Contents - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Food and Nutrition Bulletin
Charles Street Sta., P.O. Box 500 Boston, MA 02114-05()0, USA
Tel: (617) 227-8747 Fax: (617) 227-9405
United Nations University Press
The United Nations University 53-70, Jingumae 5-chome, Shibuya-ku Tokyo 1 50, Japan
Tel: (03) 3499-2811. Fax: (03) 3406-7345.
Telex: J25442. Cable: UNATUNIV TOKYO.
The Food and Nutrition Bulletin incorporates and continues the PAG Bulletin of the former Protein-Calorie Advisory Group of the United Nations system and is published quarterly by the United Nations University Press in collaboration with the United Nations ACC Sub-committee on Nutrition.
All correspondence concerning the content of the Bulletin, comments, news, and notices should be sent to the editor at the Boston editorial office address given above.
All material in the Bulletin may be reproduced freely provided acknowledgement is given and a copy of the publication containing the reproduction is sent to the editorial office.
The Food and Nutrition Bulletin is intended to make available policy analyses, state-of-the-art summaries, and original scientific articles relating to multidisciplinary efforts to alleviate the problems of hunger and malnutrition in the developing world. It is not intended for the publication of scientific articles of principal interest only to individuals in a single discipline or within a single country or region. Notices of relevant books and other publications will be published if they are received for review. The Bulletin is also a vehicle for notices of forthcoming international meetings that satisfy the above criteria and for summaries of such meetings.
The Food and Nutrition Bulletin also serves as the principal outlet for the publication of reports of working groups and other activities of the UN ACC Sub-committee on Nutrition (SCN) and its Advisory Group on Nutrition.
The SCN itself is a focal point for coordinating activities of FAO, WHO, UNICEF, the UNU, UNESCO, the World Bank, the World Food Programme, the World Food Council, the United Nations Environment Programme, and other bodies of the United Nations system that have an interest in food and nutrition.
Submissions. Unsolicited manuscripts of articles of the type published in this and previous issues may be sent to the editor at the Boston office address given above. They must be typed, double-spaced, with complete references and must include original copy for any figures used (see the "Note for contributors" in the back of this issue).
Any disciplinary or conceptual approach relevant to problems of world hunger and malnutrition is welcome, and controversy over some of the articles is anticipated. Letters to the editor are encouraged and will be printed if judged to have an adequate basis and to be of sufficient general interest.
Peer review. The Bulletin is a peer-reviewed journal. Every article submitted first receives editorial review. If it is consistent with the editorial policy and is not obviously deficient in some way, it is sent to two or sometimes three experienced and knowledgeable reviewers. Occasionally a paper may be returned to the authors by the editor with suggestions for improvement before it is submitted to the reviewers.
If two reviewers agree that the paper should be published in the Bulletin, it is accepted and either sent immediately for copyediting or returned to the authors for consideration of suggestions from the reviewers and the editor. If both reviewers agree that the paper should not be accepted, the editor writes a personal letter to the authors explaining the reason and enclosing the comments of the reviewers anonymously. If the reviewers do not agree with each other, either the paper is sent to a third reviewer or a decision is taken by the editor. In these cases, the authors are usually given a chance to respond to the reviewers' comments.
Disclaimer. It is expressly understood that articles published in the Bulletin do not necessarily represent the views of the United Nations University, the UN ACC Sub-come mittee on Nutrition, or any United Nations organization. The views expressed and the accuracy of the information on which they are based are the responsibility of the authors. Some articles in the Bulletin are reports of various international committees and working groups and do represent the consensus of the individuals involved; whether or not they also represent the opinions or policies of the sponsoring organizations is expressly stated.
The United Nations University (UNU) is an organ of the United Nations established by the General Assembly in 1972 to be an international community of scholars engaged in research, advanced training, and the dissemination of knowledge related to the pressing global problems of human survival, development, and welfare. Its activities focus mainly on peace and conflict resolution, development in a changing world, and science and technology in relation to human welfare. The University operates through a worldwide network of research and postgraduate training centres, with its planning and coordinating headquarters in Tokyo.
The United Nations University Press, the publishing division of the UNU, publishes scholarly books and periodicals in the social sciences, humanities, and pure and ape plied natural sciences related to the University's research.
Food and Nutrition Bulletin
Editor: Dr. Nevin S. Scrimshaw
Assistant Editor: Ms. Edwina B. Murray
Editorial Consultant: Ms. Sarah Jeffries
Senior Associate Editor Clinical and Human Nutrition:
Dr. Cutberto Garza, Director and Professor, Division of Nutritional Sciences, Cornell University, Ithaca, N.Y., USA
Senior Associate Editor-Food Science and Technology:
Dr. Ricardo Bressani, Instituto de Investigaciones, Universidad del Valle deGuatemala, Guatemala City, Guatemala
Dr. Abraham Besrat, Senior Academic Officer, United Nations
University, Tokyo, Japan
Dr. Hernán Delgado, Director, Institute of Nutrition of Central America and Panama (INCAP), Guatemala City, Guatemala
Dr. Joseph Hautvast, Secretary General, IUNS, Department of Human Nutrition, Agricultural University, Wageningen, Netherlands
Dr. Peter Pellet, Professor, Department of Food Science and Nutrition, University of Massachusetts, Amherst, Mass., USA
Dr. Zewdie WoldeGebriel, Director, Ethiopian Nutrition Institute, Addis Ababa, Ethiopia
Dr. Aree Valyasevi, Professor and Institute Consultant, Mahidol University, Bangkok, Thailand
Food and Nutrition Bulletin, vol. 16, no. 3
(c)The United Nations University, 1995
United Nations University Press
The United Nations University
53-70 Jingumae 5-chome, Shibuya-ku, Tokyo 150, Japan
Tel.: (03) 3499-2811 Fax: (03) 3406-7345
Telex: J25442 Cable: UNATUNIV TOKYO
Typeset by Asco Trade Typesetting Limited, Hong Kong Printed by Permanent Typesetting and Printing Co. Ltd., Hong Kong
Lawrence Haddad and Ravi Kanbur
Through the use of a stylized, age-based eligibility feeding programme, we attempt to quantify the benefit of having individual- (as opposed to household-) level food intake data when it comes to targeting food transfers on the basis of age. In this context we show how optimum age eligibility cut-offs depend on the availability of intra-household data on food intake. Second, we provide quantitative estimates of the valeue of intra-household information and of knowledge of the process of intra-household allocation of calvies. Age proved to be a good indicator of individual calorie deficit. However, this was not the case with household-level calorie adequacy, which rendered age apparently less useful as a targeting instrument, often at considerable calorie cost. Food sharing within the household, on the other hand, rendered age impotent as a targeting instrument because of within-household leakage. If age is to be used as an effective eligibility criterion for a food transfer, the implementation of that transfer has to ensure minimum leakage to other household members. This type of exploratory analysis is one step toward quantifying the usefulness of intra-household data in the design of nutrition interventions. The costs of collecting intra-household data may outweigh the benefits, but the experiments presensed begin to answer questions about the costs of not collecting them.
Many food and nutrition interventions are targeted to young children and infants. These age groups are generally considered vulnerable to undernutrition due to low-energy-density weaning foods and vulnerable to infection due to the move from breast-feeding to weaning foods and increased toddler mobility. Moreover, the functional consequences of poor health are thought by many to be more severe for these age groups. Hence, age is often used as an indicator of vulnerability to malnutrition .
The case for age-based targeting of interventions would be even stronger if the actual allocation of food and nonfood nutrition inputs was skewed away from young children. That the food and nutrition status of some household members is not necessarily a good indicator of the food and nutrition status of other household members is by now a fairly well established generalization [2, 3]. A recent review of the evidence on the intra-household distribution found some promale and preadult bias in terms of the quantity of food intake, especially in south Asia, although variation was seen within that region. The review also found strong evidence of promale bias in the region in the allocation of non-food nutrition inputs such as health care .
If information as to the intra-household allocation of nutrition inputs were available, nutrition targeting could be fine-tuned for maximum impact. Such information is, however, costly to collect (retroactively we could not get access to such cost estimates). This begs the question that we hope to answer: what is the value of intra-household food and nutrition data in designing targeted nutrition interventions?
We attempted to quantify the benefit of having individual-level (or intra-household-level) data as the costs of a departure from the optimum design of a stylized, age-based eligibility feeding programme due to reliance on household-level data. This modeling exercise was carried out under two different assumptions as to the intrahousehold distribution of food.
The stylized feeding programme we constructed used an upper age limit for eligibility [5-7], with children age 6 to 36 months especially highly targeted.
The essential design question was what this upper age limit should the objective was to minimize food poverty with given resources for the provision of food supplements. In terms of the best age eligibility cut-off, how far wrong can one go with only household-level data on nutrition? A more appropriate objective function would be to minimize the effects of food poverty. This acknowledges that the benefits to an 18-month-old of eliminating food poverty could be greater than the benefits to an 18-year-old. This omission will likely bias the optimum age cut-offs upward but should not bias our estimates of he costs of not having intra-household data, as these are generated from a comparison of two identical objective functions.
The second objective was to provide a quantitative estimate of the value of the extra information that the costlier intra-household survey provides when the objective is to design optimally targeted nutritional interventions. It is recognized in the food and nutrition literature, however, that such interventions cannot be seen independently of intra-household food allocation, since a supplement to a child can be nullified by an equivalent reduction in feeding at home . The third objective was therefore to provide a quantitative assessment of how far wrong one goes by neglecting the intra-household repercussions of a nutritional intervention.
Optimum age cut-offs for nutritional targeting: Application to Philippine data
The data set came from a household survey in the Philippines. The data and methods of collection are described fully elsewhere . Information on nutrition among 448 households in the southern Philippine province of Bukidnon was collected and averaged over four rounds to account for seasonality and other fluctuations. The distinctive feature of the data is that the food intake of each individual in the household was obtained using the 24-hour recall method . This intake can be converted into calories using standard conversion factors. In addition, we can calculate the calorie requirement for each individual based on 32 age-gender-pregnancy status categories. For this reason, the data are illustrative rather than definitive measures of individual-level nutrient adequacy. For more precise analysis, individual energy requirements would in addition be based on body weight and activity patterns. The calorie adequacy ratio, the ratio of intake to requirement, was the measure of food poverty in this application, and a calorie adequacy ratio of I was the benchmark. This is referred to as a food poverty line.
Food poverty is defined in terms of the P(a) poverty index . For a = 0, Pa becomes the head-count index, or the proportion of n individuals falling below a food poverty line (z). The larger the value of a, the greater the sensitivity of the index to the depth of food poverty. For example, at a = 0 the index gives equal weight to the individual with a calorie adequacy of 0.7 and the individual with a caloric adequacy of 0.99. For a = 1, Pa becomes the gap index, or the average shortfall of calorie adequacy below the food poverty line. For a = 2, Pa measures the average squared shortfall of calorie adequacy below the food poverty line. Squaring the shortfall gives an even greater weight to individuals with larger calorie deficits and makes the index even more sensitive to the depth of food poverty. In figures 1-4, the vertical axes refer to P when a = 1. Thus the units refer to the average calorie adequacy shortfall from 1.0. We call these shortfalls food poverty.
The daily food energy deficit in our sample, namely, the sum of the individual differences between intake and requirement, is 1,048,631 calories for the 2,880 individuals in the 448 households. Let I be an individual's calorie adequacy ratio. If we did not have individual-level data, we would be forced to assign a household's calorie adequacy ratio to each individual in that household. This variable is designated H. Figure I shows that the mean of I in an age group increases, by and large, with age, but that the mean of H does not. This insensitivity of H to age drives many of the results below. The sensitivity of I to age may suggest a prima facie case for an upper age limit to calorie supplements through feeding programmes and the like. But what is the optimum age cut-off?
To calculate the optimum age eligibility, consider what happens to overall food poverty at different age cut-offs and different food transfer sizes. Figure 2 shows that for a food transfer size of zero, there is obviously no impact on overall food poverty at any age eligibility: the average calorie deficit remains ape proximately 18%. The lowest line in figure 2 is when the overall food transfer is fixed at I million calories, just about the amount necessary to eliminate the energy deficit if it could be targeted only to those with energy deficits. But when this is not possible, the curve shows the best that can he achieved with age-based targeting. As the upper age limit increases, more individuals are brought into the eligibility net, hut the size of the individual transfer is reduced because the fixed calorie transfer is divided equally among those eligible. At low age cut-offs, overall food poverty falls, and thus the marginal effect of bringing more people into the net dominates the infra-marginal effect of spreading resources more thinly over the existing beneficiaries. However, as figure 2 shows, eventually this balance is reversed, and overall food poverty increases. The age cut-off corresponding to the lowest overall food poverty is the optimum age cut-off. This is called example 1.
FIG. 1. Mean calorie adequacy within each age group for individual and household measures of calorie adequacy
FIG. 2. Food poverty levels (a = upper age cut-offs and caloric interventions based on individual data (example 1)
FIG. 3. Food poverty levels (a = 1) for different upper age cut-offs and calorie interventions based on individual and household data (examples l and 2)
The value of intra-household information
The analysis of the previous section is based on a survey that collects information on individual nutrition within the household, but intra-household information is costly to collect, and it would he useful to know its benefits. In particular, how useful is it in targeting? With this data set, we can provide an answer to this question. As before, let / be the true individual calorie adequacy ratio and H the individual calorie consumption adequacy ratio when each individual is simply allocated the household's calorie adequacy ratio. Here the household's calorie adequacy ratio is the mean of I within the household. Another household calorie adequacy measure is the ratio of the sum of calorie intakes across all household members and the sum of requirements across all household members. We have shown, at least for this sample, that these two measures track each other very closely (i.e., only a slight negative correlation between calorie intake and calorie requirement) . Without information on individual intakes, we would be forced to use the distribution of H to calculate the optimum upper age eligibility. Call this example 2.
Figure 3 compares the behaviour of food poverty lines based on I and H for fixed transfers of half a million and for 1 million calories. It is clear that the optimum age cutoffs can be different for I and H. In general, the curves based on household-level data are flatter and lower than those based on individual-level data. Intuitively, the flatness reflects the insensitivity of H to age. The suppression of intra-household inequality as represented by H results in age being a much poorer correlate with observed food poverty and hence a poorer targeting instrument. The marginal food poverty-reduction effect dominates the infra-marginal effect until much higher age eligibility levels are reached. In addition, the lowness of the H curve reflects the shallowness of observed food poverty, in all age groups, once intra-household inequality is suppressed.
What is the cost of getting the age cut-off "wrong" through the use of household-level data? One way to estimate this is to use individual-level data to calculate the extra calories that would be required to achieve the same level of food poverty reduction with the "wrong" age cutoff based on H. as was achieved with the correct age cutoff based on 1. This idea is adapted from work on land-contingent poverty alleviation transfers . Table 1 presents the amounts of these extra calories for various values of overall transfer and sensitivity to food poverty. The costs of not having accurate individual-level calorie adequacy information by which to identify the optimum age cut-off are in the range of 2% to 35% of the original interventions. The costs are substantial precisely because actual calorie adequacy is strongly associated with age, and suppression of intra-household calorie information deprives us of a useful targeting instrument.
TABLE l. Equivalent cost (in calories) of not having individual-level data with which to target (example I versus example 2)
|Sensitivity to food poverty||Size of calorie intervention||Best cut-off (yr) (individual data)||Best age cut-off (yr) (household-level data)||Cost of not collecting intra household data as percentage of original intervention|
|Low (a = 0) (index: head count)||100,000
|Medium (a = 1) (index: average calorie deficit)||100,000
|High (a = 2) (index: average squared calorie deficit)||100,000
Intra-household allocation, leakage, and the implications for targeting
The analysis so far has assumed zero sharing of the calorie intervention that the eligible individual brings into the household. Either because the intervention is divided within the household, or through reductions in nonintervention calorie intake of the eligible member, it is highly unlikely that intervention calories add, one for one, to the total calories consumed by the eligible individual. What are the implications for the age-based targeting of calorie leakage from the eligible individual to fellow household members? Does it still make sense? In general, this depends on the extent to which there is intrahousehold calorie allocation away from the targeted group, namely, children.
Specifically, our data set allows us to answer the question, how useful is it to know the calorie reallocation outcome if age is used as a targeting instrument? As before, let I be the true calorie adequacy ratio, and let each eligible individual receive an equal
share of the overall transfer. Now, however, the individual shares the calories with the other household members.
The arbitrary rule imposed here is that the individual's preintervention share of household calories is unaffected by the intervention. (This rule can be justified, however, by reference to certain principles of bargaining theory .) Call this example 3.
Figure 4 compares the behaviour of food poverty under the three different examples. When the three functions are compared on the same vertical scale, we can see that the food poverty line for example 3 is the flattest and lowest.
FIG. 4. Food poverty levels (a= 1) for different upper age cut-offs based on individual and household data and no food sharing and food sharing (examples 1, 2, and 3)
The flatness occurs because the original sampling design required that each rural household in the Philippines survey contain at least one preschooler. With leakage, each household immediately receives calories even when the upper age eligibility is only two years. Therefore age is a good targeting instrument only if poor households contain more young children and intra-household allocations are not skewed away from them. The same analysis with a more demographically representative sample containing older, richer households with no children would produce a more curved food poverty line for this example.
The low position of the line would seem to indicate that in a community where children and adolescents are deficient in calories, food poverty is significantly reduced if calories nominally directed to children are leaked to other household members. This counter-intuitive result is generated because the objective function we have chosen to minimize incorporates the food poverty across all individuals in the sample and places equal weights in the objective function on the alleviation of infant food poverty and adult food poverty. The former assumption means that when calories designated to preschoolers are leaked to adolescents who are calorie-deficient, the food poverty index declines. The latter assumption is clearly unrealistic due to the high vulnerability of infants to undernutrition and its severe consequences for them. If we had built this into the objective function, the curve in example 3 would have been much higher.
Under these conditions, the costs of making the wrong assumption on food sharing (i.e., assuming there is no food sharing even when food sharing does take place) are virtually zero, since, due to leak-ages, age is no longer closely associated with the delivery of calories to those who need them most.
In the context of a stylized targeting experiment, we have shown how optimum age eligibility cut-offs depend on the availability of intra-household data on food intake. Second, we provided some quantitative estimates of the value of this information and of knowledge of the process of intra-household allocation of calories. For our sample, age proved to be a good indicator of individual food poverty. However, this was not the case with householdlevel calorie adequacy, which rendered age apparently less useful as a targeting instrument, at an often considerable calorie cost. Food sharing within the household, on the other hand, truly rendered age impotent as a targeting instrument because of within-household leakage. This effect was strengthened because each household contained at least one preschooler. Therefore, getting the age "wrong" in the context of the stylized food transfer had few consequences in terms of calorie costs. If age is to be used as an effective eligibility criterion for a food transfer, the implementation of that transfer has to ensure minimal leakage to other household members.
This type of exploratory analysis is one step toward quantifying the usefulness of intra-household data in the design of nutrition interventions. Possibly the costs of collecting the data outweigh the benefits, but the experiments presented here begin to answer questions about the costs of not collecting them.
1. Kennedy E, Alderman H. Comparative analyses of nutritional effectiveness of food subsidies and other food-related interventions. IFPRI-WHO-UNICEF. Washington, DC: International Food Policy Research Institute, 1987.
2. Behrman J. Intrahousehold allocation of nutrients and gender effects: a survey of structural and reduced form estimates. In: Osmani S. ed. Nutrition and poverty. New York: Oxford University Press, 1990:287-320.
3. Harriss B. The intrafamily distribution of hunger in south Asia. In: Dreze J. Sen AK, eds. The political economy of hunger. Volume 1: Entitlement and wellbeing. New York: Oxford University Press, 1990:351424.
4. Haddad L, Peña C, Slack A. Poverty and nutrition within households: review and new evidence. Washing
ton, DC: International Food Policy Research Institute, 1994.
5. Pfefferman G. Griffin C. Nutrition and health programs in Latin America: targeting social expenditures. Washington, DC: World Bank, 1989.
6. Beaton G. Ghassemi H. Supplementary feeding programs for young children in developing countries. Am J Clin Nutr 1982;34(supplement):864-916.
7. Timmons R. Miller R. Drake W. Targeting: a means to better intervention. Report submitted to the US Agency for International Development. Ann Arbor, Mich, USA: Community Systems Foundation, 1983.
8. Alderman H. Food subsidies and the poor. In: Lal D, Myint H. eds. Essays in poverty, equity and growth. Washington, DC: World Bank, 1990:172-202.
9. Bouis H. Haddad L. Agricultural commercialization,nutrition, and the rural poor. Boulder, Col, USA: Lynne Rienner Press, 1990.
10. Foster J. Greer J. Thorbecke E. A class of decomposable poverty measures. Econometrica 1984;52(3):761-6.
11. Haddad L, Kanbur R. How serious is the neglect of intrahousehold inequality? Econ J 1990;100:866-81.
12. Ravallion M. Land-contingent poverty alleviation schemes. World Dev 1989;17:1223-33.
13. Selten R. The equity principle in economic behaviour. In: Gottinger HW, Leinfellner W. eds. Decision theory and social ethics. Dordrecht, the Netherlands: D. Reidel, 1978.
Contents - Next