Amid talk of jobless growth and the urgent need for India to create employment for its millions of young men and women, the Chief Statistician of India, T.C.A. Ananth, says there is not much evidence to show that the unemployment scenario may have worsened in the last three or four years.
Ananth, who is also the secretary of both the National Statistics Commission and the Ministry of Statistics and Programme Implementation, told ThePrint that an economic slowdown was building up well before the Modi government announced its controversial decision to demonetise Rs 500 and Rs 1,000 currency notes and refutes doubts about the country’s official data.
Excerpts from an interview with Deputy Editor Anubhuti Vishnoi:
How would you describe the state of Indian statistics and its credibility?
My sense is that it’s very good. I distinguish between statistics and data. We live in a sea of data. Statistics are compilation of data which are put together adhering to certain structures, norms, and methodologies. Ultimately, everybody uses our data and what else does a statistician need for credibility? If your methodology, structure is clear, it will have credibility…by and large, in India data has very high credibility.
How do you then look at the controversy around 8 per cent GDP growth?
Growth is not a primitive statistic. Rather, growth is a derived number. The primitive underlying growth is GDP. Rather than saying that 8 per cent is right or wrong, the question is whether the primitive — that is GDP — is better or not. Then the answer in short: our current compilations are much better.
Ultimately, GDP is a compilation of partly accounting-based data and partly estimates where accounts are not available. My assessment of quality would be, what is the proportion of GDP which is anchored in accounting detail. On all these parameters, one can demonstrate the current series is better than the others.
And the back series?
Back series is a bit of challenge. We have not updated the back series and we have explained why we haven’t. Our task is to estimate GDP today in the most accurate way possible. A lot of data is available today on which analogues are not available in the past. For example, I am using data now in the new series based on the e-filings. Earlier, we did not have such complete data of company accounts. Therefore, it was a challenge to put together corporate data. For national accounts to fill up the gaps, a variety of approximations were used to measure the estimation of corporates. So, we don’t have easy analogues going back in the past…to do this statistically.
What will be the impact of GST in terms of better data?
The expectation is that it will be huge. I am saying this on the basis of the GSTN structure of reports and forms and how they propose to fill them out. Earlier we used to get data separately from service taxes, excise taxes, central excise taxes etc. Now, under the GSTN regime, we will get a single data source.
What is the state of unemployment in India?
I don’t have the exact number but it’s relatively small. It’s about 3-4 per cent of our total workforce. It’s small, largely because we don’t have social security. So they have to work to earn. Unemployment is also not an option. So, people work.
We have a very large part of our work force that is self-employed. While there is a rise in casual wage labour, there is no significant proportionate increase among regular salary wage workers. Regular salary wage, at least, from the most labour survey instruments I have seen, have remained around 10 per cent of the total workforce.
What is the trend on the proportion of women in the workforce?
Women’s participation has been declining. This is truer of women who are literate but not high school. Post-secondary education, women’s labour force participation has not been coming down. In the illiterate category, the performance is not so clear.
So, what are the other areas worry on this front?
One, aggregate unemployment for educated youth particularly in the age groups 19-25 years immediately post education, is quite high. There is a need to address that.
The other is, in terms of how many of these jobs have a quality earning capacity. Now that is harder to say. Because in earlier NSSO surveys, except for those who worked in wage labour, we didn’t have a measure for earnings. In the current survey we have launched, we hope to also measure earnings of self-employed to get a slightly better understanding in the future.
So is there an unemployment crisis?
The take on jobs you are talking about, is driven from a different set of surveys which are not labour force surveys and have been spun to make stories. I don’t think these issues are directly comparable. I would say that there are characteristics of our labour market that are long term and these characteristics would suggest employment in India has improved, wage employment and employment opportunities have also improved.
What about the impact of demonetisation?
Some answers you will get from GDP compilation when accounting data for 2016-17 for all companies becomes available. Now that will be a huge addition. September 30 is the last date. After that we will get access to a much richer picture of the corporate response to it and certainly we will cover small and medium companies as well. But usually our early assessment is quite close to subsequent assessments. So, I don’t personally expect to see much change.
But many argue that demonetisation has caused an economic slowdown?
That perception is probably flawed because first of all the slowdown which people were talking about was in the works well before demonetisation came into effect. There was a huge crash in global commodity prices between 2014 and 2015. That also showed up in the price data.
Now how much of the slowdown is because of demonetisation and how much is because of this long run process which was in play is something you can disentangle but only when you have long time series data. You cannot do it with one quarter or two quarter numbers.
What is your view on the debate around privacy and data?
Privacy is an individual attribute. Statistics is based on individual data but it’s about behaviour at the level of identifiable groups. Statisticians do collect a lot of intrusive data but we have bound ourselves by saying that we will only use it for statistical purposes. When we disseminate things we do so either in aggregates or we try to keep the data anonymous.
Now, the government has a variety of other objectives. The government is also a law enforcer, a regulator. How does government meet these obligations? What are the trade-offs with privacy? We should let the courts answer.