Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

15

POVERTY AND CALORIE DEPRIVATION ACROSS SOCIO-ECONOMIC GROUPS IN RURAL INDIA: A DISAGGREGATED ANALYSIS Abha Gupta1 & Deepak K. Mishra2 This paper examines the linkages between calorie deprivation and poverty in rural India at a disaggregated level. It aims to explore the trends and pattern in levels of nutrient intake across social and economic groups. A spatial analysis at the state and NSS-region level unravels the spatial distribution of calorie deprivation in rural India. The gap between incidence of poverty and calorie deprivation has also been investigated. The paper also estimates the factors influencing calorie deprivation in rural India. The study point out that nutritional deprivation is high among marginalized social groups and regions. It is the poor, scheduled castes, scheduled tribes, illiterate people, agricultural labourers and Muslims who are more likely to be calorie deprived.

INTRODUCTION Notwithstanding India’s relatively robust economic performance since the economic reforms in early 1990’s, significant deficits in human development parameters, most notably in health and nutrition standards, remain a cause of concern. India has the largest number of under-nourished children in the world. Not only that prevalence of child under-nutrition in India (43 percent) much higher than the world average (25 percent), its performance is worse than some of the poorest economies of the world (World Food Programme 2009).This prevalence is even higher among some socio-economic groups and regions. One of the WHO’s millennium development goal is to reduce the number of stunted, wasted and underweight children by 2015. Only few years are left to achieve this goal but in India still 38.4 percent children under the age of 3 are stunted, 19.1 percent are wasted and 46 percent children are underweight (National Family Health Survey 2005-06). There has been a sluggish decline in this percentage over a decade but this decline is unimpressive when compared across states and different socio economic groups. Besides poor performance in terms of some anthropometric measures, average per capita per day calorie and protein intake is also showing a declining trend in the post economic reforms period. Consumption and expenditure on cereal food items, which are a good source of energy has recorded a decline whereas other food items (vegetables, fruits, meat/egg/fish, oil, milk) have shown a slightly increasing share in the diet of the population. However, decline in calories is not seen as deterioration of health by some researchers rather it is viewed as a sign of improvement resulted by an increase in income, development of rural infrastructure, mechanization, urbanization, improvement in health and change in taste and preferences (Deaton and Dreze 2009, 2010; Verma et al. 2008; Rao 2000). Another group of scholars, however, links this with the increasing deterioration in health and 1

Research Scholar, Centre for the Study of Regional Development, School of Social Sciences, Jawaharlal Nehru University, New Delhi-110067. E-mail: [email protected]

2

Associate Professor, Centre for the Study of Regional Development, School of Social Sciences, Jawaharlal Nehru University, New Delhi-110067. E-mail: [email protected].

16

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

nutrition standards of the population (Patnaik 2004, 2007, 2010; Nasurudeen et al. 2006; Ray 2005:10; Mehta and Venkatraman 2000; Shariff and Mallick 1999; Mehta 1982). India’s growth ‘turn around’ has not resulted in remarkable improvements in health and nutrition outcomes, and it has raised questions on the inclusiveness of the growth process (Radhakrishna et al. 2004). The high level of undernourishment among children (46 percent, National Family Health survey 2005), the relatively high infant mortality rate (47/000 live births, Sample Registration System 2010) and signs of distress among marginalized sections of the society in a country which has witnessed remarkable growth in recent decades has been a widely discussed issue (Dubey and Thorat 2012; Reddy and Mishra 2010). However, India’s poverty measured in terms of head count ratio, which is a measure based on minimum calorie norm, has seen consistent decline during this period of growth. This evidence of declining poverty is not accepted by all and it remains a contested question (Deaton and Dreze 2009, 2010; Patnaik 2007, 2010)1. The rising gap between official head-count ratio and share of population having less than minimum calorie intake that formed the basis of official poverty line has been a matter of wide public concern and debate (Dev 2005; Sen 2005; Jones and Sen 2001). This debate surrounds over the method of poverty measurement and the focus has been on whether the official poverty line is adequate to account for rising expenditure on health and education, which, until recently, were being provided by the state. Most of the studies on poverty deal with the level of rural and urban poverty at the all India and state level. This paper attempts to unravel these issues at a more disaggregated level- at the level of NSS (National Sample Survey) regions and also in terms of various socio-economic groups. The broad objectives of this paper are outlined as follows: 1)

To examine changes in consumption of different food items in order to explain changes in nutrition level.

2)

To estimate changes in level of nutrients and deficiency of different nutrients from the recommended dietary allowances (RDA) at disaggregated level and to show the gaps between levels of poverty and levels of nutrition deficiency.

3)

To estimate probability of being calorie deprived at disaggregated level using binary logistic regression analysis.

From the policy perspective, the results of this paper have important implications for both the methodology of poverty measurement and also for providing nutrition security to the vulnerable sections of the population. DATA AND METHODS Data for this paper are obtained from National Sample Survey (NSS), 50th (1993-94), 61st (200405) and 66th (2009-10) Consumer Expenditure Schedules. These rounds of the survey, by the NSS are large scale sample surveys and provide information on consumer expenditure quinquennially as part of its “rounds”. Consumer expenditure survey gives information on quantity and value of different goods in a household with a reference period of last 30 days for each state/UT, all India and separately for rural and urban areas. Among these goods, information on 142 items of food are collected which can be converted into nutrition values2.

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

17

In this paper, average per capita per day 2400 kcal has been used to show calorie deprivation which is also used by Planning Commission to indirectly estimate head-count ratio for rural areas3. For converting monthly household food consumption into per capita monthly consumption, monthly household consumption is divided by household size. To get the per capita per day consumption, per capita monthly consumption is divided by number 30. In order to show probability of being calorie deprived across socio-economic and demographic groups, a logit model has been fitted which is 1 1 + ݁ ି௫ P = 1/1+e-z..……………. (1) Where P is the estimated probability, z is the predictor variable and e is the base of natural logarithm with a value of 2.7183. After simplification, we get ܲ=

Log z = P/1-P…………… (2) Where (P/1-P) is called odds and log (P/1-P) is called log odds or the logit of P. Thus, equation (2) becomes logit P = Z…………….. (3) The multivariate logistic function involves ‘n’ predictor variables which is represented by P = (1/1+e-b0 + b1x1+b2x2 +……… bnxn) ………… (4) Or,

logit P = (bo + b1x1 + b2x2 +…… bnxn)…………. (5)

The coefficients b1 represents the additive effect of one unit change in the predictor variable x1 on the log odds of the response variable. Whereas one unit increase in the x1, holding other predictor variable constant, multiplies the odd by the factor eb1. For this reason the quantity eb1 called the odd ratio. RESULTS AND DISCUSSION Trends in Food Consumption in Rural India Food is one of the basic needs for human survival. The variety of food that we consume determines our nutrition behaviour in terms of calorie, protein, fat and other micronutrients. In rural India, cereals have been the main constituents in people’s diet. Among cereals, rice recorded an important share in total cereal consumption followed by wheat, coarse cereals, vegetables, milk and fruits (Table 1). During 1994-2005 the biggest decline was experienced by cereal consumption. This decline was caused by fall particularly in coarse cereal consumption followed by rice and wheat consumption. Pulse and milk consumption declined slightly. As far as change in consumption of ‘other food items’ (vegetables, fruits, meat and edible oil) were concerned, highest increase was found in vegetable consumption. Other food items recorded a slight increase in their consumption. A recent round of NSS (66th Consumer Expenditure Survey, 2009-10) shows that cereals still hold the highest place among all food items mainly because of higher rice consumption. However, cereal consumption still continues to decline but the decline has been lesser during 2005-10 compared to a decline during 1994-05. The consumption of wheat, rice and coarse cereals shows a marginal decline. As far as consumption of ‘other food items’ (Vegetables, fruits, meat and edible oil) is concerned, a marginal increase is seen in the consumption of these food items. From the analysis

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

18

above, it can be argued that last 15 years, often referred to as the ‘post economic reform period’, rural India experienced a sharp decline in cereal consumption particularly coarse cereals, although the precise linkages between economic reforms and calories deprivation needs to be examined further. However, in recent five years (2005-2010) this decline has been minimal. The consumption of other food items has been slightly increasing over the years but this increase is not compensated by decline in cereals, as a result of which calorie and protein intakes are falling. Table 1 Food Consumption Pattern and its Change in Rural India: 1994-2010 (Monthly Per Capita in kg*) Year Food Items

1993-94

2004-05

2009-10

Kg Change (1994-2005)

Kg Change (2005-2010)

13.40 4.32 6.79

12.11 4.19 6.38

11.35 4.34 6.13

-1.29 -0.13 -0.41

-0.76 0.15 -0.25

Coarse cereal

1.97

1.27

0.87

-0.70

-0.40

Pulses

0.76

0.71

0.66

-0.05

-0.05

Milk Liquid (litres)

3.94

3.87

4.08

-0.07

0.21

Vegetable Fruits Fruits (nos.) Meat Egg (nos.)

4.75 0.22 2.71 0.12 0.64

5.25 0.30 2.84 0.14 1.01

4.58 0.21 2.66 0.14 0.95

0.50 0.08 0.13 0.01 0.37

-0.67 -0.09 -0.18 0.00 -0.06

Fish

0.18

0.20

0.21

0.02

0.01

Cereal Wheat Rice

0.37 0.48 0.56 0.11 Edible Oil (litres) Source: Authors' calculation from NSS 50th, 61st and 66th Consumer Expenditure Schedule. Note: unit in kg unless otherwise specified in brackets after the food-item.

0.08

Change in Nutrient share of various Food Items and level of Poverty in Rural India It is believed that food consumption in India has changed much which has caused overall decline in calories. There are various factors which affect consumption of food items such as production, availability and prices, lower level of unemployment, rise in per capita expenditure, change in taste, climate, decline in physical activity, improvement in health status, urbanization, increased awareness among consumers about food nutrients, access to safe drinking water, health care and environmental hygiene for effective conversion of food into energy (Kumar et al. 2007; WHO 2003; Bansil 2003; Viswanathan 2001; Martorell and Ho 1984). A group of scholars considers this decline in calories as a positive and anticipated development and for them this decline is not a matter of serious concern (Radhakrishna 2005; Radhakrishna and Reddy 2004; Rao 2000). On the other hand, Patnaik (2007) has argued that decline in calories leads to deterioration in health and poverty and blames Planning Commission for using faulty prices to adjust poverty in India as the reason for artificially lowering the estimates of poverty. The average per capita per day (PCPD) calorie consumption declined from 2148 kcal to 2044 kcal between 1993/94 to 2004/05 in rural India. On an average PCPD intake of protein also recorded a fall from 59.9 gm to 55.1 gm during the same period (Table 2).

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

19

Table 2 Change in share of nutrients from different food items between 1993/94-2004/05 in rural India Food Groups

Average Per capita per day intake of Calorie (kcal) Calorie 1993-94 2004-05 Change 809 755 -55 500 487 -13 220 140 -80

Average Per capita per day intake of Protein (gm) Protein 1993-94 2004-05 Change 17.5 16.3 -1.2 17.7 17.2 -0.4 6.6 4.3 -2.3

Rice Wheat Coarse cereals Cereals and cereal 1530 1382 -147 41.8 37.9 substitutes 57 60 3 1.0 1.1 Root and Tubers 103 98 -5 0.0 0.0 Sugar and honey Pulses, nuts and 106 92 -14 6.5 5.2 oilseeds Vegetables and 44 53 10 1.9 1.7 fruits 15 16 1 2.2 2.3 Meat, eggs and fish Milk and milk 132 131 -1 5.3 5.3 products 115 151 36 Oils and fats Misc. food, food 47 61 14 1.1 1.5 products and beverages 2148 2044 -104 59.9 55.1 Total Source: Authors' calculation from NSS 50th and 61st Consumer Expenditure schedule.

-4.0 0.1 0.0 -1.3 -0.2 0.1 0.0

0.4 -4.9

As it has already been pointed out a sharp decline in cereal consumption and a slow rise in consumption of other food items is observed from the analysis of secondary data. Table 2 clearly shows that calorie decline has been accompanied by a decline in protein intake. The main reason for this decline is fall in cereal calories particularly coarse cereals and pulse intake. Consumption of oil and fat contributed in total calories but these food items are lacking in protein and are rich in fat. As a result, all-India average fat intake has increased (Nutrition Intake, NSS 61st round report). Besides oil & fat, miscellaneous food and beverages also contributed much in calorie and protein consumption. Before discussing calorie deprivation and poverty at disaggregated level, it would be appropriate first to talk about the trends at rural all-India level, which helps in understanding the general situation of the poverty. The levels of calorie deprivation and poverty in rural India, as presented in Table 3, shows that around 72 percent rural population was not getting required calories (per capita per day intake of 2400 Kcal) during 1993-94 and this percent has risen to 80, an increase of 8.4 percentage points in 2004-05, whereas level of poverty has declined if we consider Planning Commission’s estimate accurate. In 1993-94, the level of poverty was 37 percent which has declined to 28.3 percent in 2004-05. The gap between calorie poverty level and planning commission’s poverty level has increased from 35 percentage points in 1993-94 to 52 percentage points in 2004-05, a 17.1 points increase. This mismatch between poverty and calorie intake continues to remain a contested issue among researchers.

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

20

Table 3 Change in Calorie Deprivation and Poverty Level in Rural India between 1993/94 and 2004/05 Method of estimating poverty Percent Below 2400 Kcal Percent Below Official Poverty Line Gap between Calorie Poverty and Official Poverty Line

1993-94

2004-05

Change between 1993/94 & 2004/05

71.60

80.0

8.40

37

28.3

-8.7

34.6

51.7

17.1

Source: Same as Table 2.

Change in level of Nutrients at disaggregated level The Planning Commission of India has officially taken recommended calories4 of 2400 Kcal PCPD for rural and 2100 Kcal PCPD for urban areas in order to estimate poverty5. Besides, 60 gms PCPD protein intake has also been recommended by ICMR for nutrition measurement4. Table 4 presents average PCPD intake of calories and protein and their change over a decade (1993/942004/05) with emphasis on deficit from RDA across various sections of the society. From a demographic point of view it is found that never married persons consume lower level of calories and protein than the married persons. In fact, this demographic group also shows highest decline in nutrition parameters whereas widow/divorced/separated group enjoys relatively better access to nutrition. Deficiency of calories is highest among never married persons showing 305 kcal deficiency in 1993/94 which increased to 400 kcal during 2004/05. On the other hand are widow/divorced/separated group whose calorie deficiency is much lower than other marital groups. As far as deficiency of protein among marital groups is concerned, it has been higher among never married persons than married. In rural India, different social classes show distinct nutrition level from one another. If we analyze family size, it is found that it is the bigger households who are suffering from lower level of nutrition. In fact as size of a family increases, deficiency of calories and protein from recommended tends to rise. Family consisting of 7-8 members showed a higher increase in deficiency of calories than smaller families. In fact protein intake is quite low in these families. Small families (1-4 members) tended to show much lower fall of calories and protein than other family sizes. Similarly lower consumption of nutrients is found among less educated persons and as education level rises, average calorie and protein intake also increases. Less educated persons show a major decline in their nutrition level. Protein deficiency was much high in this group. On the other hand are higher educated people who recorded an addition of 117 kcal in 1993/94 and lower deficit of 17 kcal during 2004/05. This group added more protein in their diet in both periods. As far as religious groups are concerned, deficiency of nutrients is high among Muslims and Christians. Least deficiency of calorie and protein was shown by ‘other’ religious people as only 223 kcal were lesser than recommendation.

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

21

Table 4 Change in level of nutrients and deficiency from recommendation between 1993/94-2004/05 among socio-economic and demographic groups in rural India Calorie Intake

Marital Status Never married Married Widowed/divorced/ separated Household Size 1-4 5-6 7-8 Above 8 Education Group Not Literate Primary or below Secondary Higher Religious Group Hindu Muslim Christian Others Social Group Scheduled Tribe Scheduled Caste Others MPCE Groups (Percentile) Lowest 5 10 20 30 40 50 60 70 80 90 95 Highest Poverty Line Below poverty Line Above Poverty Line Occupation Type Self empl in non agr Agricultural Labour Other Labour Self empl in agri Others

Deficit from RDA, 2400 Kcal 1993200494 05

199394

200405

2095 2194

2000 2081

305 206

2236

2129

2312 2088 2070 2091

Protein Intake

Deficit from RDA, 60 gm 1993200494 05

199394

200405

400 319

59 61

54 56

1 +1

6 4

164

271

61

56

+1

4

2199 2005 1954 1955

88 312 330 309

201 395 446 445

63 58 58 60

57 54 54 55

+3 2 2 0

3 6 6 5

2089 2162 2332 2517

1974 2031 2184 2383

311 238 68 +117

427 369 217 17

59 60 64 70

54 55 58 65

1 0 +4 +10

6 5 2 +5

2159 2041 1989 2307

2048 1979 2075 2177

241 359 411 93

352 421 325 223

60 57 52 69

55 53 53 62

0 3 8 +9

5 7 7 +2

1993 2023 2212

1895 1948 2097

407 377 188

505 452 304

54 57 62

49 53 57

6 3 +2

11 7 3

1324 1581 1717 1846 1964 2043 2150 2264 2405 2586 2798 3253

1369 1571 1676 1796 1881 1958 2038 2154 2287 2378 2570 3034

1076 819 683 554 436 357 250 136 +5 +186 +398 +853

1031 829 724 604 519 442 362 246 113 22 +170 +634

38 44 48 51 54 56 60 63 67 73 80 92

36 42 45 49 51 52 55 58 61 65 71 82

22 16 12 9 6 4 0 +3 +7 +13 +20 +32

24 18 15 11 9 8 5 2 +1 +5 +11 +22

1737 2388

1639 2202

663 12

762 198

48 67

44 59

12 +7

16 1

2076 1923 1958 2347 2233

2042 1849 1892 2181 2169

324 477 442 53 167

358 551 508 220 231

57 52 54 67 62

55 48 51 60 58

3 8 6 +7 +2

5 12 9 0 2

22

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

Source: Same as Table 2 Notes: Intakes are Average per capita per day in Kcal and metric Gram respectively.

This religious group, on an average, added average 2 gm protein in their diet. In rural India, Scheduled tribes (ST) and Scheduled castes (SC) are worst affected as both social groups show lower intake of calorie as well as protein and also higher decline in nutrient intake compared to other class people. The worst affected are the ST people who recorded highest level of calorie and protein deficiency followed by SC in 2004-05. Calorie and protein deficiency had been lower among 'other' social groups. In terms of expenditure classes, it is found that it is the higher consumption expenditure groups who are consuming sufficient calories and protein. The bottom classes suffer badly from lower nutrient intake as well as its sharp decline. As consumption expenditure level rises, there is more probability of consuming sufficient calories and protein. The top 20 percent showed higher intake of calorie and protein and bottom 30 percent experienced as much as more than 500 kcal and 11 gm calorie and protein deficiency respectively during 2004/05. In terms of occupation groups in rural areas, it is found that it is the agricultural labourers and ‘other’ labourers among which calorie and protein intake is quite low and in fact these occupation groups also show a sharp decline in nutrient intake over a decade. Agricultural labour and ‘other’ labourers are worst affected occupation groups as both these groups had been unable to consume recommended intake of calories and protein. The deficiency in the level of nutrients is much higher among agricultural labour followed by ‘other’ labourers during 2004/05. Self employed in agriculture enjoyed better level of nutrient intake as deficiency of calorie and protein was quite low in the same period. Thus, from the above discussion it is found that there is significant relation between lower nutrient consumption and socio-economic marginalization and deprivation. Never married persons, less educated, lower Monthly per capita expenditure (MPCE) classes, SC, ST, Muslims, Agriculture and ‘other’ labourers, big households are those sections of the society where nutrient intake is quite low and at the same time decline in nutrient intake is considerably high among these groups. Thus, the disaggregated picture of nutrition deficiency does not fit well with the argument that the observed decline in calorie intake could be attributed to the diversity in the food basket of the people as result of broader changes associated with economic development. Level of Calorie Deprivation and Poverty For reasons discussed above, methods of poverty estimation have been a widely discussed issue. Even though the poverty line ensured the consumption of the normative calorie intake in 1973-74, the rupee value of the poverty line at current prices is not sufficient for meeting the normative requirements after other essential expenditures are taken into account (Sen 2005). As against this, some scholars most notably Patnaik, have argued in favour of a ‘nutrition-invariant’ or ‘direct’ poverty estimate, by calculating the number of people not consuming the recommended daily calorie intake. Some studies criticize direct method of poverty measurement through calorie and deprivation (Deaton and Dreze, 2009; Verma et al. 2008; Dev 2005; Sen 2005; Rao, 2000). They have highlighted the absurd results that it throws up when state level poverty estimates are carried out. While the calorie-based approach has been termed as 'calorie fundamentalism' and has been criticized for its narrow focus, the official poverty line based approach has been criticized for being inconsistent with figures of calorie deprivation and malnutrition. One way of moving ahead

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

23

is to carry forward this comparison between percentage of population not having minimum calories (on which the poverty line was based) and the official poverty estimates to a more disaggregated level. This is what we have attempted here. Table 5 Level of Calorie Deprivation and Poverty (Percentage) among Socio-Economic & Demographic groups during 2004/05 Socio-Economic Demographic Groups Calorie deprivation Population Below Poverty line Marital Status 82.30 31.3 Never Married 78.10 25.4 Currently Married 75.40 25.2 Widow/Divorced/Separated Household Size 70.60 17.0 1-4 82.30 29.1 5-6 85.30 37.4 7-8 85.80 36.4 Above 8 Social Group 88.50 47.6 Scheduled Tribe 85.10 36.8 Scheduled Caste 77.10 22.7 Others Religious Group 79.70 28.9 Hindu 84.40 29.3 Muslim 80.90 16.2 Christian 69.70 15.2 Others Education Group 83.50 36.5 Not Literate 81.10 27.1 Primary or below 72.60 14.7 Secondary 59.70 5.0 Graduate or above MPCE Groups (Rs.) 99.70 100.0 0-235 99.00 100.0 235-270 98.40 100.0 270-320 95.90 80.9 320-365(poverty line Rs.356.30) 92.70 Nil 365-410 89.30 Nil 410-455 83.80 Nil 455-510 77.20 Nil 510-580 67.60 Nil 580-690 57.40 Nil 690-890 43.00 Nil 890-1155 32.80 Nil 1155 & more Occupation Type 81.60 23.5 Self employed in non agriculture 88.90 46.4 Agricultural Labour 87.40 30.4 Other Labour 73.10 21.5 Self employed in agriculture 73.80 14.0 Others Source: Authors' calculation from NSS 61st Consumer Expenditure Schedule

Table 5 clearly shows that during 2004-05 among all groups where calorie deprivation level is high, poverty level has also been higher. This analysis is based on gross effects and hence no causalities are implied. It is found that never married persons report both relatively higher levels

24

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

of poverty and calorie deprivation compared to their group categories. In case of family size, bigger the household higher is the level of calorie deprivation and poverty. Small households covering 1-4 members experience lowest poverty and calorie deprivation level. Bigger households (more than 7 members) perform worse on both counts. As far as social groups are concerned, it is found that lower social groups such as ST and SC tend to have higher concentration of poverty and calorie deprivation level, whereas the reverse is true for the 'other social group'. STs are worst affected as poverty and calorie deprivation level is highest among them, followed by the SCs. If we see deprivation and poverty level among the religious groups, we find that particularly Muslims are in a worse condition as both calorie deprivation (84.4 percent) and poverty level (33 percent) are much higher among them in comparison to others. Education wise analysis shows that it is the lower educated persons who are living in poverty and consuming lower calories than standard norm. Higher is the education level lower is the levels of poverty and hunger. Illiterate persons experience a highest level of poverty (36.5 percent) and calorie deprivation (83.5 percent) level while educated people (with graduation and above) recorded lowest level of poverty (5 percent) and calorie deprivation (59.7 percent) level. Similarly, lower the MPCE class, higher is the level of poverty and calorie deprivation. Thus, bottom MPCE classes are unable to feed themselves even the standard calories and are living in poverty. In terms of occupation groups, agricultural labourers perform worst on both counts followed by ‘other’ labourer. Thus, while the official poverty measures and calorie deprivation might show different levels of deprivation, there is a close correspondence among the two so far as the pattern of deprivation across different groups are concerned. INTERSTATE AND REGIONAL ANALYSIS Inter-state variations in levels of deprivation has been one of the persistent themes in the poverty debate in India (Deaton and Dreze 2010; Patnaik, 2007; Dev 2005). Specific to the divergence between poverty estimates and calorie deprivation is the wide difference between the two estimates in India's southern states. Many of the southern states have better human development, demographic and social development indicators, and the records of state interventions in the areas of food security, primary education and affirmative action in favour of the weaker sections are generally considered to be better in most, if not all states of south India, particularly in comparison with the densely populated north Indian states. In this backdrop, the fact that southern states generally have a lower incidence of consumption poverty but a relatively higher degree of caloriedeprivation has been an important issue in the discussion. Patnaik (2007) views poverty as being underestimated in southern states, whereas Dev (2005) argues that poverty using calorie norm in southern states give absurd results. Deaton and Dreze (2009) criticizes calorie norm as poverty method as this norm places all southern states at higher deprivation level despite a fact that these states perform better in some anthropometric measures. The incompatibility of the poverty estimates and levels of calorie deprivation is brought out sharply in Table 6. The discussion here has been widened by incorporating two additional indicators of deprivation and it is important to note that southern states particularly Karnataka, Tamil Nadu, Andhra Pradesh rank high on more than two deprivation indicators which confirm their poor performance on selected deprivation indicators. For example, Karnataka ranks 10th in poverty level, 21st in

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

25

calorie deprivation, 12th in children underweight and 13th in BMI of women. Similarly, Tamil Nadu ranks 12th in poverty level, 19th in calorie deprivation, and 12th in BMI of women. Performance of Andhra Pradesh in terms of deprivation indicators is 13th in calorie deprivation, 7th in children underweight and 11th in BMI. Kerala is the only state in southern region which perform better in all deprivation indicators. Maharashtra however record better performance in terms of anthropometric measures but poverty (14th) and calorie deprivation (18th) level is high in this state. On the other hand, Punjab, Himachal Pradesh, Jammu and Kashmir and Haryana are best performing states in all deprivation measures whereas worst performance is shown by Jharkhand, Madhya Pradesh, West Bengal, Orissa, Chhattisgarh and Bihar (Fig. 1). At the state level, a correlation among the different indicators of deprivation is low6. Table 6 Performance of States on selected Deprivation Indicators and their ranking during 2004-05 States

Below poverty Line*

Below 2400 Kcal*

Children (< 3) Under weight#

BMI below normal (Women)#

Jammu & 4.3 (1) 65.5 (1) 31.6 (2) 26.1 (6) Kashmir 9 ( 2) 68.4 (4) 29.9 (1) 14.5 (3) Punjab 10.5 (3) 83.8 (13) 40.4 (7) 37.5 (11) Andhra Pradesh 10.5 (4) 66.3 (2) 36.4 (5) 25.8 (5) Himachal Pradesh Arunachal 10.9 (5) 70.9 (5) 42.1 (11) 14.3 (1) Pradesh 13.2 (6) 67.6 (3) 41.8 (10) 32.5 (8) Haryana 13.2 (7) 75.4 (9) 31.9 (3) 14.3 (2) Kerala 18.3 (8) 74.5 (7) 45.9 (14) 36.5 (9) Rajasthan 18.9 (9) 84.8 (15) 50 (17) 41.9 (15) Gujarat 20.7 (10) 89 (21) 45.1 (12) 38.2 (13) Karnataka 22.1 (11) 85.4 (16) 41.1 (9) 39.5 (14) Assam 23 (12) 87.3 (19) 34.8 (4) 37.5 (12) Tamil Nadu 28.4 (13) 78.1 (10) 46.7 (15) 44.9 (18) West Bengal 29.6 (14) 86.9 (18) 40.1 (6) 15.4 (4) Maharashtra 33.3 (15) 73.3 (6) 49.4 (16) 37.2 (10) Uttar Pradesh 36.8 (16) 87.5 (20) 62.6 (20) 44.2 (17) Madhya Pradesh 40.6 (17) 74.5 (8) 40.8 (8) 30.8 (7) Uttaranchal 40.8 (18) 84 (14) 54.6 (18) 45.7 (19) Chhattisgarh 42.6 (19) 78.8 (12) 59.3 (19) 45.9 (20) Bihar 46.2 (20) 85.7 (17) 63.1 (21) 47.8 (21) Jharkhand 46.9 (21) 78.5 (11) 45.7 (13) 43.7 (16) Orissa Source: * Same as Table 5, # Computed from National Family Health Survey, Fact Sheets, 2005-06

The level of nutrition (Table 7) in terms of calorie and protein intake across all major states show that Karnataka, Tamil Nadu, Andhra Pradesh, Gujarat and Maharashtra are the states where calorie and protein intake is quite low and in fact these states also show maximum decline in both the nutrients between 1993-94 and 2004-5. The deficiency of calorie and protein from recommendation is quite high in all southern states. During 2004-05 deficiency of calorie was high in Andhra Pradesh (409 kcal), Gujarat (501 kcal), Karnataka (538 kcal), Madhya Pradesh (472 Kcal), Maharashtra (476 kcal) and Tamil Nadu (536 kcal). In fact deficiency of protein was also larger in these states such as Andhra Pradesh (13 gm), Assam (10 gm), Gujarat (9 gm), Karnataka (13 gm) Kerala (7 gm), Maharashtra (8 gm), Tamil Nadu (16 gm) and West Bengal (10 gm).

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

26

This above analysis shows that calories and protein deprivations are consistently high in all southern states except Kerala, whereas there are some states like Punjab, Himachal Pradesh, Jammu and Kashmir, Haryana, Uttar Pradesh and Rajasthan where calorie intake recorded a slight decline and consumption of protein is increasing during the period under consideration. In fact states showing lower level of calorie deficiency (such as Haryana, Himachal Pradesh, Jammu and Kashmir and Punjab) have performed better during 2004/05 and they have also recorded a larger increase of calorie and protein in diet during the period under consideration. Table 7 Change in level of nutrients and deficiency from recommendation between 1993/94-2004/05 across all major states in rural India Calorie Intake States Andhra Pradesh Arunachal Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Total Source: Same as Table 2. Notes: Same as Table 4

199394 2044 2126 1983 2113 1989 2486 2322 2504 2067 1956 2158 1933 2197 2414 2461 1872 2303 2210 2148

200405 1991 2316 2055 2021 1899 2212 2314 2358 1862 2113 1928 1924 2008 2219 2157 1865 2195 2065 2044

Deficit from RDA, 2400 Kcal 1993200494 05 356 409 274 84 417 345 287 379 411 501 +86 188 78 86 +104 42 333 538 444 288 242 472 467 476 203 392 +14 181 +61 243 528 536 97 205 190 335 252 356

Protein Intake 199394 50.3 61.3 49.5 60.1 55.3 78.2 70.4 75.3 54.7 50.2 62.6 54.7 52.6 74.6 78.9 46.1 70.3 54.7 59.9

200405 47.4 59.4 50.4 54.9 50.5 67.8 67.0 62.4 47.0 53.4 53.9 51.8 46.2 64.5 67.1 43.9 64.2 50.5 55.1

Deficit from RDA, 60 gm 1993200494 05 10 13 +1 1 10 10 0 5 5 9 +18 +8 +10 +7 +15 +2 5 13 10 7 +3 6 5 8 7 14 +15 +4 +19 +7 14 16 +10 +4 5 10 0 5

A state level analysis may hide the micro level variations in calorie deprivation. There is some heterogeneity within the states so far as nutrition deficiency is concerned. Hence, an analysis has also been performed at NSS region level (Fig. 2) which tries to identify the regions experiencing calorie deprivation. Out of selected 72 NSS regions, 48 regions experience higher level of nutrition deficiency (more than 80 percent). The worst performance is shown by regions of Madhya Pradesh which include Vindhyan and south western parts. Dry areas of Gujarat also exhibit higher nutrition deficiency. Coastal parts of Maharashtra and southern parts of Orissa show more than 92 percent population to be calorie deprived. The performance of regions of southern states also does not pose a better picture.

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

Fig. 1

27

28

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

Fig. 2

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

29

Inland northern parts of Karnataka, coastal northern Tamil Nadu and south-western Andhra Pradesh experiencing much higher level of nutrition deficiency which may be one of the reasons of poor performance of southern states on deprivation indicators. It is clear from the figure (Fig. 2) that the regions in south India where level of calorie deprivation is relatively high form a contiguous belt. The regions which pose a picture of relatively better nutrition sufficiency include northern and southern parts of Punjab, Himachal Pradesh, western plains of West Bengal, Jhelum Valley and mountainous parts of Jammu and Kashmir, central and western Uttar Pradesh. Table 8 Logistic Regression Analysis for Showing Probability of Getting Required Calories Variables Social Group

Religious group

Education Level

Marital Status

Household Size

Occupation Type

Poverty Line Group

Regions

Variable Categories Others (Ref)^ Scheduled Tribe Scheduled Caste Hindu (Ref)^ Muslim Christian Others Primary or below (Ref)^ Not Literate Secondary Graduate or above Currently Married (Ref)^ Never Married Widow/Divorced/Separated 1-4 (Ref)^ 5-6 7-8 Above 8 Self employed in non agriculture (Ref)^ Agricultural Labour Other Labour Self employed in agriculture Others Above Poverty Line (Ref)^ Below poverty Line Central (Ref)^ North East North East West South

Beta

0.682 0.924 1.147

Sig.@ 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

1

0.154 0.282

0.000 0.000

1.166 1.326

-0.524

0.000

0.592

-0.186

0.000

0.83

2.649

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

14.146 1 1.159 1.113 2.987 2.809 2.636 1.226

0.421 0.318 0.296 -0.159 -0.4 0.108 -0.32 -0.631 0.104 -0.28

0.147 0.107 1.094 1.033 0.969 0.204

Exponential Beta 1 1.524 1.375 1 1.344 0.853 0.671 1 1.114 0.726 0.532 1 1.109 0.756 1 1.978 2.52 3.15

Constant Source: Same as Table 5. Note:@Significance level, ≥ 0.01= 1 percent, 0.02-0.05= 5 percent, 0.06-0.1= 10 percent; ^Reference Category Dependent Variable: Calorie Intake, 1 shows below 2400 Kcal and 0 shows 2400 & above Kcal

30

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

PROBABILITY OF CONSUMING RECOMMENDED CALORIES: A DISAGGREGATED ANALYSIS In this section the factors affecting probability of consuming recommended calories have been probed through a logistic regression (Table 8). Our results clearly show that probability of getting recommended calories is quite low among all weaker socio-economic groups. For example, as the family size increases, the likelihood of consuming recommended calories declines which exhibits poor nutritional conditions of bigger households. The households covering more than 8 members in the family exhibit higher probability (odd ratio 3.15) of being calorie deprived than the small families having 1-4 members. Among social groups, ST are worst affected as the probability of consuming recommended calories is very low compared to other social groups. SC people however have lower likelihood (1.35 odd ratios) of being calorie deprived than ST (1.52 odd ratios). Regarding religious group, Muslims suffer badly as they have higher probability of being calorie deprived than the Hindus whereas Christians (0.853 odd ratio) and other religion people (0.671 odd ratio) enjoy better calorie intake than the Hindus. Education level plays an important role to determine calorie intake. It has been analysed that as the level of education increases, the likelihood of consuming calories from the norm also rises. Highly educated people show more chances of taking recommended calories than the other lower education group people. Considering the probability of calorie intake among occupation groups, agricultural labourers and other labourers have lesser probability of consuming recommended calories than the employed in non-agriculture. Self employed in agriculture and other occupation groups have more chances of becoming energy sufficient than those who are not self employed in agriculture. As far as poverty level is concerned, people below the poverty line have a much higher likelihood of being calorie deprived (14.146 odd ratios) than the Above Poverty Line category people. The probability of consuming recommended calories across different geographical regions of rural India show that compared to central region (covering states of Uttar Pradesh, Madhya Pradesh and Chhattisgarh), all regions show lower likelihood to consume recommended calories. Among them north eastern, western and southern region covering states of Gujarat, Maharashtra, Karnataka, Tamil Nadu, Andhra Pradesh and Kerala exhibit more chances of being calorie deprived from recommended calories. Northern and eastern states such as Punjab, Himachal Pradesh, Jammu and Kashmir, Haryana, Rajasthan, Orissa and Bihar show lower probability of calorie deprived than the other regions. However, these regions are prone to calorie deprivation when compared with central region. A relatively lower likelihood of being calorie deprived is resulted by higher consumption of cereals. CONCLUSION From this analysis it is found that over a decade (1994-2005) the consumption pattern of Indians has changed significantly. Consumption of cereals, particularly coarse cereals, has declined whereas consumption of other food items such as vegetables, fruits, milk and milk products, meat increased slightly which have a direct bearing on nutrient intake. Due to decline in cereal consumption and lower increase in consumption of other food items nutrient pattern in rural India has also changed substantially. Share of cereals particularly coarse cereals to total calories has declined whereas calories from oil and fat have increased. Since cereals are also a good source of protein but its decline has also led to lowering down of protein. In rural India on an average per

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

31

capita per day calorie and protein intake is falling and consumption of oil and fat is increasing. This, to some extent, is as per the expenditure of dietary transition models. However, given the relative underperformance of India in the nutrition front, this decline in cereal consumption has often been viewed as deterioration in the living standard of the poor. The disaggregated analysis of calorie and nutrition deficiency in rural India carried out in this study clearly points out that deprivation is higher among marginalized social and economic groups. It is the poor, SC and ST groups, agricultural labourers who suffer most in terms of calorie deprivation. There is much gap in official poverty and calorie deprivation level. We have estimated both poverty and calorie deprivation across social groups. Those having bigger families, less education, lower MPCE and those belonging to ST, SC, agricultural labour and other labour class, Muslims are found to have higher levels of poverty as well as calorie deprivation. Thus, in terms of distribution of deprivation across social and economic groups, there is a consistency between poverty and calorie deprivation although the levels are quite different in many cases. The interstate variations, however, does not show much consistency. The southern states particularly Karnataka, Tamil Nadu, Andhra Pradesh perform poor on more than two deprivation indicators. Gujarat and Maharashtra, considered as relatively developed states perform worse on both methods of poverty measurement. On the other hand, Punjab, Himachal Pradesh, Jammu and Kashmir and Haryana are best performing states in all deprivation measures. From a regional point of view, it is found that most of the NSS regions having majority of population being calorie deprived than recommendation fall in the southern, western and central parts of India. All the southern states except Kerala and including Gujarat and Maharashtra presents maximum decline in calorie and protein intake from the recommendation whereas Punjab, Himachal Pradesh, Jammu and Kashmir and Haryana, Uttar Pradesh and Rajasthan show lower decline in calories and in fact increase in protein intake. These states also show lower level of calorie deprivation and poverty. The exercise undertaken to show probability of being calorie deprived concludes that never married, big families, less educated, lower MPCE class, ST, SC, agricultural labour and other labour class, Muslims, people living below poverty line and southern, north-eastern and western states are some weaker sections and regions which are comparatively more prone to be poor and undernourished than their respective reference categories. The debate so far has concentrated on the observed divergence between poverty estimates and calorie deprivation. Our analysis, however, points out that it is the relatively marginalized social and economic groups who face greater calorie deprivation. Thus, there is an urgent need to focus on such high levels of deprivation among the marginalized groups and regions. _________________________________

Notes 1.

Calorie norm has officially been taken to measure poverty level in India. Per capita per day intake of 2400 kcal for rural and 2100 kcal for urban areas are the norms to estimate poverty. Planning Commission makes adjustment in Consumer Price Index for Agricultural Labourers (CPIAL) and Consumer Price Index for Industrial Workers (CPIIW) to the base year poverty line (1973-74) for estimating rural and urban poverty respectively. Planning Commission’s estimation of poverty using indirect method shows lower level of poverty whereas directly using calorie norm to measure poverty gives a much higher level of deprivation.

32

JOURNAL OF REGIONAL DEVELOPMENT AND PLANNING

2.

Food items have been converted into nutritive values using the standard units given in report no. 513(61/1.0/6) Nutritional Intake In India (2004-2005), NSS 61st round National Sample Survey Organisation, Ministry Of Statistics & Programme Implementation Government of India. For further details on measurement of official poverty line in India and changes in it, see Utsa Patnaik, 2007. Standard Calories are given in the Report of the Export Group on Estimation of Proportion and Number of Poor. Perspective Planning Division. Planning Commission, 1993 - 2400 kcal per capita for rural area and 2100 kcal for urban area and standard protein intake is recommended in report on ‘Nutritional Status of Rural Population’ by National Institute of Nutrition (1996) Indian Council of Medical Research, Nutritional Status of rural population, Report of the NNMB surveys, National Nutritional Monitoring Bureau, Hyderabad. Official poverty has been calculated using the report of ‘Poverty Estimates For 2004-05’ Government of India Press Information Bureau [Online at] planningcommission.nic.in/news/prmar07.pdf , Accessed on 12/03/2010 at Jawaharlal Nehru University. The correlation between Below poverty line (BPL) and Below 2400 kcal is 0.472 (significant at 0.05 level) which is low as compared to correlation between BPL and Children underweight below 3 (0.733, significant at 0.01 level) and between BPL and Body Mass Index of Women (0.622, significant at 0.01 level).

3. 4.

5.

6.

References Bansil, P.C. (2003) – “Demand For Food Grains By 2020 Ad”, in S. Mahendra Dev et al. (eds.) Towards A Food Secure India: Issues And Policies, Institute For Human Development, New Delhi. Deaton, A. & Dreze, Jean (2009) – “Food and Nutrition in India: Facts and Interpretations”, Economic & Political Weekly, Vol. 44, No. 7, pp. 42-65. Deaton, A. & Dreze, Jean (2010) – “Nutrition, Poverty and Calorie Fundamentalism: Response to Utsa”, Economic & Political Weekly, Vol. 45, No. 14, pp. 78-80. Dev, S. M. (2005) – “Calorie Norms and Poverty”, Economic & Political Weekly, Vol. 40, No. 8, pp. 789-792. Dubey, A. & Thorat, S. K. (2012) – “Has growth been socially inclusive during 1993/942009/10?” Economic & Political Weekly, Vol. 47, No. 10, pp. 43-54. International Institute for Population Studies (2005-06) – National Family Health Survey 2005-06, Fact Sheet, International Institute for Population Studies, Mumbai. Jones, R. Palmer & Sen, K. (2001) – “On India’s Poverty Puzzles and Statistics of Poverty”, Economic & Political Weekly, Vol. 36, No. 3, pp. 211-217. Kumar, P., Mruthyunjaya & Dev, Madan M. (2007) – “Long Term Changes in Indian Food Basket and Nutrition”, Economic & Political Weekly, Vol. 42, No. 35, pp. 3567-3572. Martorell, Reynaldo & Ho, Teresa J. (1984) – “Malnutrition, Morbidity, and Mortality”, Population and Development Review, Vol. 10, pp. 49-68. Mehta, Jaya & Venkatraman, S. (2000) – “Poverty Statistics, Bermicide’s Feast”, Economic & Political Weekly, Vol. 35, No. 27, pp. 2377-2382. Mehta, Jaya (1982) – “Nutritional Norms and Measurement of Malnourishment and Poverty”, Economic & Political Weekly, Vol. 17, No. 33, pp. 1332-1340. Ministry of Home Affairs (2010) – Sample Registration System, Report No. 1 of 2012, Statistical Report 2010, New Delhi. Nasurudeen, P., Kuruvila, A., Sendhil, R. & Chandresekar, V. (2006) – “The Dynamics and Inequality of Nutrient Consumption in India”, Indian Journal of Agriculture Economics, Vol. 61, No. 3, pp. 362-373. Patnaik, Utsa (2004) – “Republic of Hunger”, Social Scientist, Vol. 32, No. 9/10, pp. 9-35. Patnaik, Utsa (2007) – “Neoliberalism and Rural Poverty in India”, Economic & Political Weekly, Vol. 42, No. 30, pp. 3132-3150.

Journal of Regional Development and Planning, Vol. 2, No. 1, 2013

33

Patnaik, Utsa (2010) – “A Critical Look at Some Propositions on Consumption and Poverty”, Economic & Political Weekly, Vol. 45, No. 6, pp. 74-80. Radhakrishna, R. & Reddy V. (2004) – ‘Food Security and Nutrition: Vision 2020’, [planningcommission.nic.in/reports/.../bkpap2020/16_Bg2020.pdf; accessed on 9 August 2010]. Radhakrishna, R. (2005) – “Food And Nutrition Security of the Poor, Emerging Perspectives and Policy Issues”, Economic & Political Weekly, Vol. 40, No. 18, pp. 1817-1821. Radhakrishna, R., Rao, K. Hanumantha, Ravi, C. & Reddy, B. Sambi (2004) – “Chronic Poverty and Malnutrition in 1990s”, Economic &Political Weekly, Vol. 39, No. 28, pp. 31213130. Rao, H. C. H (2000) – “Declining Demand for Food-Grains in Rural India: Causes and Implications”, Economic & Political Weekly, Vol. 35, No. 4, pp. 201-206. Ray, Ranjan (2005-10) – ‘Analysis of Changes in Food Consumption and their Implications for Food Security and Undernourishment: The Indian Experience in the 1990s’, Discussion Paper, University of Tasmania. Reddy, D. N. & Mishra, Srijit (2010) – Agrarian Crisis in India, Delhi: Oxford University Press. Sen, Pronab (2005) – “Of Calories and Things Reflections on Nutritional Norms, Poverty Lines and Consumption behaviour in India”, Economic & Political Weekly, Vol. 40, No. 43, pp. 4611-4618. Shariff, Abusaleh & Mallick, A. C. (1999) – “Dynamics of Food Intake and Nutrition by Expenditure Class in India”, Economic & Political Weekly, Vol. 34, No. 27, pp. 17901800. Verma, Med Ram, Datta, K. K., Mandal, Subhasis & Tripathi, A. K. (2008) – “Diversification of Food Production and Consumption Patterns in India”, Journal of Agricultural & Food Information, Vol. 8, No. 3, pp. 87-100. Viswanathan, Brinda (2001) – “Structural Breaks in Consumption Patterns: India 1952-1991”, Applied Economics, Vol. 33, No. 9, pp. 1187-1200. World Food Programme (2009) – ‘India Tops World Hunger Chart’, [http://www.wfp.org/Countries/India/News/Hunger-In-The-News?Page=4; accessed on 23 July 2011] World Health Organisation (2003) – ‘Diet, Nutrition and the Prevention of Chronic Diseases’, Technical Report Series 916, Geneva.