The Reserve Bank of India has come out with revised guidelines, setting out a framework for timely resolution of stressed assets through the use of the Insolvency and Bankruptcy Code (IBC). They have withdrawn the existing forbearances allowed earlier , such as CDR, SDR, S4A, 5:25, etc, which is likely to result in an acceleration of NPA formation.
On the current stock of NPAs, of the `8.8 trillion, we estimate Rs 3.5 trillion is already part of the IBC under List 1 & 2 of RBI. We estimate that banks have Rs 1.6 trillion under the watchlist/SMA 2 loans, which is likely to slip to NPLs at a faster pace, given the lack of forbearances available. We estimate another Rs 2.6 trillion currently under various forbearances.
We were building this ‘slipping into NPL’ over the next two years; with the revised guidelines, the slippage could be accelerated and may slip in FY19.
Nearly 4% of system loans are already under IBC, we expect the pace of slippage to accelerate, especially from the watch list/SMA-II accounts, as the RBI dispensations are no longer available.
With the RBI pushing all cases Rs 20 billion to be resolved within 180 days from March 1,2018, we would also see some of the current NPAs being pushed into the IBC.
RBI has come out with guidelines for larger stressed assets over Rs 20 billion and will come out with timelines for the stressed assets over Rs 1 billion over the next two years.
Cases that are over Rs 20 billion, if in default as on March 1,2018, would have to be resolved within 180 days or would be referred to the IBC within 15 days. As we have seen from the second list of IBC cases, while banks were allowed up to six months to come up with a resolution plan, they have been unable to do so and most of the cases have been referred to the NCLT. Now, the resolution plan needs to be approved by all the lenders (as opposed to 60% under the JLF) and the residual debt as per the resolution plan will need independent credit evaluation by credit rating agencies (two rating agencies needed in cases `5 billion). While this may result in stronger resolution plans, the likelihood of a resolution plan being agreed upon reduces and the accounts are likely to be referred to the NCLT.
If there is a default post resolution, it will be referred to IBC within 15 days of default. Post the resolution plan, the asset needs to be performing until > 20% of the initial principal is repaid, or till at least one year from the first repayment.
Any failure in meeting timelines or attempt to evergreen stressed accounts would result in higher provisioning needs and monetary penalties.
With the IBC process still in its infancy—and the first list of 12 cases pushed through by RBI now in the bidding stage and likely to be resolved in the coming months—a large share of the second list of 28 cases, despite banks being given six months to come up with a resolution plan, has now been referred to the IBC. With Rs 3.5 trillion (~40% of system NPAs) now in the IBC and another `1-1.5 trillion likely to be referred in the coming year, the IBC is likely to be the key resolution mechanism.
With the increased supply of assets in the process and restriction on existing promoters from bidding for assets, the number of bidders may be lower for the smaller accounts and the potential haircuts may rise. In the first list of 12 cases, while interest has been high in some steel accounts (Essar Steel & Bhushan Steel), the likely haircuts on the other assets are high. (Monnet – 75% haircut, Jyoti Structures – 81%, ElectroSteel – 56% and Alok Industries didn’t see any bidders on the first attempt). The current NPL cover for the system is at 47%, however, overall stressed asset cover is lower at ~33%.
While gross NPAs are at 6-15% of loans, non-NPA stress accounts for another 5-10% of loans for the banks, with stressed asset cover low at 25-40% for most of the larger banks.
With non-NPA stress at ~6% of loans for the PSU banks, if we were to assume 70% would slip to NPL and a LGD to be contained at ~60%, the provisioning needs would be Rs 1.2 trillion and the government’s planned capital infusion of `2.1 trillion would suffice. However, if LGDs were to increase by 10% to 70%, the capital needs would increase ~`600 billion.
While we were building bulk of the residual pain to go through and coverage ratios to hit 60%+ by FY20, the same could now be largely pushed to FY19.
As a result, banks which have (1) higher stressed assets coverage levels and (2) strong capital base to absorb the provisioning requirements will be relatively better placed. For our covered banks, a 60% haircut implies a 1-21% fall in FY19 book value and a 10-200 bp fall in CET-1 ratio.
ICICI Bank and Axis Bank are relatively better placed, given that large provisioning needs are already factored in estimates and the banks have sufficient cushion of capital to absorb—even a 70% haircut without dilution. Among PSUs, BoB is better placed with an already high stressed assets cover. For SBI, while the book value impact is limited (2% estimated in the 60% haircut scenario), capital is low. However, recapitalisation by the government will aid accelerated clean-up of books. PNB, BoI and Union bank are relatively weak and could see capital needs increase.
By Ashish Gupta, Managing director and head of India research, Crédit Suisse, research division
Co-authored with Kush Shah and Anurag Mantry, research analysts with Crédit Suisse