Data can be next oil only if it flows freely

Data must be gathered and disseminated swiftly; else it will lead to proxy efforts which will create further confusion

In the 1990s, the physical notice board at the Bombay Stock Exchange used to be the only place Mumbai’s financial journalists would find news on important corporate developments.

This piece of paper that would typically be pinned up late evening could contain a wide range of information, from harmless news on corporation actions or a proposed mega merger.

Knowing when the piece of paper would go up and in some cases anticipating it became an art in itself. Indeed, many intrepid financial journalists of the day made their fame by ensuring they were there in the vicinity of the notice board to catch and break the news next morning.

Something similar has been happening with government data, except that no journalists (to our discomfiture obviously) are involved.

A debate has broken out about whether or not jobs are being added in the economy. A paper led by State Bank of India’s Chief Economist Soumya Kanti Ghosh argues that some seven million jobs were added in the last year. The source of this information and conclusion is Employee Provident Fund Organisation (EPFO) data.

Wonderful news but there is just one hitch — no one else has seen or is able to access this data.

Ghosh’s proposition has already come in for criticism from former Finance Minister P Chidambaram and economists like Mahesh Vyas and Praveen Chakravarty. But the dissenters readily acknowledge one thing, they don’t know better because the data is not in public domain.

Chidambaram questioned the veracity of the data. He said: “It took the country 70 years to create a ‘payroll’ stock of 919 lakh [91.9 million] jobs but, miraculously, in just 12 months, the country will generate 70 lakh [7 million] new ‘payroll’ jobs — that is nearly 7.5 per cent of the current stock!” He says it is evident Ghosh was given privileged access to data that is not in public domain. “So, first, the government must put all EPFO subscriber data in the public domain.”

But the asymmetry does not end there. In the Economic Survey 2018 released on January 29, the total organised sector employment number for public and private sectors is 29 million (295.8 lakh).

But the table is dated 2012, six years before the present day. The source is the Directorate General of Employment, Ministry of Labour and Employment. So clearly, the arm that is leaking data is not familiar with its official correspondence with the Ministry of Finance which puts together the Economic Survey. Or more likely no one cared to think about this.

But the jobs data set is only one instance of information not being made publicly accessible. Here are a five somewhat random data points which we at IndiaSpend have tried to source without success.

Migration data. The 2001 Census had it but not the 2011 Census.

India’s biggest health database, the National Family Health Service, was first put out in 1992-93. It came out every five years but stopped in 2005-06. The next one came out in 2015-16. So for 10 years, there was no reliable data on what was happening to the country’s health, quite literally.

State budget expenditure reports on education and health care are not available. In general, most state budget documents fail to provide the “second layer” of information.

Real estate data, registrations could be in public domain but are not. Similarly, data on land records is not publicly accessible. If most real estate transaction data are digitised, then why not put it in public domain? Would it not level the playing field considerably?

State crime records bureau data. We now have a better idea on top level crime data but not region-wise within a city or by police station. If I want to find out which area is more affected by a certain kind of crime I cannot. Similarly, there is no classification for hate crime, of the like we have been seeing in the “gau rakshak” cases.

The leaking of EPFO data might well have been a trial balloon. This way, the government gets to see the nature of public reaction and then decide whether or not to actually put the information in public domain. While maintaining deniability if something goes wrong.

But selective leaks make a bad situation worse since the only people who have received it have spun it positively.

The issue of jobs is perhaps the most critical issue facing this country and government. A failure to create jobs cannot be masked by selective data leaks, surely not for too long. Nor can the numbers acquire any credibility if only a few economists or academicians are able to access the data.

To be fair to this government and previous ones, we have come some way in data transparency. More data points are being shared and in many cases, like air quality, the government is being responsive to the needs of the general public.

But the hunger for data is insatiable. We have quickly leapfrogged from getting no data to demanding it realtime and live. Air quality is a classic case. The government is not just measuring air quality real-time but also putting up large public displays. Earlier, it would have held data back because poor air quality would suggest a governance breakdown. Which it is.

In the case of jobs or EPFO data, a fair demand would be quarterly numbers. Like with other economic parameters.

Governments are not companies who race to file quarterly reports to stock exchanges. But companies too have come a long way from the 1990s. So much so that it is impossible for a company to reveal any price-sensitive information selectively. If that happens, the Securities and Exchange Board of India will ask questions.

A government must be similarly accountable, to its citizens. Data must be gathered and disseminated swiftly; else it will lead to proxy efforts which will create further confusion. And efforts must be continually made to ensure the speed of gathering, processing and disseminating is improving all the time.

Data is surely the next oil but it must flow freely too.