Contents - Previous - Next

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

CALORIES AND PROTEIN SOURCES

An analysis of the sources of intake of calories and protein yielded an overall pattern of preferences (table 15):

First, the status or high-preference food category consisted of foods with some calories but little protein.

Second, for all docile groups, a significantly large share of both calorie and protein intake came from the medium-preference foods.

Third, basic foods do not constitute a major source of supply of either calories or protein. Similarly, low-preference foods are not the major source of calories or protein. Together, however, the basic and low-preference food groups account for a fairly large share of both calorie and protein intakes.

If we read tables 13 and 14 together, we find that basic foods account for a relatively low share of food expenditure compared to the share they contribute to either calorie or protein intake. For low-preference foods the percentages representing food expenditure and calorie and protein intake are nearly equal.

An inter-decile comparison of the contributions to calorie and protein intake by different preference categories of food showed that the contribution status foods made to calorie intake for both the D and ND groups increased rapidly as we moved up the docile groups. However, status foods made a relatively larger contribution to the calorie intake of the D households than to that of the ND households for all docile groups.

For the first to the third deciles, the contribution of status food to calorie intake for the D group was between 13 and 15 percent; a comparable level was not reached by the ND households until the seventh and eighth deciles. In contrast to status foods, the contribution of basic foods to calorie intake level as well as protein intake level declined for both the D and ND groups as we moved up the deciles. However, for D households the contribution of basic foods to calorie intake level was lower than that observed for the ND households, whose intake level from basic foods did not match the third and fourth decile levels of the D households until the sixth and seventh deciles.

TABLE 16. Nutrition component of expenditure on food (percentages)

NUTRITIONAL EFFICIENCY OF EXPENDITURES ON FOOD

The nutritional efficiency of food expenditure can be considered a mirror image of the preference component of food expenditure. When the quantities of different nutrients were multiplied by their shadow prices, the product was total nutritional worth, which can be viewed as the minimum cost of nutrition. The nutritional worth of food, when divided by the total market value of food, gave a measure of nutritional efficiency of food expenditure. Table 16 shows that for all decile groups the non-deficient households had a higher level of nutritional efficiency than the deficient ones. The differential was the largest in the ninth and the lowest in the fifth decile, in most deciles being between 10 and 12 percentage points. For every rupee the calorie-deficient households in the lowest decile spent, they obtained eight-tenths of the nutritional worth possible. In the lowest decile, the non-deficient households obtained about nine-tenths of the possible nutritional worth of their food expenditures

On the whole, the nutritional worth of food expenditures declined progressively as decile level increased for both deficient and non-deficient groups of households, at first slowly up to the fifth decile and very rapidly thereafter for both groups, although the decline was much sharper for the deficient groups. Illustrative of the differences between the D and ND groups was the fact that the efficiency level observed for the ND households in the eighth decile group was observed for deficient households in the second decile group.

These results imply that preference elasticity with respect to per capita household expenditure would be greater than nutritional elasticity with respect to total expenditures for both the D and ND households, although the preference elasticity will be higher and nutritional elasticity lower for the D than for the ND group.

GENERAL PATTERN OF CONSUMER BEHAVIOUR REGARDING CALORIE LEVELS AND CALORIE COSTS

Thus far, we have examined how levels of calorie intake, calorie costs, per capita expenditure levels, and levels of nutritional efficiency differed for two groups of consumers: those deficient in calories and those not deficient in calories. These differences can be considered special cases of a more general pattern of consumer behaviour regarding nutrition and taste. The general pattern would imply a negative relationship between calorie intake level and calorie costs regardless of whether the calorie intake level is below the recommended level or not.

We examined this general pattern of behaviour using the data presented in tables 17,18 and 19. Households were divided into ten decile groups on the basis of per capita expenditure levels. There were 110 households in each expenditure decile, with the exception of the top and bottom deciles, in which there were 115 and 111 households respectively. We further subdivided 110 households into ten decile groups on the basis of calorie intake per adult unit, and developed a matrix with 100 cells, containing about 11 households each. In each cell, we worked out household averages for la) calorie intake per adult unit per day and (b) calorie cost per 1,000 kcal (table 17). We further worked out averages for (a) per capita expenditure and (b) the share of total expenditure used for food (table 18). Table 19 gives the size of households in terms of both total persons and adult units. The composition of households is given in terms of a ratio of total persons to adult units, i.e. a composite variable.

Table 17 shows a consistently negative relationship between calorie intake level and calorie costs. Within each expenditure decile across the cells in the row we found a rapid increase in calorie intake levels. The range between the lowest and the highest was at least 1:2, and in some instances reached 1:3. This wide range in calorie intake levels persisted for all expenditure deciles. Another observation, equally interesting, was that across cells within each expenditure decile the calorie costs declined as calorie intake level rose. The range between the lowest and the highest calorie costs per 1,000 kcal was about 1:1.5 in most of the expenditure deciles; exceptions were the lowest and the two uppermost deciles, for which the range was even wider. Thus, in general, within a given expenditure decile calorie intake and calorie costs varied inversely, with a wider range of variation for the former than for the latter.

TABLE 17. Calorie intake and calorie costs

a. Calories in kcal per adult unit per day, costs in rupees per 1,000 kcal.

The persistent negative relationship between calorie intake levels and calorie costs suggested a more general pattern. Table 18 provides further evidence regarding this pattern: the total expenditure per capita was nearly invariant across calorie deciles for all expenditure deciles, the top and the bottom expenditure deciles being the only exceptions. At the same time, we found that as one moved from a lower to a higher calorie docile, the percentage of the total expenditure allotted to food usually increased as calorie level increased. The food share ranged from 60 to over 80 percent, and in most cases was over 77 percent. For a nearly invariant expenditure level within an expenditure docile, households at one end with relatively low calorie intakes had high calorie costs. They seemed to prefer expensive foods, but on the whole they spent relatively less on food. Moving along the calorie deciles, we observed that households (a) tended to spend relatively more on food out of a nearly equivalent total expenditure; (b) had higher levels of calorie intake; and (c) had relatively less expensive food-calorie sources. At one end of the spectrum we found quality-conscious households but, moving along the calorie deciles, we found a gradual transition to quantity-conscious households. Those who emphasized the quality of food seemed also to prefer a better standard of living in general and tended to spend relatively more on non-food items as well. The quality-conscious and quantity-conscious consumers of food together belonged to a continuum and represented a wide range of food preferences.

The matrices of calorie costs and calorie intake level were examined together with per capita expenditure and the share of total expenditure used for food. Moving along the columns of the matrices in tables 18 and 19, we observed that an increase in total expenditure (i.e. moving across the cells in a column} led to increased calorie intake and also to a shift in favour of more expensive foods. This general observation obtained for all columns, that is, for all calorie deciles. Within this general pattern two important observations were made. Households in the lower calorie deciles with a calorie intake lower than the recommended level did not seem to be "in a hurry" to achieve the target calorie level even when their per capita expenditure increased four- to five fold. Their calorie intake did increase, but not fast enough.

Calorie intake level also increased for households in higher calorie deciles almost as fast as it did for households in the lower calorie deciles. The former tended to add to their calorie intake, which was already above the recommended level. This pattern of behaviour has a disturbing policy implication. While there is general improvement in calorie intake as expenditure (or income) increases, a substantial number of households would continue to have a calorie intake below the recommended level despite a substantial increase in their level of per capita expenditure (or income).

TABLE 18. Consumption, expenditure, share of expenditure on food

a. Expenditure in rupees per capita per month food share as a percentage of expenditure.

This picture was alleviated by one feature: the increase in calorie costs or a shift in favour of expensive foods as per capita expenditure rose was relatively less rapid than the increase in calorie intake levels. However, for higher calorie deciles, the shift in calorie costs was more gradual than that observed for lower calorie deciles. The middle calorie deciles displayed a relatively gradual increase in both calorie intake and calorie costs.

continue

Contents - Previous - Next