Data and reality clash in poverty numbers

Pradhan Mantri Garib Kalyan Yojna (PMGKY) was among the biggest relief measures which the Narendra Modi government deployed during the pandemic

Pradhan Mantri Garib Kalyan Yojna (PMGKY) was among the biggest relief measures which the Narendra Modi government deployed during the pandemic. The most important part of this scheme was the government providing an additional 5kg of food grain to more than 800 million people -- free of cost. The scheme cost ₹3.9 lakh crore and involved distribution of 12.2 million tonnes of food grain. Both the government and independent experts believe that the expansion of food safety net was crucial in mitigating large-scale distress during the pandemic. However, if the government’s own statistics are to be believed, the scheme was not needed at all. To be sure, these numbers are more likely an indication of bad data rather than the government being wrong about PMGKY’s need. This mismatch between data and reality has important implications for the ongoing debate on India’s poverty numbers. Here are three charts which explain this in detail.

HT Image

What happened to food consumption during the pandemic?

Data from National Account Statistics (NAS) shows that food consumption – it is part of private final consumption expenditure (PFCE) – increased in real terms between 2019-20 and 2020-21. This holds not just for overall food consumption but almost all disaggregated items as well. One can argue that the increase in cereal consumption was supported by the heavily subsidized Public Distribution System (PDS) and the free provision of extra grains under PMGKY. But what explains the rise in consumption of non-cereal food items?

If one takes these numbers on face value, India’s policy makers had something comparable to the opposite of the Marie Antoinette view on the state of the masses. They believed that the people could not even eat bread (cereals) whereas the people were comfortably consuming cake (expensive non-cereal food items), as is shown in NAS data. The PMGKY money could easily have been spent more effectively on other things.

See Chart 1: YoY growth (in %) in food consumption between 2019-20 and 2020-21

Is there another explanation to rise in non-cereal food consumption during the pandemic?

To be sure, there is another logical explanation to this. If the relatively well-off sections – 800 million PMGKY beneficiaries are around two-thirds of India’s population – have a disproportionately high share in non-cereal food consumption, it is possible that even a sharp fall in incomes and hence purchasing power of PMGKY beneficiaries would not have brought down the consumption of non-cereal food items. An HT analysis of 2011-12 Consumption Expenditure Survey (CES) data shows that while the top 40% indeed have a higher share in non-cereal food consumption, the share of bottom 60% is anything but insignificant. This means that even if the consumption of the top 40% did increase, the fall from bottom 60% should have brought the overall number down.

See Chart 2: Quintile-wise share in cereal and non-cereal food consumption

How is this relevant to The Great Indian Poverty Debate?

India’s poverty numbers are based on CES data collected by the National Sample Survey Office (NSSO). An important criticism of these numbers draws its legitimacy from the fact that consumption spending from CES numbers have been showing an increasing divergence with the NAS numbers over time, with the latter increasingly becoming higher. Surjit Bhalla, who co-authored a working paper on new set of poverty estimates last year is among the most important voices championing this criticism.

See Chart 3: Private consumption in NSS as share of NAS estimates

What does it mean?

If the NAS numbers are indeed a better indicator of consumption, then one will have to accept that the money spent on a scheme such as PMGKY was a complete waste. If one accepts that the government had better intelligence about the situation on the ground and not launching a scheme like PMGKY could have led to widespread deterioration in food security, then one cannot but agree that the NAS numbers on food consumption are gross overestimates.

To be sure, some experts believe that the controversy is unavoidable. “Private consumption as measured in the NSS surveys is very different from private consumption estimated in NAS data”, Himanshu, an associate professor of economics at Jawaharlal Nehru University, told HT. “NAS data captures consumption even by non-corporate entities into private consumption, so it is bound to be higher than the numbers estimated by the NSS and unlike every new round of NSS survey which generates new insights into consumption patterns, NAS numbers interpolate numbers such as PFCE by using assumed ratios of various constituents of GDP. This is exactly why we need better and regular NSS survey on consumption to comment on the state of consumption inequality and poverty in India”, he added.