Public Distribution of Rice in Andhra Pradesh: Efficiency and Reform Options

Public Distribution of Rice in Andhra Pradesh: Efficiency and Reform Options

Surajit Deb

This paper explores various efficiency aspects of the Rs 2 a kg rice scheme, which has been relaunched in Andhra Pradesh. The statistical analysis suggests a substitution of demand between the open market and public distribution system purchase in response to changes in the open market price of rice in the state. An examination of the consumer benefit and subsidy burden involved in the rice scheme during 1983-2007 indicates that the gap between the two typically remained small. In the light of the central government’s PDS reform package, the paper suggests strategies for a cost-effective PDS management in Andhra Pradesh.

This paper is derived from the study titled “Grain Management in Andhra Pradesh: Scope for Reforms and Private Sector Participation” s ubmitted to the International Food Policy Research Institute. The author has received help from P K Joshi (IFPRI), and cooperation from the ofﬁces of the Andhra Pradesh State Civil Supplies Corporation and Food Corporation of India in Hyderabad.

Surajit Deb (debsurajit@gmail.com) is at Ram Lal Anand College, U niversity of Delhi, Delhi.

he Andhra Pradesh government has relaunched its Rs 2 per kg rice scheme in the 2008-09 ﬁscal, which could raise the cost of rice subsidy bill to the tune of Rs 19,800 million from a level of Rs 7,200 million in 2007-08 (GoAP 2008). It may be recalled that the subsidised rice scheme is not new in the state and the government in the past has perceived that the supply of cheap rice to the poor families is of fundamental importance to ensure food security. The Rs 2 rice scheme was introduced during 1982-83 in the public distribution system (PDS) and the rate was ﬁnally ﬁxed at Rs 5.25 per kg in 2000-01 after several changes during the 1990s. The government supplied rice at this rate to the below poverty line (BPL) families through its targeted PDS (TPDS) till recently. The working of the Andhra subsidised rice scheme has been critically discussed by several analysts – Radhakrishna and Indrakanth (1987), Olsen (1989), Indrakant (1997; 2000), Krishna Rao (1993), Parthasarathy (1995), Radhakrishna et al (1997), World Bank (1997), Dutta and Ramaswami (2001), and Indrakant and Harikishan (2003) – and the effectiveness of the scheme has been questioned with regard to its poor targeting efﬁciency and the ﬁscal burden of the subsidy bill.

1 Introduction and Objectives

Although, the government’s endeavour remains to implement the PDS to the best advantage of the beneﬁciaries, the working of the PDS in India has received much attention from the academic and policymaking circles mainly due to its targeting efﬁciency and its subsidy burden to the government (Hanumantha Rao and Radhakrishna 1997; Thamarajakshi 1997). It is often argued that PDS in large countries like India can hardly possess any built-in targeting mechanism because restricting the access only to the poor turns out to be administratively unachievable and expensive. Therefore, the central government in several occasions has intended to reform the existing PDS of foodgrains and restructure the food subsidy regime. Several measures were suggested in the recent past to reform the costly and largely non-selective PDS scheme (GoI 2000, 2001a, 2001b, 2002, 2003 and 2004). The suggested PDS reform package fundamentally incorporated a d ecentralised procurement mechanism and food coupon based distribution system and intended to introduce several other measures including allowing the private grain companies to procure directly from the farmers to eliminate transaction costs. Though the Andhra government has taken initiative to reform the PDS delivery mechanism by issuing the new biometric ration cards, the state has not yet adopted the decentralised procurement scheme or taken any step to direct its PDS programme to the

december 19, 2009 vol xliv no 51

targeted group. The number of BPL cards in Andhra Pradesh has	rice offtake proﬁle of Andhra Pradesh from the central pool re
increased since the government sanctioned additional BPL cards	serve as well as the rice offtake from the state PDS outlets. The
from year to year. According to the recent budget document the	examination of the cost accruing to APSCSCL and the rice subsidy
new rice scheme plans to cover more than 80% of the state popu	burden followed by an analysis of government costs vis-à-vis con
lation. As a manager of the public distribution network among	sumer beneﬁts is provided in Section 4. The food coupon system
Indian states, Andhra Pradesh currently draws huge quantities of	in Andhra Pradesh and its impacts are discussed in Section 5. In
foodgrains from the central allotment and also encounters an	conclusion, Section 6 deals with providing strategies for a cost
enormous rice subsidy bill. The subsidy incurred on rice by the	effective PDS in the state.
Andhra Pradesh State Civil Supplies Corporation Limited (APSC-SCL) reached Rs 7,169 million in 2006-07 ﬁscal as against	2 Andhra PDS Network
R s 5,428 million in 2005-06 and the government provided	Food security falls under the Directive Principles of State Policy
Rs 7,200 million in the budget towards the supply of subsidised	in India’s federal structure, whereby the central government de
rice to BPL families in 2007-08. One may doubt whether the pro	termines the states’ allocation and the state settles the fair price
posed rice subsidy scheme with an almost universal and state	shop (FPS) prices plus the amount for distribution. Earlier analy
wide coverage could be an effective policy for redistributing	sis has indicated that the Indian PDS remained untargeted and
i ncome in the state?	generated low poverty-alleviating effects (Dev and Suryanaray-
It is now increasingly being recognised that the success of any	ana 1991; Jha 1992; Ahluwalia 1993; Radhakrishna et al 1997;
food subsidy programme depends on its targeting efﬁciency	Mooij 1999; Swaminathan 2000). The 62nd round of National
meaning that the beneﬁts should be restricted to the target group.	Sample Survey (NSS) (2005-06) collected information on various
It has been observed from the experiences of food subsidy pro	types of ration cards used in the PDS, which revealed that about
grammes of developing countries that the public cost of the pro	79% of the households possessed some kind of ration cards
gramme becomes unsustainable when the subsidy is untargeted.	(BPL: 39%, above poverty level (APL) 41%, and Antodaya Anna
We, therefore, intend to inquire about the usefulness of the recent	Yojana (AAY) 4%) in rural areas, and 68% of the households pos
move to relaunch the subsidised rice scheme in the state, which	sessed the same (BPL: 19%, APL: 64%, and Antodaya: 1%) in urban
has a previous track record of being an untargeted programme	areas. Although the proportion of families with Antodaya and
with almost universal coverage. This paper therefore attempts to	BPL r ation cards tends to go down with the expenditure class,
examine the efﬁciency aspects of the subsidised rice scheme and	about 14% of the rural and 27% of urban families belonging to
consequently suggests strategies for a cost-effective PDS strategy	the top monthly per capita expenditure class still possessed such
in Andhra Pradesh. We think this issue is crucial and could not	cards. The I ndian PDS has also been criticised for urban bias of
have been ignored at a time when better grain management is	FPS, negligible coverage in areas/states with higher poverty, and
crucial for the country amid global shortages. With this objective	lack of transparent and accountable arrangement for delivery.
and the context in mind, the plan for the rest of the paper is as	The Andhra PDS scheme in the past has been categorised as
follows. We brieﬂy discuss the characteristics of the PDS network	e xtensive with almost universal coverage (Dutta and Ramas
in Andhra Pradesh and in particular provide an outline of the rice	wami 2001). When it comes to extending the beneﬁts of PDS,
distribution scheme in Section 2. Section 3 deals with the per-	Andhra Pradesh deﬁnes poverty more inclusively even today.
formance indicators of the rice scheme. For this, we examine the	Thus, while the central government measures 16% of the state
Table 1: PDS Characteristics of Andhra Pradesh (as on 31 March 2004)	population as below the poverty line, the state government con-
Projected % of Size of No of No of BPL Ration Cards (Lakh)	siders about 60% of the same in the BPL category. A comparison
Population BPL House-House-House-BPL APL AAY Total	of recent PDS-characteristics of Andhra vis-à-vis all states shows
in 2000 Popuholds holds in holds (Lakh) lation 2000 in 2000	that the number of state BPL cards are greater than the number of
(Lakh) (Lakh)	BPL households, indicating high errors of inclusion in the state
Andhra Pradesh 754.7 85.78 4.77 158.21 40.63 128.1 58.7 12.3 199.18	PDS (Table 1).1
35 states 9,969.4 – 5.51 7,203.78 652.03 650.8 1,440.2 173.1 2,264.1	The effective implementation of PDS depends on the physical
Source: Department of Food and Public Distribution.	access to the FPS network, whereas their location is determined
Table 2: Fair Price Shops in Andhra Pradesh Districts	by the district administration considering the village size and
Districts FPS Districts FPS Districts FPS (per thousand (per thousand (per thousand	suitability. There is one shop for every 1,965 persons in Andhra,
hectares) hectares) hectares)	as against the government of India norm of one FPS for 2,000
Srikakulam 3.09 Costal Andhra 1.98 Medak 1.56	persons (GoAP 2006). It was noticed earlier that there is a greater
Vizianagaram 2.06 Kurnool 1.18 Mahbubnagar 1.09	density of FPS in urban areas than in rural areas at the all-India
Visakhapatnam 1.50 Anantapur 1.21 Nalgonda 1.28 East Godavari 2.05 Cuddapah 1.05 Warangal 1.37 West Godavari 23.41 Chittor 1.79 Khammam 0.76 Krishna 2.32 Rayalaseema 1.29 Karimnagar 1.66 Guntur 2.08 Ranga Reddy 1.90 Adilabad 0.80 Prakasam 1.10 Hyderabad 41.82 Telangana 1.31 Nellore 1.37 Nizamabad 1.43 Andhra 1.52	level (Howes and Jha 1992). As on 30 September 2006, there are about 41,296 FPS functioning in Andhra Pradesh, out of which 83% are in rural areas, with each shop holding 485 family cards in the state (rural and urban). Thus, the PDS coverage appears to have remained extensive in rural Andhra Pradesh. To examine the geographical spread of the FPS network, we have derived the
Source: Calculated using data from APSCSCL and DES, Andhra Pradesh.	number of FPS per thousand hectares of area in different districts/
Economic & Political Weekly december 19, 2009 vol xliv no 51	71

SPECIAL ARTICLE

regions of the state (Table 2, p 71). There are differences in the FPS penetration among the districts and regions in Andhra Pradesh. The districts in coastal Andhra have better PDS network as compared to the districts in the Rayalaseema and Telengana regions.

The success of a PDS system also depends on the incentive structure provided at the end-point delivery, viz, FPS dealer’s margin. The functioning of the FPS in various states of India is considered unworkable due to lack of incentives in the grain distribution business, which leads to illegal erosion of PDS grains into the open market. The FPS dealers are known for raising their income through bogus cards, cheating the beneﬁciary by delivering a smaller quantity and diverting supplies to the open market. The wholesale and retail margin of PDS commodities for dealers in Andhra is low, viz, Rs 39.71 and Rs 17.11, respectively, for rice in cities. For wheat, the wholesale and retail margin in cities is ﬁxed at Rs 38.90 and Rs 18.00, respectively. The dealer’s margins in areas other than cities in Andhra are to some extent higher for both rice and wheat. Since the existing pattern of margin provided to retail dealers is not attractive, it has led to the large-scale leakage and pilferage in the Andhra PDS. The Report of the Working Group on PDS and Food Security for the Tenth Five-Year Plan has provided state level estimates of diversion of grains from the PDS, and it is found that 19% of rice and 15% of wheat in Andhra are diverted to open markets.

2.1 Rice Distribution Scheme

The Andhra rice scheme is essentially an extended PDS programme whereby the state government provides additional subsidy on the central issue price (CIP). There had been many changes with regard to the issue price, quantity allocation and quality of

Table 3: Details of Rice Distribution Scheme in Andhra Pradesh (2003-04)

As noted in Section 2, bogus cards were in circulation right from the launch of the scheme, so that the recorded number of beneﬁciary from the scheme came to be greater than the total number of households in the state.2 Dutta and Ramaswami (2001) subsequently observed that the PDS targeting in Andhra Pradesh is marked with low errors of exclusion. They, however, detected high “errors of inclusion” particularly in the rural areas, and therefore hypothesised that the urban non-poor has voluntarily opted out from the scheme for better quality grain available in the open market. It may also be noted that the number of ration cards in the state registered a sharp increase after the introduction of the Rs 2 rice scheme and continued to grow even in the recent past. Thus, while the state experienced poverty reductions over time, there has not been any reduction in the number of PDS beneﬁciaries even today. According to the recently released estimates by the Public Accounts Committee, there are about 8.55 lakh ineligible families in the Andhra PDS. Given the extent of leakage in the PDS, it is doubtful whether the rice subsidy is able to fulﬁl the primary objective of redistributing income in the state. It may be noted in this context that the impact of the rice subsidy scheme in providing food security in Andhra Pradesh has been examined by evaluating the relation between PDS and children’s nutritional status assessed through anthropometric measurements (Tarozzi 2002). The results doubted on the ability of the Andhra PDS for providing food security to poor households in the state.

3 Rice Offtake Patterns

A generally poor offtake of foodgrains has been noted as the worrying features of the PDS system in India. The high levels of CIP along with deteriorating quality

Scheme No of Cards (Lakh) Eligibility Criteria Issue Price (Rs/Kg) Scale of Distribution have often been identiﬁed as the

White cards (state government) 129.12 Below poverty Line 5.25 4 kg per member, main r easons for this phenomenon

maximum 20 kg per card (R adhakrishna 2002). At times, the Pink cards (state government) 58.72 Above Poverty Line 9.00 Maximum 30 kg per card

mismatch between the state’s alloca

AAY(central government) 9.35 Poorest of the poor 3.00 Max 35 kg per card

tion and demand or the local adminis-

Annapurna (central government) 0.93 Old/senior citizens Free 10 kg per beneficiary

trative bottlenecks have been recog-

Welfare hostels/institute 8.58 Children/women of the poor 4.00 15 kg per boarder (central government) and weaker section. Residents of nised as the possible weak points of welfare hostels and institutions the I ndian PDS (Jha 2002). The

Source: Derived from APSCSCL records, Hyderabad.

rice distribution in the scheme. The scheme – introduced during 1982-83 – enabled families having income less than Rs 6,000 per annum to get 5 kg of rice, per month per person (subject to a ceiling of 25 kg per month per family) at a subsidised rate of Rs 2 per kg. These households were issued green cards and the others yellow cards. The yellow cardholders were entitled for the same quantity of rice but at Rs 3.50 per kg. Subsequently, the green cards were converted into white cards and yellow cards into pink cards in 1991. The FPS issue price was raised to Rs 3.50 and then to Rs 5.25, while the ceiling on rice was reduced to 16 kg, then r estored back to 25 kg and ﬁnally reduced to 20 kg. The previous operational detail of the Andhra rice distribution scheme is provided in Table 3. The proposed scheme is expected to cover 651 lakh people out of the total state population of 822 lakhs. The government plans to provide 4 kg of rice per head, subject to a maximum of 20 kg per family per month, under the new scheme.

foodgrain offtake in the country declined drastically since the changeover from the universal PDS to the TPDS that took place during 1997-98. The offtake under PDS was found to decline further owing to the upward revision in the central i ssue price as well as diminishing APL sales in 2000, and subsequently the PDS offtake in India as a whole picked up since the drought year 2002-03. An examination of the NSS data indicates that the d ependence on PDS rice has increased in both rural and urban Andhra between the period 1999-2000 and 2004-05. The NSS data for 2004-05 shows that about 23% and 15% of rural and urban purchase of rice in Andhra Pradesh are from PDS as against 13% and 11% of rural and urban rice purchase in all states.

3.1 Andhra Pradesh’s Offtake from Central Allocation

The rice offtake by Andhra Pradesh under various PDS schemes ((BPL, APL and AAY) from the central pool allocation is compared with the all-India pattern in Table 4 (p 73). The total offtake of

december 19, 2009 vol xliv no 51

rice as percentage of allocation in Andhra Pradesh remained at about 66% against all-India average of 45% in recent years. The scheme-wise offtake for BPL, APL and AAY remained in the range of 100%, 47% and 98% as against 76%, 17% and 93% for all states, respectively during the triennium ending in 2006-07. The data on rice offtake patterns (offtake as percentage of allocation) under various welfare schemes of the government in 2004-05 are analysed and it is found that except for the special component of

Table 4: Andhra Pradesh’s Rice Offtake Patterns in Comparison to All-India

Triennium Ending Offtake as % of Allocation in Andhra Offtake as % of Allocation in India BPL APL AAY Total BPL APL AAY Total

1993-94 – – – 96.2 – – – 81.4

1996-97 – – – 85.9 – – – 66.1

1999-2000 95.1 91.3 -91.9 84.0 78.6 – 80.3

2002-03 78.7 36.1 66.9 56.1 60.7 17.1 74.2 41.2

2006-07 100.2 46.8 97.9 66.4 75.8 16.8 93.2 44.5

Source: Derived from Ministry of Food, Consumer Affairs and Public Distribution. Table 5: Regression Results on Rice Offtake in Andhra Pradesh (1982-83-2006-07)

Explanatory Variables Dependent Variable: Rice Distribution in Andhra OLS Cochrane-Orcutt OLS Cochrane-Orcutt Method (AR-2) Method (AR-2)

Constant 17.29 17.30 17.74 18.08 (10.81)* (4.14)* (12.71)* (6.20)*

Wholesale price of rice (lagged) 0.008 0.009 (3.63)* (1.53)*

Retail price of rice (lagged)	0.008	0.008
	(3.82)*	(1.91)*
Summary Statistics
R2 0.37	0.61 0.39	0.62

R-bar2 0.35 0.54 0.37 0.56

DW-statistics 0.63 2.01 0.67 2.10

F-statistics 13.20* 9.49* 14.59* 9.82*

* Indicates statistical significance at 10% level. Table 6: Implication of Subsidised Rice in APSCSCL Transactions (1999-2000)

Particulars Subsidised Rice (Rs Lakh) Total (All Commodities) (Rs Lakh)

1.1 Sales 75,361.9 89205.7

1.2 Subsidy on PDS 1,09,651.5 (99.5%) 110237.6

1 Trading income 1,85,028.3 200873.9

2.1 Purchase 1,77,657.0 192242.1

2.2 Trading expenses 5,353.8 (88.1%) 6080.4

2 Trading cost 1,83,010.8 (92.7%) 197335.1

3 Administrative expenses 2017.6 2692.4

Figures in brackets indicate percentage of rice in total transaction of APSCSCL. Source: Derived from Annual Report, APSCSCL, GoAP.

the Sampoorna Grameen Rozgar Yojana (SGRY) scheme, the offtake level in Andhra remained higher than the offtake level of all states. In 2004-05, the scheme-wise offtake for the welfare scheme like nutrition programme, mid-day meal, SGRY, special SGRY, welfare institutions and hostels remained in the range of 82.20%, 78.82%, 55.78%, 28.06% and 83.69% as against 45.11%, 60.30%, 44.72%, 70.85% and 29.74% for all states, respectively. Accordingly, it can be inferred that while various states in India are not able to draw the quota of rice to be delivered to their r espective PDS, Andhra Pradesh lifts a BPL quota which is higher than its allotment. The higher offtake of rice in Andhra is also reﬂected in the amount of food subsidy in the state. The data from GOI (2008) indicate that while the all-India share of food subsidy (TPDS allocation) in majority of the Indian states turns out to be lesser in relation to their all-India share of people below

Economic & Political Weekly

december 19, 2009 vol xliv no 51

the poverty line, Andhra Pradesh commands 10.2% of the all-India TPDS subsidy with 4.2% of the all-India BPL population (ibid). Thus, the state-wise ratio of the percentage allocation of subsidies to the proportion of below poverty population reveals that while the ratio is less than one for most of the states, it is highest in Andhra Pradesh. One may, however, keep in mind the diversion of the state’s central pool offtake to the open market, which is around 25% of the offtake under BPL quota in A ndhra Pradesh.

3.2 Determinants of Rice Distribution in Andhra Pradesh

It is believed that the PDS beneﬁciaries normally switch their demand from FPS to open market due to increases in the issue price. The change-over of demand from FPS purchase to the open market purchase at retail rates has been deﬁned as the “switchover effect” occurring from the rise in PDS price. According to Balakrishnan and Ramaswami (1997), the effect can be investigated by examining the relationship between PDS offtake and open market price. Thus, if the PDS offtake rises along with the market price then consumers are switching back to the PDS, and the entire change in offtake can be viewed to have materialised due to the switchover effect. This hypothesis is tested by using the distribution of subsidised rice in the state as observations. Two different proxies, viz, wholesale and retail price, have been used to capture the open market price of rice.3 Table 5 contains estimates from alternate regressions of subsidised rice distribution in Andhra Pradesh on the lagged market prices. Since the ordinary least squares (OLS) estimates revealed the presence of serial correlation (as evident from the Durbin-Watson (DW) test statistic), the Cochrane-Orcutt estimates are provided after c orrecting the serial correlation. The results indicate a positive association between the PDS offtake and open market price of rice in the state. This implies that there is some switchover effect on demand for rice from the open market to PDS occurring due to a rise in the open market rice prices. The switchover of demand in this case occurs due to the substitution of open market p urchase by the PDS purchase arising due to increasing open market price.

4 Costs and Subsidy Burden

4.1 APSCSCL Costs

The APSCSCL is entrusted to carry out procurement and distribution of rice wheat and maize in the state.4 An analysis of A PSCSCL’s earnings and its distribution pattern revealed that the cost of purchase of PDS items constituted about 98% of its total expenditure and about 55% of its total earnings were met out of state subsidy in 1999-2000. In view of the claim that the magnitude of Andhra Pradesh’s subsidised rice scheme is enormous, the total ﬁnancial transactions of APSCSCL along with its item-wise particulars involved in the subsidised rice is worked out in Table 6. It is observed that subsidised rice a ccounts for about 88% of the total trading expenses and 99.5% of total A PSCSCL subsidy. This reiterates that the crucial ﬁnancial i mplication of the subsidised rice scheme in the A PSCSCL’s ﬁ nance.

SPECIAL ARTICLE

Figure 1: Real Central and State Rice Subsidy Rates

1982-83 1984-85 1986-87 1988-89 1990-91 1992-93 1994-95 1996-97 1998-99 2000-01 2002-03 2004-05 2006-07

The various elements of APSCSCL’s cost structure on subsidised rice from 1985-86 to 2002-03 are summarised in Table 7. The three year average of the total cost is given in per unit basis (Rs per metric tonne); while the cost components are expressed as percentages of the total cost. It can be perceived that the cost of rice paid to Food Corporation of India (FCI) constitutes the largest part of APSCSCL’s expenditure on rice subsidy. Among the

Table 7: Percentage Components of APSCSCL’s Cost on Subsidised Rice

Triennium Cost of Rice Freight Storage Handling Commission Interests Administrative Total Cost Ending Paid to FCI Charges Expenses (Rs/MT)

1985-86 91.43 3.29 5.28 2,431.43

1995-96 96.91 1.76 0.07 0.20 0.03 0.44 0.58 5,962.87

2002-03 95.84 2.13 0.06 0.35 0.06 0.24 1.31 7,557.68

Cost components are expressed in percentage of the total cost. Source: Derived from APSCSCL records, Hyderabad.

major operational cost components, the share of freight expenditure and administrative expenses in the total cost have increased and the costs due to interest on borrowed capital declined in r ecent years. It may be noted that the interest cost burden of A PSCSCL is based on a concessional rate of interest granted under the “food credit”, whereby the Corporation receives credit from the banking system for the purchase of PDS requirements by h ypothecating the grain stock and the state government providing a guarantee. The storage costs to APSCSCL arise due to the hiring of storage capacity in case its own capacity falls short. Since the APSCSCL’s own godowns are located only at limited number of districts in Medak, Vizianagram, Kurnool, Guntoor, Visakhapatnam, East Godavari, West Godavari, Ranga Reddy and Chittoor, the Corporation also utilises hired storage capacity of the state and central warehousing corporations. The structure of storage cost can be seen to have remained more or less ordered during the last 10 years. In addition to the above-mentioned costs, the APSCSCL also pays 1% market fee in respect of the allotment and 4% VAT on the economic cost of FCI.

4.2 Subsidy Burden

The outﬂows on account of the welfare programmes have r emained a fastest growing item in the expenditure budget of A ndhra Pradesh, and the rice subsidy scheme happened to be the largest of these programmes in terms of costs (World Bank 1997). The CIP is ﬁxed for states to lift the FCI supplies, and state subsidy occurs due to the difference between CIP and the issue price at state outlets, besides the costs arising out of state’s handling. T able 8 provides the detail on how the rice subsidy burden in the state results due to the gap between FCI’s issue price and consumer price under different schemes.

74

The huge ﬁnancial burden on account of rice subsidy in Andhra Pradesh also surfaces due to the targeted size of beneﬁciary families. Although the scheme was originally targeted towards BPL families, the coverage remained almost universal (Dutta and R amaswami 2001). The government policy has remained committed towards sanctioning additional white cards and bringing in more BPL families within the scheme (GoAP 2003, 2007a). The subsidy involved in rice distribution under various schemes is i llustrated in Table 9. It is seen that almost half of the total rice is being purchased at APL rates and being distributed at BPL rates. The huge subsidy burden therefore results from improper targeting and the linked sale below the CIP, besides the costs involved with the APSCSCL’s handling of central allocations.

The state subsidy rate on rice distribution scheme in Andhra Pradesh has been worked out in real terms on the basis of the t otal subsidy bill and the distribution amount of rice in the state (Table 10). It is apparent that the total subsidy registered a secular rise till 1999-2000 and indicated some decline since then. Earlier analysis had indicated that the Andhra subsidy rate r emained higher than the central subsidy (Radhakrishna et al 1997; I ndrakant 2000; Dutta and Ramaswami 2001). Figure 1 plots the two (after deﬂating by the all-India wholesale price index (WPI) of rice, base: 1993-94), and their movements are found to cross each

Table 8: Scheme-wise FCI Issue Price and End-Consumer Price of Rice in Andhra Pradesh

Scheme FCI Issue Price (Rs/Kg) End-Consumer Price (Rs/Kg)

Below poverty line 5.65 5.25

Above poverty line 8.30 9.00

Social welfare institutions 5.65 4.00

Antodaya Anna Yojana 3.00 3.00

Annapurna 5.65 Free of cost

NPNSPE (mid-day meals) Free of cost Free of cost

SGRY (special component) Free of cost Free of cost

SGRY Free of cost Free of cost

Source: APSCSCL records, Hyderabad. Table 9: Scheme-wise Rice Subsidy in Andhra Pradesh (2003-04)

Scheme (No of Card in Lakh)	Quantity	FCI’s	State-	Total	PDS Price	Subsidy
	(Lakh MTs)	Issue Price	Specific	Cost	(Rs/kg)	(Rs/kg)
		(Rs/kg)	Costs (Rs/kg)	(Rs/kg)
Total white cards 129.12	24.36

BPL rate: 68.43 13.14 5.65 0.62 6.27 5.25 1.02

APL rate: 60.69 11.22 8.30 0.65 8.95 5.25 3.70

AAY: 9.35 3.92 3.00 0.59 3.50 3.00 0.59

Annapurna: 0.93 0.11 8.30 0.70 9.00 Free 9.00

Hostels/welfare institution 2.00 5.65 0.59 6.14 4.00 2.14

Source: APSCSCL records, Hyderabad.

Table 10: State Subsidy on Rice Distribution Scheme in Andhra Pradesh

Triennium Subsidy Bill Subsidy Bill Distribution Subsidy Rate Subsidy Rate Subsidy Rate Ending (Rs Lakh) (Rs Lakh) (Lakh Qtl) (Rs Per Qtl) (Rs Per Qtl) (Rs Per Qtl)

(Deflated by (Deflated by (Deflated by GDP Deflator, GDP Deflator, All-India Base: 1993-94) Base: 1993-94) WPI of Rice

Base: 1993-94)

1984-85 6.641.0 15,150.4 128.2 41.6 96.1 89.9

1989-90 22,903.7 36,698.6 202.5 113.2 181.5 188.3

1994-95 46,711.3 45,336.0 207.5 218.6 213.3 210.5

1999-2000 83,586.0 56,230.0 225.8 366.9 249.7 240.8

2003-04 34,471.7 20,327.3 212.9 162.8 96.1

december 19, 2009 vol xliv no 51

Figure 2: Consumer Benefit and Subsidy Cost (Rs/Qtl) in Andhra Rice Scheme 5 Food Coupons in Andhra Pradesh

The food stamp or food coupon can just be another method among various others for ensuring proper targeting of the food subsidy programme, viz, rationing, direct distribution, selft argeting or the food-for-work programme. The food stamp offers greater consumer freedom to the beneﬁciary since it allows purchase of any quantity at the market price over a wide range of speciﬁed items as compared to food coupon, which can be used to buy limited quantities of a speciﬁc number of items at dis-

1982-83 1984-85 1986-87 1988-89 1990-91 1992-93 1994-95 1996-97 1998-99 2000-01 2002-03 2004-05 2006-07

other several times between 1982-83 and 2006-07. The central subsidy rate remained higher till about the mid-1980s and subsequently the state subsidy rate grew at a faster rate so that it stayed above the central rate from 1987-88 to 1997-98. The central subsidy rates continued to rise at a faster rate and remained higher than the state subsidy rate since 1997-98 till recent years. The two subsidy rates came closer in 1999-2000 when their trends seemed to be converging but their subsequent movements revealed a d iverging trend since then. However, while the central rice subsidy rate seems to have remained stagnant in recent years, the state rice subsidy rates indicated a rising trend particularly since 2003-04.

4.3 Government Cost vis-à-vis Consumer Benefit

It is often argued that the escalating FCI costs remained responsible for the high cost-beneﬁt ratio of PDS in India. Jha and Srinivasan (2002) had earlier provided the beneﬁt-cost calculation at the all-India level. While the consumer beneﬁts in this study are deﬁned as the difference between market price and CIP, the d ifference of the CIP from the administrative and storage cost has been added to the procurement price for the calculation of the government cost. Their results indicated a higher cost-beneﬁt ratio for wheat in comparison to rice for the all-India distribution. Recently, a Planning Commission (GOI) study calculated the delivery cost of one rupee worth of PDS under various welfare schemes in India. We have compared the beneﬁts and subsidy costs involved on account of the rice subsidisation scheme in Andhra Pradesh. The consumer beneﬁt is deﬁned as the difference between market price and CIP, and two proxies have been used to capture the market price of rice in Andhra, viz, state wholesale and retail price of the common rice variety.5 The subsidy cost (Rs/Qt) has been arrived by considering only the A PSCSCL subsidy rates involved in the Andhra rice distribution scheme. Figure 2 plots the consumer beneﬁt along with the subsidy burden from 1982-93 to 2006-07. It is observed that though both the proxies of the consumer beneﬁt remained higher than the government cost, the latter remained quite close to the beneﬁts received in the rice scheme from 1982-83 to 2000-01. It is only since 2001-02 that the consumer beneﬁts seem to have stayed above the government cost owing to the decline in rice subsidy rate in Andhra Pradesh. It may be recalled that this decline in rice subsidy rate correspond to the upward revision of the endconsumer issue price in the state from Rs 3.25 to Rs 5.25 in 2001. Subsequently, the beneﬁt and cost appear to have again come closer since 2003-04 and continued to remain so till 2006-07

o wing to the rise in state subsidy costs.

Economic & Political Weekly

december 19, 2009 vol xliv no 51

counted price. The high delivery cost of the PDS grain coupled with costs of maintaining government’s PDS network and high targeting errors have often led to the belief that food stamp or food coupon could provide an effective means of transferring income to the poor in the country. The Tenth Plan document has recommended the introduction of a food stamp and food credit programmes for restructuring the PDS.6 The food coupon scheme was introduced in the state for rice distribution in September 1999 and came into effect from September 2000. Under this scheme, the Civil Supplies Department distributes coupons to the card-holders, who obtain their supply from FPS by submitting the coupons. The FPS dealer has to surrender these coupons to the department to get subsequent supply, and the latter has the full knowledge about the PDS offtake pattern before releasing the next period’s allocation. The system has two immediate advantages, viz, it can reduce the scope of diversion of PDS rice to the open market along with the elimination of dealer’s malpractices, and second, it can reduce the number of bogus cards including reduction in the number of members in the household cards. It is argued about 8% of bogus white cards in the state has been weeded out after the introduction of coupon system (Ravi and Indrakant 2003). A district-level study also corroborates the reduction in the number of cards as well as the amount of rice distributed after the introduction of coupons in six sample districts (Table 11).

6 Strategies for Cost-Effective PDS in Andhra Pradesh

In the context of the Rs 2 rice scheme in Andhra Pradesh, a study had earlier recommended to narrow down its targeting towards the poorest households and bring down the subsidy component to 50% of the government cost (World Bank 1997). The state g overnment has however restarted the cheap rice scheme in the 2008-09 ﬁscal. One would therefore inquire about the ﬁnancial implications, suitability and sustainability of this scheme. The analysis contained in this paper provides some useful insights into these aspects. Our examination conﬁrms that the PDS coverage in Andhra Pradesh has remained extensive particularly in rural areas and a substantial number of non-poor also receive the

Table 11: Impact of Rice Coupon Scheme in Selected Districts of Andhra Pradesh
Districts	% Reduction in Cards	% Reduction of Rice
Mahaboobnagar	1.6	1.5
Khammam	8.3	8.3
Kurnool	9.0	6.6
Chittoor	4.0	9.8
Vizianagaram	4.6	7.4
East Godavari	7.5	17.7
Source: Indrakant and Rao (2003).
		75

SPECIAL ARTICLE

PDS beneﬁts due to high errors of inclusion. The offtake of rice as percentage of allocation is the highest in Andhra Pradesh in comparison to the all-India average under the BPL scheme. We also observe a positive association between the PDS offtake and open market price of rice, which implies that there is some substitution of demand between open market and PDS purchase in response to changes in the open market rice prices. The subsidised rice scheme imposed substantial burden on the ﬁnancial operations of APSCSCL due to the rise in price paid to FCI and also faster increase in the cost components for freight expenditure and admini strative expenses. A comparison between the central and Andhra rice subsidy rate (in real terms) revealed that the rice subsidy rate in Andhra is rising since 2003-04. An analysis of the consumer beneﬁt and subsidy burden involved in the state rice scheme during 1983-2007 indicates that the gap between the two remained small in most of the sample years. According to the state government’s own data, the rice subsidy burden next year could go up by almost three times. The increase of such an enormous proportion in the state expenditure for a populist programme could create precarious ﬁnancial positions in the annual budget of Andhra Pradesh. This burden of public expenditure for rice subsidies appears unjustiﬁed considering the high errors of inclusion in the state PDS, where even the high income groups have – access to beneﬁts. Further, we doubt whether the subsidised rice scheme would be sustainable over time in view of the PDS reforms policy of the central government.

Although, Andhra Pradesh had earlier taken leads in initiating reform measures in other areas, the state’s initiative in effecting PDS reforms remained inadequate. The government claims that it started distributing computerised ration cards along with bar coded coupons with a view to provide universal coverage, elimination of duplicate cards and cards with ineligible/non-residents/ non-existing families known as bogus/ghost cards (GoAP 2007b). However, the state government has done very little to target the subsidy to the poor people and in fact seemed committed to sanction additional BPL cards from time to time (GoAP 2003). The e xperiences from the Sri Lankan food subsidy programme has shown that when the cost of the general food subsidy turns out to be high, it becomes necessary to restrict the beneﬁts only to the poorest sections of population and improve the delivery mechanism (Edirisinghe 1987). Thus, the Andhra Pradesh government in the near future would need to take steps to convert the rice subsidy programme into a poor-only scheme by redesigning the identiﬁcation, targeting and delivery mechanism. It appears that the introduction of smart cards only for the poor households and food stamps-based distribution in the state can make the rice scheme more effective.

A populist food subsidy scheme like the Rs 2 rice scheme in Andhra Pradesh can be sustained only in the short run and bears the risk of turning out to be ﬁscally taxing in the long-run. The continuation of the rice programme would also depend on the state government’s ability to overdraw rice from the central supply. It may be noted that in view of the declining buffer stock p osition, the government has taken steps to rationalise the allocation to states. A reduction in the state’s allotment and the policy of promoting decentralisation might make it difﬁcult for Andhra

SAMEEKSHA TRUST BOOKS

Global Economic & Financial Crisis Essays from Economic and Political Weekly

In this volume economists and policymakers from across the world address a number of aspects of the global economic crisis. One set of articles discusses the structural causes of the financial crisis. A second focuses on banking and offers solutions for the future. A third examines the role of the US dollar in the unfolding of the crisis. A fourth area of study is the impact on global income distribution. A fifth set of essays takes a long-term view of policy choices confronting the governments of the world.

A separate section assesses the downturn in India, the state of the domestic financial sector, the impact on the informal economy and the reforms necessary to prevent another crisis.

This is a collection of essays on a number of aspects of the global economic and financial crisis that were first published in the Economic & Political Weekly in early 2009.

Pp viii + 368 2009 Rs 350

Available from

Orient Blackswan Pvt Ltd

Mumbai Chennai New Delhi Kolkata Bangalore Bhubaneshwar Ernakulam Guwahati Jaipur Lucknow Patna Chandigarh Hyderabad Contact: info@orientblackswan.com

december 19, 2009 vol xliv no 51

Pradesh to overdraw from the central rice pool in the coming procurement mechanism and distribute these cereals in the PDS years. Further, the central government in its efforts to contain scheme. Thus, the state could do better in food security managethe rising food subsidy bill might implement a rise in issue price ment by adopting the decentralised system, since it would progiven the MSP hikes that have taken place in recent years. The vide the ﬂexibility of APSCSCL’s choice of grains and also widen subsidy bill in Andhra Pradesh could accordingly go up if the and deepen the PDS outreach by serving local needs. It therefore proposed issue price hike is implemented. Considering that appears that a decentralised PDS administration along with some coarse cereals like jowar and bajra are preferred in some regions subsidy rationalisation can presumably make the state’s rice of the state, the Andhra government can strengthen the state scheme more effective.

Notes – (2003): Economic Survey, 2003-04 (Hyderabad: – (2002): “Domestic Trade Restrictions and Food Planning Department, Andhra Pradesh Security” in Kirit S Parikh and R Radhakrishna

1 According to Cornia and Stewart (1993), the tar-

S ecretariat). (ed.), India Development Report (New Delhi: geting efﬁciency of public distribution system can – (2006): Economic Survey, 2005-06 (Hyderabad: O xford University Press).

arise from wrong inclusion of non-poor (Type E error) or wrong exclusion of poor (Type F error) Planning Department, Andhra Pradesh Jha, S and P V Srinivasan (2002): “Targeting Food

in the scheme.

Secretariat). Subsidies” in K Seeta Prabhu and R Sudarshan

– (2007a): Budget Speech for the Year 2007-08, (ed.), Reforming India’s Social Sector (New Dellhi: 2 According to Venugopal (1992), the number of

p resented on 19 February 2007. Social Science Press).

r ation cards in circulation was 139 lakhs during 1986, while the estimated number of household – (2007b): Socio Economic Survey (Hyderabad: – (2003): “On Improving the Effectiveness of the was 121 lakhs. Planning Department, Andhra Pradesh PDS in Achieving Food Security” in S Mahendra 3 The wholesale price of rice in the state is derived Secretariat).Dev, K P Kannan and Nira Ramchandran (ed.), by taking the weighted average of wholesale – (2008): Budget Speech for the Year 2008-09, op cit. price of rice (common variety) in important rice p resented on 16 February 2008. Mooij, Jos (1999): Food Policy and the Indian State markets. The markets (with varieties in the pa-Government of India (2000): “Expenditure Reforms (Delhi: Oxford University Press).

rentheses) are: Kakinada (akkulu), Vijayawada Commission, Report on Food Subsidy”, presented Olsen, Wendy (1989): “Eat Now Pay Later: Impact of (coarse), Nijamabad (coarse), Bhimavaram to Government of India on 10 July.Rice Subsidy Scheme”, Economic & Political (akkulu) and Tadapaligudem (akkulu). The – (2001a): Report of the Working Group on Public W eekly, 24(28).

source for the basic wholesale price data are Distribution System and Food Security for the 10th Parthasarathy, G (1995): “Public Intervention and “Agricultural Prices in India” (DES, Ministry of Five-Year Plan (2002-07), TFYP Working Group

Rural Poverty: Case of Non-Sustainable Reduc-Agriculture). Similarly, the retail price of rice in Serial No 66/2001 (New Delhi: Planning

tion in Andhra Pradesh”, Economic & Political

the state is interpreted as the average of retail C ommission, Government of India).

Weekly, 30(41-42). prices at selected centres (viz, Vizianagaram, – (2001b): Excess Food Stocks, Public Distribution

Radhakrishna, R (2002): “Food and Nutrition Security”

Rajahmundry, Vijaywada, Adoni, N S B complex S ystem and Procurement Policy (New Delhi: in K S Parikh and R Radha krishna (ed.), India

and M S B complex), as reported in various G overnment of India).

Develop ment Report (New Delhi: Oxford University

i ssues of Statistical Abstract of Andhra Pradesh

– (2002): Report of the High Level Committee on

Press.

(DES, Andhra Pradesh).

Long-term Grain Policy, Department of Food and

Radhakrishna, R and S Indrakant (1987): “Effects of

4 Besides cereals, APSCSCL also handles petroleum Public Distribution, Ministry of Consumer Affairs, Rice Market Intervention Policies in India: The

products, kerosene, levy sugar, edible oils, etc. Food and Public Distribution (New Delhi: Case of Andhra Pradesh” in J K Johnson (ed.),

5 See note 3 for the construction of the wholesale G overnment of India).

Evaluation of Rice Market Intervention Policies:

and retail price of rice (common) in Andhra – (2003): Buffer Stock Policy for the Tenth Five-Year Some Asian Examples (Manila: Asian Develop-

Pradesh. Plan (New Delhi: Government of India).

ment Bank).

6 It is suggested that the state government instead – (2004): Central Government Subsidies in India, Radhakrishna, R, K Subbarao, S Indrakant and

of issuing a ration card could issue a subsidy enti- Ministry of Finance, Government of India.

C Ravi (1997): “India’s Public Distribution Systlement card (SEC), which would show the

– (2008): Economic Survey (New Delhi: Ministry of

tem: A National and International Perspective”, number of members in a poor family and their Finance).

World Bank Discussion Paper No 380 (Washington age and indicate their entitlement level for food

Hanumantha, Rao, C H and R Radhakrishna (1997): DC: The World Bank).

stamps. Each family would collect the monthly

“National Food Security: A Policy Perspective for

Rao, Krishna I Y R (1993): “Andhra Pradesh: An quota of food stamps on the b asis of the SEC and

India” in G H Peters and Joachim von Braun (ed.),

Experiment in Food Security”, Economic &then use these stamps at any food supply shop to

Food Security, Diversiﬁcation and Resource

Political Weekly, 28(37).

buy foodgrains (rice and wheat) at a price below

Management: Refocusing the Role of Agriculture?,

the open market price. The retailer will then be Ravi, C and S Indrakant (2003): “Food Security and

P ro ceedings of the Twenty-Third International

reimbursed the difference by the state Public Distribution System” in C H Hanumantha

Conference of Agricultural Economists held

g overnment. Rao and S Mahendra Dev (ed.), Andhra Pradesh

in California (Aldershot: Ashgate Publishing

Development: Economic Reforms and Challenges

C ompany).

Ahead (New Delhi Manohar Publishers and

Howes, S and S Jha (1992): “Urban Bias in Indian

D istributors).

References PDS”, Economic & Political Weekly, 27(19). Swaminathan, Madura (2000): Weakening Welfare:

Indrakant, S (1992): “Public Distribution System in

Ahluwalia, D (1993): “Public Distribution of Food in

The Public Distribution of Food in India

Andhra Pradesh with Special Reference to Rice”,

India”, Food Policy, 18 (1).

(New Delhi: Left Word Press).

Indian Journal of Economics, 72(286).

Balakrishnan, P and B Ramaswami (1997): “Quality Tarozzi, Alessandro (2002): “The Indian Public Distri

– (1997): “Coverage and Leakages in PDS in Andhra

of Public Distribution System: Why It Matters”, bution System as Provider of Food Security: Evi-

Pradesh”, Economic & Political Weekly, 32(19). dence from Child Anthropometry in Andhra

Economic & Political Weekly, 32(4).

– (2000): “Food Security and Public Support:

Cornia, G A and F Stewert (1993): “Two Errors of Pradesh”, Princeton University Working Paper

T argeting”, Journal of International Development, A Study of Andhra Pradesh” in N Krishnaji and No 185, May.

5(5). T N Krishnan (ed.): Public Support for Food S ecurity (New Delhi: Sage Publications). Thamarajakshi, R (1997): “Economic Reforms and

Dev, S Mahendra and M H Suryanarayana (1991): “Is Public Distribution System Urban Biased and Indrakant, S and S Harikishan (2003): “Food Security State Intervention in Foodgrain Market”, Economic & Political Weekly, 26 July.

P r o-rich”, Economic & Political Weekly, 26(4). in Andhra Pradesh in Retrospect and Prospect”, in S Mahendra Dev, K P Kannan and Nira Ram-Venugopal, K R (1992): Deliverance from Hunger: The

Dutta, B and V Ramaswami (2001): “Targeting and

chandran (ed.), Towards a Food Secure India: PDS in India (New Delhi: Sage Publications).

E fﬁciency in the Public Distribution System: Case of Andhra Pradesh and Maharashtra”, Economic & I ssues and Policies (New Delhi: Manohar Publish-World Bank (1997): India, Andhra Pradesh: Agenda for Political Weekly, 36(18). ers and Distributors). Economic Reforms, Report No 15901-IN (Washing-Edirisinghe, N (1983): The Food Stamp Scheme in Sri

Indrakant, S and Nageshwar Rao (2003): Working of ton DC: Industry and Finance Division, South Lanka: Costs, Beneﬁts and Options for Modi ﬁcations,

Food Coupon Scheme in Andhra Pradesh Asia Region).Research Report No 58 (Washington DC: Inter-(Hyderabad: Centre for Economic and Social – (2003): Unlocking the Growth Opportunities in national Food Policy Research I nstitute). Andhra Pradesh: An Agenda to Achieve the Vision

Studies).

Government of Andhra Pradesh (2000): Twenty-Fifth Jha, S (1992): “Consumer Subsidies in India: Is Target-2020 Growth Targets, (Washington DC: Poverty Annual Report (Hyderabad: Andhra Pradesh State ing Effective?”, Development and Change, Vol 23, Reduction and Economic Management Sector Civil Supplies Corporation Limited).pp 101-28. Unit, South Asia Region).

Economic & Political Weekly

december 19, 2009 vol xliv no 51 77