Hazards of a bank runner- The persistent mystery of non-performing assets

Writing on the wall - Ashok V. Desai

http://www.telegraphindia.com/1150707/jsp/opinion/story_29966.jsp#.VZuPHNKqqkp

Non-performing loans are a wonderful Indian euphemism. When a borrower does not repay a bank, the loan to him becomes non-performing. He did not repay it even before it stopped performing; but then it was supposed to be performing because the bank had not asked back for the money. Or, the bank had given the client a loan for a fixed term; if he did not pay at the end of the term, he ceased to perform. Presumably, some time after that, someone in the bank would wake up, call up the borrower and ask him to repay. If he did not, the bank would write to him pointing out his breach of promise. If he sent no reply or replied rudely and omitted to repay, his debt became a bad debt - or, in bank jargon, turned nonperforming. The bank itself would put off calling it that as long as it could. It would first call it substandard. Then if the borrower continued to renege, the bank would call the loan doubtful. If he proved himself an incorrigible bad egg, then the loan would be renamed a loss asset.

If a borrower comes to know that the bank has labelled him a bad debtor, he will have lost his reputation and will have no further incentive to repay; so a bank will put off calling a debt a bad debt as long as possible. If it turns bad, the money the bank lent will be irrecoverably lost, and must be written off against its profits. So the bank may opt for another way: instead of condemning a loan as a bad debt, a bank may set aside some of its profits in a reserve against potential losses, without specifying precisely which debts it is supposed to set off.

Every private bank would take its own call on its clients. But most Indian banks were nationalized in 1969; after that, their losses belonged to the government, so there was no reason for them to worry about bad debts. They went on lending merrily. Their practices changed little till the 1980s. The oil boom that began in 1973 created demand in West Asia for Indian workers as well as exports; and Japan's industry was earning such huge surpluses from exports that its banks ran out of borrowers and happily lent to Indian companies. Much of that money from abroad was deposited in banks; they were flush with funds, and lent liberally to borrowers within the country. That was the heyday of Indian socialism: the government encouraged its enterprises to borrow, invest and expand.

Then in 1990, inflation together with an overvalued exchange rate led to a crisis. The enterprises went bankrupt; even if they could repay their loans from abroad, the Reserve Bank of India did not have enough foreign exchange to let them. The Centre had a minority government supported from outside by the Congress; the finance minister, Yashwant Sinha, did not have the power, even if he had had the will or the skill, to steer the country through the crisis. Severest controls were introduced to save foreign exchange; national income fell because of the squeeze.

So when Manmohan Singh replaced Sinha in 1991 and called me in for advice, we tried not just to resolve the crisis, but to make sure it did not happen again. Maybe we knew some economics, or maybe we were lucky, but it has not yet recurred. Although we got some good ideas from the International Monetary Fund and the World Bank, we wanted to ensure that our reforms were homemade as far as possible; so we appointed a number of committees to tell us what to do.

One of them was the committee on banking sector reforms, known after M. Narasimham, a governor of the RBI of 1970s vintage who chaired it. It suggested standardization of bad and doubtful debts, and invented the term, non-performing assets. If there were dues on a loan (whether of interest or of repayment) unpaid for 30 days, it was called a loan past due. If it remained past due for four quarters, it became a non-performing asset. This period was reduced by 1994-95 to two quarters. In 2001, the RBI stopped counting the "past due" 30 days; since 2004, it has classified all receivables overdue for 90 days as non-performing assets. (That applies to you and me; farmer borrowers have always been given more time, based on crop seasons.)

Suppose a bank has jumped through the definitional hoops and managed to declare a loan a non-performing asset; what then? In other countries, the bank would sue the borrower in a court and recover the loan. That works badly in India because courts do not deliver verdicts for years, if not decades. So the government enacted a special law. Indians like long and complicated terms; they will have a hard time beating the Securitization and Reconstruction of Financial Assets and Enforcement of Security Interest Act of 2002. Its content is even more convoluted than its name; but essentially, it created new courts called debt recovery tribunals just for banks to sue defaulting borrowers, gave banks powers to seize defaulters' assets without their permission and without getting into legal trouble, and created special companies whose business was to pick up banks' bad debts and recover as much as possible from borrowers by using all the tricks that SARFAESI put at their disposal.

As a result of this draconian law, bad debts ceased to be a problem for banks for a decade. But it seems to have resurfaced: Raghuram Rajan has repeatedly expressed his worry about bad debts in the last two years. SARFAESI remains in force; I suppose the DRTs are also in existence. What has gone wrong? K.C. Chakrabarty, who resigned last year as deputy governor of the RBI, was of the view that this was because banks had grown lax. He compared two periods: 2001-07 and 2007-13. New NPAs in 2001-07 were 72 per cent of the opening stock; in 2007-13, they were 90 per cent. The stock of NPAs in 2007 was 22 per cent of the stock in 2001 plus new NPAs in 2001-07; the stock in 2013 was 35 per cent of the stock and accretions in 2007-14. Of the stock and accretions, 76 per cent were dealt with one way or another in 2001-07, and 64 per cent in 2007-13; to make the same point differently, the ratio of closing stock to gross reductions was 30 per cent in 2001-07, and 55 per cent in 2007-14. Of the NPAs that were dealt with in some way, 42 per cent were recovered in 2001-07, and 34 per cent in 2007-13. The proportion of NPAs written off was about the same: 44 per cent in 2001-07, 40 per cent in 2007-13. The proportion "upgraded" went up from 14 to 26 per cent; in other words, the banks forgave many more bad debtors without recovering all the money from them. Were they thereby making the best of a bad situation? Maybe; but perhaps, the bankers were forgiving defaulters in return for some gratification. That worried Chakrabarty, and should worry Rajan.