A reality check

Ex-CEA makes a strong case for re-examining growth numbers

Recent research work from former Chief Economic Advisor Arvind Subramanian, which has now been published as a Harvard University Centre for International Development working paper, has reignited the debate over the quality and credibility of Indian data. In this paper, Dr Subramanian examines the so-called “new series” of data regarding the Indian economy and compares its estimates for gross domestic product with those under the previous system, which also had a different and earlier base year for calculating real GDP. He discovered that the new series of GDP was much more disconnected from other indicators of the economy than the old series, and that this level of disconnection, in fact, rendered India an outlier among comparable economies. The paper presented a difference-in-difference regression comparing the old and new series and data from peer economies, and estimated that GDP growth might, in fact, be several percentage points lower than estimated. It is important to note that the former CEA did not use either the “back series” recently released by the NITI Aayog for growth in the 2000s or the revision of the new series for recent growth, and argued that both of those would, in fact, further overestimate recent growth. The years affected cover both the National Democratic Alliance and United Progressive Alliance governments, so it is unfair to impute a political motive to the research. 

The specific estimates presented for the amount by which growth has been overestimated are arguable. But whether real GDP growth is 4.5 per cent or higher is not the point. The paper merely shows systematically what was already known or suspected, that the official GDP growth numbers are out of line with independent economic variables like exports and freight movement. Dr Subramanian had, in fact, made that point even when he was the CEA, arguing that it was rare to see growth in an economy where exports and credit and investment were stagnant or decreasing. This has serious implications for both investors and policy-makers, and should no longer be dismissed or denied. Officials have argued that the new series follows internationally accepted statistical methods. This may well be true but does not address the core questions of comparability and reliability. Whatever the process or method being utilised, the end results have to be subjected to a reality check — and it is clear that the current growth methods do not pass that test. 

An independent re-examination of the methodology and results being used for official statistics is the only way out if their reliability and credibility is to be restored. The Economic Advisory Council to the prime minister has termed parts of Dr Subramanian’s research report as a “most unusual exercise” and said that it would issue a “point-by-point rebuttal.” This may well be a valuable contribution to the academic debate, but it cannot be an adequate response. Dr Subramanian has provided a strong case for a re-examination of the new GDP estimation process as well as the high-growth assumptions guiding policymaking over the years. Thus, a proper response would be to accept that a cloud hangs over Indian growth data which will only be dispelled by greater transparency and investigation of its disconnect with reality.