Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Food and nutrition policy

Impact of dairy development programmes

B. Bowonder, B. Dasgupta, S. Gupta, and S. Prasad
Centre for Energy, Environment, and Technology, Administrative Staff College of India, Hyderabad, India

An article by Mogens Jul in volume 1, number 3, of the Food and Nutrition Bulletin (1979) indicated benefits for the poor from the Amul Dairy Development Scheme in Gujarat State, India; and an update by the same author was published in volume 7, number 2. However, there have continued to be doubts. This article has been accepted because it provides direct data to confirm the suggestions of the earlier articles. It leaves no doubt that well-designed and implemented dairy operations of this type can benefit medium-size as we/l as large farms, and also improve the income and diet of landless labourers.

INTRODUCTION

To assess the impact of dairy development on income and nutrition a detailed study was carried out in ten villages in three major states in India. Of these ten villages, five had extensive dairy development programmes. This paper explains the objective of the study, the methodology used, and the impact of dairy development as indicated by the results of the survey.

OBJECTIVES

The project was carried out mainly to examine the possible economic and nutritional impact of integrated dairy development programmes. We selected five control villages with no such programmes and five with government-sponsored programmes. In the first five villages, individual traders generally purchased milk from the households.

The study examined the following questions:

— whether dairy development has resulted in draining away milk from rural areas, or the effect of dairy development consumption on milk and milk products;
— whether dairy development has contributed positively to the improvement of the economic and nutritional status of persons in rural areas;
— whether the lower-income groups have received any positive benefits from dairy development in terms of food intake and income;
— whether there is any shortage of fodder in rural areas because of the lack of integration of dairy development with other economic activities needed to support dairying; and
— the major determinants of increase in income in rural households that arise out of dairy development programmes.

SELECTION OF THE VILLAGES

Five comparable control villages were selected. State-level data on milk procurement were obtained from the Dairy Development Corporation of Andhra Pradesh and Karnataka State. In the case of Maharashtra State, information was collected through non-governmental agencies working on dairy development. The five villages with dairy development programmes were selected by the following criteria:

— quantity of milk procured;
— prevalence of animal husbandry programmes (veterinary hospitals, cattle insemination programmes, animal vaccination centres);
— milk cattle distribution through government agencies;
— dairy development programmes (milk collection and chilling centres, fodder distribution programmes, mineral and feed distribution programmes, etc.);
— weaker-section development programmes (such as the Integrated Rural Development Project and self-employment schemes consisting of loans, assistance, etc.).

Table 1 gives the details of the ten villages selected in this study in terms of number of households, total population, distribution of farm holdings, and milk procurement in 1983 (May and October). As can be seen, five of the villages did not have any milk-procurement schemes. Table 2 gives the details of developmental programmes and table 3 the infrastructural facilities existing in the villages, as well as the staple foods of the households. The villages were so selected that they would form five pairs, each pair consisting of one village with dairy development and one without.

SURVEY METHODOLOGY

A detailed questionnaire was designed for conducting household surveys in the ten villages to collect data on food consumption. Five field investigators who had knowledge of the local language were selected from the area. The household-survey was carried out over two seasons, the lean months during summer and the flush months during rainy season. During the lean months (March to August) green fodder is not available and agricultural employment is low. During the flush months (September to February) the availability of green fodder increases, as does agricultural employment, In this survey, instead of sampling, 100 per cent coverage of all households was adopted.

TABLE 1. Characteristics of the 10 villages

ANALYSIS

Detailed analysis was carried out to examine the socioeconomic and nutritional differences between villages with and villages without dairy development programmes.

To examine the relationship between dairy development and economic status, three multiple-regression analyses were performed. The patterns of (i) consumption of milk and milk products, (ii) total food consumption, and (iii) income were analysed using stepwise multiple regression. The major factors affecting each of these values were included in the regression equation. Using a stepwise F ratio test, insignificant variables were rejected.

RESULTS

The 1,343 households consisted of four landholding classes. The landless group represents households having farms less than 0.04 ha per household; marginal farms are those with holdings from 0.041 to 0.99 ha; small farms have holdings from 1 ha to 1.99 ha; and large farms are farms greater than 2 ha.

All Households

The regression analysis for the three dependent variables was carried out separately for the two seasons.' During the lean season, consumption of milk and milk products correlated positively with milk yield per animal, consumption of processed pulses, and size of irrigated farm. Consumption correlated negatively with consumption of pulses, number of non-descript (ND) cows, milk sold to co-operatives, milk sold to individuals, and number of graded murah breed cows. The implication of this is that dairy development will have a positive effect on consumption of milk and milk products if milk yield is higher and if the number of non-descript cows is reduced.

In the case of income, for all households higher incomes were correlated with the irrigated farm size, number of nondescript (ND) buffaloes, and total land owned. Income correlated negatively with the number of non-descript cows. This clearly indicates that dairy development improves income since it substitutes higher-yield nondescript buffaloes for non-descript cows. The higher number of non-productive non-descript cows is a major constraint to development in rural areas since the marginal productivity of these cows is much lower than that of nondescript buffaloes.

TABLE 2. Developmental support in the 10 villages

The total food intake was a function of income and the number of ND buffaloes owned. This indicates that distribution of ND buffaloes under the dairy-development scheme has resulted in increased levels of income and increased levels of food intake.

For the flush season also, the consumption of milk and milk products correlated positively with the number of ND buffaloes, milk yield, and consumption of processed pulses. It correlated negatively with the number of local varieties of ND cows of low productivity. In the case of income, consumption was again correlated positively with irrigated land, number of ND buffaloes, milk sold to cooperative societies, and the number of hybrid cows of higher productivity. This clearly indicates that dairy development programmes have contributed to higher levels of income in households where ND buffaloes or hybrid cows have been distributed and also in households that sell surplus milk to co-operatives rather than individual commercial agents. Total food intake correlated positively with income, number of ND cows, milk yield, and total land owned, whereas it was negatively correlated with irrigated farm size, unirrigated farm size, and number of hybrid cows. During this season ND cows also contributed to the total food intake, since they provided income from their milk to supplement that gained from employment.

VILLAGES WITH AND WITHOUT DAIRY DEVELOPMENT

A disaggregated analysis of food intake in dairy and nondairy villages indicated the impact of dairy development. For this purpose, consumption data for all the households in dairy villages were pooled and compared with consumption data in all households in non-dairy villages. Table 4 gives mean food intake per household as well as monthly per capita consumption. During the lean months, total household food intake per capita per month was 23.6 kg in the villages with dairy development programmes, whereas it was only 20.3 kg in villages without dairy development. For the flush season (table 5) the figures were 21.4 and 18.9 kg respectively. The differences existed for cereals, non-vegetable foods, milk and milk products, and vegetables.

During the lean season, per capita milk consumption was 0.5 and 0.2 kg for dairy and non-dairy villages (tables 4 and 5), while for the flush season these figures were 0.5 and 0.2 kg. The consumption of cereals, non-vegetable foods, and vegetables (tables 4 and 5) showed a similar trend, indicating that the calorie intake as well as nutritional status of households in dairy villages is at a higher level.

Multiple-regression analysis was carried out for data of households in dairy villages. The milk and milk product consumption correlated positively with milk yield per animal, consumption of processed pulses, and size of irrigated farms, whereas as it correlated negatively with milk sold to cooperatives and commercial agents and number of ND cows. The average household income was dependent on number of ND buffaloes, irrigated land available, and total land owned. Income was inversely related to the number of ND cows owned. Here again, the distribution of ND buffaloes has resulted in higher incomes, Total food consumption for the lean season was related only to income. Other variables had no significant influence.

TABLE 3. Infrastructural facilities

For the season the consumption of milk and milk products correlated positively with number of ND buffaloes, milk yield, and consumption of pulses, and negatively with number of ND cows, milk sold to co-operatives and commercial agents, and consumption of processed pulses. With respect to income, results were similar to those for the lean season. Income correlated positively with irrigated land, number of ND buffaloes, total land owned, and number of hybrid cows, and negatively with number of ND cows. This again indicates that in dairy villages, distribution of ND buffaloes and hybrid cows has helped to increase income. As for total food intake, income, number of ND cows, milk yield, and farm size correlated positively. The number of ND cows correlated positively since with marginal input it increased the milk yield. The effect of income on food intake was very small.

TABLE 4. Monthly food intake in dairy and non-dairy villages: lean months (kg)

TABLE 5. Monthly food intake in dairy and non-dairy villages: flush months (kg)

Multiple-regression analysis was repeated for the households in non-dairy villages for the two seasons. The consumption of milk and milk products correlated positively with milk yield, consumption of processed pulses, irrigated land, and number of ND buffaloes. It was correlated negatively with consumption of pulses, number of non-descript cows, and farm size. The income was a positive function of total land owned, number of ND buffaloes, milk yield, and number of hybrid cows. Total food intake correlated positively with income, number of ND buffaloes, and number of hybrid cows, and negatively with land.

For the flush season, in households without dairy development the consumption of milk and milk products was a positive function of number of ND buffaloes, consumption of pulses, milk sold to commercial agents, and milk sold to co-operative societies. Income correlated positively with farm size, size of irrigated farms, number of Jersey cows, number of ND buffaloes, and milk sold to co-operatives. Total food intake correlated positively with income, number of ND cows, and number of ND buffaloes, and negatively with number of hybrid cows such as Holstein-Friesians and Jerseys. Total food intake did not increase in non-dairy villages even if hybrid cows were distributed, since there was no infrastructural support for maintaining the productivity of the animals.

These results must be examined keeping in mind factors pertaining to dairy development in the two types of villages. During the lean months, the average income may have been similar in both dairy and non-dairy households, but, as seen from tables 4 and 5, food intake was significantly higher in the households with dairy development programmes. Similarly, milk retained for own consumption, milk yield, number of ND buffaloes, and milk sold to cooperatives were also significantly higher in dairy villages. For the next season also, income, number of ND buffaloes, milk yield, and milk retained for own consumption were higher for households in dairy villages. Multiple-regression analysis indicates that number of ND buffaloes and milk yield had significant influence on household income. Hence dairy development resulted in contributing to the higher income.

In the non-dairy villages also, income and food intake were related to milk yield, number of ND buffaloes, number of ND cows, and irrigated land owned. These values were significantly lower in these households, except the number of ND cows. The number of ND cows owned also contributed to the higher levels of total food intake during the flush months in both dairy and non-dairy villages.

To summarize, as can be judged from the data for ten villages, dairy development resulted in higher levels of income, consumption of milk and milk products, and total food intake. Consumption of milk and milk products was high in households whenever the milk yields or the numbers of ND buffaloes were high. Both these are achieved through dairy development programmes. In the case of household income, irrigated land, number of ND buffaloes, and milk sold to cooperatives contributed positively to income increase. The number of ND cows had a negative influence on income, indicating that they act as a drain on the resources during the lean season. Whenever dairy development contributed to higher milk yield per animal, number of ND buffaloes, or improved marketing channels in the form of co-operatives, there was a positive influence on income. Total food intake was dependent on income and number of ND buffaloes during the lean season, whereas during the flush season numerous factors contributed, such as milk yield and the number of ND cows. For non-dairy villages, provision of more hybrid cows has not resulted in higher food intake.

Impact of Dairy Development on Various Income Levels

To identify whether dairy development has any positive influence on economic and nutritional benefits, food intake was disaggregated into four income groups based on land holding, and then analysed.

In the case of agricultural labourers and landless persons, monthly food intake per capita was 25.7 kg per household in dairy villages, and 18.5 kg for households in non-dairy villages, during the lean season (table 6). During the flush season, these figures were 20.5 kg and 19.6 kg respectively (table 7). This indicates that dairy development helped the landless population to stabilize its nutritional intake. Consumption of milk products was substantially higher in the households of villages with dairy development programmes. The consumption of pulses, non-vegetable foods, and vegetables was also significantly higher for these households.

In households having marginal farms, the total food intake was higher in households with dairy development 121.5 kg) than in those without (19.6 kg) (table 8). A similar trend was observed during the second season (table 9), 20.5 kg per month for dairy villages compared to 18.2 kg for nondairy villages.

In the case of small-farm owners, the total food intake was higher for dairy villages during the lean period, but during the flush season the food consumption values were close in the two types of villages (tables 10 and 11). In the case of large-farm owners, those in dairy villages consumed more food per capita in both seasons than those in non-dairy villages.

With respect to dairy development, in the landless category per capita income, number of ND buffaloes, milk yield, and milk retained per day were significantly higher for the dairy villages. For example, 0.03 kg of milk was retained for their own consumption by the landless population in dairy villages, against only 0.01 in non-dairy villages. Annual income was 933 rupees (dairy) and 802 rupees (non-dairy) per capita. The milk yield during the flush season for the landless population was 0.456 kg per day per household for dairy villages and only 0.159 kg for non-dairy villages. The difference was significant during the lean season; 0.235 kg and 0.063 kg per day. Dairy development helped to stabilize the milk yield during the lean months. A similar trend was observed for owners of marginal, small, and large farms. This clearly supports the hypothesis that dairy development resulted in higher milk yield, higher quantities of milk retained for household consumption, and higher income through sale of milk per household in all four landholding classes.

Multiple-regression analysis was carried for the four classes based on land holding size for dairy and non-dairy villages separately to identify factors that contributed to higher levels of economic and nutritional status.

Dairy Villages

During the lean months, for the landless class, milk and milk products consumption correlated positively with milk yield, number of ND buffaloes, and consumption of pulses, whereas it correlated negatively with milk sold to commercial agents and co-operations. Income correlated positively with number of ND buffaloes, and negatively with number of local breeds of cows and milk sold to commercial agents.

During the flush months, consumption of milk and milk products was not dependent on milk yield, indicating that higher milk yield contributes to higher consumption during the lean months only, when availability is poor. Income was positively correlated with the number of ND buffaloes and negatively with the number of ND cows. Total food intake was dependent on income and number of ND cows.

TABLE 6. Monthly food intake of landless: lean months (kg)

TABLE 7. Monthly food intake of landless: flush months (kg)

Continue

Contents - Previous - Next