Until the Ministry of Statistics and Programme Implementation released the Periodic Labour Force Survey report in late June 2019, the dominant narrative on employment from the establishment was one of denial that jobs were being lost or that there was stress in the labour markets. After the release of the PLFS such summary dismissal of the problem has declined.
This is because the PLFS itself has shown that the unemployment rate in the country in 2017-18 was at a 45-year high. There was only a brief attempt to insist that the PLFS was not comparable to earlier surveys. But, that did not stick and gratefully the argument was not pursued aggressively.
In October 2019, two papers published by researchers compare the results of PLFS with the earlier NSSO surveys. One of them, "Emerging Employment Patterns of 21st Century India - I" by Laveesh Bhandari and Amaresh Dubey is at the invitation of the Economic Advisory Council to the Prime Minister. The second "India's Employment Crisis: Rising Education Levels and Falling Non-Agricultural Job Growth" by Santosh Mehrotra and Jajati Parida is published by the Centre for Sustainable Employment at Azim Premji University.
Both papers have no hesitation in comparing the PLFS with the EUS. The former does not fuss over comparability and the latter asserts that the two are comparable.
The paper by Bhandari and Dubey for the EAC-PM shows an increase in employment between 2011-12 and 2017-18 -- from 433 million to 457 million. However, the paper by Mehrotra and Parida shows a fall in employment from 474 million to 465 million in the same period.
The difference is not small. There is a difference in both magnitude of employment and in the direction of change. First, employment estimates of Mehrotra and Parida are systematically higher than those of Bhandari and Dubey. Secondly, Mehrotra and Parida imply a 9 million decline in jobs while Bhandari and Dubey suggest an increase of nearly 24 million jobs over the same six year period.
Mehrotra and Parida have explained this difference in a separate note. They explain that while their estimates are based on the usual principal and subsidiary status of employment (UPSS), those of Bhandari and Dubey are based on only the usual principal status (UPS). Researchers who use the NSSO surveys usually use the UPSS to estimate aggregate employment in India. Laveesh has explained to this author that it is principal status that is important and subsidiary status is mostly unpaid family labour and disguised employment or under-employment. However, unpaid family labour is not small. Jajati tells me that 22 million jobs were lost here and another 26 million were lost in casual labour.
Using both estimates we could infer that while UPS employment has increased smartly (by 24 million) as demonstrated by Bhandari and Dubey, subsidiary employment has declined so sharply that it has caused a decline in overall employment as suggested by Mehrotra and Parida.
But, Mehrotra and Parida point out a second source of the difference between the two papers. This is the projected population in 2017-18 to estimate the absolute numbers of employment from the ratios provided by the NSSO. Bhandari and Dubey have used a slightly higher rate of growth in population than the projections made by Mehrotra and Parida.
Nevertheless, it seems to be clear that while employment as defined by principal status has increased, there was a massive fall in employment as defined by the subsidiary status between 2011-12 and 2017-18.
I may conjecture here that employment in the unorganised sectors also declined and this is where all the unpaid family workers were engaged.
GST came into force in July 2017 and brought grief to small and micro industries. Employment in these units is predominantly informal and these are also the ones that provide the subsidiary employment of the non-agricultural sectors.
The PLFS survey was conducted from July 2017 through June 2018, a period of substantial distress in the unorganised sectors. It is also the period when the organised sector gained market share from the unorganised sector that was cracking up. Excluding subsidiary status employment during this period would reveal this increase in market share of the organised sector but it would not tell us the rest of the story.
Here is one evidence of the fall in employment in the unorganised sectors. Mehrotra and Parida's paper tells us that employment in manufacturing declined from 59.8 million in 2011-12 to 56.4 million in 2017-18. Separately, Annual Survey of Industries tells us that employment in organised manufacturing increased from 13.4 million to 15.6 million in the same period. We can therefore deduce that the employment in unorganised manufacturing declined from 46.4 million to 40.8 million -- i.e. a fall of 12 per cent.
Neither of the two sets of researchers have spoken their last word on the subject and, more will dissect the data. Hopefully, these will not paralyse us with too many contradictory views but will thaw the establishment out of its paralysis by denial.