The Future of State-Owned Financial Institutions

How Much State Ownership, and Why?

There is increased evidence from official development institutions and private economists around the world documenting the linkage between more rapid—and stable—economic growth on the one hand, and sound financial systems on the other.

Despite numerous privatizations over the past decade, publicly owned banks and other state-owned financial institutions still serve the majority of individuals in developing countries, according to presentations by James Hanson, George Clarke, and Robert Cull of the World Bank. State-owned financial enterprises are less prevalent in developed economies, with very few exceptions, such as Germany and, to a lesser extent, the United States, with its large government-sponsored entities supporting residential home ownership that have implicit government backing, according to David Marston of the International Monetary Fund. Public ownership of these financial institutions and others has been rationalized on several grounds:

  • To counter the power of strong private sector banks or to promote the development of home-grown banks in the early stages of an economy's history—the so-called infant industry rationale. Both arguments helped justify the formation of the First and Second National Banks of the United States in the early 1800s, for example.

  • To ensure that economic growth is consistent with national objectives. This is a clear rationale for socialist economies, but even in private economies there is a view that governments have better knowledge of socially beneficial investment opportunities than private banks.

  • To ensure that underserved groups or sectors, such as agriculture and small businesses, receive credit.

  • To respond to financial crises, which have hit developed and developing countries alike. In some of these cases, government ownership is temporary, but in some cases it lasts for significant periods.

Among government officials around the world there is support for some government ownership of financial institutions based on one or more of these rationales. Economists generally, however, are skeptical of all these rationales except for the last one.

Performance of Public Sector Banks

With rare exceptions, public sector banks have performed poorly by conventional financial measures, such as returns on equity or assets, the extent of nonperforming loans, and expense levels, according to James Hanson. In principle, these banks may fare better if their broader social missions are taken into account (for example, when they finance roads or sewers), because the benefits to the entire economy may exceed those to the specific borrower. But in practice these banks tend to extend much if not most of their credit to large borrowers, which suggests that social returns are not larger than private gains.

Public sector banks often provide subsidized lending and directed credit to special industries or enterprises identified by the government. They also can burden their governments with large contingent liabilities arising from explicit guarantees or the implicit assumption that some of these banks are "too big to fail." In China, for example, nonperforming loans of major financial institutions at the end of 2003 stood at 2,440 billion RMB (the Chinese currency), equivalent to about 18 percent of the loans of these institutions and 21 percent of gross domestic product, according to a study by Nicholas Lardy of the Institute for International Economics. Furthermore, as much as 90 percent of these might be regarded as a government contingent liability. Public sector banks have also demonstrated a poor collection record with their borrowers, especially in bad economic times, and thus tend simply to "roll over" their loans.

These patterns help account for the negative relationship between economic growth and state ownership of banks found by various researchers. They also help explain why independent rating agencies such as Moody's find that, relative to their private counterparts, state-owned banks tend to be less well capitalized, less profitable, and have thinner core earnings—and thus typically have explicit or implicit guarantees to depositors regarding the safety of their funds. For all these reasons, a number of authors urge governments that own financial institutions to be more transparent in their financial results, in the amounts of explicit (and implicit) subsidies that governments extend to them, and in the government's contingent liabilities to them, according to research by James Hanson, by David Marston and Aditya Narain of the International Monetary Fund, and by Manal Fouad, Richard Hemming, Davide Lombardo, and Wojciech Maliszewski, also of the IMF.

Several types of cost arise from public sector banks, including directed lending, subsidies, and fiscal (macroeconomic) costs. For example, the two main public sector banks in Uruguay—Banco de la Republica Oriental del Uruguay (BROU), the former central bank, and Banco Hipotecario del Uruguay (BHU)—lend at subsidized rates and therefore cater mainly to poor credit risks in the agriculture sector and to a lesser extent in other sectors in crisis. The cost of BROU's and BHU's interest rate policy can be measured in the lower profitability of these banks compared to their private sector competitors: the after-tax return on equity for private banks averaged 12 percent in 1995–2000, but only 4.7 percent for BROU and 1 percent for BHU. At the macroeconomic level, the quasi-fiscal activities in Uruguay may have cost between 0.8 and 1.2 percent of gross domestic product on average between 1996 and 2002, according to Manal Fouad.

Any fiscal analysis of this type also should take into account the tendency of governments to overtax through parsimonious provisioning rules that understate the true extent of loan losses and lead to erosion in real bank capital. As a result, the fiscal burden imposed by state-owned financial institutions may be much higher than some of the contributors have calculated.

Country Experiences with State-Owned Financial Institutions

China is perhaps the best-known example of a country with dominant public sector banking. Most of its banks are owned by the central, provincial, or local governments, and nonperforming loans to other state-owned enterprises loom large in its economy. The outlook for the Chinese banking sector is mixed, according to Nicholas Lardy. On the one hand, the official returns on assets of China's state banks dropped precipitously during the 1980s and 1990s, while nonperforming loans grew rapidly during the mid-1990s, reaching 25 percent of total bank loans by 1997. On the other hand, Chinese officials in recent years have recognized the importance of dealing with the banks' problems. They therefore have reduced the government's involvement in the banks' allocation of credit and have created four asset management companies to deal with the nonperforming loans. The results so far are impressive—nonperforming loans have declined significantly (largely because many were transferred to the asset management companies), while the banks appear to be more efficient. Yet the outlook remains guarded: lending by Chinese state-owned banks soared in 2003 and early 2004 and was initially resistant to attempts to check the expansion, which raises the risks of higher nonperforming loans in the future.

The experiences of state-owned banks in other countries are also mixed. A relatively positive picture emerges from the portrait of the largest such institution in South Africa, as outlined by Lewis Musasike, an executive manager at the Development Bank of South Africa.

Indonesia had well-known problems with its state-owned banks and connected lending among its private banks before the Asian financial crisis of 1997-98. Since then, the state-owned banks have been recapitalized by the government (their bad debts taken over by a separate agency), and prospects for them may now be somewhat brighter, judging from a presentation by Hendrawan Tranggana, managing director of compliance at Bank Rakyat Indonesia. Two economists from the World Bank—P. S. Srinivas and Djauhari Sitorus—paint a somewhat darker view; although recorded profitability may have improved, large uncertainties remain about the magnitude and ultimate cost of the banks' nonperforming loans.

Ishrat Husain, governor of the State Bank of Pakistan, discussed Pakistan's government-owned banks, which have had the more common experience, particularly its agricultural development bank: high nonperforming loans, overstaffing and other inefficiencies, and poor customer service. Through much of the 1990s, however, Pakistan has been privatizing its banks while beefing up the prudential regulation of privately owned banks—so far, with good results.

Outside of China, government ownership of banks is highest in India, where it accounts for about 75 percent of all banking assets. India has used state ownership to allocate resources to both the public and private sectors and to help provide credit to the mass of low-income individuals and businesses not served by privately owned institutions, according to Urjit Patel, the chief policy officer at India's Infrastructure Development Finance Company, Ltd.

There is now clear evidence that government ownership and involvement in the banking sector has become excessive, not only costing taxpayers money (through bailouts) but also increasingly giving preference to government securities instead of to loans to their intended borrowers. As long as this situation continues, India's economic growth rate—which has surged in recent years with the opening up of the economy to foreign investment and the growth of a vibrant information technology sector—will remain lower than it could be with more private sector allocation of credit.

Privatization: The Ideal Course

From strictly an economic point of view, the optimal policy for governments that own financial institutions—given the poor performance record of these institutions throughout the world—is to privatize them. This course not only can save governments money—by eliminating subsidies—but also can improve the performance of their economies by ensuring that credit is channeled through the market rather than through government officials. The banks themselves are also likely to be more efficiently run.

There are various routes to privatization, however, each with advantages and risks. Privatized banks appear to do best when sold to a single strategic investor; but this course carries with it obvious political risks—that governments have "sold out," especially when the buyers are foreign. Excluding foreigners can prove costly, however. In Mexico's first round of bank privatizations in the early 1990s, for example, the banks subsequently expanded too rapidly—behind the wall of protection from foreign competition—and later had to be renationalized. Other countries, notably in Eastern Europe, have had similar problems, underscoring the importance of a sound regulatory framework for privately owned banks. Research by Robert J. Cull and George Clarke, senior economists at the World Bank, and Mary Shirley, president of the Ronald Coase Institute, shows that when foreigners are not allowed to bid, governments are likely to reap less revenue when they do privatize and to lose the advantage of the banking know-how and skills that foreign institutions can bring to many economies whose financial sectors lack them.

Clarke, Cull, and Shirley say that however privatization is accomplished, it should not be done partially. If the government still holds a portion of the stock, especially a majority of the stock, the banks still tend to perform poorly by conventional financial measures.

Another way to privatize banks is to sell them, in part or in whole, to employee stock ownership plans. So far, twenty countries have used this technique, though the tendency is for these plans to own only a minority interest, along with other private investors; in some cases, the state even retains some ownership, according to a presentation by David Binns, vice president of the Beyster Institute/Foundation for Enterprise Development, and Ron Gilbert, president and co-founder of ESOP Services, Inc.

Stock ownership plans predictably tend to be most successful when subsequent control is vested in a single strategic investor, which provides capital and know-how. Some experts believe that the plans should complement a strategic privatization scheme and not be considered in isolation. Many feel that foreign strategic investment provides the surest method of privatization, by offering independence from vested interests and an infusion of outside expertise.

Yet another route to privatization is by offering shares to the public through an initial public offering. Not surprisingly, developed economies that sold off their state-owned financial enterprises first used this technique; it has been used since the early 1990s by a number of developing countries. So far, thirty-three of the eighty countries outside the Organization for Economic Cooperation and Development (OECD) have privatized 156 state-owned banks. Initial public offerings account for 44 percent of all emerging country bank privatizations, according to research by Fred Huibers, global head of equity capital markets at ING Barings.

There are side advantages to privatizing through an initial public offering. For one thing, these offerings can help develop local stock markets, since the privatized institutions become some of the largest enterprises on these exchanges. This effect is not automatic, however, and the cornering of the markets in privatization vouchers marked some of these efforts. More developed stock exchanges, in turn, facilitate access to capital by other types of domestic enterprise and thereby can improve overall economic growth. Although in principle governments may also realize more revenue by selling shares to many investors than to a single strategic investor—especially if the initial public offerings are carried out when market valuations are high—empirical studies find that governments tend to underprice their share offerings in order to entice citizens to participate in shareholding. But even this outcome has a potentially significant benefit: By stimulating wider ownership of the privatized enterprises among many investors, initial public offerings effectively inhibit the ability of governments to later renationalize the institutions.

Governments damage confidence in privatizations led by initial public offerings if they retain an ownership interest in the "privatized" institutions and continue to direct their lending toward favored borrowers. Such behavior not only perpetuates practices antithetical to general economic growth but also, by damaging the earnings prospects of the institutions, clearly can harm investors who have purchased their shares. Also, if the public offerings can be (and have been) carried out in a way that limits information available to investors, then investors are likely to overpay when they do buy shares. Shareholders face even greater problems where the institutions whose shares they purchase subsequently are not effectively supervised, as happened in the first bank privatization in Mexico, according to Fred Huibers.

Experiences in Pakistan and Uganda

Ishrat Husain and Louis Kasekende, the World Bank's executive director for Uganda and former governor of the Central Bank of Uganda, shared perspectives on the comprehensive privatization programs in their countries, and the strategic vision of their countries' policymakers. Husain suggests that Pakistan's experience demonstrates the benefits of privatization for the banking sector and the economy, but he warns that it must be handled carefully. Some of the key lessons are that the legal framework is critical; the central bank must be independent; state-owned banks must be independent; the assets of state-owned banks must be easily disposed of; and fit and proper criteria for privatization must be strictly applied. Both experts note that tough measures must be taken before attempted privatization—cutting down on excess staff and overbranching—aroused countervailing political pressure. At the same time, Kasekende warns that preprivatization "beautification" should not be overdone, as it may be unduly costly and bring little long-term benefit.

Both of these experts also agree that selecting the right buyer was a critical decision in their countries. Kasekende emphasizes the importance of conducting a comprehensive market survey of viable sale options, thorough due diligence of the investor's market reputation and not just capital strength, and also of understanding the investor's future intentions to strengthen capital and management of a bank. Husain highlights the valuable role of international bank managers and professional specialists and recommends that governments actively seek their involvement during privatization. Both suggest that governments need to deal swiftly and decisively with nonperforming loans and to reverse the weak repayment ethic among borrowers from state banks. This works best when fit into a broader privatization program for state-owned enterprises and when efforts are made to deepen capital markets as alternative sources of financing for large enterprises. With a strong and stable financial system as the objective, both authors agree, revenue considerations should be secondary in planning the privatization.

Managing State-Owned Institutions

Where they exist, state-owned banks still have powerful constituencies: their borrowers, their managers, and of course, many officials in government who still see the institutions as an "off-the-books" way to allocate capital toward uses believed to enhance economic growth, despite overwhelming evidence to the contrary. For this reason, as a practical political matter, various authors in this volume—even those skeptical of state ownership—believe that many if not most state-owned banks will continue to exist.

Under these circumstances, are there "second-best" options that governments can pursue either to minimize losses associated with state ownership or—in the best of worlds—to achieve some positive results? Unfortunately, the record so far on this front has not been encouraging; by and large, efforts to replace poor managers, write off bad loans and replace them with supposedly safer government debt, reduce staff and expenses, merge troubled institutions, and improve information technology have not been successful, according to James Hanson of the World Bank. This does not mean that these steps should not be tried when divestiture is not possible, but the rare instances of success must be acknowledged.

At the very least, the institutions can and should be effectively supervised, just as if they were privately owned. This, unfortunately, is not currently the case in most parts of the world, and it must change. Research by Jonathan Fiechter and Paul Kupiec of the International Monetary Fund shows that state-owned financial institutions not only pose systemic risks to their economies but also threaten governments with potentially significant liabilities and thus the need for a rise in taxes or a reduction in other government spending—or both—to make up for such financial shortfalls. Unfortunately, the inherent conflict of interest in both owning and supervising banks is difficult to resolve.

Oversight is not likely to be effective, however, unless the overseers are independent and their missions well defined. Other codes of conduct are also relevant, including: careful limits on the activities of (and thus the risks posed by) the institutions; rules governing minimum capital (and adherence to those rules); supervision of the institutions' internal controls and other means for limiting risks and expenses; and annual, honest reports by the supervisors of what they have found. The World Bank and the International Monetary Fund can help countries implement these basic steps, perhaps beginning by organizing regular meetings of supervisors of state-owned financial institutions to exchange views and best practices.

Other Alternatives

When a state-owned financial institution cannot be either sold or successfully reformed, there are still three other options: The institution can be closed, it can be converted into a government agency, it can be converted into a "narrow" bank (a depository holding only government securities as assets), or it can be closed. A few governments, mostly in Eastern Europe, have closed banks, and the banking systems in these countries survive and even prosper without the troubled banks. Conversion to an agency makes the bank an official arm of the government and therefore makes its activities more transparent, though this conversion may complicate efforts to collect loans, as in China. Narrow banks have been supported by a number of economists in other contexts (including Robert Litan) but are not well suited for entire banking systems, especially in developing countries, where there are few if any alternative providers of credit.


State-owned financial institutions are likely here to stay in many countries, for political rather than economic reasons. But there are signs of change. In Egypt, for example, authorities are demonstrating a willingness to reduce the roughly 65 percent state ownership in their banking system by allowing at least one of the three large state banks to be sold.

Even when politically motivated, governments that own financial institutions would serve their citizens better if they made the financial commitments to those institutions more transparent, managed them more soundly, and restricted their activities to a few sectors in which the social returns from extending credit are likely to exceed the private returns.