The fear of statistical messages

The fear of statistical messages

Key indicators of Employment and Unemployed Survey (EUS) based on 66th round (2009-10) were recently released by the National Sample Survey Organisation (NSSO). Broadly, the report showed that employment creation between 2004-05 and 2009-10 has been only one million jobs against the official claims of the government of creating more than 50 million jobs. The results confirmed what has been reported earlier from the 64th round (2007-08) of the EUS which is similar in coverage and otherwise to the 66th round. In a previous column in this newspaper, results of the 64th round were highlighted (“The shadow of jobless growth", Mint, 21 July).

The results of the 66th round are no different from what had emerged from the 64th round. While the government completely ignored the warning signals based on that round, there is now a concerted effort to discredit the data from the latest round on the grounds of quality of investigators and the data collected. The attack has come from some of the highest officials including the previous chief statistician of India and the deputy chairman of the Planning Commission. The criticisms—ranging from the presentation of the data to use of contract staff by the NSSO—have raised questions not only on the credibility of the data reported but also on the institution which is otherwise widely respected in academic circles. These need to be responded as a prerequisite to a meaningful debate on the outcomes.

Unfortunately, such attempts to attack the statistical system are meant to side-track attention from real issues and are not new. A well-known example of this is the consistent criticism of the divergence between the estimates of private final consumption expenditure of the national accounts and those from the NSSO. There has, however, never been any such criticism of the employment data before the release of this report. But there is also a method in this criticism of the NSS data. The same government officials who have claimed credit for higher growth in wages (during first United Progressive Alliance government [UPA-1] trumpeted as a possible reason for the inflationary pressure), significant decline in poverty and the impressive growth rates have been quick to debunk the data of the 66th round the EUS. It is worth mentioning here that recent attack on inflation data is also part of the grand design of “malign when uncomfortable" syndrome. Given the chain of events surrounding the publication of this report, such attack on the quality of data is not unexpected.

The truth of the matter is that the NSSO also happens to be the primary agency for wage data collection that has been used to argue the success of the government in delivering growth or for that matter inclusive growth. The poverty estimates which were reported by the deputy chairman of Planning Commission in a recent Economic and Political Weekly article as a success of the government’s inclusive growth agenda are based on data collected by NSSO investigators. There was no mention there of any discrepancy in data collected or any hint that use of contract staff could have led to faulty data collection.

Not to forget, even the gross domestic product (GDP) data—considered sacrosanct—is based on ratios and estimates largely derived from NSSO surveys. The GDP base is updated every five years. Why do these base years coincide with the employment-unemployment survey “thick" rounds? The reason is that some of the crucial ratios and parameters that form the basic building block of adjusting the GDP base are derived from the NSS surveys. Has there been any question on the credibility of the growth numbers because they are based on faulty data collected by contracted inefficient NSSO investigators?

This data from the EUS of the NSSO is as good or bad as the GDP data, wage rate data or any other statistical data that is routinely used by the planners and government officials for formulating plans. The broad numbers reported by the press are not the creation of some journalists but are the same as that what was reported on page 21 of the report.

The fact that the UPA-1 did not create as many jobs as it promised or what should be expected of an economy growing at 8-9% is not debatable. These are facts. Credit must be given to the National Statistical Commission and the NSSO that they have presented the facts without fear or bias. It is not the contention of the column that there should not be any debate on the quality of official statistics or that they are sacrosanct. India has a long and rich history of such debates with active participation of leading academics and some of them continue even today. Good and credible data is essential for any meaningful analysis of the past trends and prerequisite for future planning.

The unnecessary diversion that the debate on quality of 66th round data has created is not in the right spirit of such debate. It has diverted some of us from analysing the rich content of the report on what happened to the economy in the last five years.

Himanshu is assistant professor at Jawaharlal Nehru University and visiting fellow at the Centre de Sciences Humanities, New Delhi.

Comments are welcome at