What remains to be done?

The development of a robust banking system in low-income countries will require that countries address the underlying weakness of the system—not the symptoms. This is a long and arduous process that requires consistent actions on many fronts to complement the effort on the fiscal front. For Sub-Saharan countries that means building a sound and efficient payment system (in many countries, it still takes two to three weeks to clear a check). It also means restoring the safety and soundness of the financial system (banking and nonbanking), which at a minimum provides basic banking services, particularly trade finance, to a broad segment of the population and firms. And it means introducing better accounting, legal, and supervisory systems and a major effort to upgrade skills at all levels. In Asian economies, it means continuing to restructure and privatize public banks, opening them to competition from the private sector, strengthening prudential regulation and supervision, and developing capital and money markets.

Sever the link between banks and loss-makers

In most countries, the most difficult challenge is to sever the link between nonperforming public enterprises and state banks by cutting the enterprises off from new credit and collecting outstanding loans. It has proved difficult, if not impossible, for state banks to enforce a hard budget constraint on these enterprises. State banks not only roll over loans to public enterprises as they come due, they even increase their exposure—often through the capitalization of unpaid interest. When central banks attempt to enforce credit ceilings, enterprises often build up large intercompany arrears, including those to private firms. As these arrears accumulate and threaten the solvency of banks and enterprises, the central bank is forced to relax credit again.

These failings are especially severe for large public utilities and heavy industries. Although among the most inefficient and financially strapped enterprises (often because they sell their products or services at subsidized prices, and often to public entities that do not pay their bills), they are not allowed to close because they provide essential or strategic services. At best, hard budget constraints can be effectively imposed only on private firms or small public enterprises—and these are already starved for capital and prime candidates for liquidation or privatization. The result is that state banks continue to accumulate large losses and require frequent recapitalization. The drain of big enterprises is likely to worsen as they face increased competition from imports and domestic private producers. That is why it is so urgent to privatize these enterprises and to use the proceeds to reduce high-cost government debt that crowds out the private sector and increases real interest rates. The proceeds could also be used to restore the integrity of the contractual savings systems that governments have often used to finance their deficits. But governments will be persuaded to act on these difficult issues only when the true costs of recapitalization and inaction—in terms of money and lost job opportunities— are made transparent.

Go beyond recapitalization

One of the main reasons restructuring of banking systems has often failed is that governments normally have borne the entire cost and have not changed the incentives facing the banks and their managers. Restructuring has not extended to banks' incentives system or the structure of the banking sector. Losses have seldom been shared with other stakeholders (borrowers, depositors, shareholders, and managers) or accompanied by vigorous loan recovery efforts. In most cases, restructuring has merely transferred nonperforming

One of the main reasons restructuring of banking systems has often failed is than governments normally have borne the entire cost and not changed the incentives facing the banks and their managers loans to loan recovery agencies, where they languish uncollected. Loans went from being undermanaged in the banks to unmanaged in these agencies. In Ghana, less than 20 percent of debt was recovered, and in Cameroon and Tanzania, less than 5 percent. In Kenya, the new bank established to take over the bad loan portfolio of commercial banks has itself become financially distressed—because it was forced to lend to troubled state enterprises rather than pursue delinquent borrowers. In contrast, Chile's successful bank restructuring of 1986 required the banks, under new ownership and management, to use a large share of their profits to repurchase nonperforming assets from the central bank, where they had been placed in special accounts. This gave the banks the incentive and the breathing space to work with their clients to maximize loan recovery.

In general, bank restructuring is extremely difficult and demanding, especially in countries where legal and management skills are in short supply and where the private sector lacks the means to buy the companies or their assets. This is particularly the case in Eastern Europe, where cutting off credits to public enterprises imposes, at least in the short run, unacceptably high costs on the economy through the loss of productive capacity and sharp rises in unemployment. In this case, what is needed is Chapter I type bankruptcy proceedings, which allow for a temporary moratorium on enforcement of creditors' claims, to permit an examination of the feasibility of a reorganization, and during which banks continue to lend to the enterprise so it can function during the reorganization. The challenge, of course, is to prevent the moratorium from delaying the reform process. One such approach is being implemented by the Kyrgyz Republic. It has established a solvency resolution scheme where the government budget, and not commercial banks, will provide, in a transparent manner, the financial resources required by public enterprises to support the cost of care and maintenance, orderly liquidation (in particular, funding of severance payments), or passive restructuring. At the end of the process, nonviable enterprises will have been liquidated, and successfully restructured enterprises will resume normal relations with the banking system on commercial terms and be offered up for sale immediately (box 4.4).

Reduce the role of state banks

The lesson for low-income countries is that incentives must change alongside restructuring efforts. And this involves bank privatization and changes in management—since it is much easier to establish an arm's length relationship between properly regulated and supervised private banks and their private clients. In Africa, Guinea-Bissau, Madagascar, Mauritania, and most CFAzone countries have privatized or liquidated banks as part of bank restructuring, but this has yet to happen in a large number of countries.

In general, liquidation and privatization has proved easier for the small and medium size banks, particularly when it is part of an overall reform program that allows new entry in the banking system. Pakistan and some CFA countries are examples. In 1990, Pakistan successfully privatized two of the smallest public sector banks—the Muslim Commercial Bank and Allied Bank Limited. Within three years, the banks tripled their deposits and doubled their profits, largely through improved services, cost cutting, and a vigorous collection effort. In 1991, the government allowed private banks to compete with public banks, which were still reeling under the pressure of $2.4 billion in nonperforming loans. Nine new local banks were licensed that year. Foreign banks also have expanded their business: with only 75 of 7,740 branches nationwide they have captured more than a quarter of banking system deposits. The interest rate spread of public sector banks dropped by 33 percent in less than two years because of increased competition.

It should also be stressed that the success of banking reform in Pakistan owes much to general economic policy, which has liberalized the foreign exchange market and trade, privatized large public enterprises, and attracted foreign investment. These measures in turn increased the demand for financial services by a vibrant and increasingly sophisticated and discriminating private sector.

Some countries are adopting a more measured approach in restructuring their banking system. In addition to allowing new entry, they are partially privatizing state banks. India's largest state-owned bank raised $700 million through a public equity issue to 2.3 million investors, augmenting paid-in capital and reducing the central bank's share from 98 percent to 68 percent (expect ed to fall to 55 percent after a second public offering). This approach will be replicated in other state-owned banks, which are expected to raise private equity capital up to 49 percent of their share capital, with corresponding representation of private shareholders on their boards. The banks are undertaking a major effort to collect on some $12 billion in nonperforrning loans (6 percent of GDP and 12 percent of the assets of the banking system), and intend to sell their urban real estate holdings. These measures are intended to reduce government funds needed for recapitalization.

But liquidating or privatizing the state banks that account for the bulk of banking system assets is difficult—both economically and politically. Yet maintaining their dominant position, even while allowing new entry, does little to change their performance, as Tanzania shows. An alternative is to downsize banks as a way of reducing the cost of restructuring and to seek management contracts with reputable domestic and foreign banks, preferably with a preferred equity position, to run the downsized public banks until improvements attract suitable buyers. Furthermore, as is being contemplated in Macedonia and proposed in Tanzania, the government could break the larger state banks into competing networks that service both urban and rural areas—as they were prior to their nationalization in the 1960s and 1970s— and then sell them to reputable private domestic and foreign banks.

Balance competition with solvency

Competition and privatization are not an end in themselves. They are a means to improve the quality and reduce the cost of financial services to a broad segment of the population. But a warning is in order: experiments with liberal entry without adequate supervision have failed in Africa. In Nigeria, several poorly managed, undercapitalized banks speculated in the foreign exchange market, and their losses now threaten the deposit insurance institution's solvency.

The challenge is to balance the goal of increasing competition and the need to ensure the solvency of financial institutions. Many low-income countries lack the resources to regulate their banking system. Given the high risks in the economic environment and the weak management and information systems, a system that offers extra cushions against risk is desirable. Governments might consider requiring capital adequacy ratios well above the recommended 8 percent risk-weighted capital regarded as sound by international standards (or raising liability limits for bank owners) and providing some co-insurance from depositors. Low-income countries could encourage reputable foreign or jointventure banks to help diversify and deepen their financial systems. Reputable banks, seeking to preserve their reputation, are less likely to operate objectionably even if they cannot be consistently well supervised.

A market-determined interest rate is an integral part of a competitive banking system. But to avoid the detrimental effect of high real interest rates, interest rate liberalization has to be phased in with a macroeconomic stabilization program that reduces fiscal deficits. Also essential is a prudential regulation and supervision system that requires banks to enforce hard budget constraints on their borrowers, improve the quality of their portfolio, and improve their capital structure.

Better supervision and prudential regulations. Most countries have strengthened laws and regulations, and some have improved their supervisory and enforcement powers. Indeed, IDA has been particularly active in training. But the information and skills needed to apply these standards remain limited. There are few trained banking supervisors, accountants, and auditors—and offsite and on-site inspection is ineffective. Many countries lack basic accounting and auditing standards, and even where they exist auditors are too pliable. In some West African countries, the banking sector was not covered by a standardized system of accounting and auditing rules until recently. In such cases, neither banking supervisors nor depositors and creditors can detect incipient financial distress.

Because building these skills will be slow, low-income countries should consider twinning or subcontracting arrangements with established foreign institutions such as central banks and qualified accounting firms. A few countries (Bolivia, India, and even Switzerland) are successfully using major accounting firms to supplement their own supervisory authority.

Clear exit procedures, implemented quickly at the point of insolvency. Ultimately, discipline in a competitive banking system comes from the threat of failure. Designing appropriate mechanisms for handling bank crises remains a difficult task, because it requires balancing the objectives of preservation of overall banking stability and the possibility of individual bank failure. Frequently, governments step in to prop up insolvent banks because of the possibility of contagion to other institutions, the potential disruption of the payment system, and the state's fiduciary responsibility to depositors as nominal bank supervisor. Bank supervisors become discouraged because of weak political support for their technical recommendations, and a culture of passivity and unprofessionalism seeps into regulatory agencies.

The principal lesson from many countries is that strong professional standards for bank supervision are no substitute for effective political action. Bank closure invariably escalates to the highest political level, particularly in underdeveloped financial systems with a high degree of concentration. Governments that have succeeded in dealing with this difficult problem have typically established, ahead of time, detailed mechanisms and procedures for quickly intervening in insolvent banks in a way that depoliticizes, to the extent possible, the exit process. These measures include replacement of bank management, a vigorous collection effort, sale and disposal of assets, prosecution of fraud, and allocation of losses to shareholders, depositors, and taxpayers—in that order. Quick application of these procedures minimizes chances of contagion to other parts of the financial system and reduces the cost to the government and the risk of recurrence.

Serve small borrowers

Even after the banking system is substantially restructured, small and medium-size enterprises (10-200 employees), microentrepreneurs (fewer than ten employees and including large numbers of the self-employed), and the rural poor are likely to be underserved because of high risk and transaction costs.

Many low-income countries have instituted special programs to provide financial services, particularly credit, to farmers at subsidized rates—often with the support of donor agencies, including IDA. But default rates have been high. Credit intended for the poor often has been preempted by wealthy farmers and well-connected entrepreneurs. Savings mobilization has been poor and transaction costs high. And commercial banks have been reluctant to undertake voluntary lending, while specialized rural financial institutions have not become financially self-sufficient and often have collapsed.

There has been a general failure to establish viable financial intermediaries that can mobilize resources and channel them into productive agricultural and rural enterprises. The same goes for microenterprise programs instituted and managed by government in urban areas. A recent Bank review of lending of $3.7 billion to small and microenterprises in thirty countries found that performance was best in Latin America, where loans to the individual borrower averaged $22,600 and repayment rates were 92 percent. The worst performance was in Africa, where the average loan was $143,400 and only 62 percent was repaid. Latin America had more competent banks and a more dynamic small business sector—and better program designs.

Small borrowers value safe, reliable, and convenient savings and banking services. Postal savings have been used effectively in many East Asian countries to nurture a savings culture and mobilize rural savings, particularly following reforms that changed the terms of trade in favor of farmers. These systems offer deposit instruments to help with savings and liquidity management, and payments instruments to facilitate small transactions. Similarly, small borrowers value easy access to short-term loans. Such loans and services have been successfully provided by organizations that provide savings and credit at market prices on a permanent basis—instead of short-lived subsidized credit programs, which often fail to reach the intended target group.

Many non governmental financial intermediaries and non governmental organizations—relying on local savings, careful credit screening, and lending at market rates—have reached a large number of borrowers, particularly women. These organizations reduce default risks significantly through careful selection of borrowers, provision of technical assistance, and prudent use of collateral and joint liability systems. Borrowers open savings accounts, and only those that have repaid earlier small loans are lent larger sums. Such organizations include credit and loan associations, mutual banks in Guinea, Mali, Nigeria, Rwanda, and Zimbabwe, the Grameen Bank in Bangladesh (1.5 million borrowers), Banco Sol in Bolivia (40,000 borrowers), BRI Kupides in Indonesia (1.9 million borrowers and 10 million savers), the Kenya Rural Enterprise program (5,000 borrowers), ACCION in Latin America, Women's World Banking in Africa, Asia, and Latin America, and the Fundes Foundation of Switzerland in Latin America.

These institutions have good track records in outreach and enterprise sustainability, two of the most important criteria in microfinance. But their lending is hampered by small capitalization and the high transaction costs associated with providing technical assistance to numerous small borrowers. These organizations need help in the form of seed capital and technical assistance to grow enough to exploit scale economies, evolve into licensed financial institutions (as Banco Sol in Bolivia and Corpo Sol in Colombia), or have refinancing facilities with banks. Seed capital and technical assistance should, however, be temporary and conditional—first on operational self-sufficiency (revenues covering all nonfinancial costs) and later on full self-sufficiency (no subsidies). Within five to seven years, the better institutions should be able to access capital markets and attract private investors, as some of them are successfully starting to do (box 4.6).

Improve collateral and debt recovery laws

The laws for collateral hamper sound lending and reduce access to credit for many groups. Simple changes in these laws would enable the use of collateral in obtaining credits, a boon to small borrowers (see chapter 2, box 2.4). A related issue is the functioning of the land markets, where better titling registries could improve access to loans. More attention also needs to be paid to expediting debt recovery procedures, which are now slow and cumbersome. Defaulters exploit the protection that laws afford them and creditors are unfairly penalized by the slowness of the court system. As a result, many loans are not forthcoming.