Jump to... Executive summary Introduction Objectives Survey methodology Role of women Household vulnerability Conclusions Agenda Appenidx

Food access, risk and insurance

Household food security can be loosely defined as the ability of all individuals to access an adequate supply of food, on a stable basis, and in a sustainable way. The literature on HFS boasts countless definitions, although one put forward by Reutlinger and the World Bank is perhaps the most used and quoted. It defines HFS as "Access by all people at all times to enough food (of good quality) for an active, healthy life" (World Bank 1986). Regardless of the definition adopted, four core concepts are common to most: access, security, sufficiency and time (Maxwell & Frankenberger 1992). In the course of this paper, the primary focus will be on the first two aspects, where access is defined as the entitlement to produce, purchase, exchange or receive food, and security as the balance between vulnerability, risk, and insurance (Maxwell & Frankenberger 1992). As such, household vulnerability must be assessed in terms not only of immediate access, but also of the stability and sustainability of those channels through which the household mediates its food access. Thus, risk considerations must be an integral part of such an assessment. For example, a household can improve its access to food by disposing of assets or by investing in a riskier or less sustainable activity, but its vulnerability is likely to increase as a result of such action.

A household may derive its food entitlements from its own production, income (from the sale of labour or of surpluses), and disposal/use of assets. When households are able to generate a surplus above their basic food requirements, the excess resources are diverted into assets, from which the household can draw in the event of a food crisis. Assets can be either physical or human, or merely in the form of social and institutional claims. Within this perspective, each household has a portfolio of assets and claims that are managed according to evolving HFS coping strategies; the breadth and riskiness of these options will map the vulnerability profile of each household. Hence, an analysis of the relative importance of these idiosyncratic dimensions of a household’s food security strategies and the identification of the synergies derived by the individual household’s reliance on more than one of these sources to foster food security will provide the basis for locating the most vulnerable segments of a given population. Based on this framework, the most food insecure households will be those that face the combination of shortfalls in immediately available food (through their own production or through income-generating activities), assets, and claims, all leading to a greater exposure to the risk of present or future food entitlement failure.

Idiosyncratic transaction costs characterizing food, product and asset markets in less-developed countries (LDCs) tend to vary across geographical areas, and also across household and individual levels, creating biases that constrain access to markets. As a result, poor rural households in remote areas tend to rely heavily on their own produce. Households that are unable to meet their food requirement through their own production must rely on markets, with the consequent risk of entitlement failure deriving from market shocks – the larger the share of total household resources devoted to food purchases, the greater the shock. Output shocks in food production, however, can substantially increase the vulnerability of subsistence farmers. For these households, crop and income diversification, such as seasonal off-farm agricultural labour, emerge as very important coping mechanisms to reduce the risk of entitlement failure as a result of shocks in output of staples.

Even in the presence of highly imperfect food markets, food market dependency alone does not imply greater food insecurity. In addition to producing food, a household can gain access to food by generating sufficient revenue to be able to purchase the necessary items with earnings or divested assets. Sufficient income, however, is not the only dimension through the income variable can be characterized. At equal levels of food market dependency, identical income levels can conceal differences in food entitlements and in the risk of food entitlement failure, due to several possible factors: (a) differences in the number of income sources: generally speaking, the larger the number of sources, the lower the risk of entitlement failure; (b) differences in the riskiness of individual income sources, and in the variance between different sources; (c) differences in intrahousehold control, with heterogeneous preferences and the propensity to acquire food among household members; (d) differences in household- and community-specific transaction costs in food markets; and (e) differences in the shares of income derived from asset divestment. Consequently, the level of income, even if combined with the concepts of home-produced food supply and food market dependency, would still not be sufficient to fully characterize HFS in its complexity. Several other factors that account for differences in the level of risk associated with a household’s food security strategies must also be included in order to portray a household’s vulnerability to food entitlement failure.

In the presence of uncertainty, diversification of income sources is an important part of a household’s strategy to reduce the risk of entitlement failure and, as such, must be accounted for in any attempt to measure vulnerability. At low levels of income and high levels of destitution and food insecurity, diversity in income composition may be even more important than the actual income level in securing the survival of vulnerable household members in the event of a food entitlement failure.

Furthermore, in circumstances where financial markets are highly imperfect or non-existent, asset ownership is a common form of precautionary savings among households in rural communities. For food security purposes, asset ownership must be further characterized on the basis of liquidity. Liquid assets, such as small animals, are often used by poor rural households for consumption smoothing, as a form of insurance against the risk of food entitlement failure. Liquid asset divestment, if done to cope with temporary food shortages, can be interpreted as a good indicator of moderate vulnerability (Watts 1983). Conversely, divestment of assests that are less liquid, such as land or a house, often indicate extreme vulnerability.

Asset levels and changes in asset ownership over time appear to be promising indicators of prevailing vulnerability, particularly if it is possible to clearly identify distress sales². Lack of distress sales, however, does not necessarily imply food security. Even where asset holdings are larger, households may be reluctant to dispose of productive assets to safeguard current food consumption levels because of the opportunity cost in terms of future food access (which relates to the concept of livelihood security). Preservation of assets takes priority over meeting immediate needs until the point of destitution, when all options have been exhausted (Corbett 1988). Furthermore, the decision to forego an asset sale can be related in some circumstances to the decision-maker’s perception of the gravity of the food problem. Because asset divestment is a more ‘tangible’ concept than caloric requirement or nutrition assessment, the decision will often be biased against the latter. Finally, asset ownership or transactions may not accurately reflect food adequacy, as a result of different preferences and propensities within the household regarding the sale of assets by the asset owner in order to meet emergency food needs³. Nevertheless, asset ownership and/or divestment will provide valuable information for identifying food-insecure households.

Once a definition of HFS has been adopted and the various components identified, the next step in the attempt to operationalize the HFS concept calls for choosing the particular indicator to be used in characterizing the food insecure. Consensus has still not been reached on acceptable indicators and methods of measuring HFS (Haddad et al. 1994). As expected, the variety of indicators and their chronological evolution reflect the shift in paradigms documented earlier. The choice of a particular indicator must be based on the specific objectives of the research, and the trade-offs between resource constraints and information needs.

Traditional indicators have often included both supply-side indicators, such as food production, and consumption-side indicators, such as household food consumption, total expenditure and calorie adequacy (direct measures)⁴, and anthropometric measures (indirect measures). Socio-economic indicators have become increasingly important in HFS and vulnerability assessment. Simple measures such as household access to assets are often good determinants of vulnerability (Chambers, 1989 Swift 1989a. Also, variables that are relatively easy to collect, such as household size and composition, land use and ownership, and asset liquidity, can often be successfully used as indirect indicators to locate the most vulnerable groups in terms of food security (Haddad et al. 1994).

Based on the above conceptualization of HFS, Section V sets out a number of single-factor socio-economic indicators likely to contribute to locating food-insecure households, while Section VI contains a composite indicator combining information on household food access and vulnerability to entitlement failure, relating this measure to the role of women and household participation in IFAD projects.

Box 1 gives the variables that will be analysed in Section V as individual factors credited for contributing to HFS, and will be considered for inclusion in the proposed composite indicator of HFS.

Box 1. Factors affecting HFS

Intrahousehold issues of HFS: assessing women’s role

A past approach by many development programmes to improve the food security of the target population was geared towards (i) increasing the household’s ability to produce its own food via asset redistribution and technological innovation, and (ii) increasing household income levels, thus enabling the household to acquire an adequate food supply. This approach was developed under the assumption that the potential beneficial effects of food-related programmes would ultimately trickle down to all the household members not directly targeted. Only recently has the flaw in this approach become evident. As pointed out by Pinstrup-Andersen, not only the availability (household supply) and ability to acquire food (household access) must be considered, but also the desire to acquire food, and the intrahousehold preferences and distribution (the latter being related to individual access).

In the search for a better understanding of the missing link between food-related programmes and the target recipients, one must abandon the predominant view of the household as a homogeneous decision-making and benefit-sharing unit. A clear understanding of the intrahousehold mechanisms of resource allocation and the potential conflicts arising from individual preference is of paramount importance if programmes are to reach the legitimate recipients. There is now an increasing consensus on the importance of focusing on women to reduce leakages of food and nutrition programme benefits to non-targeted recipients. Several studies have shown a strong correlation between women’s income and improved nutritional status. Empirical evidence supports the view that women show a higher propensity towards acquiring food out of incremental units of income under their control. If this is the case, household food entitlements are not necessarily equivalent to women’s entitlements; the control of income and assets along gender lines within the household therefore becomes crucial.

In their quest to improve women’s status, many development projects, including those of IFAD, have focused on activities commonly considered to be within the domain of women, such as home production, health, nutrition and reproduction. However, as the importance of women as agricultural workers and income earners have become increasingly recognized, policies and programmes have been directed towards eliminating the inequities and constraints affecting women’s productive role. These relate to: (i) women’s reproductive role, both in terms of physical strain and of time diverted from income-earning activities; (ii) limited access to factors of production and information; and (iii) asymmetrical obligations and rights (Gittinger et al,1990). By ignoring the gender issue, some past policies have actually contributed to worsening women’s bargaining power within the household by strengthening the existing gender bias, and excluding women from access to factors of production. There is often a surprisingly clear-cut division between what is perceived as being within the male or female domain regarding tasks, income and decision-making in the areas of production and consumption. Information on the division of labour and resource control is crucial if correctly targeted strategies are to be put in place in support of individuals at high nutritional risk.

In addition to the desired income control effect, women’s involvement in market activities is likely to induce secondary effects on food and nutrition security at the intrahousehold level. A greater role in income-generating activities is unlikely to release women from home chores. Many activities can be undertaken simultaneously with home production (e.g., handicraft manufacturing and the running of a store). Activities such as caring for animals and selling at markets, however, can only be done by sacrificing other tasks. For this reason, women’s activities must be evaluated not only in terms of income-earning potential and the total amount of time allocated to the activities, but also in terms of the degree of simultaneity that each activity allows. Thus, income-enhancing activities, such as manufacturing handicrafts, have the advantage that they can be performed without the necessity of leaving the house or the community. Other activities, such as agriculture production, caring for animals and selling at markets, require that women be absent from the house or the community for extended periods during the day, or even for several days at a time. The consequences can be particularly harmful in terms of the health and nutritional security of children, the most vulnerable members of the household, particularly in cases where a suitable substitute for women’s time is not readily available.

In pursuance of this intrahousehold refinement of the concept of food entitlement, Section V analyses the information collected on women’s activities and the control of resources within the household, relating them – in Section VI – to the indicators of household food entitlement. Specifically, the aim is to quantify the level of women’s participation in income-generating and non-market activities, and the degree of control retained by women over household resources. The empirical questions addressed in this section are whether there are some activities particularly associated with women’s greater control over household entitlements, and whether the fostering of women’s control over these areas effectively translates into improvements in HFS.

IFAD’s approach to HFS and women in development

Two decades of experience in mitigating food insecurity in the region, and a firm commitment to prioritizing the poorest segments of the rural population, make IFAD an important player in the challenge faced by an increasing number of Latin American households to guarantee stable and sustainable food supplies for each of their members.

During the past two decades, IFAD’s strategy towards alleviating food insecurity in the region has adjusted to respond to the rapidly changing conditions that have characterized the Latin American context. In recent years, IFAD has advocated a more diversified and comprehensive approach, with a primary focus on income-generation, in line with the evolving development theory. In the projects under study, this goal has been pursued mainly through the building of infrastructure and improved access to credit and technical assistance for productive activities. One of the objectives of this report is to assess the impact of IFAD projects on improving poor households’ access to food. In view of the definition of HFS adopted above, this assessment must go beyond a mere analysis of induced income effects to touch on other important dimensions, such as the role the projects play in reducing risk and in providing surrogates for insurance, thus reducing the exposure of poor households to food insecurity.

Recently, IFAD has been pursuing a refinement of the concept of food security promotion by including gender considerations in projects concerned with poverty and nutritional issues. The Fund’s commitment to the new development paradigm emphasizing the importance of women is clear. However, the impact of this approach at the community level has not been carefully researched. To this end, Section V analyses the data by household participation status in one of IFAD’s projects in an attempt to clarify the relationship between project participation and women’s involvement in income-generating activities, their access to resources, their responsibility in decision-making and their HFS status.

Agricultural commercialization, HFS and women

At various times during the last two decades, many Latin American countries have embarked on structural adjustment programmes, emphasizing, among other goals, the production of non-traditional agro-exports. The rationale for the promotion of NTXs in the region is multifold. It includes, at the macro level, foreign exchange needs, high foreign debt levels and depressed market conditions for traditional export crops, and, at the micro level, the desire to generate employment. In some instances, these programmes have produced the expected short-term benefits, such as increased household incomes, and also generated badly needed foreign exchange. However, research on the distributional effects of cash crop adoption and its impact on HFS and women has been sporadic, and the results often ambiguous. Initial indications point to hidden costs and inefficiencies resulting from a failure to include intrahousehold considerations in non-traditional crop adoption programmes.

Empirical evidence indicating the positive relationship between NTX adoption and household income has ignited widespread optimism about the beneficial effect of NTX policies on the well-being of poor adopters. Promising early signals have led many development practitioners and international donors to the unconditional promotion of these crops in many areas throughout the world. However, the empirical link between household income and the food security of individual household members has not proved to be very strong. There are many reasons for this, including changes in the composition and control of income within the household, and the changing mechanisms of intrahousehold resource allocation resulting from NTX adoption. Cash crops, and the income accruing to the household from their cultivation, are usually controlled by men. However, the burden of the additional work they require is shared equally across the household. Consequently, the adoption of cash crops may be responsible for widening the gender gap within a household. The widespread opinion expressed in recent studies is that while women are assuming new tasks and additional burdens, they may not be getting a fair share of the accruing economic benefits.

Furthermore, it has been argued in the NTX literature that increased agro-export production increases food dependency, both at the national and household levels. Whether food and cash crops are complementary to or whether they compete with scarce resources, is still an open question that deserves attention because of its relevance to the food security debate.

The issue of NTX adoption is relevant here since the two IFAD projects in question include a component geared towards assisting farmers in the production and commercialization of new agricultural products. Considerable effort is being made, in the NTX communities, to promote the adoption of export crops such as broccoli and cauliflower. As empirical evidence from other areas of Guatemala has shown, NTXs are not immune to problems and, particularly in the nineties, a combination of factors have rendered their production extremely risky and often unprofitable. This provides an interesting case study for briefly analysing the short-term effect of these crops on nutrition-related factors.

Using the information collected in three of the eight communities in the sample (Chochal, Torlón and Chichalum), households are compared based on whether or not they have adopted NTX crops. The goal is to provide the project with some feedback on the side effects of promotion of commercialization, since little is known about the changes created by NTX adoption in these communities. Using the dichotomy of NTX non-adopter-adopter, we will investigate the socio-economic features of the two groups. The analysis will emphasize aspects directly related to women (namely, women’s time allocation and their control of income and assets), as well as those that are key to HFS (namely, household income, crop diversification, food market dependency and farm productivity).

---

2/ Distress sales are those sales of assets and livestock reported to have been made to purchase food and basic necessities, to repay debts or to pay for expenses following illness of a family member.

3/ We will take up the issue of intrahousehold asset ownership in the section dedicated to women's resource control.

4/ Because of the preliminary nature of this study and the limited time and resources available, no attempt was made to collect information on individual caloric adequacy. This type of information is expensive to collect and is not immune to problems. Also, although we recognize the importance of including anthropometric measures in the assessment of a household vulnerability to food insecurity, we refrained from collecting this type of data due to the above-mentioned constraints.