on non-food expenditure and on savings and assets.
Definitions
Risk. The probability that an event will occur: for example, in a semi-arid area
drought sufficient to reduce crop production by 50 per cent may occur two years
in ten: that is, a risk of 0.2 or 20 per cent.
Vulnerability. If a specified event would be expected to affect a household in
a specified way, whatever the risk of that event occurring, the household is
vulnerable to that event. For example, a household might be regarded as
vulnerable to a specified level of crop failure if this would reduce the income of
a household by, say, 50 per cent, even if crop failure in that area was rare. There
are no generally agreed measures of vulnerability.
29
● The Household Economy Approach
1. The effect of shocks on households
It is logical to look at the effect of a shock on household economy in two stages:
●
What effect would we expect a particular shock, or combination of shocks, to
have on current income, given the options open to the household?
●
If we feel that as a result of the shock the household will suffer an income
deficit, what do we think it will do to overcome this? Households may have
a variety of ways in which they can compensate for a loss of income.
The effect of shocks on current income
% of annual household food needs
The direct effect of a shock will depend upon how a household normally obtains
its income: for example, a household that depends for a large part of its income
on crops would be affected by crop failure in a different way from a household
that depends chiefly on livestock – the first will be affected directly by a loss of
crop income, whereas the second might not be directly affected at all. Conversely,
an outbreak of rhinderpest would be expected to affect a pastoral household
more than an agricultural one. Similarly, a household which depended on the
purchase of food would be affected by a rise in the price of food.
Figure 11 shows the likely effect of a 50 per cent fall in crop production on
the income of households with varying levels of dependence on food crops. Note
that the total income of the households shown in the example might vary.
100
80
60
40
Deficit
20
Livestock
Food crops
0
l
ho
se
ou
H
30
d1
De
it
fic
Ho
l
ho
e
us
d2
De
it
fic
Figure 11:
Effect of a fall in crop
production on two
households
An introduction to household vulnerability ●
As described in Chapter 1, poorer households often have more diverse sources
of income than better-off ones. It follows that a household with a wider range of
income sources is likely to be less vulnerable to a shock to any one income
source: for example, a household that obtains only 20 per cent of its income from
crop production cannot directly lose more than 20 per cent of its current income
through crop failure.
Household income may also be indirectly affected by shocks: for example,
crop failure in an agricultural area where poorer people depend on better-off
people for agricultural work might lead to a situation where farmers with more
land require less agricultural labour – poor people would not only lose income
from their own crops, but also income from paid employment. If the price of
cereals increased because of a crop failure, this would indirectly affect people who
depend on cereal purchases for their food supply.
In some livestock areas that depend upon trading animals for cereals, a
drought may have direct effects on income as a result of the loss of grazing
(which will reduce milk yields and cause the condition of the animals, and hence
their value, to decline) and may have indirect effects as people attempt to sell
more animals, leading to a fall in animal prices and a loss of income from animal
sales. Conflict or unrest which restricts people’s freedom to travel to markets to
trade may also reduce household access to food.
How prices change chiefly depends upon the balance of supply and demand
for that commodity. For example, if a large number of people sell livestock and
the demand from traders for livestock does not increase significantly, the price
will fall, possibly very sharply. If the demand for food increases and the food
supply to the market supply does not keep step, the price will rise.
Figure 12 shows the numbers of livestock sold in 1982–83 before the
Ethiopian famine of 1984–85. Figure 13 shows the relative prices of a goat and
a sack of sorghum in a town in Western Sudan following a drought. In the latter
case, the value of a goat in terms of sorghum fell from approximately two sacks
per goat to three goats per sack. Both examples are from isolated areas where a
severe fall in local production and market supply was not replaced by cereal
imports from outside.
31
● The Household Economy Approach
3000
2500
Figure 12:
Birr
2000
Sales of livestock, Korem,
1500
Ethiopia, 1982–83
1000
(P Cutler, ‘Famine
No. of
livestock
sold
500
Fe
b
Ma
r
Ap
r
Au
g
Se
pt
O
ct
No
v
De
c
Jan
Ma
y
Jun
e
Jul
y
0
forecasting: prices and
peasant behavior
in Northern Ethiopia’,
Disasters 8 (1), 1984)
Sudanese pounds
1000
800
600
Figure 13:
Relative prices of goats
400
and sorghum, El Fasher,
200
0
Jan
Mar
May
July
Sept
Sorghum
Sudan, 1990
Goat
(Save the Children
Food Information
System, Darfur)
As might be expected, changes in the relative value of assets and food tend to be
greatest where:
●
the shock is severe and/or affects a large number of households, and many
people are attempting to sell assets at the same time
●
in locations where the demand for labour or livestock or other products such
as firewood, charcoal, and handicrafts tends to be relatively fixed
●
where the supply of food to the market is limited.
Increases in food prices may also be exaggerated by the tendency of households
to hoard food. Even households with a marketable surplus may choose to defer
32
An introduction to household vulnerability ●
the sale, either because they are uncertain about future supplies, or to speculate
on rising prices.
The effect of a shock on household income can therefore only be understood
if we know the exact nature of the shock and the specific economy of the affected
household.
2. Compensating for loss of income
A household may be able to compensate for an income deficit and maintain its
consumption level if it:
●
has a sufficient stock of food to cover the deficit
●
has cash savings, livestock and other assets and can exchange some or all of
them to buy food. The value of cash savings and assets in terms of food will,
of course, depend on their relative price: if the price of livestock or labour
falls and the price of food rises, the effect is to devalue the assets, possibly
severely, and to reduce the capacity of the household to compensate for
a deficit
●
can expand other sources of income or find new ones: for example, the
household may be able to increase its firewood production or seek
paid employment or collect enough wild foods to meet part or all of the
deficit
●
receives food, cash or productive resources (such as livestock) as a gift: for
example, as charity from better-off households, as state welfare or as gift.
A household may also be able to compensate (or at least survive) by reducing its
consumption of food and non-food goods. This may be by:
●
reducing their food intake
●
consuming cheaper, less desirable foods
●
reducing expenditure on non-food items such as soap, clothing, education
and healthcare
33
● The Household Economy Approach
●
reducing the number of dependent people in the household. This may
happen if people travel away to find work or it may be a deliberate action, as
when children are sent to live with better-off kin.
In practice, households may attempt to both expand income and reduce
consumption.
Household decisions about compensating for an
income deficit
A household seeking to compensate for a deficit may have a range of options
open to it, and will seek to use these options to best advantage. For instance, a
household in deficit might decide to seek additional paid employment and not
to sell livestock, or vice versa – or it might choose to do both at the same time.
The options chosen will depend upon a household’s judgement about its own
circumstances and the best strategy to pursue. Very similar households may
pursue very different strategies. The best strategy for a household with more
livestock may be to sell one or two animals and purchase food, whereas a
household with fewer livestock may pursue some other option.
In general, all we can say with confidence is that a household will obviously
try to avoid starvation. But it will go to any practical length short of starvation
to retain the land, tools, seed and livestock that will ensure its future livelihood.
The costs to a household in surviving a shock, short of outright destitution
and starvation, may be severe. The potential costs to a household include:
●
impoverishment, caused by the use of savings and the sale of assets
●
the reduction of expenditure on basic non-food items such as cloth, soap,
education and healthcare
●
the risks and social disruption involved in labour migration, which may often
lead to prostitution and begging
●
the hazards of consuming famine foods
●
hunger and under-nutrition
●
long term debt.
34
An introduction to household vulnerability ●
Coping strategies
The term “coping strategy” is sometimes used to describe people’s responses to
shocks. In fact, most poor households already exploit to some extent all the
economic opportunities open to them, and it is rare for completely new
opportunities to be available. As a rule of thumb, if people are attempting an
activity which they do not normally do (such as migration to another location in
the hope of finding work or relief ), this indicates that they are already in severe
economic difficulty.
In practice, the range of options open to the poorest households in most of
the poorer countries has become more and more restricted. The last few decades
have seen increasing constraints on the rural subsistence economy and hence on
the range of strategies available to poor families to protect their income. With a
few exceptions (such as parts of Sudan), population pressures, together with the
shrinkage of wild areas and hence of opportunities for hunting and gathering,
and the closure of borders to pastoral migration have made it much harder for
poorer rural households to subsist on their own production, much less to build
savings and reserves. Households have largely compensated for the loss of
domestic production by increasing their dependence on the exchange of
livestock, fuel and paid employment. In most parts of the world, there are now
few poorer households that could subsist without the exchange of household
produce and labour.
Also, in the last few decades, particularly in Africa, opportunities for
employment and trade have in general increased. Even in the poorest countries,
most villages are now within reach of motor roads, and opportunities for
employment have expanded to take in neighbouring countries and the richer
nations of Europe, the Americas and the Middle East. Products that would
formerly have been used for local consumption are now often traded to capital
cities and international markets.
The effects of these changes on household vulnerability are complex, but in
most situations they have probably reduced the risk of outright starvation. In
many areas, governmental and international intervention is now common
(although in Africa this still tends to happen only as the result of major shocks).
Each situation is different, but it can be argued that households less dependent
on production are now less affected by production shocks. In times of local crisis,
35
● The Household Economy Approach
people can migrate, sometimes over long distances, to find work (or survive on
remittances from migrants), and this has the effect of spreading the risk over a
wider geographical area. Livestock and other commodities may now be traded in
a wider and more stable market, and their prices may be less affected by local
fluctuations in supply and demand.
The paradox is that in many areas people have probably become poorer, in
terms of the size and diversity of their incomes and food supply, but at the same
time are less vulnerable to starvation.
Summary
The effect of a defined shock (or shocks) on a household in any location
will depend on the specific combination of:
●
the severity of the shock
●
the scale of the shock – drought and other shocks affecting larger areas
may lead to changes in the relative value of labour and other assets and
food
●
the type of shock
●
the ways in which the household normally gets its income
●
the opportunities open to the household to compensate for any loss of
income and the specific choices that the household makes
●
to the extent that the options open to the household involve market
transactions, the relative value of labour and other commodities and
food.
36
3
An overview of the household
economy approach
This chapter gives a broad overview of the household economy approach (HEA)
and the information needed to conduct an HEA analysis.
The aim of HEA is to understand how families are making ends meet under
both normal and abnormal conditions. There are five steps in conducting an
HEA analysis. They are:
1. Define the food economy or economies (the group or groups of households)
for which an analysis is required.
2. Define categories of household wealth within each food economy.
3. For each wealth category, collect information that will enable you to describe
how, in a “normal” or baseline year, a typical household obtains its income
and what the economic context is. The aim is to compile a description of the
economy that is “complete”: that is, containing all the information needed to
understand not only people’s current access to income and food but also the
potential for them to expand their income under different conditions.
4. Describe the economic context to
which the households relate.
5. Use this description as a baseline
The generic question behind all
from which to understand the likely
food economy analysis is: “how
effect of changes in the economic
does this food economy work?”
context on household income and
food supply.
37
● The Household Economy Approach
Define the food economy or economies (a group of households
of similar economy) for which an analysis is required
Define the household (a production/consumption unit)
Within each food economy define categories and proportions
of household wealth
Describe a typical household budget for a typical household
in a normal year for each wealth group (or at least three)
Use this description as a baseline from which to understand
the likely effect of changes in the economic context on household
income and food supply
Figure 14:
The five steps in
conducting an
HEA analysis
1. Defining food economy areas and
populations
Livelihoods and geography
A food economy is a specified group or population of households. It is defined
as all the households in a geographical area where most households obtain their
food and cash income by roughly the same combination of means. All the people
in a village, a refugee camp or some larger area may be involved. For a local study
there may only be a single food economy, but a large geographical area may
include households with several different types of economy.
A population of households is a defined group of households in a food
economy: for example, all the households in a defined area or a camp.
The first step is to divide the area you are studying into areas with similar
38
An overview of the household economy approach ●
Takeze N
Central Tigray
NW Lowlands
Tigray E
Highlands
Semien
Gonder/
Gojam
Lowland
Takeze
W Takeze
Tigray S Zone
WT Belg/Meher
Blue Nile
Curve
W Oromiya Lowland
Blue Nile
Wollo Shewa Highland
Tach/
Simad
Wollo
Non-Belg
N Shewa Rift
Wollo Shewa Belg
Figure 15: Food economy areas, north-eastern Ethiopia (Save the Children/
Tanya Boudreau)
economies: for example, into a population that depends primarily on agriculture
(even though they may also keep livestock and have other sources of income) and
a population that depends primarily on livestock. The number of areas or
populations and the fineness of the distinctions will depend upon what objective
you have in mind. Figure 15 shows the food economies defined for a study of
north-eastern Ethiopia.
The division need not necessarily
The finer the level of subdivision,
be by geographical area. There may
be two or more groups of people
the more the work which will be
in the same geographical area who
required to describe these – each
demonstrate sufficient differences in
area will require fieldwork to
their economy to make a distinction
describe the household economies.
necessary.
In two districts of the Fifth Region
39
● The Household Economy Approach
of Mali, for instance, the great majority of the population are cereal cultivators
with, depending on their wealth, a greater or lesser investment in livestock. But
a minority are migrant fisherfolk, whose main livelihood comes from following
and catching fish along the main Niger River and its inland delta, according to
the seasonal increase and flood of the river and the growth cycle of the fish. There
is nothing to stop us dealing with two or more food economies co-existing
within the same geographical area. We can attach our own labels for
convenience: “the agricultural food economy”; “the fishing food economy”, etc.
Whether or not a population should be subdivided, and if so, at what level,
depends upon the question being asked. All investigations must begin with a
clear objective and a clearly stated question: for example, “How much food
should be supplied to this refugee camp?” or “What effect would an observed rise
in food prices have on the households in this area?” The investigation may be
local, or it may relate to a larger area – such as an assessment of the threat of
starvation in a country or entire region affected by drought. Clearly, if we were
interested in the effect of a shock on a whole country, different populations will
be affected differently because:
●
the shock (such as drought or conflict) will vary in its effect across a large area
●
different economies will be affected differently by the same shock
●
the opportunities people have to respond to the shock will vary from place to
place.
In short, people in different parts of a country will normally face different
intensities of threat from external events, and will have a different capacity to
respond to these events: their vulnerability in times of crisis will vary. And just as
important, exactly what we need to know about this will depend upon the
question that has been asked and our capacity to respond. For example, a global
monitoring system would probably be interested in a broader (for example,
provincial) level of geographical disaggregation, with a view to initiating further
investigation. From a national perspective this would probably be useless – it
might, for example, already be clear that there is a potential problem in an area
– and a finer disaggregation (for example, to district level or below) might be
required. There is no point in collecting information for its own sake.
40
An overview of the household economy approach ●
Food economies and administrative divisions
Food economy areas are economic, not administrative, divisions. Because a food
economy area is defined in terms of the economy of households, it will not
necessarily correspond exactly, or even roughly, with administrative boundaries.
In some situations this may lead to difficulties:
●
when establishing the population of the food economy area. Population
statistics are normally available by administrative area: for example, districts.
If a food economy area cuts across administrative boundaries, it may only be
possible to estimate its population
●
when reporting results. The findings of an investigation may require action
by government. This action is often taken with reference to administrative
areas.
2. Preparing a wealth breakdown
Within each food economy area we need to identify groups of people in different
wealth categories. These categories are defined with reference to that particular
area: it is relative wealth we are interested in.
The wealth breakdown employs the indicators of wealth used by the people
themselves. In some situations, the “wealthy” are those with more land; in others,
they are people with more livestock or labour; and sometimes they exhibit a
combination of wealth criteria. There is often a precise local terminology for
wealth categories; it seems to be a universal human instinct to categorise the
members of one’s community by their wealth.
The absolute differences in wealth within particular populations may be large
or small.
The distribution of wealth within populations is often uneven: there are
usually more households at the poorer end of the scale than at the richer end
(Figure 16). If HEA is to be used to estimate the number of households in need,
we must be able to describe the shape of the wealth distribution.
41
● The Household Economy Approach
Percentage of population
50
40
30
20
10
0
Destitute Very
poor
Destitute
Economically
inactive. High
dependence
on others. Often
old/sick. Weak
social linkages.
Poor
Middle
Rich
Poor
Own: 1 – 5 acres,
0 – 2 cattle,
0 – 2 oxen, 1 – 5 goats,
a donkey
and 5 – 10 chickens
Middle
Very Poor
Own 1/4 – 2 acres Own: 5 – 15 acres,
2 – 8 cattle,
no livestock
0 – 2 oxen, 6 – 15 goats,
1 – 2 donkeys
8 – 30 chickens
Very
rich
Rich
Own: 10 – 50 acres,
10 – 30 cattle,
3 oxen, 10 – 30 goats,
2 – 3 donkeys
15+ chickens
Very Rich
Often have a
business base.
Diverse income
sources including
trade, transport.
Wider access to
resources
Figure 16:
Example of wealth group
breakdown. Kitui Lowlands,
Kenya, 2000 (Save the
Children)
The wealth category which contains the largest number of households is
described as the “modal group” (Figure 16).
3. Describing households within wealth
categories
For each food economy, we need to describe the household economy in a
“normal” or baseline year and how this differs between poorer and richer
households.
42
An overview of the household economy approach ●
Defining the baseline year
The baseline year is one that reflects the usual conditions, the way people
normally live. The reason for defining such a year is to enable comparisons to be
made when conditions depart significantly from normal. Thus, when describing
how people live, we must be sure to get information for a “normal” year. To do
this, we need to be clear about what we mean by “normal”.
In practice, of course, there never is an absolutely normal year: some aspects
of all household economies will vary from year to year. What we are talking
about is a year when crop production, livestock mortality, food, crop and
livestock prices and employment availability – in fact, everything that the
population relies upon – are more or less usual for the area. It may even be
possible to identify a specific year during which this was the case.
In some areas, household income may fluctuate wildly from year to year: for
example, in semi-arid zones crop production may be in deficit for four years in
ten, meet consumption requirement in three and be in great surplus in three. In
these areas it is usual for people to store the surplus and use it to make up the
deficit in years of low production. For these zones it may be necessary to take a
range of typical years (deficit and surplus) and average them into a “normal” year.
Describing households
By discovering how typical households in each food economy area normally
obtain their food and other income, we can build up a picture of the food
economy of a given area as an aggregate of its households. In most situations,
there are too many households for every one of them to be investigated, so we
have to try to sample the different types of household. This is done with
reference to at least three wealth groups: poor people, rich people and a middle
(often the modal) group – even if we have identified more wealth groups than
this in the population.
We can do this because the income options open to people in a given place
are usually quite limited: most households of similar standing (for example,
poorer households) within a food economy will be doing roughly the same things
to survive. There will, of course, be variation within each category of household.
43
● The Household Economy Approach
●
●
A “household” is defined as a group of people who:
contribute to a common economy
rely on the income from that economy for at least the greater part of their
food.
There is, however, a potential problem in defining “typical” households. In many
areas the model of the nuclear household (one male and one female parent and
their children) is not applicable. The family pattern may be one male and one or
more wives and their children, living in more than one dwelling. Other relatives
(and sometimes non-relatives, such as guest workers who live as family members)
may also be involved.
Household members do not all have to be resident in the same place. In some
cases the married women may live separately, and it is common to find that some
household members, such as migrant workers, spend long periods in other
places.
An example... you may be told that households normally have about four or
five milking cattle, and you may think that the typical household is composed
of a man, his several wives and their children. But if your informant is talking
about a “household” of a woman and her children only, you could greatly
underestimate the importance of milk in the household’s economy: the milk
will go much further when shared between six people than between twenty.
●
●
●
●
●
44
The household economy is described in terms of:
household food production
household cash income from the sale of household production and
employment
the seasonal pattern of income
household expenditure on food and non-food items
household assets, in terms of food stocks, livestock and cash
An overview of the household economy approach ●
●
●
the markets used to exchange different commodities
the price of all the items exchanged by the household.
Information is usually also obtained on the longer-term trends in production and
cash income.
The aim is to develop a picture of a typical household budget and assets for
each wealth category.
Information about the household is collected systematically by:
●
identifying in general the types of production relevant to the population (for
example, sorghum, maize, sheep, migrant labour)
●
establishing the sources of income (food and livestock production, labour
exchange, other production, eg, wild foods, handicrafts) for a typical wealth
group for that population
●
discovering how the household uses this income, item by item: how much is
consumed and how much sold? The amount sold, minus costs, gives an
estimate of cash income (or return from barter)
●
finding out the proportion of cash income spent on food items and non-food
items (taxes, fuel, loan repayments, education, healthcare, etc).
Collecting information in this way enables it to be thoroughly cross-checked: the
amount of food the household obtains should be equal to or greater than the
amount of food it biologically needs; the income from production and labour
sales should be equal to or greater than the amount spent; and the use of labour
should correspond with the income from labour. Talking about a typical
household with groups of people representing that wealth category allows for a
more open discussion than focusing on one particular person’s affairs.
An example of a form used for data collection is shown in Annexe 3.
45
● The Household Economy Approach
4.The economic context
This includes the markets used by households to exchange different commodities
and labour; non-market transfers of goods between households and the way in
which this varies under different conditions; and the availability of wild foods.
5. Analysis
The baseline economic description can be used to understand how a food
economy is likely to be affected by a change in the economic context. For
example, a fall in food production will lead to a fall in household income from
food crops; it may lead to a rise in crop prices, making it more expensive for
households to purchase food, and so on.
46
An overview of the household economy approach ●
Summary
The aim of HEA is to understand how families are making ends meet under
both normal and abnormal conditions. HEA involves five steps:
1. Define the food economy or economies (the group or groups of
households) for which an analysis is required.
2. Define the distribution of household wealth within each food economy.
3. For each wealth group, describe how a typical household obtains its food
and non-food income and its pattern of expenditure in a normal, baseline
year.
4. Describe the economic context – markets, non-market exchange and
access to wild foods.
5. Use this description as a baseline from which to understand the likely
effect of changes in the economic context on household income and food
supply.
The aim is to compile a description of the economy that includes all the
information needed to understand people’s current access to income and
food and the potential for them to expand their income under different
conditions.
47
● The Household Economy Approach
4
Framework and definitions
The household economy approach (HEA) is intended to provide a way of
describing and comparing a wide range of different types of economy. In order
to do this it is necessary to ensure that all the necessary information is collected,
and that this information is classified in a systematic way. In practice, sources of
household income do not always fall neatly into clear categories. For example,
cereals produced by the household may be consumed (a food crop) or sold (a
cash crop). Not all foods classify easily as cultivated or wild – there are also wild
crops which are semi-cultivated, for example by deliberately spreading seed. This
chapter describes the framework used for information collection and defines the
terms used.
1. Classifying household income
Household food production. The headings used are:
●
food crop production
●
animal products and livestock
●
hunting
●
fishing
●
collecting wild foods.
Gifts. The heading is:
●
gifts and relief food.
48
Framework and definitions ●
Income from the exchange of household production and labour. The
headings are:
●
the exchange of labour through self-employment, eg, petty trade, small
transport services, and paid employment
●
the exchange of non-food household production (for example, firewood,
pots, wild foods, livestock products)
●
the exchange of livestock
●
the exchange of crops.
This list offers a sufficient number of headings to ensure that all sources of
income are included and for useful distinctions to be made between different
aspects of household economy. For example, livestock sales and crops are both
household agricultural products that are sold – yet livestock is such an important
source of income for many households, and its management and marketing so
specialised, that it has been included as a separate heading. Similarly, fishing is a
specialised form of hunting, but in some areas it is so important that it has been
included as a separate activity.
This classification is not, however, absolutely rigid. When an HEA is used in
an area where it is known that a heading is unimportant (for example, there is no
fishing, or no market exchange), that heading can be dropped. Similarly headings
can be subdivided if a particular local activity is sufficiently important or a finer
discrimination is required in analysis. For example, paid employment may be
divided into subcategories: local seasonal employment, employment at a longer
distance within the country, and employment in a third country, which may lead
to substantial remittances. The heading “food crops” may be subdivided into
wheat, sorghum, pulses, etc.
The important thing is to ensure that no heading is excluded unless you are
absolutely certain that it is completely unimportant in the case.
Note that in HEA we are interested in household income which can be used
for consumption, net of costs. For instance, some income may be invested in
food production (for example, buying seed, inoculating livestock). These costs
are deducted before defining income. Not all food produced by a household is
necessarily consumed within that household. Some may be sold, stored or given
away.
49
● The Household Economy Approach
In the field, you will encounter a wide variety of economic activities
that create income. It is usually straightforward enough to categorise them
under the headings given above. However, some items of income may cause
problems, eg:
●
items the household produces that are potentially edible but are in fact sold.
For example, a household may produce enough wheat to meet half its
consumption needs, but may routinely sell 20 per cent of this to obtain cash.
In this case, the 80 per cent of the wheat consumed would be recorded under
“own crops” and the remaining 20 per cent under “cash crops”
●
relief food is mostly considered as a “gift” (as the household does not have to
work to obtain it). But if the relief is given as food-for-work, it would be
classified under “paid employment”.
In cases of doubt, the best way to decide where to put an activity is to think
about what has to be done to obtain the income. Moreover, misclassification will
often make no difference to the outcome of the analysis. For example, swarming
termites are often gathered for food, and would therefore be classified as “wild
food”. In some places, however, women from specific households have rights to
specific termite hills, which technically speaking makes the termites a livestock
product (meat). However, it would make no difference to the outcome of the
analysis whichever classification was used.
2. Household food production
Food crop production
In most investigations the concern is to ensure that the household meets its basic
calorific needs in the short term, rather than that it enjoys the best diet in the
longer term. Unless you are specifically interested in the nutrients in the
household diet, the description can be confined to the energy-giving foods in
the diet.
50
Framework and definitions ●
So, for example, we need to know about sorghum and milk, but not about
chillies or salt, or even vegetables such as tomatoes or spinach – which are very
low in energy, however important they may be for vitamins and minerals.
Food crop production is defined in HEA as all activities where main food
crops are produced (in a planned manner) under the direction of household
members, for consumption within the household.
Livestock production
This means animal products produced by the household for consumption. It
includes: milk and milk products including ghee, blood, meat and eggs, from
cattle, sheep, goats, camels, pigs, domestic fowl, etc; fish where these have been
farmed; and cultivated honey.
Fishing
This means all access to fish and other aquatic organisms for consumption
within the household, where the fish have not been farmed. It includes activities
such as netting, spear fishing, setting fish traps, river poisoning, line fishing,
fishing with explosives and collecting molluscs.
Hunting
This describes all access to meat and other animal products consumed within the
household, where the animals have not been managed. This includes activities
such as hunting (with net, spear, gun, poison, etc), trapping and bird-liming. It
may include collecting invertebrates such as termites and locusts, gathering wild
honey and hunting reptiles and other small animals, but, depending on local
conditions, you may prefer to put these activities under “collection of wild
foodstuffs” (see below). Hunting may also include the stealing or rustling of
domestic animals, where these are stolen for consumption.
51
● The Household Economy Approach
Collection of wild foodstuffs
This includes all collection and consumption of non-farmed vegetable foods:
roots, tubers, fruits, leaves, grains and other seeds, nuts, bark, taking grain stores
from termite mounds, etc. At your discretion, this category may also cover
invertebrates and other animal foods that are not really hunted. In some cases,
wild foods are semi-cultivated in that people may encourage production by
deliberately spreading seed (for example, the water-lily Nymphea lotus in South
Sudan): these can be included as wild foods.
A distinction should be drawn between wild foods, and “famine foods”. Wild
foods are those which are collected as part of a normal diet, and/or may be
collected in greater quantity in time of need. Some wild foods (and some
cultivated foods such as cassava) are potentially toxic and require extensive
preparation (peeling, grinding, soaking) to make them fit for consumption.
Famine foods are those eaten only in desperation often despite their toxic
qualities.
3. Gifts and relief
As a source of food, gifts and relief differ from sources listed so far in that the
household does not acquire it through its own labour. In its pure sense, a gift of
food is non-reciprocal: food is given to the household by other households or
individuals within or beyond the food economy, and the recipient household
gives nothing in return. In practice, this seldom happens. The household almost
always gives something in return for the food it receives. This may have no
“value”, in the sense that it cannot be converted into food: for example, the
household may agree to come under the jurisdiction of the household that gave
them food if there are any legal disputes. Often, however, the recipient
household gives something of value, such as labour or a share of the harvest, to
the donor household. Alternatively, they may give back the same commodity:
they may receive a milk camel and later give a milk camel in return. Informants
may well describe all these exchange transactions as “gifts”.
52
Framework and definitions ●
Hence, gifts may be difficult to categorise. If food is described as a gift, try to
discover if anything is given in exchange (not necessarily at the time of receipt,
as “payment” may be delayed). If the exchange is equal (as with the camels
mentioned above), then the “gift” is cancelled out and can be disregarded. This
will also be true in most areas for dowries or bride price, as over time a household
will probably pay out as much as it gains. If the exchange is unequal, then ask
yourself whether what is being “paid” for the food is the normal market price or
a token payment. Ask also what would happen in a bad year.
Often gifts are repaid in labour. However, the amount of labour “paid” is
much less than that which would be required to buy the amount of donated food
in the market (thus making it a non-market, or token, payment of labour); and
in a bad year, when the labour is not required by the giver, the gift of food may
well continue. In this case, although the gift bears certain similarities to
“exchange of labour” as a source of non-food income (see below), it is not
categorised as such because the exchange works by different rules.
The category of “gift” as a source of food includes gifts of uncooked food,
gifts or loans of lactating animals, gifts of cooked food (sharing meals) and
temporary fostering, where a household member is sent to live with, and be fed
by, another household.
Relief food
Relief food is food that a household acquires from sources outside the food
economy (normally government, the UN, a non-governmental organisation
(NGO) or a religious organisation). Relief food is generally obtained without any
type of payment. This category includes emergency relief food, school feeding
and feeding at clinics.
Food-for-work
Food-for-work is classified as paid employment. Note that this category only
covers relief food that is a regular source of income year on year and which the
household has come to expect.
53
● The Household Economy Approach
4. Income from the exchange of household
production and labour
Income from the exchange of household production and labour includes goods
that have value, can be exchanged and are obtained by the household through
economic or social activities. This income is mostly in the form of money,
although in some areas payments for labour are made in food.
Because income from exchange comes in a variety of forms, it is referred to in
this section as “value”, underlining its importance as income that can be
exchanged for food or other items and to distinguish it from “gifts”.
Exchange of crops
This is where crops are exchanged for “value”: ie, cash or other exchangeable
items. It may involve food crops, non-food cash crops (such as tea, coffee, rubber,
cocoa, tobacco, cotton, flowers and sisal) and other non-processed (or minimally
processed) vegetable products such as timber, copra, palm leaves, toddy and
grass. The sale of farmed fish is included in this category. Do not include
products that have been extensively processed, as these come under “exchange of
other production” (see page 53): for example, timber from woodlots is a crop,
but carved wooden tables are “other products”. You will have to use some
discretion here. You may find it useful to separate sale of food crops from all
other crop sales.
Rental income (which is sometimes in kind) from land can also be classified
in this category.
Exchange of livestock or livestock products
This covers the exchange of livestock (or their products) owned by the household
for cash or other value. Livestock are also categorised under “sources of food
income”. Livestock products include milk, meat, curd, ghee, eggs, horn, hide,
wool and honey. This category also includes manure, which can be “sold” to
farmers: the manure often comes in return for free grazing, but sometimes the
54
Framework and definitions ●
farmer will in addition pay the livestock owner to graze animals on his fields. In
this latter case, the payment should be included under “sale of livestock
products”. This category also includes renting out livestock as draught animals.
Exchange of labour
This includes all exchanges of labour for value. Labour can occur on a
permanent, seasonal or daily basis, in rural or urban settings, and the methods of
payment may differ. With agricultural labour in particular, there can be
difficulties of classification: however, you can usually be confident that the sale
of labour is occurring when family members work on land to which they would
not otherwise have access and receive cash, food or other goods in return for that
work, the quantity depending on the amount of work they have done (note,
however, the comments under “gifts” and “relief ” on pages 52–53). When labour
is paid in food, you can still include it with the creation of non-food income,
because the mechanism by which the food is acquired makes it very close to (and
subject to the same vulnerability as) the sale of labour for cash.
Exchange of other products manufactured in the household
This covers all exchange for value of products manufactured in the household, or
by household labour, where the household controls all the factors necessary to
produce the goods (that is, they are not working for somebody else). There are
many items exchanged in this way, but mats, charcoal, beer, baskets, jewellery
and pottery are amongst the most common. As with other categories, if there are
large overheads only profit should be recorded.
Exchange of fish
This covers the exchange of (non-farmed) fish for value.
Exchange of wild foods
This covers the exchange of wild foodstuffs and other non-managed, seasonal,
naturally-occurring products for value. This category includes the sale of aloe and
55
● The Household Economy Approach
medicinal herbs, for example, but not the sale of firewood or gemstones. These
latter would be included under “exchange of other products manufactured in the
household”.
Trade
Trade means the exchange of goods that have previously been brought into the
household (that is, have not been produced in or by the household) and then
re-exchanged. These goods can be crops, livestock or consumer products. The
important point is that they are bought and then resold to create income – for
example, buying a sack of sugar and selling it on in smaller parcels. Again, take
care to record profit only.
5. Savings, assets and reserves
HEA normally uses four categories of savings, assets and reserves:
●
Food stocks. The food in stock in the household over and above that
required for consumption in the current year or until the next crop. In other
words, food that could, if necessary, be used to cover a deficit in income. This
also includes the storage of wild foods and animal products.
●
Cash savings. The cash available to the household over and above that
required for current consumption. Again, this could, if necessary, be used to
cover an income deficit.
●
Livestock holdings. Those animals that are under the control of the
household and are a potential food source, or which could be sold and the
proceeds used to purchase food. Livestock include cattle, camels, sheep and
goats (often recorded as “shoats”, although a distinction may be made if there
is a large difference in the value of sheep and goats), poultry, rabbits and
guinea pigs.
●
Other potentially tradable assets. This might include corrugated roofing
sheets, radio sets and other household items.
56
Framework and definitions ●
It is important to keep in mind that some potentially tradable items including
ploughs, draught oxen and land are necessary for production. These items may
be included but they should be clearly distinguished from food and other goods
which more clearly form reserves of wealth which might be used to cover a period
of reduced income. This distinction may be necessary when findings are
analysed.
6. Distinguishing between wealth categories
There are two basic points to make about distinguishing wealth categories:
●
You, the person making the enquiry, do not define poverty or wealth. Your
informants do that. In each food economy, you should allow informants to
explain what would mark a household as poor, what would mark it as typical
and what would mark it as rich. You will then proceed with these local
definitions of the categories of wealth.
●
The basis of wealth will vary according to the major sources of income.
Among agricultural populations, a widely used indicator of wealth is the
amount of land cultivated; not necessarily the amount of land “owned”, as the
former term combines information about access to land, labour and other
inputs in one measurement. Pastoral populations (as well as some agricultural
people) may measure wealth by the number of livestock held. In some places
where paid work is the basis of the economy, the wealth indicator may be the
number of economically active members of the household.
It has been the experience of Save the Children teams that key informants in an
area will normally agree on the most important criteria.
Having the criteria defined by the informants does not prevent you from
comparing one food economy with another. The “poor” households in two
different areas may have been locally defined, but you will still know the
numbers of livestock, the levels of agricultural production, etc, of the two groups
and so be able to compare them. The aim is simply to find an agreed way of
57
● The Household Economy Approach
differentiating between poor and rich that can be used as a basis for subsequent
discussion.
When you are dealing with the poor and rich groups, you should aim to
ignore the extremities of the scale. There will almost always be a small number
of extremely rich people in a community, perhaps landlords or money-lenders;
and similarly there may be small numbers of destitute households, which are
economically inactive or are living on charity. You need to find the “typical” poor
and rich. In practice, this means identifying not how the poorest 1 per cent of
the population live, but how the poorest and richest 10 per cent live:
●
The poor. These are the poorest households that are economically active,
making up at least 10 per cent of the population. This means, in practice, that
a “model” poor household will not be in the very poorest (near-destitute)
5 per cent of the population, but will fall somewhere within the bottom
10 per cent. In some areas, this model household may in fact represent a large
proportion of the population, with most of the bottom 20 per cent or even
25 per cent of households in this group.
●
The rich. A “model” rich household will fall into the top 10 per cent of the
population, but will not be in the top 1 per cent (who tend to be very much
richer and therefore unrepresentative, even of the rich group).
To get an idea of the differences within any society, you need at least three levels
of wealth. In some places, however, you may find that people consistently and
convincingly identify more than three levels, and you should then do likewise.
58
Framework and definitions ●
7. The wider economic context of households
Households exist in, and are often dependent upon, a wider context such as:
●
markets, in which they may need to exchange products for food and other
items. We cannot say that we understand household economy unless we
understand these links: not only because disaster may affect them, but also
because they are central to the ability of many households to expand their
income and realise the value of assets when other sources of income fail
●
the local environment, which governs whether food and non-food
production – hunting, fishing and the gathering of wild foods and non-food
products that can be consumed or exchanged for food – can be expanded
when other sources of income fail
●
non-market transfers. The ability of a household to survive may depend on
food and other goods obtained from kin, other members of the community,
government and other agencies.
Markets
We need to know how the household normally uses exchange to acquire income.
For the purposes of HEA, exchange forms the link between the household food
economy we are looking at and the wider economy. So if people sell or buy
labour, livestock or mats at a place outside their own area, the information
required is a description of who sells – or buys – what and where they do it.
Information may also be required about the market itself and the way in which
this works.
In general we are interested in six categories of market exchange:
●
livestock
●
crops
●
non-food production (for example, firewood)
●
food
●
paid employment
●
self-employment.
59
● The Household Economy Approach
In HEA a market is defined as a place where exchange occurs. This place may be
a specific named market in a town or a quite large area – for example, a district
where people find agricultural work .
People often use more than one market for the same commodity. It is helpful
to know the relative importance of each market to the household.
People often travel long distances to find work, and the markets may be in
other countries, including the richer countries of the Middle East, Europe and
North America. It is important not to miss these, as the value of remittances can
be considerable.
Non-market transfers between households and the
availability of wild foods
In many places people are able to survive periods of reduced income by obtaining
additional food, livestock or cash on non-market terms from kin, or as charity,
or by collecting additional wild foods. Information on the extent to which nonmarket redistribution between households may potentially meet household
needs under conditions of reduced income, and on the potential availability of
wild foods, is also usually collected.
The seasonal picture
A household’s sources of income will often be seasonal. For example, a household
that obtains part of its income from crops harvested in December may eat
nothing but home-produced crops between January and June and nothing but
purchased cereals between June and January.
Food economy assessments should include a detailed seasonal calendar
showing when different methods of food and income are obtained. This allows
an element of seasonal interpretation.
For example, consider two areas in which the population obtains 50 per cent
of its food needs from its own crop production. In one of the areas, all this
production comes at one time: there is only one harvest because there is only one
rainy season. In the other area, the production comes from two harvests and
relies on two rainy seasons. The effect of the rains failing will be very different in
60
Framework and definitions ●
the two areas. In the first, the entire domestically cultivated component of food
income – 50 per cent of food needs – will be lost. In the second, a proportion of
the domestically cultivated component will be lost, but provided the second rains
are successful, the second harvest will make up for at least some of the 50 per
cent.
The amount of detail required in the seasonal picture will depend on the use
you will be making of the information. In many cases, it is useful to know not
only if a household will be short of food or experience some other economic
difficulty, but also when this is likely to occur.
8. Units
When collecting information, it will be necessary to use the units employed by
the people themselves. These will often be sacks, cups, tins of various sizes and
beer bottles rather than standard measures of weight or volume. A list of
conversion factors is needed to convert these local units into standard units
(kilos, litres, etc).
HEA converts all the food produced by the household (and in some
applications the cash savings, livestock and assets) into food energy units, or
kilocalories (kcal). This is so that items such as milk or meat can be compared
with cash, grain or wild foods, thus enabling comparisons to be made between
households.
Food energy units (rather than, for example, cash) are used because:
●
it is not always possible to put a cash value on some items of income, as these
are not bought and sold. Milk and meat, for example, are often produced by
a household and are consumed or given away, but not sold
●
as access to food energy is often the primary concern of HEA, it is more
convenient to think consistently in such terms. It also makes calculation
easier.
61
● The Household Economy Approach
5
Finding the information
This chapter offers guidelines on sources of information. It covers:
●
locating and using documentary (secondary) sources
●
locating key informants – individuals with a good knowledge of an area or
some aspect of an area
●
selecting and approaching communities
●
observation.
The ways in which the quality of information is verified is discussed in Chapter
7. Methods for getting information, including the use of rapid rural appraisal
(RRA) and participatory rural appraisal (PRA) tools and interview techniques are
discussed in Chapter 6.
A typical HEA enquiry will begin at the centre (the capital or regional town)
and move outwards to the field. This approach has the advantage of:
●
running parallel to the linear or hierarchical structure of most governmental
and non-governmental organisations. This can work to your advantage, as
you progress from central office (in the country capital) to regional office to
district office to subdistrict office. This observes protocol, as each individual
you meet knows that you have already met and been “cleared” by their
superiors
●
allowing you to build up your knowledge from general outlines to specific
details as you get closer to the field.
62
Finding the information ●
Remember that:
●
It is good manners to introduce yourself to the local authorities before
conducting any enquiry, whether in a capital city, district centre or village.
In some places it is a requirement (you may be ejected if you do not) and
sometimes official written permission is required.
●
People are often busy. Try to time your discussions to minimise the
inconvenience to your respondents. This may mean working in the evening
or at weekends.
●
In interviews you should always clearly explain why you are conducting the
interview, what you intend to do with the information, and what result (eg,
the supply of food relief ) the people being interviewed might expect.
Particularly if time is short it may feel like a waste of time to work with
institutional structures. It is worth doing because:
●
you might learn more from people at a distance from the field than you think
you will
●
informants in the field can be warned in advance about the timing and nature
of the discussion, and this is likely to improve the quality of the interview
●
the process of working out from a centre follows the process by which you
collect secondary information, so little is lost if you combine literature
searches with interviewing
●
working outwards from a centre tends to also mean that you move from
gathering general to gathering specific information as you go along. This
allows for you to fine-tune your understanding of an area, and so your
questioning, over time.
A general rule of interviewing is that the more you know before the interview
starts, the more productive the interview will be.
63
● The Household Economy Approach
1. Documentary sources
You are extremely unlikely to find a written source, or combination of written
sources, that contains all the information you need to construct an accurate
baseline description of the food economy. Fieldwork is always necessary. But
written sources can be extremely helpful in giving sufficient background
information to suggest the type of questions you should be asking about a food
economy, in supplying information with which to cross-check what you are told
by key informants, and in identifying key informants: the authors of interesting
reports may possess much relevant information which they did not record in the
report. Well before you go into the field, you may be able to identify some
possible food economies.
Written sources may use a number of terms to describe areas having similar
production systems: for example, farming system zones, livelihood areas, agroeconomic zones and agro-ecological zones. These sources can tell you the weather
patterns that can be expected in an area (and the likelihood of the rains failing),
what crops (if any) can be grown in the
area and the likely yields, the types of
livestock kept, etc, which may indicate
Note: Written sources are not
geographical areas offering the same
infallible. One reason is because
range of possibilities for production.
they often only deal with one
By putting this information together
aspect of an issue, and do not
and using it to draw lines on a map
have to relate this to other
you have taken the first step in
economic activities. Look out for
defining food economy areas.
the income survey which shows
Secondary sources of this type will
80 per cent of the population
often be found at national capital,
earning less than they need to
provincial and district town levels.
However, as already pointed out,
survive, and the crop yield survey
food economies seldom correspond
which is based as much on
exactly to administrative boundaries
optimistic enthusiasm as on real
at provincial, regional or even
crop production.
sub-regional level. But since most
information
is
collected
by
64
Finding the information ●
administrative unit, you will have to calculate how the administrative and
ecological areas fit together.
Below is a list, by no means exhaustive, of secondary sources, with comments
on the possible uses and drawbacks of each. Note that in some countries it is
necessary to obtain government permission before conducting research in
university or government archives. You should always check whether this is the
case before starting the literature search.
Whenever consulting a documentary source always look for the “methods”
section. How did the author get the information and how accurate is this
information likely to be?
Anthropological or ethnographical studies
These may supply some information on sources of income and the wealth and
poverty issues that are the basis of the wealth classification used in HEA, and may
be a useful source of information on social organisation. But for many areas,
particularly those that have recently experienced conflict, drought or other
traumatic events, most anthropological work will be out of date and may be
misleading.
Agro-ecological maps or surveys
These are of particular value for your initial consideration of how to divide a
country or large region into food economy areas. Usual sources are the ministry
of agriculture, the national university’s geography and agriculture departments
and agencies with a specialist interest, including UN Food and Agricultural
Organization (FAO), UN Environment Programme (UNEP) and UN
Development Programme (UNDP).
Surveys of farming systems or livestock management systems
This material can be obtained from the same sources (ie, FAO, UNEP, UNDP),
and may be helpful for investigating sources of income, as it often describes the
types and quantities of crops grown and livestock held. Farming systems surveys
can provide information by wealth category although the information is often
aggregated: it tells you about the “average” farm household, which tends to
obscure the differences between rich and poor. Always try to find out how the
sample was selected; it may not have been truly representative.
65
● The Household Economy Approach
Surveys of household budget or income
These can give general information about sources of income, particularly nonfood income and that element of food income which is gained through purchase.
But the results of such surveys are usually presented in aggregated form, covering
large administrative divisions – so much so that one often wonders why the
information was collected and what possible practical use it is ever put to.
If using household budget surveys, be sure to check the definition of income
as used in the survey: does it, for example, include on-farm production and
consumption?
Agricultural surveys
Information about yield, area under crops and production is useful for work on
sources of income, especially food crops and sale of crops. However, yield
information is often based on trial plots cultivated under optimum conditions,
and as production information may be an extrapolation of these results, you may
find wildly optimistic figures.
Livestock surveys
These are of some help when investigating sources of income (chiefly the sale of
livestock and livestock products) and livestock markets where records are
sometimes routinely kept. But surveys of livestock holdings are rare and are
notoriously difficult to carry out accurately: absolute figures should be treated
with caution unless there is clear evidence that the survey in question is
trustworthy. Even then, although surveys can be informative about average herd
sizes and compositions, they tend not to give information on the distribution of
livestock within a community; thus they are not much use for information at a
household level, which is what you are mainly looking for.
The national population census
This is, of course, vital if you wish to estimate the population of food economy
areas, but it can also be useful for information about sources of income,
particularly hints about wage labour (an imbalance of male and female
inhabitants in an area may suggest a remittance economy, where men have
moved away to work). Censuses can also be useful in cross-checking information
66
Finding the information ●
for economic balance, to establish whether towns are large enough to provide
markets for the volumes of trade described by key informants.
Censuses tend to be undertaken at ten-yearly intervals, may be published only
after a long delay and may be out of date by the time of your research. In some
areas you will need to obtain estimated rates of population growth and apply
these to the census population to calculate the current population. If you do this,
remember that national growth rates may obscure major variations in population
growth between one area and another.
Price data
These are useful for verifying sources of income, as they enable the crosschecking of prices in a normal year, and so help you to check key informant data
on normal year income (for example, from livestock sales) and expenditure on
food (from food prices, especially cereal prices). They can sometimes also give
historical evidence of how far terms of trade are distorted in a food crisis; how
the exchange value of assets such as livestock collapses against grain prices. But
published data is usually only from the largest markets, especially provincial
capitals, which may not reasonably represent a given food economy area. Also,
price data are difficult to interpret unless you have a very good historical
knowledge of the area and of what underlay the price trends. Rates of currency
inflation will usually need to be taken into account in interpreting long-term
price trends.
Non-governmental organisation (NGO) project information
This is highly variable in quality, but most useful where the NGO works on
micro-credit, agriculture or livestock projects. It can provide useful (and unique)
information on sources of income; for looking at wealth categories (since reports
sometimes include the results of wealth rankings); for understanding the assets
of a household; and on seasonality.
Government district or regional reports
These often contain census, livestock and agricultural survey data, and can be a
useful digest of other government information sources. They are often
particularly good for background information on economic activities and
67
● The Household Economy Approach
employment in an area, thus helping you with sources of income, and may also
have information relevant to drawing the boundaries of the food economy.
Relief food delivery information
This is important for recording sources of income in situations where regular
relief food is an important component of “normal” household budgets. However,
this situation is uncommon, the information is often very poorly recorded, and
difficult to obtain other than in highly aggregated form.
Tables of labour requirement in agriculture
These are useful for cross-checking the crop production and employment aspects
of sources of income.
The internet
An increasing amount of information is available on the internet, although it is
often of a general kind (Annexe 4).
2. Key informants
Most of the time, the information you require, especially that relating to
households, will be provided only sketchily, or not at all, by documentary
sources. The information you require must be gathered in the field. The starting
point for this is the key informant. A “key informant” is someone you have
reason to believe can tell you what you need to know. For instance, a trader will
be able to tell you about (amongst other things) trade routes and seasonal prices;
an agricultural extension officer could tell you about land holding sizes, crops
grown and on-farm production levels in the area where she or he works; a
veterinary officer could tell you something about the numbers of livestock in the
area. Members of households can also be used as key informants and are often
the best informants on how they get food and income. It is a matter of deciding
what you want to know, then finding someone who might be able to tell you.
68
Finding the information ●
Key informants are most useful for providing an overview of the area you are
looking at.
From secondary sources you should have acquired some idea of possible HEA
areas, and perhaps of problems with these notional areas (for example, they are
the same ecologically, but not in other ways). You can explain what you are trying
to achieve and ask informants to draw maps of what they think are food
economies. You can also ask about specific matters that you suspect may lead to
subdivision of an area: are all the people along the river cultivators? Do they all
cultivate the same crops? Do they all sell crops? Do they sell these crops in the
same market?
Your main types of key informant are likely to be:
●
individuals who work for organisations delivering services to rural
populations (government, NGOs, UN, religious)
●
members of the communities you are researching
●
other researchers (from, for example, national or foreign universities, the UN
or an NGO). Researchers may be found anywhere. You may be introduced to
them in the community; or they may be a long way from the community,
writing up research in the capital. The easiest way to find out the names and
whereabouts of researchers is by asking around research institutions and
universities, and by following up names from your review of secondary
sources.
At the same time, you should remain open to the possibility that you will find
key informants in unexpected places: the retired teacher who has lived in the
locality for decades, has her own agricultural plot and a lively interest in her
community; in a tea shop or from herders who ask for a lift – long bush rides are
as good for bouncing around ideas as for bouncing bones.
If you approach key informants properly, you will obtain good-quality
information that is otherwise unavailable. Some key informants may have spent
decades working with the people you are interested in and have a very detailed
knowledge of their subject. The information obtained from key informants is
also usually critical when you progress to interviews in communities, as you will
already have a good basic knowledge of the economy concerned.
Not all informants are of equal value. You will find that a large proportion of
69
● The Household Economy Approach
A few tips... Key Informants
●
Try to give key informants advance notice of your visit, and of the issues you
will want to discuss.
●
Arrange to meet key informants through people they know: their superiors
(if in an organisation) or people who work with them (for communities).
●
Prepare carefully. Have a good idea of the types of things you need to know,
and background information to help you ask relevant questions.
Choose a variety of key informants. Include women and people with different
social/technical backgrounds.
the useful information which is obtained is provided by a small proportion of
informants. It is worth asking to be directed to people who have a reputation for
having a knowledge in a particular subject area, and (tactfully) terminating
interviews when it is clear they are going nowhere.
Whenever possible, try first of all to contact the ministry of agriculture,
NGOs, the WFP, etc, at national level. This can go hand-in-hand with the
literature search. Explain what you are doing and ask for support. By going to
the top, you might be able to use communication networks within the
organisation to warn people about when you want to visit them in the field, and
to give them some idea, before you meet, of what you wish to talk about.
70
Finding the information ●
3. Information from the community
Most of the information required for an HEA enquiry will be obtained from the
community, typically in group discussions which are used to establish the wealth
groups in the community, and to develop household budgets for each wealth
group.
This section discusses the selection of communities and the approach to
community discussions. Methods and techniques are described in greater detail
in Chapter 6.
Selecting communities
Selecting the communities (most often villages) in which interviews are to be
conducted calls for some preparation. Ideally, the selection will be based on an
overview of an area obtained from secondary sources and key informants. This
will allow the selection of communities which represent the range of economic
situations found within the area: for example, areas which represent slightly
different agricultural conditions. The number of communities selected will
depend on the nature of the enquiry and the practical constraints.
It is not enough simply to pick a likely-looking village from the window of
your vehicle or the back of your donkey, arrive and start interviewing. If you do
walk into a community unannounced and start to ask questions about wealth
and poverty, assets and crop yields, you will probably find that you get at best
poor interviews, and at worst complete silence. The questions you are asking do,
after all, sound suspiciously like those of the tax inspector. The more sensitive the
issue (for example, about smuggling or illicit hunting), the less likely you are to
find the interview illuminating.
You should always be introduced to the community. An introduction may be
in writing from, or in the company of, district officials, or through an NGO or
church group, who can introduce you to their congregation.
The best way to get information in a village has to be discovered in each case.
In some places enquiry can be fairly direct – for example, it may be possible to
establish the wealth groups by simply discussing this with the village chief or
administrator. But it may be necessary to put a group of people together and
71
● The Household Economy Approach
develop a view of wealth groups in stages, using ranking, etc. Approach each case
with an open mind – there are no absolutely fixed ways of getting communitylevel information.
4. Observation
Every opportunity should be taken to directly observe:
●
household conditions
●
the condition of livestock and crops
●
markets, which will allow direct estimates of price, and sometimes some idea
of the quantities of commodities being sold.
Useful observation requires knowledge of what you should be observing: for
example, thin cattle may be abnormal, or it may be seasonally normal.
Conditions in the poorest households may be of extreme, even shocking poverty.
This may be how people usually live.
5. Measurement
Direct measurement is not usually part of an HEA study. An exception is local
measures. In many locations local measures are used for agricultural and other
produce. Tins, bottles and volumetric measures are the most common, as
accurate (or trusted) weighing machines may not be available. The measures used
are usually standardised (eg, a tomato tin, a sack) and widely understood within
a locality. As the calculation of food energy is based on weight, it is important to
know the equivalent weight for each measure for the more important food items.
In some cases this will be a known (eg, in northern Rwanda a basket of sweet
potatoes is equivalent to roughly 5 kg) but often the equivalence will have to be
measured directly. This can be done in a market using a portable spring scale –
small cheap spring scales can often be purchased in larger markets.
72
Finding the information ●
6. How many interviews are needed?
There is no hard and fast rule for the number of interviews you need to conduct
to characterise an economy. It is important to recognise that, in choosing the
sites, you are not attempting to construct a statistically representative sample.
Within the constraints of time, cost and access, your aim is to compile a
description of the economy that is sufficient for the specific objective of your
enquiry. It depends on:
●
how the information will be used
●
the degree of access possible
●
the time available.
Your confidence in the results will be based on:
●
the completeness of the appraisal. HEA focuses enquiry on those aspects of
economy about which you need to have information
●
the consistency of the information received. The greater the number of
independent interviews (that is, where the informants are not known to each
other) that yield consistent outcomes, the more confidence can be placed in
the findings
●
the consistency of responses from different interviews.
In extreme situations, you may not be able to gain access to a population. For
example, it may be impossible to enter a war zone, and you may have to
construct a view of the economy by speaking to key informants outside the area.
However, it may be possible to organise trained teams to sample a range of
locations thought to be representative of the range of economic situations within
the area.
73
● The Household Economy Approach
Summary
HEA information can be obtained from:
●
documentary sources: ethnographical, agricultural and other reports
●
key informants: people with a specialised knowledge of a subject or
economy
●
community interviews: the household wealth distribution, and
information of household economy
●
observation: crops, livestock, living conditions
●
measurement: establishing equivalent weights for local measures.
An investigation usually starts at the centre, the capital city or district
headquarters, and moves out to the field, allowing a picture to be built up
from a general overview to a detailed description.
In virtually all cases, HEA will require fieldwork.
There are no hard and fast rules about the number of interviews required
for an HEA analysis.
74
Contexts for Children’s Rights ●
6
Methods for getting
information
This chapter outlines some rapid rural appraisal (RRA) and participatory rural
appraisal (PRA) methods that are useful in HEA work, and describes approaches
to conducting interviews with key informants and groups.
The aim of an interview is straightforward – the interviewer is trying to find
out something which the interviewee knows. However, in many situations,
interviews can be problematic. Communication might be through a translator,
with the risk that the question will be changed; there may be misunderstandings;
the interviewee may be reluctant to respond to certain questions and so on.
The chapter covers:
●
tools commonly used in HEA
●
interview techniques with key informants and groups
●
measurement.
1.Tools
A variety of RRA/PRA techniques can be useful in conducting a semi-structured
interview. The main benefits of these techniques are:
●
to make complex and abstract questions clear
75
● The Household Economy Approach
●
●
to handle complicated information sets, eg, the proportion of households that
fall into different defined wealth categories
to rank lists of items such as foods.
However, remember that these techniques are tools to be used when necessary,
not as an end in themselves. There may be other simpler and more effective ways
in which some of these tasks can be achieved. In some places it will be found that
people have their own techniques for similar purposes. For example, the Nuer in
Southern Sudan use straw stalks to count cattle for dowries, in which case this
will do well in place of the beans which are widely used for “proportional piling”.
Some people, even those with low levels of formal education, are entirely at ease
with the idea of proportions. In some locations informants will be found who
have a complete knowledge of all the people and households in a locality, and can
give you an exact wealth ranking of the households, so that the use of tools may
be unnecessary. The use of these tools also takes time, which may be better spent
in other ways.
Ranking
Ranking can help you to establish the order of importance of a set of things. Say,
for example, you want to find out the relative importance of the different sources
of a household’s food. First you ask for a list of all the sources of food, and second
rank them in order of importance. For our needs, however, this will seldom be
enough, and you will need to use proportional piling (see page 80) to work out
the ratio of the different elements to each other, as well as their relative
importance.
There is no big secret about ranking. You can ask directly if you feel your
informant is able to give a meaningful answer. Or if you wish to spend more time
on ranking, you can use a couple of other field techniques. One is borrowed from
PRA and utilises different sizes of stick. The other, “pair-wise ranking”, is used in
several disciplines, and by a process of elimination narrows down the scope for
error.
Collect a number of sticks equal to the number of items you would like to
76
Methods for getting information ●
compare, making sure that each stick is of a different length. Explain to your
informants that you would like them to associate the longest stick with the item
that is most important and the shortest with that which is least important.
Hand the sticks to the informants and ask them to arrange them on the ground
so as to represent the order of importance of the items you are comparing. Wait
until the informants have finished arranging the sticks and discussing the order
amongst themselves before you ask about their final picture. If they wish to
make changes after your discussion, that is fine, so long as there is justification
for it.
Pair-wise ranking
Pair-wise ranking is used to overcome the difficulty people often have with
ranking more than two items at a time. When asked to think about how five or
six different items relate to each other, informants can sometimes find it difficult
to assess so many objects at once. Pair-wise ranking helps you break the process
down so that informants are only comparing two items at any one time.
An example:
You are working in an area that suffers from frequent droughts and crop failures.
The population has become used to adapting its behaviour in drought years to
overcome the threat of famine. In the course of discussions with key informants,
you record a variety of mechanisms for adapting to bad years. They are:
●
eating wild foods
●
migrating out of the area to find work
●
selling unusual numbers of livestock for food
●
rationing food consumption to one meal per day.
You would like to know the order in which a poor family in this area uses these
different mechanisms in a bad year.
Set up a matrix that lists across the top (in any order) the items to be
compared. List the same items, in the same order, down one side as set out in
Table 2 overleaf.
77
● The Household Economy Approach
W
ild
foo
ds
On
em
eal
Sal
es
live
sto
ck
Ou
tm
igr
atio
n
Table 2
Wild foods
One meal
Sales livestock
Outmigration
Now take the top item in the vertical column, and ask the informant to compare
it with the items listed horizontally across the top. In this case, the top item of
the column is “wild foods”. You cannot compare wild foods with itself, so you
move along the row to the right and ask: “Do families like this eat wild foods
before they ration themselves to one meal?”
You write the answer on the grid. Always mark the preference under the item
that is preferred. So if the informant thinks that families ration themselves to one
meal before eating wild foods, put a mark below “one meal”. If they eat wild
foods before they sell livestock, put a mark under “wild foods”. So after
comparing wild foods to the other three items, the table looks like this:
W
ild
foo
ds
On
em
eal
Sal
es
live
sto
ck
Ou
tm
igr
atio
n
Table 3
Wild foods
One meal
Sales livestock
Outmigration
78
✗✗ ✗
Methods for getting information ●
Table 3 shows that households ration themselves to one meal before eating wild
foods (one mark under “one meal”), but eat wild foods before selling livestock or
migrating (two marks under “wild foods”). This completes the first row.
Next, move down the column to the next option: “one meal”. You have
already compared one meal with wild foods, and as you cannot compare it with
itself, go straight to comparing one meal with “sales of livestock”, then with
“outmigration”, again working across the table.
Finally, compare “sales of livestock” with “outmigration”. You have now
compared all items with all other items, as in Table 4 below:
W
ild
foo
ds
On
em
eal
Sal
es
live
sto
ck
Ou
tm
igr
atio
n
Table 4
Wild foods
One meal
✗✗ ✗
✗✗
✗
Sales livestock
Outmigration
Totals
2
3
1
0
Add up the number of points that each option scored. You wanted to know what
the population would do first in a bad year; this will be the activity with the most
points. So in this area, in a bad year, people start by rationing themselves. If this
does not work, they eat wild foods, then sell more livestock than usual, and
finally migrate to find work.
You could just as well have asked which activity was “the most important”,
rather than which was “done first”, and you may well have obtained a different
set of answers. You could also have included a far larger number of options; but
the logic remains the same.
79
● The Household Economy Approach
Proportional piling
As noted above, you will almost always want to find out the relative importance
of different types of information. You are not simply asking whether people eat
more food they have grown themselves or more food they have bought – but how
much food they have grown themselves and how much they have bought. One
way to do this with key informants is by proportional piling.
First, you will need a hundred beans (or nuts, or pebbles, or anything else
readily available).
Ask your informants to put the beans (or whatever) in piles that show how
important each item is in relation to the others. It is best for informants to divide
the beans first and then tell you what the piles represent afterwards. They should
put more beans on the piles representing items that are more important, and
fewer beans on the piles representing items that are less important.
Informants must use all the beans. They have to imagine that the beans
represent all the food they eat during the year, or all the families in the
community, or all the cattle in the village, or whatever you are asking about.
Once the informants have divided the beans into different piles, you can
count them, calculating an approximate percentage for each pile.
Example
In a highly pastoral economy you have asked a group of women from average
families to estimate the proportion of households which fall into the following
wealth categories:
0 cattle
1 – 9 cattle
10 – 20 cattle
21 – 40 cattle
over 40 cattle
= destitute
= poor
= average
= rich
= very rich
Establish that the beans represent all the families in the community.
Informants may initially have trouble dividing the pile of beans into five or
six smaller piles. In this case, try starting with two piles. Divide the beans into
80
Methods for getting information ●
two piles, showing in one pile those families with more than 21 cattle, and in the
other showing those with 20 cattle or fewer.
Then ask for the informants to divide the rich pile again into two, showing
those with fewer than 40 cattle in one, and those with more than 40 in the other.
You would finish by having the informant divide the other pile into three,
representing those families with 0 cows; those with 1–9 cows, and those with
10–20 cows.
If you have trouble remembering what each pile represents, label them with
small pieces of paper.
2. Conducting interviews
Interviewing through a translator
A translator is used to translate, as accurately as possible the questions put by the
interviewer, not to insert their own opinions. Always take time before the
interview to discuss with the translator the techniques you will be using and the
general topic of conversation. Ask for a literal translation, rather than their
personal interpretation. Tell them that you would like to hear their impressions
afterwards, but that during the interview you want to hear the interviewee’s slant.
Wherever possible, try to work with the same translator throughout. A good
translator is a partner, who knows how to manage a group properly and can steer
you out of trouble if you ask an inappropriate question. The best translators will
quickly grasp the purpose and logic of the interview, and with practice should be
able to conduct interviews on their own. If a translator is really unsatisfactory
find someone more suitable – it is impossible to conduct good work with
inadequate translation.
81
● The Household Economy Approach
Semi-structured interviews
A semi-structured interview is one in which the interviewer knows exactly what
information they want but where the questions are not put in a specific order, or
even directly on the subject of interest. The questions will not follow the
inflexible format of a questionnaire. HEA usually uses semi-structured
interviews, either with individuals or groups.
It is essential to keep in mind the type of information you are after. As time
is usually short, your questions should be focused: always ask yourself whether
the question you are about to put will really elicit the information you need.
In general it is best to begin with open-ended questions (eg, what crops are
grown in this area?) which will allow the person to tell you what they know.
Leading questions (eg, is maize grown in this area?) that put ideas into the
discussion, should be left until the end of the interview, if they are asked at all.
Interviewing individuals: key informants
A checklist is useful to ensure that no subjects are missed. As long as the
informant understands your question, you should allow them to digress to a
certain extent – they are probably answering the question in their own way. So
do not adhere blindly to a checklist, but instead keep in mind what you need to
know and ensure that the questions are well understood; this should enable you
to tread the difficult path between covering the ground and allowing informants
to talk freely. The checklist is for you to be sure that you have covered the
ground. The most important point to remember is that you are attempting to get
information on specific topics, not asking specific questions.
Remember also that you have chosen the key informants because of their
detailed knowledge of matters that interest you. So it is pointless to ask
informants about matters of which they cannot be expected to have first-hand
information. You may as well make up the answers. If key informants are
unhappy about answering a question, or say that they do not know the answer,
take them at their word.
You may feel that some of the questions you ask, although within the field of
knowledge of informants, are too “difficult” for them to answer. But remember,
82
Methods for getting information ●
Some guidelines... semi-structured interviews
●
Be flexible in your approach. Let the informant explain points fully, and allow
them to “wander” if it is necessary for them to make their point.
●
Ask direct (not leading) questions: the questions to which you want to know
the answers.
●
Only ask questions to which the key informant can be expected to know the
answer.
●
Ask questions about groups of people, not about the informant.
Continually remind the informant of the terms of the conversation:
who you are talking about, what kind of year you mean, etc.
concepts that may be alien to you often form the very basis of how rural
households survive. When you talk to key informants, particularly at village
level, you really are talking to the experts: the answers to your questions consist
of basic information that underpins their lives. After all, you yourself probably
have a fairly accurate idea of how much you earn! People tend to know in minute
detail how rich and poor people are defined, how much food they grow and how
many days in a season poor households work for rich households.
In a questionnaire survey, the interest is in the interviewee; individual answers
are tallied together in an attempt to come up with the Big Picture. In HEA the
questions are not specifically related to the individual to whom you are talking.
“How many goats do you have?” is a survey question; “How many goats do most
poor families have?” is an HEA question.
Interviewing in the community: group interviews
HEA interviews conducted in the community are usually conducted with groups
rather than individuals. Group interviews differ from individual interviews in
that people within a group often have divergent views, which leads to interaction
83
● The Household Economy Approach
between individuals. It also raises the need for some management to ensure that
you remain on the subject.
Before conducting interviews in the community you should already have a
broad overview of the economy of that group – at the least you should know the
chief economic activities and have a sense of the normal social relationships. For
example, you should know if men are allowed to interview women (or the
reverse) and have a clear idea of what constitutes a household.
The more you know about a community before the interviews begin, the
more productive your interviews will be.
Keep your groups small, three to six people. The questions you are asking
require concentration, and discussion may be impossible in a very large group.
You will not necessarily get more varied opinions with a group of thirty than with
a group of three. If possible, and socially appropriate, include some men and
women in the group as they often have complimentary information. Do not let
one person dominate the discussion.
Interaction between group members should (within reason) be encouraged.
3. Specific areas of enquiry
HEA work requires that you define food economies, households, wealth groups
within communities, and household budgets for each wealth groups. This
section gives some ideas as to how this can be done.
Defining the food economies
Food economies are usually broadly outlined from documentary sources and
from key informants. The aim is to identify a population which shares a broadly
similar economy, and which will be vulnerable to the same sort of events. Some
questions which may be useful are:
84
Methods for getting information ●
●
●
●
Do people produce the same things across the area (eg, crops, livestock, trade
goods such as charcoal or handicrafts)?
Is this production carried out in the same way? For example, if people are
cultivating sorghum, are they doing it at the same time of year? Are they
getting the same kinds of yield across the area? Is all the sorghum rain-fed, or
is some irrigated? Are all groups growing it for themselves, or are some
basically sharecropping for others?
Do people trade the same things? For example, is it normal for the population
right across the area to consume most of their sorghum harvest, or are there
some people who, perhaps because they are near a market, trade it all?
An example from Uganda
A risk mapping team had outlined a provisional HEA in Rakai and Masaka
districts of Uganda. The boundaries of the food economy were based on
information gained from studying secondary sources in Kampala. Weather
patterns, crops and yields very similar across the area.
However, in the course of interviewing key informants in a variety of sites
across the provisional food economy, it became apparent that land tenure
relations in the north of the area and the south were very different. In the
north, there were large numbers of landless tenants, whereas in the south, far
more of the population had freehold-type access to land. Thus the ways that
people made a living in the two areas were very different. As a result of
talking to key informants, the initial HEA was divided into two.
85
● The Household Economy Approach
Note: Do not worry if this process sounds complicated. The definition of
food economies is an integral part of the HEA analysis, and not something
that has to be done separately from the rest of the work. Once you have
defined the possibilities, the food economies become apparent as you begin
to look more closely at how the rural economies in the area of study work.
This is because a food economy is an area where a group shares the same
risks and vulnerability because they live the same way. As your task is to find
out how people live, it will quickly become apparent if you are looking at one
food economy or at several. If people in different locations are telling you the
same things about an area, then they are all in the same food economy. If
they say people are doing very different things, you are looking at two or
more food economies.
Defining households
The challenge here is to discover which social unit within a society most closely
matches the definition of a household used in HEA work, ie, a group of people
who work together as a production and consumption unit.
Anthropological or ethnographical studies of the people with whom you wish
to work may describe the elements of their social structure. However, in many
cases it will be necessary to ask key informants. Often this will be straightforward
– a household will be a man, a woman and dependent children. In some cases
the situation may be more complex. The basic approach to defining a household
is to try to find out who produces and consumes what. For example, if you are
working in a polygamous, agro-pastoralist society:
Q. When the maize is grown, who does the work in the fields?
A. The women do it. Sometimes the man helps the youngest wife.
Q. Do all the women work together on the same piece of land?
A. No. If the man is rich, his wives may all work together growing tobacco on one
piece, but they grow maize on their own piece of land. Sometimes all the women
in the village will work together.
86
Methods for getting information ●
Q. Does that happen often?
A. No. Normally only at harvest time.
Q. After harvest, do all the wives share the maize? Do they eat it together, or put
it in the same granary?
A. No. Each woman has her own granary. They eat the food with their own children.
Sometimes the man might tell one wife to give some food to another, if she doesn’t
have enough. But this is unusual.
Q. Does each woman own animals?
A. Yes, they have their own chickens and goats.
Q. Cattle?
A. No! The cattle are looked after by the man. In the dry season, the man takes them
to pasture.
Q. And in the wet season?
A. They stay around the homestead. Most of the cattle are for the children. Each one
is for one child. If the family gets new cattle, the father will say this one is for so
and so.
Q. What if the animal is sold? Does all the money go to the child?
A. No. The father will keep it and spread it out between the wives, depending on
who needs it. But most will go to the mother of the child.
In this case, the household is the female-headed unit. This unit produces food
crops together (using communal labour at peak periods) and consumes together.
They may redistribute food to other female-headed units within the male-headed
family, but such redistribution is irregular. Livestock and livestock products,
although largely owned by the man, are allocated to individual children, and thus
to female-headed units. Although income from livestock sales is shared, the bulk
will go to the female-headed unit with the greatest share in the animal. You can
assume that, over time, the remaining money that goes to other female-headed
units is balanced by money that comes into the unit in question from the
households of other wives. To be sure of this classification, however, you would
need to find out more about how money is earned and divided, but the
discussion above should give you some directions for your questioning.
Often village-level informants will be able to tell you the indigenous names
for different types of family unit. It helps to use these names during discussions,
87
● The Household Economy Approach
to ensure that you and the informant are talking about the same groups. If any
ambiguity exists, you should explain in your report what the terms for
households mean, and the average number of persons in the household type you
have used.
Wealth ranking
You are trying to discover three things:
●
What are the criteria that differentiate poor and rich households?
●
Using these criteria, how are households grouped within the economy?
●
What proportion of the population falls into each group?
In a typical HEA enquiry it will be necessary to establish the wealth groups at
each site visited as this will lead on to interviews with groups drawn from each
category identified.
Two types of secondary source can be particularly useful:
●
NGOs and other organisations often carry out wealth rankings to help their
planning. They usually divide the population into groups according to wealth
and then describe what proportion of the population falls into each group.
This is what you are trying to do as well, so it makes sense to use any work
that has been done already. But do check that the wealth ranking is
representative of the food economy you are looking at, and not just of a
village within the food economy. In many societies (including western Europe
and north America) there is a tendency for rich people to live together in one
community and poor people to live together in another. In this situation, the
proportion of rich or poor people in any one place is not going to reflect the
proportion of rich and poor in the area as a whole. Even so, wealth ranking
that has already been done can be extremely useful, particularly if you can
discuss the information with the people who compiled it.
●
The second type of source is more diverse, and comes under the general
heading of socio-economic information. This includes data about land
holding, livestock ownership, the hiring of off-farm labour and a variety of
other topics collected by household survey. Once you have established criteria
for wealth and poverty, and key informants have given you an idea of how
88
Methods for getting information ●
many people are rich and poor according to these criteria, you can crosscheck what they say against the survey information. For example, if you are
told that the real mark of poverty in an area is owning less than one hectare
of land and that about one-third of the population is poor, you can use land
ownership surveys to check if about a third of households own less than one
hectare.
In a community it is often straightforward to establish the wealth groups.
Rural populations usually have very strong views about wealth differentiation.
You will often find that socio-economic groups are locally classified and have
local names. In Islamic societies, identifying the recipients of alms (zakat) may
be a good way to identify the poor. In some cases an individual may be found
who has a complete knowledge of all households and can both classify these
and organise them into groups. If this is not the case, the best way to divide
the population into wealth groups is to follow the three-stage method
described below.
Begin by finding out what criteria are used to differentiate between
households of different wealth. It is best to start with either the better-off or the
poor, the ends of the scale, and to fill in the middle groups as a second step. Try
asking questions like: “If you took me to a rich (or poor) household, what would
we find?” or: “How do you know if a household is rich (or poor)”. The answers
to these questions often revolve around ownership of livestock, ownership of
land, family size, hiring labour, food consumption, quality of housing and
economic activities.
The last step is to use this information to make generalisations about groups.
As suggested above, you may well find that the key informants have their own
categorisations. If not, take all the facts you have been given and use them to
divide the population of the area into categories. Always begin by defining the
poor category and the rich category, and then define the rest, who are neither
poor nor rich: it is much easier to establish the ends of the wealth distribution
than the middle.
In discussing the better-off and the poor it is important to ensure that you are
not discussing people who are rare exceptions. In some places you may find one
or two “rich” people (landlords, moneylenders) and a small number of near
89
● The Household Economy Approach
destitute poor. Keep in mind that you are discussing the top 10% and the
bottom 10%, not exceptional cases.
Example
In the ‘Lakes’ food economy of Tomboctou, Mali, only the rich own large parcels
of land. A few households own small parcels, and the majority of the population
work as sharecroppers on the land of the rich. Very few people own cattle: again,
this is the prerogative of the rich. Many households own goats. Most households
own donkeys, which are necessary for carrying firewood for sale and for domestic
use. If you don’t own a donkey, you are really poor. Because the rich own so
much land, they use the labour of others to work it. This is normally
sharecropping labour, but they also hire labour for a few days at peak periods.
Most people neither hire labour in, nor hire themselves out, but the really poor,
in addition to sharecropping, hire out their labour at peak periods. Rich
households can have up to twenty people: they benefit from having a lot of
labour, and varied opportunities to make income, and their success attracts
relatives, who come and live as part of the household. Most households have
between six and eight people. This means that they have enough labour to work
the fields, and also to send at least one person out of the area to get profitable
work elsewhere. Poor households are small, and do not have enough men to send
away to work.
Discussion revealed that:
●
The rich group are the big landowners, who have other households
sharecropping their land, who own cattle and who hire labour at peak
periods. They normally have large households.
●
The poorest group are all sharecroppers. They own no livestock. They are
almost always quite small households, or households with very little labour.
They do not send members outside the area to work, and the land they work
by sharecropping is very small. They will undertake local, paid labour at peak
periods.
You can now ask about the rest: “Are the households that are not rich and not
poor almost all the same?”
Absolutely not. They all have donkeys, but they do not all have goats. There
90
Methods for getting information ●
is a wide variation in the amount of land they cultivate. Those who cultivate
more land tend to have goats. The small landowners normally have goats too.
You can divide this group into three:
●
The poor who sharecrop less than two hectares in a year. They will seldom
have more than five goats.
●
The small landowners, who normally own between one and two hectares, but
are richer because they keep all their produce. They will normally have
between ten and twenty goats.
●
The large sharecroppers, who cultivate up to six hectares, have between
ten and twenty goats, and may hire labour in the same way that rich
households do.
So you have five basic groups: rich landowner, poorest sharecropper, poor
sharecropper, small landowner and richer sharecropper.
The final step is to discover what proportion of the population falls into each
group. This is probably easiest to do by proportional piling (see page 80). Once
you have agreed the allocation of the groups with the key informants, bring out
the beans and explain that the total number represents all the households in the
area. Each bean is one household. Can the informants divide the pile into five,
to show how many households there are in each group?
Sometimes you will find that a very large proportion of the population has
been assigned to the poorest or the richest group. If so, try asking the informants
if they can further subdivide the group. As a rule, you will not want either
extreme of wealth to account for more than 25 per cent of the population: for
example, it would be a very unusual society where half the population is at the
bottom of the heap.
When you have defined the groups, and know their relative size, you can
decide which ones you wish to discuss.
Do not simply ask how many households are “poor”. Areas that have already
received food aid may have a vested interest in saying that everyone is poor.
Instead, ask how many households have the characteristics you specifically
associate with being poor in that area.
91
● The Household Economy Approach
Defining the baseline year
Your aim is to define the most frequently occurring type of year in terms of
production and exchange. In many cases it will be possible to designate an actual
year (say, 1997) as a “normal year” and then use it as the focus of discussion. In
cases where production varies a lot between years, as occurs in many semi-arid
areas, it may be necessary to establish a range of years which taken together make
a normal period.
Note that in some areas the situation may be changing so rapidly – for
example, as the result of drought or civil war – that it is inappropriate to talk of
a “normal year”. In these circumstances use the current year as your baseline,
recognising that the baseline will have to be changed once the situation stabilises.
To decide on which year (or years) to use, count out ten markers (beans, twigs
or anything else that comes to hand) and lay them out in a line. Explain to your
informants that each marker stands for one of the past ten years: the first is 1990,
the second 1991 and so on. Ask them to divide the markers into two piles: one
for good years, one for bad years. Then ask them to divide the good pile into
good and very good and the bad pile into bad and very bad. The largest single
pile of the four represents the normal years. Ask which years were in this pile; the
informants may say, for example, 1990, 1991, 1994 and 1996. Then agree that,
throughout the interview, you will be talking about the most recent of these
normal years; in this case, 1996.
A recording form is shown in Annexe 3.
Household budgets: interviews with wealth groups
Once you are sure of the meaning of “household” in the place where you are
working, and have defined the categories of wealth, the next stage is to interview
small groups of people from each wealth group to find out:
●
What are the sources of food income?
●
What are the sources of non-food income?
●
What proportion of total food income comes from each source?
●
What proportion of total non-food income comes from each source?
92
Methods for getting information ●
To complete a full HEA interview with one wealth group takes time (typically
about two hours) and it is important that all the information required is
obtained. It is therefore important not to get too distracted by incidental matters.
HEA requires the collection of complete household budgets – there is little
point in learning about food income
but having no knowledge of non-food
income.
It is important to note that
Interviews with groups are always
production costs, eg, for fertiliser,
semi-structured. However, the form
labour or land rental should only
shown in Annexe 3 gives a good basic
be included once. If a person is
structure for a wealth group interview.
employing labour to work land,
Following the main sections of the form
this should be recorded as a
also allows the interview to proceed
cost. If fertiliser is used, the cost
methodically through the main topics –
of this is usually deducted from
sources of food, sources of other income
and expenditure – and ensures that
the amount of value gained, and
nothing is omitted.
only the profit recorded. If you
At the outset of the discussion it is
do this, do not also record
important to ensure that the group is
fertiliser as an expense.
made up of people from the same
wealth group: for example, members of
poor households or of rich households.
If people from different wealth groups are included confusion will result for both
the interviewee and the interviewer. This is especially important where the power
relationship is unequal – for example, between rich patrons and poor clients – as
people in the less powerful position may be loath to talk.
You should begin by clearly defining who you are talking about:
●
The wealth group. For example, you know that the poorest households in this
area farm less than half an acre of land and have no animals. Check that all
those present fall into that group.
●
The size of household. Establish with the group that the discussion is about
a typical household from that group, not a specific household. Define the
number of people in the typical household (eg, mother, father and four
93
● The Household Economy Approach
●
children). A way into this is to ask each member of the group to outline their
own family membership. Derive a typical household from that.
The normal baseline year, which you are discussing.
Then establish the sources of household food supply. “So, a household which
farms less than half an acre and doesn’t have animals... how do they get food in
a normal year?” List each response.
You may find it useful, when you have the lists, to rank the importance of
each item.
Once you have a list of activities, the next task is to find out the relative
importance of each activity: how much of total food/non-food income is covered
by each activity. There are two ways of going about this: adding things up and
breaking things down.
Adding things up
The procedure here is that, armed with your list of economic activities, you take
each activity in turn and find out how much food/non-food income it produces.
Once you have covered all the activities on the list, you can check that the food
you have been told about adds up to at least the household’s total consumption
needs, using the methods outlined below.
This approach depends heavily on information about production. You will
tend to move from questions about how much the household produces, through
questions about what is done with the produce, to questions about how much
they consume (which is the most important part), how much is sold, how much
stored as reserve, etc.
Here is an example. Having agreed on how to describe poor households and
a normal year, you ask your informant:
Q: What would a poor household, like the one you described, do to get food in
a normal year?
A: Well, most of it they would buy. But they would grow some food too.
As the informant hasn’t mentioned several categories that appear on your list, you
should quickly check whether they are relevant to the discussion:
94
Methods for getting information ●
Q: Do they ever collect fruits or other foods from the bush?
A: Not really. In a very bad year, when the rains fail, they might, but not in a
normal year.
Q: What about fishing? Do poor households go fishing in the rivers or the
wetlands around here?
A: Yes. Most poor families fish a little most years.
After a few more questions like this, you have a list showing crop production,
purchase, fishing and gifts as sources of food income for a poor household. You
now take each item in turn and build up information about it. You start with
crop production:
Q: What type of crops do poor families grow?
A: Mostly they grow groundnuts, sorghum and millet.
You then ask for an approximate ranking of the importance of these crops for
poor families, and begin with the most important food crop, in this case
sorghum:
Q: How many harvests of sorghum would a poor household get in one year?
A: Just one. That’s true for all crops: there is only one growing season here.
Q: So just how much sorghum would a poor household expect to harvest in most
years?
A: Between two and three sacks for the sort of family I described.
Q: How large is a sack?
A: 90 kilograms.
Q: And how about groundnuts?
A: Two sacks.
Q: Is that with or without the shells?
A: With the shells.
Q: How much would that be when they were shelled?
A: About two-thirds of a sack.
Q: And can you tell me about millet? How much millet would a poor household
harvest?
A: Probably one sack or a little bit less.
95
● The Household Economy Approach
Q: Are the groundnuts, millet and sorghum all consumed?
A: Yes, for the most part. They would eat all the sorghum and millet, but they would
eat about half the groundnuts and sell half.
You can now move on to discussing the next item, purchase, finding out how
much food is being bought, when it is bought and how much it costs.
Continue with each option in the same systematic way: look in each case at
how much food comes from each source, and how much of it is consumed, sold,
given away, etc.
When you have completed the information set for sources of food, you can
move on to asking about sources of non-food income; again, by listing the
sources of non-food income and then finding out how much income comes from
each source. In the example above, you will return to the information about
selling groundnuts when you move on to discussing sources of non-food income.
The advantages of this method are that:
●
It provides a lot of information that can be cross-checked against itself (in the
example above, does two or three sacks of sorghum and one of millet sound
reasonable for a household cultivating half an acre?), and against secondary
sources (does two or three sacks of sorghum and a sack of millet for half an
acre agree with yield tables? Does half an acre sound reasonable for a
household of six to cultivate, given the literature on labour requirements for
sorghum and the information from the informant about the amount of offfarm labour the household undertakes?)
●
It can be carried out in instalments. If you have an informant who is
particularly knowledgeable about agriculture, but doesn’t know much about,
say, wild foods, you can collect just the agricultural information and add it to
information from other specialists to build up a more complete picture.
However, the disadvantages of this method are:
●
It is time-consuming.
●
In order to construct the food economy, what you are really interested in is
food consumed. Calculating consumption by analysing production is going
the long way round, and to do it effectively, you need to have a good
knowledge of the entire household economic cycle so that you know what
96
Methods for getting information ●
happens to all the food produced; otherwise, you will be left with production
that does not seem to be consumed but which you cannot account for. An
example of this would be grain that is lost to pests during storage or is used
for brewing. You might not think to ask about this, and so might assume that
the grain is eaten. This will make it impossible to construct the food
economy, as your figures will not add up.
Breaking things down
Again, before you do this you should already have a list, which you believe to be
complete, of economic activities: food generating and non-food generating.
Begin by obtaining a broad picture of the relative importance of each food or
income source, by proportional piling if necessary, then work through each
source individually, checking that the information about it fits with the overall
information you have been given.
As applied to sources of food, this approach goes directly to what a household
consumes, and production has only peripheral importance.
Here is an example. Looking at the same household type in the same area as
above:
Q: So all the food a poor household eats in a normal year will come from
purchase and exchange, crop production, fishing or gifts?
A: That’s right.
Q. OK, what I want to know now is how important each kind of food is to that
household. So what I would like you to do is this. Imagine that this pile of
beans represents all the food that the family eats in one year: the food they
buy, the gifts, the fish and the food they grow, all together. I’d like you to
divide it into four piles, one for each kind of food, so there is one pile for food
they buy, one for food they grow, one for fish, one for gifts. The size of each
pile should show the importance of that kind of food to the household. So if
they eat a lot of fish, you make a big pile for fish, like this [demonstrates], but
if they only eat a small amount, then you only make a small pile, like this.
Divide all the beans into four piles, so that we can see how the different kinds
of food compare.
A: Like this?
97
● The Household Economy Approach
The informant divides the beans up thus:
Crops 25%
Gifts 10%
Fish 15%
Purchase 50%
The interviewer then asks questions about each source, attempting to check that
the proportion given in the exercise is consistent with the informant’s estimate of
each individual element of income:
Q: Thank you. Let’s start with the biggest pile. What is this for?
A: That’s the food that the household buys.
Q: What sort of food do they buy?
A: Maize, beans. Some tea and sugar.
Q: Anything else?
A: Not really. Sometimes they might buy some milk, if there is somebody sick, but
very seldom.
Q: How much maize do they buy in a month?
A: That’s very hard to answer.
Q: Why?
A: Because it depends. They buy more when they have no sorghum or millet.
Q: When is that?
A: Normally from December to June or July.
Q: And how much do they buy in those months?
A: About half a bag a month.
Q: What about the rest of the year?
A: They don’t really buy maize the rest of the year.
Q: So in one normal year they buy half a bag a month for seven or eight months.
About four bags in all?
A: Yes.
Q: What about beans?
98
Methods for getting information ●
A: They buy them at the same time as the maize – about half as much.
Q: Half as much as the maize? So about a quarter of a bag every month for seven
or eight months?
A: Yes. So that would be about two bags a year.
So far you have discovered that the maize and beans bought come to between five
and six sacks per year in total. That would represent between 40 and 50 per cent
of minimum food needs. From what the informant goes on to tell you about
sugar and oil, the total purchase of these roughly equals 50 to 60 per cent of the
household’s minimum food needs. This falls within 10 per cent of the
proportional piling estimate (which was 50 per cent), so, given that you can
allow a variation of up to 20 per cent around the actual figure, this is quite
acceptable. You would in any case expect a household to eat slightly more than
the absolute minimum they need to stay alive, and so the total figure for what
people actually eat (as represented in the piling exercise) and the total for what
they have to eat (which you have used to cross check the exercise) may be slightly
different.
You would go on to cross-check the other proportions from the piling exercise
in the same way.
The advantages of this method are:
●
It is far quicker than the adding things up method, and involves asking
questions about what you need to know (consumption) rather than working
through to consumption from production.
●
It is easy to cross-check the information for internal consistency as you go
along, as you have the full information set from the beginning and are
checking that each piece adds together to make a credible whole. In the
example above, for instance, if purchased food was given as 50 per cent in the
piling exercise, but the informant later said that households bought one sack
of maize per year, which would meet less than 10 per cent of energy
requirement, it would quickly become apparent that something was wrong.
By contrast, the adding things up method does not give you a picture of total
income until the interview is over.
●
There is a visual focus for the discussion, and this helps to minimise
misunderstandings between you and the informant. For the same reason, it
99
● The Household Economy Approach
works well in group situations, giving each member of the group something
to focus on.
The disadvantages are:
●
Since you are mainly trying to confirm the truth of the initial, proportional
piling estimate, you will tend to collect less information. This makes for a
quicker interview, but also means you have less to compare with secondary
sources later, which makes cross-checking more difficult.
●
The piling exercise can create confusion. The question ‘What proportion of
food comes from...” is open to a variety of interpretations, depending on how
food is measured. By quantity? By quality, as locally perceived? By quality (in
this case calories) as perceived by us?
●
This method only works if you have a key informant who knows about every
aspect of food income for a household, so that they can give each aspect a
relative value. If the informant knows a lot about crops, but very little about
fishing and gifts, and nothing at all about wage labour, you are wasting your
time asking them to position all types of food income in relation to each
other.
Things to look out for. Whichever method you choose (and you will probably
want to use a combination), you may find that asking questions about the
quantities of food produced does not get you very far. This is especially true
where food is not bagged or where much of it is eaten “green” (see page 116).
You may prefer to talk about how long the food lasts (for example, “How many
months of sorghum do you get?” instead of “How many bags...”). If you do this,
remember that a range of foodstuffs are normally eaten at the same time, and so
will “last” longer than they would if they were the only food being consumed.
Four bags of cereal are equal to 30 per cent of minimum food needs for a family
of six, which would mean that, if it was the only food eaten, the cereal would last
about four months of each year. As a general rule, one bag of cereal lasts one
household of six persons one month if it is the only food eaten. In this case,
however, the four bags of cereal will in practice be mixed with beans or
groundnuts, and so will last for perhaps six months. So do not automatically
assume that six months of cereal is equivalent to six bags.
100
Methods for getting information ●
When you move on to talking about sources of non-food income, key
informants will tend to think you are talking about the non-food income
used to buy food. This is not normally the case: you want to know about all
non-food income, regardless of how it is used. You must make this very clear at
the outset.
Where payments are made in non-monetary “value”, you should try to
express them in terms of how much staple food they would buy at normal market
price. This gives you an opportunity to compare different income sources and to
describe the proportion of non-food income each source provides.
Note that you are trying to define a “typical” household in a wealth category
when, in fact, there is some variation across the group. You will need to make a
judgement about what information should be included and what excluded.
Typical problems include:
●
Specialised activities that only some households in the wealth category engage
in: for example, brewing, gathering honey or hunting. These should be noted,
as they provide useful contextual information for any final report, but if only
a few households regularly benefit, the activities should be excluded from the
household budget. Include only items that represent a typical pattern of
expenditure.
●
Variation in expenditure. The take-up of health services varies widely, so you
will need to find a reasonably typical level of actual expenditure.
An important feature of a household budget is that it should be internally
consistent:
●
The amount of food recorded in the interview should be sufficient to allow
the household to subsist. This is discussed in Chapter 7.
●
Income and expenditure should balance. Methods for keeping a running
balance during an interview is discussed in Chapter 7.
What level of household food consumption is normal? There is no fixed level
of average energy intake per person which is “normal”. This will vary with levels
of individual activity, the ambient temperature, the availability of food, and
other factors (Chapter 1). In practice it is often found that in warmer areas where
people are not engaged in heavy work (eg, much of lowland Africa), recorded
101
● The Household Economy Approach
household food consumption is around 1,900 kcal per person per day (assuming
a household of a typical range of ages). However, this may vary from place to
place. The fact that you are getting information that indicates a somewhat lower
or higher level of average consumption does not necessarily mean that you are
being misled. Look for explanations in terms of unusual levels of activity or
environmental losses.
Markets
The name and location of all the markets used for all the main traded
commodities is easily obtained from key informants. A note should be made of
the geographical location of each markets – in many cases it will be difficult or
impossible to locate these on maps.
The current price of different commodities will normally be obtained as part
of the household interviews.
Non-market transfers and the availability of wild foods
An estimate of income obtained by non-market transfers (gift, reciprocity) as
part of normal household income (eg, gifts, relief ) will be obtained as part of the
household budget interview. If the information is to be used for understanding
the effect of a drought or other shock on households it will also be necessary to
get an idea of how this source of income may vary under conditions of economic
stress.
In many situations this information can be obtained by asking people about
what happened in previous periods of stress, such as rain failure.
Information will also be needed on the potential availability of wild foods,
hunting, and fishing – to what extent could a household cover a fall in income
by expanding these food sources? Clearly, exact quantification is impossible, but
it is usually straightforward to establish if the supply of wild foods is very small,
or if these are abundant, and to roughly quantify their likely availability. This
may be done, as with non-market redistribution, by reference to a period of stress
when people used wild foods.
102
Methods for getting information ●
4. Seasonal calendars
Seasonal calendars are a means of recording events during a normal year. They
can be constructed in a number of ways, and before, during or after a visit to a
food economy. They can include any number of topics, such as harvest and
planting times, sales of livestock, purchase of grain, movement to grazing lands,
peak fishing times, sickness among children, etc.
The advantage of a seasonal calendar is that it presents visually a number of
different activities that are carried out at the same time, enabling you to see the
correlation between them. For instance, if peak labour requirements occur at the
same time as planting and harvest, it is likely that smaller households (which you
have discovered through interviews resort to off-farm labouring as an important
income-generating activity) will not be able to devote much time to their own
plots. If the second peak fishing time, in November, coincides with the period
when animals are being moved to the river for the dry season, and also coincides
with the harvest period, poor families, who are less likely to be at the river as they
do not have cattle and who have to work as labourers during this period, are less
likely to be able to fish.
Because seasonal calendars reveal a range of possible interconnections, they
can spark off interesting debates. So you should construct one for your food
economy area as soon as possible, even if it is only very rough, to serve as an aid
to discussion.
A form for recording a seasonal crop information and seasonal labour use is
given in Annexe 3.
Constructing a calendar in a village
This exercise often works best with a group. This does not have to be a focus
group (that is, one consisting of, say, poor women only); in fact, the exercise
works better if there is a mixture of people.
Find a patch of ground and clear it of stones and other debris. Collect a
number of different objects to represent different activities (for example, stones
for harvesting of maize, beans for labour demand, leaves for fishing periods, etc).
103
● The Household Economy Approach
First, define the seasons carefully with your informants. Try to match the local
seasons to the months of the Gregorian calendar (if only in your head, but better
make a note on paper), but use the local seasons as your x-axis. Mark them out
on the ground, by drawing lines in the dirt or laying down sticks.
Start by correlating the rainy and dry seasons with the local seasons. Then
choose any one of the activities and ask your informants to place the stone, bean
or leaf representing it in the months or seasons when the activity occurs. For
instance, if you want to know when maize is harvested, ask the informants to
place stones in all the months when this happens.
Continue in the same way until all the activities or events are represented on
the ground. Your discussion can then focus on the connections between the
different activities. Do not forget to copy the calendar on to paper before leaving.
A variation on this can be used to show when an activity is at its most intense
or important. If the seasonal calendar shows you that an activity takes place
during six months of the year, take a hundred beans and ask the informant to
place them all under the months when that activity is occurring, in quantities
according to the importance of the activity during each month. This method
combines proportional piling and seasonal calendars to good effect. You can
create a histogram by proportionally piling beans (or another appropriate
symbol, such as maize or nuts) across a particular time period to show when the
activity is at its most intense. You can show, for instance, the rate of consumption
of different sources of food in this way.
Preparing seasonal calendars from documents
Secondary sources can be invaluable in understanding certain aspects of
seasonality, and are often readily available. In most places, you should be able to
find monthly rainfall charts and cropping calendars. You should also look for
farming systems surveys or seasonality surveys, which can tell you a lot about
seasonal labour use. Occasionally you may even find figures for seasonal
food consumption. Treat these with some caution: the methods used (such as
family members recalling their consumption over a 24-hour period) can be
inaccurate.
104
Methods for getting information ●
Keep the beginning of a seasonal calendar on hand when you read through
documents. Record events or activities as you come across mentions of them. In
this way you can collate information from a number of sources (and possibly
point up inconsistencies) while compiling a seasonal calendar.
See Figure 4, page 13, for an example of a seasonal calendar.
105
● The Household Economy Approach
7
Checking the accuracy of
the information
Where field information has been collected through interview rather than direct
observation and measurement, it must be verified. This is as true of the key
informant inquiry described in the last chapter as of the most rigorously sampled
questionnaire survey. Here, the household economy (HE) methodology and the
approaches to obtaining information we have looked at have an important
advantage: built-in methods for checking the accuracy of what you are
discovering. Before you begin an HEA analysis, you should be fully conversant
with these methods.
We first look at how accurate the information can be. Then we look at how
to ensure this accuracy. This is done by cross-checking: making sure that
information from a variety of different sources, or dealing with a variety of
related issues, is consistent. Each of these methods is explained separately below.
Finally, there is a discussion of some common mistakes that can compromise the
accuracy of your information.
Cross-checking takes five forms:
●
internal consistency within wealth group interviews
●
internal consistency between wealth group interviews
●
agreement between key informant information, written sources, observations
and other experience
●
agreement between key informants
●
balance between interviewers.
106
Checking the accuracy of the information ●
1.The limits to accuracy
Even within a defined wealth group, economy will vary from household to
household. There will therefore be no single “right” answer to a question. In
HEA work you will be looking at a range of answers to any one question. It is,
however, important that the ranges are within reasonable, plausible limits.
For instance, poor households in a community may reportedly survive by
consuming two to four 90kg sacks per year of their own cereals. Two sacks
provide approximately 17 per cent of the basic energy requirement for a family
of six; four sacks provide approximately 34 per cent (see Table 5). The same
families may reportedly buy nine to eleven sacks of cereals (75–92 per cent of
requirement). These two pieces of information are internally consistent, as there
is a large overlap between the two percentage ranges. In the worst case, a
household with two bags from its own production, which also purchased nine
bags, would meet (17 + 75) 92 per cent of its requirement from cereals. In the
best case, a household would have cereals equal to (34 + 92) 126 per cent of its
needs. The best estimate which we have is that in this poor group of households,
income from cereals is in the range 92–126 per cent of household energy
requirement. Methods of using interval information are discussed in Chapter 8.
2. Checking for the internal consistency of
information
The nature of the information you are gathering about how people live in a rural
society makes it possible to check for consistency. On the one hand, there is a
finite and relatively small number of economic options available to households,
so you will already know the broad parameters of the investigation. You will be
looking at how people combine a variety of known options to survive. There is
a maximum amount of a given crop that can be grown on a given amount of
land, and there is a minimum amount of food energy required to stay alive,
107
● The Household Economy Approach
under the conditions of the area, and so on. If all or most of these pieces of
information are in reasonable agreement (and we have reasonable explanations
where they are not), this indicates that we are right.
Cross-checking, then, is simply a way of confirming that all the pieces of
information you obtain fit together in a coherent way, and that the information
set enables the household to survive in a normal year.
Internal consistency within wealth group interviews
An interview in a wealth group takes the form of establishing a budget – income
from food and non-food sources and expenditure – for a typical household of a
defined size in that wealth group. The information obtained should be:
●
internally consistent: income and expenditure should balance (within the
limits of the estimates obtained, as already outlined)
●
consistent with the biological requirements of the household, and the
observed physical condition of people and their standard of living.
However, it is important that inconsistencies are picked up by the interviewer
during the interview – it is of little use if these are discovered at a later time when
it is impossible to return to repeat the interview. During an interview you should
keep a running tally of:
●
the quantity of food and cash income, and expenditure, to ensure that they
are roughly in balance. It is very difficult to fabricate a convincing balanced
household budget, and virtually impossible for this to be done by a group of
people during an interview. If people are trying to mislead, the answers are
usually so absurd that it is obvious. Where the budget does not balance, this
is often a clue that some item of information is being deliberately withheld:
for example, people may be reluctant to discuss smuggling or other illegal
activities.
If the information does not add up and it is clear that the responses are
incomplete, you should discard the data.
As far as possible you should try to reconcile the information during the
interview. If you find discrepancies you can open lines of enquiry that will
allow the budget to be balanced. If major discrepancies are found
subsequently, there is nothing to be done except to discard the information
108
Checking the accuracy of the information ●
●
the quantity of food obtained by the household from its own production,
wild foods and other sources. In many cases a reasonable average estimate of
the normal consumption for a poor person is about 1,900 kcal per person per
day (Chapter 1). The interview should indicate roughly this level of food
income or there must be a reasonable explanation of why it is lower or higher
than this: for example, better-off groups may sometimes have an average
intake greater than 1,900 kcal. People in cold environments may have a
higher energy intake. Lower intakes may be associated with extreme poverty
or with abnormal conditions.
An illustration of why this type of cross-checking is useful
A common response by women in southern Sudan to the question of what
they eat during the wet season is: “We have nothing, so all we eat is leaves.”
It is true that they eat very large quantities of leaves at this time, but it is
actually impossible to survive for long on the energy provided by leafy greens
alone. In fact, in order to obtain 1,900 kcal per person, a household would
have to consume around 40 kg of greens per day – quite impossible. Without
being able to place their response in the context of what is actually possible,
it would be tempting to take it at face value.
There are two ways to check an estimate of household food consumption. The
first is to:
●
build up a picture of all the food a family obtains in a year
●
convert all this food into energy values, using the tables (Annexe 2)
●
check to see that the total energy value is at least equal to 1,900 kcal (or the
value being used) per person per day.
This method is used at the end the end of the investigation (Chapter 8). It can
be difficult to use this method during an interview as it requires food tables and
a calculator and tends to intrude on discussion. In the field it is necessary to use
quicker less obtrusive methods.
109
● The Household Economy Approach
The second method is to use quick approximations, which allow you to keep
a running tally of food income. The trick is not to use kcal, but to have a clear
idea in mind about the sort of quantities of basic foods that are needed to provide
a household with 1,900 kcal per person per day (or the value being used). In
nearly all situations this is made easier by the fact that the diet of most
households in poorer countries is based on cereals and pulses – other foods often
provide only a small contribution to household energy consumption.
Most cereals contribute around 350 kcal per 100 grams. So, to provide 1,900
kcal you would need around 500 grams, or 0.5 kg. A household of six people
therefore needs about 3 kg cereals per day, or approximately 1 ton of cereals
per year.
For example: you discover that poor families in a location produce six 90 kg
sacks of grain each year – a total contribution of 540 kg to annual food needs.
You can see immediately that this will meet about half of the household’s annual
cereal needs.
It is possible to keep similar simple rules in mind for other common sources
of food. Some are shown in Table 5. It is worth developing rules for local use as
soon as the main sources of income are known.
Table 5: Short-cut approaches to calculating food needs
On average, each day, one person requires approximately:
0.5 kg of cereals, or
2.5 kg Irish potato, or
1.6 kg sweet potato, or
1.2 kg fresh cassava
Each year, a household of six people requires:
1 ton of cereals, or
3.5 tons of sweet potato, or
5.5 tons of Irish potato
100 kg of beans = 10% of annual household energy requirement
I litre fresh cow’s milk per day = 6% of annual household energy requirement
2 goats or sheep = (15 kg dressed weight) 1% of annual household energy requirement
20 sacks of 50 kg = 1 ton
11 sacks of 90 kg = 1 ton
110
Checking the accuracy of the information ●
Example
You are told that in a normal year an agro-pastoralist household of six buys six
90 kg sacks of grain (50 per cent of food needs). They grow two sacks of grain
each year (16 per cent of food needs), and they also get two litres of milk every
day for six months of the year from their animals (about 6 per cent of food
needs). In a normal year, they eat two goats from their own herd (about 1 per
cent of normal food needs) and three goats at feasts (about 1.5 per cent of food
needs).
The total food income, you have been told, about adds up to only 74.5 per
cent of the household’s minimum food need. There must be something you have
not considered.
You can make the same kind of calculation for any source of food. Money can
also be converted this way if you have enough information about grain prices.
For instance, say a household makes $50 a year, which is used to purchase food.
If grain is $0.50 per kg, $50 represents 100 kg, or around 8–10 per cent of
annual food income.
As you get more practice in food economy, you will find that making these
calculations as you go along, during the interview, becomes second nature. You
are continually checking for internal consistency.
Internal consistency between wealth group interviews
In some interviews it is possible to check for consistency between wealth groups.
For example, if poor people are working for the better-off rich and the better-off
are employing the poor, the income received by the poor should be in agreement
with the estimate expenditure on labour by the better-off.
Cross-checking key informant information against written
sources
Although, as we have seen, written sources will seldom tell you exactly what you
want to know, they do provide information that allows you to cross-check what
key informants tell you.
For example, several informants may say that in this area in a normal year a
111
● The Household Economy Approach
poor household consumes two or three 90 kg bags of the maize it has produced,
and sells between a half and one bag. This would mean that they are cultivating
two-and-a-half to four bags of maize per year (225 to 360 kg), which seems very
low.
The key informants are less specific about the amount of land poor
households might plant with maize. For the poorest 20 per cent of households,
some said about a quarter of an acre, most said about half an acre and one said
two acres. They all say that poor households would use all their land to grow
maize and would not cultivate any other crop.
You want to know how much land poor households cultivate with maize in a
normal year. With this figure, you hope to cross-check what you have been told
about household maize production.
The district development report shows that over 90 per cent of the
population are smallholder farmers. A small farm survey, conducted on a random
sample of households in two villages, shows that 30 per cent of the population
have less than one acre and 5 per cent have less than half an acre. Since the
population is mainly agricultural and land is the major means of production, you
would expect small land holdings to correspond to poverty. So the two written
sources seem to agree with key informants that the poorest 20 per cent will
cultivate less than one acre. They also explain why some informants said poor
households had only a quarter of an acre: some households do, but the majority
in this category (25 per cent in the survey) have half an acre or more, but less
than one acre. Consulting UN Food and Agricultural Organisation (FAO)
labour tables for maize cultivation will show you that a family of six would have
enough labour (after performing wage labour for others) to cultivate half an acre,
but not enough for one acre. Here, documentary sources explain and reinforce
key informant information.
The same survey shows that yield is low in the area: only seven bags per acre.
If this is correct, and the land held is about half an acre, than the key informant
figures for production are also roughly right.
So here the information has checked itself (land size and production add up),
key informants roughly check each other, and the use of documentary sources
has allowed you to confirm this.
112
Checking the accuracy of the information ●
Checking agreement between key informants
As you are only asking key informants for information they should know, you
should expect them to agree on matters over which there can only be a relatively
narrow range of imprecision. Some may tell you that a modal household
consumes six sacks of its own cereal a year, some that it consumes five, but as
discussed earlier in this section, this is acceptable. In general, though, you should
find that key informants are saying similar things.
It is, however, important to check one informant against another, and this
should be done as you go along: reviewing interviews day by day, seeing where
inconsistencies arise, and concentrating on them the following day.
When major inconsistencies arise, you should ask the key informants why
this is so. If there is no rational explanation, it is often best to believe the majority
opinion among informants, as long as this is internally consistent and agrees with
the written sources.
For this method of cross-checking to work, you must choose the key
informants carefully so that they represent a variety of viewpoints. Wherever
possible, take your findings back to key informants and discuss them. If this is
not possible, you should at least ensure that the findings are sent to the
informants with a request for any final comments.
Balance between interviewers
The methods of cross-checking so far discussed all concentrate on analysing the
information, or the individuals who give you the information, to ensure that it
is accurate. You should also remember that gathering information is a process in
which the researcher is intimately involved: you should therefore cross-check
your own responses and opinions as well as those of the informants. There are
two ways to do this.
The first way is, before finalising your analysis of the HEA, to send the
informants a draft and ask them if they are happy with it. Many informants,
particularly those contacted early on, will not have had an opportunity to
challenge your final assumptions.
113
● The Household Economy Approach
The second way to prevent your preconceptions from clouding the analysis is
to make sure that all research is done by groups of two or three persons. This
allows the interviewers to cross-check each other. Groups of researchers should
include individuals with different characteristics: try to mix people of different
sexes, ages, nationalities and academic or professional training.
3. Avoiding pitfalls
A major advantage of the field approach is that, since you have been crosschecking the information as you go along, most misunderstandings will come to
light during the interview.
Some other common causes of misunderstanding are outlined here. You are
bound to find many more. In the experience of Save the Children, the most
common causes of inconsistency are mistranslation and/or the interviewer and
interviewee talking at cross-purposes. The question may be ambiguously worded,
or may not have been agreed between interviewer and informant.
If you are using a translator, check with them that they understood the
question you asked.
If the question was properly translated, tell the informant that you do not
understand the answer. Explain why this is so; if the answer was inconsistent with
earlier information, or with the understanding of an issue you acquired from
other key informants, say so. In most cases, the informant will be able to explain
the inconsistency immediately. If not, it is worth checking that you are both
talking about the same subject.
The informant has moved beyond the initial boundaries of the conversation.
It is worth confirming that they are still using the same categorisations as were
agreed at the outset. Check once again that the informant is talking about:
●
a normal year
●
the agreed household type and size
●
the agreed wealth group.
114
Checking the accuracy of the information ●
An example from Turkana, Kenya
When talking about livestock holdings, many informants gave very high
figures. This was because, while other assets could be owned by the small
female-headed household unit (which was the unit under discussion) cattle
are owned by the larger male-headed unit, which comprises several femaleheaded households. It made no sense for informants to discuss the herd
sizes of the smaller unit, because the smaller unit does not have a herd.
Informants automatically switched to talking about the male-headed
household, and assumed that the interviewer recognised that they had
done so. This was not obvious to the interviewer, until a discrepancy was
noted between the number of animals held (high) and the importance of
livestock products in the diet (low). Once this was pointed out to
informants, they were able to explain that when explaining milk
consumption, they were talking about a woman and children, when
explaining herd size, they were talking about something different: animals
owned by man, wives and children. The misunderstanding had arisen
because interviewer and informants were talking about different types of
household.
Even when you choose your key informants carefully and cross-check their
responses, there will still be occasions when the information they give just does
not add up. This can occur for several reasons:
●
The informant is wrong.
●
The informant is right, but is not referring to the same thing as you or the
question is ambiguous. “How many cattle do you own?” may not yield the
same answer as “how many cattle do you manage?”. In some locations a
person may hold animals for others on a variety of terms, and only “own”
some.
●
There is confusion over units of measurement: for example, the informant
may substitute a local unit for the acres or hectares used by the interviewer or
115
● The Household Economy Approach
A food economist in Gao, Mali, was surprised at the very high yields of rice
being achieved from irrigation schemes, and more surprised that households
couldn’t account for this rice: the total of what was consumed, sold and given
away by a model household seemed to be lower than the total grown. After
some questioning, he discovered that informants were all talking about paddy
production: unhusked rice. A bag of paddy can contain as little as 60% edible
rice. This explained high yields, as he had taken the paddy production to be
rice production. It also explained the apparent discrepancy between
production and consumption. The gap was, in fact, the difference between the
husked rice weight and the paddy weight.
●
●
●
116
use a different bag size. Often interviewers assume 90 kg bags for cereals, but
you must always to check this.
There is confusion over methods of storage. Food crops are often stored in
such a way that much of their bulk is not edible: rice can be stored as
unhusked paddy, maize can be stored on the cob. This will make the
quantities given by the key informant for a crop much greater than the
quantity that is edible, which is what you are interested in. Conversely,
cassava and other root crops are often stored as dried chips. In this case, the
energy value of the food by weight is much higher than you might expect if
you were working on figures for wet cassava.
There is confusion over the method of consumption. Be careful about asking
how many “bags” of a foodstuff are produced. Cereals and other crops are
often consumed “green”: picked and eaten the same day. As this part of the
crop is never bagged, informants will, quite logically, omit to mention it
when talking about how many bags they get.
In many places, crop production may be occurring several times in a year. Be
sure to check that the figures you are getting are for the year, not just for one
season in the year.
Checking the accuracy of the information ●
Conversations with cultivators in Mannar, Sri Lanka, were often complicated
by confusion between “food” and rice. Over the course of conversation,
farmers would begin to talk only about rice, under-reporting the importance
of other (normally purchased) foods. It was necessary to continually remind
informants that the conversation was about all foods. Rice plays a central
role in the society of northern Sri Lanka. Many social systems and
relationships are modelled on, or closely related to, the economic relations
which surround rice production. In terms of the space it takes up in people’s
minds, other foodstuffs don’t compete. Nevertheless, these other foods are
important to the energy intake of the population.
You will discover quite quickly that, wherever you are, food has symbolic
importance. When you talk about food in some cultures, certain things may be
excluded. Would a key informant in Britain mention beer as food, despite its
calorific importance to a large part of the population? Probably not. Yet the same
informant would probably see soup as food. So it is important to check that you
are getting a full list of all the important energy sources, and not just those with
cultural importance as food. Check also that there is agreement on the criteria by
which the importance of foods (especially in ranking and proportional piling) is
being assessed. If you attempt to rank by quality rather than quantity, the
question of individual perceptions needs to be borne in mind.
Similar problems can occur with livestock. Small stock and fowl may not be
mentioned because they are not perceived as being sufficiently important.
117
● The Household Economy Approach
Summary
Cross-checking takes five forms:
●
internal consistency within wealth group interviews: household economic
activities should agree with the economic possibilities; household budgets
should balance
●
internal consistency between wealth group interviews: when there are
economic relationships between wealth groups, these should agree
●
agreement between key informant information, written sources,
observations and other experience
●
agreement between key informants
●
balance between interviewers: if practical, work in a group; allow
interviewees to see analysis before this is finalised.
118
Contexts for Children’s Rights ●
8
Putting the information
to work
Information gained in the household economy approach (HEA) is typically used
to assess the likely impact of a shock, such as a reduction in food production
caused by drought, on the economy and food supply of a defined population (or
populations) of households. We are trying to understand how families are
making ends meet and how they are meeting their food and non-food needs
under changed conditions. This chapter describes how to carry out an analysis.
1. How complex?
The basic principles of HEA analysis are common to all situations, but the
complexity of the analysis will vary according to:
●
the number of food economy areas involved. Studying a large area
may involve the analysis of scores of food economies, and the relationship
between them. For example, if one area is affected by crop failure, this may
increase the demand for paid employment at some distant location and have
a knock-on effect even on areas that have not been affected by drought
●
the complexity of the shock. This may involve several variables: changes in
price, production, etc, sometimes over a period of several years
●
the objective of the analysis and the quality of output required. At one
extreme, HEA can be used as a framework for rapid assessment, in which case
119
● The Household Economy Approach
●
an approximation is sufficient. At the other extreme, it can be used as the
basis for the commitment of substantial resources, in which case a more
rigorous analysis will be required
the conditions in the area being assessed. In some settings, HEA may be used
to describe the economy of a population that is known to have been severely
affected by some event; here a judgement will have to made about the
benchmark against which abnormality is to be assessed. For example, in an
area such as southern Sudan that has long been disrupted by war, the baseline
itself may reflect an abnormal situation.
HEA analysis involves a series of simple calculations. The aim is not simply to
calculate a single result, but to work systematically through the likely effects of a
shock on the economy of households, and to assess the nature and magnitude of
possible outcomes. The processes actually at work are much too complex to be
represented by a simple numerical simulation. A household does not experience
the effects of a shock as a series of logically connected stages; people may be able
to anticipate certain events and take action before their effects are felt. Some
judgements must always be made.
In practice, the analysis of HEA data describing a single population is usually
done on a spreadsheet. This chapter therefore uses a simplified case to
demonstrate the basic principles of analysis.
2. Using food economy information
There are two main steps in using food economy information:
●
Reconciling the primary data. Converting the data obtained from a variety
of interviews, which will be in kilograms, litres, etc, into food energy units
(kcal) for each category of household for which information has been
collected.
●
Using the data to conduct an analysis. For example, in a drought-affected
area, to answer the question, “What effect is a drought likely to have on the
food supply of the households concerned?”
120
Putting the information to work ●
Note that, in practice:
●
Data reconciliation is usually carried out at least partly during the interview,
and the final reconciliation is done in the field.
●
When making a real-life analysis, you will have much more information
about the context than is given in the worked example.
●
Using a spreadsheet enables you to make multiple calculations quickly,
allowing the development of multiple scenarios using a range of values where
you are uncertain about the magnitude of some variables.
3. Converting the primary data to common
units
The data collected by an HEA analysis will typically be expressed in local units
– sacks, tins, etc – and will be converted to kilograms (kg), litres and other
standard units. The other unit employed will be cash, typically the currency of
the area concerned. To reconcile the household budget, food income has to be
further reduced to common food energy units: kcal.
The following hypothetical example relates to a normal year and a “poor”
household consisting of two adults and four children. The data will usually be in
the form of a balance sheet, as shown in the following tables.
In this case it is estimated that the household energy requirement is a
minimum of approximately 1,900 kcal per person per day. A typical household
in the “poor” category is made up of six people. Household energy requirement
for a full year is therefore (6 × 365 × 1,900) = 4,161,000 kcal.
Note that minor rounding errors in calculation can be ignored.
The currency used is shillings (Sh). The price of maize is 42 Sh per kg,
sorghum 38 Sh per kg and groundnuts 80 Sh per kg.
121
● The Household Economy Approach
Income and expenditure
Income from crops
Poor households grow maize, sorghum and groundnuts.
Table 6: Income from crops
Crop
Produced
Sold
Income
Consumed % Household
(Sh/year)
food needs
Maize
188 kg
62 kg (@ 42 Sh/kg)
2,604 Sh/yr
126 kg
10.9%
Groundnuts
220 kg
189 kg (@ 80 Sh/kg)
15,120 Sh/yr
31 kg
2.5%
Sorghum
550 kg
0
0
550 kg
46.7%
Total cash income: 17,724 Sh/year
Total crops consumed: 60.1% of household requirement
As can be seen, 188 kg of maize is produced; 62 kg are sold (at 4,000 Sh per
sack), leaving 188 – 62 = 126 kg for household consumption.
Maize provides 3,630 kcal per kg; 126 kg of maize = 126 × 3,630 = 457,380
kcal.
The percentage of household energy requirement obtained from maize is
((457,380 × 100)/4,161,000) = 10.9 per cent.
The percentage of household food income from crops (sorghum plus
groundnuts plus maize) = 10.9 + 2.5 + 46.9 = 60.1 per cent. The energy
values used are for fresh groundnuts (3,320 kcal/kg) and sorghum flour
(3,530 kcal/kg).
Income from crop sales can be calculated from the same table: 62 kg of maize
is sold at 42 Sh per kg = 2,604 Sh per year; and 189 kg of groundnuts is sold at
80 Sh per kg = 15,120 Sh per year. This makes a total cash income of 17,724 Sh
per year.
Other income
The calculation here is straightforward: the number of days multiplied by the
daily rate of pay. The only complication is if some of the payment is in kind, as
in Table 7.
122
Putting the information to work ●
Table 7 : Paid employment
Type of work
Agricultural work
Wage
170 Sh/day
% paid in food,
Person days
% paid in cash
worked
50% food, 50% cash
65
or sorghum to
the same value
This can be treated as:
●
a cash payment of 170/2 = 85 Sh/day. For 65 days = 5,525 Sh/year
●
food income of 85/38 = 2.2 kg/sorghum/day. For 65 days = 145 kg sorghum,
which converts to ((145 × 3,530 kcal/kg × 100)/4,161,000) = 12 per cent of
household energy requirement.
Table 8: Other production
Cash
Food
Firewood sales
40 bundles/yr @ 200 Sh/bundle = 8,000 Sh/yr
8,000 Sh/year
Eggs
30 @ 10 Sh/egg
300 Sh/year
Wild food sale
40 kg/year @ 20 Sh/kg
800 Sh/year
Wild food consumed
50 kg/year (estimated at 1,100 kcal/kg)
Totals
1.3% of household needs
9,100
1.3%
Total income from other production = 9,100 Sh per year
Income totals
Total household income (from Tables 6, 7 and 8) is therefore:
●
as food: (from food crops ) 60.1 per cent + (paid in kind) 12.0 per cent
+ (wild foods) 1.3 per cent = 73.4 per cent of annual household consumption
needs
●
as cash: (from crop sales) 17,724 Sh + (work paid in cash) 5,525 Sh + (sale of
firewood, eggs, wild foods) 9,100 Sh = 32,349 Sh/year.
123
● The Household Economy Approach
Expenditure
Table 9: Expenditure
Food purchase
300 kg sorghum @ 38 Sh/kg
11,400
Taxes
@ 400/adult/year
800 Sh/yr
Education
2 children in primary school @ 300 each
2 uniforms @ 800 each
2,200 Sh/yr
Medical
Typical estimated expenditure
1,100 Sh/yr
Soap
52 pieces/year @ 25 each
1,300 Sh/yr
Salt
8 kg/year @ 120/kg
960 Sh/yr
Fuel
12 litres/year @ 200 Sh/litre
2,400 Sh/yr
Clothes
3,000 Sh/year
3,000 Sh/yr
Handhoe
1 hoe/year @ 750 each
750 Sh/yr
Grinding
100 kg/year @ 20 kg
2,000 Sh/yr
Total expenditure
25,910 Sh/yr
Food purchase (300 kg sorghum) is equivalent to 300 × 3,530 kcal = 1,059,000
kcal, or 25 per cent of household food needs. The final balance is therefore:
●
Total estimated household food income = 73.4 per cent (food crops, payment
in kind, wild foods) + 25 per cent (purchased) = 98.4 per cent of household
consumption needs.
●
Total estimated household income as cash = 32,349 Sh per year.
●
Total estimated minimum annual expenditure = 25,910 Sh per year.
124
Putting the information to work ●
The difference of 6,439 Sh between income and expenditure might be spent on
additional consumption, eg, meat or other higher value foods, or saved or
invested.
In this hypothetical case, the balance is close. In practice, it will be found that
the calculation requires some judgement and approximation. Both income and
expenditure will vary within a single group of people representing a wealth group
(such as in a village) or within the same household wealth categories in different
locations in the same food economy area. For example, you may find that in one
place the “poor” group obtains 20 per cent of its food needs from its own crop
production, whereas in another place it obtains 30 per cent. How this situation
is managed is discussed later in this chapter.
4. Developing a scenario
The basic steps in developing a scenario are to:
●
define the problem that has arisen and the question to be answered
●
combine the problem with the household information to establish: the effect
we think the problem will have on current household income; the likely
ability of the household to make up that deficit; the costs involved to the
household in doing so; and how this varies between households in different
wealth categories.
Define the problem
Before we can begin an analysis, we must define the problem, which will be a
change (or changes) in the economic context in which the household operates
and on which it depends for its normal income. These changes may affect any of
the “normal” household sources of income, ie:
●
household production: food crops, cash crops, milk and meat
●
food aid or other gifts in an area that depends on them for its normal income.
125
● The Household Economy Approach
The changes may affect the wider economic context that the household depends
on for exchange, eg, a change in the price of a commodity produced or
consumed by the household. Or the problem may be a combination of the
above. For example, production may have failed for only one crop out of several;
it may be more difficult for people to find work; or the price of staple foods is
expected to increase.
The reasons for these changes will vary. For example, reduced access to
markets may be caused by war, the collapse of a bridge or the unavailability of
diesel fuel. A change in price may result from some local circumstance, such as
reduced local production of cereals, or from some distant event such as a fall in
the international price for commodities such as coffee or cotton.
Defining a change in context may require you to collect additional
information.
Combine the problem with the household description
This hypothetical and simplified example is intended to illustrate the basic
principles of an analysis.
In a defined food economy area, sorghum production has fallen by about 40
per cent because of armyworm attack. The year is otherwise a normal one. What
is the likely effect of the fall in production on the economy of the area, and how
will this vary between households in different wealth categories?
The economy
A field survey has been done and the data reconciled. Four household wealth
categories have been identified: very poor, poor, middle and rich households,
making up around 20, 30, 40 and 10 per cent of the population respectively.
Note that in this case the “middle” group is also the modal group (the most
common wealth category).
Household size for each of the groups is six people, and the normal level of
food consumption is estimated to be 1,900 kcal per person at a minimum.
The detailed enquiry is based on three of these groups: the very poor, middle
and rich.
126
Putting the information to work ●
Percentage of population
40
35
30
25
20
15
10
5
0
Figure 17:
% in wealth group
Very
Poor
Poor Middle
Rich
Wealth groups
The data are reconciled and it is found that the food consumed by each type of
household in the area is normally obtained from the sources shown in Table 10.
Non-food income and expenditure are shown in Tables 11 and 12.
Table 10: Food consumed by each type of household
Food crops
% of annual household food consumption
Very poor
Middle
Rich
Sorghum
35
50
40
Maize
10
10
20
Milk/meat
5
10
15
Wild foods
5
0
0
Gift
5
0
0
Purchase
Total
40
30
25
100
100
100
127
● The Household Economy Approach
Table 11: Sources of non-food income
Very poor
Paid employment
Livestock sales
Wild food sales
Rich
Sh
%
Sh
%
15,000
81
28,000
67
0
0
3,000
16
10,000
24
25,000
39
450
3
0
0
0
0
0
0
3,800
9
39,000
61
18,450
100
41,800
100
64,000
100
Sorghum sales
Total
Middle
Sh
%
Sh = shillings
Note that in this example, income is used in terms of the money value of income
from each source.
Table 12: Estimated household expenditure (in shillings)
Very poor
Middle
Rich
Food
8,500
6,400
2,800
Education
2,000
2,000
3,500
Soap
1,200
1,500
1,800
Fuel
0
1,000
3,000
Health
4,000
6,000
8,000
Clothes
1,450
2,000
2,500
Tools
450
450
900
Taxes
850
850
850
0
16,600
30,650
18,450
36,800
54,000
Other expenses/investment
Total
Other expenditure or investment might include livestock purchase, farm costs,
alcohol, tobacco, secondary school fees.
128
Putting the information to work ●
Table 13: Estimated capital and savings
% annual household food equivalents
Very poor
Food stocks
Livestock
Cash
Middle
Rich
0
0
60
20
100
600
0
40
100
The units used for capital and savings (Table 13) are “annual household food
equivalents”. Livestock holdings of 20 per cent would be the number of livestock
which, if sold at normal prices, would purchase 20 per cent of the basic food
needs of a household of six. For example, one goat at a price of 2,000 Sh, and a
sorghum price of 20 Sh per kg would be equivalent to approximately 20 per cent
of annual household energy requirement (100 kg of sorghum).
Markets
Paid employment is mostly agricultural work obtained locally, with the poorer
groups working for the better-off.
Food crops
Sorghum is the only crop sold. Maize is harvested earlier and is normally
consumed by the household. Richer households sell some of their surplus
production to poorer families.
Livestock is sold to traders, who then sell it in the larger district capital, either
for consumption or export to the capital city.
Wild foods
Poor people obtain a small proportion of their normal food supplies from this
source, but wild foods are not abundant: it is estimated that they might supply
10 per cent of a household’s requirements in time of severe need. Famine foods
are available, but can be toxic and are time-consuming to prepare.
Gift
There are some small transfers of food and cash, largely between kin. Rich people
traditionally help poorer people when they are in difficulty.
129
● The Household Economy Approach
5.The analysis
Set the context
We know that sorghum production is estimated to have fallen by 40 per cent in
the area. The immediate question is whether this is likely to have an effect on
prices. A change in the price of goods may indirectly affect people’s income. For
example, if they are dependent on buying food to meet their basic needs, a rise
in food prices could affect them adversely.
Sources of cash income
Sources of food
80,000
100
70,000
Gift
40
Wild foods
20
0
30,000
20,000
Maize
10,000
0
Rich
Household expenditure
50,000
Taxes
700
Tools
600
Clothes
500
Shillings
Health
Fuel
20,000
10,000
Middle
Rich
Very Poor Middle
Rich
Paid
employment
400
300
Soap
200
Cash
Education
100
Livestock
Food
Very Poor
Shillings
800
30,000
Livestock
sales
Savings and assets
Other
40,000
Wild food
sales
40,000
Milk/meat
Sorghum
Very Poor Middle
50,000
60,000
0
Sorghum
sales
60,000
Purchase
60
Shillings
Shillings
80
0
Food stocks
Very Poor Middle
Rich
Figure 18: Worked example. Sources of food and other income, expenditure, savings and
assets
130
Putting the information to work ●
Sorghum is exchanged locally, and there is no established system for traders
to market grain into the area from other sources. As sorghum sales would be
expected to fall with reduced production, it is anticipated that prices will rise.
Local informants suggest that as in previous years of this kind, sorghum prices in
local markets may rise by as much as 50 per cent.
The poor sorghum crop has reduced the amount of field labour available
locally. Wages have not fallen significantly, as they were already very low, but
work is already more difficult to get; for poor households, income from this
source is estimated to have fallen by about 30 per cent.
Livestock prices are not expected to change, as these are sold into a distant,
larger market. In previous years like this, livestock prices have not moved.
Grazing is unaffected.
In summary, the change in context is:
●
a 40 per cent fall in sorghum production
●
an expected 50% rise in sorghum prices
●
a 30 per cent fall in local paid employment.
Calculate the expected effect of this change in context on
household income
Households may be directly affected by the change in context in three ways:
●
The failure of the sorghum crop may lead to a fall in:
– the amount of sorghum normally used for household consumption
– the amount of sorghum for sale (reduce cash income)
●
Subsequently, after the harvest, sorghum prices will rise
●
Unemployment may cause a fall in cash income.
The effect of the reduced production on sorghum available for consumption (see
Table 14) is calculated as follows:
131
● The Household Economy Approach
Sorghum production has been reduced by 40 per cent. Sorghum is (see Table 10,
page 127):
●
35 per cent of the food consumption of poor households, so the fall in
income would be expected to be (35/100) × 40 = 14 per cent.
●
50 per cent of middle household food consumption so (50/100) × 40
= 20 per cent.
●
40 per cent of rich household food consumption so (40 /100) × 40 = 16 per
cent.
Table 14:The effect of a fall in sorghum production on normal household
consumption
% of normal annual food income
Very poor
Middle
Rich
sorghum production
14%
20%
16%
Total
14%
20%
16%
Remaining food income
86%
80%
84%
Reduction due to failure of
The effect on household cash income (see Table 15) of the reduced availability
of sorghum for sale is calculated as follows:
●
Poor households would be unaffected, as they do not usually sell sorghum.
●
Middle households usually make 3,800 Sh from sorghum sales (see Table 11,
page 128). They will lose (3,800/100) × 40 = 1,520 Sh.
●
Rich households will lose (39,000/100) × 40 = 15,600 Sh.
132
Putting the information to work ●
Table 15:The effect on cash income (in shillings)
Very poor
Middle
Rich
Reduction due to
loss of sorghum sales
0
1,520
15,600
Rise in sorghum price
2,290
0
0
Loss of employment
4,500
8,400
0
Total
6,790
9,920
15,600
Normal cash income
18,450
41,800
54,000
Remaining cash income
11,660
31,880
38,400
The effect of unemployment is calculated in the same way (Table 15):
●
A poor household usually makes 15,000 Sh a year from daily labour (see
Table 11, page 128). This is estimated to fall by about 30 per cent:
15,000/100 × 30 = 4,500 Sh.
●
Middle households usually make 28,000 Sh a year from paid employment
(see Table 11). They will lose 28,000/100 × 30 = 8,400 Sh.
●
Rich households will not be affected (in fact, they will make some savings, as
they will spend rather less on field labour. These can be calculated but are
omitted here).
In establishing the deficit we also have to take into account the expected increase
in the price of sorghum. A price rise will increase the cost of all food bought by
the household. The effect of this will not be immediate, as prices will not rise
until the year after the harvest (see Figure 19 overleaf ).
In practice, the extent to which a household is affected by price rises depends
on when they buy their food (see Chapter 1). Cereal prices do not usually change
abruptly: in a situation such as this example, prices might be expected to be
rather higher than normal after the harvest, and then to rise more steeply
thereafter (Figure 19). A household that had enough cash to buy all its food
needs immediately after the harvest would not face such high prices as one that
had to make its purchases piecemeal, as cash became available.
133
● The Household Economy Approach
Price
Reduced availability of field
work. Fall in cash income
for the poor / middle groups
Harvest
reduced by
40%. Fall in
crop income
Increased food prices:
increased need for cash
for normal food purchase
H A RV E S T
Figure 19:
Worked example. Factors
Time
in the calculation of the
initial deficit.
In the example, we will assume that middle and rich households have
sufficient cash savings to be able to buy most of their cereal needs immediately
after the harvest, when they can take advantage of normal prices; poor people,
who are assumed to have no cash, will have to purchase food at higher prices. If
prices are expected to rise by 50 per cent, then averaged over the year, a poor
household will face a price increase of about half of this. If the normal sorghum
price is 20 Sh per kg, this would be expected to increase to 25 Sh per kg.
We can factor this in by reducing the household cash income by this amount.
The poor group normally purchase 40 per cent of their food needs (see
Table 10, page 127), equivalent to about 458 kg sorghum. The cost of this at the
normal sorghum price of 20 Sh per kg is 9,160 Sh. At the increased price, the
cost will be 40 × 25 = 1,000 Sh, an increase of 2,290 Sh in household costs.
The middle and rich groups will be unaffected.
The effect of the sorghum crop failure, taking no other factors into account,
is estimated to be as shown in Figures 20 and 21.
134
Putting the information to work ●
Very poor households
Very poor households
Sorghum
21%
Purchase
40%
Gift
5%
Rise in sorghum price
12%
Remaining
income
64%
Crop
Maize
Milk/meat 10% deficit
14%
Wild foods 5%
5%
Loss of
employment
24%
Total normal income = 18,450
Middle households
Middle households
Gift
0%
Purchase
30%
Wild
foods
0%
Milk/meat
10%
Loss of
employment
20%
Remaining
income
76%
Sorghum
30%
Lost
sorghum
sales 4%
Crop
deficit
20%
Maize
10%
Total normal income = 41,800
Rich households
Rich households
Gift
0%
Purchase
25%
Wild
foods
0%
Milk/meat
15%
Sorghum
24%
Maize
20%
Remaining
income
71%
Crop
deficit
16%
Lost sorghum
sales 29%
Total normal income = 54,000
Figure 20:
Figure 21:
Worked example. Estimated effect of
Worked example. Estimated effect of
sorghum failure on household food
sorghum failure on household cash
income
income
135
● The Household Economy Approach
Calculate the likely ability of the household to compensate
for the income deficit
Households that suffer an income deficit because of a shock can often take steps
to make it up (Chapter 2). The next stage of the calculation is to see what each
category of household might be able to do to compensate.
In the example, poor and middle households, which have suffered an income
deficit, might be able to make this up by:
●
using their remaining cash income to buy food
●
increasing their consumption of wild foods. In this case, however, supplies of
wild foods are known to be limited. At most, this source might supply an
additional 5 per cent of food needs
●
selling assets to buy food. Poor and middle households have assets (not
including land, tools and seed) equivalent to about 40 per cent of their
normal household food needs: typically, a single animal. They could sell an
animal and use the money to buy food
●
finding additional paid work. There is no extra employment available locally.
It may be possible for people to find work further away, but few people have
prior experience of this. In the example, therefore, this option has been
excluded
●
receiving gifts. The richer households are not severely affected by the crop
failure. It is known that richer households help poor households when they
are in difficulty. Local estimates are that this might be sufficient to make up
about 10 per cent of a poor household’s food deficit, and for a middle
household perhaps 5 per cent
●
reducing their food or non-food consumption.
Before making the calculations, bear in mind that there are limits to what we can
achieve. Two issues need to be settled:
●
Different households will in fact respond to a deficit in different ways (see
page 34). For example, a poor household might choose to sell an animal, or
it might choose to collect some additional wild foods and minimise non-food
expenditure. We must recognise that we cannot predict exactly what people
will do. All we can say is what, given the resources available to them, they
could do.
136
Putting the information to work ●
●
A household may have enough resources to meet its food needs, but it might
only be able to do this at a large and (from the point of view of policy or
practical intervention) unacceptable cost to its current standard of living and
its future security. The household might survive, but only by forgoing other
basic expenditure or selling assets. For example, a household could survive
but only by not using health services, or reducing its soap consumption and
other basic household expenditure, or selling assets to the point where it
became destitute. Before making the calculation, it is necessary to set the
acceptable limit to household action. This may be important in deciding
whether intervention is needed and if so, what form it should take.
In this example, we will assume that an acceptable objective is that the
households in deficit should be able to meet their food needs and their current
levels of expenditure on basic needs. The minimum expenditure that allows for
education, health, soap, fuel, clothes, tools and taxes (Table 12) is about 10,000
shillings per household per year.
The next step is to look at the possible ways in which people could make up
the deficit. Only the poor and middle groups are discussed. The rich group
have a calculated food deficit of 14 per cent, but a remaining cash income of
38,400 Sh and substantial cash savings, and therefore would clearly be able to
make up their food deficit. When conducting a fuller analysis, however, you
should include rich households, as it is important to know how well they will be
placed to give resources to poorer groups.
The first row of Table 16 overleaf shows:
●
the estimated remaining food income of the poor (76 per cent) and middle
(80 per cent) groups
●
the remaining cash income, after allowance has been made for the cost of
basic non-food needs (10,000 Sh per household per year)
●
their estimated asset holdings.
The remainder of Table 16 shows the effect of each possible step which poor and
middle households might take to overcome their food deficit, and the effect of
this on asset holdings. Each household option is calculated separately: they are
not cumulative.
137
● The Household Economy Approach
Table 16: Options open to poor and middle households to make up a food deficit
Poor
Starting income
Middle
Food
Cash (Sh)
Assets
Food
Cash (Sh)
Assets
76%
1,660
Food stocks = 0
80%
21,880
Food stocks = 0
Livestock = 20
Livestock = 100
Cash = 0
Cash = 40
Option 1
Use remaining
1,660 Sh
cash income or
= 66 kg sorghum (@ 25 Sh/kg)
20% of food needs
= 240 kg sorghum @ 20 Sh/kg
cash savings to
= 6% of food needs
= 4,800 Sh
82% of food needs met
100% of food needs met
purchase food
Outcome
Remaining cash income = 0 Sh
Remaining cash income = 17,080 Sh
Assets unchanged
Assets unchanged
Wild foods = 5% of household needs
Wild foods = 5% of household needs
81% of food needs met
85% of food needs met
Remaining cash income
Remaining cash income = 21,880 Sh
(after basic needs met) = 1,660 Sh
Assets unchanged
Option 2
Additional wild
foods
Outcome
Assets unchanged
Option 3
Sale of I goat @ 2,000 Sh
Sale of 2 goats @ 2,000 Sh each
Animal sales
= 100 kg sorghum
= 200 kg sorghum
= 9% of household food needs
= 17% of household food needs
85% of food needs met
97% of food needs met
Remaining cash income = 1,660 Sh
Remaining cash income = 21,880 Sh
No livestock remain
Remaining livestock = 80%
Gift = 10% of household needs
Gift = 5% of household needs
86% of food needs met
85% of food needs met
Outcome
Option 4
Gifts
Outcome
138
Remaining cash income after
Remaining cash income after
basic needs met = 1,660 Sh
basic needs met = 21,880 Sh
Assets unchanged
Assets unchanged
Putting the information to work ●
Of course, these options are not exclusive. A household could choose to exploit
different combinations of options. In this example, a poor household might
choose either:
●
to rely on a combination of additional wild foods (5 per cent) and gifts (10
per cent), which would bring their food availability up to about 91 per cent
of normal. Additional food purchase would allow them to bring their food
consumption up to near normal levels (97 per cent). This would leave no cash
to meet other non-food costs
●
or to sell an animal. This would bring their food availability up to 85 per cent
of needs, and with wild foods and gifts this would meet their food needs as
well as leaving rather more cash for other expenditure.
We can therefore conclude that, in this example, the poor households would
have a difficult year but would survive, and would be able to maintain
approximately the same standard of living as before the crop failure. Some
households would get poorer.
139
● The Household Economy Approach
An example of an analysis from Sudan
Baseline, predicted, and actual food sources, poor and middle households,
pastoralists, Northern Kutum, Darfur, Sudan, 1997/1998. Predicted and
observed deficit in bold type. The predicted values are based on an analysis
done following drought, in October 1997, using the “normal” baseline data
shown. The actual values are those observed in April 1998. (Impact monitoring
report, Save the Children. A/Rahim Hussein Norien, Saeed Dunkos, Ibrahim
Sulieman, October 1999)
Food sources
% of annual household food requirement
Poor households
Predicted
Actual
Baseline
Predicted
Actual
Crop production
10
10
8
15
10
8
Migration
35
20
25
17
20
25
Wild foods
15
11
11
15
11
11
Purchase
35
22.5
20
47
35
30
Milk & meat
2
7.5
7.5
3
10
11
Food stocks
0
5
2.5
0
5
5
Gifts
0
5
5
0
0
0
Food aid
3
0
21
3
0
10
Deficit
0
19
0
0
9
0
100
100
100
100
100
100
Total
140
Middle households
Baseline
Putting the information to work ●
A more severe case: possible recommendations
A detailed consideration of intervention strategies is beyond the scope of this
manual. This case is introduced to show briefly how an analysis leads logically to
the identification of recommendations for possible action.
If the worked example is recalculated after, say, a more severe drought, we
might find that the outcome was as shown in Table 17.
Table 17: Remaining food and cash income for very poor, middle and rich households
Very poor
Middle
Rich
Remaining food income
70%
63%
69%
Remaining cash income
9,450 Sh
22,340 Sh
26,700 Sh
A poor household would now be unable to maintain its basic non-food
consumption (10,000 Sh/year) and to meet its minimum food needs.
The options available to households to compensate will also be different.
A problem of this severity is also likely to have an effect on livestock prices, and
on income from livestock sales in the following year. Assuming that cereal prices
increased by 50 per cent and livestock prices fell by 50 per cent, poor households
would have the following options:
●
To increase their consumption of wild foods. This would make up 5 per cent
of food income.
●
To rely on gifts. This would make up 5 per cent of food income; the betteroff households have also been severely affected and it would therefore be wise
to assume that the level of giving will fall.
●
To sell an animal. This would yield 1,000 Sh at the reduced livestock price
and only 40 kg of cereal at the increased cereal price; in terms of cereals a goat
would be worth about 3 per cent of food income.
●
To purchase food with their remaining cash income. This would cost
5,846 Sh.
In this case, the conclusion would be that a combination of these measures
would allow the very poor to meet their food needs, assuming they continued to
141
● The Household Economy Approach
eat at the normal level, but it would leave them only about 3,600 Sh to meet
their basic non-food needs, less than half that calculated as the basic minimum
required. It would also leave them without livestock. If a household chose not to
sell its livestock they would be left with only approximately 2,300 Sh.
Assuming that the aim is to allow the poor to meet their food and basic
non-food needs, and to retain their livestock there are broadly two lines of action
open:
●
reduce household costs. There are a variety of ways in which this may be
achieved. For example:
– Taxation and school fees for health and education could be suspended. In
this case this would effectively increase the income of poor people by about
6,000–7,000 Sh (Table 12).
– Other costs might be met by government or non-governmental
organisations: for example, by the distribution of seeds and tools, soap
or clothing. To a poor household this might be worth approximately
2,000 Sh.
– Market intervention to stabilise food prices would make food more
affordable to the poor. In this case, if sorghum prices were maintained at
normal levels (20 Sh/kg) this would be worth only an additional 1 per cent
of annual food needs. If livestock prices were also stabilised at normal levels
it would be worth an extra 6 per cent of household food needs.
●
increase household income:
– Direct food assistance might be given. This would make up their food
needs and leave them with just enough cash to meet most of their nonfood needs.
– Cash might be given to selected households.
– Additional employment might be generated, for example from public
works schemes.
The particular choice of interventions will depend on a variety of considerations,
eg:
●
the policy objective
●
the range of possible interventions which might achieve a particular outcome
●
the practicality of the types of intervention in a particular location
142
Putting the information to work ●
●
●
whether external assistance will be required (which may, particularly in the
case of food aid, entail delays)
the relative costs of different actions.
6. Calculations using estimates
In some cases the data is in the form of whole numbers, eg, household income
from food crops is 30 per cent. In practice, much of the information collected is
in the form of estimates, from either:
●
information collected in a single place where food crop income is 40–60 per
cent of normal food income
●
interviews from the same household wealth category conducted at different
locations in the same food economy. For example, an interview in location A
may reveal that food crops provide about 30–40 per cent of income for a
“poor” household, whereas in location B it is 35–45 per cent and in location
C it is 25–35 per cent. The best estimate is therefore that the true value for
the food economy area is between the lowest value, 25 per cent and the
highest value, 45 per cent.
For example, we may find that the sources of food for a wealth group are as
shown in Table 18.
Table 18: An example of household sources of food
Food source
% of household
food requirement
Food crops
50–60
Milk/meat
10–20
Wild foods
5–15
Gift
0–5
Purchased food
10–20
Total
75–120
143
● The Household Economy Approach
The difficulty here is that the upper and lower estimates do not add up to 100
per cent of normal household food consumption. Food crops supply somewhere
between 50 and 60 per cent and milk and meat 10 to 20 per cent, etc, but it
could be any value in that range.
Similar difficulties often arise with the information used to define a problem
or change in context. For example, after a drought, crop production in an area
may be estimated to be 30 to 40 per cent of normal production. Estimates of
rangeland production may involve even looser approximations.
Population data may also be approximate. For instance, you may be told that
the population of a district is somewhere between 80,000 and 100,000.
There are two ways of dealing with estimates in household budget data:
●
In a simple analysis, you can use an average value that seems to reasonably
represent the findings. Where there are multiple interviewers conducting a
study, this is often best done in a group, in the evening after the interviews
have been completed. It may also be done by calculation. In the example
shown in Table 18 the average of each range can be calculated. A weighted
average of each interval estimate must be calculated: for example, the average
for food crops in Table 18, would be calculated as follows: (((50/75) +
(60/120))/2) × 100 = 58.3
●
More complicated algorithms can be used which can logically reconcile
estimates. RiskMap (Annexe 1) combines interval estimates from the
household budget data and the problem specification to maximise the size of
the range of the output.
144
Putting the information to work ●
9
Presenting the results
Household economy approach (HEA) studies are usually done in order to
communicate something: for example that a drought is likely to have a specific
effect on the food supply or economy of a population, and to make suggestions
as to the possible actions which should be taken about this. There is never
certainty that a report will lead to action. Government or donors may not
respond for a variety of political, economic or organisational reasons. However,
unless the information about the situation is clearly communicated, in terms
which can be quickly understood by non-specialists (who often make the
decisions) and in terms which are clear and convincing, there is no possibility
that anything will be done. This chapter outlines the way to present a report.
The presentation of an HEA study is usually straightforward. The way in
which the analysis is done provides a logical structure for a report, beginning
with a description of the normal economic conditions in an area and the shock
or other changes which have occurred, and proceeding logically through the
effects which this is likely to have on households of different types.
Note that a report should be written for the convenience of the reader, not to
demonstrate the knowledge and skill of the author. You may feel that the
situation which you are describing is important: your reader may have several
such cases to deal with at any one time.
Include all the information necessary to an understanding of the problem
that you are trying to communicate. Avoid including extraneous information, no
matter how interesting.
145
● The Household Economy Approach
A report should be structured along the following lines:
A summary
This should be a summary of the whole document – background, main findings,
major points of uncertainty, main conclusions and recommendations.
The summary is an important part of the document – it will be all that the
majority of your audience will read. The summary is often left until the end of
report writing and then written badly, in haste. It is better to begin by outlining
the summary and using this to form the structure of the main report.*
An introduction
This should describe:
●
the purpose of the report (“The purpose of this report is to provide an update
of the food economy of Kakuma Refugee camp, which is situated in Turkana
region of north-western Kenya”)
●
the origin of the study (“This report, and the short food economy study it is
derived from was commissioned by the UNHCR and WFP...”)
●
what it will be used for (“...and will be used as background for the Joint Food
Assessment Mission of 1999”).
The dates of the study
A brief background should be given (“Save the Children previously conducted
food economy assessments in Kakuma in October 1996 and 1997...”).
Any relevant additional information (“Simultaneous to the food economy
assessment WFP/Vulnerability Analysis Mapping was studying the immediate
external environment of the camp and the camp’s effects and impact on the local
Turkana population. The site of the camp and the refugee’s relationship with the
host community is important in determining their income options and
possibilities for self-reliance. This and the fact that Turkana district is highly
vulnerable to food insecurity, makes it important to consider the livelihoods of
the population near to the camp, and their interactions with the refugees”).
* The example is drawn from Save the Children: Kenya Refugee Study Food Economy Updates of Kakuma Camps I, II, and
III. Conducted on behalf of World Food Programme Kenya and United Nations High Commissioner for Refugees Kenya,
September 1999, by Phillippa Coutts and Alexandra France.
146
Presenting the results ●
After a few paragraphs the reader should know what the report is intended to
communicate; what sort of information it is based on; what it is to be used for
(and whether they need to read further), and why the study is important.
A main background section
This should describe:
●
where the place or area is located (“Kakuma is located in north-western Kenya
100 km south of the Sudan border.”) Give a map
●
who lives there (“The camp was established to cater for Sudanese refugees, the
majority of whom at that time were Dinka fleeing from Bor County, Upper
Nile... since then it has been expanded to accommodate Somali refugees who
were transferred from camps which closed in Mombassa.”)
●
the size of the population.
Any other information necessary to fill out a picture of the place and its people.
This may be historical information, where for example there have been major
crises in the recent past; a brief description of other actors in the place
(governments, UN, non-governmental organisations).
If the situation has been changing rapidly over time – influxes of people,
conflict, drought – it is useful to give a timeline, so that the reader can get a quick
idea of how the current situation has arisen. For example:
January 1998: Kakuma II is opened. Cholera continued to be widespread.
February 1998: Wheat and corn-soy blend substituted for oil in the general
ration.
March etc.
Where parts of the background are complex, use subheadings. For example,
in some cases an understanding of the context may require much more detail on
the demographic structure of the population, the economic conditions or some
other aspect of the case.
The aim of a background section is to ensure that the reader has a clear, and
sufficiently complete picture of the location and the people living in it, to allow
them to set the subsequent sections of the report in context. Avoid including
extraneous information at this point.
147
● The Household Economy Approach
A methods section
State clearly and succinctly where you obtained the information used in the
report. This should include major documentary sources, the field work done,
who did it (“Fieldwork was done by two teams of two people each, the author
and a trainee”), and when. Outline the time spent doing interviews and mention
any special difficulties which arose.
A summary of the HEA methodology should be included as some readers will
be unfamiliar with this. Put this in a footnote or annexe.
The aim is to give sufficient information to allow the reader to judge for
themselves the likely quality of the information which is presented in the report.
The food economy
A description of the normal economy of the population (or populations)
concerned. This can follow the logic of the study:
●
the selection of food economies
●
the basis of wealth
●
the different wealth groups within each food economy, and the proportion of
the population falling into each group
●
the sources of food and non-food income and expenditure by wealth group
and household type
●
the exchange relationships of each group.
It is helpful to show income and expenditure in the form of pie charts.
Where data is reported as an interval (eg, food crops = 30–40% of food
supply for “poor” households”) ensure that the interval is shown.
Refer to any additional relevant information in the text (“A small proportion
of poor households also obtain honey from wild hives”).
Include only descriptive information. Do not use this section to discuss
findings.
The problem
If the study is concerned with a specific event or events, such as a drought and/or
change in price, this should be described under a separate heading. Give the
sources of information used (official reports, remote surveillance, key
informants) any relevant historical evidence (“On the last occasion when drought
148
Presenting the results ●
of this severity occurred, in 1996, it was estimated locally that this caused maize
yields to fall by approximately 20–30 per cent”).
The effect of the shock on the economy
As with the description of the normal economy, work through the problem in
steps (as described in Chapter 8). Pie charts are helpful.
A discussion
Use this section to make observations on the findings and analysis. Indicate any
areas where you are not confident in the analysis, point up anomalies if these
exist.
The aim is to complete the description for the reader and ensure that they are
aware of any potential holes in the argument. It is not the reader’s job to find
these for themselves.
Conclusions and, if required, recommendations
Conclusions should be drawn directly from the evidence presented. For example:
“Ninety-five per cent of households in Kakuma access around 90 per cent of
their food needs from the WFP ration. Around 40 per cent of the population,
the “poor”, derive the majority of their cash income by selling a quarter of
their cereal ration (whole maize and wheat flour). They spend the money on
small quantities of alternative foods, like milk and firewood or charcoal. The
amount of ration “poor” households sell depends on whether or not they
receive a ration of firewood. When wood is distributed they will sell less cereal
and consume a larger proportion of their ration.”
Recommendations should arise directly and logically from the conclusions. In
the example given, a recommendation might be to maintain, or increase, the
quantity of firewood distributed.
Keep in mind the need to present your case, and any recommendations, in a
practical way. For example, it may be logical to recommend the supply of
additional firewood, but there is no point in doing this if firewood is not
available. Think through the alternatives (eg, kerosene, other fuels). Do not make
recommendations which are beyond your technical competence. Additional
expertise may be required.
149
● The Household Economy Approach
Summary
The aim of a report is to communicate information: a report should
provide all the information required by the reader in as clear and accessible
a way as possible.
A report should include the following sections:
●
summary
●
introduction
●
background
●
methods
●
discussion
●
conclusions and recommendations.
Each section should include only information relevant to that section.Avoid
including extraneous information.
The summary is important: it is often the only section which will be read.
Do not make recommendations which go beyond your technical
competence.
150
Presenting the results ●
Annexe 1
Computer software
Where you have collected information on several areas or populations, or where
you would like to develop multiple scenarios, manual analysis can be tedious.
Computers allow the data to be quickly reconciled, minimise errors and allow
multiple scenarios to be quickly developed. Two approaches are commonly used:
spreadsheets and dedicated software.
Spreadsheets
Spreadsheet models are most useful for speeding up calculations on a single food
economy area, or a few areas.
Dedicated software
Dedicated software enables you to use more sophisticated modeling, specifically
of the interaction between populations where, as is often the case, they use the
same markets to exchange some commodities and therefore come into
competition.
The RiskMap software was originally designed for analysing very large data
sets, typically describing whole countries in terms of many areas; it is not suitable
for calculations involving single areas. RiskMap was originally developed in
Windows 3.1 but will run successfully in Windows 95 or 98 (difficulties have
been experienced with installing it in Windows NT). The RiskMap model does
not permit the detailed analysis of household budgets. The program is currently
being revised.
151
● The Household Economy Approach
Figure 22: RiskMap screen
A copy of RiskMap, including a sample database, is available, price £30, from
Malcolm Newdick, Riverbank IT management, Manor Cottage, Little Milton,
Oxford, OX44 7QB, UK.
([email protected]: www.riverbank.co.uk)
152
Annexe 1: Computer software ●
Annexe 2
Energy values of foods
This summary table of food energy values has been extracted from Tables of
Representative Values of Foods Commonly Used in Tropical Countries (Platt, 1985,
see Annexe 4).
The table is useful for evaluating dietary data based on records of group
consumption. The tables are not suitable for detailed surveys of the diets of
individuals.
Food energy values are given as the amount per 100 g of edible portion.
Kcal/100 g
CEREALS
Kcal/100 g
16. Oats, dehusked
388
1. Barley, whole, dehusked
339
17. Rice, lightly milled and parboiled
354
2. Barley, pearled
351
18. Rice, highly milled, polished
352
3. Buckwheat flour, 90% extraction
348
19. Rye, 85–90% extraction
350
4. Buckwheat flour, 60% extraction
349
20. Sorghum, whole grain
355
5. Maize, whole
363
21. Sorghum flour
353
6. Maize meal, about 96% extraction
362
22.Teff, whole grains
345
7. Maize meal, refined, 60% extraction 354
23.Wheat, whole and parboiled
344
8. Maize starch (commercial),
24.Wheat flour, 85% extraction
346
25.Wheat flour, 70% extraction
350
corn-flour
9. Millet, bulrush, whole grains
352
363
10. Millet, bulrush, meal
365
STARCHY ROOTS,TUBERS
11. Millet, finger, whole grain
336
AND FRUITS
12. Millet, finger, meal
332
26. Arrowroot flour
340
13. Millet, haraka, dehusked
353
27. Breadfruit pulp
113
14. Millet, jajeo, dehusked
355
28. Cassava, fresh
153
15. Millet, various, dehusked
355
29. Cassava flour
342
16. Quinoa
345
30. Ensete
190
153
● The Household Economy Approach
Kcal/100 g
31. Plantain
128
61. Groundnut, fresh
332
75
62. Horse bean
342
33. Potato, sweet
114
63. Horse gram
338
34. Sago flour
352
64. Kidney bean
339
35.Taro
113
65. Lathyrus pea
293
36.Yam, fresh
104
66. Lentil
339
37.Yam flour
317
67. Lima bean
326
32. Potato, Irish
38.Yam bean tuber
41
OIL SEEDS AND NUTS
68. Locust bean
380
69. Mung Bean (black)
329
70. Manga Bean (green)
324
39. Almond
657
71. Pea
337
40. Brazil nut
688
72. Pigeon pea
328
41. Cashew nut
590
73. Scarlet runner bean
326
42. Coconut, kernel, mature, fresh
375
74. Soya bean seed
382
43. Coconut, kernel, immature
125
75. Soya bean “milk”
32
44. Coconut milk, ripe nut
14
45. Dika nut, kemel dried
697
77. Soya bean
363
46. Karkashi
615
78.Tepary bean
331
47. Niger
513
79.Velvet bean
351
48. Oil bean, whole seed
544
49. Pistachio nut
626
50. Pumpkin seeds, seed coat
76. Soya bean curd
76
VEGETABLES
80. Beans, eaten green in pod
34
removed
610
81. Beans and peas, fresh, shelled
51. Sesame seeds
592
82. Bean sprouts
28
83. Beetroot
45
524
84. Carrots
33
697
85. Cucumber
12
52. Sunflower seeds, seed coat
removed
53.Walnut
154
Kcal/100 g
104
86. Eggplant
22
GRAIN LEGUMES AND PRODUCTS
87. Gourd
28
54. Bambara groundnut
367
88. Leaves, high carotene, dark green,
55. Bonavist bean
351
eg, spinach, pigweed, sweet potato
56. Chickpea
368
57. Cowpea
340
89. Leaves, medium carotene, eg, chard,
58. Fenugreek
335
New Zealand spinach, purslane,
tops, kale, bledo, etc
59. Goa bean
404
cassava leaves, watercress, cress,
60. Groundnut, dry
579
squash, pumpkin, colza, etc
48
28
Annexe 2: Energy values of foods ●
Kcal/100 g
Kcal/100 g
90. Leaves, low carotene, pale green,
119. Melon, sweet
26
eg, cabbage, kohlrabi, Chinese
120. Melon, water
23
cabbage, etc
23
121. Palm fruits, Peach palm, Pejibay
52
122. Papaya
39
123. Pineapple
57
33
124. Plum
45
94. Onion and shallot
48
125. Pomegranate pulp
77
95. Palm cabbage shoot
34
126. Prickly pear, pulp and small seeds
56
127. Star apple
82
91. Leek
92. Maize, immature on cob
93. Okra
123
96. Peppers, sweet, green and red,
seeds removed
209
37
FATS AND OILS
97. Pumpkin, squash and vegetable
36
128. Butter
745
98. Radish
18
129. Fish liver oils
900
99.Tomato with skin
20
130. Ghee
828
34
131. Lard and other animal fats
891
marrow
100.Turnip and swede
FRUITS
101. Avocado pear
165
102. Banana
116
132. Margarine
765
133. Red palm oil
900
134.Vegetable oils
900
103. Cape gooseberry
48
INSECTS AND LARVAE
104. Cashew apple
56
135. Lake fly
105. Citrus, grapefruit, pommelo, etc
37
136. Larvae, dried caterpillars
372
106. Citrus, lemon and lime
36
137. Locusts, mature
134
107. Citrus, orange and tangerine
53
138.Termites, mature
148
sugar apple
93
BEVERAGES
109. Dates, dried
303
Beer, sorghum
35
110. Fig, fresh
49
Beer, European
35
111. Fig, dried
269
Palm wine (1/2–1 day fermentation)
17
289
108. Custard apple, soursop,
112. Grape
76
113. Grenadilla, flesh and seeds
92
FISH AND FISH PRODUCTS
114. Guava, flesh and seeds
58
(INCLUDING MOLLUSCS AND
115. Hog plum, spanish plum
95
CRUSTACEA)
116. Kanapy, flesh
74
Fish, freshwater, fillet
95
117. Mammy apple excluding seeds
49
Fish, sea, lean fillet
73
118. Mango
63
Fish, sea, fat fillet
166
155
● The Household Economy Approach
Kcal/100 g
Kcal/100 g
Cod, salt
125
Beef, canned, corn
227
Fish, dried
309
Eggs, hens and ducks
158
Crustaceans (lobster, crab,
Goat, carcass
145
prawns, etc)
94
Mutton, fat, whole carcass
412
Molluscs (oysters, mussels, clams, etc)
70
Mutton, moderately fat, whole
Sardines, canned in oil
309
carcass
249
Salmon, canned
170
Mutton, lean whole carcass
149
Snail, river, pond
82
Offal, heart
129
Turtle
79
Offal, kidney
127
Offal, liver
136
SYRUPS, SUGARS AND PRESERVES
Pork, fat, whole carcass
535
Honey
286
Pork, lean, whole carcass
371
Jam
260
Pork, salt, fat
781
Molasses (cane, medium)
276
Poultry: chicken, duck, turkey, etc
139
Sugar, crude brown
389
Sugar cane juice
Sugar, white
73
Rabbit
134
Veal, moderately fat
184
400
MILK AND MILK PRODUCTS
CONDIMENTS, SPICES, FUNGI,
Cheese from whole cow’s milk, hard
MISCELLANEOUS
Colanut
384
Cheese from skimmed cow’s
350
milk, soft
87
Maize and sorghum stems
58
Milk, cow, whole
64
Sugar cane stem
60
Milk, human
75
Fungi, mixed, fresh
11
Milk, buffalo
102
Fungi, mixed, dried
99
Milk, goat
Mushrooms, fresh
13
Milk, sheep
Chillies, hot, dried
291
Milk, cow, skimmed
Garlic
139
Milk, cow whole, condensed
Tamarind
304
Milk, cow, whole, condensed,
Mustard seed
544
sweetened
71
108
34
140
317
Milk, cow, whole powder
156
MEAT, MEAT PRODUCTS AND EGGS
(unmodified)
Bacon, fat, whole side
589
Milk, cow, skimmed, condensed,
Bacon, lean, whole side
362
sweetened
276
Beef, moderately fat, whole carcass
262
Milk, cow, skimmed powder
357
Beef, lean, whole carcass
202
500
Contexts for Children’s Rights ●
Annexe 3
Data collection forms
The following data collection form has been widely used by Save the Children
(UK). As sources of income and expenditure are different in different areas, the
form has to be adapted for use in each situation (eg, the types of crops and
expenditure). The example shown was used in northern Tanzania.
An electronic copy of this form (MS Word) is available from Save the
Children, 17, Grove Lane, Camberwell, London SE5 8RD, UK.
[email protected]
157
● The Household Economy Approach
Interview summary form
Zone
Division
Interviewer
Date
Village
Wealth group
Group 1:
Group 2:
Group 3:
Group 4:
Percentage of population
Household size
Animal holdings:
owned and borrowed
– cattle
– shoats
– poultry
Land cultivated
Crops grown
– to eat
– to sell
O:*
O:
O:
O:
*O: enter other categories here
Year ranking
Year
(beginning of eating green
➞ before next year’s harvest)
1999/2000
harvest of 1999
1998/1999
1997/1998
1996/1997
1995/1996
1994/1995
1993/1994
1992/1993
1991/1992
1990/1991
158
Ranking
(very poor/poor/normal/
good/very good)
Reasons for that ranking
Group 5:
Annexe 3: Data collection forms ●
Household kcal requirements (per year) =
Food
Cereals
Total
units
Sold
Other
use
Consumed
% of
kcal needs
Green maize
Maize
Sorghum
Millet
O:
O:
Total
Roots/
Tubers
Sweet potatoes
Cassava
O:
Total
Pulses
Beans
Groundnuts
Groundnuts
(green)
O:
Total
Livestock
Milk
Meat
Shoats
Poultry
Eggs
O:
O:
Total
Purchase
Rice
Sardines
Potatoes
Grain
Milk
Meat
Oil
Sugar
Total
Total carried forward
159
● The Household Economy Approach
Food
Total
units
Sold
Other
use
Consumed
Total brought forward
Labour
exchange
(types of
labour paid
in food)
FOOD EATEN AT SITE: (day/week * months/year * food/day)
FOOD TAKEN HOME TO FAMILY: (day/week * months/year * food/day)
O:
O:
O:
Total
Honey
GRAND TOTAL
Income
Number sold
Crop Sales
Maize
Sorghum
Bullrush millet
Finger millet
Sweet potato
Beans
Sunflower
Pigeon Peas
Groundnuts
O:
O:
Total
Animal
Product
Sales
Cattle
– animal
– manure
– skin
Shoats
– animal
– skin
Poultry
Milk
O:
Total
Total carried forward
160
Price
Income
% of
kcal needs
Annexe 3: Data collection forms ●
Number sold
Price
Income
Total brought forward
Agricultural
labour
(types paid
in cash)
O:
O:
O:
O:
O:
Total
Other
casual
labour
(paid in cash)
O:
O:
O:
O:
O:
Total
Brewing
Trade
Petty trade
Selling honey
O:
O:
GRAND TOTAL
161
● The Household Economy Approach
Household work schedule
For each economically active person, list: each type of work done, the schedule of work
(days/week * number of weeks/month), and the payment earned for that work.
Person 1
January
February
March
April
May
June
July
August
September
October
November
December
*Delete whichever does not apply
162
Person 2
Person 3
Person 4
Person 5
Annexe 3: Data collection forms ●
Seasonal calendar
For each crop, mark the planting, weeding, harvesting (eating green & dried) and selling periods.
Jan
Feb
Mar
Apr
May
June
July
Aug Sept
Oct
Nov
Dec
Crop 1:
Crop 2:
Crop 3:
Crop 4:
Crop 5:
Crop 6:
Expenditure
Amount
bought
(per year)
Food
Price
Expenditure
Grains
Milk
Potatoes
Vegetables
Oil
Sugar
Salt
Rice
Sardines
Fish
Meat
O:
O:
Total
Total carried forward
163
● The Household Economy Approach
Amount
bought
(per year)
Total brought forward
Social
services
Medical
(health centre)
Medical
(traditional)
Education
Water
O:
Total
Taxes
Annual tax
Livestock tax
Sales tax
(livestock,
crops)
Development
levy
Bike tax
Total
Household
items
Fuel
Cooking
Blankets
O:
O:
Total
Clothes
Employing others
Loan repayment
Agricultural inputs
Animal costs
Renting land
Honey for brewing
GRAND TOTAL
164
Price
Expenditure
Contexts for Children’s Rights ●
Annexe 4
Other sources of information
The internet
There are now many websites providing information on food security including
sites relating to institutions in developing countries and regional organisations.
The following short list of organisations provides an entry point to sources of
information which can be downloaded and links to other sites. It is always worth
searching for relevant country and regional sites.
Food and Agriculture Organisation of the United Nations: www.fao.org
Institute of Development Studies: www.ids.ac.uk/ids/
International Food Policy Institute: www.ifpri.org
International Institute for Environment and Development, London:
www.iied.org
Overseas Development Institute: www.oneworld.org/odi
USAID Famine Early Warning System: www.fews.org
World Bank: www.worldbank.org
World Food Programme of the United Nations: www.wfp.org
165
● The Household Economy Approach
Documents
There is a vast literature on food security and related topics. The following short
list of sources develop some of the themes introduced in this manual.
M Buchanan-Smith and S Davies, Famine Early Warning and Response: The
missing link, Intermediate Technology Publications, 1995.
R Chambers, Rural Appraisal: Rapid, relaxed and participatory, Discussion Paper
No. 311, Institute of Development Studies, University of Sussex, 1992.
R Chambers and BP Ghildyal, ‘Shortcut methods of gathering social
information for rural development projects’, in MM Cernea (ed), Putting People
First: Sociological variables in rural development, published for the World Bank by
Oxford University Press, 1985, pp 399–415.
R Chambers, R Longhurst and A Pacey, Seasonal Dimensions to Rural Poverty,
Frances Pinter, 1981.
P Cutler, ‘Famine Forecasting: Prices and peasant behaviour in Northern
Ethiopia’, Disasters 8(1), 1984.
S Davies, ‘Are coping strategies a cop out?’, in IDS Bulletin 24(4), pp 60–71.
S Devereux, Famine in the 20th Century, IDS Working Paper 105, 2000.
A De Waal, Famine that Kills, Clarendon Press, 1989.
BS Platt, Tables of Representative Values of Foods Commonly Used in Tropical
Countries, First published 1945, Reprinted 1985 by the London School of
Hygiene and Tropical Medicine (Department of Human Nutrition).
Proceedings of the 1985 International Conference on Rapid Rural Appraisal, Faculty
of Agriculture, Khon Kaen University, Khon Kaen 40002, Thailand, 1985.
A Sen, Poverty and Famines: An essay in entitlement and deprivation, Clarendon
Press, 1981.
World Food Programme/UN High Commissioner for Refugees, Guidelines for
Estimating Food and Nutritional Needs in Emergencies, December 1997.
166
Save the Children Development Manuals
Helping Children in Difficult
Circumstances
A teacher’s manual
A step-by-step guide on how to support
children who have been affected by
violence, with a special emphasis on the
role of teachers.
£2.95
ISBN 1 87032 242 8
Development Manual No. 1
1991
Communicating with Children
Helping children in distress
This manual aims to help those
working in conflict situations and
emergencies to develop their listening
and communication skills in order to
identify and help children with special
needs.
£5.95
ISBN 1 84187 026 9
Development Manual No. 2
Second edition 2000
Family Tracing
A good practice guide
This book discusses how to trace and
reunite families. It highlights the
importance of assessing the best
interests of the child, and of working
with families and communities.
£4.95
ISBN 1 87032 277 0
Development Manual No. 3
1994
Street and Working Children
A guide to planning
Ideas and examples for those who want
to start or improve a project working
with homeless and working children.
£7.95
ISBN 1 84187 032 3
Development Manual No. 4
Second edition 2000
Toolkits
A practical guide to assessment,
monitoring, review and evaluation
A unique and highly practical guide to
help development workers evaluate and
monitor their work in a systematic way,
in order to improve its effectiveness and
quality, and identify strengths and
weaknesses.
£6.95
ISBN 1 87032 293 2
Development Manual No. 5
1995
The Household Economy
Approach
A resource manual for
practitioners
This manual offers an introduction to
the Household Economy Approach,
Save the Children UK’s methodology
for analysing the impact of crop failure
and other shocks on household income
and access to food.
£9.95
ISBN 1 84187 029 3
Development Manual No. 6
2000