Towards New Poverty

Lines for India

Himanshu
 Introduction
• This paper presents the result of an exercise prepared for the Planning
Commission’s Expert Group to Review the Methodology for Estimation
of Poverty to draw up new poverty lines and, correspondingly, new
poverty estimates based on the National Sample Survey consumption
data.

• This paper amplifies on a suggestion made to the committee

constituted by the Planning Commission to review the existing
methodology for official poverty estimation in India.
 CONCEPTS
• Poverty Line, Poverty Line Basket and Poverty Ratio • The conventional
approach to measuring poverty is to specify a minimum expenditure (or
income) required to purchase a basket of goods and services necessary to
satisfy basic human needs. • This minimum expenditure is called the poverty
line. •
• The basket of goods and services necessary to satisfy basic human needs is
the poverty line basket or PLB. •
• The proportion of population below the poverty line is called the poverty ratio
or headcount ratio (HCR).
 Uniform Recall Period/Mixed
Recall Period
• Up until 1993-94, the poverty line was based on URP data. This involved
asking people about their consumption expenditure across a 30-day recall
period. In other words, the information was based on the recall of consumption
expenditure over the last month alone.
• Since 1999-2000, however, data are being collected according to MRP. Under
this method, data on five less-frequently used items are collected over a one-
year period, while sticking to the 30-day recall for the rest of the items. The
low-frequency items include expenditure on health, education, clothing,
durables etc. Currently, all poverty line data are compiled using the MRP
method.
• most recent estimates by the Suresh Tendulkar and Rangarajan Committees.
 BACKGROUND
• Dadabhai Naoroji , ‘Poverty and the Un-British Rule in India’. poverty line
ranging from Rs 16 to Rs 35 per capita per year, based on 1867-68 prices.
• “cost of a subsistence diet consisting of ‘rice or flour, dhal, mutton, vegetables,
ghee, vegetable oil and salt’
• 1938, the National Planning Committee (NPC) estimated a poverty line ranging
from Rs 15 to Rs 20 per capita per month
• 1944, the authors of the ‘Bombay Plan’ (Thakurdas et al 1944) suggested a
poverty line of Rs 75 per capita per yea
• : In 1962, the Planning Commission constituted a working group ….formulated
separate poverty lines for rural and urban areas – of Rs 20 and Rs 25 per
capita per year respectively.
• VM Dandekar and N Rath made the first systematic assessment of poverty in
India in 1971, based on National Sample Survey (NSS) data from 1960-61.
They argued that the poverty line must be derived from the expenditure that
was adequate to provide 2250 calories per day in both rural and urban areas.
• Alagh Committee (1979): In 1979, a task force constituted by the Planning
Commission for the purpose of poverty estimation, chaired by YK Alagh

Minimum
consumption
Area Calories expenditure
(Rs per capita
per month)
Rural 2400 49.1
Urban 2100 56.7
• Lakdawala Committee (1993)
• (i) consumption expenditure should be calculated based on calorie
consumption as earlier;
• (ii) state specific poverty lines should be constructed and these should be
updated using the Consumer Price Index of Industrial Workers (CPI-IW) in urban
areas and Consumer Price Index of Agricultural Labour (CPI-AL) in rural areas;
and
• (iii) discontinuation of ‘scaling’ of poverty estimates based on National
Accounts Statistics.
• Tendulkar Committee (2009): (i) consumption patterns were linked to the 1973-74
poverty line baskets (PLBs) of goods and services, whereas there were significant
changes in the consumption patterns of the poor since that time, which were not
reflected in the poverty estimates;
• (ii) there were issues with the adjustment of prices for inflation, both spatially (across
regions) and temporally (across time); and
• (iii) earlier poverty lines assumed that health and education would be provided by the
State and formulated poverty lines accordingly.
• It recommended four major changes:
• (i) a shift away from calorie consumption based poverty estimation;
• (ii) a uniform poverty line basket (PLB) across rural and urban India;
• (iii) a change in the price adjustment procedure to correct spatial and temporal issues
with price adjustment; and
• (iv) incorporation of private expenditure on health and education while estimating
poverty.
• The Committee recommended using Mixed Reference Period (MRP) based estimates, as
opposed to Uniform Reference Period (URP) based estimates that were used in earlier
methods for estimating poverty.
 SUGGESTIONS OF PLANNING
COMMISSION'S EXPERT GROUP
1. Poverty in India is measured purely on the consumption dimension
and that all other dimensions, including calorie norms on which
present poverty lines were originally constructed, are incidental and
only of historical significance.
2. Poverty lines used across space and time should represent equivalent
purchasing power parity (PPP) at whatever reference consumption
level is taken to be the cut-off for basic minimum needs.
3. In order to maintain continuity of presently accepted notions of the
minimum standard of living required to avoid absolute poverty, it is
desirable that the reference cut-off be anchored to some aspect of
present practice.
 Contd.
4. Since official urban poverty estimates are less controversial, our
choice of reference consumption cut-off is the equivalent of MRP the
present official all-India urban poverty line.
5. National Sample Survey now uses MRP in most of its surveys. It is
necessary to have poverty lines that correspond to MRP distributions.
6. Although the new poverty lines are not based on any particular norm
of basic needs, and are outcome of a purely technical exercise to
calculate cost of living indices relevant around the present all-India urban
poverty line, this choice can be defended normatively.
 Current official poverty lines
• Step 1: to identify a norm for food expenditure. Recommended
Dietary Allowances (RDA) (2,400 calories per day per capita -rural
areas , 2,100 calories per day per capita-urban areas)
• Step 2: to identify monthly per capita consumption expenditure
(MPCE), including non-food expenditure, corresponding to the
respective RDA, (all-India poverty lines fixed).
• Step 3: state-wise poverty lines for 1973-74 -calculated from these all-
India poverty lines by applying Fisher Indices of state prices relative to
all-India
• Step 4: updating these state-wise poverty lines, (CPI-AL for rural areas
and CPI-IW for urban) (Himanshu, 2010)
Cont…..

• calorie norms -----only as a peg to derive MPCE norms with

which to measure consumption poverty.
• “It neither applied calorie norms to obtain 1973-74 state-
specific poverty lines nor suggested their use to track
poverty changes over time.”
• Once all-India poverty lines were fixed using calorie norms
for 1973-74, all further adjustments were only for
variations in cost of living across states and over time.
 Deaton’s Approach
• uses price data from the consumption expenditure surveys of NSSO + Fisher’s
index for both the interstate price differential and the rural-urban price
differential.
• uses the rural-urban price differential to correct imbalances in poverty lines
across rural and urban areas of each state.
• Starting point: 1987-88 consumption expenditure survey. (all India rural
poverty line)
• involves use of rural-urban indices along with interstate price indices, there
can be only one normative bundle in this kind of exercise.
 Problems in Deaton’s Approach
• his indices are based on unit values from the NSSO consumption expenditure
surveys and therefore limited to only food items
• share of food in total expenditure has been declining secularly over the years

FINDING
• rural-urban price differential obtained using unit values from NSS consumption
expenditure surveys was only 15% (higher for urban areas) (in 1999-2000).
 New poverty lines
(points of departure from Deaton’s
approach)
• The basic approach of setting the new poverty line is similar to the exercise attempted by
Deaton. The last year for which Deaton’s exercise is reported is 1999-2000. However, for all
calculations in this paper we have used the consumption expenditure survey of 2004-05.
• Use of 61st round of data- most recent
• price data from NSS surveys rather than the CPI price data.
• the interstate and rural-urban price indices used in the calculation of the new price, indices
are much more comprehensive (includes non food item also like clothing fuel, footwear etc.)
• we use the all-India urban poverty line of 2004-05 as the starting point of other state poverty
lines in urban areas and also in rural areas.
• budget shares used in our calculation of price indices are based on mixed recall
• period.
 New poverty line
• The new poverty lines correlate very well with Deaton’s across states but show
more urban-rural difference, mainly on account of wider coverage of non-food
items.
• existing official poverty lines exaggerate urban-rural price difference,
spuriously place rural poverty as less than urban in as many as nine major
states, and also correlate poorly with interstate variations in cost of living.

• Different from Deaton: estimated rural poverty – from 28.3% to

41.8%.

• Why? present official all-India urban poverty estimate as reference

• Because: adequacy of poverty lines to reflect nutrition need and validity
of procedures to update poverty lines benchmarked far back in 1973-74. (both
applicable in case of urban poverty estimates)
 Calculation
• price data obtained from the consumption expenditure survey - food,
intoxicants, fuel, clothing and footwear.
• median prices of each item in each state and sector. (not in Deaton’s
approach)
• PDS items (rice, wheat, sugar and kerosene), the prices obtained -
aggregating the PDS and non-PDS items together.
• clothing and footwear - prices obtained from the 365-day estimates of these
commodities.
•
 DROPPED ITEMS
• The cut-off for selecting these items is Rs 0.15 per month (Rs 1.8 per year for
365-day items).
• If these items have lower than this consumption in six or more states out of a
possible combination of 70 values (35 states and union territories for rural and
urban areas each) they have been dropped.
• 11 items dropped.
• For those states and sector where the item has been consumed by less than
five households, it has been treated as not being consumed in that state.
• These state-wise price and quantity data was used to calculate Fisher’s index
 EDUCATION AND HEALTH
• Education, (no such unit values like NSSO). Therefore, we have used the EUS
of NSSO for obtaining the index of cost of education.

• calculated the cost of education per school going child in each state for rural
and urban areas for children in the age-group of 5-15 from the 61st round
(2004-05) of EUS.
• Health expenditure we have used the 60th round (January-June 2004) which
focuses on health expenditure. For construction of the index, we have
calculated health expenditure per treatment in case of non-institutional
medical care and health expenditure per case of hospitalisation for
institutional medical care.