People's Democracy

(Weekly Organ of the Communist Party of India (Marxist)


No. 32

August 11 , 2013

Incorrect Method of Poverty Estimation Gives False Results

Utsa Patnaik

The Planning Commission�s incorrect estimation method shows spurious poverty decline because the standard for measuring poverty is lowered continuously. In reality poverty is high and rising.

IT is a notable paradox that poverty estimation is lucrative business, employing hundreds of economists in developing countries and in international institutions � the latter in particular are paid truly princely salaries. If all this public money spent, resulted in a correct understanding of how poverty is changing, it might be worthwhile. Unfortunately, the completely wrong method universally used by economists whether they are in the World Bank or in governments and institutions, means that their poverty estimates and claims of poverty decline are false and not worth the paper on which they are printed. Let us start by looking at our own official poverty estimates in India.

The Planning Commission has once again embarrassed us with its recent claims of decline in poverty by 2011-12 to grossly unrealistic levels of 13.7 percent of population in urban and 25.7 percent in rural areas, using monthly poverty lines of Rs 1000 and Rs 816 respectively, or Rs 33.3 and Rs 27.2 per day. These amounts will pay for one urban male haircut in a lower middle class barber shop, while they are supposed to meet all daily food and non-food living costs. The poverty decline claimed is huge, a full 8 percent points fall in rural areas over the mere two years since the 2009-10 official estimate of 33.8 percent, and a 7 percent points fall in urban areas from 20.9 percent , even though these two years saw the after- effects of drought, low employment growth and very rapid food price rise. The incorrect estimation method that the Commission continues to use makes it an absolute certainty that in another few years when the 2014-15 survey results become available, it will claim that urban poverty is near zero and rural poverty is only around 12 percent. This will be the case regardless of any rise in actual deprivation and intensification of actual poverty.

All official claims of low poverty level and poverty decline are spurious, solely the result of mistaken method, and in reality poverty is high and rising. By 2009-10 after meeting all essential non-food expenses (manufactured necessities, utilities, rent, transport, health, education) 75.5 percent of rural persons could not consume enough food to give 2200 calories per day, while similarly 73 percent of all urban persons could not access 2100 calories per day. The comparable percentages for 2004-5 were 69.5 rural and 64.5 urban, so there has been substantial poverty rise. Once the NSS releases its nutritional intake data for 2011-12 we can see the change up to that year, but given the high rate of inflation and sluggish job growth, the situation is likely to be as bad, if not worse. Our figures are obtained by applying the Planning Commission�s own original definition of poverty line. Against a background of rapidly rising cost of privatised health care, education and utilities (electricity, petrol, gas) combined with high food price inflation and exclusion of the majority of the actually poor from affordable PDS grain, it is hardly surprising that the bulk of the population is getting more impoverished, and its nutritional level is declining faster than before.


What is the basic problem with the Planning Commission�s method which produces its low and necessarily declining estimates, regardless of ground reality? The Commission in practice gave up its own definition of the poverty line which was applied only once, for getting the 1973-74 estimate. After that, it has never looked over the next 40 years even once for deriving poverty lines, at the actual current spending level which will allow the population to maintain the same standard of living in terms of nutrition after meeting all non-food costs � even though these data have been available in every five-yearly NSS survey. The Commission instead simply applied price indices to bring forward the base year monthly poverty lines of Rs 49 rural and Rs 56 urban in 1973-74. The Tendulkar committee did not change this aspect, it merely altered the specific index.

Price indexation does not capture the actual rise in the cost of living over long periods and those doing the poverty estimates would be the first to protest if their own salaries were only indexed through dearness allowance. A well-paid government employee getting Rs 1,000 a month in 1973-74 would get Rs 18,000 a month today if the salary was only indexed since the Consumer Price Index for Urban non-Manual Employees has risen eighteen-fold. (This index has been recently discontinued � it gave almost exactly the same price rise as the Consumer Price Index for Industrial Workers. The reader does not have to bother looking up indices. Simply dividing the official urban poverty line for 2010-11, Rs 1,000 by the base year poverty line of Rs 56 will give us the rise since 1973-74 in the price index officially used, namely a 17.8 fold rise. Similarly for rural India it is Rs 816 divided by Rs 49 or a 16.7 fold rise). The fact that indexing for rise in prices and paying dearness allowance � which can come to exceed the basic pay � still does not capture the actual rise in the cost of living, is recognised by the government itself by appointing Pay Commissions every decade which push up the entire structure of salaries. An employee in the same position today gets not Rs 18,000, but a four times higher salary of over Rs 70,000. Yet those doing poverty estimates continue to maintain the fiction that the same standard of living can be accessed by the poor by merely indexing the original poverty line, and they never mention the severely lowered nutritional access at their poverty lines, which by now are more properly destitution lines.

The fact is that official poverty lines give command over time to a lower and lower standard of living. With a steadily lowered standard, the poverty figures will always show apparent improvement even when actual deprivation is worsening. A school child knows that if last year�s percentage of students passing the annual examination is to be compared to this year�s percentage, then the pass mark has to be the same. The school principal cannot quietly lower the pass mark without informing the public, say from 50 out of hundred last year to 40 this year, and then claim that school performance has improved because 80 percent of students are recorded as �passing� this year at the clandestinely lowered pass mark, compared to 75 percent of students passing last year. If at the same pass mark of 50, we find that 70 percent of students have passed this year, we are justified in saying that performance far from improving has worsened. If the school is allowed to continue with its wrong method, and lower the pass mark further next year, and again in the next year, so ad infinitum, then it is eventually bound to record 100 percent pass and zero failures.

The case is exactly the same with the official poverty lines as with the pass mark: the poverty lines are being lowered continuously below the standard over a very long period of forty years. �Poverty� so measured, is bound to disappear from India even though in reality it may be very high and worsening over time. The Commission�s monthly poverty line for urban Delhi state in 2009-10 is Rs 1040 � but a consumer spending this much could afford food giving only 1400 calories per day after meeting all other fast rising expenses (see the table). The correct poverty line is Rs 5,000 for accessing 2100 calories, and a staggering 90 percent of persons have been pushed below this compared to 57 percent below the correct poverty line of Rs 1150 in 2004-5 at which they could access 2100 calories after all non-food expenses. Given the very high rate of food price inflation plus rising cost of privatised medical care and utilities, it is not surprising that people are being forced to cut back on food, since the other items are not under their control. The average calorie intake in urban Delhi has fallen to an all-time low of 1756 compared to 2072 in 2004-05, while protein intake has fallen to 54 gm from 61.3 gm five years earlier. While a high-visibility minority of households with stable incomes is able to hire-purchase multiple cars per household, enjoy other durable goods, eat out and take holidays, the vast working underclass which is invisible to the rich, is struggling to survive.


Fifty-five percent of the urban population in Delhi could not access even 1800 calories by 2009-10, compared to less than a quarter in that position a mere five years earlier. The relevant NSS data for urban Delhi are given in the Table. The Reports for 2009-10 divide the consumers into ten spending groups in terms of monthly per capita expenditure on all goods and services, ranging from the poorest to the richest. The spending limits of each class have been chosen in these reports in such a way that each group has one-tenth of all consumers. The first two columns show the upper spending limit of each group and the cumulative total percentage of persons falling below the spending levels. The third column shows the average spending per head on all goods and services in each group, and the fourth and last show the average energy intake in calories and the protein intake, from the quantities of foods obtained by the spending on food. Spending refers not only to market purchases but also includes the value of food items produced and consumed directly as by farmers, or obtained as wages in kind.



NSS REPORTS 538 & 540


Percent of




value of







per capita









the value
























































Why, it may be asked, do the highly trained economists in the Commission ignore reality and continue with their incorrect method. Surely they can see as we do, that their Rs 1040 poverty line gives access to a bare-survival 1400 calories. Part of the answer is that the ramifications of using the wrong method extend globally, for the World Bank economists have for decades, based their poverty estimates on the local currency official rural poverty lines of developing countries including India and China, while these poverty lines are falling more and more below the actual cost of living. In China, the same universally used wrong method was followed once market oriented reforms started in the 1980s. A nutrition based poverty line was correctly applied only once in the base year 1984 and then merely price indexed, giving by 2011 an absurd poverty line of 3.5 yuan daily which could not buy even one kilogram of the cheapest rice.

The World Bank claim of poverty decline in Asia is as spurious as the individual country claims, since the Bank uses an average of the local currency rural poverty lines of the poorest countries, which comes to around half a dollar per day at the prevalent exchange rates. This figure is adjusted upwards for purchasing power by a factor lying between two to three (close to 2.5), to arrive at its global $1.25 daily poverty line. But these local currency poverty lines themselves as we have seen, represent a lower and lower standard of living over time. The Bank�s $1.25 daily global poverty line applied to an individual country like India at the present exchange rate of Rs 60 to a dollar, is Rs 75 but this would now be adjusted the other way, downwards to take account of purchasing power, by the factor 0.4 namely two-fifths or Rs 30 would be taken as the poverty line for 2013. The World Bank applying this line necessarily produces a poverty percentage for India which is very close to the official one, and claims poverty decline.

In reality under the regime of poor employment growth and high food price inflation, poverty in Asia has been rising. To admit this would mean that the entire imposing looking global poverty estimation structure employing hundreds of economists busy churning out wrong figures, would come crashing down like a rotten termite-eaten house. The rest of the answer for continuing to follow a patently incorrect method, is that since it automatically produces numbers showing spurious poverty decline, this has proved to be very convenient for arguing that globalisation and neo-liberal policies are beneficial for people. Truth will always be out, however, for the people at least to see, which is what matter.