Where Are Official Statistics on Employment?

It is good news that the Labour Bureau will revive its establishments-based Quarterly Employment Surveys or QES, using a larger sample. Since the Periodic Labour Force Surveys or PLFS collects data from households, the proposed quarterly survey of jobs will collect data from establishments. But it is advisable to review the multiple existing employment and other labour market-related databases and identify the statistical voids, deficits and shortcomings in them before launching one more which does not add much value.

Existing Statistics on Employment

Presently, data on employment is collected from multiple sources. Under the Employment Market Information (EMI) and Employment Exchanges (Compulsory Notification of Vacancies) Act, 1959, data on employment is collected from the organised sector. The Economic Census periodically (not at fixed intervals) collects data from households and establishments covering the unorganised and the organised economic segments but excludes crop production, plantation, public administration, defence and compulsory social security.

Once in a decade, the Office of the Registrar General and Census Commissioner collects information on many variables including workers from the households. The National Sample Survey Office (NSSO) was conducting a quinquennial survey on several variables from households including employment, which it classified as “self-employed”, “regular wage/salaried” and “casual workers”. Now, the annual PLFS carries out what NSSO had been doing. The Annual Survey of Industries or ASI also collects information on numerous variables including employment in organised establishments in the factory sector. Occasional surveys of the unorganised sector by the NSSO also take place. Administrative data is generated under various labour laws by both central and state governments .

Ad Hoc Supplementary Labour Market Surveys

The global financial crisis of 2009 arguably impacted the Indian economy. The Labour Bureau initiated the QES the same year and between 2009 and 2015 it conducted 28 surveys to assess the impact of the slowdown on employment. In 2017, it started a new series “to generate high-quality accessible data on the labour market for effective implementation of policies and welfare of labour.”

The new series, QES-LB, relates to July-September 2017 and published in March 2018. It was based on a survey of eight core sectors, that is, manufacturing, construction, trade, transport, education, health, accommodation and restaurant and IT/BPO. Since September 2017, the central government has been using the registrations in the Employees’ Provident Fund Organisation or EPFO, under the EPF and Miscellaneous Provisions Act, 1952, to provide estimates on job creation in the formal sector (known as “payroll data”). According to EPFO, the data relates to the workers who register with it and make their first non-zero contribution in a particular month. The payroll data, including those who subscribe to the Employees’ State Insurance or ESI showed greater employment generation as opposed to the meagre employment generation shown by QES-LB. For example, the QES-LB showed that 1.36 lakhs jobs were created during July-September 2017, as compared to the previous quarter. On the other hand, Payroll Data showed that over 35 lakh jobs were created between September 2017 and March 2018. Though the two periods are not the same, the government celebrated the report that showed a more favourable picture of formal sector employment.

Since payroll data has been painting a much-wanted rosy picture on the employment front, the NDA government, which is facing criticism for not fulfilling its grand poll promise of creating 20 million [2 crore] jobs, grabbed this data and dumped the QES-LB. However, payroll data was heavily criticised by labour economists as it only reveals a partial picture of what is happening on the jobs front.

Why Another New Data Initiative?

Notably, the central government has not so far framed a National Employment Policy, even though India has ratified the International Labour Organization or ILO’s Employment Policy Convention, 1964 (C.122). In fact, the Sustainable Development Goal 8 of the United Nations aims to “promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work for all.” Further, the labour market in India has historically sought to tackle two issues—jobless growth via creating jobs, and formalising informal employment.

On the other hand, unemployment has shot up in recent years. The PLFS for 2016-17 showed a 45-year high in the rate of unemployment at 6.1% which dipped marginally to 5.8% in 2018-19. Covid-19 disrupted the economy as a whole, impacting the labour market substantially and all the stakeholders are anxiously waiting for pre-Covid normalcy to return soon. We must note here that the pre-Covid unemployment levels were higher than the historical rates. During the Covid-19 period, media and academics have relied on the high-frequency database of CMIE on employment and unemployment notwithstanding its limitations.

The unemployment rate hit a peak of about 24% in April and though it cooled in later months, the rate has been stubbornly stuck at 7% to 8% in the last few months. Further, the employment rate, even as it fluctuated, was lower than pre-Covid-19 levels. Unemployment and informal workers have hogged the limelight during the COVID-19 period, and migrant workers in particular, for the wrong reasons.

Reportedly, the Centre proposed to launch three separate surveys to collect data relating to wages and social security of migrant labourers; domestic workers; and others such as lawyers, doctors’ assistants and receptionists. However, there is no update on this. There is a severe dearth of information on employment in the unorganised sector and in the emerging service-sector segments. No doubt, the government is under tremendous pressure to create a database on employment and there are lacunae in the existing databases. The over-reliance on private agencies’ data in case of the emerging sectors is one important concern.

No More Data under Employment Market Information

The Directorate General of Employment and Training (DGE&T) publishes data collected under the EMI in outlets such as Quarterly Employment Review, Quick Estimates of Employment in the organised Sector (Quarterly), Employment Review (Annual), amongst others. This was the sole source of data relating to the organised sector. However, the quarterly and the annual reports published by DGE&T are not easily available, physically or on its website.

Economic Surveys also provide macro statistics on employment, by public and private sectors, and on industries. However, this was stopped after data for 2012 was published in the Economic Survey for 2016-17. The Economic Survey for 2017-18 repeated the 2012 data and recent Economic Surveys, such as for 2019-20, do not carry these statistics. Is the data blackout especially on important variables like employment reflective of a deteriorating employment situation in the organised sector, or is it due to the failure of administrative data machinery (as I have shown elsewhere) or both? Given the attempts to suppress the release of PLFS 2017-18 by the government agencies, the data blackout raises serious concern.

Administrative Data for Service Sector

The Shops and Establishment Act (Shops Act) is a state law that typically includes commercial establishments that are in the business of advertising, commissions, forwarding or commercial agencies, banks, brokerages and so on; or “shops” carrying on any trade or services, or rendering services to customers such as offices, store rooms, godowns, restaurants, hotels, theatres, and so on. State governments are responsible for collecting and disseminating statistics regarding establishments registered under the Shops Act. But these are not published in any statistical outlets save those published by the Labour Bureau as administrative reports, which are limited in scope as they do not provide data at the disaggregated levels (industry-wise, using the National Industrial Classification or NIC series). However, as state labour departments supply defective or incomplete data to the Labour Bureau, their macro statistics are not reliable (see footnote to Table II). Administrative data sources are pitiably and even incurably defective and utterly useless.

Private Agencies Estimate Employment for Emerging Sectors: Where is Official Data?

The business press often goes gaga over jobs created by start-ups and “new generation” firms in the services and the digital sectors, such as Paytm, Byju, Delhivery, Udaan, Phonepe, Unacademy, Bigbasket, and so on, apart from the Unicorns and Soonicorns who get media attention. For instance, the Economic Times recently reported that around 40,000 jobs were created during October 2019-October 2020 which it said constitute “green shoots” in the labour market. Another report cited Naukri.com’s report in October which suggested e-commerce and retail are “bouncing back”.

There are endless reports on employment generation, which are by the thousands in the modern service sector outfits, while around 40 crore workers are employed in the informal economy and around 3 crore are unemployed. While on principle every job created is a reason for celebration, given the extent of unemployment and underemployment, these typically urban stories of employment creation—which are in the region of thousands or even lakhs of jobs—are hardly a legitimate “feel-good” factor.

The emerging sectors like the digital, gig economy, platform economy and modern retail and e-commerce segments are important, but do we have official data on any of these segments? For example, estimates of Ola and Uber drivers vary from 9 lakh in 2016 (3,50,000 at Uber and 5,50,000 at Ola), to 15 lakh in 2018. Some have estimated that there are 30 lakh temporary workers including independent contractors, online platform workers, contract firm workers, and on-call workers in the gig economy in India in 2020.

Assocham estimated that “e-tail and the allied sectors, like logistics, warehousing, IT/ITES [are] expected to create direct employment for around 1.45 million [14.5 lakh] of workforce by 2021, a significant jump from the 23,500 jobs which existed in 2012.” Further, it says the “overall e-tail logistics and warehousing sector to directly employ more than 1 million [10 lakh] people by 2021, becoming the largest contributor in terms of employment opportunities created by e-tail.” It also has estimated the growth potential of the gig economy at an annualized 17% growth rate to reach US$455 billion in 2023.

Similarly, NASSCOM has been providing estimates of people employed in the IT and ITES and knowledge technology sectors. All stakeholders including academics use them. However, it must be borne in mind that these are employer organisations. In other words, they are “pressure groups” in a pluralistic democracy. Hence, their data, especially on employment, export and growth potential could reflect this political aspect of lobbying to influence economic and labour market policies in their favour or seeking sops. The political economy of this kind of exercise must be kept in mind while using their databases. One wonders how they arrive at these estimates given that the gig and the platform economies are wide and deep? If these sectors are important and promise so much and the government desires their growth, where is the government data on it?

Need for Comprehensive, Reliable, Official Service Sector Data

Private and public establishments in the other sectors, such as finance, transport, insurance, are registered under different laws. Hence, we do not have a single source under which statistics relating to the wide range of segments in the service sector are available. There is a dire need for the central government to think of setting up an annual survey of the service sector, similar to the ASI, either comprehensively or on a sample basis. It is shocking that even though the organised manufacturing sector employed around 9 million [90 lakh] (as per PLFS 2017-18 data) and the informal segment of it employed around 48 million [4.8 crore] we have in place a sturdy and well-established database in ASI, which covers this segment of the factory sector. Academics and policymakers use ASI data extensively. On the other hand, the organised service sector employed in 2017-18 around 32 million [3.2 crore] (little more than three times of the manufacturing sector) and the informal segment around 114 million [11.4 crore]. As argued earlier, many of the emerging sectors like e-commerce/tail, gig and platform economies, IT and IT-enabled services, start-ups populate this sector and yet we do not have an official database relating to this sector?

Need for Periodical Surveys on “Flexi” Workers

Employment of “flexi” workers such as casual workers, temporary workers or trainees, who are directly employed by the user enterprise or primary employer; and contract workers, who are indirectly employed through the labour contractors, has increased substantially. While launching the Atal Bimit Vyakti Kalyan Yojana under the ESI in December 2018, the Centre explicitly accepted the nature of this labour market arrangement when it observed that the “current scenario of employment in India...has transformed from long-term employment to short-term engagements in the form of contract and temping…”

Now the Industrial Relations Code, by catering to the demands of employers’ organisations, has legalised fixed-term employment or FTE as a legitimate and legal type of “flexi” category employment, so this category of workers will significantly grow. The central and state governments and local bodies have been employing contract and casual workers in large numbers, yet we do not have any survey to fully capture the extent of employment of “flexi” workers in the non-agricultural sectors. The ASI’s statistics deal with contract labour only in the organised factory sector and only provides information on total contract labour employment, person-hours worked and earnings.

The recommendations of Prof Trilok S Papola in a 2014 study on labour statistics in India are pertinent here: “There is no regular/ad-hoc survey that can give a comprehensive picture of employment of contract labour. NSSO-EUS [NSSO’s Employment and Unemployment Survey] provides information on self-employed, regular and casual categories, but not on the category of contract labour.”

We do not have data on contract and casual or temporary workers employed in the government sector or in the service sector. The Labour Bureau has conducted occasional surveys on contract labour in a few industries, but they provide limited information.

No Reliable Independent Estimates of Unorganised Workers

Prof. Trilok S Papola especially mentions the absence of a separate estimate for unorganised sector employment, from any source. Researchers estimate this date either by deducting the organised sector employment (under EMI) from the total workers estimated under by NSS/PLFS or they use the definition of unorganised sector provided by the National Commission on the Enterprises in the Unorganised Sector (NCEUS). Hence, the estimates of the unorganised sector vary. He also highlights the complete absence of any survey concerning the construction sector. The woeful experiences of migrant workers this summer exposed the complete absence of any database on migrant workers despite the Inter-State Migrant Workmen Act, 1979 existing on the statute book. Now, the Occupational Safety and Health and the Working Conditions Code (OSHWCC) provides for maintaining a database of migrant workers through the registration process.

India has ratified ILO’s Labour Statistics Convention, 1985 (No. 160) in 1992 and is bound to generate detailed statistics on economically active persons among others. The ILO Conventions and SDGs require India to look at labour statistics in a comprehensive and meaningful manner. The government must carefully analyse the existing data-collection exercises, rationalise them where needed and improve the administration of its statistical tasks, which are woefully and criminally inefficient. It should accordingly establish systems that will provide comprehensive and periodic reliable databases to help frame meaningful policy suggestions and make lawmaking a meaningful exercise too. The Union Labour Minister Santosh Gangwar’s remarks when a new logo was launched for the Labour Bureau assumes huge relevance in this context. He reportedly underlined the importance of databases in policymaking and called for increased use of Information Technology to make data more accessible and accurate. He said the government needs such data, for it is essential to its work in providing employment.

The author is a professor at XLRI, Xavier School of Management, Jamshedpur. The views are personal.