The government has been facing a lot of flak both inside and outside Parliament for the lack of figures on a whole gamut of issues. The Opposition has sharpened its knives to attack the government over this apparent absence of data, with Congress member of Parliament Shashi Tharoor even coining a new acronym for the NDA: ‘No Data Available’.
The government has been reluctant to share details inimical to its own interests. It is not just on migrants’ deaths, the government doesn’t have figures on how many small enterprises have been affected by COVID. The problem is compounded by the fact that India has a vast number of enterprises/entities that are unregistered and, therefore, difficult to track, which suits the official narrative as the government can easily brush aside queries on job losses, effects of demonetisation on small enterprises and kiranas, and even the underestimation of GDP contraction in the first quarter of this fiscal, besides, of course, the deaths of migrants during lockdown.
Related News: COVID highlights the urgent need to reimagine capitalism
In this backdrop, the news that the Ministry of Statistics and Programme Implementation (MoSPI) is taking steps to ensure that the vast informal sector gets adequate representation in the macro-economic data henceforth is welcome.
A committee formed under former chief statistician Pronab Sen has been tasked with formulating norms to conduct an annual survey of ‘unincorporated’ enterprises in India.
Till now, such a survey was conducted once every five years. This meant the data on ‘unincorporated’ sector (non-agri ‘unincorporated’ sector forms nearly a third of the GDP) came with a significant lag.
To set this problem right, the government had decided to start an annual survey. The first such survey was slated to begin from July 1 but was delayed due to the pandemic and some issues with sampling.
‘Unincorporated’ sector: While we often refer to the lack of adequate representation of the informal sector in calculating GDP and other macroeconomic indicators, such a term does not exist in the statistical system. What does exist is the ‘unincorporated sector’, which encompasses activities of entities that are not registered under the Companies Act.
Enterprises in the unincorporated sector do not present audited balance sheets. The ‘informal sector’ is only a subset of this ‘unincorporated’ sector and means enterprises not registered with any public entity. Till now, the ‘unincorporated’ sector (other than manufacturing, where plants registered under Factories Act were included annually) was measured only once in five years, and the report came out in the form of the ‘NSS Survey on Unincorporated Sector’, Sen told The Federal.
Data collection every five years meant we had to assume that the rate of growth of the corporate and unincorporated enterprises was the same for at least the next five years after the survey, Sen pointed out. This methodology worked fine for most of the years, but in some years it went terribly wrong. “In 2008 (during the global financial crisis), the bulk of the damage was in the corporate sector; the ‘unincorporated’ sector wasn’t that badly hurt but our methodology did not accurately reflect that. Then, in 2017 post demonetisation, the opposite was true. The ‘unincorporated’ sector was hit much harder than corporates and, again, this has not been reflected in the data so far. The economic census after demonetisation is yet to be completed, so we do not yet know the complete impact on the unincorporated sector,” he said.
New Panel: Sen said the earlier system of a five-year estimation for the ‘unincorporated’ sector was not working well and the proposal now is for the survey to be done every year by the MoSPI. “But, it is still a work in progress. The decision to do an annual survey was taken last year and the committee was set up. Our mandate is to plan out and supervise the annual survey on ‘unincorporated’ enterprises. This is different from the economic census, which is also conducted by the MoSPI, since in the economic census no financial questions are asked. The survey will have a detailed financial schedule.”
Sen also cautioned that while the survey will help with the robustness of the macroeconomic data, it will not affect quarterly estimates because of inadequate time. The biggest issue before the committee as of now is the large number of “births and deaths” in the ‘unincorporated’ sector. “There are a large number of births and deaths (of companies) in the ‘unincorporated’ sector, unlike in the corporate sector. We have seen that, by and large, every year, about three million new enterprises are born and about two million die. Such a large number of companies dying skews the sample significantly. We will have to over-sample to make a provision for a lot of sample companies expected to die. We are debating on how to make the sample even better.”
Jobs: Sen defended the existing data on employment put out by the government, saying there is an annual estimation now besides a quarterly report on urban employment. And that our jobs data is “fairly robust”. On private agencies like the CMIE coming out with weekly and monthly employment stats, Sen said CMIE collets data by telephonic interviews. “Such data doesn’t show characteristics – what sector people are employed in, how much they earn. Government surveys are much more detailed. The MoSPI is also now trying to move to a telephonic collection system but there are some issues here about time and the number of questions.”
Migration: Collecting robust data on internal migration is tough, says Sen. Most survey data collection is based on what is known as a “sampling frame” or a count of the population through listings of households and enterprises. MoSPI gets households from the national Census and enterprises from the economic census. “In migration data collection, we would have to ensure that there is a sufficient number of households in the sample which have either received a migrant or sent one out….This means a question on migration has to be included in the national Census, it currently isn’t. So, if a migrant from Bihar is sitting in Delhi, he will be counted as a Delhiite in the current Census format.”