Understanding the Distribution of BPL Cards: All-India and Selected States

Understanding the Distribution of BPL Cards: All-India and Selected States

F Ram, S K Mohanty, Usha Ram

Using the recent National Family and Health Survey-3 data, this paper examines the distribution of below poverty line cards. The possession of BPL cards by the households in various economic and social settings index is analysed. The findings suggest that about two-fifths of the BPL cards in India are with the non-poor households. On the other hand, in many of the states a majority of households in abject deprived groups do not possess a BPL card. The extent of misuse is higher in Andhra Pradesh, Karnataka and Kerala, while it is lower in Tamil Nadu. In economically weaker states like Orissa and Bihar, a higher proportion of non-poor households possess a BPL card.

F Ram (director@iips.net), S K Mohanty and Usha Ram are all at the International Institute for Population Sciences, Mumbai.

number of welfare schemes funded by federal and state governments are aimed at the upliftment of poor and the reduction of inequality.

1 Background

In order to identify the poor, the concept of “below poverty line” (BPL) is used. The local agencies with the guidelines of the ministry of rural development of the government of India carry out the exercise of identifying the BPL households and then distribute the cards. The households that possess a BPL card beneﬁt from such welfare schemes. Over the years, these schemes have been diversiﬁed, ranging from ration cards under the public distribution system which entitled them to obtain foodgrain, kerosene, cooking gas, etc, at highly subsidised rates, free housing, old age pension, free/subsidised healthcare services, etc. The provision of edible oil at subsidised rates to the BPL cardholders is a recent addition to the already existing long list of free or subsidised items earmarked for the poor. These beneﬁts account for a very large part of the national and state government expenditures and therefore their proper utilisation is often debated.

The BPL cards are distributed after identifying the households based on population based surveys in each state in accordance with the guidance provided by the ministry of rural development. So far, three BPL surveys – 1992, 1997 and 2002 – have been conducted throughout the country using three different methodo logies. The BPL survey conducted in 1992 used the household income criteria with a limit of Rs 11,000 annually. The 1997 BPL survey used two schedules, namely, schedule A and schedule B for identifying the poor households. Schedule A used the exclusion criteria based on any of the ﬁve variables, namely, ownership of a pucca house, annual income of the family being more than Rs 20,000, ownership of more than two hectares of land, ownership of consumer durables such as a television, refrigerator, ceiling fan, motorcycle/scooter, three-wheeler and ownership of farm equipment such as a tractor, power tiller, or combined threshers/harvester. The underlying assumption was to exclude the “visibly non-poor” families. Schedule B canvassed only the non-poor families (identiﬁed using schedule A), about the consumption expenditure in the last 30 days along with other demographic and social information. If the monthly per capita household expenditure was less than the Planning Commission’s estimates of the poverty line, the household was categorised as poor, otherwise not. This methodology was subject to intense criticism for the exclusion criterion, non-availability of ofﬁcial poverty line for all states, inconsistency of results of ofﬁcial poverty estimates

february 14, 2009 vol xliv no 7

Economic & Political Weekly

SPECIAL ARTICLE

of the Planning Commission and absence of any provision for inclusion of persons who have subsequently become poor to the BPL list (Sundaram 2003).

Before launching the 2002 BPL survey, an expert group was constituted in November 2001 with senior ofﬁcials of the central government, academia, state government representatives and other professionals to advise on the methodology and analysis of data. The committee recommended a score-based ranking of each household indicating the quality of life. With a score ranging from 0 to 4, a set of 13 socio-economic indicators which included the size of the operational landholding, type of house, availability of clothes, food security, sanitation, ownership of consumer durables, literacy status, status of household labour force, means of livelihood, status of school-going children, type of indebtedness, reason of migration and preference of assistance, were used. The total score ranged from 0 to 52 and the states were given ﬂexibility of deciding the cut-off points. However, the 2002 BPL lists were not operationalised until mid-February 2006 due to a stay order passed by the Supreme Court on a writ petition ﬁled by the People’s Union for Civil Liberties. The petitioner alleged that the new methodology would reduce the number of persons identiﬁed as BPL and a large number of the poor families would lose their entitlements.

2 Need for the Study

Besides the ongoing debate on the methodologies used in various BPL surveys, there has been some amount of discontent among the people with respect to the sincerity in identifying and the distribution of BPL cards. It is often believed that the distribution of BPL cards is politically inﬂuenced and the rich probably beneﬁt more than the poor irrespective of the criterion adopted in identifying the poor households. A recent study conducted by the National Council of Applied Economic Research (NCAER) in six states found that 40% of the BPL cards have been issued to people who are above the poverty line (APL), ranging from 84% in Assam, 43% in Uttar Pradesh, 50% in Rajasthan, 40% in Bihar and 38% in Chhattisgarh. Even the Antodaya Anna Yojana (AAY), initiated during 2000, to provide super-subsidised foodgrains for the poorest among the BPL category also showed similar misuse (Bihar Times, 27 December 2007).

The NCAER study was the ﬁrst of its kind which provided an insight into the extent of misuse of the BPL cards in India. However, the study was limited to six states. In the present paper we therefore attempt to throw more light on some of these issues using data from the National Family and Health Survey-3 (NFHS-3).

3 Objectives

The broad objective of the study is to understand the social and economic context of the distribution of the BPL cards in India. The speciﬁc objectives are: (1) to examine the consistency in distribution of BPL cards by selected social and economic characteristics of the households; and (2) to estimate the extent of BPL cards among non-poor in India and a few selected states.

4 Data and Methods

The three rounds of NFHS, NFHS-1 conducted during 1992-93, NFHS-2 conducted during 1998-99, and NFHS-3 conducted during 2005-06 are large sample surveys, conducted under very scientiﬁc

Economic & Political Weekly

february 14, 2009 vol xliv no 7

sampling design and survey procedures. All these rounds of survey are nationally representative and cover more than 99% of the country’s population. There were improvements in coverage of topics, methodology and sample size in subsequent rounds and a number of additional issues (such as family life, education, prevalence of HIV/AIDS, etc) were covered in the recent round of the survey. The NFHS-3, for the ﬁrst time, collected information on the possession of BPL cards along with other socio-economic characteristics of households in the household schedule. The survey covered a total of 1,09,041 households in all the states of the country. The survey asked the question: “Does the household have a BPL card?”. The responses were coded as “yes”, “no” and “do not know”. The “do not know” category constitutes 0.5% of the total cases and has been added with “no” in the present analysis. It may be mentioned that the missing cases (less than 0.5%) were not included in the analysis.

To understand the association with economic characteristics of households, the variables such as wealth index and other economic indicators are used. The wealth index based on consumer durables, housing quality, water and sanitation facility of the household has been largely used as a proxy for economic status in many demographic and health surveys (Montgomery et al 2000; Vyas and Kumaranayake 2006). It reﬂects the household’s long term economic status and is used as a proxy for the economic well-being of households (Filmer and Pritchett 2001).

In the present analysis, the wealth index, as constructed in the NFHS-3, has been used as the key economic variable to classify the households as poor and non-poor. The wealth index is constructed using the principal component analysis. Wealth quintiles are obtained such that each quintile consists of nearly 20% of the surveyed population. However, the wealth index is computed only at the national level and no separate index is computed for the states and hence for state level analysis national weights have been applied. The details of the variables in construction of wealth index may be found in NFHS-3 report (IIPS and ORg Macro 2007).

In the present analysis, we conceptualise that the households belonging to the third, fourth and ﬁfth wealth quintiles are not poor and in the subsequent analysis we refer to them as nonpoor households. This is logical given the fact that as per the Planning Commission estimates based on consumption expenditure data on uniform recall period, 21.8% and on mixed recall period (Planning Commission 2007) 27.5% of the country’s total population lives BPL.

Additionally, in the analysis, we have included individual items like the type of house, motorcycle, scooter, car, tractor, television, refrigerator, landline telephone, mobile as well as the size of landholding. It may be mentioned that many of these variables were used for exclusion of the visibly non-poor in the 1997 BPL survey. Since the 2002 BPL survey was not operational at the time of NFHS-3, the BPL cards were more likely to be based on the 1997 BPL survey. A number of composite economic indicators; such as the own pucca house with three or more sleeping rooms (deﬁned as households living in pucca houses owned by them and having three or more sleeping rooms), possession of

SPECIAL ARTICLE

any motorised vehicles (motorcycle/scooter/car/tractor) and telephone (landline/mobile), have been obtained to assess the economic status of the households.

The social attributes used in the present analysis are the educational status of the household head, presence of any adult literate member in the household, caste and religion of the household. The analysis has been carried out for the entire country and for some selected states. Bivariate analysis has been carried out to examine differentials in possession of BPL cards and its socioeconomic associates. The estimation of the number of BPL cards by the poor and non-poor has also been worked out for India as a whole and a few selected states.

5 Results

The analysis has been presented in two sections, the national level and the state level results.

5.1 National Level

Table 1 presents the percentage distribution of households possessing a BPL card by selected economic indicators and wealth quintiles. About 27% of households in India hold a BPL card which is close to the Planning Commission estimates on poverty in 1999-2000 (26% based on mixed recall basis). The possession of BPL cards by wealth quintile suggests that 39% among the poorest households, 37% among the poorer, 32% among the middle, 21% among the richer and 7% among the richest hold BPL cards. On combining the ﬁrst and second quintile (poor) and the third, fourth and ﬁfth quintiles (non-poor), we notice that nationally, only two-ﬁfths of poor households and one-ﬁfth of non-poor households possess a BPL card. In other words, a majority of the poor do not possess a BPL card in the country. The pattern is similar in rural and urban areas with a relatively higher percentage in rural than urban areas possess a BPL card.

The possession of BPL cards by households with other speciﬁc economic characteristics is also given in the table. With respect to housing characteristics, it may be noted that about 11% households living in their own pucca houses having three or more sleeping rooms possess a BPL card. This is even higher in rural India (15%) than urban areas (8%). The possession of BPL cards by those owning motorised vehicles as well as those with access to means of modernity, such as households having a television or refrigerator, also shows a similar pattern. For example, about 10% of the households owning motorised vehicles and 8% of those owning both television and refrigerator possess BPL cards. This suggests that a substantial number of BPL cards are with the non-poor section of the society.

Landholding is a critical indicator in rural India. With respect to the size of landholding in rural India it is found that while 38% landless own a BPL card compared to 16% among those with 10 acres or more land. It is evident that there is gross misuse of BPL cards, to the extent that 20% among the non-poor in India avail this facility. At the same time, a large proportion of poor have been deprived from availing the BPL card.

We have also examined the possession of BPL cards by households based on the criteria of abject deprivation. Abject deprivation has been deﬁned as a situation where a household does not have any adult literate member, lives in a kaccha house in rural areas and in kaccha or semi pucca in urban areas, no land in rural areas and no toilet facility in urban areas, no drinking water facility of his or her own, not owning any consumer durables such as a bicycle, television or, radio and no electricity for

Table 1: Percentage of Households Possessing a BPL Card

(According to Selected Economic Characteristics, 2005-06)

Characteristics	Combined	Rural	Urban
%	N	%	N	%	N
Housing characteristics
No separate kitchen	30.6	34,612	35.0	24,929	19.2	9,683
Kaccha house	38.6	14,933	38.7	14,033	36.8	900
Semi pucca house	34.5	43,500	35.2	37,879	29.6	5,622
Pucca house	17.7	50,029	24.8	21,131	12.5	28,897
Own a pucca house with
three or more sleeping rooms	11.3	7,694	14.5	3,794	8.2 3,899
Ownership of motorised vehicle
Motorcycle or scooter	10.2	18,758	14.0	7,932	7.4 10,854
Car	5.0	2,894	10.8	734	3.1	2,160
Tractor	7.7	1,876	8.0	1,707	4.8	168
Any of the above three	10.1	19,489	14.1	8,165	7.1 11,323
Ownership of television and refrigerator
Television (any)	18.6	48,156	25.0	22,132	13.1	26,023
Refrigerator	8.1 16,722	14.6	4,821	5.4 11,901
Either television or refrigerator	18.7	49,848	24.9	23,417	13.1	26,431
Both television and refrigerator	6.8 15,599	11.4	4,081	5.2 4,081
Ownership of telephone and computer
Landline	9.4 15,338	14.7	5,851	6.1 9,487
Mobile	9.9 18,326	15.8	5,407	7.5 12,919
Computer	4.4	3,283	9.2	426	3.7	2,856
Any of the above three	10.8	25,701	16.1	9,363	7.8 16,338
Ownership of landholding
None	27.6	59,260	38.1	30,465	16.5	28,795
Less than 2.5 acres	30.4	31,572	32.3	27,945	15.5	3,629
2.5-5 acres	25.2	8,503	27.4	7,291	12.4	1,211
5-10 acres	19.8	5,540	21.8	4,583	10.2	957
10 acres or more	14.1	3,734	16.0	2,946	7.1	789
Wealth quintile
Poorest (Q1)	39.1	22,490	39.1	21,370	37.5	1,121
Poorer (Q2)	37.2	21,611	38.0	19,277	31.2	2,335
Middle (Q3)	31.5	21,646	33.0	16,477	26.7	5,169
Richer (Q4)	21.4	21,361	23.0	11,098	19.8	10,264
Richest (Q5)	7.1 21,903	9.2	5,219	6.4 16,684
Poor (Q1 and Q2)	38.2	44,102	38.6	40,646	33.2	3,456
Non-poor (Q3, Q4 and Q5)	20.0	64,909	25.8	32,794	14.0	32,115
Abject deprivation	40.0	1,763	40.4	1,508	38.0	255
All	27.3 1,09,011	32.9	73,441	15.8	35,573

his/her house. The deprivation score ranges from 0 to 6 where the score 0 is termed as abject deprivation (Srinivasan and Mohanty 2002). It may be noted that 60% of the households in the abject deprivation group do not have a BPL card. This is equally true in both rural and urban areas. This is a clear indication that the BPL card and the welfare schemes implemented based on BPL card have failed to reach the majority of the poorest of the poor group.

Table 2 (p 69) presents the percentage distribution of households possessing a BPL card by social characteristics. With respect to education, two of the variables, namely, the presence of any adult literate member and educational level of the head of the

february 14, 2009 vol xliv no 7

Economic & Political Weekly

SPECIAL ARTICLE

household is given in the table. It is found that among households Table 3 gives the estimated number of BPL cards in each of the without an adult literate member, only 37% have a BPL card. On wealth quintile as well as the percentage of non-poor holding a the contrary, among households with high school and above BPL card in India. The population estimate of the expert group education, about 7% have a BPL card. The possession of for the year 2006 has been used along with the other informa-BPL cards by caste groups indicates that it is higher among the tion from NFHS-3 (Ofﬁce of the Registrar General and Census Commissioner 2006). The proportion of households holding a

Table 2: Percentage of Households Possessing a BPL Card

(According to Selected Social Characteristics, 2005-06) BPL card declines with wealth quintile. Our estimates show that

Characteristics Combined Rural Urban

in India, the total number of BPL cards in 2006 were about 61

% N % % N %

million, of which 34 million were distributed to the poor house-

Educational level of household head No education 36.9 40,909 39.1 33,538 27.1 7,373holds while the remaining 27 million were distributed among Primary 31.5 20,265 35.0 14,848 22.0 5,418non-poor households. In other words, about 44% of BPL cards Secondary 19.9 38,007 24.6 22,073 13.3 15,934are distributed to the non-poor in the country. In other words, Higher 7.1 9,617 12.9 2,833 4.7 6,782 four out of every 10 BPL cards are going to the non-poor house-Sex of the household head

holds. Our estimate is close to that of NCAER estimate (of 40%)

Male 26.8 93,357 32.6 62,486 15.2 30,871

based on a study conducted in six states of India.

Female 30.4 15,654 34.8 10,954 20.3 4,701

Caste of the household head

5.2 State Level

Scheduled caste (SC) 34.9 20,970 39.5 15,193 22.9 5,777

In this section we provide the possession of BPL cards by economic

Scheduled tribe (ST) 42.0 9,189 44.3 81,68 23.8 1,021

and social characteristics of households in some states of India.

Other Backward Classes (OBC) 27.5 43,211 31.6 29,819 18.3 13,392 Others (Non-SC/ST/OBC) 18.9 35,641 25.3 20,260 10.5 15,381 The estimates of the number of BPL cards and the percentage of Religion of the household headnon-poor having BPL cards are given for a few selected states. We Hindu 27.9 88,947 33.5 61,145 15.5 27,803

have also used the similar assumption that the households in

Muslim 25.1 13,639 28.9 8,406 19.1 5,233

third, fourth and ﬁfth wealth quintile are visibly non-poor in the

Christian 24.1 2,952 30.9 1,677 15.1 1,275

states of India.

Sikh 16.5 1,710 21.5 1,200 4.9 510

Table 4 presents the percentage of households possessing a BPL

Others 30.9 1,762 43.5 1,012 14.0 750

card by economic variables as discussed earlier. It is found that

No adult literate in the household 36.7 21,623 38.3 18,589 26.4 3,034

the relatively better-off households possessing a BPL card is highest in Andhra Pradesh followed by Karnataka; incidentally both

Table 3: Estimated Number of BPL Cards by Wealth Quintiles (2006)

Wealth Quintile Number of Household Proportion of the Estimated Number of the states are considered to be progressive states and levelled as Having a BPL Card Households Having Households Having a BPL Card a BPL Card (in Millions)

Table 4: Percentage of Households Having a BPL Card

Poorest 8,784 0.2948 17.94 (by Selected Social and Economic Characteristics of the Households, 2005-06)

Percentage Households with a BPL Card

Poorer 8,050 0.2702 16.44

States Live in Own Who Have a Who Own Own More Than Middle 6,826 0.2291 13.94 Pucca House with Motorised Vehicle Television and Five Acres of Three or More Refrigerator Agricultural Land

Richer 4,580 0.1537 9.35

Sleeping Rooms

Richest 1,553 0.0521 3.17

Andhra Pradesh 39.5 37.1 29.2 66.3 All 29,793 1 60.85

Assam 5.5 3.6 1.0 9.2 Total population as of 2006a (in million) : 1,112.12

Bihar 12.7 13.1 7.0 8.7 Estimated number of total households (in million) : 222.88

Gujarat 11.2 9.8 6.4 18.2 Percentage of households having a BPL card (2006): 27.3

Haryana 1.8 4.7 5.3 1.3 Estimated number of households having a BPL card (in millions): 60.85

Himachal Pradesh 6.8 2.0 4.8 NC Households having a BPL card in 1st and 2nd wealth quintile (poor in million): 34.38

Jammu and Kashmir 18.2 6.5 6.0 NC Households having a BPL card in 3rd, 4th and 5th wealth quintile

Karnataka 29.2 23.3 15.8 33.1 (non-poor in million): 26.47

Kerala 23.3 10.6 8.3 NC Percentage of BPL cards with non-poor households: 43.5

Madhya Pradesh 10.8 10.2 5.5 16.4

a Based on estimated population for 2006 by the Office of the Registrar General and Census Commissioner of India 2006. The number of households has been estimated by dividing total Maharashtra 8.5 7.4 5.2 15.1 population by 4.99; assuming average household size at 4.99 persons per household.

Orissa 21.6 17.4 8.1 29.4

scheduled tribes (STs) followed by scheduled castes (SCs), Other Punjab 12.2 9.3 10.9 3.9 Rajasthan 3.7 3.9 2.7 10.0

Backward Classes (OBCs) and others. There are very little variations

Tamil Nadu 3.5 1.7 0.4 4.0

in possession of BPL cards by major religion; for example, about

West Bengal 5.3 5.8 2.1 2.2

24 to 28% of the households belonging to Hindu, Muslim and

Uttar Pradesh 3.6 4.4 2.7 6.1

Christian possess a BPL card. The rural-urban differences in the

Chhattisgarh 8.7 7.1 4.1 13.2

possession of BPL card by these characteristics are notable and

Jharkhand 8.8 9.2 2.0 27.8

signiﬁcant. There are a few categories where the urban-rural

Uttarakhand 11.1 7.0 4.9 30.9

differences are remarkable, for example, percentages of ST

North-eastern states 10.1 7.9 6.8 16.7

households possessing a BPL card in rural areas was 45% com

All-India 11.3 10.1 6.8 17.5 pared to 24% in urban areas. NC: Not computed due to small sample size.

Economic & Political Weekly

february 14, 2009 vol xliv no 7 69

SPECIAL ARTICLE

“good governance states”. This is true for all other economic variables as well. On the other hand, it is lowest in the case of Tamil Nadu. For example, among households owning a pucca house with three or more sleeping rooms, about 40% in Andhra Pradesh, 29% in Karnataka, 23% in Kerala and 22% in Orissa possess a BPL card. About 10 to 20% households living in own pucca houses with three or more living rooms in Bihar, Gujarat, Jammu and

Table 5: Percentage of Poor, Non-Poor and Abject Deprived Households Who Own BPL Card (2005-06)

States All Poor Non-Poor Abject Deprived Households Households Households Households (Q1 and Q2) (Q3, Q4 and Q5)

Andhra Pradesh 65.1 79.6 58.9 79.2

Assam 18.6 26.0 11.1 19.6

Bihar 38.6 48.6 23.3 49.5

Gujarat 25.4 46.3 19.6 22.7

Haryana 12.3 22.7 10.3 28.6

Himachal Pradesh 15.0 34.3 12.6 NC

Jammu 27.1 43.7 23.9 NC

Karnataka 46.8 66.2 37.0 60.0

Kerala 32.1 60.0 30.3 –

Madhya Pradesh 26.6 33.5 15.1 36.7

Maharashtra 19.5 37.5 12.7 34.6

Orissa 47.8 58.8 29.7 55.6

Punjab 17.7 30.4 16.5 –

Rajasthan 14.5 22.1 8.9 3.3

Tamil Nadu 7.8 15.2 4.8 14.8

West Bengal 24.5 34.9 14.7 32.3

Uttar Pradesh 14.4 20.4 7.8 22.2

Chhattisgarh 30.3 37.1 14.9 55.6

Jharkhand 32.3 42.4 11.7 NC

Uttarakhand 24.5 47.7 17.4 NC

North-eastern states 22.3 31.6 18.5 35.0

All-India 27.3 38.2 20.0 40.0

NC: Not computed due to small sample size.

Table 6: Percentage of BPL Cards with Non-Poor Households for Selected States of India (2005-06)

Variable Andhra Pradesh Karnataka Rajasthan West Bengal Uttar Pradesh

Population of the state in 2006 as per RGIa (in million) 80.72 56.26 62.28 85.22 183.3

Average number of person per household (2006) 4.0 4.8 5.5 4.7 5.9

Estimated number of households in 2006 (in million) 19.94 11.67 11.40 18.18 30.83

Percentage of households possessing a BPL card, 2005-06 65.1 46.8 14.5 24.5 14.4

Estimated number of households holding a BPL card (in million) 12.98 5.46 16.53 4.46 44.40

Percentage of poor households who possess a BPL card 0.369 0.477 0.645 0.695 0.745

Percentage of non-poor

households who possess

a BPL card 0.631 0.523 0.355 0.305 0.255

Estimated number of households (in million) with a BPL card in 2006 among Poor 4.79 2.60 10.66 3.10 33.07

Non-poor 8.19 2.86 5.87 1.36 11.33

Percentage of BPL card with non-poor households 63.1 52.3 35.5 30.5 25.5

a Based on estimated population for 2006 by the Office of the Registrar General and Census Commissioner of India 2006.

Kashmir, Madhya Pradesh, Uttarakhand and north-eastern states own BPL cards. On the other hand, the percentage of such households is fewer than 10% in the remaining states, for example, in Haryana (2%), Tamil Nadu (4%) and Rajasthan (4%). It may be pointed out that in Orissa where the poverty level is highest in the country, about 22% of non-poor households possess a BPL card. The pattern remains quite similar for other composite economic indicators, namely, the possession of a motorised vehicle, owning both televisions and refrigerators and having more than ﬁve acres of land. Among households with a motorised vehicle or both televisions and refrigerators or more than ﬁve acres of land, the percentage of those holding a BPL card is highest in Andhra Pradesh followed by Karnataka and is lowest in Tamil Nadu.

Table 5 presents the percentage of households in the poor (ﬁrst and second quintile), non-poor (third, fourth and ﬁfth quintile) and in abject deprivation possessing a BPL card. The ﬁndings suggest that the percentage of households possessing a BPL card is the highest in Andhra Pradesh (65%) followed by Karnataka and Orissa (47-48%) and is the lowest in Tamil Nadu.

The possession of BPL cards among the poor and the non-poor indicates that Andhra Pradesh once again tops the list as far as possession of BPL cards among non-poor households is concerned (65%). This is followed by Karnataka (37%) and Kerala (30%). It is as high in the poverty-ridden states like Orissa and Bihar. In the economically progressive states of Gujarat, Punjab, Haryana and Maharashtra, about 10-12% of the non-poor households possess BPL cards. In the economically weaker states such as Bihar, Orissa and Jharkhand, for example, only 32-48% of poor families possess a BPL card. It may be inferred from the data that, if the gap between the poor and non-poor with respect to percentage of households possessing a BPL card is narrow, the misuse is high and vice versa. Some of the states falling in these categories are Assam, Madhya Pradesh, Uttar Pradesh and Rajasthan. What is more shocking is that a majority of the population in many of the states that are experiencing abject deprivation including Bihar, Madhya Pradesh, and Uttar Pradesh, do not have a BPL card.

In Table 6, we have provided the numerical estimates of BPL cards and the percentage of non-poor holding a BPL card in some selected Indian states. The selected states are Andhra Pradesh, Karnataka, Rajasthan, West Bengal and Uttar Pradesh. They have been selected keeping in mind that about 30-50% households in these states fall in the ﬁrst or second wealth quintile. These states also reﬂect the geographical heterogeneity of the country. As noticed in previous discussions, the extent of misuse is higher in some of these states.

The ﬁndings in Table 6 suggests that the number of non- poor households having a BPL card is higher than the number of poor households having such cards in Andhra Pradesh and Karnataka. The distribution of BPL cards among poor and nonpoor households indicates that about 63% of the BPL cards in Andhra Pradesh and about 52% of them in Karnataka are with non-poor households. Contrary to this, the percentage of such households in Rajasthan is 36%, 31% in West Bengal and 26% in Uttar Pradesh.

february 14, 2009 vol xliv no 7

Economic & Political Weekly

SPECIAL ARTICLE

6 Discussion and Conclusions

The analysis suggests that a large number of BPL cards in India are distributed to the non-poor households. This also conﬁrms the fact that households in possession of expensive assets such as pucca houses with three or more sleeping rooms, a motorised vehicle, both televisions and refrigerators and landholdings of 5 acres or more, also possess a BPL card. It is estimated that, about 44% of the BPL card (27 million) were distributed to the non-poor households in the country. On the other hand, about three-ﬁfths of the poorest (those in abject deprivation) do not possess BPL cards.

The state-level analysis suggests that the extent of misuse of BPL card is highest in Andhra Pradesh followed by Karnataka and Kerala. Even in poorer states like Bihar and Orissa, a sizeable proportion of non-poor households possess a BPL card. On the other hand, the possession of BPL cards among the non-poor is lower in Tamil Nadu. However, in many of the states, the poorest of the poor (those living in the abject deprivation) do not hold a BPL card. This is probably an indication that the welfare schemes meant for the poorest are not reaching the poor and downtrodden.

What are the implications of such a trend? First, whatever methodology may be adopted in identifying the poor, there is a need for a more vigilant and transparent mechanism to exclude the non-poor. In case of false reporting, stringent action should be taken against all involved so as to make the poverty reduction successful. Second, the omission of the poorest of the poor (the abject deprived group) in the distribution list of BPL cards is a serious concern. It is an indication that the poorest of the poor do not have a voice and in many states are omitted from government welfare schemes. Third, there is a need to explore multiple options to assess the eligibility of beneﬁciaries for various welfare schemes rather than exclusively depending on any one scheme, such as the BPL card, for example. Finally, we suggest vigilant and concurrent evaluation by autonomous institutions to reduce the misuse of BPL schemes.

References

Filmer, D and L H Pritchett (2001): “Estimating Wealth Effects without Expenditure Data or Tears: An Application to Educational Enrolments in States of India”, Demography, 38 (1), 115-32.

IIPS and ORG Macro International (2007): National Family Health Survey (NFHS-3), 2005-06: India, Volume 1 (Mumbai: International Institute for Population Sciences).

Montgomery, M R, M Gragnolati, K A Burke and E Paredes (2000): “Measuring Living Standards

with Proxy Variables”, Demography, 27: 155-74.

Ofﬁce of the Registrar General and Census Commissioner, India (2006): “Population Projection for India and States 2001-26”, Report of the Technical Group on Population Projections Constituted by the National Commission on Population, India.

Planning Commission, Government of India (2007): “Poverty Estimates for 2004-05”, accessed on line: http://planningcommission.nic.in/news/prmar07.pdf.

Srinivasan, K and S K Mohanty (2002): “Deprivation of Basic Amenities by Caste and Religion in India: An Empirical Study Using NFHS-Data”, Economic & Political Weekly, XXXIX (7): 728-35, 14 February.

Sundaram, K (2003): “On Identiﬁcation of Households Below Poverty Line: Some Comments on the Proposed Methodology”, Economic & Political Weekly, XXXVIII (9): 896-902, 1 March.

Vyas, S and L Kumaranayake (2006): “Constructing Socio-economic Status Indices: How to Use Principal Component Analysis”, Health Policy and Planning, 21 (6), 459-68.

february 14, 2009 vol xliv no 7