We cannot say whether Prime Minister Narendra Modi has yet been successful in his promise to boost job creation in India
For some time now, there has been conventional wisdom afoot that growth in India, especially in recent years, has resulted in relatively little employment creation. This hypothesis is commonly referred to as ‘jobless growth’. However, most recently, there is a burgeoning pushback against this received wisdom.
Thus, in conversation with me on 29 June 2017 at the Reserve Bank of India (RBI), at a seminar chaired by RBI governor Urjit Patel, the then NITI Aayog vice-chairperson, Columbia University professor Arvind Panagariya, opined that the notion of jobless growth was a “little bit bogus".
He added: “The point that I have to make is that when such growth happens, 7-8%, it cannot be jobless, jobs are being created. Unfortunately, we are not counting them because there are no credible surveys of recent vintage."
Panagariya is spot on that data on employment creation in India is woefully out of date. What is more, official data, coming from the National Sample Survey Office (NSSO), are—as the name of the organization implies—based on surveys, rather than on measured data, raising legitimate questions about the extent to which surveys match reality.
Far better would be if one could rely on what economists call “revealed preference"—in other words, what people actually do, rather than what a subset of them tell surveyors that they do. Thus, it is laudable that Pulak Ghosh of the Indian Institute of Management (Bangalore) and Soumya Kanti Ghosh of the State Bank of India have brought “Big Data" analytics to bear, assembling a large data set on new registrations to the Employees’ Provident Fund Organisation (EPFO).
Unfortunately, rather than capitalizing on the fact that there is a rich new quasi-real time data set on EPFO registrations, the debate has been twisted by the authors’ untenable and implausible claim that new EPFO registrations, in particular of younger workers, largely represent net new additions to employment. Thus, the Ghosh-Ghosh claim that some 5.5 million new jobs are created in India every year, far above what is conventionally believed, has captured the narrative and detracted from their genuine contribution in bringing forward this important new data set.
As for the claim of some half a million or more new jobs every year, the Ghosh-Ghosh analysis has been surgically dissected by former Congress minister Jairam Ramesh and my IDFC Institute co-author Praveen Chakravarty. The bottom line of their persuasive critique is that new registrations to the EPFO are just that, and one is riding a wing and a prayer if one believes that these also represent net new job creation. Far more plausible, given the data set and time period the study considers, is that the large bump in EPFO registrations represents, rather, the induced formalization impacts of, first, demonetisation and, then, of the goods and services tax (GST).
All this furore masks what scholars have long known: that credible measurements of what economists call the “employment elasticity of economic growth" (EEEG)—which basically means the percentage increase in employment for a one per cent increase in growth of gross domestic product (GDP)—have shown this important statistic to be declining over time in India, and indeed globally in both advanced and emerging economies.
Thus, economist Steven Kapsos of the International Labour Office, in a 2005 research paper, showed that, for instance, for the US, the EEEG declined from 0.71 in 1991-95 to 0.43 in 1995-99 and then 0.20 in 1999-2003, with similar patterns in other economies.
As for India, a 2014 RBI research paper, by economists Sangita Misra and Anoop K. Suresh documents secular decline in the EEEG. Thus, from a high point of 0.57 in 1972/1973-1977/1978, it fell to an astonishing low point of 0.01 in 2004/2005-2009/2010. After that, the EEEG has hovered around 0.20, based on extant data. The one outlier is an incredible EEEG of 0.50 from 1999/2000-2004/2005.
It’s rare when data matches up with the party in power, yet that’s the case here: Estimates based on official data show virtually jobless growth under the first innings of the United Progressive Alliance (UPA) government, while they show the highest by far (in recent times) employment impact of growth during the first administration of the National Democratic Alliance (NDA).
In simpler terms, NDA-1 created a lot of jobs, UPA-1 practically none—based on estimates coming from official data.
The truth is, we simply do not have real time, credible data for NDA-2 to tell us what the EEEG has been since 2014. In other words, we cannot say whether Prime Minister Narendra Modi has yet been successful in his promise to boost job creation in India through schemes such as “Make in India", the legitimate intent of which is to attempt to generate large-scale labour-intensive manufacturing, which the country sorely needs.
One point is crystal clear: India needs to create a lot of good, productive, and well-paying jobs—according to the conventional estimate, about a million or so new potential workers enter the labour force every month—and this is, therefore, a legitimate and vital policy goal.
Unfortunately, despite the hoopla surrounding the Ghosh-Ghosh study, the jury remains out on whether this is yet happening or not.
Vivek Dehejia is a Mint columnist and resident senior fellow at IDFC Institute, Mumbai. Read Vivek’s Mint columns at www.livemint.com/vivekdehejia