Counterview DeskIn an interview to Counterview, well-known economist Prof Amitabh Kundu* seeks to look at the reasons behind the controversy surrounding the working papers of the World Bank and the International Monetary Fund (IMF) which report low or no poverty for India in the years of the pandemic or before that.
He says, “Unfortunately, no attempt is made to determine any poverty basket and check if the poverty line is adequate to buy that. Meeting the basic human needs, thus, is not the basic concern in the current poverty debate as it is argued that such needs cannot be decided in any robust manner.”
Text:
Q: How do you look at the low figures of poverty in India in the context of deprivation figures for Indian population in different spheres?A: India witnessed lakhs of migrants walking thousands of kilometres with children in their hands from India’s urban centre to their homes in rural areas during the first year of the pandemic. Nor is it a secret that thousands died on the streets and around the hospitals due to lack of beds and oxygen. Yet there have been claims, particularly by international organisations, that the level of poverty in India has considerably gone down, or it is almost negligible. Clearly, there is an anomaly.
The problems faced by common people, particularly those at the bottom rung of the society, are those of management and governance and cannot be attributed to poverty, it is being argued. But what about the National Family Health Survey recording the infant mortality rate (IMR) to have gone down only marginally from 40.7 to 35.2 per thousand between the two rounds covering 2015-16 and 2019-21? Or, the percentage of anaemic children aged 6 to 59 months, pregnant and non-pregnant women and adult males have gone up significantly during the period?
These indicators, we are told, are no proxies for poverty. Their linkages with nutritional indicators are considered tenuous, as these can be explained in terms of intra-household distribution, poor dietary habits, improper water/sanitation facilities etc. Not without reason, there is uproar about the working papers of the World Bank and IMF that report low or no poverty for India in the years of the pandemic or before that.
The real question is what does the poverty index measure or attempt to capture? The construction of the index involves complex calculations to identify a poverty basket of consumption, working out price indices for updation of poverty line, and then applying these to the income or consumption of households for determining the poverty status. Unfortunately, no attempt is made now to determine any poverty basket and check if the povery line is adequate to buy that. Meeting the basic human needs, thus, is not the basic concern in the current poverty debate as it is argued that such needs cannot be decided in any robust manner.
Q: How has the delinking of poverty from nutrition has taken place in Indian poverty literature?
A: A nine member Working Group set up by the Planning Commission proposed the poverty line at Rs 20 per capita per month in early 1960s, loosely ensuring adequacy of minimum requirements. VM Dandekar and Nilakantha Rath (1970) went into details of minimum calorie needs, based on the average consumption pattern. During 1980s and 1990s, it was realised that this linkage is getting blurred due to changes in the consumption pattern, changing micro environment for living, withdrawal of the state in provision of basic necessities etc.
PV Sukhatme argued that the emphasis on calorie and nutrition is misplaced as the absorption of nutrients in the body depends on physical health, particularly presence or absence of gastrointestinal diseases. Water and sanitation facilities were noted as important in determining the poverty line in the absence of which, out of pocket expenditure of the household on these and health would shoot up.
Indeed, engagement of the state in providing the basic necessities free or at subsidised rates would lower the household expenditure required to cross over the poverty line. BS Minhas’ exhaustive work too suggested that calorie/nutrient requirements are largely socially and culturally determined and even the biological needs vary from person to person within a household.
It was accepted that the state through poverty intervention cannot and should not try to guarantee adequate nutrition or health to people. The Tendulkar Committee formally announced delinking of nutritional norms from poverty in 2009. Subsequently, efforts to bring dietary factors into poverty domain by the Rangarajan Committee received lukewarm response from the policy makers.
As a young Assistant Professor at the Jawaharlal Nehru University (JNU), I have observed the transition taking place. I recall Dr DT Lakdawala beseeching Dr B. S. Minhas at a seminar “not to remove the pegs on with the whole poverty debate hangs.” He thought that poverty has a special punch and force in policy discourse at national and global level, because people consider it to be linked with human deprivation, suffering, and absence of basic necessities. If it is reduced to a number which is temporally and cross sectionally comparable but lacks in content of suffering, it will cease to be an instrument in the hands of researchers and civil society to pressurise the government for appropriate welfare policies.
Q: How has the Working Paper, prepared by the IMF computed poverty figures for the recent years when there has been no Consumption Expenditure Survey whose results have been accepted by the government?
A: The results of the NSS consumption expenditure for 2017-18 have been rejected by the government as being not comparable with those of previous rounds. The computation of poverty has thus become extremely challenging in the absence of the data on consumption expenditure, as is the case in India today, as also in several developing countries. With a view to providing inputs into policy making, researchers have evolved ingenious methods of estimating the data, using the past datasets and those where the samples are not representative of the households in the country.
One must compliment Surjit Bhalla, Karan Bhasin and Arvind Virmani who have taken up this challenge in their IMF Working Paper of developing a method of interpolation and extrapolation of the consumption expenditure of the NSS survey of 2011-12 and build a series up to 2019-20. They come out with definite numbers. It would only be fair to consider these seriously to examine the extent to which these maintain temporal and cross section comparability and, more importantly, what exactly do they capture or can convey to policy makers.
They use the growth rate of the private final consumption expenditure (PFCE), but bring in the distributional changes by allowing the households consumption to grow as per the growth in nominal per capita income in each state.
Computation of poverty has become extremely challenging in the absence of data on consumption expenditure
Rural urban price differences are also introduced through separate poverty lines for rural and urban households. The method is reasonable -- except that it assumes the distributions to remain unchanged, both within the rural and urban segments in each state over the period 2014-20.
Objections have been raised regarding updating household expenditure by the growth in PFCE, as it is estimated as a residual in national income accounts and therefore includes many other components such as expenditure of unincorporated enterprises. Also, with regard to the extent to which government hospitals and non-profit organisations transfer benefits to households, there will be double counting.
It may, on the other hand, be noted that the expenditure data collected from households do not include rental payments. However, considering the pluses and minuses, updating consumption expenditure by PFCE enhances the estimated value. One wonders if there could be other multipliers for updating the figures, free from these problems .
Q: What could have been done to bring in distributional changes realistically in rural and urban areas within the states in the IMF Working Paper?
A: The first suggestion would be to multiply the figure of consumption expenditure of each household by the growth rate of income in the economic activity with which the household is associated, since the authors consider using income growth as a possible basis for projection. Secondly, the growth rates of different commodities in the PFCE are significantly different and hence adjustments can be done commodity wise, to bring focus on items of consumption by the poor, across the states.
Q: What do you think of bringing in the subsidised and free food grains given to poor households, in the poverty calculations in the IMF paper?
A: The most significant contribution of the study is its bringing in differential engagement of the state in provisioning of the essentials to the poor into poverty calculations. The public distribution system (PDS) off take of the poor has doubled after implementation of the National Food Security Act after 2014. The provisioning of extra food grains per person per month free of cost since March 2020 has also been built into the real expenditure of the poor.
This, however, opens up the possibility of bringing in the changes in the level of state engagement in provisioning of essentials, including free gas cylinders etc. within the framework of calculations. However, this must be supplemented by an assessment of the disengagement of the state in crucial social sectors such as education and health. Incremental expenditure of the poor on health and education in recent years, as estimated through NSS data, would have to be considered by revising up the poverty line.
Q: What is your opinion regarding the methodology and poverty calculations in the Working Paper of the World Bank?
A: The paper by Sutirtha Sinha Roy and Roy van der Weide for the World Bank, released recently. explores the possibility of using Centre for Monitoring Indian Economy (CMIE) data in poverty calculations after correcting for the unrepresentative character of the panel data, by modifying the weightages of households for aggregation. Indeed, the asset position and level of basic amenities are generally higher in the CMIE sample than in more robust national sources, basically due the former’s sampling procedure.
These adjustments carried out to remove the non-convergence of the CMIE data with others at macro level have resulted in a poverty figure of 12 per cent. This, many people find more acceptable not because of the methodology but the magnitude. One does not know whether the poverty estimate would be a bit higher had the adjustments been carried out for a few other parameters and also at the state level.
Q: Any concluding observation?
A: The efforts of all the scholars in building up appropriate data sets, given that the decade 2011-21 has been disastrous for macro data, may be commendable. But let it be clear, there is no royal road to statistics. Let these exercises be not similar to the search of the proverbial smart man for his wallet under a lamppost. Being asked, he answered that he dropped it somewhere near the dark gutter but insisted on not searching there since there was no light.
---
*Senior Fellow with the Sustainable Cities and Transport program at WRI India; former professor, Jawaharlal Nehru University, New Delhi
Objections have been raised regarding updating household expenditure by the growth in PFCE, as it is estimated as a residual in national income accounts and therefore includes many other components such as expenditure of unincorporated enterprises. Also, with regard to the extent to which government hospitals and non-profit organisations transfer benefits to households, there will be double counting.
It may, on the other hand, be noted that the expenditure data collected from households do not include rental payments. However, considering the pluses and minuses, updating consumption expenditure by PFCE enhances the estimated value. One wonders if there could be other multipliers for updating the figures, free from these problems .
Q: What could have been done to bring in distributional changes realistically in rural and urban areas within the states in the IMF Working Paper?
A: The first suggestion would be to multiply the figure of consumption expenditure of each household by the growth rate of income in the economic activity with which the household is associated, since the authors consider using income growth as a possible basis for projection. Secondly, the growth rates of different commodities in the PFCE are significantly different and hence adjustments can be done commodity wise, to bring focus on items of consumption by the poor, across the states.
Q: What do you think of bringing in the subsidised and free food grains given to poor households, in the poverty calculations in the IMF paper?
A: The most significant contribution of the study is its bringing in differential engagement of the state in provisioning of the essentials to the poor into poverty calculations. The public distribution system (PDS) off take of the poor has doubled after implementation of the National Food Security Act after 2014. The provisioning of extra food grains per person per month free of cost since March 2020 has also been built into the real expenditure of the poor.
This, however, opens up the possibility of bringing in the changes in the level of state engagement in provisioning of essentials, including free gas cylinders etc. within the framework of calculations. However, this must be supplemented by an assessment of the disengagement of the state in crucial social sectors such as education and health. Incremental expenditure of the poor on health and education in recent years, as estimated through NSS data, would have to be considered by revising up the poverty line.
Q: What is your opinion regarding the methodology and poverty calculations in the Working Paper of the World Bank?
A: The paper by Sutirtha Sinha Roy and Roy van der Weide for the World Bank, released recently. explores the possibility of using Centre for Monitoring Indian Economy (CMIE) data in poverty calculations after correcting for the unrepresentative character of the panel data, by modifying the weightages of households for aggregation. Indeed, the asset position and level of basic amenities are generally higher in the CMIE sample than in more robust national sources, basically due the former’s sampling procedure.
These adjustments carried out to remove the non-convergence of the CMIE data with others at macro level have resulted in a poverty figure of 12 per cent. This, many people find more acceptable not because of the methodology but the magnitude. One does not know whether the poverty estimate would be a bit higher had the adjustments been carried out for a few other parameters and also at the state level.
Q: Any concluding observation?
A: The efforts of all the scholars in building up appropriate data sets, given that the decade 2011-21 has been disastrous for macro data, may be commendable. But let it be clear, there is no royal road to statistics. Let these exercises be not similar to the search of the proverbial smart man for his wallet under a lamppost. Being asked, he answered that he dropped it somewhere near the dark gutter but insisted on not searching there since there was no light.
---
*Senior Fellow with the Sustainable Cities and Transport program at WRI India; former professor, Jawaharlal Nehru University, New Delhi
Comments