Missing the big picture: Arvind Subramanian and Josh Felman on the “Great Slowdown”
In a working paper winning much attention, Arvind Subramanian and Josh Felman correctly recognise the growth slowdown afflicting the Indian economy, but fail to see it as the result of a deep structural crisis worsened by neoliberal reform that constrains demand. Attributing the slowdown to external factors and an unintended consequence of demonetisation, they recommend resolving the bad debt problem while persisting with the neoliberal framework. That would reward defaulters and big business interests, but would not prevent the crisis from persisting or recurring, argues C. P. Chandrasekhar.
In the latest of many critical interventions since he demitted office as Chief Economic Advisor to the Government of India, Arvind Subramanian (in collaboration with Josh Felman, together hereafter referred to AS-JF) has argued that the recent sharp deceleration in India’s growth is not an ordinary slowdown, but “India’s Great Slowdown, where the economy seems headed for the intensive care unit.” The paper that makes this assessment goes on to argue that most existing analyses of the slowdown have not got it right, and claims to ‘make a contribution’ by offering “a different diagnosis of the problem” and providing “a prescription, identifying a path that could lead the economy out of the current slowdown”.
Balance sheet deficits
To summarise the diagnosis provided by AS-JF, the current crisis is the result of a Balance Sheet Crisis that arrived in two waves of twin deficits in the balance sheets of the corporate sector, the banks, non-bank finance companies, and the real estate companies. The resulting balance sheet crisis has frozen the credit pipe, and adversely affected investment and consumption leading to a growth collapse.
The first of the twin balance sheet deficits is seen as having been triggered by Global Financial Crisis (GFC) of 2008, when the world economy slowed and “the infrastructure projects started during India’s infrastructure boom of the mid-2000s began to go sour”. The consequences were that infrastructure companies found their earnings were far from adequate to meet their interest and amortisation payment commitments on loans they had taken from the banks to finance investment, and the banks were soon saddled with non-performing assets (NPAs) that were not fully recognised. Credit to the corporate sector shrank and together with poor export performance due to the global slowdown, ensured that “investment and exports, the two engines propelling rapid growth” (during the 2003 to 2008 period) began to sputter. The infrastructure companies were burdened with losses and unpaid debts and the banks that had lent hugely to infrastructure began to see a sharp increase in their NPAs.
Growth should have remained low in the aftermath of these developments, but actually registered a rebound after the global financial crisis. This does not trouble AS-JF, because they find other fortuitous factors that they see as having countered the underlying trend towards a slowdown. One was a global revival that pulled up India’s exports; another was the sharp fall in international oil prices which “boosted growth by 1 to 112 percentage points during 2015-17”; a third was a hidden fiscal stimulus resulting from expenditures financed with off-budget debt recorded in the books of the Food Corporation of India and National Highways Authority of India and not reflected in the fiscal deficit; and finally an “unexpected credit stimulus” when demonetization-induced deposits with banks and mutual funds were lent to NBFCs, which then lent the funds to the real estate and housing sectors and for purchases of durables, especially automobiles. In sum, “the structural underpinnings of India’s growth were damaged after the GFC, but the economy was buoyed by cyclical factors, most notably the large fiscal-cum-credit stimulus in 2017 and 2018.”
This growth, AS-JF argue, could not last because the credit stimulus was unsustainable and was discovered to be so after the shock of the ILFS collapse in September 2018. ‘Markets’ woke up and began to reassess the NBFC sector to find that “much of the NBFC lending had been channelled to one particular sector, real estate. And that sector itself was in a precarious situation.” This was the second of the twin balance sheet deficits: that of the NBFCs and the real estate companies. It led to a second credit squeeze as mutual funds withdrew their investments in NBFC paper and banks held back on further lending to the NBFC sector, especially as loans to the sector were a significant proportion of their loan books. This adversely affected real estate and the automobile sectors, triggering the Great Slowdown.
Demand-constrained growth and the 2003 revival
Before turning to the policy arguments that flow from this understanding of the factors underlying the boom during the pre-GFC period, the reasonable growth after the recovery from the GFC-induced downturn, and the more recent bust, it is necessary to assess that understanding, which is wanting in many respects. Prior to the post-2003 period, the Indian economy had moved to a higher growth trajectory during the 1980s and 1990s because of higher government spending financed largely with borrowing. This was possible because increased external commercial borrowing and post-liberalisation access to foreign finance allowed resorting to imports to prevent any rise in domestic prices resulting from excess demand in sectors where domestic production was supply-constrained. Debt-financed public spending boosted demand and allowed growth higher than during the late-1960s and 1970s. However, not only was this 6-per cent or so growth rate not seen as adequate but by the turn of the millennium there were signs of the economy once again slowing down. Surprisingly, it was in this background that the government decided to adopt the Fiscal Responsibility and Budget Management Act in 2003, which required it to adopt a conservative fiscal stance and reduce the fiscal deficit to 3 per cent over time. It was then that the post-liberalisation contraction in government spending began, aggravated by an absence of buoyancy in tax revenues. When combined with the weakening of the growth stimulus by the end-1990s, this would have led to a severe slowdown in growth, because of serious demand constraints.
Given this context, the fact that India experienced an unprecedented boom in the years after 2003 is indeed puzzling. AS-JF provide an explanation that is not altogether convincing. In that period, they argue, “there was a major global upswing, with the world economy expanding at its fastest pace in forty years”, and “as this occurred, firms and consumers became convinced that India’s time had come, that the long-awaited payoff from the post-1991 reforms had arrived, and the country was bound to be the next China, growing at 10 per cent per annum for the next several decades.”
In their view: “This combination proved a powerful tonic. As global demand surged, India’s exports boomed, growing at their fastest rates since independence. And as confidence in the future increased, investment – particularly in infrastructure – soared, climbing by 11 percentage points of GDP within 4 years, reaching an unprecedented 38 per cent of GDP. Accompanying this investment boom was an extraordinary expansion of credit: within three years non-food credit doubled, while in 2007/08 alone capital inflows exceeded 9 per cent of GDP.”
This explanation for perhaps the best growth years in post-Independence history, which attributes the boom to exports and the confidence it generated, is, to say the least, unconvincing. India is known if anything for its failure to win a share of global exports of goods (not services) and turn exports into a driver of growth as some east Asia economies did. Manufactured exports have remained small relative to GDP, and net exports of goods have been consistently negative. Moreover, if exports are to stimulate investment they would in the first instance be in sectors linked to those exports, and not in infrastructure, which is an area into which private investors would not be keen to enter given their long gestation lags and low returns, especially if other investment alternatives exist.
An alternative explanation
It is indeed true that investment rose in those boom years and a chunk of it went to infrastructure. But the AS-LF take does not seem a plausible explanation for that trend. An alternative explanation must start with the sentence that ends the explanation from AS-JF quoted above, in which they refer to the boom in both credit and capital flows. Besides the unexpected acceleration in growth, the other distinguishing feature of the years after 2003 was a sharp rise in cross-border capital inflows into India, resulting from a global surge in cross-border flows and India becoming a favoured destination for foreign financial investors, encouraged by a major concession in the form of an abolition of the long term capital gains tax. Total foreign investment flows rose from $4.2 billion in 2002-03 to $13.7 billion in 2003-04, fluctuated between $13 billion and $15.5 billion in the following three years, only to spike to $42.3 billion in 2007-08. This large infusion of dollar liquidity into the system had as its counterpart two trends. First, a direct increase in rupee liquidity when the dollar flows were converted into rupees and spent within the country. Second, pressure on the Reserve Bank of India to buy up dollars to prevent excessive appreciation of the Indian rupee, as that would hurt India’s export competitiveness. The increase in the foreign exchange assets of the central bank had to be matched by an increase in its liabilities, which meant an increase in money in circulation. The increased liquidity had to settle as deposits in the banking system, resulting in a sharp increase in the ratio of bank deposits to GDP.
Since banks cannot accept deposits on which they pay some interest, the banking system was under pressure to lend and invest to cover the costs of capital and the costs of intermediation, besides earning some profit. The magnitude of that pressure can be gauged from the fact that the ratio of outstanding advances of the scheduled commercial banks to GDP, which fluctuated in the 20-22 per cent range during the 1990s and early 2000s, rose sharply after 2003 to touch more than 50 per cent. It should be clear that a credit surge of that kind cannot be ensured without entering into new areas. One set of areas banks chose during those years fell under the broad rubric of retail lending to individuals, for purposes varying from investing in housing to buying automobiles and durables to financing educational expenses. Even here not all the lending could be done by the banks directly. Instead, banks lent to or invested in the paper issued by the non-bank financial companies, who then executed a part of the lending that made up the credit boom, by lending for housing investments not just by individuals but by real estate companies attempting to profiteer from the housing boom that the credit surge had triggered.
But hugely enhanced lending to the retail segment was inadequate to exhaust all the liquidity in the system. This context led to an unusual coalescence of interests of the government, on the one hand, and the banking sector on the other, especially the public banks that the state-owned. Trapped in its own fiscal conservatism, the government found itself unable to finance the large infrastructural investments it saw as needed to drive growth in the economy. Financial liberalisation had also shut down for all practical purposes the development finance institutions which earlier could have been mandated to fund large, capital intensive infrastructural projects. They had either been wound up or converted into commercial banks like the ICICI Bank and the IDBI bank.
Some new source of financing for infrastructure was called for if the private sector was to be enticed into making up for the reduced presence of a cash-strapped government in that area. The money was with the banks, but commercial banks mobilising short term capital from depositors wanting liquidity, had kept away from investing in long gestation and highly illiquid infrastructure projects, because the maturity and liquidity mismatches would make them vulnerable to default. Despite this feature, after 2003, banks were somehow persuaded into diverting their excess liquidity to the infrastructural sector. The share of infrastructure in commercial banks’ lending to industry rose from less than 5 per cent to more than 30 per cent in the years after 2003. Funds were provided to steel and power plants, for roads and ports, for civil aviation, and even luxury hotels treated as ‘tourism infrastructure’. Possibly, the fact that the government was keen on organising funding for private and public-private -partnership projects in the infrastructural sector, sent out a message that there was some kind of sovereign guarantee associated with lending to the area. The government was perhaps expected to ensure the profitability of the infrastructure projects and, failing that, to ensure bank loans were serviced or compensate the banks for any losses they may suffer.
In practice, that did not happen. To start with infrastructural investments were decided upon and implemented without tying up complementary requirements (such as coal supplies for power plants), obtaining environmental clearances, and properly assessing the market, to check whether the demand needed to render projects viable would be forthcoming at the prices they would charge. Projects did not earn the revenues to meet their interest payment commitments and begin repaying their loans, resulting in the accumulation of non-performing loans on the books of banks.
The NPA problem and flow of credit
Banks burdened with non-performing assets suffer losses if they are recognised and provisions made to absorb them. The strain on profitability this results in and the fear that new lending would increase NPAs, would in normal circumstances lead to a decline in credit flow. This, however, did not happen immediately for two reasons. First, after the GFC, which for a short time hit lending in India quite severely, governments and central banks in the developed countries turned to “unconventional monetary policies”, which involved huge injections of liquidity at extremely low-interest rates into the system. “Quantitative easing” and low (and even negative) interest rates became the cornerstones of monetary policy. One consequence was that, after the GFC shock, cross border flows of capital rose sharply, including to emerging markets, as investors borrowed cheaply at home and invested abroad for high returns. India was a major beneficiary, with foreign investment flows rising from a low of $8.3 billion in GFC-affected year 2007-08 to $50.3 billion in 2008-09. Over the decade ending 2018-19 such inflows fell short of $30 billion in only one, exceeded $50 billion in three, and touched $73.4 billion in 2014-15. This huge external injection of liquidity only increased the pressure on the banking system to sustain high levels of lending, at a time when non-performing assets were increasing. This delayed the response in terms of reduced lending that should have accompanied the increase in NPAs.
The second reason for the delayed lending contraction was that, realising that recognising as NPAs the loans provided to poorly performing, capital intensive projects would challenge their solvency, the banks chose to restructure bad assets, in the hope that this would help stressed borrowers to return to profitability and begin meeting their debt-service commitments. The restructuring involved some combination of interest rate reduction, loan term extension, conversion of a part of the debt into equity and provision of additional funding. The regulator helped this process by allowing such restructured assets to be designated as “restructured standard assets”, rather than identifying them as bad assets that invite provisioning for default. This too postponed the expected lending response, till as late as December 2015, when the then Reserve Bank of India governor initiated an Asset Quality Review, with new guidelines for recognition of bad assets. Prior to this, in April 2015, the practice by which debts restructured under the Corporate Debt Restructuring scheme could be treated as standard with a small provision of 5 per cent was brought to an end. Banks had to provide for 15 per cent once the debt is restructured. But after the asset quality review was initiated in December, many of these debts could not even be treated as restructured debts but had to be recognised as non-performing.
The impact was dramatic. In 2015-16, public sector banks recorded an 87 per cent increase in provisions and suffered losses totalling close to Rs. 18,000 crore. That was only the beginning, as the volume of NPAs and the NPA ratios rose sharply in the following years. The credit bubble had finally to give, in the absence of which both investment and consumption demand slumped, as investments in housing, purchases of automobiles and durables, and capital spending plunged. This fall was all the more severe because of the crisis that overwhelmed the NBFC sector, partly because projects that they had lent too did not earn the expected returns, but also because of a “liquidity squeeze”. NBFCs were borrowing short term and lending long term, so that returns on their lending could not clear all of their short term loans when they fell due. They needed to keep renewing those loans and obtain new loans to sustain and expand their business. Because of the liquidity problem resulting from the banks turning tight-fisted, firms like DHFL were squeezed out of the market for credit, and were unable to secure the funds to meet payments commitments and keep business going. This problem was so severe that, DHFL, burdened also by other weaknesses, collapsed. NBFCs were important intermediaries in reaching credit to small business, aspiring homeowners and household wanting to make big-ticket purchases. When they faced trouble, credit for these purposes dried up.
Neoliberal reform and the boom-bust cycle
This perspective on the boom-and-bust cycle identifies neoliberal reform as an underlying cause in multiple ways. Fiscal reform had eroded the government’s ability to ensure adequate infrastructural investments, liberalisation of the capital account led to surge in capital inflows into India, banking reforms had given the public banking sector the flexibility to diversify into both retail and infrastructural lending and increase exposures to individual firms and business groups, the corporate debt restructuring mechanism allowed NPAs to remain concealed, and fiscal reform made it difficult to deal firmly with the NPA problem when it first arose.
The boom-and-bust cycle has also to be seen in the context of long-term trends in the economy. By the late 1990s it was becoming clear that with agriculture languishing, earnings in lower-income deciles stagnant or declining, and public expenditure held back by tax leniency and fiscal conservatism, the India economy’s growth was set to contract. The economy was faced with a policy-induced structural crisis. With institutional reforms that could create a mass-market having been abjured, growth depended on the stimulus provided by public spending. But with the government unwilling to mobilise the required resources through direct taxation of surplus incomes, such spending was being financed with borrowing. With the adoption of neoliberal fiscal policies that legislatively capped the level of government borrowing, that stimulus became ineffective and growth severely demand-constrained.
To an economy trapped in this structural bind, the credit surge of the mid-2000s offered a reprieve. It resulted in a sharp increase in consumption and investment, generating a boom riding on a credit bubble, that made India one of the fastest-growing economies in the world. The boom was not the result of an export stimulus and the confidence that generated, as AS-JF argue, but of an unsustainable lending and borrowing spree. That had to end, and it did. Restoring growth requires reviving demand with public expenditure increases in the short run and asset and income redistribution in the medium term to ‘make the market’ that can sustain viable growth. This would also require recapitalising the banks after they write off bad debts that cannot be recovered so that credit flow can resume, which too would require changing the current official attitude to increasing public debt to finance the process.
Missing the credit bubble
It must be said that AS-JF do emphasise the role of credit flow and its end as proximate determinants of the post-2008 growth and the recent slowdown. But they do not recognise the role of the credit bubble in driving the pre-2008 growth process. Rather they see the credit in that period only supporting a boom that was anyway occurring. When they do speak of an unsustainable increase in credit, they see it as having come after demonetisation, when deposits with banks surged triggering lending, often mediated by the NBFCs, to real estate companies and the housing sector, creating India’s version of the housing bubble that went bust. That line of reasoning gives the two different forms of the twin balance sheet deficit problem. The first occurred because global developments after the crisis slowed growth, which adversely affected infrastructural investments made during the boom and the banks that financed it. Nobody in India was really to blame. To quote AS-LF: “After the GFC, growth was much slower, interest rates much higher, and the exchange rate much more depreciated …, all of which wreaked havoc with firms’ financial projections. Profits consequently collapsed, making it difficult for many companies to service the debts they had contracted during the boom.” The second twin deficit developed when the NBFC-led and bank-supported real estate financing boom led to a housing glut and damaged the real estate companies and the NBFCs, with collateral damage in the banking sector and elsewhere (such as the market for automobiles).
This reading of the crisis makes the NPA problem in the financial sector the prime and almost ‘independent’ cause of the crisis. Explanations varying from those which see in the slowdown a reassertion of the fundamental demand constraint limiting Indian growth to those who trace the crisis to imaginary constraints created by rigidities in the labour market and restrictions on private sector operations are, according to AS-LF, inadequate or plain wrong. Going by this perspective, resolving the current crisis requires resolving the bad loan problem. This requires the government to find ways of (i) recovering dues through means similar to the hitherto ineffective Insolvency and Bankruptcy Code, though they do not seem optimistic that this could yield much; (ii) writing off NPAs that cannot be resolved and recapitalising the banks and NBFCs; and (iii) in some areas like the power sector, where wrong decisions have created assets that cannot be sold, AS-LF suggests, “the plants would need to be ‘warehoused’—until they can be returned to the private sector. To do this, the government could create a holding company, which would purchase the assets and manage them.”
The “solution” being offered here is to write off the bad debt and bad assets with government resources or taxpayers’ money, though how the money required would be generated immediately within the neoliberal fiscal framework is not fully explored. The private sector would emerge as the biggest beneficiary of crisis resolution. Banks would take large haircuts since private assets cannot be adequately drafted to repay loans gone bad, private assets (in power) would be taken over by the government, restored to health and returned at favourable prices to the private sector.
In this resolution recommendation, there is no engagement with the possibility that the NPA problem could recur. Clearly, the assumption being made by AS-LF is that the Indian economy is in good shape and its growth secure, and the Great Slowdown is an aberration resulting from global factors and some unusual events like demonetization that spiked bank deposits and lending. However, if as seems to be the case the Indian economy is trapped in a long-term structural crisis which was concealed by an unsustainable boom riding on a credit bubble, then unless that problem is resolved this crisis could recur.
Meanwhile, AS-LF offer a suggestion that would allow the government to wipe its hands of any banking problem in the future. Once the current problems are addressed, AS-LF believe, public banks should be privatised. To quote: “Private banking has proved imperfect but public sector banking (PSBs) has proved to be decisively flawed. The age-old problem of political interference and decision-making inertia is well-known. But the problems go much deeper. We must finally acknowledge that India’s public sector banks lack—and will always lack--the basic risk management framework to conduct any semblance of prudent banking. And once the imprudent loans turn bad, public sector bankers have little incentive to resolve the problems.” In their view, of the available solutions to this problem they identify, “the simplest, most direct but politically most difficult would be to amend the Bank Nationalization Act to allow majority private sector participation in the PSBs.”
This privatisation of public banks would be another way in which big business would benefit. But then, just as the absence of the development finance institutions undermined a source of long-term finance for infrastructure, the absence of public banks would undermine the availability of finance for investments crucial to growth. Only deep faith in the essential dynamism and inner resilience of the Indian economy can see this path as one that can resolve the crisis labelled the “Great Slowdown”.
Read the paper by Subramanian and Felman :