A few observations on India's banking crisis

Good Budget or not, there is nothing to suggest a change from the view, consistently held by this blog for nearly two years now, that the biggest immediate challenge facing the Indian economy is the resolution of its banking crisis. And yes, it is an as yet unrecognized crisis, not a mere problem. Fundamentally, India's banking sector is clearly in the red, with negative equity. And without addressing this, all other efforts are only tilting at the windmills. 

Consider four graphics from Credit Suisse, the go-to source on banking sector problems. The first shows the latest on all types of stressed loans, an estimated 13.4% of all loans.

The second is a table which captures the acuteness of stress, with stressed loans at 230% of public sector banks' capital and stress asset cover of just 16%,

The third one estimates the capital infusion requirements till 2018-19 to be in the range of $34-53 bn, of which just $8 bn is budgeted,

And the final graphic appears to indicate that things may worsen still, given the rising corporate indebtedness. It finds that, among its sample of 3700 listed non-financial companies with cumulative debt over $500bn, the share of companies having interest coverage ratio less than one rose to 41% of the sample companies debt at the end of Q3 2016. 

Five observations below.

1. All these problems were no secret to keen observers of the economy, and therefore, it is indeed surprising that the banking regulator has chosen to act with the current alacrity only now. The banking regulator was behind the curve both in preventing the problem and its recognition, and is now behind in its redressal too. 

2. A far higher magnitude of recognition and haircuts may be required, especially in some sectors. And this may need to happen fairly quickly, or risk increasing the scale of the insolvency and raise the resultant resoluction costs. In such cases, there is no point in waiting in the hope that market prices will rise and demand will improve, since the scale of indebtedness is so high that there cannot be any light at the end of the tunnel. The Credit Suisse, for example, estimates that many of the largest steel firms have debt per tonne of production which is many times more than their replacement cost or unit EBITDA even with the minimum import prices. In fact, almost 86% of steel sector debt is stressed, against just 10-20% currently recognized as NPAs.  

3. This requires "deep surgery" that goes beyond asset disposals and recapitalization. It involves governance reforms to public sector banks, revisiting infrastructure contracting assumptions and principles, and anchoring growth expectations at realistic levels. It demands hunkered down growth targets and strategies. 

4. In an eco-system where fraud and malfeasance are never far away, and which is also gripped by decision paralysis, the regulator needs to be wary of the potential incentive distortions associated with such resolution activity. For example, currently, apart from the direct recognition process, the RBI already has the Corporate Debt Restructuring (CDR), Strategic Debt Restructuring (SDR), and the 5-25 takeout financing schemes. In both the latter two, the banks have been using Security Receipts (SRs), issued by Asset Reconstruction Companies (ARCs) in return for stakes in assets disposed as part of SDRs, and covertible preferential shares (CPS) that banks take in regular bilateral restructuring and SDR process. There is the very strong risk that SRs and CPSs could end up becoming the parking grounds for hiding bad assets within bank balance sheets, instead of their recognition, haircuts, and disposal. 

It is no surprise that the RBI recently warned banks about the possibility of promoters of companies establishing shell companies to repurchase their assets which are auctioned off at much lower prices.

5. What complicates the problem is the unique nature of the sector, where information failures can play havoc. While the true extent of the problem needs steps for immediate recognition, its resolution may have to be prudently phased over time. A shock therapy of sudden disclosure, disposals, restructuring, provisioning, and recapitalization will most certainly parlayze the credit markets and even bring down many banks. Instead, recognition has to be followed by a clear (to the extent possible) and predictable reform path that reassures the markets. All along, once recognized, the RBI needs to put in place stringent information reporting requirements, and have the information rigorously scrutinized by its forensics and analytics division. All this requires credible and far-sighted leadership at multiple levels.  

Weekend reading links

1. In the context of the Jat agitation, Christophe Jaffrelot makes the argument that it is a reflection of the inadequacy of private sector employment opportunities and the relatively high wages in public sector, 

In the private sector, the average daily earnings of the workers was Rs 249 in 2011-12, according to the Labour Bureau, and those of the employees at large, Rs 388. By contrast, in the public sector, the figures were respectively almost three times more at Rs 679 and Rs 945... Understandably, the young Jats, Patels, Kapus and Marathas who do not find good jobs in the private sector fall back on the government. The search for government jobs among these castes is also influenced by their particularly skewed sex ratio... With fewer girls compared to boys in these castes, there is competition in the marriage market. However, there are fewer government jobs these days. There were 19.5 million jobs in the public sector in 1992-93 when India’s population was 839 million. While there are 1.2 billion Indians now, the number of jobs in the public sector has shrunk to 17.6 million. In states that have aggressively implemented the liberalisation policy, government jobs have almost disappeared. For instance, the government’s share in employment in Gujarat is only 1.18 per cent whereas it is 16 per cent in Kerala.

2. Livemint has six graphics drawn from IMF data illustrating how badly India fares on general public finance indicators and expenditures on health and education in comparison to other far less developed countries. General government revenues as a share of GDP is the lowest among the comparison group,

3. The second consecutive weak monsoon and declining farm prices mean that rural distress is compounding general economic woes. Rural incomes have been falling, as reflected in the declining tractor sales,

4. The latest addition to the very large body of evidence that India's middle class is disturbingly small comes from Livemint - just 51 million households with annual income above $10000 out of 267 m families, with the vast majority living at extremely vulnerable income levels. 

And as a reminder to those constantly playing the China theme, the contrast is night and day,

5. Livemint reports that the experiment of having an exclusive tax bench in the Supreme Court, hearing only tax matters, appears to have been a success. The two-judge bench disposed-off 170 cases, a four-fold increase over previous years. 

6. Business Standard draws attention to the just released data on investment proposals, which at Rs 3.11 trillion for 2015, touched an 11 year low. Actual investments too were down from Rs 787.47 bn in 2014 to Rs 779.72 bn in 2015. 

As the Livemint grahic below shows, while the metrics are smaller in investment proposals, the actuals materialized have stayed the same over the past four years. Further, ten industries accounted for 65% of all investment implemented and 10 states for 80% of all proposals. 

The most worrying sign is about the level of investment proposals. Given the aggressive courting of investors and the widespread euphoria, coupled with economic weakness across the world, it was only to be expected that the investment proposals increase. This would especially be so given the low base effects legacy from a weak economy and decision-paralysed governance of 2012-14. But even here, the decline in investment proposals over 2012-15 has been about 40%. 

7. Distressed corporate balance sheets and rising bank NPAs have created a self-reinforcing downward spiral of bank credit squeeze to private firms, especially those outside the larger corporates

8. While capital investment as a share of total central government expenditure has been rising slowly to nearly a fifth, its distribution has increasingly been towards social services - housing, labor welfare, and rural works - whereas the share of transportation has declined sharply.

Interestingly, and underlining the importance of co-operative federalism, just a third of the capital expenditure of the central government is executed by itself, with the major share being transfers and loans to state and local governments. 

In fact, in 2014-15, the central government's capital expenditure was 1.78% of GDP or Rs 1.92 trillion, to the state government's 3.54% of GDP or over Rs 4 trillion.

9. Interesting observation on the changes in India's tax-to-GDP ratio over time,

In the 25-year period from 1965 to 1990, India’s tax-to-GDP increased steadily from 10% to 16% while GDP increased 2.8-fold. In the subsequent 25-year period from 1991 to 2014, India’s tax-to-GDP stayed roughly constant between 16% and 17% while GDP increased 4.5-fold. It is puzzling to us that just as India broke away from its clichéd Hindu rate of growth post the 1991 economic reforms to grow much more rapidly, its tax-to-GDP ratio stayed constant, belying those who would have predicted an increase. That, curiously, India’s rate of tax revenues did not grow commensurate with its GDP growth post the 1991 reforms is inexplicable.

What could possibly be the reason? Given that direct taxes at 2-3% of GDP is marginal, most of the changes would have revolved around indirect taxes. Evidently, in the 1965-90 period, state expanded its indirect tax base, while in the 1990-2014 period, it did little to expand the direct tax base. The later is difficult to rationalize, given the massive income growth, concentrated especially at the top of the income ladder, both individuals and corporates, who, in any case, form the lion's share of direct taxation. In other words, these incomes at the top may have grown much faster than the proportionate rise in their income tax payments. Among all public policy priorities, this demands an urgent examination by the government.    

10. The positive side of the high inflation was that it kept the tax revenues high and ensured that the debt-to-GDP ratio was kept under control. Now with low inflation, nominal GDP growth has fallen, even touching the government's average interest cost, thereby raising questions about the country's debt sustainability.

11. While this is well-known within the government, but rigorous evidence has been scarce,

Since 2009, Accountability Initiative’s PAISA study (Planning, Allocations and Expenditures, Institutions: Studies in Accountability) has been tracking money flows in elementary education. In not one of the eight districts across six states that we have studied did we find a district that received its entire allocated budget within the financial year. For the money that does reach, much of it makes its way to the district half way through the financial year. Even districts in well-functioning states like Maharashtra and Himachal Pradesh face this unpredictability. Once money reaches the district, it can take two to six months to travel from the district bank account to the schools, where expenditure actually takes place... for over 50% schools in India, even small grants that schools are expected to receive annually for essential purchases, are credited to their bank accounts somewhere between November and December, well over halfway into the school year. 

12. For all talk of decentralization, local governments share of revenues and expenditures, on a variety of metrics, remain marginal.

13. This analysis of the CAG report, which paints a very dismal picture of Indian Railways,

As of March 2014, the Indian Railways had spent around Rs.92,000 crore on 479 projects, including some dating back to the 1970s and 80s. But due to shoddy contract and project management, costs have more than doubled, and the Railways needs an additional Rs.183,000 crore just to finish these ongoing projects.

And this analysis of the ticket prices,

The Indian Railways' average revenue per passenger km for ordinary second class is a mere 13.80 paise, 14.54 paise for suburban trains, 27.47 paise for second class mail and express trains, and 109.47 paise for upper class, making it possibly the cheapest rail transport system in the world.

The assumptions underlying the optimism presented in the Railways Budget may not exactly be forthcoming. 

14. Finally, in order to relieve crowding, and boost revenues, Disney introduces surge pricing in its Florida and California theme parks during holidays and some weekends, raising prices by about 20%, 

At Disneyland, located in Anaheim, Calif., which attracts roughly 17 million visitors annually, single-day tickets now cost $99... “Value” tickets, for Mondays through Thursdays during weeks when most schools are in session, will drop to $95. “Regular” tickets (most weekends and many summertime weeks) will climb to $105. “Peak” tickets (most of December, spring break weeks, July weekends) will cost $119. At Disney World in Orlando, Fla., which includes four major theme parks, the price changes are more complex, because they vary by park. At the most popular Disney World park, the Magic Kingdom, which handles nearly 20 million visitors annually, single-day prices will remain at the current level, $105, for value periods. Prices will rise to $110 for regular periods, and to $124 for peak... The largest proportion of days at both Disneyland (46 percent) and Disney World (49 percent) fall in periods designated as regular; peak days account for 27 percent of the year at Disneyland and 29 percent at Disney World.

State capability 3.0 - The Finance Ministry veto?

In addition to the standard examples of state capability weakness, reflected in weak public service delivery, there are atleast two other far less discussed forms. One is decision paralysis due to excesses by judges, RTI activists, auditors, and vigilance officials. The second, barely discussed contributor to state capability weakness is the de-facto veto exercised by Finance Ministries. 

Consider the example of financial inclusion. Business Correspondents (BCs) were introduced with the intention of expanding access by taking banking services to the customer's door-step in remote rural areas. Its success was evidently contingent on the commercial viability of the business model as reflected in the income earned by the BCs. But as Sumita Kale points out, this may have been compromised by the hard charging Finance Ministry,

While the Report of the Task Force on an Aadhaar-Enabled Unified Payment Infrastructure had recommended a 3.14% transaction processing charge to the banks, in reality the rates allowed by the central and state governments have been 1-2%. In January 2015, the finance ministry fixed DBT commissions for banks: for urban schemes, at the National Electronic Funds Transfer/Aadhaar Payment Bridge rate; but for rural schemes, the rate was fixed at 1%, subject to an upper limit of Rs.10 per transaction. Detailed costing analysis fromconsulting firm MicroSave in May 2015 shows that 2.63% is the break-even charge: the break-up of this across the three main constituents in the disbursement chain came to about 0.96% for business correspondent (BC) network managers, 0.85% for business correspondent agents, and about 0.82% for the banks. The analysis also revealed that the savings to the government through lower administrative costs and leakages are significant. Clearly, the government can well afford adequate compensation to the banks and agent networks for their role in the disbursements.

The story is the same everywhere, across programs, both in central and state governments. Ministries formulate schemes and proposals after elaborate stakeholder consultations, with detailed costing taking into consideration program sustainability and commercial viability factors, and run them through Finance Ministry to the approving authority (Cabinet or Chief Ministers). As would be expected given the scarce resources and large competing demands, the Finance Ministry cuts down on program allocations. It would have been perfectly fine, indeed necessary, if this was all. 

Unfortunately, in most cases, the Ministry goes far beyond and offers its wisdom on program commercials (a unit rate compensation of Rs x instead of Rs y), procurements (why not through the local government agency or SHGs), financials (why not leverage resources from Corporate Social Responsibility funds or PPP), contracting strategies (one model of PPP over another), and even on manpower deployment (one administrative structure over another, x number of people to do a task instead of y). And each of these prescriptions are drawn from mindless generalization of outlier and misleading thumb-rules and 'best practices'. 

If it were only offering suggestions, the implementing Ministry could have examined and taken necessary action. But in India's bureaucratic rules of the game, such prescriptions assume the force of a veto. This is all the more so since Finance Ministry's recommendations are invariably in line with the bureacratically correct practice of lowering public expenditure. Disagreement with it runs the future risk of a malicious complaint or a RTI query or an adverse audit comment and a potential vigilance investigation with its public humiliation.  

Its manifestations are felt across programs, especially at the last-mile of implementation, and become the difference between success and failure. The commonest form of veto is by way of skimping on transaction costs (transportation and storage charges for PDS), remunerations (mid-day meal workers), construction costs (unit costs for ), operation and maintenance expenditures (school and toilet maintenance in SSA, hospital maintenance in NHM), leverage from other sources (CSR, NREGS etc), and so on. Doubtless some of these are unavoidable, forced down by the acute scarcity of resources and political economy considerations that demand the butter be spread evenly and universally. But many others like the one for BCs are plain arbitrary and flippant. 

At a very objective level this is as much a transgression of jurisdiction as kritarchy or tyranny of presumptive valuation. Consider this. A professionally competent agency (the Ministry concerned), with the responsibility of implementation, and democratically empowered, formulates a program following the due process, and circulates for approval. En-route, another Ministry, with neither the contextual knowledge nor the professional competence, and at best, with accounting and budgeting competence, picks apart program components and details, and that too in a manner that seriously undermines its successful implementation prospects. Where else can you have this except in a 'flailing' state! 

An update on DISCOM Bonds

As part of the UDAY scheme, the state governments across India will assume three-quarters of discom debts over the next two years. Apart from significantly lowering their borrowing cost (current cost of 12-14% to 8-9%), it is hoped that this would relieve the discoms off the crippling debt burden. As part of the scheme, the state governments are expected to issue bonds worth about Rs 3.2 trillion (75% of Rs 4.3 trillion total discom debt) over the coming two fiscal years, with two-thirds in the first year.

In this context, Andy Mukherjee pointed to the $48 bn challenge of markets ability to absorb these bonds. Consider this. The total net market borrowings of the central and state governments for 2014-15 were Rs 4.53 trillion and Rs 2.07 trillion respectively, and Rs 4.4 trillion and Rs 2.2 trillion respectively till February 12, 2016. In other words, if all state governments join, and most the largest have already done so, the cumulative UDAY bond issuance in 2016-17 would, at Rs 2.15 trillion, be larger than the state government issuance for 2014-15.

This coupled with a higher fiscal deficit could put a massive strain on the bond markets and quickly drive up the yields on these bonds. In fact, in the first round of auctions for Rs 21,445 Cr worth 10 year State Development Loans (SDLs), yields touched 8.88%, higher than expected, and indicative of the strains likely as more bonds flow into the market. As a reflection of the widening spreads, 

The SDL auction found takers at yields between 8.63-8.88% for 10-year maturity bonds. The 10-year benchmark G-sec (central government bond) yield is currently hovering around 7.82%, indicating a spread of 80-105 basis points. The spreads are sharply higher than the average spread of 39-45 basis points at the start of 2016.

One way around this, and not a very desirable one, is to summon the perennial buyer of last resort in India's financial markets, LIC. And unsurprisingly, Livemint reports that LIC and EPFO have been asked to assist in mopping up the restructuring bonds that will be issued. 

The mixed story with school choice and vouchers

Free-marketers hail school choice through vouchers as the most effective strategy to foster competition and improve school standards. However, the evidence on school vouchers has been, at best, mixed. I have blogged earlier, using the logic of Schelling's chessboard experiment, to argue that school choice is likely to lead to 'emergent outcomes' that may be far less benign than expected. 

The Economist draws attention to a study of a very large school voucher program initiated in New Orleans in the aftermath of Hurricane Katrina whose findings its describes as "underwhelming". In 2014, the Louisiana School Program (LSP) assigned more than 6000 students from low-income families from 12000 applicants to 126 private schools through a lottery system. The study by Atila Abdulkadiroglu, Parag Pathak, and Christopher Walters compared learning outcomes for lottery winners and losers in the first year after the program's statewide expansion and find,

This comparison reveals that LSP participation substantially reduces academic achievement. Attendance at an LSP-eligible private school lowers math scores by 0.4 standard deviations and increases the likelihood of a failing score by 50 percent. Voucher effects for reading, science and social studies are also negative and large. The negative impacts of vouchers are consistent across income groups, geographic areas, and private school characteristics, and are larger for younger children. These effects are not explained by the quality of fallback public schools for LSP applicants: students lotteried out of the program attend public schools with scores below the Louisiana average. Survey data show that LSP-eligible private schools experience rapid enrollment declines prior to entering the program, indicating that the LSP may attract private schools struggling to maintain enrollment. These results suggest caution in the design of voucher systems aimed at expanding school choice for disadvantaged students.

Admittedly the results ought to be read with caution and need to be observed over the coming years. But it cannot be denied that the reality with vouchers is at least far more nuanced. Incidentally, the Economist report also points out that post-Katrina, New Orleans' public schools improved dramatically on the back of enlightened leadership.

That scarce trait is more likely than fancy innovations to improve school education in countries like India. Further, in its absence, even innovations are likely to flounder during implementation. Delegating powers to district and local governments, despite all its concerns, is one way to facilitate the emergence of such bright spots. 

Weekend reading links

1. The banking sector and corporate balance sheets constitute the Indian economy's two biggest immediate problems. The ebitda-to-interest ratio of the median company with market capitalization of more than $100 m four years ago is lowest for Indian corporates.

2. This blog has held the view that Indian economy is currently investment demand constrained, and, therefore, unlikely to respond to supply-side measures. In this context, Jahangir Aziz highlights the challenge,

Ask any corporate (entity), and in private they will all tell you the reason to have shelved their expansion plans is not because they can’t get land, or that labour reforms have not been implemented or that cost of capital is high—instead, they will all say there is no visibility of demand, and until there is visibility of sustained demand over the medium term, regardless of reforms on the cost side, it is very difficult for them to invest.

About the priorities for public policy to re-ignite demand,

You will need to now start spending on restructuring and reforming the areas where Indians save the most—for their children’s education, housing, daughter’s wedding and healthcare. If you look at out-of-pocket expenses on health and education, they are astronomically high in India. The government needs to start attacking the areas where precautionary savings are the highest. Education and health are readily and easily observable areas where people put in massive amounts of savings and, therefore, the government should be putting in money here, rather than infrastructure. We need a better balance between infrastructure and pushing money into education and health. The biggest expenses are for higher education and not for 12 years of schooling—where you spend ridiculous amounts for private colleges and universities. This is because the centre has not met its responsibilities of building places of higher education.

And why the current priorities, even in infrastructure may be off the mark,

If you look at infrastructure design in India, we are expanding ports and connecting them to hinterlands, we are expanding airports, we are connecting the metros by expanding the Golden Quadrilateral. But no one talks about, say a 10-lane highway from Kanpur to Coimbatore. There is no way I can go from Kanpur to Coimbatore without going through either East Coast or West Coast. These large investments make sense only if we believe that the export-led model of growth that we had in the early part of 2000s will come back. But if exports won’t come back, why are we even bothering with new ports.? From infrastructure design to mindsets, all has to be changed if we need to find new sources of demand, and this new source is domestic consumption. This new demand will get me the corpus to start investing, and that will generate growth.

3. A fascinating feature on how baseball has become the ultimate socio-economic mobility ladder in the Dominican Republic. As of opening day 2015, Dominicans made up 83 of US major league baseball’s 868 players.

4. Thanks to tax inversions, the Tiebout theory would bind even more strongly for corporate tax in the years ahead. Nice article in the Times on the recent spate of tax inversions in the US. Addressing the issue of tax arbitraging should top any multilateral agenda. By the way, it is surprising why the beggar-thy-neighbor, low-tax policies of countries like Ireland does not attract the same level of indignation that currency manipulation does. 

5. Martin Wolf analyzes Japan's problem as fundamentally one of weak demand - an aging population and declining demand shrinks investment opportunities, leaving corporates with growing surpluses. In this, as has often been said, Japan may be a portend for many developed economies in the years ahead. But despite this serious headwind, the Japanese economy has done remarkably well, having the highest growth rate of GDP per working person for the 2000-15 period among all G-7 economies.

Martin Wolf's prescription is for policies to slash corporate surpluses by encouraging them to raise wages (to boost demand) and impose taxes. I am not sure. This works under the presumption that demand is currently suppressed. A country with worsening demographic balance needs a little less of everything each passing year. Higher wages are therefore only likely to be saved. A more compelling suggestion would be to ease immigration and activate that demand channel.

6. Japan may also be in the vanguard of secular stagnation trends, of which Larry Summers is ever more convinced,

With appropriate caveats about the complexities of drawing inferences from indexed bond markets, it is fair to say that inflation for the entire industrial world is expected to be close to one percent for another decade and that real interest rates are expected to be around zero over that time frame. In other words, nearly seven years into the U.S. recovery, markets are not expecting “normal” conditions to return anytime soon.

Apart from cheap capital goods, this explanation for investment demand being constrained is interesting,

The new economy tends to conserve capital. Apple and Google, for example, are the two largest U.S. companies and are eager to push the frontiers of technology forward, yet both are awash in cash and are under pressure to distribute more of it to their shareholders. Think about Airbnb’s impact on hotel construction, Uber’s impact on automobile demand, Amazon’s impact on the construction of malls, or the more general impact of information technology on the demand for copiers, printers, and office space. And in a period of rapid technological change, it can make sense to defer investment lest new technology soon make the old obsolete.

Having said all this, Summers appears to refute his original assertion and claim that it is, after all, possible to get back on the previous growth path,

Although developments in China and elsewhere raise the risks that global economic conditions will deteriorate, an expansionary fiscal policy by the U.S. government can help overcome the secular stagnation problem and get growth back on track... An expansionary fiscal policy can reduce national savings, raise neutral real interest rates, and stimulate growth.

This argument is surprising and runs contrary to his own argument about investment demand being constrained. How would fiscal policy address the headwinds of technology and capital conservation? As Japan has been finding out over nearly a quarter century, public investments in infrastructure can only get you so far. It can, at best, ameliorate some of the pains of a secular stagnation.

7. A very good exploration of the divide in the Keynesian camp between those advocating continuation of monetary accommodation (Krugman, Summers, De Long) and those (Fed insiders) preferring to proceed with raising rates.  

Resolving bad loans and reconstructing assets

I had blogged earlier that addressing India's banking crisis would require both resolving bad loans and reconstructing assets. And both would have to be complemented with devolving complete operational autonomy and massive recapitalization. 

The first step would be to classify stressed assets into completed infrastructure projects, ongoing projects, and all remaining retail, credit card, and commercial loans. In the case of the last category, equity holders should be stripped and assets auctioned off to private Asset Reconstruction Companies (ARCs). 

As regards completed infrastructure projects, where it is possible to monetize the revenue stream, the current Strategic Debt Restructuring (SDR) may be the best course of action. But banks ability to dispose-off these assets remains questionable. An alternative, therefore, may be to auction them off to ARCs, stripping equity holders and with haircuts on banks, with a cascaded and backend clawback of some share of future revenues to the banks and equity holders. This could avoid the political backlash likely in case the asset generates windfall revenues once the economy recovers. 

Finally, ongoing projects would need financial reconstruction. A vast majority of them are likely to be commercially viable once completed, but may require further equity infusion. Further, the construction risks associated with them make them less attractive for long-term investors like infrastructure debt and equity funds. And, in any case, such risks are best borne by the government. In the circumstances, a preferable strategy would be to value them and sell off to a public entity like the IIFCL. The IIFCL, by itself or through the newly created National Infrastructure and Investment Fund (NIIF), can raise three or four dedicated infrastructure debt funds, leveraging long-term foreign patient capital, to finance these purchases. These funds, with professional project management units, should manage the completion of these projects. The financing patterns can even be restructured once the construction is completed. A distinction may have to be made between public good assets like roads and private assets like power and steel plants, in terms of the extent of public financing. 

All this would have to be done quickly and simultaneously. The entire process may be concurrently audited and all requisite clearances taken to pre-empt post-facto audit and vigilance objections. As aforementioned, it would have to go with clearly defined operational autonomy as well as an equally clear recapitalization schedule. The operational autonomy would have to include eschewing the urge to saddle banks with various social obligations without sufficiently compensating them. 

The biggest uncertainty with this approach rests on the supply side. Does India's credit and capital markets have the capacity to absorb such scale of transactions? Will there be enough buyers for these assets? The gross NPAs are estimated to reach Rs 5.5 trillion by end-March 2016, and maybe double that by including all the other badly stressed assets. To put that in context, the total incremental non-food bank credit in 2014-15 was Rs 5.46 trillion, new bond and equity issuances Rs 0.17 trillion, new PSU bond issuance Rs 0.38 trillion, net CP issuance Rs 0.87 trillion, and all disbursements by public financial institutions Rs 1.03 trillion.

Update 1 (22.02.2016)

Links to articles that explain how an ARC works, problems faced by ARCs, more on problems, high ARC asset acquisition costs and banks' risk aversion, systemic problems, and disturbing relationships between bankers, promoters and ARCs.

Update 2 (26.02.2016)

Corruption is never far away with such deals. Livemint reports of a circular by the RBI on,

... fears that promoters of companies acquired by banks after they failed to repay loans may be using shell entities, in India and elsewhere, to buy back these assets at much lower prices... If that is the case, it would also allow unaccounted-for or black money stashed by Indian businessmen overseas to come back into India.

And on the progress with the Strategic Debt Restructuring (SDR) scheme,

Since June 2015, when SDR rules were introduced, lenders have converted debt to equity in a number of firms including Electrosteel Steels Ltd, Ankit Metal and Power Ltd, Rohit Ferro-Tech Ltd, IVRCL Ltd, Gammon India Ltd, Monnet Ispat and Energy Ltd, VISA Steel Ltd, Lanco Teesta Hydro Power Pvt. Ltd, Jyoti Structures Ltd and Alok Industries Ltd. Of these, the only known case where lenders are closing in on a sale is Electrosteel Steels.