Journal cf Social Development in Africa (19S9) 4, 2, 7-25 
Food, Aid and Relief Development Strategies • 
ROGER W HAY+ 
ABSTRACT 
This paper explores the usefulness and feasibility of using emergency aid and food aid to 
stimulate development processes during periods of famine in Africa. The proposition that 
relief aid can be utilised in creative ways that directly benefit individual households by 
increasing the value of their assets is made in a forceful way. Diversified sources of rural 
income resulting from this strategy reinforce the capacity of the most vulnerable households 
to protect themselves against the effects of future drought. 
This may seem a somewhat ambitious objective. However, even under considerable 
personal stress people have a tremendous will to survive and to invest in their future if given 
the chance. If this is true, then the bulk of emergency food aid supplied to African countries 
is used inefficiently because it does not offer families this possibility. 
Introduction 
The 1980's have tragically demonstrated the vulnerability of African families to famine. 
Indeed they may be more vulnerable now than they were 10 years ago, despite massive 
development efforts. The real tragedy is that Africa's food crisis was predicted and yet it was 
not prevented. Even after a massive injection of foreign emergency assistance, the conditions 
which generated it remain virtually unchanged. This being the case, the vast amount of 
emergency food aid that has been provided to African countries south of the Sahara is unlikely 
to have a significant beneficial effect on the long-term future of the families most at risk. 
Indeed it may well be that emergency food aid is keeping alive families that have little chance 
of an independent economic future. In this sense, emergency food aid may actually be 
increasing the future risk of famine. 
This paper's point of departure is the assertion that the primary cause of famine is not 
drought. It suggests instead that the vulnerability of African families to famine lies primarily 
in their susceptibility to income collapse. This occurs when income fluctuations cannot be 
* an edited form of a paper presented to the World Food Programme/African Development Bank Seminar 
on Food Aid for Development in Sub-Saharan Africa 8-11 Sept 1986, published as WFP Occas Paper No 8. 
+ Director, Food Studies Group, Queen Elizabeth House, University of Oxford International Development 
Centre, 21 St Giles, Oxford OX1 3LA, UK.
8 R W Hay 
stabilised by inter-household transfers or the liquidation of household assets, and the ability 
to maintain an adequate level of demand for food is lost! Employment and income are 
therefore central issues in this analysis. 
Furthermore, this paper argues that increasing agricultural labour productivity is not 
enough to provide adequate and stable levels of income for those families most at risk. The 
conclusion is therefore that international resource transfers must be directed at stimulating 
investment processes that lead to sustained rises in rural incomes, from non-agricultural as 
well as agricultural sources. 
This implies a broadly based 'relief-development strategy' (RDS) that discriminates 
between the risks to income collapse which different groups of households face and which 
make the most of the energy and vitality of African peasant families to invest in their survival. 
The paper argues that this strategy should dominate the allocation of international emergency 
resources and that the provision of emergency food aid is a vital part of such a strategy. 
The traditional 'relief, 'rehabilitation', 'development' sequenceis an inadequate framework 
for these objectives because, in practice, the bulk of resources are absorbed by the 'relief 
phase leaving little in the way of impetus or resources to do much more. The 'more' is essential 
if emergency aid to to have a beneficial impact on the future of African families. It should not 
only sustain and replace the stock of human and capital assets by stimulating investment 
processes likely to benefit those families most at risk, but actually expand it. 
compatible with the objective to sustain life. There are as follows: 
- Investment in economic assets at the household level. This will almostcertainly include 
investment in diversifying household economic activities as well as increasing and 
- Investment in both new economic and social assets at the community level, of the kind 
which will generate flows of benefits to the most vulnerable families. 
- Investment in the expansion of human capital in the form of new skills and in new 
More specifically, its use should be directed towards four main goals which are entirely 
stablising household level food production. 
possibilities for earning a living. 
These goals are both necessary and, within certain constraints, feasible. However, the 
viability of the strategy proposed here rests on two critical assumptions. The first is that 
African families are willing to invest in their future and are able to make sensible choices. The 
second is that communities have thecapacity to organise the implementation of projects which 
incorporate these objectives. What is required is a multitude of small-scale activities, of great 
range and diversity, spread throughout areas prone to drought The scale required is well 
beyond the capacity of any government to implement by conventional means. The only 
feasible alternative is to make a 'leap of faith', which many will find difficult, and to place the 
resources required in the hands of communities themselves. 
There are also a number of other issues which would have to be tackled if food aid could 
be used to support activities of this kind. These include the perceptions of food aid donors 
about the nature and causes of famine, which dictate the rules under which emergency food 
aid is distributed; the delays in having food aid and other resources on the ground in time to 
prevent widespread social and economic dislocation; and the willingness of government 
agencies to support community programmes which might utilise food aid to its best 
advantage.
Relief-Development Strategies 9 
The argument for relief-development strategies 
The argument for proposing a 'relief-development strategy' has four main strands. The first 
is that people who are prone to famine typically operate in an environment of uncertainty and 
the risks they face can be specified. The second is that, although this uncertainty shows up 
most obviously in unpredictable inter-seasonal and inter-annual fluctuations in agricultural 
output, it is actually fluctuations in household income which limit a household's ability to 
maintain a stable level of demand for food. If a downturn in income becomes irreversible, 
people die from starvation. The implication is that, while it is important to reduce the 
variations in household output, it is even more important to increase the level of household 
income and to reduce its instability. The third strand to the argument is that if the main goal 
is a reduction in the risk of famine, and if resources are scarce, then the expansion of household 
income, a reduction in its variability, or both, should absorb a substantial amount of time, 
effort and the resources available. The argument is completed by the assertion that this is a 
possible, indeed a necessary, use of emergency resources. 
All households operate in an environment of uncertainty. Rural households, and people 
in the cities who have no formal employment, face greater risks than most because they 
operate a household economy in an environment of great uncertainty from which they have 
little protection - other than the obligation of kinship, which should not be underestimated. 
However, two main sets of conditions lie behind most famines - low average levels of 
household income and large variations in household income. Neither inter-seasonal nor interannual 
fluctuations in income would matter very much if a family had access to stocks on 
which they could draw in order to maintain a level of current household income sufficient for 
their necessities, including food. 
Stock-income relationships 
Three elements can be seen to dominate household food security. The first is the average level 
of household income. The second is the magnitude and probability of seasonal and annual 
fluctuations around this average. The third is the value of stocks a household can maintain 
and the form in which they are held. 
These three factors are inter-related. Low incomes not only endanger current food 
consumption but also reduce the possibility of accumulating food security stocks. As income 
rises, both the amount and the proportion of income available for savings also rise. Current 
consumption is thus assured and the possibility of building up food security stocks also 
increases. The value of stocks required to achieve food security depends on both the 
probability of variation and its magnitude. If income is stable, and assuming it is adequate, 
there is very little need to hold stocks. As the probability and the magnitude of fluctuations 
increase, the value of adequate stocks must also increase. 
Low-income families therefore face a double jeopardy. First, low average household 
income, close to the' famine level', does not allow the accumulation of a surplus in good years. 
Therefore, families who are unable to accumulate much from current output are at more risk 
than those who produce enough in most years to more than fulfil their consumption
10 RW Hay 
requirements. Second, families whose income varies with large amplitudes from year to year 
in comparison with average income are more at risk than those whose income is relatively 
more stable. 
Many families in Africa have a low long-term average income and face large interseasonal 
and inter-annual income fluctuations. These families are the most at risk to famine. 
Off-farm employment and income 
Although the bulk of rural income in Africa is derived from farming, it would be a mistake 
to imagine that African families have no access to other sources of livelihood. Some family 
members may spend part of the year in wage labour, or petty trading and family-based 
industry such as beer brewing and food processing. In addition, many African families have 
a member employed in the city and depend to a large extent on this source of income to 
maintain food security. Furthermore, kinship ties frequently involve obligations to assist 
when times are hard. The importance of these intra- and inter-household transfers should not 
be underestimated. In Botswana, for example, one estimate2 suggests that the demand for 
food in rural areas increases to replace lost' food income' by 40 per cent during drought years. 
The main source of the 'new income' was remittances from family members working in the 
towns or, in the case of Botswana, in South Africa. However, comparatively speaking, rural 
African families are more dependent on farming income or income directly related to 
agriculture than some other geographical areas. Rural industries and services are poorly 
developed and offer little opportunity for employment. This increases the vulnerability of the 
rural economy as a whole to the effects of sharp declines in agricultural income and makes 
rural families more susceptible to irreversible income collapse. 
Secular declines in income and employment 
There is considerable evidence to suggest that rural incomes in Africa are declining and the 
number of people who are unable to earn an adequate income is increasing. Although there 
has been a good deal of publicity recently about declining levels of per capita food production 
in Sub-Saharan Africa, the evidence for this is rather slight Much less is said about secular 
declines in income and employment which is, by far, the more important issue. 
Neither rising nor falling aggregate food output is necessarily significant with respect to 
household food security. Many, if not most, families, and indeed most countries, do not 
produce all the food their people eat. What matters most is whether or not people earn enough 
money to buy the food they need to eat but do not produce. It is therefore most significant that 
such circumstantial evidence as exists tends to suggest that there has been a decl ine in the value 
of incomes and that opportunities foralternative off-farm employmentare contracting relative 
to population growth. If this is the case, it pinpoints precisely why the risk of fam ine in Africa 
may be increasing. Unfortunately, hard data is di fficult to come by. However, there are at least 
three pieces of evidence which support the impression that both income and employment 
opportunities are declining in Africa. 
The first is that population growth exceeds the capacity of any economic sector to provide 
work for new entrants to the labour force. If we assume that the potential labour force is
Relief-Development Strategies II 
increasing by three per cent per annum, that the elasticity of demand for labour is around 0.5 
for typical agricultural activities, farming and the like, and that (optimistically) there is equal 
access for everyone to new jobs, this would mean that a sustained increase in the value of 
agricultural output of six percent per annum would be required just to absorb new labour force 
entrants. This is well above the most optimistic estimates of the performance of agricultural 
sectors in countries south of the Sahara. 
The second piece of evidence comes from internal migration patterns. It is clear that many 
people in drought-prone areas of Africa do not see the future of their families to be in farming. 
The high premium placed on education to prepare children for wage and salaried employment 
suggests that people are prepared to take great risks and to live in appalling conditions around 
cities waiting for formal employment Urban growth rates in Africa in the last decade or so 
are the highest in the world. All this suggests that people are 'voting with their feet' against 
farming as a viable means of earning a living and that they see the risks of seeking alternative 
employment to be less than the risks they face farming. Of course, they may be wrong, but 
it is hard to find convincing arguments that they are. 
The third piece of evidence for declining rural incomes and employment opportunities 
comes from the relationship between land and the African rural population. The African 
countryside is commonly described as 'land abundant, labour abundant and capital scarce'. 
The last two features are undoubtedly correctly described. However, the first requires 
qualification and out of this qualification arises further evidence of secular declines in rural 
livelihoods. 
In the first place, land of good quality, capable of producing reliably from rainfall is not 
as abundant as it is often thought to be. A major study of land capability (UN, 1982) suggests 
that, although the combined natural resources of the 51 countries in Africa could feed 50 per 
cent more than their population projected for the year 2000, the population of 22 of them had 
already exceeded the potential carrying capacity of the land at low (and probably existing) 
levels of technology in 1975, and 28 would be in this position in the year 2000 if the level of 
technology were not raised. 
In the second place, even if land potential could be fully realised, the nature of African 
society means that people do not move freely. The assumption that they can and do is 
absolutely critical to a full utilisation of natural resource potential. The reality is more often 
that ethnic or other barriers confine people to over-populated areas where considerable inputs 
of labour and capital would be required to raise levels of land productivity to a point which 
would provide reasonable incomes for the people farming the land. There is an abundance 
of labour butagreatscarcityof capital. In its absence, the capacity of the landto yield iseroded 
and exhausted. These over-populated areas of Africa in many cases are also those most prone 
to drought and to famine because the incomes of families who farm the land are too low to 
allow the necessary accumulation of stocks with which to buffer income fluctuations. 
The magnitude of these problems has been masked by a series of good years enjoyed by 
many countries in Africa between 1977 and 1981. There is considerable evidence to suggest 
that the recent drought has resulted in a sharp increase in 'permanent' rural unemployment, 
in the worst affected areas and countries. Botswana and Northern Ethiopia come to mind as
12 RW Hay 
examples. Although every recorded drought has resulted in migration to local towns and cities 
by a group of people who have been put out of farming permanently, the scale of the problem 
now is unprecedented and it is not unreasonable to suggest that there is increasing 'structural' 
unemployment in many parts of rural Africa which has been exacerbated by the recent 
drought. 
For this reason it is not enough to assume that 'everything will go back to normal' when 
the rainy seasons have reestablished themselves. The 'crisis' (an employment crisis rather 
than a food crisis according to this analysis) will continue until households at risk find secure 
sources of employment and income. 
The conclusion which must be drawn from this analysis is that incomes and employment 
must be at the centre of a 'relief-development strategy' and that, while investment directed at 
increasing agricultural land and labour productivity remains important, for many of the most 
vulnerable families a secure livelihood must be sought outside agriculture. This implies a 
radical shift in relief (and development) priorities. A major effort is urgently required to 
diversify sources of rural income and create rural economies in which agriculture is linked 
with rural services and industries. 
The main elements of a relief-development strategy 
The general conclusions arising from this analysis suggest that a strategy that purports to 
address the causes and the effects of famine should attempt to achieve simultaneously a 
number of results. First, income is required urgently by families who face irreversible income 
collapse. Second, there is a need to pursue the following objectives: 
- a sustained increase in average household income 
- a decrease in inter-annual variations in household income 
- an increase in the value of household stocks. 
Therefore, emergency strategies which arc designed to have a permanent effect will 
combine a transfer of income to families short of food for immediate needs with support for 
one or more of the following four elements: 
- investment at the household level making itpossible for vulnerable families to sustain 
more stable food consumption patterns within and between years 
- investment in the provision of new and better skills to create greater opportunities for 
families to diversify their sources of income 
- investment in both social and economic assets, at the community level, of a kind which 
will generate Hows of benefits to the most vulnerable families. 
It is possible to use international resources to address immediate needs and also and at the 
same time to further long-run objectives. The traditional 'relief-rehabilitation- development' 
sequence is not an appropriate framework to construct and implement a strategy of this kind. 
Firstly, it is wasteful of resources as the bulk are allocated to the 'relief phase with little 
consideration for the underlying problems. Opportunities for preserving life while reducing 
the threat of future famine are thereby lost. Secondly, 'rehabilitation' generally means the 
replacement of lost household assets. The presumption is that the restoration of the ante status
Relief-Development Strategies 13 
quo is all that is required. This is not so. Thirdly, the 'relief-rehabilitation-development' 
strategy fails to utilise most effectively the single most abundant resource available during a 
famine, labour. Although the availability of surplus labour is both a consequence of famine 
and a reason for it, it is also an opportunity to be turned to the maximum use. The most graphic 
symbols of famine, relief camps, representing government and international impotence to 
prevent the worst of human suffering, also indicate the extent to which rural labour and skills 
become unemployed when household income collapses. 
Development objectives in an emergency context 
A joint relief-development approach 
The joint relief-development strategy proposed here focuses on creating opportunities for 
unemployed labour to work in productive ways which result in the types of investment 
required to enhance household food security. The resources for this investment must come 
largely from the people themselves. The strength of the strategy is that it achieves more than 
one objective at the same time. It utilises the resource which is most abundant, ie household 
labour; it addresses the most immediate effect of famine, which is an overwhelming increase 
in the number of unemployed people; it creates the possibility of transferring income to 
support failing household food demand by payments made for the work done; and it prevents 
nutritional deterioration, if put in place early enough. 
Moreover, if relief of this kind is provided in time and to people where they live, it should 
prevent the social and economic disintegration which accompanies mass migration for food, 
the dependence on emergency aid and the unacceptable conditions of relief camps, where all 
too frequently deterioration is so advanced that people are virtually unemployable. 
Although this strategy depends on payment for work, it differs in a number of significant 
respects from conventional public works schemes. In the first place, to be effective in the long 
run, the work done must create productive or social assets whose benefits are enjoyed by the 
most vulnerable families. An example may help to clarify this point. The income paid to a 
person who works to clear the land of a large farm may prevent his family from starving, but 
the benefits derived from the cleared land will flow to the owner of the land, who is not likely 
to be most at risk to famine. Although the income from public works schemes will meet a 
very real need if it makes a significant contribution to maintaining food demand, the 
better effect will be derived if income transfers can be made efficiently to households most 
at risk and if the work they do makes a lasting contribution to their own household food 
security 
This raises a number of serious questions about the choice of projects which fulfil the 
criteria set out here. It is true that Africa needs more roads, water sources, markets and forests, 
but it is more difficult to decide who will benefit first and most from many of these amenities. 
It is unlikely that the resources available to pay for the work done will be sufficient to achieve 
everything and, in order to meet the criteria appropriate to emergency-development projects, 
some hard choices may have to be made.
14 RW Hay 
A second way in which this strategy may differ from a conventional public works 
programme is that although some projects may be best organised at a community level, the 
criteria for maximum effectiveness in an emergency context admits to the possibility that 
people will be paid to improve their own assets - particularly land, but also personal skills - 
and that the most important contribution will be the mobilisation of community labour to get 
work done that a single family would be unable to do alone. 
A third distinction is that a major part of the investment should be designed to yield rapid 
returns. This is essential if the investment is to result in immediate protection from future 
income collapse. 
A fourth way in which this strategy differs from the usual public works scheme is that it 
should be designed to expand rural skills. We should not be under the impression that the only 
thing people in the African countryside know how to do is to farm. There is an enormous 
variety of skills available - masons, carpenters and artisans of every kind. Relief-development 
strategies should result in these skills being improved and transferred to others. If, sadly, relief 
camps are inevitable, they constitute an opportunity, which has been virtually ignored, for 
providing training (including improvements in literacy and numeracy) in skills which are 
economically and socially useful. This paper argues that a diversification of the rural 
economy is essential if the massive and increasing rural under-employment and unemployment 
that makes people vulnerable to famine is to be reduced. A relief-development strategy would 
be incomplete if it did not make a contribution to creating new, economically viable, 
opportunities for people to earn a living. 
Fifth, therefore, relief-development strategies should result in new, economically viable, 
opportunities to earn a living in the countryside. The flow of emergency resources into an area 
affected by famine constitutes an important opportunity to provide capital for setting up new 
small-scale industrial activities These are likely to be in the main related to agriculture and 
will supply wage goods and services for which there is reason to believe demand will increase 
rapidly as rural incomes rise. 
Finally, the strategy proposed here differs from the conventional one in that the choice of 
activities and the organisation of the work will need to be primarily in the hands of the affected 
communities. There are reasons for this. The first is that the scale of the problem is so great 
that it is beyond the capacity of any government or voluntary or international agency to 
organise andmanage. The secondis that if one takesabroad view of theprocesses which bring 
about change (and here we are speaking of the fundamental changes in the structure of the rural 
economy required to generate protection against famine) then changes which are' successful' 
generally arise from the initiative of the people facing the problem. It is true that they may 
need outside resources and outside ideas, but fundamentally solutions arise from within 
communities. 
The strategy being proposed here is only valid if local wisdom, initiative and organisational 
skills exist and only feasible if they can be mobilised. It is based on the belief that they do and 
can. Itproposestohandoverresourcestocommunities with simplebut minimum administrative 
control to use in the ways they think best.
Relief-Development Strategies 1$ 
Three caveats 
Emergency assistance which mobilises labour should not result in a diversion of labour from 
essential agricultural tasks. Projects should generally be sited within the communities where 
people live, particularly if the long-term benefits of the assets created by them are to be 
enjoyed by people at risk to famine. Moreover, the projects should either be completed before 
the season when people are fully occupied on their farms, or be organised so that they can be 
interrupted to allow people to work their land. If relief assistance results in new employment 
opportunities which are permanent, these should employ labour which is surplus to agricultural 
requirements, or people who are genuinely underemployed throughout the year. 
Second, this strategy does not exclude entirely the need for income transfers by food 
grants. It is probable that labour scarce households (with large numbers of children) will need 
income supplements which could take the form of food supplements3. 
Third, the strategy does not imply wholesale grants to local communities without control, 
nor that .no attention will be given to sound design, technical feasibility and economic 
viability, nor that 'outsiders' have no place at all, although their role may be very different 
from that of a 'city-based' advisor or many expatriate emergency teams. 
Priorities for relief-development activities 
The three technical objectives which dominate this discussion have already been specified: 
increased income, more stable income and increases in the value of the stock of household 
assets. It is, of course, not possible to specify every case for which activities are going to show 
the highest returns with respect to these objectives. However it is possible to suggest some 
guidelines. 
Three common situations will be encountered. The first arises from a collapse in 
agricultural incomes in an area where it is feasible to increase or stabilise agricultural incomes 
by investment in land improvement and technological change. In this case the clear priority 
is to mobilise labour for land clearing, terracing, drainage, the construction of small-scale 
irrigation systems and the like. 
The second arises from a collapse in agricultural incomes in an area where land is scarce 
and, despite improvements in land and labour productivity, it is unlikely that employment and 
incomes for the most vulnerable families can be secured from agriculture alone. The priority 
in this case is to identify families or individuals whose future can only be secured if they have 
access to alternative sources of income. While resources may usefully be employed in 
measures to improve land productivity, attention should be given to the provision of new skills 
and the start-up capital for new enterprises with strong links to agriculture. 
The third poses the most problems. Here agriculture does not provide a stable source of 
income and employment for the majority of the people. This may have arisen as a result of poor 
land endowments, land degradation or increasing population pressure on scarce land resources. 
Many economists would argue that migration is the only and best answer. Others would 
propose resettlementprogrammes. The first may not be theanswer because labourmovements 
are constrained by ethnic boundaries. The second has an extremely poor record of success.
16 RW Hay 
However a RDS must address the difficult issues involved. Retraining for non-agricultural 
employment may be possible in some cases and this may relieve the pressure on land resources 
sufficiently for sustained agricultural recovery to be initiated. However, in many cases the 
problems may be almost intractable. 
Emergency food aid and a relief-development strategy 
Emergency food aid is, theerfore, not the most important input required for a reliefdevelopment 
strategy, although it may have a place. A conventional relief strategy would be 
content to see that food transfers result in adequate food intake. A relief-development strategy 
has more ambitious objectives. It is concerned to see the extent to which income transfers 
permit households to invest. There is a growing evidence to suggest that they do - even out 
of income transfers provided for survival. A question of particular importance in the context 
of a 'relief-development' strategy is therefore whether cash or food transfers result in greater 
investment by recipient households. 
As one might imagine, the evidence is sparse. However, data collected from evaluations 
of a 'cash-for-food' project in Ethiopia, suggest that cash income supplements result in higher 
rates of nutritional recovery than do cash supplements. If this is true it is explicable on the 
grounds that families (or the senior male in the family) decide to forgo current consumption 
even if it places the families' children at nutritional risk, in order to secure the families' longterm 
future. 
Many people view the sales of emergency food aid with distaste - or even with alarm. This 
view tends to restrict the use of food aid as a general resource with broad applications and 
limits its support for the kinds of activities central to a RDS. There are situations where the 
sale of emergency food aid is most desirable. In some cases national food deficits may 
constitute a claim on food aid. In others, local shortages may need to be corrected, although 
in this case it is almost always better to sell imported food centrally and to provide an 
equivalent value of domestically produced food in local markets if possible. In other cases, 
food aid should not be supplied at all. 
A particular consideration in this regard is to decide the extent to which supply failure or 
demand failure predominates. Most often, at least in Africa, both occur, as so many rural 
peopleare farmers who produce for themselves and for the market- so crop failure affects both 
household income and market supply. It is also important to try to establish whether or not the 
restoration of food demand can be met by local supply, and, in the case of supply failure, 
whether this affects the national market or local markets. In the case of supply failure, 
emergency food aid may be a valuable way of providing temporary support to ensure that the 
demand for food is satisfied. 
Food aid has a more specific use in market exchanges when it is used to defend local terms 
of trade. The effect of drought on the rates at which products are exchanged in the countryside 
areprofoundlyimportantfor both producer andconsumer.Itis most importantthat the income 
of producers is protected, in order to maintain stable food consumption, and also to protect the 
stock of household assets. For example, a drought which reduces the carrying capacity of the 
range, results in unusually large numbers of animals being sold, a sharp decline in animal 
prices and, frequently, a sharp increase in grain prices due to scarcities in nearby producing 
areas. This means that pastoralists* terms of trade are affected adversely by not one but two 
influences.
Relief-Development Strategies n 
The provision of relief grain in exchange for animals at 'fair rates of exchange' achieves 
at least three objectives simultaneously4. It preserves the value of animals and thus the value 
of the pastoralists'income. Second, it accelerates the offtakeofanimals, so helping to preserve 
long run range carrying capacity. Third, if the animals can be slaughtered and the meat 
preserved, it transforms deteriorating food stocks into food stocks with an extended 'shelflife'. 
In the case of the Kenya scheme, the meat was sundried and used in school feeding 
programmes, but equally the meat could have been retained as community food security 
stocks. Thus, the decline in pastoral income was reduced and die reduction in the value of the 
stocks being drawn down was minimised. 
Some examples of activities of the 'relief-development' kind 
During the recent African 'food emergency' many people in national governments and 
international agencies were thinking along the lines set out in this paper. It is not possible to 
provide a complete inventory of recent experience. However, Table 2 offers a list of annotated 
examples to draw attention to interesting woik which has been done and which deserves 
further assessment and analysis. It is vitally important that this experience is not lost There 
can be nothing surer than the occurrence of continued periods of drought in Africa. Despite 
the lessons of the 1970's there was very little experience brought forward for application to 
the famines of the 1980's. This should not be allowed to happen again. 
A full inventory of 'emergency' activities with development objectives would make a 
fascinating contribution to collective knowledge and experience. Although the list of 
activities in Table 2 is incomplete, it demonstrates some interesting points. For example, it is 
almost certain mat not more than 10 per cent of the total emergency resources which have 
flowed into Africa in the last four years have been applied to activities designed to enhance 
household investment, to support declining rural-rural terms of trade, to enhance the value of 
household food security stocks, to increase the value of human skills, to increase the value of 
household assets, or even to increase the value of national investment This being the case, the 
list of activities offered here, incomplete as it may be, points to ways in which emergency 
assistance has been used and could be more effective if applied on a larger scale. 
It is possible to meet urgent human need and the underlying causes of famine at the same 
time, if transfers of income to households facing income collapse and starvation are made in 
ways which protect household assets, mobilise labour, enhance skills, create new employment 
possibilities and result in increased investment at the household and community level. 
We turn now to examine a set of strategic and capacity issues which will help to determine 
the allocation of resources in particular circumstances. 
Strategic and capacity issues 
There are four parameters which generally determine the broad shape of the appropriate 
'relief-development* strategy, viz supply or demand failure, or both; rates of 'permanent' 
rural unemployment; the condition of the people; and capacity issues. Supply failure in a pure 
form is exceedingly rare as a cause of famine and is only likely to occur if there is complete 
interruption to international trade. Demand failure generally predominates. The remedy is 
income transfers and, if national food production is sufficient to meet the enhanced demand 
for food, food aid is not indicated. The conventional view of famine is that its effects are
18 RW Hay 
temporary. This paper has suggested that this may not be the case and often is not Drought 
generally reveals an increasing level of permanent under- and unemployment in the African 
countryside. The allocation of resources is also influenced by the extent to which hunger has 
already resulted in economic and social dislocation and a deterioration in health and nutrition. 
The fourth parameter which determines a feasible strategy and therefore influences the 
allocation of emergency resources, is the complex of factors which come under the rubric of 
'capacity'. This includes transport, organisation, management, technical skills and the like. 
Many would argue that famine relief, even in its simplest form, overwhelms the local capacity 
and more cannot beexpected. It is true that famine relief imposes enormous additionalburdens 
on a national civil service and the technical agencies of a government. Imports by volume may 
double; transport patterns are turned upside down, inefficiencies abound; new agencies are 
formed and reformed; more people are hired and still everything goes wrong. 
However, there are three additional points to bear in mind. The first is that unless 
emergency resources are used more efficiently, Africa will become more and more vulnerable 
to famine on an increasing scale. The second is that 'famine relief has caused considerable 
fragmentation of government capacity. This is not unconnected with the international view, 
current a few years ago, that a 'super-ministerial agency' should be established to deal with 
famine. This might be a rational strategy when skills and resources are relatively abundant as 
it provides a centre of power which can allocate resources without competition. It is less 
rational when skills are scarce and when famine relief is viewed as an attack on long-term 
problems. In this case, the allocation of institutional responsibility becomes very divisive. The 
third point to be made here is that a 'capacity' which exists has been almost entirely ignored. 
It is the capacity of people and small communities to organise themselves in their own interest 
The fact is that country people are amazingly experienced in organising smallscale activities 
and managing their own affairs. 
The plain truth is that the magnitude of the problem is so enormous that there is no 
government or international agency which could possibly implement, in conventional terms, 
the variety and number of smallscale activities which are required to impose a lasting change 
on the face of rural Africa.This does not mean that all control is abandoned. An accounting 
is required. Technical advice will be sought if there is justifiable faith in the quality of the 
advice offered. Progress can be monitored and problems discussed. Joint evaluations - 'joint' 
meaning the community, the government and the donor - are valuable for all concerned, not 
least the donor. They provide a potential instrument for transferring experience from one 
community to another and for allowing communities to compare their experience and 
progress. 
RDS and 'disaster preparedness' 
An integrated 'relief-development strategy' places 'disaster preparedness' in a new light. 
While it is true that information systems that provide advance warning of crop failure and 
income collapse need to be developed (with particular attention to the 'demand side'),
Relief-Development Strategies 19 
information was not a limiting constraint to relief action in Africa during the 1980' s. However, 
it was a constraint to the implementation of activities which would have a developmental 
impact It is not difficult to identify groups of people who face greater than usual threats to 
survival from famine. 
The successful implementation of activities designed to expand employment potential in 
an emergency setting requires at least three major prior unconventional undertakings. First, 
contingency plans which identify potentially viable alternative employment opportunities for 
people in drought-prone areas should be costed and in place before famine causes major social 
and economic dislocation. These ideas cannot easily be grasped out of the air 'on the day', 
although many people living in threat of famine abound with ideas themselves. Second, 
therefore, extensive consultation is required with households likely to be affected by drought 
in order to establish their views of the future. Third, there needs to be a total integration of rural 
development strategies with 'disaster preparedness' as well as 'drought recovery' strategies. 
Famines result from low rural incomes and income instability. The objectives of rural 
development and 'relief-development' strategies, therefore, coincide. Attention to these three 
activities would transform the usefulness of 'disaster preparedness' and liberate additional 
resources for the prevention of future famine. 
The assessment of emergency programmes 
One way of drawing together and summarising the issues covered in this paper is to consider 
thecriteriaby which the 'success' of faminerelief mightbe judged. The criteria proposed here 
take into account the need to provide relief for human suffering and the need to contribute to 
household and community investment The criteria are listed in Table 3. 
Nutrition and health objectives 
The objective is clearly to maintain the people affected by famine in good health and nutrition 
or to restore them to this condition as quickly as possible. There are a number of data sets 
available which document the nutritional recovery of people receiving relief assistance in a 
variety of forms. It would be of great value if these records could be compared as this analysis 
might suggest how the costs of these programmes (administrative as well as material) 
compared with the rapidity of nutritional recovery. The relative efficacy of food and cash 
transfers needs to be clarified. • 
The prevention of income erosion 
The movement of rural terms of trade are measured rather simply by observing changes in the 
prices at which transactions occur in the countryside. Some data exist but their collection 
should be routine. Changes in the rates at which pastoralists exchange animals for grain are 
particularly important. It follows that the extent to which adverse price changes can be 
prevented is an important measure of the effectiveness of famine relief.
20 RW Hay 
The prevention cf'whole family' migration 
The search for work by adult family members may indicate stress on the family budget but 
may also indicate the effectiveness of household strategies to supplement income. The 
freedom with which this migration can occur and the availability of alternative employment 
are therefore important indicators of success. However, the migration of whole families in 
search of food is quite a different matter. Famine relief must seek to prevent this. A measure 
>f its effectiveness is the extent to which it has succeeded. 
Consumption objectives 
major objective is the maintenance of stable food consumption patterns. Income transfers, 
is we have seen, may be achieved as food grants, commodity wages or cash wages. The 
:ontribution that various elements of a relief programme make to the consumption of the 
families most in need, taking into account the different requirements and constraints facing 
'amities with high and low dependence ratios, is therefore an important indicator of the 
contribution famine relief has made to household income. Typically, the flows of new income 
ire complex and varied. School feeding programmes, food ration distribution, clinic ration 
istribution and wages from public works schemes may all make varied contributions to the 
ncome of different family groups. It is worth working out which families are deriving the 
naximum benefit and from which sources. 
Employment objectives 
liese fall into three main categories: temporary employment opportunities for those who will 
eturn to a previous occupation, new seasonal employment opportunities, and permanent 
ulltime employment opportunities. Until fairly recently there was little emphasis placed on 
Ixis important aspect of famine relief and, even now, the resources allocated are pitifully 
mall. Nevertheless there is some experience, eg in Ethiopia, Botswana and Mozambique, 
'hich deserves closer examination. It would be well worth investigating the extent to which 
articipants in public works schemes gained new skills later used to generate income, or 
ssessing more carefully the extent to which the recent drought resulted in permanent 
nemployment. 
tvestment objectives 
his paper has repeatedly emphasised the need for local investment to result from famine 
tfief. It has emphasised the need for investment in assets which will lead to higher and more 
able incomes for those most vulnerable to drought. Investment at the household level is 
ven pride of place as most likely to contribute to future household income. Remarkable as 
may be, the evidence from Ethiopia quoted here suggests that families in the most straitened 
rcumstances have a high propensity to invest in their future. Famine relief may also improve 
e value of household assets - either physical assets or human skills - through the work done, 
lird, investment may be achieved as a result of community work projects that improve land
Relief-Development Strategies 21 
or create new water sources, or the like, to which families most at risk have access. Investment 
of this kind might usefully be assessed according to whether it is likely to contribute to higher 
future incomes, more stable future incomes, a higher value of household food security stocks 
(physical or capital), or all three. 
At the second (community) level, investment may be a result of work done to improve the 
economic infrastructure or social assets. This reasoning suggests that an assessment of this 
investment should include an examination of the extent to which the benefits are likely to flow 
to families most at risk of income instability and collapse. 
At the third (national) level, stocks of national assets may be improved as a result of 
famine-related work projects. It is, therefore, important to examine the future benefits 
carefully before concluding that, in addition to contributing to current consumption, there has 
also been a contribution to future household food security. 
Conclusion 
This paper has argued for a radically different approach to famine relief by drawing attention 
to the fact that famine is essentially a problem of incomes and employment, and that famine 
relief should be concerned with short and long-run solutions to low and unstable incomes and 
declining job prospects. The strategy proposed here has a number of institutional implications, 
both for governments and international donors. 
First, no single relief input can stand alone. Food aid, by itself, is incomplete. There is 
need for much greater coordination of donor inputs. The most 'successful' famine relief 
programmes are coordinated by effective government control, and donors should support this 
development and not feel usurped. 
Second, the arbitrary divisions between relief and development assistance are unhelpful 
in practice, although useful as ways of soliciting additional resources. When it comes to the 
use of the resources, great damage is done if donor rules are not flexible. This line of reasoning 
makes a case for considerable liberalisation of the rules under which food and other 
emergency aid is administered. 
Third, the nature of the programming skills required for the administration of international 
assistance in support of a 'relief-development' strategy differ from convention. This is not 
to suggest that everyone involved needs to be a development economist, but they do need to 
understand the underlying processes which make people vulnerable to famine. 
Fourth, 'disaster preparedness', if the logic of this paper is sound, has less to do with 
creating new institutional capacity which will stand idle for seven years out of ten, and more 
to do with identifying particular groups of people who have been declared by history to be 
vulnerable to repeated famines and by concentrating effort on investment strategies which 
will increase and stabilise their incomes. This, and not the capacity of a central government 
agency to predict their next misfortune, will make them 'drought-proof. 
Finally, the eradication of famine on other continents has been one of the remarkable 
achievements of the last two decades. Both China and India stand as notable^xamples of the 
effort. Poverty still exists but the cataclysmic periods of hunger which left thousands dead 
every few years appear to be a part of history. Africa has much to learn from this experience 
and there is very little sense that this is recognised in either national or international fora.
22 RW Hay 
The task demands an alliance: an alliance of the will and resources of, first and foremost, 
the newte together with their governments, international donor agencies, and now a new 
S r n S t o e \ 5 p t e of industtialised countries who want to contribute to the solution of 
S S ; £ £ % » i > no longer room for insularity, irratior^adnunis^e arsons at 
CSSLOMI or international levels, or sectoral bias. The problem will be solved bya 
combination of good poUcy at international levels, a better understanding of thenatoeofthe 
problem, SUSIainW flows of resources (as opposed to flamboyant temporary exhibmons of 
g^rosi^andnK^ofallasenseofhopemongthepeoplemostatnsk which"ff***™ 
Mtiative, imagination and effort to work their way out of trouble knowing they have the 
support they need. 
FOOTNOTES 
1. Tins assenion is closely related to Sen's analysis. Indeed it draws from it u does most of what follows. See 
for example SenAK Poverty and Famine: An Essay on Entitlement and Deprivation, Oateadon Pre*s, 
Oxford, 1981. 
2. 'Income' means both cash income derived from sales and the value of commodities retained for consumption 
by die fanner's family. 
3. Recent experience in Botswana, where the government has mounted a most imaginative drought relief 
programme cxmbiningpubfccworksidiemes with jchool and dinicb«sediuppl^ 
has much to offer. A more thorough analysis of costs and benefits is required, but see HayRWetal "A 
Socaoeconomic Analysis of Drought Relief in Botswana", Biter-Ministerial Drought Coordinating Committee, 
Ministry of Finance and Development Planning, Government of Botswana, 1985. 
4. The alpha-factor proposed by Shkmo Reutlinger does exactly this. 
References 
EeleGE and Funk P (1985) Updating the Food Accounting Matrix for Botswana, 
Ministry of Finance and Development Planning, Government of Botswana and Food 
Studies Group, Oxford. 
HayRWetal (1985) A Socioeconomic Analysis of Drought Relief In Botswana, Inter- 
Ministerial Drought Coordinating Committee, Ministry of Finance and Development 
Planning, Government of Botswana. 
Kumar B Gopalakrishna (1985) The Ethiopian Famine and Relief Measures: An 
Analysis and Evaluation, mimeo (prepared for UNICEF, Addis Ababa), Balliol 
College, Oxford. 
McArpine MB (1985) Famine Relief Policy in India: Six Lessons for Africa, 
Working Paper No 2, Alan Shawn Feinstein Hunger Programme, Centre for the 
Comparative Study of Development, Brown University. 
MellorJohn (1986) "Agriculture on the Road to Industrialisation" in Lewis John P and 
Kallab Vakriana (eds) Development Strategies Reconsidered, US Third World Policy 
Perspectives No 5, Transaction Books for the Overseas Development Council, New 
Brunswick NJ.
Relief-Development Strategies \ 
Morris David (1975) "Needed - A New Famine Policy" in Economic and Political 
Weekly, Annual Number. 
Relief and Rehabilitation Committee (1985) Evaluation Report of Cash for Food 
Projects, RRC/UNICEF, Addis Ababa. 
SenAK (1981) Poverty and Famine: An Essay on Entitlement and Deprivation, 
Clarendon Press, Oxford. 
United Nations (1982) "Potential Population Supporting Capacities of Lands: Africa" in 
World Population Trends and Policies: 1981 Monitoring Report, Vol 1. 
Valdes Alberto (ed) (1981) Food Security for Developing Countries, Westview, Boulder. 
Table 1 
Integrated Emergency-Development Objectives 
Objectives Some Means 
Immediate 
Income transfers to households Feeding Programmes 
facing income collapse Cash grants^lash for work 
Temporary Employment 
Defence of rural terms of trade Animal-grain exchange 
Other support to peasant terms of trade 
Support for market supply if nee Food sales in local markets 
Food sales in national markets 
Long-term 
Investment in household economic Land improvement 
assets New storage facilities 
Training in new skills 
Start-up capital for new enterprises 
Investment in community New roads 
economic assets New markets 
New storage facilities 
New water sources 
Investment in community New health facilities 
social assets New schools
24 RW Hay 
Table 2 
SOME EXAMPLES OF RDS-TYPE ACTIVITIES 
Support for declining rural term* of trade 
Animal-grain exchange: Kenya, Mali 
Hen-grain exchange: Ethiopia 
Commodity-grain exchange: Mozambique 
Restoration of household assets and agricultural inputs 
Seed distribution programmes: Ethiopia, Kenya, many more 
Distribution of agricultural implements: Ethiopia, Mozambique 
(These constitute income transfers as grants. Examples of income transfer by cash grants are not known.) 
Income Transfers 
Feeding programmes of many varieties in many countries. 
(The effect of feeding programmes on household investment is not known, is worth investigating 
but is likely to be slight The effect on human capital likely to be considerable. The comparative costs 
and benefits of these schemes, which are based on a variety of principles, deserve a more thorough 
analysis in the African context) 
Cash for work projects 
'Cash for food' projects in Ethiopia 
(The projects undertaken include feeder roads, improved water supply, land terracing, smallscale 
irrigation.conitruction of buildings, brick production, school construction, reafforestation, fish pond 
construction, flood protection. 
The effect on household investment is likely to b e considerable: the effect on community level 
investment is also likely to be considerable, and the effects of these projects, particularly over the next 
few years, would be worm monitoring.) 
'Labour-based drought relief programme' in Botswana 
(Projects undertaken include land clearing, development of new water sources, road building, airstrip 
construction,protection of crop land from animals, small building projects, brickmaking, grain milling. 
The effect on household investment is not known but is likely to be slight; the effect on community 
and national level investment is likely to be considerable.) 
'Cash for work' element in Mozambique resettlement support project 
(details not known). 
Food for work projects 
Expanded land improvement projects in Ethiopia 
Programmes to expand skills 
Incorporated into some 'cash for food' activities in Ethiopia 
Projects designed to create permanent new employment opportunities 
None known, except that some community work projects in Botswana may result in participants starting 
their own small enterprises.
Relief-Development Strategies 25 
Table 3 
CRITERIA FOR ASSESSING FAMINE RELIEF 
Short-Term Objectives 
Contribution to nutritional welfare 
Contribution to health 
Prevention of internal migration 
Defence of rural-rural and rural-urban terms of trade 
Support to household food security strategies 
Contribution to household consumption 
Contribution to employment 
Long-term objectives 
Contribution to new employment opportunities 
Contribution to household investment 
Contribution to community investment 
Contribution to national investment