The credit cycle, the NPAs and capacity utilisation point to a decelerating economy
In May 2014, the NDA government inherited a difficult economy with slowing growth, high inflation and rising NPAs in the banking system. It had barely managed to wriggle out of a difficult balance-of-payments situation triggered by the “taper tantrum” in May 2013. Prime minister Modi had a historical mandate and an economic backdrop to seize the opportunity for recreating another ”1991 moment” or bring in “Thatcherian reforms” to lift the economy on a higher growth path. Instead, the new government chose to focus upon faster decision-making by wiring in bureaucrats in the line ministries to the prime minister’s office (PMO) as an antidote to the ‘policy paralysis’ reflected in multiple ministerial committees of the previous UPA government.
But the emphasis upon speed and transparency met with little success in reviving private investment. That could go down in history as the regime’s biggest failure. Two of its important reforms, i.e, insolvency and bankruptcy code (IBC) and goods and services tax (GST) were rolled out only in the second half of its tenure; the benefits will accrue only in the years ahead. Yet, the claims to economic success in terms of GDP growth and job creation are extraordinary and possibly unparalleled in India’s economic history. These have been equally contested by critics—such counter-claims have recently degenerated into accusations the government has either manipulated data or suppressed it to buttress its own claims.
This, therefore, is an attempt to recapitulate the claims and counter-claims about the important macroeconomic parameters and flag concerns that could cloud future policy space. At the heart of the controversy are the CSO’s new national accounts estimates that show the economy grew a healthy average 7.6% during the 5 years of the NDA government—the fastest growing major economy in the world. But the robustness of these estimates was questioned from different aspects of inconsistency, such as: –
– What explains the acceleration in GDP growth since 2013-14 (last year of UPA-2 government) in a macroeconomic setting of increased monetary tightening, rising real interest rates and fiscal consolidation across governments? And if we were to add a mere 1.4% average annual export growth for these five years, the macro puzzle deepens further.
– If one did a health check of these three critical indicators to assess the stage of the business cycle the economy was operating at—the credit cycle, size and direction of non-performing assets in banks and capacity utilisation of industry—an unambiguous conclusion should be that of a decelerating economy.
– Finally, in the Phillip’s curve analogy, falling inflation, subdued wage rate growth and rising unemployment should have confirmed the economy was in deep trouble and operating way below its potential.
The question is how has the CSO managed to produce such astounding GDP estimates against this overwhelming macroeconomic evidence? The immediate needle of suspicion was pointed at the new MCA-21 database used for the rebased national accounts series (2011-12) published since January 31, 2015.
This large database with information about nearly 5.58 lakh active and registered private corporations is disturbingly becoming a black-box with each passing year as it shows higher value addition with each revision round—first, second and third—as more returns are filed for the respective years. If the process of uploading returns by companies is random, it is a mystery why growth estimates should exhibit a directional bias? In other words, why should a GVA estimate based upon a sample size of one lakh firms be so significantly different than a sample size of, say, two lakhs from the same population of 9.87 lakh firms! If it is, there is a problem.
From a macroeconomic standpoint, such estimates have raised a more pertinent question: Why is the periphery (i.e, the majority of 5.25 lakh unlisted, smaller companies) consistently outperforming the core (the 2,000-odd listed firms tracked by RBI which was the primary database until the new series) in terms of growth rates since 2013-14? It defies logic that in the last six years, smaller firms have become so much more efficient relative to larger ones and possibly grown in size at a time when cheap Chinese imports are flooding Indian markets!
As the MCA-21 database is not readily available to researchers, no one has been able to offer any conclusive commentary. The professor Goldar-led subcommittee report that recommended using the MCA-21 data had found only marginal errors in its validation exercise of 500 companies out of a total of 31,636 filing returns in the extensible business reporting language (XBRL). The report was careful to record, “however, for unlisted companies, no alternative information is available in the public domain. Hence, any kind of validation exercise is not feasible there” (page 14).
The report strongly recommended that the ministry of corporate affairs should evolve a validation system to ensure the accuracy of online data reporting through MCA-21, but we simply do not know if MCA has a robust data validation mechanism, especially for the large number of unlisted smaller companies, and what their share in total value addition is. Further, if such estimates are being used to blow up data for the remaining 4.29 lakh odd companies that don’t file their returns, it is not certain that such a large sample of unvalidated data is necessarily a robust source for a prudent statistical exercise and inferences about GDP!
What about ASI data that also showed a sharp acceleration in manufacturing GVA growth? Critics have underlined the fact that CSO uses a single deflator for estimating GVA in manufacturing activities which created an upward bias in a context of significant divergence in input-output prices from 2014-15. Similarly, it is possibly using inappropriate sectoral deflators for several service activities, which created an unintended upward bias in the real GVA estimates for years in which divergence between CPI and WPI inflation rates widened. The CSO has never acknowledged this, nor explained why it is not using double-deflation in estimating manufacturing GVA.
What is undeniable though is that both WPI/CPI are inappropriate price deflators. It is a sad commentary that both the industry ministry and CSO have been unable to compile sectoral producer price indices (PPIs) for manufacturing and services; in theory, PPIs are the most appropriate price deflators used internationally. The Report of the Working Group on Producer Price Index, August 2017, led by professor Goldar, stated that PPI is regarded as being conceptually more consistent with the System of National Accounts (SNA) as a deflator; it recommended the government compile PPIs with base year of 2011-12. This would certainly remove such anomalies in real GVA estimates. It is hoped the government is making utmost, urgent efforts to compute these PPIs!
In the last couple of years, doubts have also been raised about the estimation of employment and unemployment statistics. The only but significant difference is that the ‘doubter’ in this case is the government itself! It claims that if growth is accelerating then jobs must have been created, offering in support an additive version of ‘job counting’ that appears the least scientific. The critics counter by saying that if employment has declined then growth is being overestimated!
Between these claims and counter-claims, very few have cared about potential policy errors. If growth is not as robust as new GDP estimates show, then we are missing out on appropriate monetary-fiscal policy responses. In the same breath, if unemployment is not as stark as the (not yet published) NSSO survey hints, then the government could end up being burdened with unsustainable social expenditures. Hope the new government will invest in an objective assessment of the ground reality!
-The author is New Delhi-based macroeconomist