6.1 High Growth Continues After GFC

Nagaraj (2013) notes that the growth before the GFC was credit-led. While the credit did expand rapidly (and all high growth requires credit expansion), it was eminently sustainable in terms of both the current account and the fiscal deficit. We also know that in growth acceleration in emerging economies that have been sustained (to take the economies over to complete the economic transformation), has always been accompanied by credit expansion, especially in the ELG economies. It is difficult to see high-speed growth without credit expansion. What is pertinent though if the same was inflationary due to demand exceeding capacity output, then reigning in the same is necessary.

That inflation increased over the period is certain. But whether therefore the growth was inflationary requires deeper analysis which we do later. Essentially, most of the commentators did not distinguish between a supply-side and demand-side inflation, being carried away by the fashion of the day. This was the case with the RBI as well which desperately tried to fight the supply-side inflation just before the GFC arguing that the inflationary expectations had been rising. Later, we present evidence that the data covering core inflation in the CPI was spurious anyway.

6.2 Estimates of the Stimulus

The fiscal stimulus was quick off the ground, because a few months prior to the re-election of the government it had a plan in the form of enhanced MGNREGS allocations and rural infrastructure, to bring about greater inclusiveness. Once the crisis happened, and after re-election,the government added to the expenditures and also gave tax cuts on both service taxes and excise duties, reducing the same by between 20 and 40%.

In an ex post realized sense, the part of the fiscal stimulus that provided the counteraction on account of revenue expenditure increases rose from 8.5% in 2007–08 to 10.7% in 2008–09 and then to 10.8% in 2009–10 and then to 10.5% in 2010–11, before it slowed down to 10.0% in 2011–12. See Fig. 6.6 again. The capital expenditures which had been at a high 2.4% in 2007–08, actually fell to 1.6% in 2008–09 and then rose to 1.7% and to 2.0%. Since the expenditures were on a trajectory of decline, and may have been front loaded coming early in 2007–08, we can say that they moved up by around 0.1% approximately on an ex post basis giving a total central government expenditure thrust on an ex post basis to be 2.25% rise on account of revenue and 0.1% on account of capital as a proportion of the GDP, i.e., to total of 2.35%. Since a multiplier of roughly 1.5 may be assumed over the short period to these expenditures, the ex ante fiscal thrust would have been around 2.35/(1 − 0.04 − 0.0235 * 1.5), i.e., approx. 2.57% of GDP.Footnote 1

Similarly, the tax revenues of the Center fell from 8.8% to 7.1% in 2008–09 and to 7.5% in 2009–10 giving an average decline over the 2 years of 1.5%. This again would have ex ante been about 1.66% giving a combined ex ante thrust of around 4.23% of GDP.

Kumar and Vashisth (2009) estimate direct additional counteraction measures of 2% (consisting of reduction in indirect taxes, spending on infrastructure, and export incentives) in 2008–09 amounting to about US$ 80 b, which had followed MGNREGS and rural expenditures of 4.23% of GDP, giving a 6.23% of GDP. The latter 4.23% were made for political considerations and consisted of loan waivers, additional allocations on MGNREGS, Bharat Nirman and PMGSY, besides fertilizer and electricity subsidies. However, all helped to provide counteraction. If about half the 4.23% and the additional 2% is taken as having helped to counteract from the trajectory post the shock, then these estimates are approximately the same as combined tax and expenditure estimates of 4.23% that we have estimated. Mundle et al. (2011) use an estimate of US $ 4.1 billion as the initial thrust (Rs 20,500 crores at exchange rate of Rs. 50/USD c.2008) to carry out their simulation. This was followed by a second thrust of roughly the same amount (Kannan undated C. 2009).

Kumar et al. (2009) note a direct fiscal stimulus in the first year of the crisis of 1.3% of GDP (para 5.2), which in its view was quite small in comparison to that of other countries, but justified by the starting point of high fiscal deficit. The paper noted correctly that India was already on a slowdown, but assumed it was due to business cycles, rather than to the severe monetary tightening and the allowed appreciation of the rupee that happened in the last year before the GFC.

Kumar and Soumya (2010) writing a little later noted that the government expenditure had gone up even before the GFC due to the elections, which may have come in handy as an expenditure that maintained expenditure pressure even after the negative shock of the GFC. In addition, they note a counteraction amounting to 1.8% of GDP (through three stimulus packages) and an effect of the budgetary expansion of nearly 3% of GDP during 2008–09 and 2009–10.

Kapur and Mohan (2104) estimate that the fiscal stimulus was overdone and was responsible for the growth overshooting to result in inflation. While a marginal overshooting was possible, the inflation as we argue in the next chapter was largely supply side and partly spurious. The problem with the CPI had been pointed out by Dholakia (2018) as well. During the GFC, the growth had crossed 9.75% over at least two quarters. The very sharp rise in the investment rates with gross capital formation rising to over 37% in a matter of 4 years (with a large part of that being on account of the rise in private investments and the savings rate rising) meant that the capacity output could keep pace with the demand determined level of output. Inflation rose largely due to the supply side of rising oil and commodity prices. Post-GFC until 2014, before China slowed down the inflation in India was largely an imported inflation as oil continued to be high and a structural and necessary inflation in the rising food prices. As the supply side eased from 2012 to 2013, the inflation also came down. For a discussion on the specific nature of inflation, see the next chapter.

6.3 Policy Paralysis

We have already seen that the period since 2011–12 had been problematic with slowing down of growth, and with only a short-lived bounce up in 2018 from the demonetization, which was not sustained. Nagaraj (2020) in a quick piece brings out the broad dimensions of the slowdown and also points to demonetization, crony capital, and inadequate public investments.

Why has the growth been low? We list several factors some of which are shocks and some due to policy or even non-response to shocks which could have been easily countered.

The so-called “policy paralysis” was certainly important in having contributed to the slowdown from 2011 to 2012 onwards. Especially impactful was the ban on iron-ore mining and the series of court decisions that increased the uncertainty- for instance, the ban on sand mining. And the Supreme Court's indictment of the coal allocation process. The mess created by the shoddy way in which coal blocks were allocated, could be “resolved” only with the coming of the Modi government in 2014. The CAG pointed to the lack of proper bidding in the allocation of licenses and spectrum for the 2G services. The fairer more transparent and fee maximization with which the 3G auctions were conducted seemed to suggest much loss of revenue to the government, when indeed earlier the policy was to not to allocate the spectrum separately. The so-called “2G Scam” hurt the Congress government, and contributed immensely to the growing uncertainties, since it seemed to suggest that courts and audit organizations could reverse decision of the government. At that time, the sharp decline in output, and especially that of capital formation was discussed exclusively in terms of policy paralysis. The PMEAC highlighted the “policy paralysis” to be the reason for the fall in the growth. However, the sharp tightening which happened just when the fiscal stimulus was being withdrawn was perhaps more important. Since the RBI was responding to a supply inflation, the negative effect on the demand was anticipated then.Footnote 2 When we realize that the rate hikes had started happening as early as April of 2010, the kickdown happened because of the RBI.

6.4 PMEAC’s Analysis

C. Rangarajan, the Chief Economic Adviser to the PMEAC, argued that the sharp rise in the incremental capital output ratio (ICOR) was reflective of the large hold up of assets under construction. He also noted that the growth fell more sharply than what had been caused by the fall in investment. Moreover, “investments did not commensurately translate to output” was the claim. (FeBureau, 2013). However, we know that when investment spending falls then with immediate effect given a large demand multiplier of the order of 2, output should fall increasing the ICOR. The productive aspect of new investment is not in the picture in the short run. However, the report also noted the sharper than expected fall in the core inflation, suggesting room for monetary expansion. [Later, we will see that the earlier rise as well as the fall in 2013 may have been spurious—due to the movement of “housing rentals” rather strangely computed by the CSO]. We know that this was a period of massive tightening though by the RBI even as the stimulus had been pulled back, and further collapse of investments took place. I could anticipate given the ferociousness of the rate hike, to “fight” largely a supply side inflation, and that too in the wake of a stimulus that engendered largely investments in infrastructure which tend to be leveraged that the impact would be sudden and would be very resistant to recovery.Footnote 3

6.5 Monetary Tightening

We must not underestimate the macroeconomic developments especially the non-response or the angular response of the RBI in tightening when the fiscal stimulus of 2019–10 and 2010–11 was withdrawn, as we have already brought out. The continued tightening in the management of shocks whether it was in response to external shocks like the taper tantrum, or the pressures for depreciation of the currency, were perhaps the most important of the factors.

The real interest rates were high, the gap between Indian and global rates were very large. While these large gaps could temporarily attract some FDI and FII investment, there is no spending out of the flows. Net flows are known to depend upon growth differential between the host country and the rest of the world, and spending is not necessarily co-terminus with investment inflows.

Typically, high rates, high differentials and high real interest lower the investment spending by both domestic and foreign businesses. Tightening when due to domestic credit shrinkage as compared to fall in NFA can have differential spending effects as between domestic capital-dependent firms and those having access to foreign flows during times of an otherwise exuberant economy. The entire period up to 2018 has been one of high interest rates (actual) as measured by the corporate bond yields and the MCLR/PLR/Base rates. While the latter are relevant to corporates which have the scale to raise funds in the capital markets, the vast bulk of the MSMEs and even many large firms it is the lending rates of banks with is important.

It is only since 2018 that the RBI brought down rates, but by then, after years of punishment due to inadequate liquidity which had kept interest rates up, had already brought down growth and demand through the fall in capital formation. Renewal of capital formation cannot take place merely on account of a fall in rates. When rates declined, finally, there were no “animal spirits” left, to respond to the lower rates in 2018.

Housing investments which are known to be sensitive to demand conditions and respond to low interest rates (even when expectations of future incomes are not bright, unlike productive investments), could possibly have responded, but the cuts in interest rates were not deep enough. And soon enough, lenders to the real estate sector began to see their loans being rolled over. RBI’s insistence that these be recognized as NPAs, without any attempt to revive housing demand, in the face of large unsold inventory further contributed to the mess.

The NBFCs which lend to smaller units too being badly affected, contributed to the ineffectiveness of low rates in 2018 and thereafter, to be able to raise investment or for that matter any kind of spending.

6.6 Market Rises and Fall

Mr. Modi’s government was ushered in, in 2014 with a huge mandate, and there was near unanimous expectation, almost tantamount to conviction, that the government would go about setting the economy right. Early in June 2014, the government took charge. From early February itself when it was clear that the advantage lay with Mr. Modi, the NIFTY had risen from about 5935 to 7623 before the government was sworn in. In anticipation of the positive actions to help the economy and business, that nearly all believed would come, the market rose further to 8763 by March 2015, at the time the second budget was presented. The overall rise of 47% happened within a year, when there was no marked improvement in the economy on the ground; so large were the positive vibes generated by the Mr. Modi. However, when the anticipated measures to improve the investment climate and the demand conditions did not materialize the market rather reluctantly accepted the situation, and slowly gave way. The belief in Mr. Modi’s intentions and capability to set the economy right continued all through the period, certainly even into the early COVID-19 lockdown phase. In other words, the market has been ever ready to respond positively to the government.

6.7 Positive Vibes But Underspending

Over the 9 months that was the tenure of the first budget, the spending was actually less than what was budgeted! Underspending in democracies is rare, and the matter does require an explanation. The underspending as indicated by the revised estimates when compared to the budget estimates for Non-Plan expenditures presented in the 2015–16, budget for the year 2014–15 (first 9 months of Modi-1) was as much as 1.05%, i.e., Rs. 18,863 crore. But the underspending on Plan Expenditure was by as much as Rs. 112,356 crore which was close to 6.26% of the budget estimates. And on total expenditures, it was Rs. 131,219 crore or 7.31% of the budget estimated of 2014–15. On capital expenditures—largely a part of Plan expenditures it was short by Rs. 30,100 crore, i.e., 1.68% less than the budgeted figures! Capital expenditures are known to have large multiplier effects of well over 2.0 and Plan expenditures of around 1.5 (Bose & Bhanumurthy, 2015). These overall expenditure shortfalls amounted to over 1.04% of GDP of 2014–15 at factor, and with an immediate multiplier effect of 1.5 overall, the hit to GDP growth would have been in the range of 1.5% per annum! It was only in the expenditures of 2015–16 that the budgeted expenditures were reached. See Table 6.1. It was contented by many senior civil servants that the lag in expenditures from budgets and approvals had increased due to the extreme centralization by the Prime Minister's Office (PMO).

Table 6.1 Estimates of revenue and expenditure and actuals as assessed in March 2016, of 2014–15 (Rs. crore)
Full size table

Many mainstream departments despite having the approvals for expenditures nevertheless actually expended only when they were sure that the PMO was fully conscious of these expenditures.

6.8 Unconditional Pursuit of Fiscal Deficit Targets

The commitment to containing the fiscal deficit was almost total and overriding without any nuance conditioning it on the state of the economy. The government may well have been misguided by the growth estimates in this period based on the new GVA 11–12 series which clearly over reported the growth in the period from 2011–12 onwards perhaps till 2017–18. See Morris and Kumari (2019). However, there were other clear indicators that were ignored—the index of industrial production, credit growth, and exports, besides physical transaction indicators that all pointed to a slowdown. Even the new series showed a decline in the rates of gross capital formation to GDP. Yet the stance of the government was that growth must be high! The newly introduced Periodic Employment Survey (PES) was available to the government from 2016 to 2017, but these were not released to the public since the high unemployment and low employment growth in these would have gone against the government’s posturing.Footnote 4 The CMIE’s employment surveys could be dismissed as being “unofficial’, never mind its extensive coverage and high reliability. Perhaps very importantly the belief that the Finance Minister understood it all, prevented any nuanced understanding.

One could not get away from the impression that the government was more interested in scoring points, like in a school debate. There was enough indication that the new series may not be capturing the reality even within the state system. Thus, “… the Economic Survey of 2014–15 cautioned the reader to the use of the new series in computing growth rates and hoped that with some years the issues would be ironed out. It noted that: ‘The upward revision in manufacturing growth in the new series also owes to inclusion of trade carried out by manufacturing companies in the manufacturing sector itself, which was earlier part of the services sector. (p. 5. Economic Survey 2014–15, See Morris and Kumari (2018). The RBI (2015) too noted problems with the new series.

The approach of the government to its budget was dysfunction in being that of a householder to his home budget, where frugality, and tightening during adversity are recommended, and spending within “means” is nearly always appropriate. The government seemed to have attached an unconditional value to reducing the fiscal deficit, and took pride in achieving its fiscal deficit targets. It may have been playing to its own limited perception of what mattered to international rating agencies.Footnote 5

6.9 Scaling Down MGNREGS

The BJP government’s DNA was quite against the MGNREGS. The scaling down of the same without its replacement by any other spending program would have reduced overall spending to put downward pressure on demand. The usual assumption that reduced government spending crowds out private spending, was not true during much of Modi-I since the “animal spirits” were down, despite the hope, and the economy was wilting under declining demand. In the year 2012–13, under the UPA, MGNREGS allocations has been scaled down from 0.43 to 0.383% of GDP. It was scaled down further to 0.26% for 2013–14. And in the first 2 years of he Modi-I it was as low as 0.23–0.26% of GDP. It was only in 2016–17 that it reached some 0.374% at which level it remained until it was scaled down again in 2020–21 (budget estimates).Footnote 6 See Table 6.2.

Table 6.2 Trends in the expenditure on the Mahatma Gandhi National Rural Employment Guarantee Scheme (MNREGS)
Full size table

This scaling down of the MGNREGS which had been scaled up to 0.5% of GDP after the GFC, was one of the important reasons for the slowdown. The aggressive reduction that followed in the last years of UPA and the early years of Modi-I further exacerbated the demand shortfall. By the time it was sought to be revived from 2016 to 2017 onwards, the other expenditures, especially investment had been reduced to near stagnancy.

6.10 Demonetization and Delayed Effects

The demonetization was instrumental in reversing a moderate rise in the growth that had begun sometime in 2016. As we had already seen from the data, the index of industrial production, fell sharply. But then there was a quick rebound (2018Q1), which was also short lived. Demonetization reduced the incomes of many petty producers who had customers with whom they could not have operations on a credit basis when there was little cash in the economy. Thus, vendors in public places like railway stations, busy street corners, etc., had to pull in their businesses. Similarly, those with high transaction intensity using cash were badly affected for a shorter period. The demonetization was known to the government to have been a blunder almost a few days into its announcement, and the government through the RBI went about restoring the cash in the system as quickly as it could.

The RBI without much thought replaced the 500 and 1000 denomination demonetized notes with notes of 2000 denomination, thereby perhaps aiming to optimize the available capacity of impressions per unit of time, ignoring the transaction potential of the system of notes now with 2000. The note of 2000 has poor transaction intensity (velocity of circulation) when without notes of 500 and 1000, since these would be used in exchange when purchases of below 1500 are made with a 2000 note. A mix of 2000, 1000, and 500 would have been transactionally more efficient (Joshi & Mukherjee, 2017), and the need of the day. It was the rise in speed of circulation of 100 notes, thanks to locally emergent systems of collection and circulation by note carriers that saved the days. The net result was that the pain of demonetization was prolonged due to the RBIs naivety.

As argued earlier demonetization affected consumption adversely with a delay of around 12–18 months, in 2019–20. Because of this delay, the NIFTY index of consumption, and the prices of shares of FMCG and white goods companies began to fall only in early 2020. We have already reviewed the pattern through the IIP for consumer goods both durables and non-durables.

6.11 Authorities Enhancing Risks and Uncertainty

The government’s penchant for high centralization, and going beyond the voice of reason or practicality, with all good intentions, as in the case of the demonetization was evident in other episodes as well. Thus, the push towards electric mobility was not nuanced, nor based on any reasonable trajectory. High degree of commitment, in the government’s thinking, meant tougher and harsher measures to push the economy to shift, forget about recognizing any of the nuances that are involved. The Indian automobile sector facing the regulatory jump from Bharat IV directly to Bharat VI in 2020 had responded very well, with vast investments to make the transition on a very hard target of April 2020. Earlier, the Indian oil companies had already invested to produce cleaner fuels meeting Bharat VI requirements. In September 2018, the government announced the move to electric vehicles, while the Bharat VI assets were being created or had yet to be sweated! In August 2019, it announced a series of concessions for electric vehicles 5% GST (normal vehicles had 28% GST), a reduction in direct taxes for those buying electric vehicles up to Rs. 2.5 lakhs. Yet, there was little clarity on the policies or measures required to bring about a charging infrastructure. So despite the incentives no major shift could have taken place.

Later, the government had to announce that it will not ban diesel vehicles since by then loose statements on banning diesel by senior politicians had been picked up by the media. In no country has the shift been accompanied by bans or deadlines. Bans and unrealistic deadlines for obsolescence create far too great a loss of social value. Bans, unlike incentives and taxes which price goods and services, are difficult for markets to value, and result in heightened uncertainty. Investments in the sector which had slowed up except for meeting Bharat VI all but seized up. Even consumers felt the uncertainty, which resulted in delaying or putting off purchases. Government further multiplied the hurt by raising the registration rates by insisting on collecting 3 years tax at one go, which raised the initial cost of ownership quite considerably. In some cases, the hike could be as high as 400%.Footnote 7

Thus, the sharp decline in auto sales from April 2018 and which has continued to date is not difficult to account for. See Figs. 2.8 and 2.9 again. Motor vehicles are an important industry of the manufacturing sector which along with its immediate input and user industries account for some 45% of manufacturing in India. These adverse policies have been to affect demand, except for the bounce up in early 2018 after the demonetization dip.

6.12 Not External Factors

A few commentators have called attention to “external shocks” that pushed growth down. But the period since quantitative easing was put in place by the US and the EU has been one of steady if moderate growth globally. Oil prices have remained quite benign, rising only marginally before the COVID-19 Crisis. The Chinese economy had slowed down after the GFC still grew at 7.5–8.5% till 2014–15, and slowed down thereafter to about 6–6.5%. While the “taper-tantrum” was severe it was entirely on the financial market side, and need not have affected the real economy, had the central banker provided the compensating liquidity, instead of allowing the rates to go up to shockingly high levels. The crisis in global trade by the Trump-Xi trade wars could have affected performance only in 2019 or thereafter.

Neither could the IT and Services sectors be blamed.  Service exports did grow from 2011–12 to 2018–19, at nearly 7% in dollar terms, but not at the very high rates seen during the “Tiger” period. See again Fig. 3.9. Service exports were not so dependent on pricing given the “absolute” advantage that India has in many services especially ITES. Export of goods nearly stagnated over the same period but varied much with some dependence on the REER. Exports of manufactures could easily have been put on a higher growth path with a more aggressive pricing of the rupee. But depreciation has not been on the agenda of either the Government or the Central bank. Instead, as we have seen, they have sought to uphold the exchange rate and prevent its depreciation, through raised interest rates. And even one Commerce Minister seems to think that exchange rates are determined entirely in markets, and hence missed recognizing the instrumentality of the exchange rate.Footnote 8

6.13 Real Estate Construction and Growth

The real estate sector has been in deep trouble. Again here besides the high taxes suddenly brought about by the reform that included the sales in the sector under GST, high interest rates and a series of ‘shock’ regulations have been important. The Real Estate Regulatory Authority/Act (RERA) while well intentioned, and having many positive aspects, nevertheless meant increased capital mobilization on the part of the builders, since now they could not divert funds from project to project to (mis)use client funds. This created a demand for credit which came at a time when the credit was being curtailed. Not all of them had the credibility or equity to borrow to make up for the additional requirement. The problems of slower income growth, higher interest rates which arrested consumer demand, and the liquidity shortage which they faced, made them delay deliveries which further compounded the problem since buyers were unwilling to make purchases without seeing substantial construction on the ground. Inventories increased even as construction slowed down. Buyers had been hit by rising rates over 2011–2015, and when the rates did not fall after 2015 the slowness in income growth had its effect.

The real estate sector in India has large problems of a structural nature that range from absurd regulations on layout, very low regulated FSIs, very little land being released for urbanization especially in metros, problems in land aggregation, requirement of Non-Agricultural Clearance (NAC), non-existence of even a modicum of coordination between layout and land use on the one hand and articulation infrastructure on the other, large risks of title, and so on. (Morris, 2017). Yet the immediate slowdown, and the situation of many of the players reaching illiquid situation can be attributed to the overall slowdown, and the problems with the NBFC and the banking sectors, compounded by the NPAs arising out of the real estate and infrastructure sectors. See also Varma and Morris (2020) who argued for an immediate resolution of the problems in the sector.Footnote 9

6.14 Infrastructure and NPAs of Banks in the Slowdown

Bad lending practices by the public sector is no doubt an important reason for the erosion of their capital and high NPAs. This is well recognized as a problem. However, the reasons for the same as arising out of the lack of autonomy (from government), attempts to micromanage by the government, and an overbearing regulation that removes from them the responsibility to be accountable for performance and risks taken, are not widely recognized as being the core underlying reasons. The political pressures that make them give credit to risky clients favored by the government of the day, as a reason is however well recognized. Perhaps more that the latter the former is important. It is also what makes them vulnerable to political pressures, and therefore it is ultimately the dysfunctional design of their interface with the government and the regulator that is responsible. The RBI in itemizing good practice, prevents the embedding of responsibility. See Morris (2019).

But above all, the role of the slowdown in amplifying the NPAs of the banks, is not generally recognized. NPAs are expected to be countercyclical, rising during periods of recession and slow growth, only to improve later when the growth picks up. The suddenness of the decline in gross capital formation and collapse of construction from 2013 onwards was instrumental in increasing the NPAs. Had the economy recovered quickly perhaps these current very large back breaking NPAs would not have materialized despite the structural (managerial) weaknesses of the PSU Banks. The continued slowdown brought not only PSU banks (2018) but also many NBFCs (2019) and private banks, to their knees. Poor and delayed responses by the RBI and the government to the crisis in real estate, infrastructure, and in the NBFCs both being linked together (as when government chose to ignore the Dewan Housing Finance collapse, and the collapse of the ILFS), were important contributors. Similarly, the government delayed the inevitable requirement to re-capitalize the PSU banks worsening their problems.

Weakness in assessment of loans by PSU banks, may also have been due to the poor financial structuring of PPPs, which allowed interest rate risks to be borne by developers. When there is rise in rates they have an incentive to shift the same the adverse effects of the same on to the PSBs, when government could not be cajoled into making concessions. See Morris (2019) for the details of how private infrastructure capital could play the government and the public sector banks to limit their downside risks, allowing them to take on unwarranted risks as many of them chased rents. Even reputed capital may have been caught in the maelstrom since when some developers could bid aggressively taking on risks others were also forced to do so, or accept being out of the business.

Consider the growth rate in NPAs measured in percent per annum of NPA in nominal terms. See Figs. 6.1 and 6.2. When the economy recovered with high growth rates, the growth rates in NPAs of all Scheduled Commercial Banks, fell from rates close to 10% to about −10% by 2006. The same is true of all banks, though in the case of both Private Sector Banks (PVSBs) and Foreign Banks (FBs), there is considerable volatility. And as the CRR began to be raised in 2006, the rates climbed up again for the PSBs more slowly than for the PVSBs or FBs. This could reflect a slower recognition of NPAs by PSBs relative to the others. But the rates of growth continue to rise, for PSBs even during the period of the fiscal stimulus after the GFC. For PVSBs, there was lull from 2010 to 2014 after which they too began to rise. Thus, the problem of NPAs of banks is very strongly correlated with interest rates and growth, and hence the argument of macroeconomic policy of adverse interest rates and slow  growth having driven NPAs of the banking sector including the private banks (other than foreign private banks) has much support.

Fig. 6.1
figure 1

Growth rate of NPAs (% per annum)

Full size image
Fig. 6.2
figure 2

NPA share of total advances of scheduled commercial banks and growth rate of advances (%, and % per annum resp.)

Full size image

The proportion of NPAs in all assets though would be higher for the weaker PSBs. By 2016–17, the weaknesses in lending and the adverse selection interacting with the structural weaknesses of the banks, pushed the NPA growth to even higher levels of 60%, before falling to about 20% in 2017. With large NPAs, the lending would also break and this is evident in the slow growth of credit, seen earlier. Visit Figs. 3.1 and 3.2 again. We may also see that NPA reductions take place when credit growth is high, and when credit growth has suddenly fallen, especially after 2011–13, the NPA’s as proportion have risen.

6.15 The GST Effect

India has been going through tax reforms, the success and significance of which is generally underestimated. Some key distortions though continue—viz. disparate rates and a most painful (at least to honest taxpayers) tax administration. Reforms brought indirect taxes on the value-added principle and fewer chapters. And now with the GST, the integration of taxes across the state and the center, and across both goods and services is essentially complete.Footnote 10

It is a remarkable political achievement to have made GST a reality. This was especially so since the benefits to the states are quite asymmetric. Being a destination oriented scheme, taxes, accrue to the government of a region based on the (final) consumption in that region. As such production-oriented states (states with much production and not too high a level of consumption—Gujarat, Chhattisgarh, Andhra Pradesh, Tamil Nadu) would stand to lose much. Gujarat had opposed the GST and was asking for a revenue neutral rate of 23% which would have (at least in an accountant’s calculation where feedback effects on demand are not considered) seemingly protected its current revenues.Footnote 11 There was commitment of the center to protect the revenues of losing states like Gujarat—essentially ensuring a revenue growth of 15% annually. The shortfalls below this trajectory were to be made good by the center over the first 5 years, had already been made. But it was clear that Gujarat would lose after this period.Footnote 12 It was only once Mr. Modi became prime minister that Gujarat’s interest could be kept aside for the larger gain of the country. The much feared inflationary impact proved a damp squib, since the GST rates mimicked the existing rates with very little change from earlier on a product by product basis.Footnote 13

Therefore, the opportunity to bring about fewer chapters and rationalization was missed. But this was not such a debility as most think since in the future convergence to fewer rates is always possible. The big achievement was to get the states to agree.

The GST as was widely anticipated would reduce the avenues for avoidance, since compliance was built into the process, and paying entities would demand GST-paid certificates from their input suppliers in order to get the input tax credit. In any case, one leg of any transaction including a good or service produced or bought by a GST-registered entity would be captured even if the other entity is not registered, thus increasing the probability of capture of inter entity transactions and hence improving compliance. Improved compliance in a sudden and significant way, is equivalent to an increase in the effective tax rate. And any increase in the tax rate has the effect of reducing demand. The point is that the GST should have been implemented with a somewhat reduced rate if this increased compliance effect could have been worked out ex ante, i.e., the effective rate should not have unwittingly increased. Else the GST should have been implemented with other counteracting measures on demand. Such a move of combining structural reform with macroeconomic policy in a positive way is necessary for the reform to be successful, and for growth to be maintained if not to improve. The merit of the GL of 1991–92 and 1992–93 was that it combined in a positive way structural reform with the stabilization specifically demand management to result in an increase in the growth rate to 6.7% from the pre-reform growth of about 5.5% after a dip over just about a year due to the pull back in dysfunctional public spending. Similarly, the design of the NHDP resulted in the economy being kick started from 2003 to 2004 after it had entered into a slow growth period from 1997 to 1998 onwards. It is not clear whether this was actually intended.

GST rates as much as the combined excise and sales tax rates earlier (CENVAT and VAT together prior to the implementation of the GST) can hardly be said to be in any way optimal. Many rates which are high (28% and some of the 18% rates) could actually be higher than the even the revenue maximizing rates! Short-run price elasticities being low or negligible may have acted to obfuscate this issue, for a government whose senior officials are forever fire-fighting.Footnote 14 There is an opportunity for the government to actually lower taxes in some major goods like automobiles (especially two wheelers) which can actually be revenue enhancing. However, it is unlikely, given the accountant’s approach to taxes that the government would see this opportunity. The GST Council has approached the problem of taxation without regard to price elasticities. It has not yet commissioned any studies to assess the price elasticity of demand. Assumptions of what is luxury, and what the “common man’s need” are, have driven rates. There is another perhaps unintended consequence of the GST and the instrumentality of the GST Council. An across the board tax cut in indirect taxes would be difficult since now the GST Council would have to agree. An ability to make sudden cuts or increases is necessary for macroeconomic management of demand when the economy is hit by large demand shocks.Footnote 15 Today the only way would be for the center to reduce its rates, which would then destroy the neatness of half the collections being to the center, and also affect businesses since they will have to worry about two distinct rates.

6.16 Tax “Terrorism”

When the NDA/BJP came to power, it was aware of the uniquely Indian phenomenon of “tax terrorism”, an aspect of dysfunctional governance that had been going on for many years ever since the tax officers has been made accountable for collections fixed at the level of the finance minister. It is the illegitimate demand for taxes made by the tax department (typically made at the senior levels but not limited to them). The Central Board of Direct Taxes at the highest levels face demands for a certain tax target that is usually quite large in relation of the previous year’s collections, typically by over 15%. If the nominal growth of the economy is around 12% or more, the target is not too difficult to realize. But when GDP has slowed down as it did from 2011 to 2012 onwards, targets of 15% were far too onerous. The CBDT instead of going back to the finance minister with realistic targets would be cowed down to accept these targets, however, irrational they were. They would then get about collecting the same from the PSUs, especially the oil companies and other large entities like SBI, BHEL by sending them illegitimate tax notices, typically around February to meet the year end targets. Then of course the victim PSUs would file cases, which the tax authorities would not seriously contest, allowing the PSU to get back much (but not all) of the extra tax payments it had made. The dead weight losses in this uniquely Indian “tamasha” are not inconsiderable.Footnote 16 The cycle would of course repeat itself, the demands becoming onerous particularly in a year of slow growth in the nominal GDP. Mr. Arun Jaitly as soon as he assumed office as Finance Minister, promised to put a stop to the same. But the demands for direct tax collections continued to be very high while the economy had slowed down and the official growth figures were not bringing out the extent of the slowdown fully. The tax officers with continuing tax demands on a slow economy, now extended the “terrorism” to even the private corporate sector, and started raking up old cases. The damage done to “animal spirits” was considerable, and the impression of the Indian government being parasitic and arbitrary could only have been strengthened, never mind the expected “business friendliness of the government”. Similar, behavior may have begun in the indirect tax departments as the GST targets of 15% increase have become absurdly high in an economy with a nominal GDP growth of under 10% during the last year and half.

6.17 Taxes in the Growth Experience

From Fig. 6.3, notice that the growth in tax revenues at a high 20% during the “Tiger” Period, when the nominal growth of GDP would have been around 13.5%. (Real growth of 8.5% and core inflation of no more than 5%). Yet the taxes grew faster showing an elasticity of nearly 1.5%, from around 2003 to 2004 to the eve of the GFC. Once growth was restored from 2009 to 2010, taxes may have grown at 10% toll about early 2013, after which it fee marginally to rise to a growth of around 11–12% during the bounce back the demonetization; and had then fallen to about 3% from about mid-2018 onwards when growth fell off. The taxes here include all—both direct and indirect—taxes, and as expected bear a strong relationship with the broader patterns of growth. The period immediately following the GFC, though it had high growth over 2 years, shows a lower tax growth, since this growth was not so much due to private spending as much as due to public investments and spending. Yet we do see a rise even if very briefly centered around 2010–11.

Fig. 6.3
figure 3

Growth in net tax revenue (central government receipts) % per annum at current prices

Full size image

The question of whether or not GST affected the demand is an interesting one. Earlier we had suggested that conceptually the GST, when rates are unchanged and morphed from the earlier service, excise and VAT rates to the GST without much change in the overall incidence, then prima facie there is no GST effect is to be expected. But GST also meant vastly improved compliance given its design, the GSTIN backbone, with its all India basis. Essentially, even if one leg of any payment/purchase came for a GST-registered entity, the other leg whether or not registered would be revealed and splitting the entities would not help as well, since there were rather tight limits on the sales of the entities that could avoid registration. Data, covering major items, in a way that we can build series over a time period that covers the GST period as well as the period before is not available. Adding excise, service and GST taxes of the center, though we can get a consistent time series of the tax collected by the center on the same base/s. We have taken the moving average over four quarters of such figures and divided the same by the four quarter moving average of the GDPMP (spliced series at 2011–12 that chains with the 2004–05 series). This ratio is reported in Fig. 6.4.

Fig. 6.4
figure 4

Ratio of quarterly moving average of Excise + Service taxex or GST to GDPMP

Full size image

Observe that as the massive fiscal stimulus of 3% (exFootnote 17post) of GDP was put in place from 2008 onwards, which continued to 2011–12 in part, the ratio fell from 3.6 to 2.6% and then rose to about 3.1% and continued to rise over 2015 and 2016 as the economy had slowed down. But the sharp rise from mid-2016 onwards which coincides with the introduction of GST provides support to the expectation that the GST (even though there was little change in rates) brought greater compliance and higher “effective” rates. The tax pressures may also have been compounded by the effects of the demonetization which certainly brought down the denominator in the measure. The sharp but short-lived bounce of the economy in 2018 results in a fall in the ratio, a rise back from 2018 onwards. Figure 6.5 computes the YoY on quarterly data of tax collected by the central government on the same items—excise, service, and GST (Fig. 6.6).

Fig. 6.5
figure 5

Growth rate in production taxes (accruing to centre) Excise + Service + GST (since July 2016)

Full size image
Fig. 6.6
figure 6

Central government revenues and expenditures as % of GDPMP at current prices

Full size image

Both the increased “diligence” and the tax “terrorism” mentioned earlier, besides the natural step up in compliance from a shift to GST would most certainly have increased tax collections in relation to GDP to result in downward pressures on demand, and hence a slowdown. This is unfortunate and could have been avoided by reducing the GST rates especially on those goods and services (given their high price elasticity of demand) that would have had rates (and hence prices) well above their revenue maximization rates.

6.18 Government Expenditure

Government expenditure rose as the stimulus was put in place to overcome the demand shock posed by the expected fall in exports following the GFC. The Central government’s expenditure rose from 8.5% of GDPMP to as much as 11.0% in 2009–10 and 2010–11, before it began to fall. Capital expenditure in contrast went up by less than 0.5% and may have remained stagnant thereafter. From 2011 to 2012, capital expenditure was marginally lower in relation to GDP. Despite the stimulus having been substantially scaled down by 2011–12, the revenue expenditure by GDP continued to be high. In 2014–15, it reached the pre-crisis level. However, this slowness in the fall is more because of the denominator, i.e., the GDPMP falling, since the monetary side was tight as we have already seen. In a major strategic thrust to the demand, a much greater thrust to capital expenditures could have been given. That it was based largely on account to revenue expenditure increases would have reduced its efficacy. Recall that the start of the “Tiger” Period was due to the rise in expenditures brought about by the NHDP and the PMGSY. Since then, there have been no such transformative second-generation reforms that have had positive demand- and supply-side effects.

From Table 6.3, see a rise in the share of Overall Combined Government Expenditure prior to 2003–04 which provided the fiscal kick –both through revenue and capital expenditure. Once growth picked up the ratio declined. It rose again in the wake of the GFC as the stimulus was put in place. When the stimulus was given up it, however, fell very little since the economy had also slowed down thanks to the monetary tightening. From 2015 to 2016 onwards, government expenditure has been at higher levels, particularly in 2018–19 and 2019–20. The share of the government in capital expenditure rose, because from 2015 to 2016 onwards, despite the government’s fiscal conservatism, the overall capital formation had fallen so significantly that the government was doing an increasing part of the same. The growth rates in government expenditure confirms these observations. After the GFC as the stimulus was active, government expenditure increased in its rate, to slowdown since then with a bump up in 2016–17 and 2018–19 in a weak way. The weak pick-up in capital expenditure (earlier it had short up during the “Tiger” period) again in 2016–17 whose growth effects fizzled out may be seen. The fiscal counteraction in 2010–11 and 2011–12 was less due to capital expenditure. See Fig. 6.7.

Table 6.3 Ratio of combined expenditure of central and state government to GVA and of capital expenditure to gross capital formation (all at current prices %)
Full size table
Fig. 6.7
figure 7

Expenditure growth rates centre and states combined CAGR %

Full size image

A good package to overcome any one of the many public health or urban problems could have provided the country with the opportunity for fiscal stimulus all through 2011–12 to the present. It is not surprising that the NIPFP finds the capital expenditure multiplier to be as high as 2.88 immediately and over 4 in the long run.

6.19 The Fiscal Deficit

The fiscal deficit reduced rapidly over the “Tiger” period from around 6% it had reached around the end of the slow growth period (1997–98 to 2002–03) to as low as 2.5% on the eve of the GFC. Since the GR, the fiscal deficit has always fallen during periods of high growth, and high growth more than cuts in government expenditure had been important. However, from 2013 to 2014 onwards and including the Modi-I, despite the slowness of growth, fiscal deficit targets were pursued with a vehemence that may have contributed to the slowdown itself. The fiscal deficit had soared during the first 3 years since the GFC when the stimulus was given to the economy. The primary deficit closely following the same. See Fig. 6.8.

Fig. 6.8
figure 8

Fiscal and primary deficits as % of GDPMP

Full size image

Since the GL, it is quite remarkable that the center had in fact during the peak of the “Tiger” period brought down the same to actually achieve a surplus of 0.2 and 0.9% of GDP during 2006–07 and 2007–08, just before the GFC. This points to the great opportunity provided by rapid growth (based on private investments and exports) to overcome structural problems. There is little basis to the critique that the high growth was dysfunctional being credit-led and inflationary in this view.