To be able to solve the job riddle, govt needs to improve payroll data

Govt should make EPFO data available to researchers to find the truth about employment numbers

Rapid economic growth and widespread job creation were the two central promises of the National Democratic Alliance government as it assumed power in 2014. However, ever since growth started decelerating from the start of the 2016-17 fiscal year, there have been widespread concerns about the lack of jobs in the economy. In the absence of credible data, this narrative grew, especially in the wake of demonetisation in November 2016 and the rushed introduction of the goods and services tax in July 2017.

But a new study conducted by Pulak Ghosh, professor, Indian Institute of Management, Bangalore, and Soumya Kanti Ghosh, group chief economic advisor, State Bank of India, has gone against this conventional wisdom and suggested that 7 million jobs will be added by the end of the current fiscal year. Instead of depending on survey data, Ghosh & Ghosh focused on payroll data. They combined enrolment numbers from three administrative databases — the Employees’ Provident Fund Organisation (EPFO), the Employees’ State Insurance Corporation and the National Pension System — from April to November 2017 and then extrapolated the trend to arrive at the annual figure.

While this result has come as a shot in the arm for the government — Prime Minister Narendra Modi has said the finding is proof that his policies have worked — many others have claimed that the study overstates job creation in the economy. For instance, experts point to the fact that the total number of members (across all ages) in whose EPF accounts contributions were received grew by 5 million between 2013-14 and 2015-16. However, the Ghosh & Ghosh study claims that over 5.5 million will be added in 2017-18 alone and that too in the age group of 18-25 years. In the absence of other evidence suggesting rapid jobs growth in the economy, many have taken this claim with a pinch of salt. A big part of the problem is that there is a significant overlap and duplication across various social security schemes. As such, without conducting de-duplication, such an aggregation across different databases is likely to overstate the final number. Also, with more and more organisations formally coming under the EPFO, it becomes difficult to distinguish between the creation of new jobs and the effect of formalisation of the economy. On their part, the authors of the study have said that the assumptions they have made are a careful and conservative way of calculation with no effect on the conclusion of their study.

The point is that such studies are definitely a step forward. For far too long, India has lost out because it based its policy on surveys that were not even conducted in a timely manner. The surveys themselves have inherent flaws, ranging from definitional issues to problems of duplication. Analysing payroll data is a good way to go beyond the survey-based data that has been the focus of debate so far, and to that extent the study by Ghosh & Ghosh should be welcomed as it moves the debate in the right direction. More such debates are needed to get at the truth about employment. Further, resolving the questions raised will be facilitated if the government takes the lead and makes available to researchers in general the detailed payroll data maintained by the EPFO and others. A key imperative towards bringing more clarity would be for the government to take steps that improve payroll data and create a unique identifier to weed out duplication across databases.