Are the latest employment estimates by the Central Statistics Office (CSO) factually correct? No. They are off the mark, and confined to the economy’s organised or formal sector, accounting at best for 15% of the workforce. Is there a paradox in high output growth rates and the marginal effect on employment? Probably not, if one acknowledges that GDP estimates (after the latest revision a few years ago) have apparently overstated domestic output growth on account of the infirmities in the methods applied and datasets used.
Between September 2017 and April 2018, says a CSO media release last week titled “Payroll Reporting in India: An Employment Perspective – April 2018”, the economy added 4.1 million new jobs in the formal sector. Apparently, these estimates supplement the data on organised sector employment that are reported regularly in the Economic Survey since the early 1960s, compiled by the Directorate General of Employment, Ministry of Labour and Employment.
The CSO release defines jobs as ones that provide at least one government financed (or mandated) social security benefit such as Employees’ Provident Fund (EPF), National Pension Scheme, or Employees’ State Insurance Scheme. NITI Aayog and official economists have also put out similar estimates since early this year, inviting widespread scepticism among knowledgeable people. As the formal sector accounts for just about 8-15% of India’s workforce (depending on the definitions used), the official estimates are completely silent about the majority of the workforce engaged in the informal sector.
Lack of credibility
Ignoring the tiny size of the formal sector for the moment, how reliable or credible are the official estimates? Very little and for two reasons: the estimates are based on administrative records of implementing the social security schemes, whose completeness, consistency and accuracy are unknown; and since a formal (organised) sector worker, in principle, can legitimately access (or subscribe to) more than one social security scheme, double counting is a distinct possibility. The release does not explain how the problem is addressed in the database. Therefore, rightly, experts have demanded the release of the administrative data for independent verification.
Moreover, the official data suffer from a conceptual problem. The social security databases, by design, are lists of workers enrolled in the schemes, as an entitlement or as voluntary subscribers — not employment registers. These schemes are applicable to establishments above a certain size (of employment), and to certain kinds of enterprises. For instance, in the factory sector, those employing 20 or more workers are mandated to provide EPF to all the workers (with a matching contribution by the employer). So, if in a factory, employment goes up from 19 to 20 workers, it comes under the purview of the EPF, to be provided to all the 20 workers. Thus, the EPF enrolment increases by 20 workers, but the additional job created is just for one worker. Herein lies the fault.
Historically, evasion of EPF by employers is widespread, given poor enforcement of labour laws. The present government, in its efforts to formalise employment, has incentivised employers to enrol workers under EPF by offering to make employers’ contribution to the social security scheme for three years, thus boosting enrolment. Maharashtra, for instance, as an additional labour welfare measure, has widened the ambit of EPF to include all power-loom workers (irrespective of the size of the enterprise), boosting formal sector employment. Such measures, however temporary, may enlarge the formal sector size but cannot be counted as new jobs created. Hence, the official measure is flawed.
The formal sector stands at the apex of India’s labour market pyramid, agriculture being at the bottom, employing 50% of the workforce. The remaining workers are in the non-farm informal sector, spread across rural and urban areas. In fact, it is this sector that has grown in recent decades at the expense of the other two sectors mentioned above. Moreover, nearly half of the informal labour workers are self-employed in household (or own account) enterprises, often engaging unpaid family labour. Varying degrees of under-employment or disguised unemployment are the defining feature of informal labour markets.
Since 1972-73, the five-yearly Employment-Unemployment Surveys (EUS) conducted by the National Sample Survey (NSS) have been the mainstay for analysing labour market trends. Though infrequent, the database has served a valuable purpose of capturing the complexities of the labour market; access to household-level data lately has spawned rich and granular analyses of the informal economy.
As the last round of the EUS was in held in 2011-12, there is no reliable way of updating employment trends. The EUS has been replaced with an annual Period Labour Force Survey, and a time use survey. However, it will be a while before these data series come up with stable and credible estimates.
The Labour Bureau under the Ministry of Labour and Employment has been carrying out household surveys somewhat similar to the EUS since 2010-11. They show a decline in worker-population ratio between 2013-14 and 2015-16, suggesting a deteriorating employment situation recently (the estimates for 2016-17 are apparently withheld).
Why are so few jobs created when the economy is said to growing at over 7%? An answer is that GDP growth figures are probably overestimated on account of the mis-measurement of GDP in the new National Accounts Statistics (NAS) series (“ Why 7.6% growth is hard to square ”, The Hindu , February 12, 2016). The economy is probably growing much slower. The faulty barometer of economic well-being seems to be misleading the nation.
For instance, in manufacturing, in the last few years, the growth rates reported by the Index of Industrial Production (IIP), and the Annual Survey of Industries (ASI) are consistently lower than those reported by GDP in manufacturing, suggesting an overestimation of manufacturing value added in the NAS. Demonetisation and the Goods and Services Tax (GST) seem to have dented informal sector production and employment, which official data sources seem in no position to capture.
The CSO’s press release has claimed that 4.1 million new jobs were created in the economy’s formal sector during eight months since last September. It is based on enrolment of additional workers under state-provided (or mandated) social security schemes. The claim seems questionable on empirical and conceptual considerations. Empirically, there is a possibility of double counting of workers. Consistency and accuracy of the administrative datasets have not been publicly established. Conceptually, the data are about enrolment in the social schemes, not about new jobs generated. The recent rise in the enrolment is probably on account of an expanded coverage of the social security measures by incentives to employers’, and the mandated expansion of the coverage in States like Maharashtra.
The official claim is about the formal or organised sector, which at best counts for 15% of India’s labour force. In the absence of a nation-wide, time-tested EUS data for recent years, there is no way of knowing what is happening in agriculture and the non-farm informal sector, accounting for 85% of the workforce.
The paradox or mystery of poor employment growth and high GDP growth disappears if one concedes that the recent output growth rates are probably overestimated after the latest revision of the National Accounts Statistics a few years ago, on account of the questionable methodologies and databases used.
R. Nagaraj is with the Indira Gandhi Institute of Development Research, Mumbai