Food insecurities

The most efficient way of addressing the problem of food inflation is to expand and deepen the public distribution system.

THE persistent nature of the food inflation over the last year has brought the acuteness of food insecurity in India into political focus. Data until June 2010 show that while annual point-to-point food inflation rates show signs of moderation, the price index of food articles has continued to increase. The recent increase in fuel prices has added a new dimension to the problem of inflation. The Reserve Bank of India (RBI), in a statement dated July 2, 2010, said: The recent increase in fuel prices will have an immediate impact of around one percentage point on WPI [Wholesale Price Index] inflation, with second round effects being felt in the months ahead.

Inflation is essentially a tax on the incomes of poor people, who are essentially net buyers of food. In India, while landless labourers and urban consumers are almost fully dependent on market purchases of food, a substantial number of small and marginal peasants do not produce enough food for their household requirements. When food prices increase, these households spend a larger share of their incomes on food purchases. In many cases, households reduce food consumption itself or shift to cheaper, less balanced, diets. In other words, food insecurity has close linkages with nutrition security as well as the health status of the poor.

It should be obvious that the most efficient way of addressing the problem of food inflation in India today is to expand and deepen the public distribution system (PDS). However, revealing a mindset that defies the obvious, the United Progressive Alliance (UPA) II government is going ahead with a phased dismantling of the PDS itself. The government's proposed Food Security Bill has been criticised severely by the Left parties and right-to-food activists. It is argued that the Bill, if passed in its present form, will restrict the overall supply of subsidised food in India, exclude large sections of the needy from the ambit of the PDS, and exacerbate insecurities around hunger and malnutrition.

A welfare instrument

A strong case for deepening and expanding the present PDS can be made by alluding to India's experience with the earlier universal form of PDS. The PDS was established in the 1960s with the aim of (a) maintaining stability in the prices of essential commodities across regions; (b) ensuring food entitlements to all sections at reasonable and affordable prices; and (c) keeping a check on private trade, hoarding and black-marketing. The 1960s also marked the beginning of the Green Revolution. While it contributed significantly to the attainment of national food self-sufficiency, these gains were regionally biased in their spread. As a result, the achievement of local food security in the food-deficit regions and the growing urban regions came to depend significantly on external supplies from the food-surplus regions.

It was clear from historical experience that the market could not be a substitute for concerted state action in the procurement and distribution of foodgrains. A national food policy, thus, became critical for food security in the deficit regions. Beginning from the fourth Five-Year Plan period (1969-74), the government institutionalised the policy of procuring foodgrains from farmers on the basis of a procurement price and distributing them to various parts of the country. There were two objectives to this strategy: equitable regional distribution of foodgrains at reasonable prices and the provision of a fair price to farmers. The universal PDS was introduced in 1965 as the most important vehicle for the distribution of procured grains in the deficit regions.

In the period after 1965, the performance of the state as an intervening agent in the food economy was mixed. On the one hand, the Green Revolution had led to a massive regional concentration of foodgrain production. On the other hand, this had the potential to exacerbate the regional disparities in foodgrain consumption. However, studies have shown that the presence of a universal PDS acted as a check against this.

The only major source of information on the use of the PDS before the 1990s is for 1986-87, when the National Sample Survey Organisation (NSSO) conducted its 42nd round sample survey on the utilisation of the PDS. Two conclusions can be drawn from this survey. First, a substantial share of the total quantity of different food items purchased by India's population was from the PDS. In other words, subsidised purchases from the PDS acted as an important supplement to other sources of purchase of food items. The share of purchase from the PDS in the total quantity purchased was higher in urban areas compared with rural areas. The fact that the PDS, with all its infirmities, played a role in keeping in check regional disparities in foodgrain consumption showed that it had the potential to be an instrument of welfare.

Secondly, what was also clear was that the PDS was not serving the vast majority of the country's population even in 1986-87. The performance of the PDS, indicated by the extent of offtake of foodgrains, varied considerably across States. Southern States such as Kerala, Andhra Pradesh and Tamil Nadu ranked the highest in respect of per capita offtake, while the northern States such as Bihar, Madhya Pradesh, Orissa and Rajasthan ranked the lowest. The share of rural households that reported no purchase from the PDS in 1986-87 was about 98 per cent in Uttar Pradesh, Bihar and Orissa. At the same time, about 87 per cent of rural households in Kerala purchased foodgrains from the PDS.

Targeted PDS

Hence, the challenge in the 1990s was to expand considerably the reach of the PDS. However, official policy in the 1990s took the PDS to a completely different trajectory.After the neoliberal turn to economic policies after 1991, primacy was accorded to the logic of fiscal prudence, which entailed drastic reductions in subsidies, such as that on food. Thus, in 1997, the government decided to abolish the universal character of the PDS and convert it into a targeted scheme. Following the introduction of the Targeted PDS (TPDS), the population was classified into Above Poverty Line (APL) and Below Poverty Line (BPL) categories. Only those households classified as BPL were made eligible for subsidised purchase of commodities from ration shops. In the first phase, the APL households were eligible to purchase commodities from ration shops, but had to pay the full economic cost of the handling of commodities.

There are, of course, two immediate issues here. First, there are major problems associated with having a classification of households based on a survey in one year, and then following that classification for many years. The reason is that incomes of rural households, especially rural labour households, fluctuate considerably. A household may be non-poor in the year of survey, but may become poor in another year because of insecurities in the labour market.

Also, the poverty lines that are used are incomes at a near-destitution level. A household that earns an income just above the destitution-poverty line cannot be judged as not requiring social security assistance ( Weakening Welfare by Madhura Swaminathan, LeftWord Books, New Delhi, 2000).

Secondly, the fundamental critique of the concept of targeting has centred on the higher weight that neoliberal reforms give to errors of inclusion than to errors of exclusion. The increase in efficiency realised by excluding all the ineligible persons was given more emphasis compared with the inclusion of all the eligible persons. The errors of inclusion have only financial implications, but errors of exclusion have social costs, which have to be weighted higher.

The experience after 1997 showed that the fears of massive exclusion of the needy from the PDS were to be truly realised in practice. A widespread complaint from many parts of rural India after the introduction of the TPDS has been the existence of a major mismatch between households classified as BPL by the government and their actual standard of living. As noted in the report of the High Level Committee on Long Term Grain Policy (chaired by Abhijit Sen) submitted in 2002, the narrow targeting of the PDS based on absolute income-poverty is likely to have excluded a large part of the nutritionally vulnerable population from the PDS.

In 2004-05, the NSSO conducted another round of survey on the functioning of the PDS. The results from the 2004-05 survey are instructive in the discussion on errors of exclusion. Table 1 shows the distribution of households by types of ration cards possessed in select States; it shows that less than 30 per cent of the Indian rural population was classified as BPL in 2004-05, and thus eligible for the TPDS. The remaining 70 per cent of the rural population was de facto APL' and hence not covered under the TPDS. Further, in poorer States such as Bihar and Rajasthan, fewer than 20 per cent of households had BPL cards.

Let us now consider agricultural labourers, who form the most marginalised section of society. Only 48 per cent of agricultural labourers in rural India possessed either a BPL or an Antyodaya card in 2004-05 (Table 2). In Bihar, fewer than 30 per cent of agricultural labour households possessed either a BPL or an Antyodaya card.

Thus, while the TPDS was ostensibly aimed at reducing the errors of wrong inclusion, it invariably enhanced the errors of wrong exclusion with the attendant social costs.

Estimation vs Identification

Apart from the errors of wrong exclusion, the TPDS had yet another strange operational feature. As mentioned, under the TPDS, the population was divided into BPL and APL households with different allocations and prices. The problem here was how to classify the population into BPL and APL. There was no system to ensure that the estimation and identification of BPL households go hand-in-hand. Hence, a totally unscientific and arbitrary method was followed by the Central government, which further complicated the inherent contradictions of the TPDS.

The total quantity of commodities supplied by the Food Corporation of India (FCI) to each State was fixed by the Centre using a particular criterion. The Government of India used estimates from the quinquennial sample surveys of the NSSO to arrive at the proportion of poor households in each State. The quantity of commodities supplied to the States was fixed as per the allocations for this estimated proportion of poor households. The States, thus, had to design separate census surveys in order to identify these poor households. However, the census surveys conducted by the States, on many occasions, threw up a higher number of poor households than what the NSSO's sample surveys showed.

Given that the total number of poor households was predetermined by the Centre, the costs of providing subsidised foodgrains to the rest of the poor households had to be met by the State governments. The crisis in State finances did not provide any room for States to incur such additional expenditures. The poor households that had not been counted in were classified as APL and kept outside the ambit of the TPDS. The TPDS was thus transformed into a Procrustean bed by neoliberal reforms.

The above discussion makes it clear that any efficient PDS has to be universal in character, as the inherent deficiencies in the methods of targeting result in enormous social costs. It is here that the proposed Food Security Bill of UPA-II fails to meet the mark. The framework in which the Bill is being discussed is rooted firmly in narrow targeting. The suggestions for a reversion to universal PDS are dismissed on the grounds of high fiscal costs. As fiscal rectitude is given as the primary reason for not having a universal PDS, it is important to look at how much a universalised PDS will actually cost.

Cost of Universal PDS

The M.S. Swaminathan Research Foundation (MSSRF), in its recent Report on the State of Food Insecurity in Rural India, attempts to analyse the economic feasibility of a universalised PDS. The calculations in this report are attributed to Madhura Swaminathan and are reproduced from the final report of the National Commission for Farmers (NCF). The MSSRF analysis has the following assumptions:

a) Universalisation implies coverage for at least 80 per cent of the population of India. The universal PDS would exclude (through self-selection) the richest 20 per cent of the population;

b) Universal PDS would provide the prevailing BPL allocations of 35 kg of wheat and rice at Rs.4.15 a kg and Rs.5.65 a kg respectively to 80 crore persons;

c) A provisioning of rice and wheat in the ratio of 2:1; and

d) The current economic cost borne by the FCI and an average family size of five are taken as given (NFHS-3 estimates the average size to be 4.8).

The Centre for Budget and Governance Accountability (CBGA), New Delhi, has also attempted a similar estimation. The assumptions in this analysis are:

a) Coverage of all the 23.96 crore households of the country with 35 kg of foodgrains at the Central Issue Price (CIP) of Rs.3 a kg.

b) The prevailing Minimum Support Price (MSP) and economic costs of wheat and rice are taken as given and the provisioning of rice and wheat is in the ratio of 2:1.

In a recent research project based at the Tata Institute of Social Sciences, Mumbai, we used these two methodologies and calculated the amount of financial resources required for universalising the PDS in 2010-11 (for a summary, see Awanish Kumar and Aditi Dixit, Is a Universal PDS Financially Feasible in India? at https://www.tiss.edu/announcements/attachments/res-brief-PDS.pdf). Table 3 shows the estimates as per the NCF approach, that is, assuming 80 per cent coverage and the CIP meant for BPL population applied to the entire population that is covered.

With the current population estimated at 115 crore, 80 per cent coverage would imply 92 crore individuals. With an average family size of around 4.8, the total number of households to be covered would be 19.17 crore and the total amount of foodgrains required would be 805.1 lakh tonnes. The offtake of foodgrains under the TPDS in 2008-09 was only 348 lakh tonnes.

Thus, the provision of 35 kg of foodgrains to 80 per cent of households in India, at the currently applicable BPL Central issues prices for rice and wheat, will require an additional amount of only about Rs.42,237.9 crore (assuming MSP and economic costs remain constant). This additional amount would amount to only 0.64 per cent of India's gross domestic product (GDP). The total food subsidy required would amount to just 1.48 per cent of the GDP.

If we calculate the total financial allocation required for universal PDS according to the CBGA approach, making some simple modifications, we get a slightly different picture. The added assumption here is that foodgrains are supplied at CIP for Antyodaya households, since we want to arrive at a reasonably practical range of calculations for various possible arrangements under a truly' universal PDS. In other words, through these two sets of calculations, we attempt to present the upper and lower limits of expenditure required for universalising PDS.

Table 4 attempts to calculate the total food subsidy required as per the second method. The additional food subsidy required here would amount to Rs.91,922 crore, which amounts to 1.39 per cent of India's GDP. The total food subsidy required would amount to 2.23 per cent of the GDP.

Table 3 and Table 4, viewed together, present a range' for the possible financial commitment that the government would need to undertake to revert to a universal system of PDS. Table 4 shows a higher additional cost of Rs.91,922 crore since it assumes that all 24 crore households of the country would buy 35 kg of foodgrains.

Compare these required costs with the revenues forgone of the UPA-II government. The UPA-II government has forgone an amount of Rs.414,099 crore in terms of tax revenue and other exemptions for 2008-09 and Rs.502,299 crore for 2009-10, which amounts to almost 79 per cent of the aggregate tax collection in the fiscal year 2009-10 (R.E.) and nearly 8 per cent of India's GDP.

Further, the effective tax rate of the corporate sector, at 22.78 per cent (in itself much below the statutory tax rate of 33.99 per cent), was significantly lower than that of the public sector companies. Even the said amount of revenue forgone is an underestimate since the concerned budget exercise operates only on a sample of 90 per cent companies.

In sum, given the large human and social costs associated with the TPDS owing to the errors of exclusion, the cost of a universal PDS is negligible. If the government could divert a part of the revenue foregone from corporate houses this fiscal year, a universal PDS can be established easily in the country.

Whether the neoliberal dogma that has afflicted the UPA-II government will allow it or not is to be seen.

R. Ramakumar is an Associate Professor at the Tata Institute of Social Sciences, Mumbai.