MUMBAI: Elections have long been won and lost on economic progress. Chief among them is GDP growth, which pretty much determines the success rate of governments. Come Thursday, growth data for FY18 will be released, but there are enough reasons why you shouldn’t take these estimates at their face value.
One can hear the drumbeats with the NDA government boasting about India’s world-beating 7 per cent growth every year since it took over in May 2014.
But in reality, neither the government nor officials are aware of the actual growth rate under its tenure. That’s because, for any given financial year, the Central Statistics Office (CSO) -- the custodians of GDP data -- issue five estimates spread over nearly three years with the final estimate released after 34 months!
The final revised estimate for FY15, the first full year under NDA’s tenure is not yet out! The second revised estimates are available only for two fiscal years and even these estimates have been off the cuff from that of the initial projections.
While initial estimates pegged growth at 7.4 and 7.6 per cent for FY15 and FY16 respectively, they printed at 6.9 and 7.9 per cent. In fact, an analysis of GDP data between FY05 and FY17 shows that advance and provisional estimates of all 13 financial years were either underestimated or overestimated.
Because an early release of GDP aggregates helps prepare annual budgets, the CSO first publishes what’s called an Advanced Estimate (AE) in January every year. This sets the macroeconomic tone and paints the first overall economic picture. But economists believe AE’s underestimate the current year’s growth as they are extrapolated based on Provisional Estimates (PE) of the previous year. Going by the past 13 year’s data, underestimation is close to 0.5 per cent.
A month later, ie., in February, comes the Revised Estimates (RE), followed by First RE, Second RE and Third RE, which is considered the final estimate of a given year.
The time gap between the AE and First RE is 10 months and the difference is driven by the transition from using high-frequency indicators to actual sectoral data.
The time lag between AE and Second RE is nearly two years and incorporates actual expenditure figures available from budgets and also by replacing the high-frequency IIP data with data from Annual Survey of Industries and others.
The First RE, usually, corrects the underestimation or overestimation of the AE and conveys the direction in which the economy is heading. But analysis shows that there have been significant variations in First and second revised estimates.
Initial estimates of GDP, published in the first quarter, are based on high-frequency indicators like electricity, water and gas, capturing economic activity across different sectors and used as a basis for extrapolation.
Since extrapolated values of previous years don’t depict the current state of the economy, these estimates are revised when actual data is available.
Also, the methodology for computing quarterly GDP estimates is different from annual estimates, making it difficult to compare revisions across different base year series. Unlike in developed nations, quarterly estimates are based on high-frequency indicators and aren’t revised later.
Developed countries also rely on long-time series, which is unavailable in India. Due to the time lag with which the true picture of the economy becomes clearer, the final estimate loses relevance as it’s unavailable when needed for policy decisions.
A sector-wise analysis also shows that revisions at the sub-sector level may be different compared to overall GDP growth rate. The extent of revision in any sector remains unpredictable, but do not lead to major corrections in the overall growth rate.
Sneak peek
For every financial year, CSO issues five estimates spread over three years.
Final revised estimate of economic growth for FY15, the first full year under NDA’s tenure, is not yet out Between FY05-17, advance and provisional estimates of all 13 financial years were either underestimated or overestimated.