What exactly is the poverty line?

Poverty is a normative concept. Its definition differs across countries and over time. The concept of poverty line, therefore, cannot be delinked from its historical discourse. It has to be seen in reference to the social, political and economic context in which poverty was initially defined and the methods that were used to measure it over the years.

Measuring poverty

Poverty can be classified into two groups: absolute poverty or relative poverty. Relative poverty is a variant of the measure of inequality in a society. Absolute poverty, on the other hand, is a measure of deprivation of food, clothing, shelter, health care and education (or any other human need considered essential) in a society. For example, in the United States, official poverty line is defined as three times the cost of a minimum food diet in 1963, which translated into $22,811 for a family of 4 (2 adults and 2 child) in 2011. By this measure, nearly 15% of the population was poor in the US in 2011.

At the current exchange rate (~61), the official poverty line in the US would translate into an income of Rs. 13.84 lakh, which would likely lie in the top 5% income bracket in India. (Still people complain that salaries in India are becoming comparable to the developed world and that we would soon lose our competitiveness—an argument I am unable to fathom).

Defining poverty in India

In India, the current definition of poverty has its roots in the definition adopted in 1977 by the task force headed by Y. K. Alagh. The poverty line was defined on the basis of minimum calorie intake—2400 calories per person per day in rural areas and 2100 calories in urban areas—coupled with some non-food consumption considered absolutely necessary to get the required calories (see Table 1). The task force used the NSSO’s all India survey to estimate the level of expenditure that would translate into meeting the required calorie norm. The poverty line was estimated to be Rs.49.09 in rural India and Rs. 56.64 in urban areas.

The poverty line was, thus, “observed” from real expenditure and consumption pattern of the poor. This definition, as stated by the Lakdawala committee, “was partly normative and partly behavioural”. Though anchored in the calorie norm, it measures the purchasing power necessary to meet the minimum calorie standard rather than evaluate actual malnourishment or under-nourishment. The calorie norm serves only as a first order approximation for measuring the poverty line.

Table 1: Definition of Poverty Line in India

Year	Poverty Line (expenditure per month per capita in Rupees)		Definition
	Urban	Rural
Working Group, 1962 (at 1960-61 prices)	25	20	Minimum standard of living, excluding spending on health, education and shelter, which were supposed to be provided by the state
Dandekar and Rath, 1971 (at 1960-61 prices)	22.50	15	Calorie norm of 2250 per person per day
“Task Force” of Planning Commission, 1979 (at 1973-74 prices)	56.64	49.09	Average calorie intake per person of 2435 calories in rural and 2095 calories in urban areas—based on estimates by the Nutritional Expert Group

There were two major attempts to revise the poverty line, with the Lakdawala committee in 1993 and the Tendulkar committee in 2009.

The Lakdawala Committee: Focus on state-level estimates
The Lakdawala committee focused its attention on three aspects of poverty estimation: the base-year consumption basket, the choice of price index for adjusting the poverty line over time and estimation of poverty at state level.

On the first issue, it decided to retain the existing definition to retain comparability over time.

To update the poverty line, it recommended constructing a new price index for the poor by using CPI (-IW, -UNME and -AL) instead of using the price deflator from National Accounts Statistics estimates of private consumption.

State level poverty estimates were becoming increasingly controversial, primarily because the centre’s financial assistance to states depended on it and many states claimed that the planning commission grossly underestimated their poverty status. The committee agreed that the current methodology needed to be revised. It recommended adjusting the base year (1973-74) poverty line for observed differences in cost of living across states. It then updated it for future years using the consumption pattern of the poor and the state-level consumer price indices.[1]

Tendulkar Committee: Taking a slightly broader view of poverty

While clearly stating that there is “an inevitable element of arbitrariness in numerically specifying the poverty line” and that the “in the interest of continuity” it would be best “to situate it in some generally acceptable aspect of the present practice”, the Tendulkar committee made four radical departures from the past, of which two are especially important:

1.     It did not pay too much heed to the nutritional requirement. The Tendulkar committee decided to ignore the calorie norm. Lakdawala committee had put the head count ratio at 25.7% in urban areas in 2004-05, using the calorie norm as the anchor. Tendulkar committee estimated the total per capita consumption expenditure associated with it. The purchasing power represented by this per capita total consumption expenditure was taken to be the “poverty line basket”, to be applied to rural areas after allowing for rural-urban price differential. Note that with this method the rural-urban differential in calorie norm disappeared.

Inevitably, the application of the new method meant lower calorie consumption by the poor as it had to allow for expenditure on various other heads, including health and education. However, the committee accepted this based on the notion that per capita calorie intake in India has declined over the years, which suggests a change in preferences of consumers. Further, as per FAO recommendations, 1800 calories per day may be enough for a person in urban area to function without any decline in productivity, and this matched the committee’s calculation of what the poor were consuming.

2.     They changed the price deflator. The Tendulkar committee used the NSS implicit price deflator to update the poverty line, both at the all-India and state level. This has been considered a major correction from the past as it refers to the same consumption basket that the poor consume rather than apply a generic price deflator.

The poverty estimates of Tendulkar committee are much higher than in the past (see table 2), which shows that the poverty line was revised substantially.

Table 2: Estimates of Poverty Line and the Head Count Ratio

Year	Poverty Line (expenditure per month per capita in Rupees)		Head Count Ratio (Percentage of population below poverty line)
	Urban	Rural
Official estimates of Poverty
1972-73	47	41	51.5
1977-78	69.9	60.6	48.3
1983-84	117.5	101.8	37.4
1987-88	152.1	131.8	29.9
Lakdawala Committee
1973-74	49.63	56.96	54.9
1977-78	56.84	72.5	51.8
1983	89.45	117.64	44.8
1987-88	115.43	165.58	39.3
1993-94			36.0
2004-05			27.5
Tendulkar Committee
1993-94	-	-	45.3
2004-05	578.8	446.68	37.2
2009-10	859.6	672.8	29.8
2011-12	1000	816	21.9

Source: Planning Commission, various reports.

Two primary concerns

Two concerns of the various expert groups on estimating poverty have been: (1) Identify poverty based on some objective notion of what it means and how to interpret it using (NSS) data; (2) Consider the historical narrative and not create a new notion of poverty in vacuum.

They needed to retain some way of comparing the numbers over the years. Inter-temporal comparisons allow us to understand whether our policies are working and to what extent. For instance, Tendulkar committee’s estimates suggest that poverty has declined faster in the years after 2004-05 than in the previous decade. This could have related policy implications for growth and anti-poverty programmes.

Conclusion

Defining the poverty line was a very important step in the Indian context as it set the agenda for a model of development that reduced poverty over time. It provided an objective framework for estimating poverty, researching the causes and consequences of poverty, and designing targeted anti-poverty programmes.

An interesting feature of the work of expert groups has been that both the (Lakdawala and Tendulkar) committees significantly revised upwards the estimates of poverty. This shows that the poverty line has been revised upwards, in keeping with changing social norms. Yet, a link with the past was always retained.

While we definitely have the choice to redefine the poverty line to suit our new preferences, delinking it completely from its historical narrative will mean that we will lose the option to evaluate the effectiveness of our policies and programmes. Revising it substantially upward would also have the implication of spending more on anti-poverty programmes and increasing the resource transfer from the centre to the states.

Finally, it is ironic that Professor Tendulkar had often argued that our poverty estimates were too low. It is ironic that he is now being cast in the shadow of the devil who defined the poverty line too low; despite a fairly big jump in their estimate of poverty in India.

---

[1] A more detailed explanation cannot be attempted here. Readers are advised to read the report of the expert group: Planning Commission, “Report of the Expert Group on Estimation of Proportion and Number of Poor”, July 1993 < http://planningcommission.nic.in/reports/publications/pub93_nopoors.pdf>.