ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

On Calibrating the Poverty Line for Poverty Estimation in India

This paper examines the issue of the appropriateness of using National Sample Survey 28th round data for setting the official base year poverty line. The poverty lines, as per the alternative base years according to the data of NSS 27th and 26th rounds, which are based on much larger sample sizes, are found to be significantly higher. The paper also brings out the fact that the rural-urban difference in the average daily per capita calorie requirement of the Indian population as per the official statistical system narrows significantly with the suggested change in the classification of worker population, which is by no means unrealistic or unwarranted.

Aspects of Poverty and Employment

On Calibrating the Poverty Line for Poverty Estimation in India

This paper examines the issue of the appropriateness of using National Sample Survey 28th round data for setting the official base year poverty line. The poverty lines, as per the alternative base years according to the data of NSS 27th and 26th rounds, which are based on much larger sample sizes, are found to be significantly higher. The paper also brings out the fact that the rural-urban difference in the average daily per capita calorie requirement of the Indian population as per the official statistical system narrows significantly with the suggested change in the classification of worker population, which is by no means unrealistic or unwarranted.

G C MANNA

I Introduction

I
nformation on headcount ratio (HCR), i e, the percentage of people below the poverty line, is of vital importance to planners and policymakers. The government often requires the state-level estimates of HCR to decide budget allocations, grants, etc, to be made to the states. The state governments also need reliable estimates of state-level HCR at periodic intervals to monitor the poverty situation. Given this perspective, it is needless to emphasise the importance of calibrating the relevant poverty line accurately so that the estimate of HCR may reflect the underlying actual poverty situation prevailing at a given point of time. The official poverty lines for India are anchored on a calorie norm of 2,400 calories per capita per day for rural areas and 2,100 calories per capita per day for urban areas. The Task Force1 set up by the Planning Commission, defined the poverty line for the base year (1973-74) as the per capita total consumer expenditure level corresponding to the above calorie norms on the basis of the all-India size distribution of monthly per capita consumer expenditure (MPCE) for 1973-74. These all-India poverty lines for 1973-74 were estimated to be Rs 49.09 and Rs 56.64 per capita per month at 1973-74 prices for rural and urban India respectively. This estimation was done on the basis of the consumer expenditure data for the 28th round of National Sample Survey (NSS) conducted during the year.

To get the poverty lines at current prices for subsequent years, the poverty line for the base year thus obtained is updated by indexation, thus taking care of the temporal change in the price level of items in the underlying consumption basket. For obtaining the state-specific poverty lines, the all-India rural and urban poverty lines for various years are further adjusted to reflect the observed differences in the rural and urban cost of living across states.2 Such poverty lines have been worked out for the years 1977-78, 1983, 1987-88, 1993-94 and 1999-2000, apart from the base year (1973-74). These years refer to the corresponding survey periods of various NSS quinquennial rounds, viz, NSS 32nd, 38th, 43rd, 50th and 55th rounds respectively on household consumer expenditure and employment-unemployment. Table 1 gives the all-India poverty lines (“implicit”) for these years as released by the Planning Commission.

Once the poverty lines for India as a whole or any state for that matter, are obtained, the HCR is estimated from the size distribution of MPCE for the corresponding year based on the NSS quinquennial round on household consumer expenditure as the proportion of persons below the poverty line.

From what has been stated above, it is clear that the poverty line and hence the corresponding estimated HCR for any given year primarily depends upon three factors, viz, (a) the choice of the base year, (b) per capita calorie norm to be considered for defining

Table 1: All-India Poverty Lines for Various Years

Sector All-India Poverty Line (Rs at current prices) 1973-74 1977-78 1983 1987-88 1993-94 1999-2000

(1) (2) (3) (4) (5) (6) (7)

Rural 49.63 56.84 89.45 115.43 205.84 327.56 Urban 56.96 72.50 117.64 165.58 281.35 454.11

Sources: Planning Commission (1993): Report of the Expert Group on Estimation of Proportion and Number of Poor, Perspective Planning Division, July; Press Releases by the Planning Commission on Poverty Estimates for the years 1993-94 and 1999-2000.

Table 2: All-India Poverty Lines for Different Base Years

Sector Poverty Line for Different Base Sample Size (Number of Households) Years at Current Prices (Rs) Involved in the Corresponding Distribution of Persons by MPCE Class

Alternate Alternate Official Alternate Alternate Official
Base Base Base Base Base Base
Year: 1 Year: 2 Year Year: 1 Year: 2 Year
(1971-72) (1972-73) (1973-74) (1971-72) (1972-73) (1973-74)
(1) (2) (3) (4) (5) (6) (7)

Rural 36.83 45.07 49.09 11,468 72,270 15,467 Urban 47.78 54.67 56.64 19,459 52,820 7,881

Economic and Political Weekly July 28, 2007

the poverty line, and (c) the price deflator to be used for updating the base year poverty line. In this paper, we examine critically the practice being followed so far as the first two factors are concerned in order to judge their appropriateness in the context of poverty measurement. The paper is organised as follows: Section II deals with the issue of appropriateness of using the NSS 28th Round data for setting the official base year poverty line and also examines how the base year poverty lines would differ if NSS 27th Round data were used for setting the official base year poverty line instead. Section III looks into the details of various aspects of working out the average calorie norms for rural and urban poor of India and shows how the estimated average calorie norms might differ if procedural changes (viz, change of the classification of worker population that is by no means unrealistic or unwarranted) were made. In Section IV, the average calorie norms for India as a whole based on the Task Force procedure and the suggested alternative procedure have been compared to show that the rural-urban difference of the estimated average calorie requirements narrows noticeably when the modified procedure is used. Further, an estimate of how much of the reduction in the ruralurban difference of average calorie norms may be due to the changes in the classification of worker as per the modified procedure is presented. The results of the state level analysis are discussed in Section V. The paper is concluded in Section VI.

II Appropriateness of Official Base Year and Alternatives

As already mentioned, the year 1973-74 has been taken as the official base year for estimation of poverty in India. To work out the poverty line for the base year, NSS 28th Round (1973-74) data on reported household consumption, both in quantity and value terms, were used. Using appropriate conversion factors,3 average (per capita per day) calorie content of the quantities of food items for each MPCE class was calculated separately for rural and urban areas. Thereafter, applying the linear interpolation method to the data on monthly per capita total consumer expenditure and the associated average calorie content of food items in each total MPCE class separately for the rural and urban areas, the official base year poverty line in terms of MPCE was obtained as Rs 49.09 for rural India and Rs 56.64 for urban India at 197374 prices. In other words, interpolated MPCE for rural and urban population was found to correspond to the average calorie consumption of 2,435 and 2,095 per capita per day for the rural and urban areas respectively during 1973-74.

Since the base year poverty line is adjusted with the help of suitable price deflators to obtain the poverty lines for the subsequent years, the choice of base year is important. In this context, it may be mentioned that the NSS 28th round survey period was only of nine months duration (October 1973-June 1974). More importantly, the number of households surveyed for the consumer expenditure enquiry in this round was much less (less than onefourth for rural India and less than one-sixth for urban India) as compared to that of NSS 27th round (October 1972-September 1973) survey – the first of the series of the quinquennial surveys on household consumer expenditure and employmentunemployment introduced in the NSS. Thus the results based on NSS 28th round that have been used for deriving the poverty line for the base year are likely to be subject to higher sampling error as compared to those based on NSS 27th round.4

It should be both interesting and useful to compute the poverty line for the year 1972-73 ( (i e, year preceding the official base year) based on the consumer expenditure data of NSS 27th round survey that adopted a much larger sample size and see how this differs from the official poverty line. However, since the year 1972-73 was affected by famine, we would also like to derive the poverty line for another alternative base year (viz, 1971-72) using the NSS 26th round (July 1971-June 1972) data, which too had a larger sample size for urban India.

The average MPCE and average calorie content of food consumed by various MPCE classes based on NSS 26th, 27th and 28th rounds are presented in Statement A1. The poverty lines for the two alternate base years that we obtain, after applying the linear interpolation method to the data on average MPCE and the associated calorie content of food items (per capita per day) in the class separately for rural and urban areas, along with

Table 3: Monetary Equivalence of Alternative Poverty Linesat Official Base Year Prices

Sector Consumer Price Index (CPI)* Monetary Equivalence of

(CPIAL for Rural; CPIIW Alternative Poverty Lines (Rs)

and CPIUNME for Urban) at Official Base Year (1973-74) Prices

Alternate Alternate Official Alternate Alternate Official

Base Base Base Base Base Base

Year: 1 Year: 2 Year Year: 1 Year: 2 Year

(1971-72) (1972-73) (1973-74) (1971-72) (1972-73) (1973-74)

(1) (2) (3) (4) (5) (6) (7)

Rural 199.67 235.42 290.44 53.57 55.60 49.09
(36.83) (45.07)
Urban: 195.08 224.00 273.00 66.86 66.63 56.64
(IW) (IW) (IW) (47.78) (54.67)
183.17 203.75 236.67 61.74 63.50 56.64
(UNME) (UNME) (UNME) (47.78) (54.67)
189.125 213.875 254.835 64.38 65.14 56.64
(Av) (Av) (Av) (47.78) (54.67)

Notes: Figures within brackets denote the alternative poverty lines for two alternate years at respective current prices.

* Obtained as the simple average of corresponding monthly indices as per the CSO’s Monthly Abstract of Statistics. ‘Av’ means simple average of CPIs for IW and UNME.

Table 4: Average Calorie Norm in Use in Poverty Estimation inIndia and the Corresponding Weighting Diagram of Population

Age-Sex-Activity Group Weighting Diagram Calorie Norm
Rural India Urban India (Kcal Per Day)
(1) Children
(a) Less than 1 year 2.72 2.49 650
(b) 1 to 4 years 8.21 7.66 1200
(c) 4 to 7 years 7.99 7.53 1500
(d) 7 to 10 years 7.47 7.66 1800
(e) 10 to 13 years 7.46 7.66 2100
(2) Adolescents: 13-15 years
(a) Boys 2.37 2.36 2500
(b) Girls 2.22 2.23 2200
(3) 15 years and above
(i) Workers: Male
(a) Heavy 22.63 3.83 3900
(b) Moderate 2.13 7.72 2800
(c) Sedentary 2.66 14.10 2400
(II) Workers: female
(a) Heavy 10.88 1.47 3000
(b) Moderate 0.93 1.40 2200
(c) Sedentary 0.59 2.87 1900
(4) Non-workers:
(a) Male 3.92 8.62 2400
(b) Female 17.82 22.40 1900
Average calorie norm 2432 2094

Source: Datta and Sharma (2002), p 4.

Economic and Political Weekly July 28, 2007 the poverty line for the official base year are presented in Table 2.

The poverty lines presented in Table 2 for the alternate base years as well as the official base year are based on the same calorie norm (viz, 2,435 and 2,095 per capita per day for rural and urban areas respectively). It may be seen that the poverty lines for three alternate base years, namely, 1971-72, 1972-73 and 1973-74, come out to be Rs 36.83, Rs 45.07 and Rs 49.09 respectively for rural India and Rs 47.78, Rs 54.67 and Rs 56.64 respectively for urban India.

To examine if the above differences in the poverty lines for alternate base years with those for the official base year are explained by the difference in price level, we have inflated the alternative base year poverty lines using the Consumer Price Index for Agricultural Labourers (CPIAL; Base: 1960-61 = 100) for rural areas and both Consumer Price Index for Industrial Workers (CPIIW; Base: 1960 = 100) and Consumer Price Index for Urban Non-Manual Employees (CPIUNME; Base: 1960 = 100) for urban areas. Table 3 summarises the results.

Based on Table 3 results, following observations may be made:

  • (a) The poverty lines as per the two alternate base years are closer to each other at constant 1973-74 prices, and
  • (b) These poverty lines for each of the alternate base years, which have been derived from the NSS data with much larger sample sizes,5 are significantly higher than the official base year poverty line for both rural and urban areas at constant 1973-74 prices.
  • III Average Calorie Norms and Differences Due to Procedural Changes

    Now we turn back to another important aspect, namely the average per capita daily calorie requirement, considered as the norm for defining the poverty line. The official poverty lines for rural and urban India have an average calorie norm of about 2,435 and 2,095 calories per capita per day, respectively (which have been rounded off to 2,400 calories and 2,100 calories respectively). These are based on the recommendation of the Task Force on Projections of Minimum Needs and Effective Consumption Demand [Planning Commission 1979] (henceforth to be referred to as the ‘Task Force’). The Task Force used the calorie norms of the Indian Council of Medical Research (ICMR) for various ‘relatively homogeneous’ person categories and the corresponding age-sex-activity distribution of the population as the weighting diagram to obtain these average calorie norms.

    To explain further,6 the said weighting diagram, given in Table 4 is based on (a) the projected age-sex structure of the population for 1982-83 (III projection) of the Expert Committee on Population, 1977, (b) 1971 census occupational structure and

    (c) participation rates based on “usual activity status” from the NSS employment-unemployment data of 27th round (1972-73). The weighting diagram of population coupled with the calorie norms yielded an average calorie requirement of 2,432 calories per capita per day for rural areas and 2,094 calories per capita per day for urban areas. Adding the extra calorie needs of pregnant and lactating women, the average daily calorie requirement came to be 2,436 calories for rural areas and 2,097 calories for urban areas. For the sake of simplicity, these were rounded off to 2,400 kcal for the rural areas and 2,100 kcal for the urban areas.

    The Task Force made the following assumptions7 while classifying workers (Table 4): (a) heavy workers included persons engaged in cultivation, agricultural labour, mining and quarrying and construction; (b) moderate workers included persons engaged in livestock, forestry, hunting, plantations, orchards and allied activities, manufacturing, servicing and repairing;

    (c) sedentary workers included persons engaged in trade and commerce, transport, storage, communication and other allied services; and (d) calorie requirements for adult non-workers were the same as those for sedentary workers.

    It is worth noting that the Task Force considered all persons engaged in cultivation, agricultural labour, mining and quarrying and construction as heavy workers. It is quite likely that jobs pursued by a certain percentage of such workers might actually be only either supervisory or clerical in nature. Hence the calorie requirements of such workers would not be as high as the stipulated one (i e, 3,900 Kcal/day for men and 3,000 Kcal/day

    Table 5: Classification of Different Types of Work:Task Force Classification vis-a-vis Proposed Classification

    Type of Work Coverage in Terms of NCO (1968) Codes Task Force Classification Proposed Classification

    (1) (2) (3)

    Heavy 60, 61, 63, 65, 71, 95 and 97 52-56, 570, 574, 579, 63-68,

    71-84, 87-99 Sedentary 00-59 and 98 00-51, 571, 572, 573, 59 Moderate Remaining Remaining

    Note: See NCO (1968) for descriptions of various NCO codes.

    Table 6: Age-Sex-Occupation Groups/Calorie Norms as Per Expert Group (1988) and Weighting Diagram/Average Calorie Requirement as Per Proposed and Task Force Classifications of Work

    Age-Sex-Occupation Calorie Weighting Weighting Group Norm Diagram: Diagram: (Kcal Per Day) 1999-2000 1999-2000 by Expert (As per (As per Group Proposed Task Force Classification Classification of Work) of Work) (1988) Rural Urban Rural Urban

  • (1) Infants (less than 1 year) 700* 1.84 1.59 1.84 1.59
  • (2) Children:
  • (a) 1 to 3 years 1240 7.36 5.67 7.36 5.67
  • (b) 4 to 6 years 1690 8.37 6.56 8.37 6.56
  • (c) 7 to 9 years 1950 7.49 6.35 7.49 6.35
  • (3) Boys: 10 to 12 years 2190 4.44 3.91 4.44 3.91
  • (4) Girls: 10 to 12 years 1970 3.79 3.36 3.79 3.36
  • (5) Boys: 13 to 15 years 2450 3.30 3.33 3.30 3.33
  • (6) Boys 16 to 18 years 2640 3.32 3.67 3.32 3.67
  • (7) Girls: 13 to 18 years 2060 5.69 6.24 5.69 6.24
  • (8) 19 years and above:
  • (I) Workers: Men
  • (a) Sedentary 2425 2.93 11.99 3.88 14.95
  • (b) Moderate 2875 9.82 1.52 3.32 7.05
  • (c) Heavy 3800 11.16 11.45 16.71 2.97
  • (II) Workers: Women
  • (a) Sedentary 1875 0.53 2.12 0.76 2.96
  • (b) Moderate 2225 3.94 0.33 1.26 1.24
  • (c) Heavy 2925 5.39 2.61 7.84 0.87
  • (9) Non-Workers:
  • (a) Men 2425* 3.17 5.76 3.17 5.76
  • (b) Women 1875* 17.48 23.51 17.48 23.51 Average calorie requirement 2286 2249 2350 2142
  • Note: * Assumed for calculating average calorie norm. Expert Group (1988) recommended 108 kcal/kg and 98 kcal/kg respectively for infants in the age-group of 0-6 months (body weight 5.4 kg) and 6-12 months (body weight 8.6 kg).

    Source: For age-sex-activity groups/calorie norms, see ICMR, 1988, p 129.

    Economic and Political Weekly July 28, 2007

    for women, as in Table 4). Similarly, among persons categorised as “moderate workers” by the Task Force, there are likely to be a fraction of persons whose actual type of work might be either “heavy” or “sedentary” in nature and accordingly their calorie requirements are likely to be different.

    We define heavy work to be “manual work” as per the practice followed in the National Sample Survey (NSS). In these surveys, a manual work is defined as a job essentially involving physical labour. In our view, it is this category of people who should be actually classified as “heavy workers” in terms of calorie requirements recommended by the ICMR. The exhaustive list of occupations and the corresponding 1968 National Classification of Occupation [NCO 1968] codes, which constitute manual work as per the NSS, are the following:8

  • (i) Cooks, waiters, bartenders and related workers – domestic and institutional (NCO Code 52).
  • (ii) Maid and other housekeeping service workers (NCO Code 53).
  • (iii) Building caretakers, sweepers, cleaners and related workers (NCO Code 54).

  • (iv) Launders, dry cleaners and pressers (NCO Code 55).
  • (v) Hairdressers, barbers, beauticians and related workers (NCO Code 56).
  • Table 7: Alternative Estimates of Average Per Capita Daily Calorie Requirement by Indian Population for Different time Periods/Approaches

    Sector Average Per Capita Daily Calorie Requirement

    by the Indian Population
    1973-74 1993-94 1999-2000
    (Derived by As per As Per As Per As Per
    Task Force) Task Force Proposed Task Force Proposed

    Classification Classification Classification Classification of Work of Work of Work of Work

    (1) (2) (3) (4) (5) (6)
    Rural 2435 2356 2280 2350 (-3.5) 2286 (-6.1)
    Urban 2095 2140 2244 2142 (2.2) 2249 (7.4)

    Notes: (1) Figures within parentheses denote percentage change as compared to the year 1973-74.

    (2) Estimates in columns 3 to 6 are based on the latest calorie norms as per Expert Group (1988).

    (vi) Fire fighters (NCO Code 570).

    (vii) Watchmen and gatekeepers (NCO Code 574).

    (viii) Protective service workers not elsewhere classified (NCO Code 579).

  • (ix) Agricultural labourers (NCO Code 63).
  • (x) Plantation labourers and related workers (NCO Code 64).
  • (xi) Other farm workers (NCO Code 65).
  • (xii) Forestry workers (NCO Code 66).

    (xiii) Hunters and related workers (NCO Code 67).

    (xiv) Fishermen and related workers (NCO Code 68).

  • (xv) Production and related workers, transport equipment operators and labourers (NCO Divisions 7, 8 and 9 excluding Code 85: electrical fitters and related workers; and Code 86: broadcasting station and sound equipment operators and cinema projectionists).
  • It is interesting to note that occupations listed above under items
  • (i) – (viii) would get classified as sedentary work according to the approach adopted by the Task Force. We however continue to consider “sedentary work” as per the coverage adopted by the Task Force, that is, trade and commerce, transport, storage, communication and other allied services but excluding those occupations covered under our definition of heavy work. The rest of the occupations would be considered as moderate work under the proposed classification. Table 5 specifies the differences in the coverage of heavy work, moderate work and sedentary work as per the Task Force Classification and our proposed classification if we express three categories of work in terms of NCO codes.
  • The most recent recommendations on dietary allowances for different “homogeneous groups” of people are made by the Expert Group of ICMR constituted in 1988, hereinafter to be referred to as the “Expert Group (1988)” [ICMR 1988]. There are certain differences in the structure of age-groupings and in the calorie norms for different groups of persons as per the Expert Group (1988) when we compare them with those in Table 4 used for deriving the official average calorie norms. In terms of daily calorie requirement, the Expert Group (1988) has identified sixteen “homogeneous groups” of persons – two for infants of less than one year (viz, infants of age less than six months and

    Table 8: Grouping of States/UTs According to Average Daily Per Capita Calorie Requirement as Per the Proposed Classificationof Work (1999-2000)

    Average Daily Per Capita Names of States/UTs Names of States/UTs Calorie Requirement (kcal) (Rural Areas) (Urban Areas)

    (1) (2) (3)

    2150-2200 Arunachal Pradesh, Uttar Pradesh, Lakshadweep, Nagaland, Arunachal Pradesh, Jammu and Kashmir, Mizoram, Rajasthan Meghalaya, Bihar, Manipur, Tripura, Assam,

    2200-2250 Manipur, Haryana, Nagaland, Uttar Pradesh, Mizoram, Rajasthan, Lakshadweep, Jammu and Kashmir, Meghalaya, Chandigarh, Himachal Pradesh, Delhi, Haryana, Bihar, Sikkim, Himachal Pradesh, Punjab Madhya Pradesh

    2250-2300

    Assam, Delhi, Andaman and Nicobar Islands, Sikkim, Andhra Pradesh, Gujarat, Orissa, Karnataka,

    West Bengal, Dadra and Nagar Haveli, Tripura Punjab, Daman and Diu, Dadra and Nagar Haveli, Maharashtra, West Bengal, A and N Islands, Kerala, Goa

    2300-2350 Madhya Pradesh, Orissa, Kerala, Goa, Gujarat Tamil Nadu, Pondicherry 2350-2400 Maharashtra, Karnataka – 2400-2450 Andhra Pradesh, Pondicherry – 2450-2500 Daman and Diu, Tamil Nadu, Chandigarh –

    States/UTs with urban average daily per capita calorie requirement exceeding the rural requirement: 1999-2000: Haryana, Mizoram, Punjab, Rajasthan, Sikkim, Uttar Pradesh, West Bengal, Andaman and Nicobar Islands,Lakshadweep 1993-94: Arunachal Pradesh, Haryana, West Bengal, Andaman and Nicobar Islands, Chandigarh

    Note: The newly formed states of Chhattisgarh, Jharkhand and Uttaranchal are covered under the erstwhile states.

    Economic and Political Weekly July 28, 2007 of age six months or more but less than one year); three for children in the age-group one-nine years; five for boys and girls in the age-group 10-18 years; and six for men and women of age 19 years and above engaged in heavy, moderate and sedentary work.

    With these sixteen homogeneous groups, we add two more groups, viz, men non-workers of age 19 years and above and women non-workers of age 19 years and above, to account for the total population. But we merge the two groups for infants into one group of age less than one year since age-distribution of infants over these two groups is not readily available. The detailed composition of the seventeen groups along with their calorie norms as per the Expert Group (1988) is given in Table 6.

    With the proposed classification of work (Table 5) and by considering the age, sex and occupation [NCO 1968] distribution of persons (occupation according to ‘principal usual status’) as per the data available for the latest quinquennial NSS largesample round on the Survey on Employment-Unemployment, namely 55th round,9 we have obtained the weighting diagram

    Annex Statement A-1: Average MPCE and Per Capita Per Diem Intake of Calories (All Items) by MPCE Classes for NSS 26th, 27th and 28th Rounds

    MPCE 26th Round 27th Round 28th Round
    (July ’ 71 (October ’72 (October ’73-
    June ’72) September ’73)** June ’74)@
    (Rs) Av MPCE Calorie Av MPCE Calorie Av MPCE Calorie
    (Rs) (Rs) (Rs)
    (1) (2) (3) (4) (5) (6) (7)
    Rural India
    0-13 1177 10.38 776 10.57
    13-15 14.01 1055 14.04
    15-18 1554 16.49 1220 16.81
    18-21 19.46 1398 19.67
    21-24 (22.43) 1813 22.41 1586 22.45
    24-28 (25.97) 1923 25.91 1743 26.03
    28-34 (30.92) 2179 30.86 1944 30.98
    34-43 (38.26) 2497 38.14 2210 38.38
    43-55 2810 48.24 2538 48.56
    55-75 3259 63.20 2929 68.56
    75-100 3720 85.26 3439 85.31
    100-150 118.00 4110 119.42
    150-200 4987 170.32 5221 169.85
    >= 200 342.81 6991 284.69
    All 2168 44.17 2266 53.01
    Urban India
    0-13 971 9.27 624 8.47
    13-15 13.95 894 13.56
    15-18 1244 16.64 1051 17.01
    18-21 19.63 1208 19.58
    21-24 1445 22.60 1379 22.77
    24-28 1568 26.01 1438 26.19
    28-34 1704 30.98 1610 31.18
    34-43 (38.45) 1882 38.39 1802 38.51
    43-55 (48.74) 2117 48.58 1980 48.91
    55-75 2333 63.73 2266 64.04
    75-100 2651 86.00 2563 85.94
    100-150 119.60 2879 107.60
    150-200 3150 170.98 3533 171.82
    >= 200 309.53 4150 287.82
    All 2179 63.33 2107 70.77

    Notes: ( ) Figures within brackets obtained as the simple average of the entries in columns 4 and 6 since average MPCE figures are not available from the corresponding NSS report (No 238).

    * Obtained as [(Calorie per consumer unit) × (no of consumer units per hh)]/[av Hh size], NSS Report No 238. (Col 7 left blank since not available from the report).

    Sources ** Sarvekshana, Vol VI, Nos 3-4; January-April 1983. @ NSS Report No 240.

    of the population for the year 1999-2000 corresponding to the round (Table 6). By applying this weighting diagram on the different calorie norms of Expert Group (1988) and assuming that (a) the average calorie requirement of infants is 700 kcal/ day and (b) the calorie requirements of adult (age 19 years and above) non-workers are the same as those for adult sedentary workers (the same being assumed by the Task Force also), we have worked out the average calorie requirement for the rural and urban areas (Table 6). To facilitate comparison, we also present the average calorie requirement as per Task Force classification of work.

    IV Comparison of Average Calorie Norms

    For having additional inputs, we have computed average per capita daily calorie requirement based on employmentunemployment data of another round namely, NSS 50th round (1993-94), exactly in the same manner as done for NSS 55th round (1999-2000). Table 7 presents the alternative estimates.

    Our estimate of average calorie requirement for Indian population as per the proposed classification of work comes out to be 2,286 calorie per capita per day for rural India and 2,249 calorie per capita per day for urban India during 1999-2000. The corresponding estimates for the year 1993-94 are also close, namely, 2,280 calorie and 2,244 calorie, respectively. If we make allowances for additional calorie requirements for the pregnant women (300 additional calorie per day) and women in lactation (550 and 400 additional calorie per day respectively for the lactating period up to 6 months and 6-12 months) as recommended by the Expert Group (1988), about 2,290 calories per person per day for rural areas and 2,250 calories per person per day for urban areas would be adequate.

    It may be mentioned that our estimated average calorie norms for rural and urban India are quite reliable with percentage (relative) standard error being less than 0.1 per cent (see Section V and Statements A-2 and A-3). It is of interest to note that in our suggested average calorie norms, rural-urban divergence is not so pronounced, as it exists in the official norms. The suggested norm is much less than the official norm (2,400) for rural India but significantly higher than that (2,100) for urban India. But with the Task Force classification of work, rural-urban divergence in the average calorie requirement prevails (columns 3 and 5, Table 7) although the extent of divergence diminishes when we compare it with the official rural-urban calorie norms in use.

    Let d73-74,TF denote the divergence between the rural and urban per capita average calorie requirement for the official base year. Similarly, let d99-00,TF denote the said divergence for the year 1999-2000 as per Task Force classification of work and d99-00,PPdenote the divergence for the year 1999-2000 as per the proposed classification of work. According to Table 7, the values of d73-74,TF; d99-00,TF; and d99-00,PP are found to be 340, 208, and 37 respectively.

    The total change/decline in the value from d73-74,TF to d99-00,PP,i e, (d73-74,TF – d99-00,PP) can be decomposed into two distinct parts namely, (d73-74,TF – d99-00,TF) and (d99-00,TF – d99-00,PP). Undoubtedly, the effect of ‘change in the proposed classification of work’ can be measured by computing the ratio R defined as

    R = (d99-00,TF – d99-00,PP)/(d73-74,TF – d99-00,PP).

    The value of R works out to be 0.56. In other words, about 56 per cent of the overall effect on the reduction in the divergence

    Economic and Political Weekly July 28, 2007

    between the rural and urban average per capita daily calorie requirement is attributable to the changes in the proposed classification of work.

    V Variations in Calorie Requirements Across States

    It may be useful to assess the average daily per capita calorie requirement at the state/UT level specific to the age-sex-activity distribution of the people for the state/UT. The same would be of interest if one is willing to assess the actual proportion of persons below this average daily calorie requirement for any particular state/UT. Accordingly, following the age-sexoccupation grouping (Table 6) and proposed classification of work (Section 3/col 3 of Table 5), we have estimated the average daily per capita calorie requirement for each state/ UT by analysing the age-sex-occupation distribution of the people for the respective states and UTs based on the data of NSS 50th and 55th rounds. We have also examined the validity of these estimates by computing relative standard errors (rse) of estimates as per the procedure described in the succeeding paragraph.

    Let c1, c2 and c0 denote the estimated average daily per capita calorie requirement for a state/UT based on the data of odd sub-samples (i e, sub-sample 1 in case of NSS 50th round and sub-samples 1, 3, 5 and 7 in case of NSS 55th round), even sub-samples (i e, sub-sample 2 in case of NSS 50th round and sub-samples 2, 4, 6 and 8 in case of NSS 55th round) and all sub-samples10 respectively, excluding the revisited samples in case of 55th round.11 Then estimate of rse of the pooled estimator corresponding to c0 is obtained as:

    var(c )o

    o

    ×100rse(c ) = c

    o 12 2

    where var(co) =[(c1 −c0) +(c2 −c0) ]

    2

    The state-specific estimates of average daily per capita calorie requirement as per both the proposed classification and Task Force classification of work, the rses of the respective estimates and the sample size i e, the number of sample persons based on which the analysis have been made are presented in Statements A-2 and A-3 for NSS 50th round (1993-94) and 55th round (1999-2000) respectively.

    It may be seen that the rses of state level estimates of average daily per capita calorie requirement are very small (less than 1 per cent in most of the cases) implying thereby that these estimates are quite reliable. State level estimates of average daily per capita calorie requirement as per the two rounds/time periods are closer to each other except in a few cases where the rses of the estimates are relatively higher at least for one time period. Thus

    (1993-94)

    Statement A-2: State-wise Average Daily Per Capita Calorie Requirements and the Relative Standard Errors (rse)of the Estimates (Per Cent)

    State/UT Rural As Per As Per Task Proposed Force Classification Classification of Work of Work Calorie rse Calorie rse (1) (2) (3) (4) (5) Andhra Pradesh 2411 0.41 2464 0.87 Arunachal Pradesh 2213 1.13 2449 2.14 Assam 2257 0.27 2300 0.21 Bihar 2245 0.16 2328 0.03 Goa 2340 1.84 2322 1.11 Gujarat 2339 0.18 2394 0.14 Haryana 2181 0.41 2224 0.21 Himachal Pradesh 2227 0.04 2385 0.10 J and K 2155 0.27 2240 0.02 Karnataka 2323 0.09 2413 0.35 Kerala 2319 0.28 2229 0.45 Madhya Pradesh 2289 0.08 2443 0.11 Maharashtra 2359 0.42 2451 0.58 Manipur 2193 0.00 2308 0.47 Meghalaya 2286 0.43 2549 0.72 Mizoram 2188 0.57 2445 0.46 Nagaland 2178 1.13 2315 0.93 Orissa 2302 0.07 2376 0.41 Punjab 2241 0.09 2288 0.33 Rajasthan 2209 0.23 2340 0.02 Sikkim 2283 0.95 2367 0.31 Tamil Nadu 2413 0.16 2434 0.14 Tripura 2250 0.92 2250 0.90 Uttar Pradesh 2183 0.19 2295 0.02 West Bengal 2249 0.19 2267 0.59 A and N Is 2275 0.46 2228 0.35 Chandigarh 2276 3.32 2084 2.67 D and N Haveli 2353 1.10 2286 1.70 Daman and Diu 2398 2.57 2260 0.00 Delhi 2309 5.75 2126 5.55 Lakshadweep 2235 0.11 2157 0.35 Pondicherry 2345 3.39 2338 2.57 All-India 2280 0.05 2356 0.17 Sample Size (No of Persons) (6) 21420 5133 16706 36575 643 11849 6137 9808 4626 14202 11856 28795 22107 5457 5074 2350 2520 16653 11148 16846 2066 16253 6975 52149 23454 2656 336 1201 396 244 391 325 356351 Urban As Per As Per Task Sample Proposed Force Size Classification Classification (No of of Work of Work Persons) Calorie rse Calorie rse (7) (8) (9) (10) (11) 2265 0.08 2157 0.10 15912 2259 1.61 2155 1.30 834 2206 0.02 2121 0.40 3784 2198 0.09 2113 0.10 10617 2280 0.34 2167 0.80 848 2251 0.11 2124 0.07 10959 2220 0.59 2113 0.61 3124 2216 1.16 2153 1.40 1472 2151 0.49 2094 0.26 2553 2242 0.02 2154 0.87 11552 2301 0.73 2180 0.55 8181 2220 0.33 2133 0.31 15924 2263 0.21 2149 0.24 24277 2179 0.27 2188 0.23 3696 2135 0.66 2066 0.23 1897 2176 0.31 2241 0.10 4371 2143 0.29 2102 0.03 1171 2254 0.63 2163 0.60 4525 2222 0.47 2119 0.52 8779 2204 0.59 2122 0.37 8440 2238 2.02 2141 0.80 571 2309 0.36 2174 0.45 16405 2184 0.63 2112 0.32 2316 2177 0.08 2104 0.10 23098 2295 0.22 2163 0.13 13509 2290 0.30 2176 0.15 1504 2306 0.66 2138 0.66 537 2250 1.80 2178 2.48 334 2191 1.61 2077 0.75 347 2239 1.06 2131 0.84 4047 2185 1.12 2118 1.18 1343 2251 1.88 2155 1.50 1462 2244 0.003 2140 0.03 208389
    Economic and Political Weekly July 28, 2007 3113

    we would summarise our observations based on the results of NSS 55th round that corresponds to a more recent time period (1999-2000). We would focus our analysis by considering the average calorie requirements based on the proposed classification of work.

    State level average daily per capita calorie requirement as per the proposed classification of work varies significantly across states in both rural and urban areas. For rural areas, with average daily per capita calorie requirement for all-India being 2,286, the state level estimates vary between 2,161 (Arunachal Pradesh) and 2,475 (Chandigarh). For urban areas, the estimates vary between 2,150 (Nagaland) and 2,313 (Pondicherry) as against the all-India average of 2,249. The state level estimates of average daily per capita calorie requirement, sorted in ascending order of calorie requirement, along with the rses of the estimates, separately for rural and urban, are presented in Statement A-4. Table 8 gives a summary.

    VI Concluding Observations

    The choice of base year has a pivotal role in the estimation of headcount ratio (HCR) at any given point of time. This is because the poverty line as per the base year is adjusted with the help of suitable price deflators to estimate the poverty lines, and accordingly the HCR, for the subsequent years. With likely change in the consumption behaviour of the people, the year 1973-74 is too old a period to be still considered as the base year for poverty estimation.

    Further, we have seen substantial differences (Section II) between the poverty line as per the official base year (1973-74) and those as per the two alternative base years, namely, 1971-72 and 1972-73 based on much larger sample sizes. The poverty lines as per the alternative base years are closer to each other and are significantly higher than the official base year poverty lines (Rs 49.09 for rural and Rs 56.64 for urban). The poverty line for the alternative base year 1972-73 exceeds the official base year poverty line by about 13 per cent for rural India and 17 per cent for urban India12 at 1973-74 prices.

    Moreover, according to a more realistic classification of different types of work (Section III) and the latest (199394 and 1999-2000) age-sex-occupation distribution of persons and latest (1988) recommendations on calorie norms by the Expert Group of ICMR, the average daily per capita calorie requirement by the Indian population is found to be significantly different (Section IV) from the official calorie norms (2,400 for rural and 2,100 for urban) being in vogue in poverty estimation. Our estimated average calorie requirement for the Indian population comes out to be about 2,290 calorie per capita per day for the rural areas and 2,250 calorie per capita per day for the urban areas. We are of the view that these average calorie norms, which, in our opinion, reflect better

    (1999-2000)

    Statement A-3: State-wise Average Daily Per Capita Calorie Requirements and the Relative Standard Errors (rse)of the Estimates (Per Cent)

    State/UT As Per Proposed Classification of Work Calorie rse (1) (2) (3) Andhra Pradesh 2425 0.79 Arunachal Pradesh 2161 0.62 Assam 2252 0.45 Bihar 2238 0.05 Goa 2332 0.69 Gujarat 2335 0.35 Haryana 2207 0.36 Himachal Pradesh 2244 0.04 J and K 2217 0.47 Karnataka 2372 0.15 Kerala 2318 0.26 Madhya Pradesh 2312 0.83 Maharashtra 2363 0.25 Manipur 2202 0.95 Meghalaya 2218 1.31 Mizoram 2187 0.91 Nagaland 2210 0.24 Orissa 2312 0.43 Punjab 2248 0.30 Rajasthan 2194 0.49 Sikkim 2240 0.60 Tamil Nadu 2467 0.46 Tripura 2285 0.11 Uttar Pradesh 2172 0.31 West Bengal 2277 0.58 Andaman and Nicobar Islands 2268 0.43 Chandigarh 2475 0.11 D and N Haveli 2281 0.36 Daman and Diu 2467 0.29 Delhi 2263 0.63 Lakshadweep 2176 3.80 Pondicherry 2428 0.17 All-India 2286 0.08 Rural As Per Task Force Sample Size Classification (No of of Work Persons) Calorie rse (4) (5) (6) 2476 0.69 22236 2315 1.39 4074 2265 0.54 18742 2303 0.03 39782 2261 1.24 866 2412 0.29 12813 2258 0.20 6490 2332 0.13 7845 2336 1.20 8000 2449 0.09 14328 2222 0.04 12139 2439 0.46 28533 2449 0.20 20468 2350 1.40 3928 2468 0.79 4773 2436 1.08 2116 2419 0.47 2390 2367 0.41 17052 2291 0.24 11875 2311 0.62 18834 2367 0.18 5185 2458 0.44 17214 2265 0.58 4677 2261 0.12 56929 2280 0.71 23233 2257 0.34 1662 2443 6.45 674 2355 0.84 908 2307 0.17 771 2190 0.83 967 2090 1.96 455 2371 0.09 1229 2350 0.06 371188 Urban As Per As Per Task Force Sample Size Proposed Classification Classification (No of of Work of Work Persons) Calorie rse Calorie rse (7) (8) (9) (10) (11) 2256 0.16 2152 0.21 16449 2158 2.58 2122 2.61 898 2197 0.58 2125 0.55 3476 2185 0.29 2115 0.23 11430 2295 0.58 2190 0.64 1173 2256 0.51 2150 0.38 12896 2232 0.10 2130 0.66 3690 2227 1.20 2145 1.12 3408 2163 1.97 2118 0.73 4965 2263 0.14 2163 0.11 11230 2293 0.15 2166 0.20 9155 2233 0.04 2143 0.10 16261 2277 0.04 2148 0.18 24005 2193 0.23 2176 0.03 3420 2167 0.30 2097 0.05 1636 2201 0.68 2191 0.47 4082 2150 2.51 2110 2.31 1107 2257 0.80 2162 1.34 4584 2266 0.90 2145 0.66 8418 2209 0.79 2133 0.69 10019 2254 0.75 2143 1.10 1080 2309 0.32 2172 0.37 16606 2195 0.19 2124 0.26 2400 2200 0.38 2106 0.27 24832 2281 0.14 2150 0.39 14167 2282 2.73 2148 2.35 765 2226 0.93 2115 0.63 3165 2274 1.45 2163 2.12 381 2270 1.08 2098 0.10 959 2230 0.56 2128 0.21 4946 2212 0.20 2110 0.74 2314 2313 1.46 2182 0.55 1583 2249 0.08 2142 0.04 225500
    3114 Economic and Political Weekly July 28, 2007

    approximation to real situation, be considered for the purpose of poverty estimation.

    The official estimate of poverty line for the year 1999-2000 is Rs 327.56 for rural India and Rs 454.11 for urban India. The corresponding estimate of HCR has been placed at 27.09 per cent and 23.62 per cent respectively for the rural and urban areas. These are based on the consumer expenditure data of NSS 55th round on a “30-day recall basis”.13 Keeping other issues of estimation including the average calorie norms unchanged, had one considered the year 1972-73 instead of 1973-74 as the base year, the estimated poverty lines for the year 1999-2000 on 30-day recall basis would have been Rs 371.00 for rural India and Rs 534.20 for urban India.14 Accordingly, the HCR for the year 1999-2000 roughly works out to be 38.18 per cent and 34.41 per cent respectively for the rural and urban areas, i e, higher than the released estimate by about 11 percentage points.

    Finally, with 1972-73 as the base year and with our suggested average calorie norms, i e, 2290 calories (rural)/2250 calories (urban), the poverty line for the year 1999-2000 is estimated as Rs 334.23 for rural India and Rs 614.46 for urban India. And the corresponding estimates of HCR are found to be about

    28.76 per cent and 44.18 per cent for the rural and urban areas respectively.

    We do not, however, present the estimate of either the poverty line or the HCR based on the consumption basket as

    Estimated Average Daily Per Capita Calorie Requirement

    as Per Proposed Classification of Work, 1999-2000

    Statement A-4: States Sorted in Increasing Order of

    Rural Areas Urban Areas
    State/UT Per Capita rse State/UT Per Capita rse
    Calorie (Per Calorie (Per
    Require- Cent) Require- Cent)
    ment ment
    (1) (2) (3) (1) (2) (3)
    Arunachal Pradesh 2161 0.62 Nagaland 2150 2.51
    Uttar Pradesh 2172 0.31 Arunachal Pradesh 2158 2.58
    Lakshadweep 2176 3.80 J and K 2163 1.97
    Mizoram 2187 0.91 Meghalaya 2167 0.30
    Rajasthan 2194 0.49 Bihar 2185 0.29
    Manipur 2202 0.95 Manipur 2193 0.23
    Haryana 2207 0.36 Tripura 2195 0.19
    Nagaland 2210 0.24 Assam 2197 0.58
    J and K 2217 0.47 Uttar Pradesh 2200 0.38
    Meghalaya 2218 1.31 Mizoram 2201 0.68
    Bihar 2238 0.05 Rajasthan 2209 0.79
    Sikkim 2240 0.60 Lakshadweep 2212 0.20
    Himachal Pradesh 2244 0.04 Chandigarh 2226 0.93
    Punjab 2248 0.30 Himachal Pradesh 2227 1.20
    Assam 2252 0.45 Delhi 2230 0.56
    Delhi 2263 0.63 Haryana 2232 0.10
    A and N Is 2268 0.43 Madhya Pradesh 2233 0.04
    West Bengal 2277 0.58 All-India 2249 0.08
    D and N Haveli 2281 0.36 Sikkim 2254 0.75
    Tripura 2285 0.11 Andhra Pradesh 2256 0.16
    All-India 2286 0.08 Gujarat 2256 0.51
    Madhya Pradesh 2312 0.83 Orissa 2257 0.80
    Orissa 2312 0.43 Karnataka 2263 0.14
    Kerala 2318 0.26 Punjab 2266 0.90
    Goa 2332 0.69 Daman and Diu 2270 1.08
    Gujarat 2335 0.35 D and N Haveli 2274 1.45
    Maharashtra 2363 0.25 Maharashtra 2277 0.04
    Karnataka 2372 0.15 West Bengal 2281 0.14
    Andhra Pradesh 2425 0.79 A and N Is 2282* 2.73
    Pondicherry 2428 0.17 Kerala 2293 0.15
    Tamil Nadu 2467 0.46 Goa 2295 0.58
    Daman and Diu 2467 0.29 Tamil Nadu 2309 0.32
    Chandigarh 2475 0.11 Pondicherry 2313 1.46

    per the latest quinquennial NSS rounds, which is beyond the scope of the theme of the present article.

    EPW

    Email: mannagcm@yahoo.com

    Notes

    [The author is grateful to Suresh D Tendulkar for his valuable comments and suggestions on the draft of this article. The author is also grateful to Dipankar Coondoo and R Panda of for their valuable guidance in preparing this article. The author is thankful to the union ministry of statistics and programme implementation for making available unit level data of NSS 50th and 55th rounds required for the present study. Views expressed in the article are of the author and not of the organisation to which he belongs.]

    1 Task Force on Projection of Minimum Needs and Effective Consumption Demand, Perspective Planning Division, Planning Commission [Planning Commission 1979].

    2 For the procedural details, see Planning Commission (1993). 3 See Report of the Task Force for further details. 4 It is, however, difficult to comment on the extent of actual sampling

    errors of results based on these rounds due to non-availability of unitlevel data on the electronic media.

    5 Only the rural sample size for the alternate base year 1 (1971-72) is somewhat lower than the corresponding sample size for the official base year (1973-74).

    6 See the Report of the Task Force [Planning Commission 1979] and Datta

    and Sharma (2002). 7 See Report of the Task Force [Planning Commission 1979, p 6]. 8 See Instructions to Field Staff, Volume I, p A-20, NSS 61st Round,

    July 2004-June 2005 [National Sample Survey 2004].

    9 We have not considered data of “revisited samples” to give equal weight, in terms of sample size, to each of the four sub-rounds. See Instructions Manual, NSS 52nd Round for the concept of revisit [NSS 1995].

    10 In the NSS, sample villages/blocks are selected in the form of a few (normally two) independent sub-samples.

    11 In NSS 55th round, a set of sample villages/blocks was revisited for collecting data on employment-unemployment.

    12 Considering CPIIW in case of urban for price adjustments.

    13 In the 55th round, consumer expenditure on clothing, footwear, medical (institutional) and durable goods were collected using a reference period of 365 days. For the remaining non-food items, a 30-day recall period was adopted. For food items, two reference periods of last 30 days and last 7 days were adopted for collecting the related information from the same set of households.

    14 Obtained by inflating the official all-India poverty line (“implicit”) for the year 1999-2000 by the ratio of figures under column 6 (considering CPIIW for urban) to column 7 of Table 3.

    References

    Planning Commission (undated): ‘Press Releases on Poverty Estimates for the Years 1993-94 and 1999-2000’.

  • (1979): Report of the Task Force on Projections of Minimum Needs and Effective Consumption Demand, Perspective Planning Division, January.
  • (1993): Report of the Expert Group on Estimation of Proportion and Number of Poor, Perspective Planning Division, July. Datta, K L and Savita Sharma (2002): Facets of Indian Poverty, Concept, New Delhi. ICMR (1988): Nutrient Requirements and Recommended Dietary Allowances
  • for Indians, Indian Council of Medical Research, reprint, 1992.

    NCO (1968): National Classification of Occupation.

    NSS Report No 238, Volumes I/II: Calorie and Protein Values of Food Items Consumed Per Diem Per Consumer Unit in Rural/Urban Areas, NSS 26th Round.

    Sarvekshana, Vol VI, Nos 3-4; January-April, 1983 (For Results on NSS 27th Round).

    NSS Report No 240: Tables on Consumer Expenditure, NSS 28th Round.

    NSS Report No 457: Level and Pattern of Consumer Expenditure in India, 1999-2000.

    NSS Report No 471: Nutritional Intake in India, 1999-2000.

    NSS (1995): Instructions Manual, NSS 52nd Round, National Sample

    Survey. NSS (2004): Instructions to Field Staff, Vol I, NSS 61st Round (July 2004-June 2005), National Sample Survey.

    Economic and Political Weekly July 28, 2007

    Dear reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Comments

    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top