You are on page 1of 5

Food security in the time of inflation

If the expansion of the Public Distribution System results in an increase in market prices, it
may counterbalance the benefits of food subsidy

half plate:While PDS has begun to play a far greater role in household food consumption in recent years, it only covers less than 50
per cent of a households cereal intake. Photo: AFP
Passage of the National Food Security Act (NFSA) has put the Public Distribution System (PDS) at the core of
the national mission to feed the hungry. The PDS, operated via fair price or ration shops, will distribute up to
5 kg of rice at Rs. 3 per kg., wheat at Rs. 2 per kg, or millet at Rs.1 per kg per person per month to 75 per cent of
the rural population and 50 per cent of the urban population.
Will the PDS be able to handle this scale of distribution? Will availability of practically free grains overcome the
potential inconvenience of buying from a single shop? Will this allocation meet all the households cereal
needs? Some of the data from the India Human Development Survey (IHDS) conducted by the National
Council of Applied Economic Research and the University of Maryland provides a guide to the recent past.
About 42,000 households were surveyed in 2004-05, and once again in 2011-12, and provide an interesting
description of how PDS has grown in importance for ensuring household food security in an era of rapid food
price inflation.
PDS cards come in three flavours Above Poverty Line (APL), Below Poverty Line (BPL) and the Antyodaya
Anna Yojana for the poorest of the poor. Here we combine BPL and Antyodaya card holders. While the Central
government decides on the proportion of a States population that is eligible for BPL status, States identify
which particular households should get which card using their own criteria.
Substantial efforts were made between 2006 and 2012 to improve targeting of BPL cards and expand the
number of Antyodaya cardholders.
Did retargeting of PDS work? Comparing access to BPL cards across different income categories (using 2012
constant prices) paints a picture of the glass both half full and half empty. In 2005, 44 per cent rural and 31 per
cent individuals with monthly incomes of Rs. 300 or below had a BPL card; by 2012 this proportion had risen
to 56 per cent (rural) and 36 per cent (urban). But while an effort was made to include the poorest, affluent
households also benefitted. Among households with monthly per capita incomes of Rs 2,000-2,500, the
proportion of households with BPL cards grew from 25 per cent to 43 per cent in rural areas and from 15 per
cent to 29 per cent in urban areas.
Inefficient targeting
This low relationship between income and type of ration card is worrisome because it suggests that it is difficult
for us to identify the poor and target subsidies. NFSA plans to provide subsidised grains to 75 per cent of the
rural households so it should cover most of the poor; but for urban households where only 50 per cent are
expected to receive BPL cards, this inefficient targeting could be highly problematic with poor households being
excluded from receiving subsidised grain while some middle income households benefit.
Apart from retargeting of PDS, the years 2009-2012 also saw runaway food price inflation. So looking at the
IHDS data from 2005 and 2012 allows us an opportunity to examine the role of PDS in household food security
during times of high inflation.
The biggest change took place in people who actually purchased grains in PDS, termed PDS off take. Only 25
per cent of the respondents purchased grain from the ration shops in 2004, but over 50 per cent did so in 2012.
Food price inflation accounts for much of this increase but perhaps increasing efficiency of PDS shops could
also account for some of the increase. People with APL cards are supposed to receive grains at the market price
and only 12 per cent APL households purchased grain from PDS shops in 2005. By 2012 this proportion had
risen to 29 per cent. Even if they had to pay the full PDS price, in an era of rapidly rising prices, full PDS price
was still lower than the market price. This increase was particularly large in the cities. However, since the
quantities that can be purchased via PDS by BPL households is limited, only about 4.5 kg of grains per month
per person was purchased from PDS, forming about 45 per cent of the total grain consumption. Once
households decided to buy from fair price shops, the amount they purchased was determined by the allowance
and did not vary between two survey years or across different income groups.
These observations create an interesting quandary. On the one hand, PDS has begun to play a far greater role in
the household food consumption in recent years and with the expansion of its role under NFSA, will be even
more important. On the other hand, it only covers less than 50 per cent of a households cereal intake. If the
expansion of PDS results in an increase in market prices, it may have a substantial negative impact on
household budget and may well counterbalance the benefits of food subsidy.
Vast regional differences
While we have painted a national picture here, regional differences in the importance of PDS in household food
budget are vast. Whereas in north-central India only about 40 per cent of the people buy grains from ration
shops, nearly 85 per cent in the south use food from ration shops.
These observations suggest three major challenges for the country as we move forward. First, identifying the
poor is likely to remain a problem, particularly for the urban areas. Second, since PDS covers less than half of
the households cereal budget, if the expansion of NFSA affects cereal prices it will impact all households, even
those covered by NFSA. Third, inter-State disparities will continue to persist given the complex equation
between the Centre and the State in providing food subsidies.
(Sonalde Desai is with the National Council of Applied Economic Reasearch and the University of Maryland.
This is Part 3 of the five-part NCAER series.)

Identifying the poor is likely to remain a problem, particularly for the urban areas
An inclusive growth policy
The impressive gain by rural households in spite of the favouritism towardsnon-primary
activities appears real

The Indian economy has moved on a high growth path since the mid-1980s. After a blip in growth between
1990-92, liberalisation, initiated for aligning the Indian economy with the world in 1991, not only put the
economy back on a higher growth path but also sustained this growth till the 2000s. During the last few years,
India has been the second fastest growing economy in the world.
Despite the high growth over the past two decades, concerns have been raised over the growth not being equally
distributed. Policy makers responded to these concerns arguing for inclusiveness in the 11th Five Year Plan in
2007. How has the rapid growth during the 11th Five Year Plan period helped in improving the income levels of
the most vulnerable Indian households?
Sharing of growth
The aggregate estimates routinely brought out by the Central Statistical Organisation (CSO) show a feel good
factor that real per capita income has been growing rapidly. But there is little evidence on (a) how this
growth has been shared among households in rural India versus urban India and (b) whether households
belonging to different socio-religious groups have grown together. Three rounds of the National Sample Survey
Consumer Expenditure (NSS CE) surveys carried out between 2004-05 and 2011-12 suggest an unprecedented
rise in household expenditure and a consequent decline in poverty. These estimates imply that some benefits of
growth have been shared by vulnerable households. But these data do not clarify whether poverty has declined
because of new social safety net programmes or because vulnerable households have participated in the general
economic growth.
The recently-concluded India Human Development Survey (IHDS) a nationally representative survey of
about 42,000 households conducted by researchers from the National Council of Applied Economic Research
(NCAER) and the University of Maryland examines changes in the incomes of the households during the period
of rapid economic growth, 2004-05 and 2011-12. It is the only nationally representative panel survey covering
the same households. During the two rounds of IHDS, besides a range of outcome indicators, data on
household income and its sources have also been collected.
Though validation of the data is still underway, we present some pointers based on preliminary analysis. The
median real income of the households from all sources had been about Rs. 28,200 in 2004-05; this increased to
about Rs. 37,500 in 2011-12, which is an average of 4.7 per cent annually. Unlike aggregate growth figures
released by the CSO, IHDS data allows calculation of household income by the place of residence of households.
Those IHDS calculations show for the first time that the real average household income in rural India has
increased 5.0 per cent annually almost twice the 2.6 per cent annual growth in urban India. This has resulted
in a significant narrowing of the gap in household income from 2.26 times in 2004-05 to 1.97 in 2011-12.
These figures are consistent with the growth of per capita expenditure calculated from the respective NSS CE
(61st and 68th rounds) monthly per capita expenditure growth in the rural and urban sectors.
When we normalise the household median income by the number of members in the household, the growth of
income in rural India is even more impressive an average annual median per capita income increase of 7.2
per cent, which is more than twice the rate experienced by urban households (3.2 per cent annually). This story
of growth at the aggregate level is fascinating in itself because most of the changes during the liberalisation
phase have favoured the growth of non-primary activities. But the impressive gain by rural households in spite
of the favouritism towards non-primary activities appears real and requires further investigation.
Further proof of growth
We note similar differences in median income growth across different socio-religious groups that provide
further confirmation of the inclusiveness of the recent economic growth. In IHDS surveys, we have defined six
social and religious groups high caste Hindus, Other Backward Classes, Dalits, Adivasis, Muslims and Other
Religious Minorities. The highest growth in the median per capita incomes is reported for Dalits (7.8 per cent
annually) and OBCs (7.3 per cent), while the real median income of high caste Hindus grew only at 4.6 per cent
annually. The average income growth of other vulnerable groups was also higher than that of high caste
Hindus. The income of Adivasis grew at 5.7 per cent annually while the income of Muslims grew by 5.4 per cent.
A working plan
Our preliminary results point towards the largest gains for the traditionally vulnerable households rural
areas, Dalits, OBCs, Adivasis and Muslims. This narrowing of group differences is all the more remarkable in
the face of a slightly diverging overall income distribution. Our preliminary calculations of per capita income
inequality suggest a small increase from a Gini ratio of 53 in 2004-5 to 55 in 2011-12.
The relatively greater progress of vulnerable sectors despite this growing inequality seems to suggest that the
inclusive growth policy implemented during the 11th Five Year Plan may have been working. While a much
more rigorous analysis is required to delineate the factors that have led to this, our conjecture is that some of
the social sector schemes like the Mahatma Gandhi National Rural Employment Guarantee Act, Janani
Suraksha Yojana, the National Rural Health Mission et al. may have contributed to this inclusive growth.
(Amaresh Dubey is professor at the Centre for the Study of Regional Development, Jawaharlal Nehru
University and Reeve Vanneman is professor at the University of Maryland. This concludes the five-part
series of the IHDS-II findings. The views expressed here are personal.)