While we are positive on the sector’s fundamental momentum, we observe that spreads are relatively tight. We also expect significant issuance over the next two years as 57% of hard-currency Indian bank bonds will mature by 2020.
- Indian bank regulations have come on apace and government support for the sector is not in question. The country now has a working time-limited insolvency regime and a regulator with the power to force banks to work out stressed assets.
To cope with the costs of the legacy cleanup operation, the Indian government is recapitalising public-sector banks to the tune of INR 2.1 trillion, at the high end of expectations.
Public-sector banks have much weaker balance sheets than their private counterparts due to their higher exposure to stressed sectors. They also dominate the sector, accounting for over 70% of banking assets. This explains why the government’s remedial measures are aimed more at them than at the private sector, which is less in need of assistance.
We therefore expect private sector banks to continue to gain market share and outperform their public peers for the next year.
We note that the Indian bank bond market is of exceptionally short duration, with 77% of hard-currency senior bonds maturing in the next four years. This has resulted in squeezed valuations in the short end.
We may therefore advocate looking further along the Indian bank curve for value and sticking to established private banks, as there is no significant discount in bond pricing for accidents, despite the execution risk inherent in bailing out large troubled lenders (eg, IDBI’s NPL ratio stands at 28%).
Structure of the market
The Indian financial system is growing and well-balanced. While the overall system is stable at 136% of GDP, little changed over the past seven years, it has doubled in nominal terms over that period and is rapidly becoming more diverse. The market shares of non-bank intermediaries, such as mutual funds, and private banks are growing from a low base, while banks’ share of credit flows dropped significantly from 50% in FY 2016 to 38% in FY 2017, as corporates increasingly replace bank funding with private sources. The size of the system and its growing diversification are both healthy, as is its limited cross-border interconnectedness; cross-border lending and borrowing stand at just 10% and 14% of GDP respectively, insulating Indian banks somewhat from external financial shocks.
Banks dominate the Indian financial system, accounting for 60% of system assets. One key aspect of the Indian system is the high level of interconnectedness between the government and the banks. Some 70% of system assets are held by the Public Sector Banks (PSBs), which are majority owned by the State. Moreover, all banks must hold at least 20% of their assets in government securities. Banks must also allocate 40% of net credit to Priority Sector Lending. Priority sectors include Agriculture (at least 18% of loans), Micro Small & Medium Enterprises (MSMEs – at least 7.5% of loans must go to the Micro segment), export credit, education, housing, social infrastructure and renewable energy. While all banking activities are carried out commercially, the government clearly does direct the flow of loans to its priority areas, somewhat distorting the market. However, the government remains very supportive of the banks, with a long history of providing capital support when necessary, the latest instance of which is the final instalment of the balance-sheet clean-up operation that began in 2016 and should be completed next year.
The big issue – legacy NPLs and new regulation
From the top-down, India’s financial system therefore seems fairly balanced. It also benefits from a very stable base of excess customer deposits, making funding comfortable. The big challenge for the sector over the past few years has been a high stock of non-performing and restructured loans. This problem is finally being tackled head-on after several years’ regulatory build-up, making us confident that the banks will finally clean up their balance sheets over the next 18 months.
Problems arose following the build-up of high corporate leverage to support infrastructure investments in the 2000s, which was largely financed by the PSBs. Deteriorating global conditions, weak domestic demand and structural bottlenecks (e.g. delays in environmental clearances) from FY 2014 hit banks hard, particularly in their Metals, Engineering, Transportations and Infrastructure loan books, leading to a significant deterioration in asset quality.
A further imbalance was created through widespread use of the 5:25 scheme to fund long-term infrastructure assets, whereby a lack of long-term funding meant that loans were granted as bullet structures for five years at a time on the understanding that they would be rolled for 25 years. When asset quality in these crucial sectors deteriorated, the authorities’ first response was to create various restructuring schemes to obviate banks’ need to recognise these bad assets; these included the Strategic Debt Restructuring (SDR), Flexible Structuring of Project Loans (5:25) and the Scheme for Sustainable Structuring of Stressed Assets (S4A). Multiple restructurings and/or evergreen finance were possible.
However, as banks’ financial positions continued to deteriorate, curtailing credit growth, reducing government dividends and necessitating a drip-feed of capital infusions, regulators decided from 2014 to tackle the problems head-on. This resulted in a raft of regulatory measures transforming the banking landscape. The main planks of the plan were as follows:
In February 2014, the RBI set up the Central Repository of Information on Large Credits (CRILC), capturing all exposures of banks above Rs 50 million. The data was accessible to the RBI and to banks themselves, adding transparency to the system to enable all to see how large exposures were marked on various books.
In 2015, the Indradhanush Plan introduced critical governance innovations in the PSBs, such as the creation of the Banks Board Bureau to improve the quality of Board candidate, splitting the roles of Chairman and Chief Executive and attracting qualified private-sector candidates for top positions.
At the same time, the RBI carried out an Asset Quality Review on the banks, prompting a spike in NPLs as more bad assets were recognised. PSBs fared markedly worse than the better-run private banks on this score.
PSBs' NPL ratio development
Private banks' NPL ratio development
Finally, in 2015, general forbearance, known as “standard forbearance” was forbidden. The only forbearance now allowed were the schemes listed above: SDR, 5:25 and S4A.
In 2016, the Insolvency & Bankruptcy Code (IBC) was enacted. It envisages timely resolution of borrower defaults through collective decision-making by creditors. The code not only provides a process to deal with insolvency but also lays down strict timelines for insolvency resolution, failing which the borrower will be liquidated.
The IBC obviated the need for forbearance mechanisms, as troubled loans could now be resolved. Consequently, the Banking Regulation Act was amended to authorise the RBI to direct banks to refer specific cases to the IBC where necessary. The RBI formed the Internal Advisory Committee (IAC) to carry out this task.
Since July 2017, 41 NPL accounts (35-40% of all system NPLs) have been referred by the IAC to the National Company Law Tribunal (NCLT) for insolvency proceedings under the IBS. If these accounts cannot be resolved within 270 days, they will be liquidated. The process consists of two tranches: NCLT1 started in July 2017 and covers the 12 largest NPLs, accounting for around 25% of system NPLs. NCLT2 started in December 2017 and covers the next 29 largest NPLs (10-15% of system NPls). Most banks have 60% coverage against NCLT1 accounts, which include three steel companies (Bhushan Steel, Electrosteel Steels and Monnet Ispat & Energy), one auto company (Amtek Auto) and one engineering firm (Jyoti Structures).
While these measures proved effective at accelerating NPL work-out rates and recognising troubled assets, they proved expensive for the banks, where heightened provisioning costs and lower interest income hit profitability and regulatory capitalisation. Over the past nine months, the authorities finally bit the bullet:
In October 2017, the government announced an INR 2,11,000 crore ($32 billion) recapitalisation of the banking sector, at the higher end of expectations. The money will come in three tranches and will be front-loaded: INR 1,35,000 crore from recapitalisation bonds, INR 18,139 crore from budgetary provisions and the balance of INR 58,000 crore from capital raising by banks from the market while diluting government equity (which should go some way towards solving the problem of high levels of government investment crowding out the private sector). The chart below shows how much capital each bank has been allocated:
Selected banks' capital infusions
In February 2018, the RBI launched its revised framework for the resolution of stressed assets. Key provisions include the need to report loans that are overdue even by just one day, the need to update the CRILC monthly, the abolition of the remaining forbearance categories and the codification of the 180-day limit for the work-out of bad loans.
- Finally, in April 2018, the RBI revised its Prompt Corrective Action (PCA) framework for banks in line with international best practices. Banks will now be categorised in one of three risk levels depending on capitalisation, profitability and leverage, with mandatory remedial actions becoming more stringent if banks’ financial positions weaken. All details are available here.
India has therefore moved from an environment where the lack of an insolvency regime led to mass forbearance of bad debt to one where an effective bankruptcy code now exists, is enforced, is timely and where capital injections have been provided to ease the balance-sheet cleanup operation. Bank oversight and governance have been improved and a transparent early-warning system has been created. Progress has been really impressive.
So where does this leave banks?
NPL ratios are set to rise…
PSBs are in a weaker position than their private counterparts, mainly because they have been exposed to struggling lumpy infrastructure debt. Reported NPL ratios for PSBs stand at about 12.9%, whereas they are “merely” at 4.7% in the private sector. On top of that, banks still hold significant restructured loans, which will be recognised in the near future. The RBI has given banks until August 2018 to resolve their restructured loans or classify them as non-performing and act accordingly. These comprise 1.8% of “standard restructured” loans (i.e. legacy loans from previous forbearance schemes), most of which will become NPLs; and another 1.8% of loans restructured under now-defunct schemes, the bulk of which will also become NPLs (with the exception of some performing 5:25 exposures). Private banks have a similar issue, though on a smaller scale: reported NPLs stand at around 4.7% of total loans, with a further 1.1% of loans restructured under the abolished schemes and 0.6% of internally classified stressed loans.
...While higher provisions will dent profitability
We expect provisions to rise as assets migrate from the stressed to the non-performing buckets. The change of rules regarding restructured loans required provisions of 40%, up from the current 5%. Profitability will be further impacted by the discovery in February of INR 126bn of fraudulent transactions at Punjab National Bank (c.0.25% of system loans), which are likely to be written off. We also note that the rise in yields is causing substantial mark-to-market losses on banks’ securities holdings, although the RBI has given them one-time leeway to spread these losses over four quarters. Overall, we see little chance of current credit costs of 2.5-3.0% subsiding in this financial year, and weaker banks are likely to post further net losses.
However, fresh capital and a friendly regulator should help the system get through this period of restructuring to emerge much stronger on the other side. Once again, private banks are likely to outperform the PSBs.