India’s banking system is creaking. If the country doesn’t find the capital to revive its weak public-sector banks and to address the mounting problem of nonperforming loans, its banking system will founder, with devastating consequences for the broader economy. The government must thus undertake to reform and consolidate the sector while encouraging strong local institutions and foreign investors to increase their presence in India.

To fend off insolvency, Indian banks will need to find fresh capital of $9 billion to $15 billion—representing 2 to 4 percent of the country’s gross domestic product and 50 to 90 percent of the current capital of Indian banks—during the next five years, which is roughly how long it will take to reform the system. Up to 60 percent of this sum will be required to write off irrecoverable loans, 18 percent to finance productivity improvements, and the rest to support growth (Exhibit 1). Some 35 domestic banking institutions, accounting for about 40 percent of the sector’s assets, are particularly fragile because of the poor quality of their assets. It is from these troubled banks that the vulnerability of India’s banking system largely arises. In addition, the recent crisis caused by weak investment-management processes at the Unit Trust of India (a state-owned institution with $12 billion of assets under management) threatens to suck $1 billion or so of liquidity from the banking system in order to fund redemption pressure. A further $500 million to $1 billion of capital could be needed to ensure Unit Trust’s solvency.

The dimensions of such a recapitalization are by no means huge compared with the needs of Indonesia, Malaysia, and South Korea when financial crisis struck them in 1997; each required capital of 10 to 15 percent of its GDP. But the challenge is enough to force a resource-strapped Indian government to set about strengthening the sector and encouraging structural change.

India’s credit risk problem is actually worse than it seems, because it has been disguised by accounting norms less stringent than those of developed economies and by the tendency of banks to roll over past-due loans. Nonperforming loans are reported to be 8 percent of the loans of banking institutions, but international equity analysts and credit-rating agencies say that the real figure is more likely to be 12 to 13 percent. This level of nonperforming loans is close to those of Malaysia (13 percent) and South Korea (11 percent) when the Asian crisis erupted in 1997.

Furthermore, government-directed lending obligations in India are higher than those in many Asian economies, and this increases the burden of nonperforming loans because so much of the lending has ended up in poorly managed companies. Foreclosure and bankruptcy laws are weak and have failed to provide an effective framework for the speedy recovery of bad loans.

The need to improve productivity through investments in technology accounts for one-third of the capital requirements of Indian banks. Most of them—public and private alike—suffer from low productivity, measured by the number of transactions relating to deposit and loan accounts per hour for each employee. Public-sector banks currently function at 20 percent of the productivity level of the best Indian bank and at 10 percent of the productivity of US banks.

Recapitalization won’t be easy. As the largest shareholder in the banking system, owning almost half of India’s banks, the government has in the past recapitalized financial institutions to the tune of about $5.5 billion. But its need to control the fiscal deficit—already 6.1 percent of GDP in 2000 and 11.4 percent including the deficits of India’s states—will prevent it from allocating extra capital to meet future requirements.

Capital markets are unlikely to come to the rescue, though they have been open to banks for the past six years. Indian banks have cumulatively raised only about $1.6 billion through domestic and international capital markets, which show little interest in most of the country’s government-owned and private banks because of their weak performance.

In short, the government will have to look to private domestic and foreign investors to revive the sector. Such a move could serve them well given the underlying size and growth of the Indian financial-services market. In particular, McKinsey research suggests that revenues from personal financial services will grow by more than 10 percent a year in real terms over the next five years—more than double the expected growth rate for the sector in Asia as a whole.

Private investors could capitalize on current and future reform if they were to buy into the best-run private banks and into distressed public-sector banks with strong customer franchises, thus aiding the most attractive part of India’s banking system and restoring to health important pieces of it. To complete the surgery, the government must allow unviable banks either to merge with stronger ones or to die, while treating the weakest banks’ weakest point: namely, nonperforming loans.

The government has started to tackle the problem of these loans by considering the idea of converting a large financial institution into a centralized bad-asset management company. This could be the right approach, but it won’t be enough unless it is accompanied by the acquisition of skills drawn from the private sector, by changes in the foreclosure laws to speed up the repossession of assets, and by the removal of archaic legislation that protects defaulting companies. In another positive move, regulators are addressing faulty corporate governance in Indian banks, by, for instance, obliging them to strengthen their boards. Weak corporate governance has contributed to the high level of nonperforming loans and to low productivity.

Among the most encouraging trends, but one that must be accelerated, is the government’s acceptance of some degree of consolidation—for example, the acquisition of small banks by strong local institutions. One case in point was the purchase, by the Life Insurance Corporation of India, of a 27 percent stake in the profitable government-owned Corporation Bank. A proposed amendment to the Bank Nationalization Act will reduce the minimum government share-holding in banks to 33 percent, from 51 percent, certainly a long overdue and welcome development. This step means that the government will give up its controlling stakes, but it will also have to accept changes in the character of these institutions’ governance.

The government is also tacitly approving foreign ownership of banks. Apart from letting overseas investors increase their ownership stakes to 49 percent, it has allowed foreign banks to acquire Indian ones on a case-by-case basis. Thus ING has taken a 20 percent stake in the Indian private-sector regional Vysya Bank, and Chase Capital has bought 15 percent of HDFC Bank (which is the banking arm of India’s private Housing Development Finance Corporation). Indeed, banks with strong corporate governance and sound asset quality, such as HDFC Bank and ICICI (Industrial Credit and Investment Corporation of India) Bank, already command a substantial premium over the rest of the sector (Exhibit 2).

In our judgment, however, full-scale foreign ownership of banks that are currently owned by the government is unlikely to come for at least three years, given the lack of political consensus on the issue and opposition to privatization. Rapid banking consolidation is also being held back by restrictions on the voting rights of bank shareholders other than the government and by the requirement that the central bank approve transfers of shares amounting to more than 5 percent of a bank’s capital. Furthermore, securities market rules prompt mandatory open offers when 15 percent of a bank’s shares are acquired. Such regulations must be eased, and strong local and foreign institutions must be permitted to own majority stakes in banks if India is to attract more investment into its struggling banking system and, ultimately, to defuse a significant threat to its economy.