Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

The evolution of Chile's main nutrition intervention programmes

Isabel Vial, Eugenia Muchnik, and Juliana Kain


As is the case in most countries, the Chilean government absorbs a high proportion of the cost of implementing health and nutrition intervention programmes. The aim is to provide minimum adequate levels of health and nutrition to low-income families by increasing consumption levels above those resulting from market demand and supply. Such programmes involve economic transfers to the beneficiaries that take the form of explicit or implicit subsidies.

This paper focuses exclusively on explicit food subsidies undertaken by government agencies that involve budget outlays. The analysis of the level of implicit food subsidies to consumers that results from farm-price interventions is less important in Chile than elsewhere because of the transition from general subsidies to targeted interventions that has occurred since 1974 [1].

Government efforts to improve human nutrition have a long history in Chile. In fact, the complementary feeding programme (CFP), the major food programme operating in the country, began in 1924. Since then, it has constituted a very important part of the Chilean welfare system, lasting through many different political regimes. It targets infants and preschool children (0-5 years of age) and pregnant and lactating women, totalling around 1.3 million beneficiaries at an annual cost (in 1990) of approximately US$50 million.

The school feeding programme (SFP) was initiated in 1963 for low-income schoolchildren, providing them with on-site food rations. In 1988, more than 517,000 children received complete benefits (break fast plus lunch) at an annual cost of approximately US$35 million.

The day-care centre food programme (DCCP), under the management of the National Board of Day Care Centres, provides on-site food rations to low-income infants and preschool children who attend these centres. In 1990 it covered over 60,000 children at an annual cost of approximately US$16 million.

Other small-scale food programmes are run by voluntary agencies, the Catholic Church, and community-based organizations. By investing systematically over a long period of time to enhance human capital formation, all programmes have been instrumental in reducing the prevalence of malnutrition, low birth weight, and infant mortality. In fact, Chile is one of the top three Latin American countries in terms of favourable indications of maternal and child health as well as the nutrition status of children. The rate of undernutrition has decreased substantially, from 15.9% in 1976 to 8.8% in 1982. Since then, it has fluctuated between 8.2% and 9.1% except in 1983, when it reached 9.8%. Severe undernutrition has virtually disappeared, so that most of the currently existing cases correspond to mild undernutrition [2].

The reduction in infant mortality has been even more dramatic, decreasing from 120.7 per 1,000 live births in 1954 to 65.2 in 1973 and 18.8 in 1988. According to Castaņeda [3], this trend has been the result of improved education and sanitary conditions, as well as participation by pregnant women in the CFP.

Data on birth weight also show that the percentages of low and insufficient birth weight (less than 2,500 g, and between 2,500 and 2,999 g respectively) have also declined, particularly the former. In 1988 the proportion of low birth weight was 7.2%, a decrease of around 40% since 1976. This could be assumed to be a direct consequence of the integrated nature (health and nutrition) of nutrition intervention programmes, reflected in the continuing increase in coverage of primary health activities, the increase in educational activities to enhance adequate use of existing health and nutrition services, and better targeting of public programmes to reach low-income households.


The complementary feeding programme (CFP)


The creation of the National Health System (NHS) in 1952 was fundamental to the expansion of the existing food-distribution programmes. This organization was conceived as a system that would provide integral health care, both preventive and curative, to blue-collar workers, indigents, and their respective families. As a result, the existing milk-distribution programme was integrated with the provision of health services, and expanded to also benefit children under 6 years of age, and pregnant and nursing mothers [4].

Preventing undernutrition among the most vulnerable groups and promoting health through periodic medical visits, immunizations, and education on the use and demand of services among low-income families were the main objectives of the CFP. The distribution of milk not only involved provision of free food but was also used as a device to attract beneficiaries to demand health and nutrition services. This was accomplished by channelling the distribution of milk through the infrastructure of the primary health care system, which presently includes 286 urban health clinics, 1,010 rural health outposts, and 1,281 rural health stations. In 1970 milk was distributed to the entire population of pregnant and nursing mothers and children under 12 years of age, regardless of income.

This untargeted programme was modified again in 1974, shifting from a general subsidy to one targeting the lowest income groups. This modification was achieved by restricting the number of potential beneficiaries by lowering the age limit for children from 12 to 6 and by tying the distribution of milk to beneficiaries who are regular participants in health examinations at the health centres of the NHS. This meant targeting food allocation to the clientele of the health system, namely low-income households, both urban and rural [5].

Modifications introduced to the CFP have been mainly in two directions. The first was increasing the programme's efficiency through better targeting of the benefits, decentralizing delivery of services to the local level (municipalities), and shifting the manufacture of the milk products over to the private sector.

The second included changing the type and quantity of milk products distributed. The most important changes in this direction have been: (1) replacing powdered milk (12% fat), previously delivered to all beneficiaries, to milk products containing 26% fat for infants and nursing mothers; (2) requiring industries that produce milk for the programme to sell the same product in the market, whereby the image of "a product for the poor" has disappeared; (3) introducing milk-cereal mixtures instead of powdered milk for children between 1 and 5 years of age, so as to increase the intake of vitamins and minerals, diminish food leakages within the family, and use cheaper raw materials (e.g. soybeans); (4) providing additional foods as a preventive measure (since 1980) to low-weight pregnant women, undernourished children, and those nutritionally at risk.

The concept of targeting at-risk groups for the supply of a larger food basket and more frequent health examinations changed in 1983 with the introduction of two criteria: Biomedical risk is determined by insufficient weight increases between medical visits; this is considered the most sensitive indicator to be used in the prevention of undernutrition. Socioeconomic risk is determined on the basis of family characteristics, the employment status of family members, and housing conditions [2].

The beneficiaries were assigned to one of two sub-programmes, the basic programme for healthy children or the reinforced programme for children with one or both risk conditions present. Because of budget constraints, only biomedical risk has been used as the criterion for assigning beneficiaries to either programme since 1985.

Table 1 shows the type, quantity, and nutritional contribution of food distributed in 1988 by each of the CFP subprogrammes. The basic programme contributes more than 100% of the protein requirements of beneficiaries up to I year old but substantially less for children 1 to 2 years old and for the preschool age group. The caloric contribution is quite low for all groups except infants up to 1 year old. It can be argued that for the children who are well nourished, the milk-distribution programme is used to attract them and their mothers to health controls.

A different situation is observed with the reinforced programme, which contributes more than 100% of the protein recommendations for beneficiaries up to 24 months of age and 59% for those of preschool age. The caloric contribution is around 100% for beneficiaries up to 12 months of age and 37% of the recommendation for preschoolers, which is 27% higher than the basic programme.

In terms of total milk distributed, the reinforced programme has increased over time relative to the basic programme. In fact, 30% of the food distributed in 1983 was under the reinforced programme, with this share increasing to 60% in 1984.

TABLE 1. Complementary feeding programmes: food supplements distributed and energy and protein contributions of basic programme (for normal population groups) and reinforced programme (for groups at biomedical risk), 1988


Foods (kg/beneficiary/month)



Powdered milk (26% fat)

Milk-cereal mixture

Powdered soup



% of RDA day


% of RDAg

Basic programme
0 5 mo 2 0 0 0 327 55 17.6 170
6-11 mo 2 0 0 0 327 38 17.6 102
12-23 mo 0 1 0 0 167 14 5.3 25
2-5yr 0 1 0 0 167 10 5.3 19
Healthy pregnant and nursing mothers 1 0 0 0 133 6 11.5 19
Reinforced programme
0-5 mo 2 0 1 1 591 98 23.7 180
6-11 mo 2 0 3 1 880 104 31.7 180
12-23 mo 0 2 3 2 1,007 86 26.9 120
2-5 yr 0 2 0 3 574 37 14.9 59
Pregnant women
Undernounshed 2 0 0 4 809   27.3  
Low weight gain 2 0 0 4 642   18.1  
Nursing mothers
Undernourished during pregnancy 0 0 0 2        

Source: Ministry of Health, Chile.
a. 1985 FAO/WHO/UNU recommendations.
b. Low-fat powdered milk.

Programme coverage

An important point to stress concerning the development of the programme since its inception is the fact that it has survived many different political regimes. This is due mainly to the fact that food distribution and the provision of health services were initiated to ameliorate malnutrition per se and were not perceived as "belonging" to the political platform of any specific party programme. In fact, researchers and university professors have participated throughout in the design and development of policies [6].

Table 2 shows the evolution of the quantity of milk distributed by the CFP from 1954 to 1988. The programme grew considerably in 1965, and again with the inclusion of the milk-cereal mixture in 1976. Because of the economic recession of 1981-1982 and the resulting pressures on fiscal expenditures, the authorities felt that food distribution could be reduced, especially given the fact that prevailing levels of undernutrition among children were low. However, this decision resulted in an increase in the rate of undernutrition during 1983, indicating the importance of such programmes. Food distributed by the CFP was subsequently increased over the 1982 levels.

TABLE 2. Quantity of milk distributed by the CFP (millions of kilograms), 1954-1988

  Milk   Milk   Milk
1954 1.4 1966 15.2 1978 22.2
1955 1.5 1967 20.2 1979 25.7
1956 2.0 1968 16.4 1980 25.2
1957 1.7 1969 14.8 1981 24.6
1958 2.9 1970 17.1 1982 28.8
1959 7.6 1971 19.0 1983 17.0
1960 8.4 1972 19.3 1984 27.8
1961 8.1 1973 20.3 1985 30.3
1962 8.6 1974 20.8 1986 32.1
1963 8.0 1975 23.6 1987 30.0
1964 7.3 1976 30.4 1988 38.6
1965 11.6 1977 30.8    

Source: Ministry of Health, Chile.
Since 1979 the quantity of milk delivered by the CFP has included the milk-cereal mixture.

National figures regarding coverage rates of preschool-age children, both urban and rural, according to income quintiles, are available for 1985 (table 3). As shown, coverage exceeded 80% in the three lowest quintiles of the urban population, increasing to around 90% in the rural sector [7]. The programme reached a high percentage of families in the highest income group in the rural areas (74%), but reached only 33% of high-income families in the urban sector.

Table 3. Urban and rural coverage of the CFP by income quintile, 1985

Quintile Urban Rural
Beneficiaries Non-benef Cartes Beneficiaries Non-benef Cartes
N % N % N % N %
1 276,251 91.7 24,888 8.3 82,081 95.1 4,241 4.9
2 219,516 86.1 35,292 13.9 66,900 94.2 4,111 5.8
3 161,933 80.1 40,301 19.9 45,965 88.1 6,231 11.9
4 100,536 68.0 47,304 32.0 31,565 85.5 5,353 14.5
5 40,452 32.7 83,027 67.3 16,812 74.1 5,888 25.9
Total 789 788 77 77 230 812 22 4 4 243.323 90.4 25,824 9.6

Socio-economic analysis of the CFP

Empirical evaluations of feeding programmes may take at least three forms [8]: (1) impact analysis, which establishes links between a project and certain objectives; (2) cost-effectiveness, which goes a step farther to compare alternative projects to determine the least costly method of achieving a predefined objective; and (3) cost-benefit analysis, which makes it possible to compare all project profitability.

Projects designed to provide free food have usually been evaluated using impact analysis or analysis of cost-effectiveness because of the difficulty using prices to value benefits. Nevertheless, conceptual and methodological developments during the last 15 years have made it possible to estimate prices of goods for which there are no formal markets. The CFP, being the most important programme in Chile, has undergone all three types of evaluation.


Impact analysis

The initial assumption of this study was that diminishing infant mortality is a desirable objective of social policy [3]. The analysis focused on different socio-economic policies with the purpose of evaluating their contribution to the observed decline in infant mortality.

Multiple regression analysis made it possible to quantify the impact various factors have had across regions and over time on reducing infant mortality from 1975 to 1982. The analysis led the researchers to the conclusion that one of the sources of the reduction in infant mortality was the decrease in regional differences in mortality rates. This took place in conjunction with a decrease in regional differences in birth rates, extended coverage of the CFP, a larger number of medical health examinations of infants, a larger number of health examinations by nurses, and greater urban provision of potable water and sewage disposal in urban areas.


Cost-effectiveness analyses

The impact of the CFP on the nutrition status of children was evaluated by comparing the outcome with that from an equivalent income transfer [6]. Two main research questions were addressed in this study: (1) to what degree is the food given to families consumed by the direct beneficiaries, and (2) to what degree have the economic benefits and nutritional benefits associated with the programme provided an improvement in the nutrition status of the population from a long-run perspective? The main objective was to provide quantitative estimates of the impact of the CFP.

Estimation of a caloric demand function was used to test the hypothesis that programme participation improves calorie consumption beyond the change induced by an equivalent income transfer, but this hypothesis could not be confirmed by the analysis. In addition, the study analysed the hypothesis that the nutrition status of the beneficiaries is significantly different from that of the rest of the population, once factors such as income, rural-urban location, sex, and age have been controlled. In terms of weight and height for age, no significant differences were discovered between CFP participants and non-participants.

Another set of regressions was estimated to determine calorie and protein distribution in the family of each of the three daily meals. Controlling for income, sex, urban-rural residence, family size, and calorie requirements by age-group, the results indicate that male adults (over 15 years of age) receive the largest share, followed by female adults, and finally by children. When food consumed between meals is included, nutrient distribution within the family is in accordance with requirements.

Finally, the weight and height of former participants between 14 and 30 years of age are significantly above those of non-participants, which was interpreted by the authors as a long-term effect of the programme.


Cost-benefit analysis

To determine the benefits of the CFP, Torche [8] examined the effects of feeding and regular health examinations on beneficiaries. Next, a model was built to relate food intake and health examinations to the nutrition status of the beneficiary. Finally, the benefits associated with changes in nutrition status were established.

Food intake was measured in terms of calories, protein, and other nutrients. Nutrition status was measured by weight and height indices with respect to age; given the existence of a health-nutrition synergism, any analysis that tries drastically to separate the impact of food intake from health examinations is extremely difficult. The study therefore analysed both effects of the CFP simultaneously. The statistical results indicated significant differences in food intake, health examinations, and nutrition status only in the last two quintiles, that is, in households of the lowest socioeconomic levels.

Changes in nutrition status are a form of investment in human capital that generates a stream of future benefits. The study considered improvement in birth weight, improvement in school performance and future productivity of infants, and changes in infant mortality and morbidity.

The recovery of children born with subnormal weight implies the use of medical resources. Saving those resources can be considered the lower bound estimate of the benefit derived from improving the food intake of mothers during pregnancy. The effects of the CFP are double: it diminishes the number of children with low birth weight, and it increases birth weights even if they remain below the range considered normal. Only the first of these two effects was considered.

To assess the benefits derived from better school performance, it was necessary to compare the behaviour of a "typical" child participant with a "typical" child non-participant. Better school performance eliminates one of the reasons students repeat grades. Thus, the economic value of this benefit is that fewer educational resources will be used per year of schooling provided. Another expected consequence is a decline in early school drop-outs. In economic terms, this represents a cost and a benefit simultaneously. The cost is the larger use of educational resources required for those who remain in school, and the postponement of entrance into the labour force. The potential benefit is the increase in labour productivity that will manifest itself once the individuals start working.

The survey did not provide conclusive results on morbidity among children subscribing to the CFP. Therefore, the savings in resources from a decrease in the number and intensity of sickness episodes was not quantified. But the effect of the programme on infant mortality was included when measuring the benefit from better school performance by correcting the income profiles to account for the survival probability at different ages for both groups, with and without CFP. Finally, the expected benefit in terms of greater adult productivity was not quantified.

Costs were calculated on the basis of the social value of products distributed by CFP and the social value of the health examinations provided for a typical child of the lowest income quintiles, from the fourth month of the mother's pregnancy until 6 years of age. In doing this, (1) the social value of the product distributed was estimated on the basis of their CIF prices; (2) the social value of health examinations was calculated on the basis of the marginal cost of providing the service; (3) a social discount rate of 17% per year was used; and (4) a lag of three months between product purchase and their consumption was assumed.

The evaluation is presented in terms of net present value. It was concluded that CFP produced a positive net social profit. It was also determined that returns to the programme could be improved by redistributing resources from the highest income quintiles toward pregnant women and to the lowest income quintiles.

In synthesis, the economic analysis of the CFP has established that this programme is a profitable investment from the social standpoint, and has a significant positive impact on the nutrition status of preschool children.

Contents - Previous - Next