• India
  • Jun 20

Ex-CEA accused of cherry-picking data

Rejecting the claims of former Chief Economic Adviser Arvind Subramanian regarding over-estimation of GDP growth after 2011, the Economic Advisory Council to the Prime Minister (EAC-PM) said his analysis is based on “cherry-picking” of high-frequency indicators while ignoring data on services, agriculture and robust tax collection.

The EAC-PM also accused him of blindly trusting data provided by a private agency, CMIE, and distrusting government institution Central Statistics Office (CSO).

In a paper, EAC-PM said India’s GDP estimation methodology stands at par with its global standing as a major and responsible economy. The paper was authored by Bibek Debroy (chairman, EAC-PM), Rathin Roy (Member, EAC-PM), Surjit Bhalla (economist), Charan Singh (economist) and Arvind Virmani (ex-CEA).

The EAC-PM, however, conceded that GDP calculation has always been an imperfect art which involves extrapolation, estimation and sometimes even guesstimates based on past trends.

What are the issues raised by Subramanian?

The former CEA had claimed that India’s economic growth rate had been overestimated by around 2.5 percentage points between 2011-12 and 2016-17 due to a change in methodology for calculating GDP.

India’s GDP growth rate between this period should be about 4.5 per cent instead of the official estimate of close to 7 per cent, he said in a research paper published at Harvard University.

Manufacturing is one such sector where the calculations have been largely mismeasured, wrote Subramanian, who quit as the CEA in August last year before his extended tenure was to end in May 2019.

Subramanian had stated in his paper that he was not including tax collection data in his analysis because their relationship to GDP was unstable.

What are the arguments of EAC-PM?

Subramanian’s report “would not stand the scrutiny of academic or policy research standards” as it lacked “rigour in terms of specific data sources and description and alternative hypothesis”, said the EAC-PM.

In the former CEA’s paper, “high-frequency indicators are cherry picked”. Subramanian used 17 indicators to express his scepticism about the growth rates post 2011-12. A majority of the 17 indicators have been taken directly from Centre for Monitoring Indian Economy (CMIE), a private agency that is not a primary source of information but collects it from different sources.

A cursory look at the indicators suggests a strong link with industry indicators (a sector that contributes an average of 22 per cent to India’s GDP), while the representation of services (60 per cent of GDP) and agriculture (18 per cent of GDP) is as good as missing.

IIP (Index of Industrial Production) like any other indicator is not the most precise indicator of industrial growth. To improve IIP, a new series was introduced when the base year was revised to 2011-12.

Subramanian’s logic of not using tax data appears to be a convenient argument meant to avoid inconvenient conclusions based on hard facts. Unlike many indicators, tax data is not collected through surveys or by agencies through arcane techniques. These are hard numbers and should be an important indicator of growth.

There have been no major changes in tax laws until the end period in the analysis of Subramanian.

How did the statistics ministry respond?

The Ministry of Statistics and Programme Implementation (MoSPI) had also dismissed the contention of Subramanian and said that it follows accepted procedures and methodologies for arriving at projections of national income.

The estimation of GDP in any economy is a complex exercise where several measures and metrics are evolved to better measure the performance of the economy, the MoSPI said.

The ministry also pointed out that the methodology of compilation of macro aggregates has been discussed in detail by the Advisory Committee on National Accounts Statistics (ACNAS) comprising experts from academia and institutions like the RBI.

In India, the base year of the GDP series was revised from 2004-05 to 2011-12 and released on January 30, 2015, after adaptation of the sources and methods in line with the System of National Accounts 2008 (2008 SNA).

Notes