In this featured article, Utsa Patnaik, Professor of Economics (Retired) at Jawaharlal Nehru University in New Delhi, argues that the Indian government massively undercounts the poor because the consumption standard against which poverty is measured has itself been allowed to fall over time. If the original definition of poverty line is used then three-quarters of India’s population today are poor. China also would see a dramatic increase, and the World Bank’s global poverty numbers would be revised far upward.
Poverty estimates from India’s Planning Commission recently stated that a person was to be considered ‘poor’ in 2009-10 only if his or her monthly spending was below 22.4 Rupees per day – the equivalent of 48 U.S. cents — in rural areas, and 28.7 rupees per day in urban areas. This has exposed how unrealistic ‘poverty lines’ are.
Some television channels assumed that these figures covered food costs alone and showed how they could not meet even a fraction of a persons minimal nutrition needs at today’s prices. These paltry sums, however, are supposed to cover not only food but all non-food essentials, including clothing and footwear, fuel for cooking and lighting, transport, education, medical costs and rent.
Even a school child knows that working health cannot be maintained, nor basic necessities obtained, by spending so little. Amazingly, however, 350 million Indians subsist below these levels. They can hardly be said to ‘live’ in any true sense: their energy and protein intake, and consumption of cloth and other necessities, is far below normal. They are underweight, stunted, and subject to a high sickness load, but without the means to obtain adequate food or medical treatment. The official poverty lines do not measure poverty any more; they measure destitution.
The outcry against calling these destitution lines, ‘poverty lines’, is justified, for true poverty lines are much higher and show 75 percent of all Indians to be poor. Per head energy and protein intake, as well as cloth consumption, has been falling for the last two decades. Despite a good growth rate of per capita income, this result is because of worsening income distribution, with a small minority monopolizing all the gains while the majority has suffered loss of purchasing power. With 80 million tons of unsold public food grain stocks — 50 million tons in excess of normal levels — piled up by July 2012, the sensible policy is to do away with targeting and revert to a universal distribution system, combining it with an urban employment guarantee scheme.
Unfortunately the neo-liberal policy makers today ask the wrong question: ‘How can we reduce the food subsidy?’, and not the right question: ‘How can we lift the masses of India from the current level of the lowest food consumption in the world, even lower than the least developed countries?’
The poverty line for rural India has been revised upwards by 15 percent by the Tendulkar Committee, whose brief in 2009 was to review poverty measurement, but it still remains absurdly low. Members of the Planning Commission and the Tendulkar Committee are academic experts, so how have such laughable figures of minimum cost of living emerged from their statistical labors? The fact is that over thirty years ago the then-Planning Commission made a mistake of method, and the present Commission stubbornly continues to cling to that mistake despite its being repeatedly pointed out, including by this author (The Republic of Hunger, 2004). The mistake was to change the definition of poverty line and de-link it from nutrition standards.
The original definition of ‘poverty line’ was a sensible one, based on an Expert committee recommendation in 1979, and was applied to the National Sample Survey data on consumption spending. The NSS presents, every five years, data on 12 groups of spenders ranging from the poorest, able to get only about 1300 calories per day to the richest with about 3000 calories, The poverty line was defined as that particular level of total spending per capita on all goods and services, observed from these data, whose food spending part satisfied the nutrition level of.2400 calories of energy intake per day in rural India, and 2100 per day in urban areas. The rural norm was scaled down soon to 2200 calories. Both nutrition norms are modest. The Commission accepted the Expert Committee’s nutrition-based definition but applied it only once, to the 1973-4 data, to obtain correct monthly rural/urban poverty lines of 49 and 56 rupees per day. At those levels, 2200 and 2100 calories, respectively, were accessible. Some 56 percent of rural and 49 percent of urban persons spent less than this, and so were poor.
Then the Commission, for reasons unknown, changed the definition in practice, and it never again directly looked at the total monthly spending on all goods and services which permitted nutrition norms to be accessed. This was despite the fact that every five years the required information on this for every spending level was available – the physical quantities of food intake, and the corresponding daily average intake of energy, protein and fat. The definition the Commission actually adopted was that the 1973-4 poverty lines were to be adjusted for inflation using a price-index, regardless of whether the lines so obtained still allowed nutritional standards to be met. This amounted to taking a fixed 1973-4 basket and merely adjusting its value for price change.
The 1993 Expert Committee also opted to continue with this mistaken method. Price index adjustment to the fixed basket has been followed for nearly 40 years since 1973-4, including after the minor modification by the Tendulkar committee. It produced the absurdity of 22.4 and 28.7 Rupees as the rural/urban daily poverty lines by 2009-10. Indexing for two more years raises it to Rs. 26/32 per day for 2011-12. But Rs.32 today will not buy even a single kilogram of sugar on the urban open market.
Why these economists should have such faith in the ability of price indices to capture the long-term rise in the cost of living is not clear. Price indices are useful for short period dearness allowance calculation, but they do not reflect the actual rise in the cost of living over longer periods of time. This fact is of great importance for all labour unions and occupational associations.
As an example from personal experience, the starting gross monthly salary of an Associate Professor in a Central University – those established by Acts of the Indian Parliament — in 1973 was Rs. 1,000. This was quite adequate, since ration cards could be used; on this income one even ran a car. Applying the Consumer Price Index for Urban Non-Manual Employees, which had risen seventeen-fold by 2011, the equivalent monthly salary for an Associate Professor joining now should be Rs.17,000 (US Dollars 340) on the Planning Commission’s logic. But this would not support the most modest professional life-style of four decades earlier. The newly appointed Associate Professor’s actual salary today is three times higher, thanks to the Pay Commission, which every ten years has hiked salary grades to maintain living standards of government servants and employees in publicly funded institutions.
Yet, denying all experience and evidence, these economists assert that mere price-index adjustment is enough to obtain current poverty lines from those of 40 years ago. No wonder they have created such a mess with their unrealistic estimates. By 2005, a rural person needed Rs.19 per day total spending to access 2200 calories; at the official Rs.12 per day poverty line, she could obtain only 1820 calories. (The Tendulkar Committee merely tinkered with the problem, raising the Rs.12 marginally to Rs.13.8). An urban consumer needed Rs.33 per day in 2005 to meet 2100 calories, whereas the official Rs.18 permitted 1790 calories. At the 2009-10 official poverty lines of Rs.22.4/28.6 rural/urban, the minimal cost of living is even more seriously understated.
Yet, in March 2012, the Indian Planning Commission once more claimed a decline in the percentage of those living in poverty. It said in a press release:
The all-India HCR (head-count ratio) has declined by 7.3 percentage points from 37.2% in 2004-05 to 29.8% in 2009-10, with rural poverty declining by 8.0 percentage points from 41.8% to 33.8% and urban poverty declining by 4.8 percentage points from 25.7% to 20.9%.
This claim is as false as all previous claims by this and earlier Commissions on poverty decline, because the standard has been continuously lowered over time. A person living at the official 2009-10 poverty lines would have been able to consume 1880 calories in rural areas and 1720 calories in urban areas. Compare that to the 1993-94 figure of 1980/1885 calories, which already had fallen well below the 2200/2100 norms for minimal caloric sufficiency, satisfied by official poverty lines only in 1973-4.
State poverty lines vary in India, and in a number of states, the daily energy intake the unrealistic official poverty line can command has fallen below 1500 calories. For example, the official poverty line for urban Delhi state for 2009-10 is Rs.1,040 monthly, or 34.7 per day. That trivial sum permitted only 1400 calories daily energy intake while the observed spending level on all goods and services satisfying the nutrition norm was five times the official one.
We cannot accept a claim by a school of improved academic performance citing that the percentage of failed students has declined over time, if we discover that the pass mark has been lowered substantially, and at the original pass mark more students have actually failed. For any valid comparison, the standard has to be kept unchanged. Actual poverty, measured by keeping the standard unchanged, far from declining, has risen, with 76/73 percent of the rural/urban population unable to reach the modest nutrition standard of 2200/2100 calories by 2009-10, compared to 68.5/58.5 percent five years earlier.
Not only energy, but also protein intake has been falling for all except the top population decile and so has cloth consumption. The rise in poverty is not surprising, given the twin impact of global recession from 2008 and severe drought in 2009-10. Per capita annual food grains consumption for all uses (including feed for producing animal products, processing and so on) at 174 kg. in India by 2008, touched the lowest level in the world, lower than the average for Least Developed Countries. People are being forced to cut back on food and other necessities, including the poorest deciles of consumers, not only owing to high food price inflation when their money incomes are not rising in tandem, but also owing to higher cost of health care and utilities as these are privatized.
China’s official poverty lines are equally absurd, and for the same reason as in India. A nutrition norm was applied in 1984 to obtain a correct 200 yuan annual rural poverty line, which thereafter was merely indexed, giving 1274 yuan by 2011, or 3.5 yuan per day. This is supposed to cover all living costs but would not have bought even a single kilogram of the cheapest rice (information provided by China residents). Actual poverty in China is far higher than claimed.
One wonders if we will ever see honest estimates from official sources anywhere, since by now hundreds of economists are closely imbricated within a vast global poverty-estimating structure with the World Bank at its apex, producing increasingly misleading estimates every year in its glossy Reports. The World Bank’s global poverty line is an equally large underestimate, for it is derived using “purchasing power parity conversion” from local currencies to US dollars, of these very same absurdly low local-currency official rural poverty lines of developing countries, including India and China. The World Bank claim of poverty decline is equally unsound.
What are the realistic poverty lines in India today applying the Planning Commission’s own original definition and method, namely directly observing what level of current spending satisfies the modest nutritional norms? From the 66th Round, 2009-10 nutrition data of the National Sample Survey, the actual spending levels on all goods and services allowing per head daily norms of 2200/2100 calories rural/urban to be met in that year, were Rs.1,090 per month (Rs.36.3/day) and Rs. 2100 per month (Rs. 70/day). The percentage of persons falling below these poverty lines was 76 in rural and 73 in urban India. This high level of deprivation is the rationale for going back to a non-targeted, universal food distribution system, but this will not be enough. The purchasing power of the poor has to be raised at the same time through employment generation schemes. The proposed draft National Food Security Act, if it stays with the 46 percent figure of total population to be covered on ‘priority basis’ will be arbitrarily excluding from its ambit 350 million poor persons.
Note: An earlier version of this article was published in The Hindu in 2011. This revised version incorporates the Planning Commission’s March 2012 estimates.