Now is the time for a bad bank

Indian banking has been going through a rough patch on the asset quality front for many years now. Gross NPAs (GNPAs) of scheduled commercial banks, which used to range between Rs 500 billion and Rs 700 billion in the 2000s, started picking up from the late 2000s and crossed Rs 1,000 billion in the year FY12. Post the asset quality review (AQR) of 2015, GNPA of all scheduled commercial banks (SCBs) increased more than 18 times to reach Rs 10,397 billion as of March 2018 from Rs 566 billion in March 2008. The good thing is that it declined in FY19 post the implementation of the Insolvency and Bankruptcy Code (IBC) and the aggressive efforts by banks to clean up their balance sheets.

The idea of a bad bank has been around for some time now though it was abandoned at one point. One of the reasons for jettisoning the idea was the view that a bad bank can do little in the absence of an effective resolution mechanism. Now that a resolution mechanism through the IBC has been put in place, the time could be ripe to revive the idea. Indeed, with around Rs 10 trillion of stressed assets in the banking system, we believe the time has come for a professionally-run bad bank. 

A bad bank is like a normal entity where the bad loans of a bank are parked. Such a model has been used in the past by countries such as Japan, Sweden, the UK and Slovenia (first initiated in 1988 by Mellon Bank). This allows the bank to continue with its usual functions without getting constrained by finance, resources or management bandwidth. This is particularly true for public sector banks in India. Besides faster economic recovery, the sentiment boost that will accrue to banks as their NPA burden is removed will be significant.

The potential advantages of having a bad bank are enormous. Having a single entity as the sole owner of a stressed asset — as opposed to coordinating with several individual creditors — will allow for binding resolutions even outside the NCLT framework — say regarding Inter Creditor Agreement (under the aegis of as Sashakt India Asset Management Ltd) that eludes consensus most of the time. A bad bank, being manned by seasoned industry professionals and distressed debt specialists, would be focused on the recovery aspect. Hence, it would be quite appropriate to revive the idea of a bad bank to quarantine and absorb our stock of poisoned assets. 

Critiques can argue that the banking system has asset reconstruction companies (ARCs) to support it in the case of bad debts, but we cannot forget that ARCs have been in India for more than 15 years, and in all these years they have not been of much help in realising the objective of reducing bad loans of banks or of reviving them. Further, these ARCs are resource constrained and do not have big wallets to buy large assets.  

However, there are primarily two issues with a bad bank. First, the lifecycle of the proposed bad bank, and second, the capital required to create such a bank. If both the bad bank and IBC were to coexist, it would send a strong signal to the market, contrary to the popular perception of the bad bank folding up once the purpose is served. Next, capital. Like in the case of bank recapitalisation, a bad bank could be initially capitalised by the government through, say, specially floated distress bonds, through which it could pay for its distressed asset purchases from banks. The size of such distressed bonds could be initially around Rs 60,000 crore with which the bad bank could be adequately capitalised for at least two years. For example, Rs 10 trillion stressed assets with 76 per cent current provision coverage ratio of banks implies Rs 2.4 trillion (net of Rs 7.6 lakh provision cover) are still to be provisioned, of which 25 per cent, or Rs 60,000 crore must be provided by banks over two years strictly as per the provisioning norms of the Reserve Bank of India.

After recapitalisation, bad bank can offer to buy the stressed assets from all the banks at a mutually approved price. Say for a Rs 100 notional stressed asset, Rs 43 could be the right market price, that is, 57 per cent discount to notional (the current recovery rate as per IBC). In the normal course, individual banks can choose to decline such a bid. However, they would then have to provide for and mark down the bad assets to say Rs 43, well below the bid price. Hypothetically, if the bad bank were to recover Rs 70 from the asset over time, that is, Rs 27 more than their purchase price of Rs 43, they could pay a fixed percentage of the surplus (assume 75 per cent of Rs 27/Rs 20) to the original bank. On the flip side, losses, if any, would stay with the bad bank. Given the carrot of participation in the recovery upside, alongside the stick of higher provisions, it is more than likely that banks would be inclined to dispose of their stressed assets to the bad bank.

With the benefit of hindsight, benefits of the “good bank-bad bank structure” include a renewed focus on the long-term core operations of a good bank without the distraction of troubled assets. The most important benefit of setting up a bad bank would be that the PSBs would have a provision coverage ratio of around 76 per cent (up from 62 per cent in FY18). This implies that the PSBs would have provided for most of the bad assets and a wholesale transfer of the bad assets to the bad bank is just a technical issue and the process of recovery and resolution could be carried out much better. 

Additionally, removing troubled assets will relieve pressure on capital, enabling the institution to engage in more profitable and growth-oriented business activities including lending. Finally, removing troubled assets from the balance sheet would have a positive impact on the view of credit rating agencies, investors and potential investors, lenders, depositors and borrowers. 

The author is DMD & CFO, State Bank of India. Views are personal