Informal non-farm survey: A lost opportunity

The “Annual Survey of Unincorporated Sector Enterprises (ASUSE)”, released by the National Sample Survey Office (NSSO) on June 14, 2024, best exemplifies the shortcomings with official statistics. It reveals partial data provides no context or perspective and what it doesn’t reveal are actually relevant for its stated purposes than what it does reveal.

First a few basic facts about this report:

(a) It comprises of two annual reports, ASUSE 2021-22 (survey period of April 2021-March 2022) and ASUSE 2022-23 (survey period of October 2022-September 2023) and cover informal non-farm sectors in manufacturing, trade and other services (excluding construction).

(b) The stated purposes of these surveys are: (i) “to gauge the demand-side employment scenario” in the “unincorporated non-agricultural sector” to (ii) “help in meeting the requirements” of different ministries dealing with MSMEs, textile, labour and employment etc. (iii) with data “essential” for policymakers, researchers and other stakeholders to “better understand and support this crucial segment of the economy”.

Making sense of annual ASUSE

The two surveys reveal the following:

(i) In 2021-22, there were 59.7 million establishments – which increased to 65 million in 2022-23. That is, an overall growth of 9%; manufacturing establishments grew by mere 2.2%, while that of “other services” grew by high 15%.

(ii) In 2021-22, 97.9 million workers were employed in these establishments – which grew to 109.6 million in 2022-23 (growth of 12%).

(iii) Family members comprised of 86% and 85% workers in 2021-22 and 2022-23, respectively. The rest were hired hands.

(iv) Of the hired hands, 88% were informal workers with no social security cover in 2021-22 – which went up to 92.9% in 2022-23.

(v) Wages to hired workers fell from ₹1,26,243 in 2021-22 to ₹124,842 in 2022-23.

(vi) GVA share of these establishments fell from 12.4% in FY21 to 6.3% in FY23.

Such data makes little sense without context and perspective. For example, FY21 is not a normal year as its first quarter was hit by the second wave of the pandemic – leading to shutdown of establishments and job losses. Comparing it with 2022-23 (October 2022-September 2023) gives an exaggerated picture of growth because a large part of establishments and workers would have simply resumed business after the pandemic impact ended.

To make better sense (perspective) of the above numbers one would have to look at the ASUSE of 2015-16 – called “Key Indicators of Unincorporated Non-Agricultural Enterprises (Excluding Construction) in India, 2015-16”.

The current report doesn’t mention it or provide a comparative picture despite covering the very same informal non-farm sectors with no discernible change in the methodology that could have made such comparison inappropriate. But the reasons would become clear soon.

The 2015-16 survey (period of July 2015-June 2016) shows, the number of informal non-farm establishments were 63 million, employing 111 million workers and contributing 9.2% to the GVA.

Now a comparison with the ASUSE of 2022-23 (over that of 2015-16) would mean (a) rise in the number of establishments by 2 million (b) fall of jobs by 1 million and (c) fall of GVA share by 2.9 percentage points.

By omitting the 2015-16 survey, the ASUSE escapes the need to explain the job loss and fall in GVA share. It is, however, no secret that informal economy was hit hard by three big shocks after the 2015-16 survey (July 2015-June 2016): Demonetisation of November 2016, GST of July 2017 and the pandemic of 2020.

The ASUSE of 2021-22 and 2022-23 also failed in many other ways.

These surveys could have mapped the very establishments covered during the 2015-16 survey to know the impact of the three shocks (demonetisation, GST and pandemic) – which would have helped in working out the precise fiscal and policy support needed to revive or boost productivity and job creations in manufacturing, trade and services other than construction. These measures could then have come into the full budget for FY25 to be presented next month (July 2024).

That is because the 2015-16 survey was elaborate and presented the following elements which are missing from the ASUSE of 2021-22 and 2022-23:

· Distribution of establishments and workers – sector-wise, ownership-wise, state-wise and gender-wise.

· Availability of skilled manpower, waste management and toilets in each establishment.

· Original investments in plant and machinery, enabling classification of those establishments as MSMEs and others.

The 2015-16 survey report ran into 153 pages. In comparison, the current ASUSE report (comprising of two annual one surveys) is skeletal – the factsheet runs into 13 pages and the press note into 6 pages.

Now, look back at the stated purposes of the ASUSE mentioned earlier (gauge the demand-side employment scenario; meet requirements of MSMEs, textile, labour and employment ministries with data “essential” for policymakers, researchers and other stakeholders to understand and support informal non-farm sector).

Going by the two annual ASUSE reports, none of it is possible in absence of details. One may argue that the Centre would have relevant details which have not been made public. But that makes little sense because now is the time for finalising fiscal roadmap and preparing the final budget for FY25. If “essential” data is not in public no researcher or other stakeholders would understand or be able to assist in policymaking.

There is actually a disturbing pattern to the collection and dissemination of statistics.

Data is public good

The actual size of Indian workforce is not known despite six annual PLFS reports since 2017-18. None of these six gives that actual estimate of total workforce – formal or informal. It was the Azim Premji University’s 2019 study which analysed the unit level data of the PLFS 2017-18 to estimate the workforce at 465.1 million (and a loss of 9 million jobs for the first time in Indian history, from 474.2 million in 2011-12).

There is no way of knowing how many workers are engaged in informal sector – which comprises of both informal non-farm (ASUSE) as well agriculture. Agriculture is known to employ 45.8% workers (PLFS of 2022-23) but not in absolute number. The Ministry of Labour and Employment still carries the NSSO survey data of 2009-10 to claim that informal workers constitute 94% of total workforce (“46.5 crore comprising around 2.8 crore in the organised sector and the balance 43.7 crore workers in the unorganised sector”). The ASUSE gives absolute numbers but only for non-farm sectors for 2021-22 or 2022-23.

In short, nothing is known about the size of Indian workforce, distribution, wages and social security cover with any degree of accuracy. How can a meaningful policy be framed?

This is despite the National Statistical Commission (NSC) raising a red-flag more than a decade ago in 2012. Its “Report of the Committee on Unorganised Sector Statistics” said: “It is increasingly realised that lack of reliable statistics on the size, distribution and economic contribution of the sector...has been a major constraint in providing a realistic understanding of the significance of the Indian economy, leading to its neglect in development planning”.

Yet nothing has changed. India does not have a national policy to address the job crisis – work on which began in the UPA-II years. Nor has the Centre fixed a new minimum wage regime and provided universal social security cover (including gig workers) despite passing four new labour codes promising these in 2009 (wage code) and 2020 (codes for social security, industrial disputes and health and safety).

A new industrial policy, which was in the works for two years and a draft of which was circulated in December 2022, was put on the back burner in December 2023 as the government decided to continue with the PLI-DLI incentive model to raise its GDP and job shares.

While the number of jobs PLI-DLI model has created is not known, Raghuram Rajan and Rohit Lamba have pointed out the chinks in this approach vis-à-vis the Micron’s semiconductor plant in Gujarat. India is subsidising this plant by giving 70% of the project cost or about $2 billion as DLI. Since Micron’s plant would generate 5,000 jobs, the DLI of close to $2 billion would effectively mean Indian government spending ₹3.2 crore ($40,000) for creating each job! New Heavy Industry Minister HD Kumaraswamy was flagging this to question the government’s approach to job creation on June 15, 2024, before making a U-turn.

India’s job crisis is real. Had that not been the case, quality jobs (regular wages/salaried) wouldn’t be progressively falling and workers wouldn’t continue fleeing to low-productive low-paying informal agriculture (as six PLFS reports show). It is also reflected in rising inequality in income – which can be best addressed by creating enough quality jobs. High GDP growth has swelled the ranks of billionaires – adding 94 new ones to the list of in 2023, second after the US’s 132, to take the total to 271. Yet, 813.5 million Indians are receiving “free” ration, 102.7 million “subsidised” LPG cylinders, 110 million cash transfers of ₹6,000 and 129 million listed as “active workers” for the menial low-paying MGNREGS works.

The problem is not limited to employment alone.

Recall the consumer expenditure survey (HCES/MPCE) of 2022-23, released on February 24, 2024. It was partial just like the ASUSE – only value data was released, not volume data and hence, the impact of inflation can’t be measured. The previous HCES/MPCE of 2017-18 was junked for showing a fall in ‘real’ consumption (an indication of impoverishment). The PLFS of 2017-18 showing a 45-year high unemployment was withheld until the 2019 general election was over.

Worse, there is no sign of Census 2021 even three years down the road; there is no Economic Census after 2016; other key economic indicators like the IIP, CPI and WPI are based on more than a decade old data and the poverty line estimate is even older. Can we expect good quality policy based on bad quality data?