Agflation and the Public Distribution System

surya@igidr.ac.in

COMMENTARYmay 3, 2008 EPW Economic & Political Weekly14had an Antyodaya card, 10.5 per cent had BPL cards, 55.6 per cent had ordinaryPDS cards and 33.1 per cent had no card (Table 1). Thus, the PDS is predominantly rural in terms of absolute as well as percentage number of households pos-sessing ration cards. It benefited a major-ity of the households in the sense that more than two-thirds in the urban and four-fifths in the rural areas possessed ration cards. The profile observed for the country as a whole generally held across major stateswith variations around the national average. The states of Andhra Pradesh, Karnataka and Orissa stood out for a high proportion of BPL card households. Andhra Pradesh had the highest propor-tion of households with BPLcards (rural: 54 per cent; urban: 27 per cent) as against an estimated rural household poverty of 10 per cent and urban at 30 per cent. Karnataka had BPL card households of 42 per cent (rural) and 14.4 per cent (urban) of total households in each area as against household poverty estimates of 17 per cent (rural) and 26 per cent (urban). Orissa provided the BPL cards to 42 per cent of rural households and 11.8 per cent of urban households as against the estimated poverty ratios of 45 per cent (rural) and 39 per cent (urban) (Table 1). As regards the socially vulnerable groups in rural India, 5 per cent of the scheduled tribes (ST), 4.5 per cent of the scheduled castes (SC), 2 per cent for other (including other backward classes (OBCs)) households possessed the Antyodaya card. As regards the BPLcard, 40 per cent of the ST households, 35 per cent of the SCs, 25 per cent of the OBCs and 17 per cent of the other households held these cards. In sum, three-fourths of theSTs and more than four-fifths of the SCs,OBCs and other social groups possessed ration cards in one form or the other (Table 2).The urban profile was slightly different: More than 1 per cent of the SC andST households possessed Antyodaya card while the percentage was less than one for other groups. The percentage of house-holds holding the BPL card was the highest for the SCs (17 per cent), followed byST and OBC households (about 14 per cent each) and others (5 per cent). In the urban sector, about one-half of the ST households and more than two-thirds of the SC, OBC and other social groups had access to the PDS outlets (Table 2). 4 Reliance on PDSThe most important issue that should cause concern is not that of universal cov-erage but the quality of foodgrains sold, availability and transaction costs, which deter the cardholders from fully utilising thePDS. For instance, though 81 per cent of the rural households in the country as a whole had ration cards, only 24 per cent reported rice consumption from the PDS and only 11 per cent wheat. As regards urban India, 67 per cent of the households had ration cards but only 13 per cent and 11 per cent bought rice and wheat, respec-tively, from the PDS (Table 3, p 15). The extent of their dependence on the PDS for rice and wheat was very limited in the sense that on an average a rural/urban household obtained about/less than 10 per cent of its consumption of rice/wheat from thePDS (Table 4, p 15). Among the major states where the PDS operation is pronounced and significant, the following features may be noted: – Though 90 per cent of the rural and 77 per cent of the urban households in Tamil Nadu had ration cards, 79 per cent of the rural and 48 per cent of the urban house-holds reported having obtained rice from the PDS for their consumption; as regards wheat about 10 per cent of the rural and urban households used the PDS also (Table3). However, the extent of their reliance on the PDS was much less: on an average, a rural household obtained only 40 per cent of its rice/wheat consumption from the PDS while the urban household got 28 per cent of its rice and 23 per cent of its wheat consumption from the PDS(Table 4). – Other states where the PDS played an Table 1: Distribution of Households by RationCard Type and Incidence of Household Poverty (major states, 2004-05)State Rural Sector Urban Sector Percentage of Households with Household Percentag e of Households with Household Antyodaya BPL Card Other Card No Card Poverty (%) Antyodaya BPL Card Other Card No Card Poverty (%) Card Card AndhraPradesh 2.8 541628 9.61 1.5 26.618 5422.95Assam 0.6 1263 2519.75 0.2 3.2 40562.79Bihar 2.3 15602338.06 0.8 4.7 4252 26.46Chhattisgarh 4.4 35 32 29 35.61 2.1 15.2 40 43 35.55Gujarat 0.8 36501314.72 0.1 8.4 67 2411.29Haryana 2.6 16681311.17 1.5 9.9 6128 10.67Jharkhand 3 23512340.84 0.8 7.5 335815.43Karnataka 9.6 4226 2317.45 2 14.4335126.31Kerala 1.8 28 57 1310.83 0.9 19.8601914.57Madhya Pradesh 3.3 31 38 28 33.12 1.9 12.7 43 43 37.04Maharashtra 4.4 31461924.95 0.3 8 67 2525.49Orissa 2 42233344.97 1.3 11.8295839.32Punjab 0.1 1276127.65 0 3.9 66304.64Rajasthan 2.8 1678 4 15.34 0.6 2.4 82 1525.68Tamil Nadu 1.5 19 69 11 20.09 0.6 12.8 64 22 19.01Uttar Pradesh 2.8 14 65 19 28.58 0.7 7.2 57 36 22.49West Bengal 3.2 27 61 8 24.33 0.8 8.8 71 20 9.89All-India 2.9 26.5 51.8 18.7 24.5 0.8 10.5 55.6 33.1 20.13The term “major state” refers to a state, which had a population of 20 million or more as per the 2001 Census.Source: GoI (2007b, p 13) and estimates of household poverty were made using the NSS unit record data and state-and sector-specific poverty lines published by the Planning Commission [GoI 2007a].Table 2: Distribution of Households by RationCard Type and Incidence of Household Poverty (all India, major social groups, 2004-05)Social Group Rural Sector Urban Sector Percentage of Households with Household Percentag e of Households with Household Antyodaya BPL Card Other Card No Card Poverty (%) Antyodaya BPL Card Other Card No Card Poverty (%) Card Card ST 5.039.630.824.643.351.313.637.647.529.81SC 4.4 34.9 43.7 17.031.73 1.6 17.349.831.333.12OBC 2.324.554.518.722.910.914.451.533.224.22Others 1.9 17.363.017.7 13.01 0.4 5.2 61.832.612.06Total 2.9 26.551.818.7 24.50 0.8 10.555.633.120.13Source: GoI (2007b; p A-198 and p A-216) and estimates of household poverty were made using the NSS unit record data and state-and sector-specific poverty lines published by the Planning Commission [GoI 2007a].

COMMENTARYEconomic & Political Weekly EPW may 3, 200815important role and the proportions of households, which reported consumption from thePDS, are as follows: For rice, Andhra Pradesh (rural: 62 per cent; urban: 31 per cent), Karnataka (rural: 59 per cent; urban: 21 per cent) and Kerala (rural: 35 per cent; urban: 23 per cent). The esti-mated percentages of households which used thePDS wheat also for consumption and which were pronounced pertain to the states of Karnataka (rural: 46 per cent; urban: 15 per cent), Gujarat (rural: 29 per cent), Maharashtra (rural: 26 per cent), and Madhya Pradesh (rural: 20 per cent; urban: 10 per cent) (Table 3). In all these states, the dependence on the PDS, that is, corresponding percentages of purchases form thePDS were generally much less (Table 4).5 Targeting EffectivenessThe targeted versions of the PDS, in spite of the good intentions, involved errors of both omission (of the eligible poor households) (called Type I error in targeting) and excess coverage of the ineligible non-poor house-holds (called Type II error) even today.2 The NSS reports size distribution of house-holds across 12 percentile classes of monthly per capita consumer expenditure (MPCE) for both rural and urban sectors. The estimates show that the poorest four MPCE classes, which accommodated the poorest 30 per cent of the population, did not exhaust the set of AntyodayaandBPL cardholders. More than 50 per cent of the households in these MPCE classes did not have the AntyodayaandBPLration cards (Table 5, p 16). Households possessing the Antyodaya and the BPL ration cards, though generally declining in percentage, were found across higher percentile classes of expenditure in both rural and urban sectors (Table 5). In rural India, even the richest percentile class (consisting of the richest 5 per cent of the rural population) included households possessing the Antyodayaand the BPL ration cards: Nearly one (0.8) percent of the richest households had the Antyodaya card and about 11 per cent had the BPLcard. In urban India, at least one-hundredth of the richest 5 per cent had the BPL card. In other words, there is considerable scope for reducing both Type I andType II errors in the targeted versions of PDS. What is important to note is that the ex-tent of dependence on the PDS by the Antyodaya and BPL card households is limited and about the same across percen-tile classes of expenditure (as observed in the profile for the country as a whole (Table 5)). While the Antyodaya andBPL card households in the poorest 5 per cent of the rural households obtained 30 per cent of their consumption of rice from the PDS, the percentage varied around 27 per cent for Antyodayaand BPL beneficiary house-holds falling in the up-per expenditure groups. As regards urban India, the dependence on the PDS for rice of the Anty-odaya and BPL card-holders varied from 42 per cent among the poorest to 25 per cent among the richest. As regards wheat, the ex-tent of dependence on the PDS varied around 28 per cent in both ru-ral and urban sectors (Table 5). This would imply that even the An-tyodaya and BPL card households depended to a proportionately larger extent on the open market and hence, the levels as well as sta-bility of open market prices matter to ensure food security of the poor. It was only in Karnataka that the PDS met more than three-fourths of the rice/wheat consumption of the rural Antyodayaand BPL card households and about two-thirds of those of their urban counterparts (Table 4).6 Implications and IssuesWhat does the discus-sion so far imply? What are the issues that call for policy attention? (1) If by universalisation of the PDS, the emphasis is on access to the PDS as a mar-ket source for foodgrains, then the system is already virtually universal. All the more so in the rural sector because a significant proportion of the rural households are cultivator households and hence, do not rely on the market for foodgrains. (2) Still only a minor subset of the ration card holding households depended upon thePDS. In general, even the Antyodaya Table 3: Percentage of Households Reporting Consumption from PDS and from Any Source(major states, 30-day period, 2004-05) State Percentage of Households Reporting Consumption during a 30-Day Period RiceWheat From PDS From Any Source From PDS From Any Source Rural Urban Rural Urban RuralUrban Rural UrbanAndhra Pradesh 62.2 31.1 96 94 0.6 0.7 27 56Assam 9.0 2.3 100 93 0.2 0.3 56 78Bihar 1.0 0.7 100 98 1.7 1.5 96 97Chhattisgarh 21.7 13.2 99 97 5.3 5.4 31 76Gujarat 31.5 7.2 94 94 28.7 6.8 83 91Haryana 0.1 0.0 82 92 4.0 5.2 99 98Jharkhand 4.4 2.8 99 92 4.3 2.0 71 89Karnataka 58.5 21.0 98 91 45.6 14.6 74 81Kerala 34.6 23.3 98 93 12.2 12.1 60 72Madhya Pradesh 17.9 8.7 80 93 20.3 10.3 94 98Maharashtra 27.5 6.0 93 93 25.8 6.9 86 91Orissa 21.5 5.8 98 93 0.2 1.0 36 68Punjab 0.1 0.1 74 85 0.3 0.6 100 96Rajasthan 0.0 0.2 41 70 12.7 1.9 87 95Tamil Nadu 78.9 47.7 97 93 8.9 10.7 29 61Uttar Pradesh 5.8 2.1 96 96 5.6 2.6 99 98West Bengal 12.8 5.4 99 94 9.0 3.5 61 82All-India 24.4 13.1 92 92 11.0 5.8 72 83Source: GoI (2007b; p 18).Table 4: Reliance on the PDS (by category of households, major states, % of consumption from the PDS, 2004-05) Rural Sector Urban Sector RiceWheatRiceWheat A and BPL All A and BPL All A and BPL All A and BPL All Hhs Hhs Hhs Hhs HhsHhs Hhs HhsAndhra Pradesh 33.74 23.26 5.16 3.16 36.49 14.55 2.78 1.28Assam 26.04 3.90 0.48 0.15 30.27 1.74 0.00 0.08Bihar 1.75 0.54 5.97 1.08 2.36 0.61 9.33 1.35Chhattisgarh 23.33 11.54 30.85 16.19 36.50 10.75 41.67 5.12Gujarat 31.36 13.35 41.06 14.81 28.67 3.24 22.98 3.07Haryana 0.00 0.03 12.99 2.43 0.00 0.02 29.76 4.69Jharkhand 5.55 1.54 17.56 4.07 14.35 1.99 11.18 1.53Karnataka 73.61 46.51 83.55 54.38 60.21 18.17 63.73 13.68Kerala 42.37 20.18 40.95 26.67 40.39 15.99 47.98 21.74Madhya Pradesh 30.71 15.71 32.12 11.97 29.49 5.05 38.29 7.02Maharashtra 48.08 21.25 61.02 26.21 37.97 4.63 43.09 6.83Orissa 13.85 7.01 1.96 0.56 15.06 2.94 3.52 0.94Punjab 0.53 0.10 1.57 0.22 0.00 0.38 0.69 0.60Rajasthan 0.00 0.00 45.26 10.49 1.33 0.16 25.43 1.73Tamil Nadu 51.77 40.37 61.64 40.24 48.90 28.05 51.17 23.38Uttar Pradesh 18.89 3.50 12.56 2.29 15.71 1.84 10.90 1.23West Bengal 5.87 2.29 45.67 15.93 7.64 1.87 29.48 3.50All-India 27.40 13.16 28.16 7.32 34.95 11.24 28.08 3.82A and BPL hhs = Households with Antyodaya and BPL cards.Source: Computed from different tables in GoI (2007b).

MANOHAR

COMMENTARYEconomic & Political Weekly EPW may 3, 200817call for additional procurement of foodgrains and this would end up virtu-ally mopping up the entire marketed sur-plus. This will have adverse implications for foodgrain prices in the open market [Suryanarayana 1995], which is the major source food for even the Antyodaya cardholders. In sum, what is needed now is not a demand for universalisation of thePDS but a revision of the food security norm, a more BPL-friendly system and its efficient functioning. Notes1The term “agflation” refers to food inflation caused by increases in demand for food resulting from (i) human consumption; (ii) use as a biofuel.2 See Cornia and Stewart (1993) for details regard-ing errors in targeting.ReferencesBhattacharya, Nikhilesh, Dipankor Coondoo, Pradip Maiti and Robin Mukherjee (1991): Poverty, Inequality and Prices in Rural India, Sage Publica-tions, New Delhi.Cornia, Giovanni Andrea and Frances Stewart (1993): ‘Two Errors of Targeting’,Journal of International Development, Vol 5, No 5, pp 459-96.Government of India (2002):Tenth Five-Year Plan (2002-07) Volume II: Sectoral Policies and Programmes, Planning Commission, New Delhi. – (2007a):Poverty Estimates for 2004-05, Press Information Bureau, New Delhi. – (2007b):Public Distribution System and Other Sources of Household Consumption 2004-05, Vol-ume I, NSS 61st Round (July 2004-June 2005), National Sample Survey Organisation, Ministry of Statistics and Programme Implementation, New Delhi.Suryanarayana, M H (1995): ‘PDS: Beyond Implicit Subsidy and Urban Bias’,Food Policy, Vol 20, No 4, pp 259-78.Suryanarayana, M H and Dimitri Silva (2005): ‘Is Targeting the Poor a Penalty on the Food Inse-cure? Poverty and Food Insecurity in India’, Journal of Human Development, Vol 8, No 1, pp 89-107.Email: bernard@epw.in