SPECIAL ARTICLE Trends in Diversion of Grain from the Public Distribution System
Reetika Khera
This paper estimates the proportion of grain diverted from the public distribution system to the open market in the past decade by matching official offtake figures with household purchase reported by the National Sample Survey. At the all-India level, diversion of PDS grain remains a serious issue; however there are interesting contrasts at the state level. Based on trends in monthly per capita purchase of PDS grain and estimated diversion, states are categorised into three groups – “functioning”, “reviving” and “languishing” states. The paper also discusses the possible reasons for the improvement in the PDS in the reviving states and questions the assessment of the PDS as uniformly and irreversibly dysfunctional.
I thank Angus Deaton, Jean Drèze, Himanshu for comments and suggestions on an earlier draft and Abhiroop Mukhopadhyay, Chris Oldiges and Sanjeev Sharma for help with data-related queries. I also thank Rajeev Jaiswal, Chandrasen Kumar, V S Moni, K Raju, Swaran Singh and Alok Shukla for sharing state-level data and insights on the PDS and FCI.
Reetika Khera (reetika.khera@gmail.com) is at the Department of Humanities and Social Sciences, Indian Institute of Technology, Delhi and a visiting scholar at the Centre for Development Economics, Delhi School of Economics.
1 Introduction
I
The current discussion on diversion relies on data from the 61st round of the National Sample Survey (NSS) (pertaining to 2004-05) or older data, or worse, no data. For instance, Bardhan (2011) describes the PDS as “dysfunctional” on the basis of “a rough estimate that less than a quarter of the subsidised foodgrains reaches the poor” (no source provided). Banerjee (2011) cites an outdated diversion figure, from a Planning Commission report published in 2005. Basu (2011) relies on estimates of leakage calculated using NSS data from 2001-02, ignoring the 2004-05 estimates. The latest calculations are those by Jha and Ramaswami 2010 (who stop in 2004-05) and Himanshu and Sen (2011). The use of outdated estimates has led to the portrayal of the PDS as uniformly and irreparably defunct which, in turn, a llows one to begin with an assumption of a dysfunctional PDS and to make a case for replacing it with “cash transfers” (Basu 2011, Bardhan 2011, S omanathan 2011, Banerjee 2011).
In this paper, I extend the series of diversion estimates to 2007-08, the latest year for which data are available. I find that there are divergent trends across states insofar as per capita purchase of PDS grain (wheat and rice) and proportion of grain d iverted are concerned. These estimates suggest that the states can be grouped into three broad categories: in a handful of states, PDS purchases are relatively high and diversion of grain is not a major concern (“functioning” states). Other states (“reviving” states) have seen a marked increase in PDS purchases in recent years, starting from a low base, associated with a decline in d iversion. In yet others, the situation remains grim where low purchases, high diversion, and little improvement over time (“languishing” states). This suggests that it would be incorrect to label the PDS as uniformly and irreparably “dysfunctional”.
A reassessment of the PDS is required for another reason. Statelevel innovations in food policy (e g, the proportion of households covered by the TPDS, or the difference between market and PDS price of grain) have been initiated in several states from 2007 onwards. These reforms have reinforced the trend towards a r evival of the PDS. Section 5 documents some of these policy changes and how they may have contributed to the revival of the open market. In this paper, the conventional interpretation of PDS in some states. d iversion is adhered to, as there are no reliable estimates of losses
due to other reasons. This suggests that these figures should be 2 Data and Methodology treated as the “upper-bound” on diversion. Other reasons these “Diversion” estimates presented in this paper include leakages estimates may provide the upper limit on diversion are discussed due to corruption, transport losses, losses due to spoilage and so in Section 4.1. on. The general practice has been to attribute all such losses to The analysis here roughly covers the past decade, including the illegal sale of PDS grain, meant for ration cardholders, on the two “thick” rounds of the NSS: the 55th (1999-2000) and 61st
(2004-05) rounds and three “thin” rounds: the 57th (2001-02),
Table 1A: Per Capita Purchase of PDS Rice (kg/month)
63rd (2006-07) and 64th (2007-08). The 64th round pertaining
Rural Urban
to 2007-08 is the latest round for which data are available from
1999- 2001- 2004- 2006- 2007- 1999- 2001- 2004- 2006- 2007
2000 02 05 07 08 2000 02 05 07 08 the NSS. Though the 63rd and 64th rounds are “thin” rounds, it is Andhra Pradesh 2.3 1.8 2.57 3.17 3.48 1.21 1.1 1.39 1.9 1.85 possible to generate reliable estimates at the state level.
Assam 0.71 0.6 0.48 1.11 1.01 0.57 0.3 0.18 0.47 0.03 The proportion of PDS grain that is diverted from the system Bihar 0 0.0 0.04 0.1 0.07 0.07 0.1 0.04 0.06 0.06
can be estimated by matching data on “offtake” by state govern-
Chhattisgarh NA 0.4 1.45 2.4 3.17 NA 0.3 0.94 1.08 1.2
ments from the Food Corporation of India (FCI) with data on
Gujarat 0.38 0.2 0.24 0.3 0.26 0.19 0.1 0.06 0.08 0.05
household purchase from PDS shops, collected by the NSS.
Haryana 0 0.0 0 0.12 0.09 0 0.0 0 0.16 0.04
“Offtake” refers to the amount of grain that the states take from
Himachal Pradesh 1.52 1.6 2.05 2.33 2.67 0.67 0.4 0.68 1.12 1.56
the FCI for distribution through the PDS. These data are available
J&K 2.1 0.8 2.78 4.52 3.05 4.11 2.5 4.65 5.2 4.68
from the department of food and civil supplies which publishes
Jharkhand NA 0.1 0.15 0.22 0.31 NA 0.0 0.12 0.07 0.08 Karnataka 1.2 2.39 1.85 0.3 0.92 monthly data (state-wise, for rice and wheat separately) in the
1.4 2.16 0.85 1.01 0.56 Kerala 4.14 1.8 1.71 2.04 2.24 3.48 0.8 1.22 1.54 1.54 Monthly Foodgrains Bulletin. Madhya Pradesh 0.22 0.1 0.36 0.29 0.46 0.09 0.1 0.1 0.17 0.19 On the other hand, the NSS collects data on monthly purchase
Maharashtra 0.47 0.4 0.61 0.77 0.85 0.24 0.1 0.13 0.17 0.13 of rice and wheat from the PDS as part of its consumer expendi-
Orissa 1.53 1.2 0.9 1.29 1.93 1.34 0.2 0.31 0.71 0.97 ture surveys. This can be aggregated up to the state level, by Punjab 0 0.0 00.05 0.01 0 0.0 0 0 0
multi plying per capita per month purchase by total population
Rajasthan 0 0.0 0 0.03 0.04 0 0.0 0 0.11 0.07
for the relevant year. The Census of India publishes projected
Tamil Nadu 3.16 3.4 4.13 4.7 4.84 2.14 1.8 2.46 3.36 3.19
population which can be used for scaling up per capita purchases
Uttar Pradesh 0.11 0.1 0.14 0.43 0.64 0.08 0.0 0.05 0.16 0.14
to the state level.
Uttarakhand NA 0.2 1.04 1.1 1.14 NA 0.1 0.09 0.19 0.43
I aggregate the data on monthly offtake from FCI to match the
West Bengal 0.23 0.2 0.27 0.35 0.4 0.23 0.1 0.14 0.16 0.14
reference period of the NSS surveys (i e, from July to June).2 The
India 0.7 0.84 1.05 1.18 0.4 0.54 0.75 0.69
difference between offtake and total purchase gives an estimate Table 1B: Per Capita Purchase of PDS Wheat (kg/month)of the amount of grain that is diverted.
Rural Urban
1999- 2001- 2004- 2006- 2007- 1999- 2001- 2004- 2006- 2007
3 Purchase and Diversion of PDS Grain
2000 02 05 07 08 2000 02 05 07 08
Andhra Pradesh 0 0.0 0 0.01 0.01 0.27 0.1 0.01 0.04 0.05
3.1 Per Capita Purchase of Rice and Wheat from the PDS
Assam 0 0.0 0 0.01 0.02 0 0.0 0 0.02 0.01 Bihar 0.15 0.0 0.06 0.07 0.09 0.16 0.0 0.08 0.06 0.09 Tables 1A and 1B report the per capita purchase of wheat and rice
Chhattisgarh 0.1 0.08 0.07 0.03 0.0 0.14 0.03 0.03 for rural and urban areas separately for each of the relevant
Gujarat 0.55 0.5 0.54 0.48 0.43 0.3 0.3 0.16 0.18 0.13 years. In rural areas, though average per capita purchase of PDS Haryana 0 0.0 0.23 0.64 0.55 0 0.1 0.36 0.57 0.27 grain is low, there is a lot of variation in the trends at the state-Himachal Pradesh 1.27 0.5 1.25 1.67 2.46 0.7 0.2 0.36 1.19 2.01
level. Average per capita purchase of grain is approximately 1kg
J&K 0.66 0.1 0.31 0.47 0.67 0.97 0.5 0.46 1 0.94
per month in the case of rice, and even lower (less than 500
Jharkhand 0.1 0.11 0.18 0.14 0.0 0.09 0.07 0.05
grams) in the case of wheat. The highest per capita purchase of
Karnataka 0.3 0.3 0.42 0.35 0.36 0.31 0.1 0.17 0.18 0.1
rice is recorded in Tamil Nadu (4.84kg/month in 2007-08).
Kerala 0.44 0.1 0.17 0.3 0.31 0.54 0.1 0.17 0.29 0.29
Himachal Pradesh (HP) records the highest monthly per capita
Madhya Pradesh 0.16 0.5 0.91 0.62 1.1 0.15 0.4 0.55 0.41 0.62
purchase of wheat (2.46kg).
Maharashtra 0.67 0.5 0.86 0.88 0.88 0.32 0.2 0.28 0.3 0.19 Orissa 0 0 0 0.0 0.09 Before proceeding, a few clarifications may help. One,
0.00.010.34 0.02 0.05 Punjab 0 0.0 0.02 0.05 0.54 0.11 0.0 0.05 0.06 0.32 though the targeted PDS (or TPDS) was introduced in 1997, in
Rajasthan 0.2 0.3 0.89 0.6 0.63 0.18 0.1 0.16 0.48 0.54 effect, the subsidy for above poverty line (APL) households Tamil Nadu 0.14 0.0 0.08 0.12 0.14 0.34 0.1 0.12 0.2 0.17 ended in 2000-01.3 Two, between 1999-2000 and 2001-02, PDS
Uttar Pradesh 0.18 0.2 0.19 0.3 0.41 0.19 0.1 0.1 0.2 0.15 entitlements for BPL households increased from 20kg/month/ Uttarakhand 0.4 0.83 0.75 1.01 0.0 0.07 0.05 0.44
household to 35kg/month/household. Between 2001-02 and
West Bengal 0.15 0.1 0.15 0.17 0.22 0.39 0.1 0.08 0.1 0.14
2007, PDS entitlements remained the same.4 Three, the propor-
India 0.2 0.35 0.31 0.39 0.1 0.17 0.22 0.21
tion of BPL households was fixed by the central government
Source: Author’s calculations using NSS data. Chhattisgarh, Jharkhand and Uttarakhand were created as new states in 2000, therefore there are no estimates for those states for 1999-2000. when the TPDS was introduced in 1997. These proportions were
Economic & Political Weekly
EPW
SPECIAL ARTICLE
Figure 1: Purchase of PDS Grain Rural
— Reviving states – . – Functioning states ---- Languishing states
PDS grain, kg/capital/month
5
4
3
2
1
0 1999-2000 2001-02 Source: Calculated from NSS data for relevant years.
based on the Planning Commission’s poverty estimates using 1993-04 NSS data. Again, these proportions remained largely unchanged until 2007 when several state governments began expanding coverage. PDS grain allocated to states for BPL households has therefore not undergone any change. Fourth, even after the subsidy for APL households was phased out, the central government continued to allocate grain for APL households. APL allocations have been periodically revised on the b asis of the concerned state’s offtake history (Himanshu 2011). Finally, in this paper, per capita PDS purchase refers to the average over all households (i e, APL and BPL).
Based on per capita purchase of PDS grain, states can be grouped into three broad categories. The first category is that of “languishing” states (see Figure 1 and Figure 1A (p 110)). Roughly speaking, these are states where per capita purchases have remained below 1kg/month. This group consists of several states of north India (Haryana, Punjab, Rajasthan) and eastern India (Assam, Bihar, Jharkhand and West Bengal). The eighth state in this category is Gujarat, the only state where there has been a decline in per capita purchase of PDS grain over the past decade.
The second group consisting of five states (Figure 1 and Firure 1B (p 110)) is that of “reviving” states where per capita purchases were below the 1kg/month benchmark at the beginning of the period being studied, but have risen since then. This is a mixed group: Orissa in the east, Chhattisgarh and Madhya Pradesh in central India, Uttarakhand and Uttar Pradesh (UP) in the north. Note, however, that in the case of UP the signs of revival are only barely perceptible – average purchase has risen from 0.29kg/ month/capita, at the beginning of the period, to 1.02kg/month/ capita in 2007-08 (see Tables 1A and 1B). The most remarkable improvement is in Chhattisgarh where per capita purchase in 2007-08 was 3.2kg/month, putting it among the top five states on per capita purchase (after Himachal, Tamil Nadu, Jammu and Kashmir (J&K) and Andhra Pradesh).
The remaining seven states, where per capita purchase of PDS grain has been greater than 1kg/month throughout the p eriod u nder study and has also improved over time, are the
HP
TN
JK
AP
CH
KE
KA UTT
OR MA
,1',$
INDIA
MP
UP
AS GU
HA RJ WB
JH PU
BI
2004-05 2006-07 2007-08
“functioning” states (see Figure 1 and Figure 1C (p 110)). These include all four southern states (Andhra Pradesh, Karnataka, Kerala and Tamil Nadu) and Maharashtra. This category includes two states from the mountainous north (Jammu and Kashmir and Himachal Pradesh). Note that the “functioning” states are primarily rice-consuming states.
The trend of per capita PDS purchase emerging from Kerala requires further comment. Average purchase of rice crashed from 4.1kg/capita in 1999-2000 (the highest across states in that year) to 1.71kg/capita in 2004-05, and then recovered partially in subsequent years to 2.24kg/capita at the end of the reference p eriod. By 2007-08, however, Kerala loses its top position to Tamil Nadu (where the monthly average was 4.84kg/capita). The initial decline in per capita purchase in Kerala is partly b ecause PDS coverage in Kerala suffered with the introduction of the TPDS. In the pre-1997 period, Kerala was the only state in which access to the PDS was close to universal in principle and in operational terms.5 (In most other parts of the country, though PDS coverage was in principle universal, in rural areas access r emained poor.)6 When the TPDS was introduced in 1997, Kerala was forced to reduce its PDS coverage and this contributed to the decline in per capita purchase (Tables 1A and 1B).
In general, the PDS seems to deliver more in rice-consuming states. As far as wheat is concerned (see Table 1B), per capita p urchases remained low throughout the reference period ( bet ween 300 and 400 grams per capita). The only exception is Himachal which has been doing well both in terms of level and trend. Average purchase rose from 1.27kg/capita in 1999-2000 to 2.46kg/capita in 2007-08.
Average per capita purchase in urban areas is nearly half of the corresponding figures for rural areas (0.69kg/capita for rice and 0.21kg/capita for wheat). In urban areas, the same (as in rural areas) rice states do well: Tamil Nadu, Andhra Pradesh, Himachal Pradesh, Jammu and Kashmir and Chhattisgarh. In the case of wheat, only Himachal Pradesh crosses the 1kg/capita barrier (average purchase of wheat in urban Himachal was 2.01kg/capita).
3.2 Diversion of Rice and Wheat
Table 2 reports the estimated proportions of wheat and rice d iverted in the two “thick” NSS rounds (1999-2000 and 2004-05) and three “thin” rounds (2000-01, 2006-07 and 2007-08).7 This section discusses the main trends in diversion rates.
First, comparing performance of the states over this period, one finds that across states things got worse between 1999-2000 and 2004-05 (see Table 2). At the all-India level, the leakages from the PDS increased from 24% to 54%. The deterioration at the beginning of the period was visible even among the “functioning” states of the south (for instance, Kerala). However, since the last “thick” round (i e, 61st, pertaining to 2004-05) and more recent “thin” rounds of the NSS (63rd and 64th rounds), one n otices some signs of improvement, though only marginal, at the all-India level. The overall diversion of grain has come down from 54% in 2004-05 to 44% in 2007-08 (see Figure 2, p 111). In spite of this marginal improvement, diversion rates were higher in 2007-08 than in the pre-TPDS period (i e, in 1999-2000 when 24% of wheat and rice were diverted). Diversion rates in the “languishing” states remained high throughout the period. Among the “functioning” states, diversion rates were lower than the all-India average but even here two states, Karnataka and Maharashtra, have unacceptably high rates of diversion (see Table 2). Not surprisingly, the lowest diversion at the end of the period is in Tamil Nadu (4.4%) with Himachal Pradesh not far behind (13.2%).
Second, the most interesting pattern is in the reviving states where there has been an impressive decline in diversion rates in the past few years (see Figure 2). In Chhattisgarh, diversion has declined from half to zero between 2004-05 and 2007-08 (but see discussion in Section 4.3). In UP too the diversion rate has plunged from 58% to 27% in the span of three years. Though rates of diversion in 2007 remain high (see Figure 3, p 111), the experience of
Table 2: Diversion of PDS Grain1
the reviving states shows that leakages can be reduced. Possible explanations for these improvements are discussed in Section 5.
Third, note that there is a regional dimension to PDS purchase and diversion rates, with the bulk of the functioning states falling in the peninsular region, whereas, barring Rajasthan and Gujarat, the languishing states are concentrated in the eastern part of the country (Assam, Bihar, Jharkhand and West Bengal).
Fourth, the level of diversion of rice is lower than wheat in each of the years. In 1999-2000, about one-tenth of the rice was diverted, whereas nearly half (49%) of all wheat was diverted. However, the proportion of rice that is diverted increased b etween 1999-2000 and 2004-05 – from just 9.9% in 1999-2000 to 18.2% in 2001-02 and to 41.3% in 2004-05. Since 2004-05, there has been a marginal decline (approximately five percentage points) in the rice that is diverted. In 2007-08, a little over one-third (37%) of rice was diverted.
Finally, there is no clear trend in the indicators for two states (Haryana and Punjab). Per capita purchase of grain is very low in these states (less than 1kg/capita/month). Diversion rates fluctuate widely over the past decade. One plausible explanation for this is that while what the PDS delivers (in terms of per capita purchase) in these two states has not undergone much change, the allocation and offtake have gone down, which has led to an improvement in terms of diversion.
4 Methodological Issues
Four important methodological issues related to the estimation of diversion rates are briefly discussed here.8
4.1 Possible Under-Recording of PDS Purchase in NSS Data
First, it is possible that there is some under-recording of PDS grain purchase in the NSS data. There are two possible reasons why this may need investigation: One, Deaton and Drèze (2009)
| 1999-2000 | 2001-02 | 2004-05 | 2006-07 | 2007-08 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Rice | Wheat | Foodgrain | Rice | Wheat | Foodgrain | Rice | Wheat | Foodgrain | Rice | Wheat | Foodgrain | Rice | Wheat | Foodgrain | |
| Andhra Pradesh | 15.2 | 14.4 | 15.2 | 12.3 | -210.8 | 11.2 | 22.3 | 93.0 | 23.2 | 16.1 | 66.9 | 17.0 | 19.2 | 50.3 | 19.6 |
| Assam | 54.7 | 100 | 65.3 | 69.4 | 98.1 | 74.9 | 83.5 | 100.0 | 88.7 | 72.4 | 98.4 | 76.6 | 73.0 | 97.5 | 77.5 |
| Bihar | 94.6 | 75.2 | 80.2 | 77.3 | 91.6 | 88.3 | 84.8 | 92.8 | 91.0 | 83.6 | 84.4 | 84.0 | 92.4 | 85.1 | 89.5 |
| Chhattisgarh | 45.8 | 33.4 | 43.2 | 45.1 | 82.6 | 51.8 | 28.9 | 65.3 | 30.9 | -3.1 | 57.0 | -1.5 | |||
| Gujarat | -23.9 | 8.2 | -2.5 | 35.6 | 27.3 | 29.8 | 52.7 | 51.3 | 51.7 | 66.1 | 39.6 | 53.2 | 73.0 | 53.3 | 63.1 |
| Haryana | 0 | 100 | 100 | 94.0 | 94.0 | - | 82.7 | 82.7 | 39.5 | 29.4 | 31.4 | 61.8 | 48.8 | 51.1 | |
| Himachal Pradesh | - | - | - | 26.0 | 43.8 | 31.2 | 7.0 | 46.2 | 27.0 | 11.6 | 32.4 | 21.8 | 12.9 | 14.3 | 13.6 |
| Jammu and Kashmir | -1.4 | -80.3 | -12.3 | 54.1 | 79.0 | 60.7 | -8.9 | 79.4 | 23.0 | -36.5 | 66.4 | -1.0 | 7.6 | 59.1 | 24.3 |
| Jharkhand | 71.5 | 83.0 | 79.1 | 82.3 | 87.9 | 85.2 | 86.4 | 80.9 | 84.4 | 83.3 | 85.2 | 84.0 | |||
| Karnataka | 17.1 | 21.0 | 18.0 | 47.0 | 53.7 | 48.4 | 25.8 | 41.7 | 28.7 | 32.6 | 34.4 | 32.9 | 42.2 | 33.4 | 41.0 |
| Kerala | -44.7 | 5.9 | -36.9 | -28.6 | 66.9 | 0.0 | -1.9 | 78.9 | 25.6 | 0.8 | 55.3 | 14.8 | 3.5 | 55.6 | 16.2 |
| Madhya Pradesh | 59.3 | 18.2 | 46.9 | 50.8 | 46.4 | 47.4 | 12.9 | 56.7 | 50.1 | 52.8 | 64.0 | 61.1 | 20.8 | 39.9 | 35.5 |
| Maharashtra | 24.4 | 33.3 | 29.9 | 40.0 | 53.2 | 48.3 | 46.5 | 51.0 | 49.3 | 44.6 | 38.5 | 41.4 | 40.7 | 44.1 | 42.5 |
| Orissa | 26.8 | 87.5 | 36.7 | 21.4 | - | 21.0 | 74.1 | 99.0 | 76.3 | 53.4 | 91.5 | 57.0 | 46.2 | 97.1 | 50.2 |
| Punjab | 100 | -107.0 | -52.9 | 92.5 | 87.7 | 87.9 | 100.0 | 93.1 | 93.2 | 71.9 | 81.1 | 78.5 | 17.6 | 18.4 | 18.4 |
| Rajasthan | 100 | 53.0 | 53.4 | 76.1 | 75.8 | 75.8 | 100.0 | 93.9 | 93.9 | 69.8 | 83.5 | 81.9 | 75.7 | 82.0 | 81.2 |
| Tamil Nadu | -12.3 | -21.7 | -13 | -79.2 | - | 0.0 | 9.4 | -86.7 | 7.3 | 2.4 | -105.6 | -0.7 | 8.7 | -186.1 | 4.4 |
| Uttar Pradesh | 46.6 | 17.4 | 31.1 | 77.4 | 67.1 | 69.7 | 85.4 | 36.7 | 58.0 | 72.3 | 7.8 | 50.5 | 52.9 | -14.5 | 26.7 |
| Uttarakhand | -109.8 | -810.0 | 0.0 | 44.2 | 84.8 | 59.4 | 44.2 | 88.3 | 63.3 | 33.3 | 70.9 | 48.5 | |||
| West Bengal | 23.8 | 70.9 | 57.3 | 42.4 | 84.0 | 67.3 | 70.4 | 85.0 | 80.6 | 72.4 | 80.4 | 76.8 | 70.8 | 77.9 | 74.8 |
| India | 9.9 | 48.6 | 23.9 | 18.2 | 66.8 | 39.0 | 41.3 | 70.3 | 54.0 | 39.6 | 61.9 | 46.7 | 37.2 | 57.7 | 43.9 |
1 Proportion (%) of grain offtake from FCI, that does not reach households. 2. For a discussion of "negative" diversion estimates, see Section 4.3. Sources: “Offtake” data from Monthly Foodgrains Bulletin published by Government of India. I use total (BPL, APL and Antyodaya) offtake figures. For 2004-05, population projections for 1 March 2005 have been used. Similarly for 2006-07 (2007-08), projected population on 1 March 2007 (2008) has been used. Population projections have been taken from Office of the Registrar General and Census Commissioner (2006).
Economic & Political Weekly
EPW
SPECIAL ARTICLE find that for some years the NSS somewhat understates total Figure 1A: Per Capita Consumption of PDS Grain (Rural): ‘Languishing’ States c ereal consumption in India, compared with total availability 1.2
Assam
according to Ministry of Agriculture (see Figure 6, Deaton and
1
Drèze 2009). If under-recording of PDS purchase (distinct from
Rajasthan
Gujarat
total cereal consumption) is an issue, using the method outlined
0.8
Per capita purchase of PDS grain (kg/month)
above would lead us to automatically attribute this underesti
mation, or some part of it, to diversion.
Second, NSS data on consumption is based on recall for the
past 30 days which require investigators to spend time carefully
reconstructing purchase figures for each item. It is plausible that
there is under-recording of PDS purchases in NSS data.9 This
0.6
West Bengal
0.4
Haryana
Punjab
Jharkhand
0.2
Bihar
0
s uggests that crosschecking PDS purchase against other data
1999-2000 2001-02 2004-05 2006-07 2007-08
sources might be a useful exercise.10 In the case of the mid-day
Figure 1B: ‘Reviving’ States
meal scheme, the proportion of children benefiting from the mid
3.5
day meal are much lower according to NSS than from other data
sources (see for instance, De et al, forthcoming and Pratham
2011) and independent field evidence.11 Field evidence indicates
that the improvement in the Chhattisgarh PDS began after 2004,
when PDS reforms were initiated. As the discussion in Section 4.3
suggests, these improvements are only partially reflected in the
2007-08 data. Similarly, according to two informal studies of PDS
purchase in Orissa conducted in 2010, three-quarters of sample
households reported getting their full quota of rice from the PDS
(see Rai 2011 and Office of the Commissioners of the Supreme
Per capita purchase of PDS grain (kg/month)
3
2.5 Chhattisgarh
Uttarakhand
2
Orissa
1.5
Madhya Pradesh
1
| 0.5 | ||||
|---|---|---|---|---|
| Uttar Pradesh | ||||
| 0 | ||||
| 1999-2000 | 2001-02 | 2004-05 | 2006-07 | 2007-08 |
Figure 1C: ‘Functioning’ States
Court 2010). It would be interesting to see whether data from the next thick round of the NSS (pertaining to 2009-10) validates the
results of these smaller field studies. 6
Per capita purchase of PDS grain (kg/month)
Jammu and Kashmir
5
Tamil Nadu
Kerala
4 Himachal Pradesh
3
Andhra Pradesh
Karnataka
2
Maharashtra
1
4.2 Choice of Multiplier: NSS vs Census Projections
Another issue to bear in mind is the choice of multiplier for ag
gregating per capita PDS purchase up to the state level. There are
two obvious choices for this: census population projections and
NSS population multipliers. The norm, insofar as estimation of
diversion is concerned, has been to use census population projec
tions. In addition to that, I use NSS multipliers to get estimates of
total rural and urban population. Using NSS “population” multi
pliers can provide a useful check for the reliability of population projections made by the Census of India.12 Here again, estimated diversion using both methods is not very different (see Table 4, Khera 2011c).
4.3 Contribution from the ‘State Pool’ and Accounting Practices
The centre allocates grain to states in accordance with the proportion of BPL families fixed by the Planning Commission. Where states have expanded the PDS (e g, Andhra Pradesh, Tamil Nadu, among others, see Table 3, p 112) the actual number of BPL households is higher than that sanctioned by the central government. When coverage of BPL households is increased, states have three options to fulfil additional grain requirements. First, they can spread the centre’s allocation (of 35 kg per BPL household) over a larger number of households by reducing per household entitlements. For instance, the Government of Tamil Nadu gives only 20kg per household but covers to all households. Second, states can purchase the additional grain r equired, e ither from the FCI or locally (i e, within the state,
110
0 1999-2000 2001-02 2004-05 2006-07 2007-08
through the state civil supplies corporation). This is being done, for instance, in Chhattisgarh. Third, states can take grain a llocated for APL households at higher prices and use it to meet the needs of the additional BPL households (as is done in Andhra Pradesh among other states). By a combination of these measures, the Government of Tamil Nadu has been able to run a u niversal PDS.
Chhattisgarh, Tamil Nadu and Andhra Pradesh (and possibly a few other states too) augment centrally allocated grain (through Open Market Sales or OMS purchases or “state pool” contributions). Where states augment foodgrain supply, using the allocation and offtake figures reported in the Monthly Foodgrains Bulletin would be incorrect because that only reports allocation and offtake from what is called the “central pool”. This would lead to underestimation of grain diversion.13 To get an accurate estimate, in such cases one would have to add to the offtake figure, the grain allocated to the PDS by the state from local procurement and other sources.
may 21, 2011 vol xlvi no 21 EPW Economic & Political Weekly
Proportion (%) PDS grain diverted
Figure 2: Decline in Estimated Diversion of PDS Grain, Selected States (2004-05 vs 2007-08) entitlements in each state is not compiled accurately
90
80 76.3
2004-05 entitlements relies on state government websites,
70
2007-08 online newspaper reports and so on. Data on the
60
58 states’ contribution to the PDS (mentioned above)
54
51.8 50
50.2
50.1
are also not readily available. There is no easy way
43.9
40
of finding out which states are augmenting FCI sup
35.5
30 26.7 27
25.6
23.2
19.6
20
13.6 10
7.3
which used to be available on the website of the de
4.4
0
-1.5 partment of food and civil supplies until 2005 is no
OR INDIA MP UP AP KE HP TN Chh -10
longer publicly available. These and other data gaps
Source: Author's calcuations, see Table 2. For Chhattisgarh, see discussion in Section 4.3.
need to be addressed urgently. As information on which states currently “top up” FCI allocations is not readily available, I take a closer look at these two issues using
5 ‘Reviving’ and ‘Functioning’ States
data from the Government of Chhattisgarh for 2007-08. This is the To understand the revival of the PDS in some states, one needs to first year in which Chhattisgarh began contributing rice from the look at state initiatives in the field of food policy. Some such inistate pool in order to run an expanded PDS. The department of food tiatives for selected states are listed in Table 3. When the TPDS of the Government of Chhattisgarh provides data on rice and wheat was introduced, the central government began allocating subsiby source: the state’s contribution as well as allocation and offtake dised foodgrain for households classified as below poverty line from the central pool of the FCI. (BPL). The proportion of BPL households in each state, in turn,
The first point to note is that there is a small discrepancy in was decided on the basis of the 1993-94 poverty estimates of the “offtake” data as reported in the Monthly Foodgrains Bulletin and Planning Commission. This compelled state governments to by the respective state governments. Though the reason for this “downsize” their PDS in accordance with what they got from the discrepancy is not clear, it could be attributed to leakages, transit central government once the TPDS was introduced. However, a losses, lags in accounting and so on. few years into the TPDS, several state governments realised that it
The second important point is that when we add Chhattisgarh had led to the exclusion of many poor households. For instance, state’s contribution to estimate diversion, the proportion of according to NSS data from 2004-05, only 53% of rural housefoodgrain diverted in Chhattisgarh in 2007-08 jumps from -1% to holds belonging to the poorest monthly per capita expenditure 37.7%.14 Note, however, that such revisions would be required (MPCE) quintile had a BPL ration card.17 These large exclusion only in those states where FCI allocations are “topped up” by state e rrors can be attributed to the small TPDS coverage (in other government. Given the manner in which data is compiled today, words, the Planning Commission’s poverty estimates were too it is not easy to tell if there are any other states for which such an low) and also to the poor design and implementation of the BPL exercise is required. survey. As a partial remedy to this (at least) since 2006, state gov-
In the case of Tamil Nadu, the department of food provided ernments began to spend state resources to increase the coverofftake and “liftment” from the central pool.15 “Liftment” refers to age, or provide additional subsidy to those households covered by the data that is sold at the PDS outlet. Note that Tamil Nadu is prob-the TPDS. As Table 3 shows, in several states, the PDS is now quasiably the only state in the country that has a system for tracking sales universal (Tamil Nadu, Andhra Pradesh, Chhattisgarh, Karnaat the PDS outlet. 16 Using FCI offtake and sales data provided by the taka, Himachal Pradesh). In many others, states governments state government, diversion figures for Tamil Nadu
Figure 3: Estimated Diversion of Grain (2007-08)
rise from 4.4% to 8% (see Khera 2011b, Table 6).
Proportion (%) of grain diverted Finally, note that for some states, using the -5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
above method, the estimated proportion of grain Bihar Jharkhand
diverted is negative. This means that more grain is
Rajasthan bought from ration shops than is supplied accord-Assam
West Bengal
ing to the Monthly Foodgrain Bulletin. Using lags Gujarat Haryana
(i e, offtake of grain allocated this month may
Orissa happen in subsequent months) does not resolve Uttarakhand
India
this issue. This also suggests that further stream-Maharashtra Karnataka
lining of this methodology is required and diver-
Madhya Pradesh sion figures must be treated as approximations. Uttar Pradesh
Jammu and Kashmir Andhra Pradesh
4.4 Data Gaps Punjab
Kerala Several data gaps hinder an exercise of this sort. Himachal Pradesh
Tamil Nadu
Estimates of transport and storage losses are not Chhattisgarh available. Even basic information on coverage and Source: Author's calcuations, see Table 2. For Chhattisgarh, see discussion in Section 4.3.
Economic & Political Weekly
EPW
| SPECIAL ARTICLE | |
| have expanded the coverage (e g, Rajasthan) and/or reduced | the PDS issue price. This has had the effect of increasing the |
| prices (e g, Jharkhand). The additional subsidy burden is met ei | i nterest of both the state government and BPL cardholders in tak |
| ther by reducing quantities (e g, from 35kg or 20kg in Madhya | ing as much as possible from the central pool and PDS shop re- |
| Pradesh) supplied to eligible households or by putting in addi | spectively. As market prices of wheat and rice have risen, grain |
| tional state resources or both. | sold at APL prices by the FCI has become cheaper than the open |
| Out of seven “functioning” states, four (Andhra Pradesh, | market. This is one of the reasons why state offtake of APL grain |
| Himachal Pradesh, Kerala and Tamil Nadu) are states where the | has been high in recent years.20 |
| PDS remained (or became) universal or quasi-universal because | The expansion of the PDS has been accompanied by supply-side |
| of extra commitment of resources by the state government. In | PDS reforms (e g, computerisation of records, deprivatisation of |
| Tamil Nadu the PDS is universal in rural and urban areas; in | PDS shops, upward revision of official commissions for ration |
| Himachal Pradesh APL families continue to get wheat and rice | dealers).21 Last and certainly not the least, renewed political |
| from the PDS, albeit at a higher price than BPL households. | commitment to the programme has been an important part of |
| A ndhra Pradesh runs a quasi-universal PDS: it covers many more | the reform in these states. |
| households than the centre provides for and like Tamil Nadu pro | |
| vides an additional subsidy on the issue price (only Rs 2/kg for | 6 Concluding Remarks |
| rice).18 Each of these functioning states (in 2007-08) have shown | In this paper, I estimate the “diversion” of PDS grain (wheat and |
| a consistent improvement over the reference period. At least two | rice) in India. I also discuss the methodological issues which |
| (Himachal and Tamil Nadu) of these states have a well estab | make the precise estimation of such leakages difficult. This dis |
| lished record of good performance in social welfare schemes.19 | cussion, including the fact that all leakages (storage and trans- |
| What could be behind the improvements noticeable in the re | port losses included) have to be clubbed with leakages due to cor |
| viving states? Khera (2011b) discusses two alternative sets of ex | rupt practices, indicates that estimates of diversion estimated in |
| planations: demand-side and supply-side factors that could affect | this manner must be treated as an upper bound on illegal diver- |
| PDS purchase by households. A common feature across reviving | sion of PDS grain. Imprecise as these estimates may be, comput |
| states is that the PDS has been expanded (e g, by increasing cover | ing these proportions over a period of time is still a useful exer |
| age, or the implicit subsidy) in recent years. For instance, in | cise because one can get a sense of which direction things are |
| Chhattisgarh the state provides an additional subsidy and has in | moving in. Looking at the overall proportion of grain diverted, |
| creased the coverage of the PDS to approximately 80% of rural | between 1999-2000 and 2007-08 (i e, the 55th and 64th rounds |
| households (from just 45%). Field-based evidence from the reviv | of the NSS), the situation is far from encouraging. At the begin |
| ing states suggests that when the issue price is lowered, say to | ning of the period, 24% of grain was diverted. The situation got |
| Re 1/kg or Rs 2/kg, people are unwilling to let go of their grain. | worse until 2004-05 when 54% of grain leaked but since then |
| These are referred to as “demand-side” factors that contribute to | there has been a reversal of that trend. At the end of the period |
| enhancing voice. Another measure that enhances voice is the in | (in 2007-08), 44% of PDS grain was diverted at the all-India level. |
| creased difference between market price and PDS price in recent | Studying trends disaggregated by state, the picture is very mixed. |
| years. The gap between market and PDS price has grown because | Based on level of per capita PDS purchase and trends, I classify |
| of a rise in market prices and because many states have lowered | states into three broad categories: “functioning” states, “ reviving” |
| Table 3: Some State-Level Initiatives in the PDS |
| State | Year | BPL Cards (lakhs) | BPL | APL | |
|---|---|---|---|---|---|
| Allocated | Actual | ||||
| Andhra Pradesh | Price reduction, 7 April 2008 | 40.63 | 175.5 | 20kg rice max (Rs 2/kg) | 30kg rice (Rs 9/kg) |
| Chhattisgarh | Price reduction 2007 and December 2008; | 18.75 | 36 | 35kg (Rs 2/kg), | - |
| expansion in 2007 | |||||
| Himachal Pradesh | - | 5.14 | 3.17 | 35kg (Central Issue Price) | 35kg (Central Issue Price) |
| Jammu and Kashmir | - | 7.36 | 4.8 | 35kg (Central Issue Price) | 35kg (Central Issue Price) |
| Jharkhand | Price reduction in October 2010 | 23.94 | 14.76 | 35kg (Rs 1/kg rice) | - |
| Karnataka | April 2005 | 31.29 | 76.77 | 25 kg rice/wheat (Rs 3/kg) | 13 kg rice, 3 kg wheat (Rs 6.7/9/kg) |
| Kerala | At least since 2006 | 15.54 | 14.82 | 25kg rice (Rs 3/kg) | 35kg (Rs 8.9 rice; Rs 6.7 wheat) |
| Madhya Pradesh | Price and quantity reduction in April 2008 | 41.25 | 52.65 | 20kg (Rs 3/kg for wheat and | - |
| Rs 4.50/kg for rice) | |||||
| Orissa | 1 July 2008, reduction of issue price | 32.98 | 37.63 | 25kg/month (Rs 2/kg rice) | - |
| Rajasthan | Price and quantity reduction, expanded coverage, | 24.85 | 35.57 | 25kg (Rs 2/kg) | 35kg (6.50) |
| 1 May 2010. | |||||
| Tamil Nadu | Price reduction 15 September, 2008; universal | 48.63 | 181.9 | Max 20 kg rice and wheat: 5-10kg | |
| since 2006 | (Rs 1/kg rice and Rs 7.50/kg wheat) |
Central Issue prices are APL Rice Rs 8.30; Wheat Rs 6.10. BPL Rice Rs 5.65; Wheat 4.15. Sources: Number of BPL cards from the Monthly Foodgrain Bulletin (as on 30 June 2009). PDS Entitlements data from state government websites. Andhra Pradesh: http://www.aponline.gov.in/ quick%20links/departments/consumer%20affairs%20food%20&%20civil%20supplies/a%20p%20state%20civil%20supplies%20corporation/about/index_old.html#file2; Chhattisgarh: http:// cg.nic.in/khadya/documents/CM%20Khadya.pdf; Himachal: http://admis.hp.nic.in/ehimapurti/function.htm; Karnataka: http://des.kar.nic.in/ecsurveynew/chapter5_eng.pdf (2008-9) and http:// www.kfcsc.com/pds.htm; Kerala: http://www.civilsupplieskerala.gov.in/PDS_details.aspx; Orissa: http://www.orissa.gov.in/foodsupplies/index.htm; Tamil Nadu: See also http://www.tn.gov. in/policynotes/pdf/food.pdf; Data on starting date of state-level initiative are from The Hindu (2005), The Hindu (2007), The Hindu (2010), Mohan (2008), Thai Indian News (2008), Sify News (2008), Economic Times (2006).
may 21, 2011 vol xlvi no 21
states and “languishing” states. The seven states in the first category The evidence on diversion presented in this paper shows are not surprising, as they have a good record on implementation of that it is no longer possible to label the PDS as a “dysfunctional” various social welfare schemes. Apart from the southern states, this programme – it amounts to ignoring the performance of the category includes HP and JK in the north and Maharashtra. The eight PDS in seven major states and the signs of revival in at least “languishing” states are not surprising either. The group includes As-another five states. Informal field evidence from after 2007-08 sam, Bihar, Jharkhand, West Bengal – states that have a poor record (the latest year for which data are available), suggests this on other socio-economic indicators too. trend towards a revival has been further consolidated. There is
The most interesting category is that of the five “reviving” clearly a need for further research that would help understand states, which includes Chhattisgarh, Madhya Pradesh, Orissa and the consistently poor performance in some states and reasons Uttar Pradesh, which at the beginning of this period were like the behind improvement in others. The future viability of the PDS “languishing” states but by 2007-08, seem to be catching up with depends a great deal on the scope for extending these improvethe “functioning” states. ments across the country.
Notes resorts to open market purchases of rice for the Social Democracy in India”, Economic & Political 1 Government of India (2005a) defines diversion
PDS, in 2007-08 it did not. Weekly, Vol 46, No 1, pp 39-43. differently from the definition used here and in 16 Data on sales from PDS outlets is available from Basu, Kaushik (2011): “India’s Foodgrain Policy: the other studies. the Tamil Nadu Civil Supplies Corporation. Tamil An Economic Theory Perspective”, Economic & 2 I also try introducing lags between offtake and
Nadu has been a pioneering state as far as compu-Political Weekly, Vol 46, No 5, 29 January, purchase, but that does not affect the main re
terisation of stocks down to the ration shop level pp 37-46. sults, so those tables are not reported here. is concerned. Other states have also begun to Bhatty, Kiran (2006): “Social Equality and Develop3 There has been no revision of APL prices since
do this. ment: Himachal Pradesh and Its Wider Signifi2000-01. Meanwhile market prices have risen, so
17 See Drèze and Khera (2010a). See also Swami-cance”, PhD thesis submitted to the London even APL grain is provided at a subsidy today.
nathan (2008) on social exclusion from the PDS. School of Economics.
18 Andhra Pradesh never really followed the central De, Anuradha, Reetika Khera, Meera Samson and hold entitlements (see discussion below). Some of
4 Since 2007, some states have reduced per house
directives on conducting a BPL survey in 1997 and A K Shivakumar (forthcoming): PROBE Revisited these changes are tracked in Table 3. 2002. (New Delhi: Oxford University Press). 5 See R Krishnakumar (1997) and M Suchitra
19 See Drèze (1997 and 2003), S Vivek (2010), Bhatty Deaton, Angus and Jean Drèze (2009): “Food and (2004) on the recent history of the PDS in Kerala.
(2006). N utrition in India: Facts and Interpretations”, 20 Another important reason for high APL offtake Economic & Political Weekly, Vol 44, No 7, 14 Feb
6 See Jha and Howes (1992 and 1994) for more on urban bias in the PDS.
rates is that in 2006, FCI fixed APL allocation for ruary.
states based on the state’s offtake record prior to Drèze, Jean (1997): “Primary Priorities: Learning Lessons recording of PDS purchase, choice of multipliers
7 Three caveats related to these estimates – under
that year (Himanshu 2011). from Himachal Pradesh”, Times of India, 23 July. and accounting practices – are discussed in the 21 The revival of the PDS in Chhattisgarh has re-– (2003): “Where Welfare Works: Plus Points of the next section. ceived some attention. See Dhand et al (2008), TN Model”, Times of India, 21 May. Shukla (2009), Drèze (2010 a & b), Drèze and
8 For a full discussion of these methodological ca-– (2010a): “The Task of Making the PDS Work”, The
Khera (2010b), Patnaik (2010) and Himanshu and
veats, see Khera (2011c). Hindu, 8 July.
Sen (2011).
9 In an informal conversation with an NSS investi-– (2010b): “Losing Their Nerve”, Hindustan Times, gator, it has been learnt that some investigators 14 October. mark items on the NSS schedule that are con-Drèze, Jean and Reetika Khera (2010a): “The BPL Censumed by the household during their household sus and a Possible Alternative”, Economic & Politi-
References
visit, but data on actual quantities consumed in- cal Weekly, 27 February, Vol 45, No 9, pp 54-63.
volves an element of informed guesswork (Jean Adhikari, Anindita and Kartika Bhatia (2010): “NREGA – (2010b): “The PDS Turnaround in Chhattisgarh”, Drèze, personal communication). If this is a wide-Wage Payments: Can We Bank on the Banks? Eco-The Hindu, 13 November.
spread practice, recorded PDS purchases may not nomic & Political Weekly, Vol 45, No 1, 2 January. Dhand, Vivek Kumar, Dinesh Kumar Srivastav, A K Sobe accurate. Further, in such cases, sudden and
Banerjee, Abhijit (2011): “A Platter of Choices”, Indian masekhar and Rajeev Jaiswal (2008): “Compumarked changes in the performance of the PDS,
Express, 28 March. terisation of Paddy Procurement and Public Disare likely to be reflected in the NSS data with a Bardhan, Pranab (2011): “Challenges for a Minimum tribution System in Chhattisgarh”, available lag, as knowledge of such “jumps” may take time
to become widespread enough to inform guesses.
10 In Khera (2011c), alternative estimates of diversion using data from the Indian Human Development Survey (IHDS) conducted in 2004-05 gives comparable levels of diversion (see Table 3 there) and similar patterns at the state-level.
11 According to De et al (forthcoming) and Pratham (2011), mid-day meals were served in 84% (in 2006) and 83% (in 2010) of schools. According to NSS data (2004-05), only 22.5% of rural households and 8% of urban households were benefiting from the scheme (Government of India 2007).
12 The NSS does not rely entirely on census projections. The NSS multipliers are based on their own methods of house listing, etc, and are likely to be more reliable (Angus Deaton, personal communication).
13 Note, that in most states that have expanded the PDS, the tendency is to spread existing FCI allocations over a larger population. This is done, for instance, in Madhya Pradesh, Rajasthan, Kerala, Karnataka (see Table 3).
14 Note that estimates of diversion reported in Himanshu and Sen (2011) present Chhattisgarh as a turnaround. This could be because they fail to take into account the contribution to the PDS from the “state pool”.
15 Though at times the Government of Tamil Nadu
| INDIA AND THE ILO March 5, 2011 | |
|---|---|
| India and the ILO in Historical Perspective India, the ILO and the Quest for Social Justice since 1919 Indian Officials in the ILO, 1919-c 1947 | – Sabyasachi Bhattacharya, J Krishnamurty, Gerry Rodgers – Gerry Rodgers – J Krishnamurty |
| Employment in Development: Connection between Indian Strategy and ILO Policy Agenda Fundamental Principles and Rights at Work: India and the ILO | – T S Papola – Kamala Sankaran |
| For copies write to: Circulation Manager, |
Economic and Political Weekly,
320-321, A to Z Industrial Estate, Ganpatrao Kadam Marg, Lower Parel, Mumbai 400 013. email: circulation@epw.in
Economic & Political Weekly
EPW
SPECIAL ARTICLE
o nline at http://www.iceg.net/2008/books/3/26 _216-223.pdf
Economic Times (2006): “Rice Cheaper for Kerala BPL Families”, Economic Times, 11 Janauary, available online at http://articles.economictimes.indiatimes.com/2006-01-11/news/27435361_1_bplfamilies-kerala-bpl-rice. Last accessed on 9 March, 2011.
Government of India (various issues): Monthly Foodgrains Bulletin (New Delhi: Ministry of Food and Consumer Affairs).
Government of India (2002): “High Level Committee Report on Long Term Grain Policy” (New Delhi: Ministry of Consumer Affairs, Food and Distribution ).
Himanshu and Abhijit Sen (2011): “Why Not a Universal Food Security Legislation”, Economic & Political Weekly, Vol 46, No 12, 19 March.
Howes, Stephen and Shikha Jha (1992): “Urban Bias in Indian Public Distribution-System”, Economic & Political Weekly, 27 (19), pp 1022-30.
– (1994): “Public Distribution of Food in India – A Comment”, Food Policy, 19 (1), pp 65-68.
Jha, Shikha and Bharat Ramaswami (2010): “How Can Food Subsidies Work Better? Answers from India and the Philippines”, Asian Development Bank, Working Paper No 221.
Khera, Reetika (2011a): “UID and Social Welfare Schemes”, Economic & Political Weekly, Vol 46, No 9, 26 February.
Krishnakumar, R (1997): “A Successful System under Threat”, Frontline, Vol 14 (21), 18-31 October, available online at http://www.hinduonnet.com/ fline/fl1421/14210250.htm
Mohan, Gopu (2008): “Rice at Re 1/kg in TN from September 15”, Indian Express, 31 August, available online at http://www.indianexpress.com/ news/rice-at-re-1-kg-in-tn-from-sept-15/355474/. Last accessed on 9 March 2011.
Office of the Commissioners of the Supreme Court (CWP 196/2001) and Office of the Advisors to the Supreme Court Commissioners, Orissa (2010), Status of Implementation of Food Schemes in Orissa, May.
Office of the Registrar General and Census Commissioner (2006): “Population Projections for India and States 2001-2026”, Report of the Technical Group on Population Projections constituted by the National Commission on Population, Census of India 2001 (New Delhi: Office of the Registrar General and Census Commissioner).
Parsai, Gargi (2011): “Survey in Delhi Shows Slum Women Prefer Ration to Cash Transfers”, The Hindu, 17 April, available online at http://www. hindu.com/2011/04/17/stories/2011041764571000. htm, last accessed on 22 April 2011.
Patnaik, Biraj (2010): “Chhattisgarh’s Food for Thought”, Live Mint, 21 April.
Pratham (2011): “Annual Survey of Education Report”, ASER 2010 (New Delhi: Pratham).
Rai, Parshuram (2011): “Performance Audit of Food Security Schemes in Orissa and UP”, Supported by Action Aid India, Centre for Environment and Food Security, New Delhi 2011.
SEWA (2010): “Do Poor People in Delhi Want to Change from PDS to Cash Transfers?”, mimeo, SEWA Delhi, available online at http://www. sewabharat.org/Executive%20summary%20 english-cash%20transfer.pdf. Last accessed on 22 April 2011.
Shukla, Alok (2009): Available online at http://www. expresscomputeronline.com/20090831/egovernancechampions09.shtml
Somanathan, E (2011): “Money Where the Mouth Is”, Hindustan Times, 23 March.
Suchitra, M (2004): “Undermining a Fine System? I ndia Together”, available online at http://www. indiatogether.org/2004/jan/pov-keralapds.htm. Last accessed on 9 March 2011.
Swaminathan, Madhura (2008): “Public Distribution System and Social Exclusion”, The Hindu, 7 May.
Sify News (2008): “BJP Offers Rice at Re 1 per kg”, 11 December, available online at http://www.sify. com/news/bjp-offers-rice-at-1-re-per-kg-newselection-jegv1wfjhic.html
Thai Indian News (2008): “Madhya Pradesh to Provide Wheat, Rice at Subsidised Rates”, available online at http://www.thaindian.com/newsportal/politics/madhya-pradesh-to-provide-wheatrice-at-subsidised-rates_10026700.html, 12 March. Last accessed on 9 March 2011.
The Hindu (2005): “Government to Supply Rice at Rs 3/kg”, The Hindu, 21 April, available online at http://www.hinduonnet.com/2005/04/21/stories/2005042110910400.htm. Last accessed on 9 March 2011.
Vivek, S (2010): “Understanding Public Services in Tamil Nadu: An Institutional Perspective”, Doctoral dissertation, Syracuse, NY Syracuse University. http://www.inewsone.com/2011/04/14/delhi-slumdwellers-want-better-ration-shops-not-cash/43319
