STAFF PAPER SERIES
SMALL FARM LIVELIHOOD SYSTEMS AND FOOD
SECURITY: ADDRESSING DIVERSITY
Peter E. Hildebrand and Amy J. Sullivan
Staff Paper SP 02-7 August 2002
UNIVERSITY OF (FLORIDA
Institute of Food and Agricultural Sciences
Food and Resource Economics Department Gainesville, Florida 32611
SMLL FARM LIVELIHOOD SYSTEMS AND FOOD SECURITY:
Peter E. Hildebrand and Amy J. Sullivan
Food and Resource Economics Department
University of Florida
Gainesville FL 32611-0240
Poor farmers or extractivists constitute a significant portion of the world's population. In order to help solve the continuing scourge of food insecurity in ever more fragile rural livelihood systems of the world, the tremendous diversity in which these smallholders produce food for themselves and others must be addressed directly. These people face a continuous struggle to meet their needs with available resources in the face of constantly changing conditions. They must devote their limited resources to the production of food and fiber for subsistence as well as commercial enterprises for cash generation. This balancing act is further complicated by fluctuating needs for cash and food, and constantly changing household composition. This paper describes how researchers and practitioners can model the diverse, dynamic, and complex livelihood systems of these limited resource farmers, and how ethnographic linear programming can simulate the livelihood strategies of diverse households within a given livelihood system.
Key words: ethnographic linear programming, participation, modeling.
LIVELIHOODS, FOOD SECURITY A.ND DIVERSITY IN LIMITED RESOURCE, LANDED HOUSEHOLDS Peter E. Hildebrand and Amy J. Sullivan Yet it is clear that within the limits of the next decade peasant
labor farms will, nevertheless remain an unalterable fact in
numerous countries, including the U.S.S.R. We who are concerned with the practice of agriculture must construct its future forms
from the existing forms of peasant farming; therefore, we are, in
practice, concerned with the deepest possible study of the
4 peasant farm. A.V. Chayanov, 1925
Most of those who are today seeking to understand the
economic behavior of the peasantry seem to be unaware that they
are traversing much the same ground trod from the 1860s onward by
several generations of Russian economists. D. Thorner, 1966
Understanding the system is important for identifying problems.
It is even more important for shaping solutions.
Michael Collinson, 1998
As we begin the third millennium, nearly a century and a half from the
time Russian economists began studying peasant farms, we still appear at a loss as to how these farms function, remaining on the edge of poverty and
struggling for food security, yet continuing to survive. The only
explanation is that we have not made a concerted effort to really try to
understand these farms- even though they have been and continue to be one of
the most important single sectors in most countries of the world, and,
perhaps, in the entire world. In 1936, Whittlesey (quoted in Mosher, 1969) estimated that "subsistence farms" supported about 60% of humankind (about
1.6 billion people). Wharton (1969) estimated (presumably in the mid 1960s) that subsistence farms covered some 40% of the cultivated land of the world
and supported 50-60% of humankind (1.7-2.0 billion people). While the number
of subsistence farms today is not well estimated, various estimates still place it between one arid two billion. According to the UN's International
Fund for Agricultural Development (IFAD), in the year 2000, 75% of the
world's 1.2 billion "extremely poor" are rural. This accounts for about 900 million people who are the poorest segment of limited resource rural households. If these "extremely poor" were half to two-thirds of the limited resource rural households in the world, this would mean between 1.35 and 1.8 billion people are surviving in limited resource rural households. No matter the estimate, these households are still a very important part of society and the food production systems of the world.
Chayanov and other Russian economists recognized that the "deepest possible study of the peasant farm" required a multidisciplinary approach. Starting about mid way through the 2 0th century we abandoned a multidisciplinary, whole farm perspective in favor of a highly specialized and reductionist scientific approach to agricultural development. Commercial crops were left to agronomists, animals to animal scientists, and forests to foresters, few of whom understood people, while costs and returns were-left to economists, and peasants to anthropologists, few of whom understood farming.
Over the last 15 to 20 years at the University of Florida a group of faculty and their graduate students from several multidisciplinary programs have been working together to modify the reductionist pattern. The impetus for these activities was based in the Farming Systems, Agroforestry, and Tropical Conservation and Development Programs. Participation has come from such diverse disciplines as Agricultural Economics, Agronomy, Animal Science, Anthropology, Forestry, Geography, Latin American Studies, Political Science, Soil Science and Wildlife. Studies have been conducted in many countries, mostly in Latin America, the Caribbean and Africa. While the specific nature of each of the studies has been as diverse as the interests of the students and faculty involved in them, because of the frequent interaction in country and on campus of those involved, we became more and more aware of emerging
commonalities. First, virtually all the studies involved both natural resources and the people who used them. We also began to realize that food security and sustainable livelihoods were the primary concerns of most of the people who were subjects of our research and for whom the "farm" was not just a production unit, but first and foremost a home.
In order to help solve the continuing scourge of food insecurity in
ever more fragile livelihood systems in the world, the tremendous diversity in which smallholders produce food for a significant portion of the world's population must be taken directly into account. The use of averages and aggregated data by development agencies in attempts to reach broadly adaptable solutions simply masks the nature of the situations and conditions in which these smallholders produce or earn their food and other necessities. As we have begun to analyze the impact of the dynamics of family development and associated diverse changes in household composition, we have been able to see that averaging these even for a community shifts family development to conditions of lower stress (more food secure) than most households actually exhibit. Conclusions based on these lower stress, composite households will inevitably distort reality and lead to unanticipated and undesirable results.
Smallholders and Food Security
Food insecurity (in its simplest form: not always being able to count on having enough food to meet locally acceptable levels of consumption)
results from various factors that differentially affect diverse groups of both rural and urban people. In order to analyze food security issues, and to make sensible recommendations to alleviate food insecurity, it is necessary first to define the category being addressed. Distinguishable categories of potentially food insecure include 1) urban poor, 2) rural
landless, and 3) rural people with access to land, either as farmers (including either or both crop and livestock products) or as extractivists. For our purposes, access to bodies of water for fishing is considered as being landed. Within the third category, which is the group of interest to us here, Are three sub types of households that must be distinguished: 1) those whose resources at any moment in time generally allow them to produce a surplus beyond locally acceptable minimum livelihood and food security levels; 2) those whose resources at any point in time are generally or always insufficient to allow them to reach locally acceptable minimum livelihood and food security levels and therefore suffer chronic shortages (probably being the "extremely poor" in the IFAD report); and 3) an intermediate type on the margin between sufficiency and insufficiency of resources. These are not permanent categories-households can and do move from one to another as they pass through different stages in their lives. Unfortunately, no good data exist on the relative numbers of people or households usually found in each of the three rural, landed categories (Barbara Huddleston, FAO, personal communication) Nevertheless, there are certainly a large number of nonurban households in the categories to be addressed. Beyond this designation, no further effort will be expended on defining the target sectors, but it is certain that they are the same as the group defined by the Russians as peasants.
Poverty Proliferates Poverty
Even for the landed poor households who in their stage in life have the resources generally, but not always, to achieve socially acceptable minimum levels of livelihood, poverty proliferates poverty. Three factors exacerbate the effect of the vortex that draws these poor into increasing poverty and
food insecurity. One is the need to produce many of the diverse things needed for their livelihoods from their own resources even when specialized production could increase the overall value or volume of their efforts. Second, the inescapable need for cash for necessities not produced on the farm, and often at specific times of the year, forces these people to direct their resource use into activities that meet this need even if it reduces the household's overall value or quantity of production. Third, in households with many mouths to feed and the least amount of labor available (as with young families with several young children), the household requires the most nutritionally dense, and therefore most expensive diet, in terms of cash or resource use. Later in life, when such households have adolescent children they often move into a phase of relative abundance and low stress, and relative food security, particularly if the older adolescents help in production activities or free adult labor from reproduction activities so they can participate in production tasks. Unfortunately, when households in a community are averaged, the average family inevitably has adolescent children so therefore does not reflect the high stress situation.
Specialization Versus Diversification
In commercial, family agriculture (that is, capitalized but not industrialized agriculture) most efficient resource use, and therefore highest profits are normally achieved by combining two or more activities in the farm operation. Economists demonstrate this using what is called a Production Possibilities Curve, Figure 1. When two enterprises (activities) are combined on a relatively well-endowed family farm with abundant resources, the opportunities curve bows away from the origin so the greatest revenue obtainable from the resources of the farm is achieved with a
combination of the two enterprises. In Figure 1, by producing approximately 260 units of Product 1 and 380 units of Product 2, the combined value of the two products would be approximately $820 if the price of Product 1 were $1.42 and the price of Product 2 were $1.00. This may reflect the case of the consistent surplus category farmers above. Farmers in this category often allocate their resources to the combination of enterprises that bring them the greatest revenue. Food production is of secondary importance and some or all is purchased with the income generated from the sale of products.
Abundant and Scarce Resource Farms
X! I N
0 200 i
100 4-4- Value
0 100 200 300 400 500
Figure 1. Production possibilities curves and revenue lines
For farmers on the margin of food security, a number of factors
influence the nature of the production possibilities curves shown in Figure
1. The result can be the exact opposite of farmers in better resource situations. To the extent that these farmers have access only to lower quality resources (for example, steeplands as opposed to valleys), their production functions shift downward, resulting in less product for the same amount of input. Also, more of the resources on marginal farms are fixed in quantity at the farm level (cannot be purchased, so cannot vary in quantity). This means that as more enterprises or activities are undertaken, each of
them receives a smaller proportion of such scarce resources as manure, family labor or management time, further reducing the level of productivity.' This can result in an opportunities curve that bows toward the origin. In this situation, the farmers would be better off (able to achieve more value from their resources) by specializing in one enterprise rather than in diversifying into two or more. For the same prices, the farmer with scarce resources would be able to achieve more value (value of crop production would be approximately $325) by producing only Product 2, Figure 1. But for these farmers, food security and therefore food production is of primary importance. To specialize, they would have to depend on the market, other infrastructure such as transportation, governmental that would assure them relative stability, and confidence that they could sell their products and purchase the quantity and quality of food they needed when they needed it and for a reasonable price. Most of these conditions are missing for the marginally food secure families considered here. For this reason, they are virtually forced into enterprise diversification in order to assure themselves an adequate diet. If the resource-restricted farmer illustrated in Figure 1 needed at least 50 units of Product 1, then the shift of resources from Product 2 would reduce production of Product 2 to about 175 units and the value of crop production would be only about $220. Thus, forced diversification to achieve a nutritionally adequate diet for farmers with a production possibilities curve that bows toward the origin means that the combined value or quantity of crop and animal production they can achieve is lower than otherwise might be possible. Thus, poverty proliferates poverty. This effect is seen in most of the cases studied.
A more complete explanation is in: Hildebrand, 1986.
The Need for Cash
Not all these households' food nor other of life's necessities are
acquired directly by production or extraction. Virtually all households in the world at the end of the 20th Century require cash for such necessities. (Chayanov recognized this need among the Russian peasants at the beginning of the 20th Century.) This need further complicates the livelihood strategies of farmers who are only marginally food secure. In some cases, cashproducing activities such as Brazil nut gathering in the Amazon must be undertaken even if this activity competes with food production and produces relatively little cash. These cash producing activities can take the form of extraction, off farm work, hunting or fishing, or the production of something like tobacco that cannot be used for food. Many sources of cash are from non-traditional activities such as eco-tourism or previously unknown crops, and shifts in labor demands among household members to carry out these nontraditional activities can create additional burdens on certain members of the household and reduce further the productivity of traditional food sources. Again, poverty proliferates poverty.
Family Development and Household Composition
Among many other factors affecting food security of landed, rural
households is household composition. The following stylized figures are a way to visualize the impact of these factors on food security. Consider a couple just setting up a nuclear family household (in year zero). For a short time, both husband and wife can work in production, or food- and cashacquiring activities, because the wife has a relatively limited number of household duties without children. There are two mouths to feed and two
workers in the household to contribute to production activities and food acquisition. However, as soon as the woman has a child, her duties shift noticeably and she must spend more time in the home attending reproduction activities. Thus, at the same time that the number of mouths to feed increases, the work force available for production activities decreases (Figure 2).
Factors Affecting Food Security
0 5 10 15 20 25
Years in Life Stage
-a- Family Size Workers
Figure 2. Factors affecting production and consumption
Young children require a diet that has a higher nutrient density (nutrients per kilogram dry weight) than adults (Figure 2). That is, children require more proteins, minerals and vitamins per kilo of food eaten. A diet rich in these nutrients (meat, milk, legumes) requires more labor and/or more cash (either for inputs or purchased from the market) to acquire. This means that even more effort must be put into the family food budget, Figure 3, during the years when the woman must spend more time in reproduction activities so less total labor is available to the household for food production activities. One way of showing this food budget is by adding nutrient density to number of people in the family, Figure 3.
Factors Affecting Food Security
0 5 10 15 20 25
Years in Life Stage
- Family Size Workers Rel Nut Dens Req'd Food Budget
Figure 3. Food budget
The combined effects of an increasing number of mouths to feed, increased nutrient density required in the diet, and reduced number of workers in production activities, puts a great deal of stress on the household during the first ten or so years of family development, Figure 4.
Factors Affecting Food Security
25 -- I ---- d -20
- 10 ] -L- q -d
0 5 10 15 20 25
i- Family Size Workers Rel Nut Dens Req'd Food Budget iStress on Household
Figure 4. Household food security stress related to life stage.
Stress here is measured as food budget index divided by the index of the number of workers available for production activities and is similar in concept to Chayanov's consumer-worker ratio. Stress is very low until the 10
first child is born. It peaks when there are a number of small children in the household and only the two parents available for reproduction and production activities. This can shift a household from food secure to food insecure. Stress again declines as the oldest children begin to take on some of the production and/or reproduction activities. Thus, the household can shift again, from food insecure to food secure.
A number of factors can have a marked influence on these stylized
curves and may be variable for different cultures. For instance, if another adult joins the household, many of the curves shift markedly. If the adult is male, it usually affects the number of workers for production activities as well as the number of mouths to feed. However, the relative shift in production resources is probably greater than the shift in consumption requirements. Also, the relative nutrient density of the diet shifts downward. All these factors shift the stress curve downward at the point when the adult enters the household. If the new adult is female, shifts ordinarily will be distributed differently than if the adult were male. Also, if the labor resource of the first child to reach adolescence is used to directly or indirectly enhance production activities, this can reduce the amount of stress and enhance the recovery into a low stress phase, particularly if the male adolescents help the father and the female adolescents help the mother. While these figures are based on generalized situation, it is clear that in high stress household compositions, poverty
Potential for Aggregate Analysis
Frustrating national and donor attempts to alleviate food insecurity for these categories of households is their tremendous diversity-a necessary
strategy for these same households. The major donors and their food security programs desire broad solutions, felt necessary for reasons of efficiency. Almost universally this translates into the use of data averaged across a community, watershed, county, region or even country. The same people who advocate and understand the need for conservation of blo-diversity as a critical component for sustainability often ignore livelihood diversity across communities, watersheds, counties, regions or countries. The problem related to a search for potential solutions to food insecurity is that an "average" household of two adults and three children aged 7, 11 and 15, for example, reflects the household composition of few, if any households in that community. Based on Figure 4, the average household described above would be coming out of the high stress phase and have resources available to consider new options. They may have resources available, for example, to consider an agroforestry program with fruit trees that might not come into production for three or more years. Their discount rate for longer term potential income may be sufficiently low, compared to a high stress household, that they view favorably the use of current resources to provide for a higher level of living in the foreseeable future. Using the average household as a basis for making project decisions may lead to frustration on the part of the donors. Likewise, using households that have recently moved out of the high stress phase as examples of successful projects seriously overstates the impact such programs have.
All of the factors described above can be incorporated in a linear
programming analysis of food security and many of the cases in our research use this method of analysis. Seasonality of food and cash needs, and household composition are the drivers of the livelihood strategies of these categories of rural, landed households. Members of these households choose their livelihood strategies to attempt to satisfy seasonal food and cash needs within the constraints of the resources they have available to them.
By using linear programming models of livelihood systems, various household compositions easily can be incorporated in an aggregate model reflecting, for example, a community, that also can be aggregated to a larger scale. But the larger scale can also reflect the diversity at the lower scale so that technologies, projects and policies can be tailored for each of the household types at the community level.
Ethnographic and Participatory Linear Programming
Our methodology is based upon recognizing and understanding the structure of complex livelihood systems and the diverse strategies of different households within a livelihood system. This recognition and understanding comes from:
" Going to the field;
" Using ethnographic and participatory methods and tools for data
" Creating models that describe existing livelihood systems;
" Simulating the strategies of diverse households within a given
" Testing alternatives (e.g., technology, infrastructure, or
" Estimating responses and formulating recommendations.
Recognition and understanding of diversity in limited resource households depends on examining them from the appropriate point of view, and choosing the appropriate level of analysis.
" The appropriate point of view is that of an insider.
" The appropriate level of analysis is the livelihood system.
Livelihood systems are the appropriate level of examination because they naturally differentiate diversity by:
" Accessibility to technology (e.g., tractors, irrigation).
These kinds of factors can impose a particular livelihood system on a group of households by governing, dictating or modifying the full range of activities (crop and livestock production, crafts, off-farm work, remittances, etc.) available to the individuals in the households to contribute to their survival and well being. When the livelihood system to be modeled has been selected, diverse kinds of household composition subject to the system are included in linear programming analyses. These multiple evaluations then represent the diversity of strategies found in the livelihood system. By weighting the kinds of household compositions analyzed, an aggregation representing all households in the system can be achieved.
Hildebrand, P.E. (Ed). 1986. Perspectives on farming systems research and extension. Lynne Rienner Publishers, Inc. Boulder.
Mosher, A.T. 1969. The development problems of subsistence farmers: A preliminary review. Chap. 1 In: Wharton, Jr., C.R. (Ed). 1969. Subsistence agriculture and economic development. Aldine, Chicago
Shanin, T. 1966. Chayanov's message: Illuminations, miscomprehensions, and the contemporary "development theory." Forward In: Thorner, D., B. Kerblay and R.E.F. Smith (Eds). 1966. A.V. Chayanov: The theory of peasant economy. University of Wisconsin Press, Madison.
Thorner, D. 1966. Chayanov's concept of peasant economy. Chap 2 In: Thorner, D., B. Kerblay and R.E.F. Smith (Eds). 1966. A.V. Chayanov: The theory of peasant economy. University of Wisconsin Press, Madison.
Wharton, Jr., C.R. (Ed). 1969. Subsistence agriculture and economic development. Aldine, Chicago.