Contents - Previous - Next

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

Unexpected benefit from a dairy project

Mogens Jut
Department of Food Preservation, Royal Veterinary and Agricultural University, Copenhagen, Denmark.
Chairman, United Nations University Task Force on Post-harvest Technology


One of the world's most successful dairy developments has been that of the AMUL dairy in Anand in the State of Gujarat, India. Its operation is based on daily milk collections from about 225,000 milk producers in the Kaira district. They are joined in co-operatives that together own the dairy in Anand, a very large and modern milk processing plant. These activities have resulted in considerable income and the generation of employment in the area. Experience with the Anand dairy was so encouraging that the Indian Government embarked on a project whereby the same successful system could be introduced in the milkshed areas for Bombay, Calcutta, Delhi, and Madras. The project started with the objective of furthering dairying in these areas and of organizing milk supplies and milk distribution for these four cities.

For assistance in financing the project, the Indian Government applied for UN assistance, especially from the World Food Programme (WFP). WFP was able to donate considerable amounts of dried skimmed milk and butter oil to the project. These constituents could be recombined in existing dairies in India, the resulting milk sold to the public, and the proceeds from the sale used for financing the building up of a dairy system including improvement of milk cows, the establishment of collection systems and dairies in the countryside, and construction of transport facilities to the cities and the distribution of dairy products to urban areas.

Before making any commitment, the UN reviewed the project and changed the objectives in order to place the major emphasis on the supply of milk to vulnerable groups in Indian cities. The following were the revised objectives:

1. to make milk available for vulnerable groups;
2. to satisfy consumers' needs and get producers a larger share of price;
3. to improve productivity of dairy farming;
4. to remove cattle from cities;
5. to form a basis for a national dairy industry.

The data presented below were collected after about six years of operation, partly in connexion with a UN evaluation team.

Milk Consumption in Cities

Because of the project's urban emphasis, it was appropriate to evaluate milk consumption among low-income urban groups. Fig. 1 suggests that in all four cities concerned, milk consumption is very low among the lowest-income segments. As the project had not met its specific objective in this regard, the matter needed further study.

Income distribution in the cities, particularly in Calcutta, is revealing (Table 1). Assuming a family size of six, with two wage earners per family, the lowest-income group constitutes about 25 per cent of the population. This group earns less than 150 rupees (about $US 21 ) per family per month, or five rupees per day. If approximately 60 per cent of the income is spent on food, the family food budget would be about 3 rupees per day.

Table 2 gives the food prices for the same period in Calcutta, showing the amount of energy (in kJ) in foods that can be purchased, per rupee. If a six-member family's food energy requirement is set at 50,000 kJ per day, only by buying wheat atta, the flour used for chapaties, will a family be able to obtain an energy intake as high as 47,000 kJ per day. Such a diet would be unreasonably monotonous, nutritionally unbalanced, and could not be prepared without some oil. Yet, it would still not meet energy requirements. If, in this case, the purchase of milk is proposed, the smallest unit, one-half litre of buffalo milk, would cost approximately 1 rupee, but would result in the daily energy intake per family being reduced to about 32,000 kJ, i.e., 64 per cent of requirements. Clearly, in this desperate situation, populations could not divert any of their very modest resources to the purchase of milk. Nevertheless, one quarter or more of the population live under these or worse conditions. Their only recourse is to resort to a reasonably balanced diet composed of the least expensive items available in the market, e.g., wheat atta, some cooking oil, and some legumes.

FIG. 1. Milk Intake Per Capita Per Day According to Household Income Group.

Reducing Milk Prices

An obvious conclusion might seem to be that milk prices should be reduced. However, Table 2 illustrates that the reduction would have to be very substantial for milk to be competitive with other so-called protective foods such as legumes or fish. The dairy system itself is designed to provide milk transport and distribution at the least possible cost. Therefore, any reduction in price would have to be achieved by lower payments to milk producers in the country, and the price would have to be reduced substantially, as that component of the price which represents distribution costs could not be reduced. Clearly, this would be to ask presumably poor rural producers to subsidize equally poor urban dwellers, but the price sacrifice would have to be so great that in all probability no milk would be collected. Besides, because milk consumption is much larger in the

TABLE 1. Monthly Income Distribution, Calcutta, 1970, in Rupees

No income 67.08%
Less than Rs 75 8.56%
Rs 75 - 1 50 7.42%
Rs 150 - 300 7 03%
Rs 300 - 500 4.02%
Rs 500 - 750 1.52%
Rs 750 - 1000 0.90%
Rs 1000 - 1500 0.48%
Rs 1500 - 2000 0.19%
More than Rs 2000 0.19%
Undetermined 2.62%

References to the data in the tables and figures are given in the report of the evaluation mission: UN/FAD World Food Programme, "Second Interim Evaluation of WFP-Assisted Project India 61B - Milk Marketing and Dairy Development." Eight rupees (Rs) = approximately $US 1. higher-income groups, this would mean a sacrifice in income for the rural poor to subsidize foods for better-off people in cities.

Free Distribution of Milk

It might be concluded that milk should be distributed free to the most vulnerable groups. Table 3 gives the prices for various foods that could be used in supplementary feeding programmes. The lowest amount of milk that could be considered for such distribution, i.e. a portion corresponding to 2,000 kJ in the form of double-toned milk, would cost 0.87 rupee per day. The population of the four cities here considered totals approximately 20 million. The economic situation is probably worse in Calcutta than in the other three, yet an estimate of 10 per cent of this population's falling within the lowest-income, very deprived groups, seems to be a conservative one.

Distributing a portion of milk to 2 million people per day would, in milk and distribution costs, amount to well over 1,000 million rupees annually, i.e., an expenditure considerably in excess of the total amount of resources expended for the first seven years of the project. Clearly, Indian public funds would not be sufficient for such an undertaking, and foreign aid would not likely be available. It could seriously be asked whether, in such a situation, milk would be the preferred food for free distribution. Table 3 suggests that such foods as the Indian multi-purpose food or bal ahar might serve an equally useful purpose at a much lower cost. These stable products are much easier and cheaper to distribute than is perishable, easily spilled milk.

TABLE 2. Lowest Observed Food Prices in Calcutta, 1970, and Calculated Energy in kJ

  Rs/kg kJ/kg KJ/Re
Rice (open market) 1.97 15,050 7,640
Rice (control) 1.26 15,050 11,940
Wheat atta (open market) 1.03 15,050 14,610
Wheat atta (control ) 0.96 15,050 15,675
Legumes 1.50 5,850 3,900
Mustard oil 5.12 37,000 7,230
Milk (buffalo) 2.02 2,720 1,350
Potatoes 0.65 2,930 4,500
Vegetables 0.70 1,000 1,430
Fish 2.35 4,200 1,790
Meat 2.83 8,700 3,090

TABLE 3. Prices of Various Foods, in Rupees

  2000 kJ 15g protein
Standard milk 0.97 0.67
Double-toned milk 0.87 0.37
Extruded vegetable food 0.61 0.50
Multi-purpose food 0.30 0.11
Rice 0.28 0.46
Home-prepared balanced food 0.25 0.25
Bal ahar 0.18 0.20
Wheat 0.15 0.21
Groundnut flour 0.14 0.06

TABLE 4. Food Intakes among Low-income Groups in Calcutta

  Energy (kcal) Protein (g) Vitamin A (IU) Calcium (mg)
Intake Requirement % Intake Requirement % Intake Requirement % Intake Requirement %
Pre-school child 500 1200 42 15 18 54 300 900 33 200 450 44
School child 850 1900 45 25 37 44 1000 3000 33 230 650 35
Adult 1200 2700 45 35 50 46 1500 3000 50 275 450 61

Mall nourishment

It could be argued that milk has a special nutritional value and that its content of calcium, vitamin A, and protein would justify the extra expenditure for an individual ac quiring milk. However, Table 4 and Figs. 2 and 3 suggest that the population groups concerned are so desperately undernourished that their first need is for food energy. Table 4 and Fig. 2 also suggest that the diet is reasonably balanced, because intakes low in energy are generally equally, but not more, deficient in protein, vitamin A, or calcium. Thus, these people do not need a better-balanced diet, they simply need more food.

Rural Areas

At this point in the study, some became quite disappointed with the performance of the project. However, it seemed necessary to consider also the effect of the project in rural areas. The populations in the milkshed areas concerned number about 12 million. Table 5 shows that the economic situation is just as bad for rural populations in India as for urban groups. Even if some concession is made for some considerable intake of home-grown foods where a person has access to any land at all, or to food donated, pilfered, etc., Table 5 indicates that rural people may be every bit as malnourished as their urban counterparts.

However, one of the objectives of the project was to organize milk collection in rural areas. India has very large herds of cows and buffaloes, but they are mainly used for their value as draft animals and for the benefits of their manure. Nevertheless, where an organized milk collection system Calcutta, 1969. is put in, the milk also becomes of some value. A nondescript local cow may produce some 250 litres of milk per year. The procurement price may well be 2 rupees or more. This means that having just one cow producing milk may add more than 50 per cent to the income of a person in the lowest-income bracket.

FIG. 2. Daily Per Capita Intake as Per Cent of Requirements,

Such economic improvements have, in fact, taken place in areas with organized milk collection. Thus, Table 6 shows that those persons in rural areas who adopted the new system of milk collection co operatives doubled their dairy income. Peculiarly enough, they have also, as the table indicates, doubled their non-dairy income. It appears that introducing organized dairying brings with it a number of other economic and social improvements.

Scale Bias

Much agricultural development has been criticized because it benefits mainly large agricultural producers. For this reason, it became highly relevant to consider whether this project had such scale bias. Table 7 suggests that, to a very large extent, the project benefits small farmers and even landless farm labourers. Out of 328 persons in the village, 122 landless farm labourers or small farmers had milk animals. In Kaira, Gujarat, 97 per cent of landless farm labourers and 74 per cent of farmers with less than 1 hectare had at least one milk animal. Corresponding figures for Ludhiana, Punjab, are 76 and 39 per cent; and for Madurai, Tamil Nadu, 34 and 43 per cent.

Effect on Nutrition

The above considerations indicate that Organized milk collection systems can double income in rural areas and that they also benefit the smallest producer and even the landless farm labourer. One may then question what effect this increased income has on nutrition.

Fig. 3 indicates that among very low-income rural groups, the income elasticity of the intake of food and energy and some other nutrients is rather high - approximately 0.2 to 0.3. This means that doubling income will result in a 30 per cent increase of food-energy intake and a similar increase in most other nutrients. This, in effect, means that the project could bring about a considerable nutritional improvement among the rural producers and their families, and those affected may total 12 million individuals. The project may thus well be the development project with the greatest impact on nutrition in India.

Food Supply Considerations

FIG 3. Daily Per Capita Intake as Per Cent of Requirements, Gujarat (1961 - 62)

In much of the project area, there were complaints that it was difficult to provide fodder to the dairy cattle, and that green feed products should be produced on suitable, i.e., normally irrigated land. The rule of thumb to many was that one third of an acre of irrigated land growing green fodder such as alfalfa should be set aside for each cow or buffalo. The reason is that an indigenous cow or a buffalo on good feed management, or a crossbred cow that could thrive only on efficient feed, will produce much more milk than a cow left to its own devices for feed. Table 8 shows that, in economic terms, it would be very profitable to obtain green fodder, and in addition, an adequate amount of concentrates for feed purposes. However, this could have a negative effect on food supply. Producing green fodder means that production of food grains on that area of land would be foregone. Also, the concentrates fed to animals would often be grains, oilseeds, etc., that could be used directly for various food preparations. Thus, when resources put into feed production, or into the feed directly, and the food value of the milk are all considered, the result is a calculation like the one presented in Table 9.

TABLE 5. Distribution of Disposable Personal Income in India, 1974

Disposable annual
income (Rs)
Per cent of households
up to 1,000 23.5 8.5 18.5
1,000 - 2,000 36.6 21.5 34.5
2,001 - 4,000 27.4 33.2 30.2
4,001 - 6,000 6.41 4.4 6.8
6,001 - 10,000 3.71 1.2 5.0
10,001 - 15,000 1.6 5.5 2.4
15,001 - 20,000 0.5 2.8 1.9
20,001 - 30,000 0.2 1.8 0.4
30,000 and up 0.1 1.1 0.3

TABLE 6. Annual Farm Income (in Rupees) - Madiad Village (Kaira District)

  Total Dairy %
(<5 acres) 1,511 1,151 78
(5 - 10 acres) 3,353 1,917 57
(>10 acres) 10,492 3,851 37
Average 4,097 1,985 48
(<7.5 acres) 752 453 60
(7.5 - 15 acres) 2,081 1,212 58
(>15 acres) 6,002 2,556 43
Average 1,192 675 57

This table suggests that an uncritical introduction of dairy farming in India could have a catastrophic effect. For example, India has 230 million head of cattle, including buffaloes. If each were to require one third of an acre planted with green fodder, 69 million hectares would be required, plus the land needed to produce feed concentrates. This figure should be compared with the present 75 million hectares planted with food grains in India. These calculations clearly indicate that wide introduction of dairying must be based on utilizing dairy cattle's ability as ruminants to use grass, straw, forage, and agricultural by-products such as sugarcane tops, rice bran, and the like for feed and convert it into a high-grade food, in this case milk.

In addition, India has many areas with favourable agricultural conditions. Here, intercropping of a food crop with a feed crop can be done, or, on irrigated land, two food crops could be raised, plus two feed crops in periods between the food crops, thereby obtaining feeds with no sacrifice of food production. India is fully aware of these considerations. Co operative dairying has probably advanced farthest in the State of Gujarat. Table 10 indicates that, combined with very large increases in milk collection, Gujarat achieved a remarkable increase in both food crops and other agricultural crops. Thus, dairy farming has not encroached in any way on production of other foods in this area.

TABLE 7. Households with Milk Animals, Village of Sabarkanta District in Gujarat

number of
Number having
milk animals
Landless labourers 82 22 (27%)
Small farmers 131 100 (76%)
Medium and large farmers 115 100 (87%)

TABLE 8. Annual Milk Production and Net Income Derivable from It in India

Indigenous cow on current feeding 200 175
Indigenous cow on improved feed 475 700
Crossbred cow on cultivated green    
fodder and concentrate feed 2500 1300
Exciting labour    


There is a general tendency to evaluate achievements of a development project on the basis of the original objectives. In this case, it was clear that the original objectives were not achievable. During the course of development, the project nonetheless managed to achieve some remarkably useful goals.

TABLE 9. Net Annual Food Caloric Balance of Cows' Milk Production in India (kJ)

  Input Output Balance
Indigenous cow fed on agri-cultural waste in addition
to wayside grazing
0 565 565
Indigenous cow      
1 kg concentrate feed/day during
lactation in addition to wayside
grazing, etc. 880 745 135
Crossbred cow
600 kg concentrate feed  
annually 4,800    
and   5,670 -10,830
crop from 1/6 ha land 11,700    

TABLE 10. Agricultural Development in Gujarat (in thousands of tons)

Year Total
grain crops
other crops
1955/56     11
1956/57     14
1957/58     21
1958/59 177 83 28
1959/60 145 90 23
1960/61 137 94 24
1961/62 188 77 35
1962/63 184 98 50
1963/64 205 132 62
1964/65 216 73 61
1965/66 215 73 66
1966/67 262 106 72
1967/68 317 141 81
1968/69 233 105 113
1969/70 414 171 124
1970/71     118
1971/72     133
1972/73     148
1973/74     112
1974/75     130


Contents - Previous - Next