Enabling poor rural people
to overcome poverty
On the basis of the findings and concerns raised in the course of producing this report, the following suggested areas for further research have been created:
The simple indicator proposed must be considered merely as a starting point in the search for a more comprehensive indicator of a household’s vulnerability to food insecurity. The main scope of this presentation was to offer some preliminary thoughts on the issue and suggest a simple conceptual framework for further elaboration. In particular, the use of different variables should be explored in an attempt to capture the idiosyncrasies of the food security strategies of different (groups of) households in distinct regions and contexts. Also, the use of different methods of aggregation could and should be further investigated. Finally, attempts to relate this type of relative indicator to some more absolute measure of food security should be pursued further.
Some caution must be used in interpreting the findings with regard to the impact of project participation. To quantify this effect in absolute terms, one must consider the possible existence of selection bias in the projects’ recruitment process. It may well be that the differences observed between beneficiaries and non-beneficiaries are not entirely the result of the project's activities but of pre-existing heterogeneity between the two groups, thus reflecting beneficiaries’ better endowments from either a human or physical capital standpoint, dating back to the pre-project period. A quick comparison between non-beneficiaries and (future) beneficiaries in the pre-project period with respect to human capital (education) and physical assets (land) revealed how this may have been the case in some of the communities surveyed. In Tituque, just as in Buena Vista and San Martín, the projects appear to be working with better-endowed households in their effort to promote coffee and small livestock production, respectively. While this may be logical from a purely economic rationale, it may also reflect the inability of projects to target the poorest of the poor through productive investments. However, in order to disentangle the two effects, the collection of panel data including baseline information would be required.
Furthermore, to validate the proposed indicator and test its soundness, the availability of benchmark measurements, such as anthropometric measures of children and/or women, would have been extremely helpful. Due to the limited scope of this study, no attempt was made to collect this type of information. However, some anthropometric measurements, such as the body mass index (BMI) of non-pregnant women of childbearing age, are relatively easy to take, and often provide a good indication of nutritional performance. Given the projects’ emphasis on income-generating activities for women, the BMI, in particular, could be successfully used to capture the net impact, after comparing the presumed income benefits and any potentially negative effect resulting from women’s time reallocation and/or increased workload following their involvement in market activities.
