A Bad Bank Won’t Work Without A Good Design: Why Government Must Own Most Of It

The Finance Minister has proposed the setting up of a “bad bank” to take over a chunk of stressed assets in the banking system so that (mostly) public sector banks can clean up their balance-sheets.

In her budget speech last Monday, Nirmala Sitharaman said: “The high level of provisioning by public sector banks of their stressed assets calls for measures to clean up the bank books. An Asset Reconstruction Company Limited and Asset Management Company would be set up to consolidate and take over the existing stressed debt and then manage and dispose of the assets to Alternate Investment Funds and other potential investors for eventual value realisation”.

The details on how this bad bank will be set up, and who will finance it, are not yet clear, but here’s the point: a bad bank will not work if it does not have a good design. Some preliminary thoughts on the bad bank have come from Debasish Panda, secretary in the Department of Financial Services, and Tarun Bajaj, secretary in the Department of Economic Affairs, and the signs are not good.

Apparently, the government is not going to put in a single rupee for the “bad bank”, whose capital will then be contributed by public and private sector banks. The idea is that the “bad bank” will operate on the 15:85 principle, where bad debts of over Rs 500 crore will be handed over to it, on which 15 per cent will be paid out to the bank transferring the non-performing asset (NPA). Some Rs 2.25 lakh crore of NPAs involving 60-70 accounts will thus be out of bank balance-sheets.

The problem is not the 15:85 rule, or any modification of the ratio proposed by the bad bank’s managers after it is set up. It is the flawed expectation that it will function well without government equity or control.

Existing asset reconstruction companies have not had a great impact on the problem of bad debts primarily because they do not give public sector bank managements the comfort that can come only from a government-owned entity.

If, say, a public sector bank were to transfer a bad asset worth Rs 100 crore and receives Rs 15 crore for it, most banks would be happy to do so if they have already provided for 85 per cent of it in their books. Even if it slightly less, they can write off the additional provisions in a quarter with otherwise good profits.

The question is: why haven’t banks which have already made those provisions not done so?

The answer may be this: there is no guarantee that a private asset reconstruction company that buys off assets cheaply at 15 per cent of face value will not make more from it by finding a good buyer. It could also realise more than 15 per cent at some point by patiently turning an asset around with the help of good managers.

Public sector managers often hesitate to take such major writeoff decisions because they can later be accused of writing off loans to favour cronies or big business houses. Often these assets may be bought on the cheap by another private sector entity, which may sometimes be linked to the same management that let the loan go into default.

A majority public sector-owned bad bank would help managements cross this mental hurdle. They can hand over their assets on the cheap, and in case an asset yields more at a later date, this can be justified by pointing out that the profits remain in the public sector.

A 15 per cent payout on Rs 2.25 lakh crore worth of assets works out to all of Rs 33,750 crore. Putting in that kind of equity means paying out less than Rs 17,000 crore for a 51 per cent stake. Public and private sector banks can then be asked to take up the balance. Even if they don’t, government can easily put in the entire amount through the issue of recapitalisation bonds, which is what it has anyway been doing so far.

Recapitalisation bonds are cash-neutral, as government buys equity in banks which then purchase an equivalent amount of government bonds, with the actual burden on the fisc being limited to the annual interest costs. On Rs 33,750 crore, the annual interest burden would not be more than Rs 2,500 crore at current interest rates on sovereign debt (yields on 10-year government bonds are far below seven percent right now).

Also, when bad assets turn good, or deliver more than the 15 per cent, the government can use the surpluses to buy back these same recap bonds, retiring them ahead of time.

To repeat, a bad bank will function sub-optimally if the government does not own it substantially. Government ownership of the bad bank is a better idea than expecting banks themselves to carry the can with their own money. If that was about to work, why would public sector banks not write off the loans that they have already provided for anyway?

The bad bank is a good idea only if it addresses the unstated fear in the minds of public sector bankers that they may be later taken to task for handing over a potentially good loan for a song.