How badly India needs to fix its employment (and unemployment) and jobs data problem can be seen from the furious debates that break out each time a new study on job creation comes out. The latest is the just released Employee Provident Fund (EPFO) data, which shows 3.1 million subscribers were added in the six month period between September 2017 and February 2018. Some economists, who support the government, have pounced on this to come to the conclusion that 6.2 million jobs were created last year. Others have rubbished those conclusions, pointing out that the data set is too small and the month by month variation in the data is too wide, and finally the conclusion of 6.2 million jobs in the formal sector flies in the face of other economic evidence.
Other slices of data available lead to different conclusions. The survey done by CMIE on a regular basis shows that job creation is anaemic at best. The CMIE survey is a household survey, and it generally captures the state of job availability in the informal sector. Meanwhile, the KELMS data on jobs and productivity that is available on the Reserve Bank of India website, shows that jobs actually shrank in between 2014 and 2016, when the economy was actually growing extremely fast. The Ministry of Labour is reworking its survey now, but in the last few years, it showed that job creation was fairly low in the limited number of sectors it was tracking.
Niti Aayog has submitted its Task Force on Improving Employment Data draft some time ago for public comments, and it makes a number of important suggestions including using a mix of household surveys (which capture the number of people in a household who are working), enterprise surveys (which captures how many people are employed in the firms being surveyed), as well as other methods, including using EPFO and other data as proxy for estimating jobs being created in the formal and informal sector. It also discusses the strengths and weaknesses of each method of estimating jobs, employment and unemployment, and also pitches for the use of more technology to capture data better and analyse it better. It also briefly touches upon the use of the GST filing data to capture more information on employment, which is an eminently sensible suggestion.
However, when the government is looking at improving its statistical coverage and capture job data better, it needs to widen the scope of survey to also capture salaries and wages at which jobs are being created. That would give a better picture of the kind of jobs being created. In the US, for example, while unemployment is showing a downward trend in recent times, economists estimate that real wage growth has not improved for many years. One estimate by Brookings said that wages in 2017 in the US were only 10 per cent higher than they were in 1973, with an annual real wage growth of barely 0.2 per cent.
In India, despite the measures introduced by the government, real growth in rural wages - both farming and non-farming - has been fairly low. There is no estimate of how wage growth has been in urban areas, and whether high end jobs are being created or low end ones.
At any rate, there is one point every economist agrees on. The number of jobs being created by the Indian economy growing at its current rate is still far lower than the number of people joining the workforce every year. The difference is opinion is on how big is the gap between jobs being created and jobs that require to be created. The World Bank recently came up with an alarming estimate that India will need to grow at 18 per cent if it needs to create enough jobs. That is clearly an impossible feat - but it does give the Indian government some idea of the problem at hand. The demographic dividend is not turning out to be a dividend at all, because we simply cannot create enough jobs or grow fast enough to employ the people coming into the job market.