After a year and a half of work, the 16- member panel headed by Niti Aayog Vice-Chairman has failed to come to any decision on India’s official poverty line. Instead, it has recommended setting up yet another group to do the job (Hindustan Times, September 12, 2016). The last eight years have seen three expert committees set up to define the poverty line. But each effort has been mired in controversy.
The magnitude of poverty and its alleviation carries enormous importance in public policy. Understandably, almost every government in India in the past few decades has constituted an expert committee for setting up the poverty line so as to facilitate designing programmes and policies for targeted intervention. Justifiably, there was disquiet in the present government, which came to power with the promise of employment generation, meeting the basic needs of housing sanitation, health and education and ushering in inclusive development, for not making any specific statement on poverty line, trends and targets. The setting up of an expert committee to determine the poverty line, on the recommendation of the Niti Aayog, was in response to this expectation.
Poverty at the household level is measured based on information on income, consumption expenditure, socio-economic capabilities, physical assets, housing conditions and access to amenities (Planning Commission (2012a) and Alkire (2011). Until recently, the expert groups have generally relied on information made available by the National Sample Survey Organisation (NSSO) for this task.
Interestingly, Planning Commission is not the only agency in India which has computed and released poverty estimates for use in the policy domain and mobilisation of public opinion. Institutions such as the Ministry of Rural Development (2009), National Commission for Enterprises in the unorganised sector (NCEUS, 2008) and a host of individual researchers (Deaton and Dreze 2009) have attempted determination of the poverty line and provided estimates on poverty. The estimates given by them differ widely; this is reflected also in the uncertainty and lack of definitiveness on the part of the government. The latter responded to public resentment on the estimates by setting up yet another committee. All this has resulted in a plethora of methodologies and widely varying estimates. Consequently, it is not very clear which of these would be used by a government department for identification of beneficiaries at the ground level. Given this background, establishing an Expert Group to define poverty by the new government was long overdue.
Methods adopted by the Planning Commission to estimate poverty
The expert group headed by D T Lakdawala (EGL) proposed a methodology for determination of the poverty line basket (PLB) of goods and services based on ‘certain normative calorie requirement’, using actual information of consumption expenditure at the national level (Planning Commission 1993). Poverty lines, defined by the EGL in terms of per capita total consumer expenditure at 1973-74 market prices, separately for rural and urban areas, were adjusted over time across states for changing prices, by keeping the PLB unchanged. The consumption basket of the households which met the minimum calorific requirement (2400 for rural and 2100 for urban areas) was taken as the PLB. Understandably, the basket included all non-food items of these benchmark households, along with their food items. Households having consumption expenditure less than that were considered poor. Once accepted by the Planning Commission, it became the basis for poverty calculations and programmatic interventions by different ministries and government departments for over a decade.
The assumption underlying the methodology was either the social services such as health and education would continue to be supplied by the state or that the amount spent by the benchmark households in 1973-74 on these items would be adequate. The withdrawal of the state from provisioning of public services and significant increase in private expenditure on these over the past three decades has rendered the PLB completely irrelevant. Non-inclusion of substantial private expenditure on health and education in the poverty line elicited sharp criticism from civil society. Anchoring poverty to per capita calorie norm alone was dismissed also because research showed no relationship between calorie intake and nutritional or health outcomes.
Recommendations of the Tendulkar Committee
The expert group headed by S D Tendulkar (EGT) set up by the Planning Commission (2009) proposed an alternate framework by avowedly delinking the poverty line from normative calorie requirement. It proposed a computational procedure for poverty measurement, taking the changes in the pattern of consumption, price configurations, differences in state engagement in provision of basic services over the past decades and availability of these to different sections of population into consideration.
The EGT assumed that the consumption basket in urban areas at national level, associated with the urban poverty figure of 27.5 in 2004-05 (as determined earlier by EGL and accepted by Planning Commission) met the minimal needs (as there was no major controversy around that), including education and health-care for the poor. This was used as the basis for computing poverty line for rural and urban areas at state level using appropriate price indices. It recommended shifting from uniform recall period for consumption expenditure for all items to mixed recall period to avoid recall error, since the periodicity of buying different items (such as bread and milk on one hand and shoes and clothing on the other) happen to be different (Planning Commission 2009).
The use of the urban PLB for rural areas, as proposed by EGT, resulted in a significant upward revision of the rural poverty figure from 29 per cent to 41.8 per cent, although urban poverty in 2004-05 remained pegged at 25.7 per cent. The latter did not change simply because the methodology did not allow it to be changed.
Debate and controversy
The concept of poverty, seen as manifestation of hunger and denial of minimum biological requirements, has earned it a special space in human rights discourse. Delinking of poverty line from the calorie norm resulted in loss of its universal acceptability and poignant appeal. As a result, a large number of activists and researchers spoke of providing the basic necessities within human rights framework rather than targeting the poor (Srinivasan 2007). They pleaded for universal coverage of one and all and argued that a poverty alleviation scheme should not have an entry barrier or mandatory exit provisions.
The Tendulkar Committee estimates of poverty have come under severe attack for erring on the lower side. It was argued that EGT adopted a conservative approach, owing to the government’s unwillingness to allocate more funds for poverty alleviation. Civil society appreciates scholars and committees that give high poverty figures and recognises them as progressive and pro-poor. Unfortunately, such efforts to place the poverty line at a higher level helped the middle class to get bracketed with the poor and corner a large part of the benefits, allotted in the name of the poor, through their social and political linkages.
The assertion that urban poverty figure has not changed over time has also been questioned based on certain disturbing trends in urban areas. It is well recognised that elitist planning in most urban centres, particularly the large cities, has pushed the poor to the peripheries and segmented settlements. This has resulted in their increased distance of daily travel, higher requirement of calories and larger transport expenditure. The drudgery of their work also appears to have gone up relatively due to increased informalisation of labour force and extendable working hours.
The EGT notes that the actual calorie intake of those near the poverty line in urban areas is 1776, much below the calorie norm of 2100 per capita per day. It, however, did not see this as a problem since the Food and Agriculture Organisation (FAO 2004) has revised the norm to 1770 per capita per day for Asian countries. This, however, has been dismissed by researchers and civil society as unscientific and politically motivated. There is general agreement that the concern to ensure the basic calorie requirement to vulnerable sections of the society cannot be dismissed, at least for some time to come.
The normative level of expenditure on education and health care in rural areas, built into the poverty line is noted to be less than the actual private expenditure by 14 per cent. The figure for urban areas was even higher—22 per cent. Clearly, the poverty line normative expenditure on these, grossly underestimates the actual needs of the people.
More importantly, self-reported morbidity level, which has been considered for computing normative medical expenditure, would be much below the ailment based on medical examination. The 60th Round of NSSO, shows that there has been an increase in the percentage of ailing persons by 4.5 percentage points in urban areas, against 3.3 point increase in rural areas over a period of seven years. This is understandably because of an increase in the awareness of people on health issues. It would be logical to assume that the real rate of ailment is even higher than that reported in the 60th round and is likely to go up with greater knowledge and modernity. The NSSO data further suggest that the level of morbidity rises systematically with rise in consumption expenditure. The incidence of ailments is the lowest among the STs followed by SCs. It is the highest among the upper caste population who report high consumption expenditure, in both rural and urban areas. It is thus evident that reported health care expenditure of the benchmark households, which constitutes the basis for validation of the normative expenditure for poverty line, grossly underestimates the actual requirement of the people.
Incidentally, only 11 per cent of the households in the highest income groups depended on government facilities, the figure being 26 per cent for the poorest class. Similarly, as against the national average of 19 per cent urban households going to government facilities, 30 per cent do that among the scheduled tribes and 24 per cent among the scheduled castes. The percentage of households going to private facilities in urban areas for treatment has gone up from 76 per cent to 81 per cent from 1986-87 to 2004, while the dependence on private facilities in rural areas has remained unchanged. Furthermore, the dependence on government facilities for hospitalisation has come down from 60 per cent in both rural and urban areas in 1986-87, to 38 per cent in urban India and 42 per cent in rural areas by 2004. Given these disturbing trends, one would argue that the poverty line of the EGT would not make sense unless there is a change in the perspective of the government to extend medical facilities to poor and vulnerable sections of population and reverse the trend of their shifting from government to private facilities.
Perspective of expert group headed by Rangarajan
Owing to the academic criticisms and massive social mobilisation against the poverty estimates of EGT, another expert group, chaired by C. Rangarajan (EGR), was appointed. The latter proposed a significant departure from the EGT and, in a certain sense, reverted back to the EGL perspective.
EGR recommended using separate poverty line baskets for rural and urban areas, as opposed to keeping the urban poverty line basket central to poverty calculations, as proposed by EGT. Furthermore, EGR decided that the poverty line should be anchored to certain levels of adequate nourishment. It computed the average requirements of calories, proteins and fats based on the norms of Indian Council of Medical Research, differentiated by age, gender and activity levels of people in rural and urban areas. A food basket which meets the normative requirements of the three nutrients constitutes a part of the PLB.
Interestingly, EGR decided to include certain non-food expenses including clothing, house rent, conveyance and education into poverty calculations. The new poverty line works out to monthly per capita consumption expenditure of INR 972 in rural areas and INR 1, 407 in urban areas in 2011-12, much higher than what would be obtained as per the EGT in this year. The poverty lines estimated by the EGR, thus, work out as 19 per cent and 41 per cent higher in rural and urban areas compared to that of EGT. It estimated that 30.9 per cent of the rural population and 26.4 per cent of the urban population lived below the poverty line in 2011-12. The all-India ratio was 29.5 per cent. The poverty ratio was, however, noted to have declined from 39.6 per cent in 2009-10 in rural India and from 35.1 per cent in urban India.
EGR considered poverty to be a multi-dimensional phenomenon but decided against using multiple indicators for measuring poverty. Given the inherent subjectivity involved in selection of indicators and their aggregation by giving weight ages, it stayed with per capita consumption expenditure for articulating deprivation. It, nonetheless, proposed an alternate approach to estimate poverty, based on the ability of households to save.
EGR, unfortunately, left a host of issues and controversies concerning the norms of intake for different nutrients, substitution of one for the other and norms for non food items, unanswered. Given the openness of the methodology, it elicited no controversy. Nonetheless, it failed to provide a clear road map to measure poverty.
Ground level identification of poor in rural and urban areas
Given the difficulties of applying the methodologies proposed by different expert groups, as discussed above at the field level, the government appointed two expert groups chaired by N C Saxena (EGS) and S R Hashim (EGH). The former was for proposing criteria for identifying the poor in rural areas and the latter for urban areas. The database for their operationalisation was to be built through the Socio Economic Caste Census (SECC) 2011.
Taking a cue from the BPL Census methodology adopted in 1997, EGS proposed a three stage procedure for (a) automatic exclusion, (b) automatic inclusion and (c) scoring of the remaining households. A household is to be excluded from the list of the poor or deprived, if it owned: (i) a three or four wheeled vehicle, (ii) mechanised farm equipment, such as tractor, power tiller, thresher or harvester and (iii) land above twice the district average, if irrigated and three times, if unirrigated. Furthermore, households that had a member as (a) a government employee, (b) earning more than INR 10,000 per month or (c) paying income tax, are not to be considered for inclusion in the poverty list. Correspondingly, households with any of the following deprivations would automatically be included in the list, such as being : (a) primitive tribal, (b) Maha Dalit (c) dependent on alms, (d) legally released bonded labourers, (e) not having a shelter and (f) headed by a child or disabled.
The remaining households are to be ranked based on a composite deprivation score, obtained on the basis of a set of indicators and their weightages, as noted below:
- SC/ST household (3 points), denotified and ‘most backward castes’ (2 points), Muslims/OBC (1 point);
- Landless agricultural households (4 points), labourers with some land (3 points), casual workers, artisans and fishermen (2 points);
- No adult member between the ages of 16 and 59 (1 point);
- A member with disability, TB, leprosy, mental illness, AIDS (1 point);
- All members above 30 years of age being illiterate or neo literate viz education up to class 5 (1 point).
The priorities among the households securing identical scores in 0 to 10 scale, as presented above, are to be determined in the following sequence (1) SC/ST landless labourers, (2) Other landless labourers, (3) SC/ST labourers with some land and (4) Other agricultural labourers with some land. The Gram Sabha’s judgment is to be given due weight age in deciding about their inclusion.
The expert group headed by Hashim (EGH) recommended an identical three-stage methodology for identification (Planning Commission 2012b). Automatic exclusion of households from BPL list is to be based on the following criteria: (a) possession of a house with four or more dwelling rooms, (b) possession of any of the following: motorised vehicle, air conditioner, computer or laptop with internet, (c) possession of any three of the following assets: refrigerator, land-line telephone, washing machine and two wheeler motorised vehicle. The remaining households are to be screened for automatic inclusion as per three types of vulnerabilities:
Residential vulnerability: homeless or residing in a house made of plastic/polythene or in one room unit with walls and roof of grass, thatch, bamboo, mud, unburnt brick or wood;
Occupational vulnerability: no household income, having a member engaged in a vulnerable occupation like begging/rag picking, domestic work, sweeping or gardening and all earning members engaged on daily basis; and,
Social vulnerability: child-headed household, no able-bodied person in the 18-60 age group and all earning members either disabled, chronically ill or aged more than 65 years.
The residual households are to be assigned scores in stage 3. They are to be considered for inclusion as per their deprivation score. The households with score zero are not eligible to be in the BPL List.
The release of the SECC data in early July 2015 has, unfortunately, led to sharp reactions in political as well as academic circles. The first line of criticism is that bringing caste data into the public domain was purposely delayed, as was the Census data on religion. Although a part of the socio-economic data has been brought into public domain, the caste specific information has not yet been released on grounds of inaccuracy and inconsistency. Secondly, it has been pointed out that, owing to public knowledge that this Census was being conducted to identify the poor for anti-poverty programmes, households, have purposely reported higher levels of deprivation.
Repeated attempts at developing a methodology to study poverty has led to the realisation that univariate analysis or determination of poverty based on one indicator or one poverty line would meet with resentment and resistance from professionals, researchers and social activists who hold that poverty has multifarious manifestations. Given the lack of robustness in the data on consumption and changing role of the state in providing basic amenities, inclusion of social and physical indicators relating to caste, quality of housing, physical disability and the like for inclusion and exclusion of households at micro level is gaining currency. Devising a simple methodology of scoring based on a limited number of indicators, with low measurement error, therefore, is the need of the hour.
Alkire, S. 2011. Country Briefing: India, Multidimensional Poverty Index at a Glance, Oxford Poverty and Human Development Initiative, Oxford.
Deaton A. and J. Drèze. 2009. Food and Nutrition in India: Facts and Interpretations, Economic and Political Weekly, 14 February.
Food and Agriculture Organisation, Statistics Division. 2004. Minimum Dietary Energy Requirement (kcal/person/day), Jointly released by FAO, WHO and UNU, Rome.
Ministry of Rural Development. 2009. Report of the Expert Group to Advise the Ministry of Rural Development on the Methodology for BPL Census for 11th Five Year Plan, Ministry of Rural Development (Saxena Committee), Government of India, New Delhi.
National Commission for Enterprises in the Unorganised Sector. 2008. Report on Conditions of Work and Promotion of Livelihoods in the Unorganised Sector (Arjun Sengupta Committee), Government of India, New Delhi.
National Sample Survey Organisation. 2006. Morbidity, Health Care and the Conditions of the Aged, Government of India, New Delhi.
Planning Commission. 1993. Report of the Expert Group on Estimation of Proportion and Number of Poor (Lakdawala Committee), Government of India, New Delhi.
Planning Commission. 2009. Report of the Expert Group to Review the Methodology for Estimation of Poverty (Tendulkar Committee), Government of India, November.
Planning Commission. 2012a. Press Note on Poverty Estimates 2009-10, Government of India, New Delhi.
Planning Commission. 2012b. Report of the Expert Group to Recommend the Detailed Methodology for Identification of Families Living Below Poverty Line in the Urban Areas (Hashim Committee), Government of India, New Delhi.
Saxena N. C. 2009. The Moving Bar of Poverty Statistics, Business World, September 25.
Srinivasan T. N. 2007. Poverty Lines in India: Reflections after the Patna Conference, Economic and Political Weekly, vol 42, no 41.