State-Owned Banks, Stability, Privatization, and Growth: Practical Policy Decisions in a World Without Empirical Proof

Contributor Notes

Author(s) E-Mail Address: mike@amandrews.ca

This paper provides an overview of the possible linkages between state-owned banks, privatization, and banking sector crises. Data on privatizations in over 65 countries is used together with data from the banking crisis literature to consider the relationship between state-owned banks and financial sector stability. The paper draws on the existing literature to provide guidance to policymakers regarding bank privatization.

Abstract

This paper provides an overview of the possible linkages between state-owned banks, privatization, and banking sector crises. Data on privatizations in over 65 countries is used together with data from the banking crisis literature to consider the relationship between state-owned banks and financial sector stability. The paper draws on the existing literature to provide guidance to policymakers regarding bank privatization.

I. Introduction

Does it matter whether the state owns some or all of a country’s commercial banks? There have been relatively few attempts to answer this question directly, but the extensive literature on state-owned enterprises more generally implies that it does matter. More tellingly, policymakers in many countries at all stages of development have opted for bank privatization over the last 25 years, reflecting a growing consensus that state-owned banks are less desirable than privately owned banks. This growing preference for private bank ownership may be due to an expectation of greater financial stability and higher economic growth.

The propensity of countries to privatize banks after a systemic crisis reinforces the widely held view that state-owned banks are bad for financial stability. Stability and growth, of course, are not independent. Banking crises result in significant fiscal costs and even more significant losses in output. Governments in many postcrisis countries have clearly decided that one way to avoid the fiscal burden of repeated recapitalizations of state-owned banks is to privatize. Privatization is frequently part of the package of policy measures intended to strengthen the financial system, reducing the likelihood of future crises and the associated output losses.

Even if full blown crises are avoided, the distortions introduced by state-owned banks can make the financial sector less able to contribute to growth. There are multiple dimensions to these distortions. State-owned banks may be explicitly required or implicitly expected to finance loss-making state-owned enterprises, or provide financing on noncommercial terms to regions or sectors, or extend credit based on political connections rather than risk assessment. State-owned banks may be inefficient, providing opportunities for inefficient private sector banks to thrive in less than competitive markets, or alternatively permit efficient banks to earn extraordinary profits. State-owned banks may have a cost of funds advantage over privately owned banks due to an implicit or explicit government guarantee. If these funds are used to finance inefficient state-owned enterprises, the result can be a crowding out of private intermediation.

This paper provides an overview of the issues and the existing literature addressing linkages between state-owned banks and growth, privatization, and banking crises. The paper’s new contribution is data on bank privatizations developed from multiple sources (Appendix I), which is used in conjunction with data on systemic crises to consider the following questions:

  • Do state-owned banks cause banking crises?

  • Does privatization of state-owned banks cause banking crises?

This data, together with details of the nationalization of banks during recent banking crises, is also used to consider a third question:

  • To what extent is privatization of state-owned banks after a banking crisis simply a return to the precrisis market structure?

The data indicate that privatization of banks nationalized during crises accounts for only about one-third of the bank privatizations occurring concurrently or within five years of the end of a banking crisis. This suggests that policymakers in post crisis countries have an increased preference for private ownership of banks.2

A recurring theme in this paper is that financial sector stability issues, which include the general preconditions for a sound financial sector as well as the strength of the supervisory apparatus and the soundness of banks themselves, are intertwined with the growth and fiscal issues that seem typically to drive policy decisions regarding state-owned banks. Some suggestions are provided throughout the paper for further research into a topic that will remain important to policymakers for many years to come given the continued prevalence of state-owned banks in many countries.

The balance of the paper is organized as follows. The next section of the paper provides an overview of the case for and against state ownership of banks, noting that regardless of the state of the academic debate, the decision by policymakers to privatize is evidence that government ownership has fallen into increasing disfavor. Section III of the paper uses the data on privatizations presented in Appendix I together with the dates of banking crises and details of bank nationalization during crises to consider linkages between state-owned banks and financial sector stability. Sections IV and V draw on the literature on bank privatization, and privatization more generally, in a discussion of issues of particular concern for policymakers. The penultimate section of the paper provides policy suggestions to mitigate the negative influence of state-owned banks in circumstances where privatization cannot be quickly achieved, and the final section contains brief concluding remarks.

II. The Rationale for and Pitfalls of State-Owned Banks

The number of bank privatizations around the world since the mid 1970s is evidence of how state ownership of banks has fallen into disfavor with many policymakers (Appendix I includes over 235 privatizations in more than 65 countries).3 However, views on government ownership of banks and other enterprises have evolved over time, and policymakers in the post-World War II period were generally much more inclined toward state-ownership.4 As a result, through the middle of the twentieth century governments in many countries became more actively involved in the ownership of enterprises and provision of goods and services of all types. Thus, even with large numbers of privatizations, state-owned banks still play a major role in the financial system of many countries.

In the debate over the proper role of government, banking was commonly included in a list of key sectors or functions that should be government controlled. In developed countries, the premise for government ownership of the financial sector was to control the “commanding heights” of the economy, ensuring among other things that the growth of regions or sectors was not impeded by market failures. In developing countries, additional factors influenced the tendency toward greater state-ownership in the financial sector.

The classic “development view” of state-owned banks is that government ownership can stimulate growth when economic institutions are not sufficiently developed for private banks to meet financing needs. This view, combined with the belief that government should control the strategic sectors of the economy, was “adopted around the world as governments in the 1960s and 1970s nationalized the existing commercial banks and started new ones in Africa, Asia, and Latin America.”5 This trend was reinforced in some newly-independent countries by a resentment of colonial institutions including foreign-owned banks, which were often viewed as favoring the economic interests of their shareholders and large multinational clients at the expense of indigenous individuals and businesses.6

An alternative premise for state-owned enterprises and banks in particular, is that political objectives take primacy over the quest for growth and development. In this “political view,” politicians use state-owned banks and other enterprises “to provide employment, subsidies and other benefits to supporters, who return the favor in the form of votes, political contributions and bribes.”7 State-owned banks are particularly desirable as instruments for the distribution of political largess because their lending activities can influence all sectors of the economy and banks frequently operate large branch networks spanning all or most regions of a country. In addition, the information asymmetry between banks and outsiders makes it relatively easy to disguise political motivations for loans, and the full costs of such loans may be deferred for some time until a state-owned bank recognizes losses on loans made on political rather than commercial terms.8

In both the political and development views, the rationale for state-owned banks is to finance projects that otherwise would not be funded. The difference is that in the development view, the motivation is the more laudable objective of funding economically desirable projects that the private sector neglects due to market failures, rather than more crassly funding politically desirable projects without regard to economic viability. Either case, however, provides strong motivation for the establishment and maintenance of state-owned banks.

Despite these motivations, state-owned banks no longer enjoy the popularity of the 1960s and 1970s. Development successes proved elusive (Box 1). By the 1980s, many of the African governments that nationalized banks in the previous decades faced acute economic crises.9 Latin American crises in the 1980s preceded privatization of many state-owned banks. The general trend toward privatization in Europe in the 1980s included the divestiture of state-owned banks in many countries including France, Italy, Portugal, and Spain. On balance, the costs of state-owned banks came to be seen as outweighing the benefits. In formerly planned economies, transformation of the banking system was central to the transition to a market economy.

Many of the costs and drawbacks to state-owned banks are the same for state-owned enterprises more generally. In addition to susceptibility to partisan political influence, a mandate for commercial viability may conflict with social and development objectives. These can include considerations such as providing nationwide service regardless of economic viability, supporting economic activity in certain sectors or regions, and providing employment opportunities. While this may comprise all or part of the rationale for state-owned banks, even if there is a sound governance structure in place to insulate state-owned banks from direct political pressure, the need to achieve various government policy objectives may preclude the efficiency that a privately-owned firm would achieve in financial intermediation. This can be reflected in explicit subsidies to state-owned enterprises, or more commonly, poorer financial performance than would be expected from a pure commercial entity.

Explicit or implied requirements to finance inefficient state-owned enterprises or directly finance government deficits can further impair the ability of state-owned banks to operate on commercial terms. For example, one of the challenges to bank reform in China is the volume of nonperforming loans extended to state-owned enterprises that are unable to repay the loans, and must themselves be restructured. Use of the deposits raised by state-owned banks for the support of state-owned enterprises may crowd out potentially more productive use of savings by privately-owned intermediaries.

What Can the Numbers Tell Us? State-Owned Banks and Development

Interventionist and inefficient governments, and poor protection of property rights are among some of the most usual suspects in cross-country empirical studies of growth and development. Since state-owned banks so often are prevalent in countries scoring poorly on these measures, it may be difficult to determine whether state-owned banks are really a cause of lower economic growth, or simply a common feature of countries with poor government and institutional infrastructure.

One of the few empirical studies explicitly exploring this relationship, La Porta, Lopez-de-Silanes, and Shleifer (2002) constructs a database of government ownership of banks in 92 countries. This is used with various measures of economic performance, the quality, nature and role of government and the legal framework, and financial sector structure and performance, to consider four questions.

  • How significant is government ownership in various countries?

  • What types of countries have more government ownership of banks?

  • Does government ownership of banks promote subsequent financial growth?

  • Does government ownership of banks promote subsequent economic growth and how does it affect factor accumulation, savings and growth of productivity?

The paper finds that countries with higher levels of state-bank ownership tend to have lower levels of per capita income, underdeveloped financial systems, interventionist and inefficient governments, and poor protection of property rights. While slower economic growth is associated with higher levels of historical state-ownership of the banking system, the paper does not empirically infer causality.

Two other empirical studies fail to find support for the development role of state-owned banks. Building on the data on the La Porta, Lopez de Silanes, and Shleifer data on state-owned banks, Dinç (2002) examines the lending behavior of state-owned banks using financial data for state-owned and private banks. The paper finds that the greater lending and restructuring activities of state-owned banks in election years supports the “political view” of state-owned banks, rejecting the hypothesis that state-owned banks play a beneficial development role. Examining Italian banks, Sapienza (2002) finds that while some behavior such as lower interest rates and favoring particular regions may be consistent with development objectives, only the political view is consistent with these behaviors as well as the influence of election results and political party affiliation.

In addition to the general problems of quantification and measurement of factors influencing development, the state-bank ownership data used in these studies are subject to some limitations. The La Porta, Lopez-de-Silanes, and Shleifer data uses indicators of state-bank ownership in 1970 and 1995. To the extent that state-owned banks are indicative of institutional factors that change only slowly, use of data for only two dates may well be indicative of the influence of state-owned banks on growth. However, when examining specific countries use of data for two dates may not provide an indication of some important episodes in a country’s banking history. For example, the nationalization of most Mexican banks in 1982 and subsequent privatization in 1991–92 falls in between the two dates. Similarly, the nationalization of major banks in response to a banking crisis can be a major blip in the long-term trend of declining state-ownership of banks, and in many countries, 1995 was far from an end point in bank privatization.

III. State-Owned Banks and Financial Sector Stability

Problems in state-owned banks can arise from the same three generic causes that affect private sector banks. Microeconomic causes of problem banks are generally poor banking practices that lead to losses through inadequate management of credit and other risks, or fraud. Macroeconomic shocks such as the 1970s oil crisis, or imprudent fiscal or monetary policies, can lead to losses at privately or state-owned banks. Similarly, structural problems such as an inadequate legal system for the enforcement of contracts can affect bank performance regardless of its ownership.

Even if state-owned banks have only the same vulnerability as privately-owned banks to these three sources of problems, state-owned banks may be more exposed to solvency-threatening losses. This is because the profits of state-owned banks are lower than they otherwise might be,10 reducing the availability of earnings as the first line of defense against unexpected losses, and lessening the ability to generate capital through retained earnings. The macroeconomic environment is the same for private and state-owned banks in a given country, so observable differences between banks with private and state ownership must be attributable to bank-specific factors. There could be a range of factors contributing to the poorer performance of state-owned banks, including objectives other than profit maximization, less competent management, overstaffing and other operational inefficiencies, and less well developed risk management. In addition, state-owned banks may be more or less rigorously supervised leading to lower likelihood of detection of emerging problems and initiation of remedial measures by the supervisory authority.

The lower resilience of state-owned banks can be compounded by the greater vulnerability of state banks to losses on loans and investments made for policy or political reasons rather than on pure commercial terms. While privately-owned banks may be subject to moral suasion, state-owned banks are subject to the directives of the shareholder, which might include support for inefficient state-owned enterprises, for either development or political purposes. Not only does this increase the risk of loss to the banks, it also is a misallocation of capital within the economy.

A. Do State-Owned Banks Cause Banking Crises?

The three empirical studies to examine the relationship between state-owned banks and banking crisis find little or weak evidence of a causal link. This finding may be somewhat surprising given the evidence from studies of bank privatizations indicating that state-owned banks have poorer financial performance than private banks. However, individual bank problems do not necessarily lead to systemic crisis, and also the lack of a causal link may be because state-owned banks tend to be more prevalent in countries with other policies and weak institutions that may be more important in causing crises. Thus, a high prevalence of state-owned banks tends to be associated with crises even though there is no empirical proof of causation.

Barth, Caprio, and Levine (2000) find little evidence of a causal link between state-owned banks and the likelihood of crises in an examination of 66 countries. The finding is not consistent with their prior assumption, and they hypothesize that “state-owned banks that encounter difficulties may receive subsidies through various channels, so that the banks are never identified as being in crises.”11 The study uses data on government ownership as of 1997 and, thus, may not capture any effect from changes over time in the percentage of a country’s banking system controlled by state-owned banks. In addition, the focus of the study on regulation and ownership may not adequately control for many other factors that may contribute to banking crisis, particularly the necessary preconditions for an effective regulatory regime. The nature of regulation specific to the financial sector may be much less important than the quality of the general legal infrastructure and government institutions.

La Porta, Lopez de Silanes, and Shleifer (2002) find only a weak relationship between the level of government ownership of the banking system and measures of financial instability in their examination of 92 countries. They hypothesize that this “may be because such factors as the general interventionist stance of the government, its efficiency and the security of property rights may be more important correlates of government bank ownership than are the assorted crises.”12 Their findings indicate that countries with higher levels of government ownership of the banking system “are more backward and statist. They are poorer and have more interventionist and inefficient governments, and less secure property rights. Countries with less developed financial systems also seem to have higher government ownership of banks.”13 Although the study does consider the levels of bank ownership in 1970 and 1995, the findings rely on equations including only the 1995 levels, as the authors note a high correlation between the 1970 and 1995 levels. Use of a single point for government ownership, or even two points, may miss important developments within countries. These can include a cycle of nationalization and subsequent divesture in response to a banking crisis, for example, in the Nordic countries in the 1990s. In addition, the trend toward privatizations may not be fully captured as in many countries significant privatization occurred after 1995. However, if institutional factors are determining factors and government ownership only a by-product of the state structure in countries with poorer infrastructure, similar findings would be expected using additional and more recent data on government ownership of the banking system. While government ownership can be divested relatively quickly, the quality of institutions changes only over time.

The third empirical examination of the relationship between government ownership of the banking system and the likelihood of banking crises also considers whether the severity of crises is increased by higher levels of government ownership. Caprio and Martinez Peria (1999), use the La Porta, Lopez-de-Silanes, and Shleifer data on government ownership of banks in a sample of 64 countries. They find that greater government ownership does increase the likelihood of banking crises, although the model does not control for potentially important institutional factors such as the rule of law, property rights, and government efficiency. The finding that greater government ownership of banks increases the costs of banking crises is not statistically significant, and as the authors note, is subject to significant difficulty in measuring the costs of crises. The use of data at two points in time rather than a series may not reveal a relationship if government ownership of banks, or changes in the percentage of the banking system held by government-owned banks, is a proximate cause of crisis. For instance, poorly-handled privatization might precipitate a crisis, as could nationalization. However, as with the La Porta, Lopez-de-Silanes, and Shleifer findings, if institutional factors that change only slowly are the causal factors, then similar results might be expected using additional data points for government ownership.

Further empirical work addressing some of the difficulties in measurement and modeling identified by these studies might provide evidence of a stronger link between state-ownership and banking crises, but even undisputed empirical proof would likely have little impact on policymakers. The broad trend toward bank privatization has emerged in the absence of such proof, likely because the empirical work does not contradict the theoretical arguments against state-owned banks and the anecdotal experience with state-ownership in many countries. While not providing strong support for a causal link to banking crises, the findings of all three empirical studies are consistent with the body of literature indicating that institutions are more important than other factors as determinates of economic development.14 Recent empirical work indicates that poor institutional structure is more important than the specific regulatory framework for the financial sector.15 Put another way, countries with poor institutional structure are more likely to have state-owned banks and weak public sector governance, and thus are more prone to banking crises. Improving the institutional structure, including reducing the direct intervention of government in economic activities, usually involves reducing government ownership in the banking sector.

B. Does Privatization of State-Owned Banks Cause Banking Crises?

Bank privatization programs, or more accurately, shortcomings in the design or execution of the programs, are sometimes cited as contributing to future banking crises. Mexico and Chile in particular are cited as examples (Box 2).16 However, when the data on bank privatizations are juxtaposed with the dates of banking crises, only a handful of countries are identified where major bank privatizations took place within five years prior to the onset of a banking crisis (Table 1). Some of these countries tend to have experienced “serial” crises over a period of years, indicating great difficulty in addressing the fundamental problems of the banking sector. Failure to establish preconditions for effective banking supervision, deficiencies in the regulatory framework and its enforcement, lack of capital and inadequate managerial capacity were all proximate causes of the ensuing banking crises. A successful privatization may deal with issues of managerial capacity, but a simple change in bank ownership is not enough to address broader financial sector problems.

Table 1.

Countries Experiencing Crises Within Five Years of Bank Privatization

article image

See Appendix II for details on crisis dates.

Kenya (Commercial Bank of Kenya) is an instance where government initially sold only a small ownership stake, retaining majority control for a number of years following the initial move to privatization. In Korea, the privatization of Kookmin had no causal link to the 1997 crisis, although the need for strengthened prudential supervision and hidden weaknesses in the banking sector because evident after the onset of crisis. In the Ukraine, the initial privatizations were through share distribution, a method of privatization that fails to bring new capital or expertise to the bank. Privatization of banks is only rarely associated with banking crisis, but the few instances suggest that partial privatizations are not effective in addressing the weaknesses of state-owned banks, nor are privatizations that do not bring new capital or management skills to the bank.

C. State-Owned Banks and the Management of a Crisis

Once a banking crisis occurs in a country, does the presence or absence of state-owned banks have any impact on the authorities’ reaction to the crisis? Quantifying the extent of problems to develop a viable strategy can be more difficult if state banks have not been subject to the same standard of banking supervision as applied to privately-owned banks. While there is an international consensus that state-owned banks should be subject to the same prudential oversight as private banks, in practice it often happens that regulatory forbearance is applied to state banks. This can be due to an implicit assumption that government ultimately backstops the risk to depositors, mitigating the need for prudential oversight, or because bank supervisors prove unable or unwilling to require that state-owned banks adhere to regulations. This may mean that the asset valuations and reported profitability of state-owned banks may be less-reliable than similar data for privately-owned banks.

The presence of state-owned banks may be a benefit in attempting to stabilize a crisis as state-owned banks may be less susceptible to runs as government may be seen as more able and willing to provide financial support than the shareholders of private banks.17 Nevertheless, in about one-third of banking crises in countries where state-owned banks accounted for 75 percent or more of the banking market, the authorities introduced a blanket guarantee as part of the crisis management strategy,18 suggesting that in the absence of a specific government commitment, depositors may run even from state-owned banks. State-owned banks can be useful in dealing with runs on insolvent private banks if the ability to honor deposits is undoubted. State-owned banks could act as the paying agent under a blanket guarantee or deposit insurance scheme for the deposits of closed banks, and would not require cash or liquid assets for these deposits to the extent that they were retained rather than withdrawn by depositors.

Privatization Precedes Banking Crises in Chile and Mexico

Chile: As part of a broad reform program initiated in 1973, 19 of 20 state-owned banks were sold to private investors in 1975. The bulk of these banks were acquired by financial conglomerates, which were required to make only a 20 percent initial down payment toward the purchase price. Many conglomerates then used loans from the banks to make down payments on nonfinancial state-owned enterprises being privatized. Some changes to the legal framework for banking supervision were implemented in the late 1970s, but were inadequate to deal with the rapidly changing financial sector. By 1981 the banking system was in crisis, leading among other things to intervention in eight banks, central bank liquidity support, and in 1982 and 1984, purchase by the central bank of nonperforming bank assets. In 1985, the central bank participated directly in recapitalizing banks, five of which were subsequently returned to private ownership in 1986. As part of the response to the crisis, the prudential framework was strengthened, and the funding and staffing of the supervisory agency increased.

Mexico: Government sold controlling stakes in 18 banks over 14 months from June 1991 to July 1992. Although some deregulation had taken place, sale of the banks at generally high multiples of book value indicates the purchasers expected the Mexican banking market to continue to be characterized by limited competition, providing opportunities for large profits. Initially this was the case. As spreads widened, however, this tended to mask lack of operating efficiency (Gruben and McComb, 1997). Competition increased more rapidly than expected, when numerous new domestic banks were chartered beginning in 1993, and new regulations in 1994 pursuant to the North American Free Trade Agreement permitted greater foreign competition. In addition to facing increasing competition with little improved operating efficiency, the banks also engaged in significant amounts of related parties’ transitions, and some used derivatives to take risky and leveraged currency positions. Banks came under increasing pressure following the peso devaluation in 1994, and in 1995 a special recapitalization program was introduced to deal with a number of problem banks.

Neither in Chile nor in Mexico can the privatization of banks be singled out as the cause of the ensuing crises. In both cases, privatizations occurred in the early stages of major liberalization programs. Stronger prudential frameworks and better supervision could have mitigated subsequent problems, for instance by restricting insider transactions and imposing more stringent limits on credit and currency risks. In both cases the transformation from a banking system dominated by state-owned institution to privately-owned institutions took place quickly, but it is not clear that there would have been any advantage to extending the privatization program over a longer-time period, apart from the opportunity to make further progress on other needed reforms.

One common tool to deal with crises is the nationalization of privately-owned banks as an alternative to permitting them to fail (Table 2) This has occurred in systems that were dominated by privately-owned banks prior to the crises, as well as in cases where there was already significant state ownership. In times of crises the private sector may be unwilling or unable to provide capital to support the banking system. Faced with a widely-insolvent banking system, many governments have opted to use public funds for bank recapitalization, acting as the “owner of last resort” to preserve essential banking functions.

Table 2.

Nationalization in Response to a Banking Crisis

article image
Source: Updated from Hoelscher and Quintyn (2003).

Privatization has also been a common policy response to banking crises. Of the 65 countries undertaking bank privatizations documented in Appendix I, 39 have also experienced banking crises (Appendix II). Of these 39 countries, 23 undertook one or more bank privatizations concurrently with the crisis or within three years of its end. This is due in part to the divestiture of banks that had been nationalized as part of the immediate response to the crisis, but this explains the majority of privatizations in only about one-third of these 23 countries. For the other cases, the crisis appears to have provided a political impetus for privatization. In some instances, this has been influenced by conditionality attached to IMF programs or World Bank loans. Another factor, however, is that politicians will be in favor of privatization when the political cost of maintaining state ownership outweighs the benefits.19

The political benefits of state-owned banks may be reduced, or less easy to realize, in a post-crisis period. Banks emerging from serious financial distress will be less able to afford credit decisions made on political basis or in furtherance of policy objectives that may conflict with commercial objectives. The need to restructure and rationalize may limit opportunities to provide employment in regions or to political supporters by maintaining unneeded positions. Even if state-owned banks emerge from a crisis financially able to deliver the political largesse, the heightened scrutiny and enhanced governance that may follow large expenditures to recapitalize state-owned banks can prohibit such transactions, or make readily apparent their political motivation. Finally, there may be a “never again” factor. When the electorate has been critical of the costs of resolving the banking crises, politicians may be attracted to privatization, particularly to a strong foreign investor, as a means of ensuring that they will not have to approve future expenditures to support state-owned banks.

Further research, particularly case studies, of privatization following crises could be enlightening. It would be useful to systematically examine the benefits to government of rapid divestiture of banks nationalized during a crisis as opposed to a longer-term approach to returning banks to private sector ownership. The example of Sweden (1992–94) indicates that rapid divestiture is possible and desirable. The general problems with governance of state-owned institutions, potential strengthening of management and risk management through sale to a well-regarded private bank, and a desire to return the banking sector to a normal footing as quickly as possible are among the reasons supporting the quick sale strategy. However, country authorities are often attracted by the potential financial upside of longer-term government ownership, arguing that the government stake will appreciate in value until finally sold after several years of profitable operation.

IV. Bank Privatization: Toward A Policy Consensus?

The world-wide trend of bank privatization less reflects a single consensus than waves of policy decisions with similar outcomes, often reached for very different reasons. While bank privatizations often occur during or shortly after banking crises, bank privatizations are frequently part of a more general trend in a country and, thus, generally share many of the same objectives as privatization. In most countries, some or all of the following objectives have motivated privatization,20 which in many cases have specific considerations for state-owned banks.

  • Raise revenues for the state. The importance of privatization revenues extends well beyond development and transition economies. Privatization revenues have been important for some countries seeking to meet the Maastricht criteria. British privatization proceeds in the 1980s substantially reduced government debt.

  • Promote economic efficiency and reduce government interference in the economy. Government ownership often has not been effective in meeting development goals. Policymakers may expect privatized enterprises to be more responsive in meeting consumer demand. Efficiency gains can eliminate the need for subsidies, freeing up fiscal resources for other priority spending or debt reduction. The potential fiscal burden of subsidizing the credit and operating losses inefficient state-owned banks can provide political motivation for privatization, as even substantial costs to clean up a bank’s balance sheet to make it attractive to investors may be less burdensome than continuing subsidies. A more efficient banking system will benefit the economy overall by reducing the costs of intermediation.

  • Promote wider share ownership. Initial public offerings (IPOs) are a frequent means of privatization, with provisions such as shares being sold in small allotments or restriction on foreign participation commonly being used to promote ownership by individual domestic investors. These provisions are frequently viewed as a tool to promote the development of capital markets. However, policymakers often have a preference for a strategic partner to acquire a controlling or significant interest as the bank is divested, as opposed to widely dispersed ownership, particularly where there are concerns about the quality of management and systems of a state-owned bank.

  • Provide the opportunity to introduce competition. In some of the former socialist countries, such as Russia and Poland, large state-owned banks were transformed into a number of smaller banks before or during privatization. Using privatization to encourage foreign bank entry can lead to an overall enhancement of management skills in the banking sector.

  • Subject state-owned enterprises to market discipline. Privatizing banks can be particularly helpful in achieving this objective, as private banks are less susceptible to moral suasion or explicit directives to provide preferential financing to state-owned enterprises. Thus, bank privatization may help to improve the efficiency of other state-owned enterprises as they would have to be able to obtain credit on commercial terms rather than relying on the support of state-owned banks.

The evidence of many bank privatizations in a diverse array of countries makes it clear that policymakers believe that it does matter whether government owns banks. While there may not be conclusive empirical evidence of causation, it is clear that state-owned banks are associated with “bad” growth and development outcomes. These can be attributed to inefficiency on the part of state-owned banks, or less benignly to political interference. Even if lacking empirical proof, many policymakers have concluded that private sector banks are more efficient, and privatization removes the irresistible cookie jar of state-owned bank largess from the reach of politicians. Thus, the trend to privatization of state-owned banks is likely to continue.

V. Bank Privatization: Issues for Policymakers

The lessons to be drawn from the experience of bank privatizations are in the form of broad principles rather than a “how to” checklist, since each case has unique features. There are many similarities between the privatization of banks and privatization of nonfinancial enterprises, however, there are some key differences. The failure of a privatized bank is potentially more damaging than the failure of a nonfinancial enterprise because of the potential loss of depositors’ funds, disruption to the payments system, and possible domino effects on other banks. For these reasons, it is important that there be an appropriate institutional structure in place, including a sound framework of general commercial law and effective banking supervision. The bank supervisory authority should play a key role in the privatization process, as it would in reviewing and approving the proposed change in ownership of any bank.

State-owned banks may enjoy real or perceived special privileges. For example, depositors may consider their deposits implicitly guaranteed by the state. It may be necessary to introduce a form of limited deposit insurance prior to privatization of state-owned banks as a means of clearly signaling the end of an implicit guarantee. Failure to deal with the special positions of state-owned banks prior to privatization runs the risk that markets continue to perceive state support, which both provides the privatized bank with a competitive advantage, and increases the potential political pressure for a bail-out should the privatized bank subsequently experience difficulties.21

Bank privatization in transition economies presents a special case that is significantly different from privatization of nonfinancial enterprises. Cement companies can still produce and sell cement, but the services of socialist banks as bookkeepers for the planned allocation of resources are in little demand in a market economy.22 Thus, state-owned banks require even more fundamental reform than nonfinancial enterprises because their basic economic function has to be completely overhauled in preparation for privatization. Key considerations and general guidelines drawn from the case studies of bank privatization, with support from the literature on financial sector development and privatization more generally, are presented below. The issues are sequenced from broad policy measures to specific concerns for individual bank privatizations (see Box 3 for a summary).

A. Institutional Infrastructure

Change in ownership alone will not address many of the factors contributing to poor performance by state-owned banks. Institutional factors, the most important of which are captured in the preconditions of the Basel Core Principles for Effective Banking Supervision, have to be conducive to sound banking. These factors include sustainable macro-economic policies, legal infrastructure, particularly with respect to contract law and measures for pledging collateral and enforcing security agreements, and appropriate and widely-used accounting standards. In countries where these preconditions require strengthening, successful bank privatization must be part of a broader program of reforms.

In an ideal world, it would be possible to complete each part of a major reform project without the complications of how to deal with other issues either concurrently or sequentially. Everything cannot happen at once, and even if there is a clear view on whether privatization should precede or follow key reforms, the ideal sequencing may not be possible. In practice, many elements of the reform are undertaken concurrently, or subsequent to privatization. Policymakers in many countries have proceeded with bank privatization before the necessary legal infrastructure and framework for effective banking supervision has been put in place. In these circumstances, private ownership, motivated by potential loss of investment may be better able to minimize exposure to the banking risks arising from inadequate infrastructure, although in such circumstances government ownership might rather be seen as a greater advantage. In some cases private investors have preferred a continued government minority stake as a possible means of influencing favorable outcomes in an unpredictable legal system, or increasing the likelihood that important state-owned enterprises honor the commercial terms of their contracts with the privatized bank.

B. Public Policy Objectives

The drive to privatize banks frequently comes from the belief that private ownership will contribute to financial stability and longer-term growth. However, some or all of the objectives of privatization generally—raising revenues, promoting efficiency, encouraging wider share ownership, enhancing competition, and introducing market discipline—will also apply in bank privatizations. These public policy objectives are likely to influence the preparations for privatization and the design of the transaction itself, which can come into conflict with financial stability concerns.

Key Considerations in Bank Privatization

Institutional infrastructure: Preconditions are vital to sound banking. These include macro-economic stability, legal infrastructure, accounting standards and an appropriate safety net (lender-of-last-resort facilities, and, possibly, deposit insurance). Sound banking supervision is required to review proposed privatizations from a prudential perspective, and to subsequently oversee the privatized banks. Perfect infrastructure and banking supervision will never be in place, so it will generally be preferable to privatize in conjunction with other reforms rather than to wait for ideal circumstances.

Public policy objectives: A safe and sound financial system is not the only objective to be met in privatization. There will be inevitable trade-offs, and other objectives such as supporting national champions or maintaining employment are likely to have broad political support. Prudential issues should not be sacrificed to other policy objectives due to the potentially far-reaching impact of subsequent bank failures.

Preparing a bank for privatization: An “as is” sale is preferable, if possible, as it can be completed quickly and does not entail major public investment in preparing a bank for privatization. However, state-owned banks are frequently in such poor condition that financial restructuring is required if reputable private investors are to be attracted.

Methods of privatization: Almost all successful bank privatizations have been some form of share sale. Attracting a reputable financial institution as a strategic investor, often with a significant public share float, has generally proven more successful than privatizations resulting in widely-held ownership. Government retention of a majority shareholding for an extended period has often been unsuccessful, thwarting true reform and leading to a need for additional recapitalization. There is empirical evidence that foreign bank entry improves the function of national banking markets, so attracting a foreign bank as strategic investor may be particularly desirable.

Prudential review: As with any change in bank ownership, the supervisory authority should only approve the transaction if the new owners are fit and proper, management is competent and experienced, the source of capital is verified, and the business plan is viable.

A desire for “national champions” and maintaining domestic control of the largest financial institutions are two related public policy objectives that frequently influence privatizations. While there may be a desire to acquire expertise and/or foreign capital to enhance stability and growth, policymakers are often reluctant to lose domestic control of large institutions. One of the motivations for both objectives is an element of pride or nationalism associated with having strong domestically-owned institutions. In addition to this emotional concern, which may have a very important political impact, economic arguments are also presented in favor of maintaining domestic control. While there is certainly no consensus on these issues, arguments presented include:

  • A national economy may be diminished in the long run if it becomes merely a “branch plant” without the benefits of the headquarters functions of international firms.23

  • Domestically-owned banks may establish stronger relationships with domestic industry, thus, providing more favorable and consistent trade financing to the nation’s exporters and importers than will foreign banks.24

  • Domestically-owned banks are arguably less likely to favor foreign business over domestic customers if faced with capital constraints, and are more susceptible to the exercise of moral suasion by government.25 Similarly, domestic banks cannot withdraw from a market in the same way that a foreign-owned subsidiary might curtail certain activities or even withdraw completely from a country as a result of a change in the strategic focus of the parent bank.

Other policy concerns are likely to include the maintenance of services in all areas served by state-owned banks prior to privatization, continued servicing of specific sectors, and preserving employment. These concerns can conflict with the desire to increase efficiency, as new private owners typically look to close unprofitable locations, eliminate policy-influenced lending to small business or state-owned enterprises, and improve operating efficiency through staff retrenchments.

Since privatization is a political process, regardless of the state of the economic debate regarding national champions, maintaining service to all regions and sectors and preserving employment, these issues are likely to be raised in the policy debate over bank privatization, and thus will influence the process.

C. Preparing for Privatization

A crucial question is whether to restructure a state-owned bank prior to privatization, or to try to sell the government stake essentially on an “as is” basis. In the rare case of state-owned bank operating efficiently on a commercial basis, there is little need for operational or financial restructuring as part of the process of government divestment. However, in the more typical case, state-owned banks require significant restructuring to become fully competitive with privately-owned banks.

The case for financial restructuring is often clear-cut as deeply insolvent state-owned banks are not very attractive to private sector owners. Since investors are unwilling to pay enough to “fill the hole” created by bad assets, government as owner has to find a way to provide the bank with a sufficient quantity of good quality assets to equal its liabilities in order to attract new equity investors. Methods of restructuring can vary. One frequently used model is the “good bank-bad bank” split,26 with nonperforming loans left in the bad bank, and government providing the good bank with assets, usually bonds, to fill the balance sheet hole.27 A variation on this approach, which has been used in Ghana, Tanzania, and Uganda, among other countries, is to transfer the bad assets to a specialized asset management company (AMC) rather than leave them in the bad bank. When the volume of bad assets is smaller, or if the decision is made that the bank should work out the problem loans itself, government as shareholder may subscribe to new equity issues. A further variation, which is only available if the bank to be privatized is already on a reasonably sound financial footing, is to issue subordinated debt to bolster the capital base prior to privatization.

While the need for financial restructuring is often clear-cut, the timing of such restructuring is not. When the state-owned bank is insolvent, delayed recapitalization can serve to increase losses and the ultimate cost. An insolvent bank can lack sufficient income from its earning assets to cover its costs, and without the new earning assets acquired through recapitalization, it may not be possible to return to profitability regardless of the amount of operational restructuring undertaken. However, when a bank has been recapitalized, failed operational restructuring and long privatization delays can lead to the need for further recapitalization expenses when the bank is finally ready for divestiture. For this reason, it is often recommended that recapitalization be closely linked to the privatization transaction.28 One attempt to balance the need for earning assets provided through recapitalization with the need to ensure effective restructuring is to provide recapitalization in stages, contingent on meeting restructuring objectives.29

Even when it is clear that operational restructuring is required, it may not be clear whether it is better for this to happen under government ownership, or if it is ultimately more cost-effective to sell the bank on an as-is basis. The “as is” sale price may be higher than the sale price for a restructured bank net of ongoing operating losses and one-off charges for staff retrenchments, branch closings, and other restructuring costs. This is because restructuring costs may not be fully recovered in subsequent divestiture, as management or consultants retained to assist are unlikely to achieve the exact branch alignment and staffing that a new owner would prefer. While new owners may pay more not to have to deal with an operational restructuring plan already underway, there may also be situations where new owners are reluctant to take on the burden of staff reductions and branch closures. Particularly where strong political pressure is anticipated, new owners may require certain closures or lay-offs to occur prior to privatization.

D. Methods of Privatization

The literature provides several taxonomies of privatization methods,30 but almost all bank privatizations can be categorized as share sales, asset sales or voucher privatizations. The vast majority of bank privatizations take some form of share sale, with a phased privatization often involving first an IPO or private placement, followed by subsequent secondary offerings (Appendix I). “Privatization is a process, not an event,”31 so while it is common to categorize by type of transaction, there are many decisions that lead to the final choice about how to divest government’s ownership stake. These decisions are influenced by policy objectives and political and fiscal constraints.

The use of voucher privatizations, where individuals received vouchers that could be exchanged for shares in various state-owned enterprises, has been almost exclusively limited to the transition economies of the former Soviet Union. The attractiveness was the speed of government divestment, and intended egalitarianism of distributing ownership of state assets to individual citizens. The process does not raise funds for the state, and thus is not suitable for meeting the government financing objective that is often one of the driving forces for privatization. Voucher privatizations brought no new equity into the bank, and at least initially resulted in a widely-held ownership structure, precluding a strategic investor taking a keen interest in the operational restructuring and governance of the bank. In some cases this changed over time as investors acquired significant holdings of shares originally distributed through voucher privatizations. Voucher privatizations have generally been unproductive, and when employed for banks have not led to healthy banks. However, as the transactions took place while the countries were in the throes of massive reform with governments using the voucher method having few options, it is difficult to see how more successful privatizations could have been completed at the time.

There are few cases of bank privatization by asset sales. One example is the disposition of banks nationalized in Finland in response to the Nordic banking crisis. In 1993 the government sold to four commercial banking groups equal tranches of the assets of the Savings Bank of Finland, which had been formed from an amalgamation of savings banks during the crisis. The creation of many new banks from the branches of Zhilsotsbank in Russia could be considered a form of asset sale, as branch managers were allowed to choose the assets that would constitute the new banks, essentially acquiring state assets at a zero price. Similarly, the good-bank bad-bank split could be considered a form of asset sale, as the good assets of the bank are repackaged for sale. A variation on this method is the disposition of the assets of closed banks by a centralized AMC, such as the Indonesian Bank Restructuring Agency. Aside from these examples, it is difficult to find cases of privatization by sale of state-owned banking assets as opposed to the sale of shares in a state-owned bank.

By far the most common type of bank privatization involves the sale of shares, which can be either a public offering, or a tender or auction process. Virtually all cases included in Appendix I are some form of share sale. The choice of share sale method is typically influenced by a range of sometimes conflicting objectives.

Maximizing government revenues may be achieved by a phased privatization, however, this has to be balanced against the likely difficulty in instilling market-oriented governance and management in banks when government retains a large ownership stake.32 Continued state-ownership carries with it the risk of recurring credit losses or operating losses, leading to a need for additional recapitalization before final divestment.

A widely-subscribed IPO can be politically attractive as a means of preserving domestic ownership, avoiding the pitfalls of lending to parties connected to significant owners of the bank, and may also serve to foster capital market development by providing a large listing for the local stock exchange. Widely-held ownership has the drawback of not providing strong oversight of management by a significant shareholder, and also does not provide the natural conduit to strengthen management and the bank’s internal systems that would arise from sale of a controlling interest to a strong bank.

Privatization by IPO can be disappointing in countries with small and emerging capital markets.33 Underdeveloped institutional structures, such as inexperienced investment banks, limited broker networks and trading mechanisms, have led to market manipulation at worst, or inefficient share distribution at best. Countries seeking to use bank privatizations as a catalyst for capital market development may be disappointed with the pricing of the IPO, and still have markets with limited depth and liquidity due to inadequate institutional structure and low investor interest.

Evidence from case studies suggests that better financial performance is achieved when privatization involves a strong financial institution as a significant shareholder.34 Ensuring that there is a suitable significant investor can be achieved through a sale by tender, or in an IPO, by reserving a controlling percentage for a prequalified investor. However, such a transaction can be politically difficult in developing and transition economies, as the only suitable strategic investors are likely to be foreign. There is empirical evidence to support the hypothesis that foreign bank entry can make domestic markets more efficient by forcing local banks to operate more efficiently, providing long-run benefits for banking customers in the form of lower intermediation and service charges.35 This suggests that a reputable foreign bank is particularly desirable as a strategic investor when privatizing in markets dominated by domestic banks, notwithstanding possible political opposition to sale to foreign interests.

E. Prudential Review

There are many cases where the subsequent financial difficulties of a privatized bank could have been avoided if an appropriate prudential review had been undertaken prior to privatization. Owners and managers lacking banking experience or fitness and probity, investors lacking the promised capital, and unviable business plans are common causes of failed privatizations that should be identified in a prudential preview.36 The privatization should only proceed if the supervisory authority is satisfied in all respects. Pressure to approve a transaction despite prudential concerns, lack of capacity on the part of the supervisory authority to undertake a suitable review, or proceeding with privatization without any involvement of the supervisory authority has resulted in the need for subsequent intervention in failed privatizations in Croatia, the Czech Republic, Mozambique, and Uganda, among others.

VI. Theory Meets the Real World: Obstacles to Privatization and Half-Way Measures to Enhance Governance

Even when privatization of banks is viewed as a good policy option, implementation may be problematic. Many bank privatizations have been long delayed or aborted. Key issues to be managed include the cost, sequencing of other reforms, and achieving political consensus. Even if privatization is not possible, policymakers have some options to help avoid the vicious cycle of repeated recapitalizations or forbearance to deal with recurring losses of inefficient state-owned banks.

The cost of making weak banks attractive to private investors may far exceed the revenues from privatization. This is not an uncommon situation, and even when long-term cost savings are substantial, the immediate fiscal burden of making weak banks attractive to private investors can be greater than the immediate costs of continued state-ownership. This is particularly true if the recognition of the costs of state-owned banks is deferred through supervisory forbearance. Banks may appear sound and profitable if loan loss provisioning requirements and capital adequacy requirements are not enforced. This creates a strong incentive for the “wait and hope” strategy. Unfortunately, experience around the world is that the condition of weak banks is more likely to deteriorate than improve unless decisive action is taken.

India presents a case in point of the real difficulties in proceeding with privatization, and some of the half-way measures that can be undertaken (Box 4). Among other obstacles, the cost of restructuring weak banks to make them attractive to private investors was seen as prohibitive. The costs are not limited to dealing with nonperforming loans, but extend to needed rationalization of branch networks and head office staffing. Quite apart from the monetary costs of severance and branch closures to achieve efficiencies, there are significant social costs, and a political cost to downsize the unionized workforce. As an alternative to privatization, India has pursued bank reform with the following key components:

  • reduction in barriers to entry to foster greater foreign competition

  • ensuring private sector-quality boards of directors and senior management

  • voluntary retrenchment schemes to facilitate needed staff rationalization

  • gradual strengthening of prudential norms

The combination of exposure to increasing competition, relaxation of some of the more restrictive elements of the regulatory regime, strengthened governance and prudential oversight was intended to improve the performance of the state-owned banks, while retaining majority government ownership and at least some elements of the social commitment to finance priority sectors.

China provides an illustration of issues of sequencing and the extended time that can be required for other reforms. Although China has indicated an intention to privatize all but the largest state-owned enterprises, thus far the number of divestitures has been small.37 This is in part because of the need for broad structural reforms. The large state-owned banks historically served to allocate credit in a planned economy, so not only has there been a need to introduce basic commercial banking concepts such as credit risk assessment, there has been the much broader need to reform the state-owned enterprises unable to service debt on commercial terms, and to introduce a prudential framework and effective bank supervision.38

India: Experience With State Bank Reform.

India’s banking sector has evolved considerably since the beginning of a reform program in 1991. Public sector banks (PSBs), which accounted for about 90 percent of the banking market in 1991, now have 75 percent of total banking assets. None of the 27 PSBs has been privatized, although 15 have tapped the capital markets and have minority shareholdings ranging from 25 percent to 45 percent. The policy of gradually tightening prudential regulations and at the same time increasing competition in the market by removing restrictive regulations and permitting new entrants has contributed to improved efficiency in the PSBs. Asset quality and profitability have converged toward the average for commercial banks in India.

Weaknesses within the PSBs were broadly known within policy circles, but until the introduction of more stringent accounting and prudential standards in 1992–93, the extent of the problems was not evident in the banks’ financial reporting. Interest accrued but not paid could be recognized as income, and banks were widely under provisioned in the absence of specific prudential requirements.

New banks very quickly took advantage of liberalized entry rules, with 24 new private banks, including 15 with foreign ownership, beginning operations in India between January 1993 and March 1998. The new prudential standards quickly brought to light longstanding problems in the PSBs. In 1992–93 the PSBs, all but one of which had been profitable the previous year, collectively recorded a net loss, and half reported negative net worth. This prompted government to make capital injections into 19 of the PSBs in 1993–94, with many receiving further support in subsequent years. The capital support was contingent on recovery plans, but a number of banks made little substantive progress, in part because of the expectation, subsequently confirmed, that government would continue to provide capital injections.

A 1999 review of the PSBs identified as chronically weak rejected merger and closure options. Privatization was viewed as attractive to eliminate the need for future government recapitalizations, but impractical due to cost of needed restructuring and the likely inability to attract private investors. Instead, renewed efforts at restructuring, including harder looks at staff reductions and branch closures was recommended. These renewed efforts ultimately bore fruit, with all PSBs meeting the 9 percent capital adequacy requirement in 2003.

The Indian approach to date has been to reform state-owned banks without privatizing and retaining some noncommercial mandates such as lending to priority sectors. The Reserve Bank of India as banking supervisor has been extremely active in driving the restructuring, which has been undertaken concurrently with efforts to strengthen governance and management practices throughout the Indian banking sector. The list of changes to the legal framework for banking supervision and improvements to its practical implementation is impressively long. The greatly strengthened prudential regime is intended to ensure that other government policy objectives do not overwhelm the need for PSBs financial viability, but it remains to be seen if this is achievable over the medium to long term.

This reform process has been underway since the early 1990s, combined with a measured opening of the banking market to foreign competition. Recapitalization of state-owned banks in 1998 and the creation of asset management companies have not truly addressed the banks’ fundamental problems of governance and management, so there is a continuing flow of new problem assets, notwithstanding the very rapid growth in the loan portfolio and weak provisioning rules, which have helped to minimize reported nonperforming loan levels. A further complication in reforming the banks is the need for a new social welfare mechanisms to replace the housing, medical and other services historically provided to employees and retirees by state-owned institutions. These functions need to be removed from the state-owned banks if they are ever to be privatized.

In cases where privatization in the short term cannot be achieved, there are measures that can enhance the performance of state-owned banks. Vulnerability to explicit or implicit political interference, and the potential difficulty in reconciling various government policy objectives with prudent commercial banking practices, can leave a state-owned bank with an unclear mandate, or unable to fulfill conflicting elements of its mandate. If the commercial banking operations are not to be privatized, then three important half-way measures are (i) a mandate to operate on a commercial basis; (ii) a governance structure to insulate, so far as possible, state-owned banks from overt political influence; and (iii) implementation of the same supervisory regime that is applicable to private banks. These measures can make government ownership a sustainable state as well as paving the way for ultimate privatization.39

State-owned banks should be required to operate on a commercial basis. This requires competent staff and efficient internal systems, operating free from political influence. The governance measures cited above can ensure that competent senior management are retained with the mandate and freedom to implement the same kinds of systems and controls that would be adopted by any prudent commercial bank. A key component of this commercial operation is credit risk assessment, both for state-owned enterprises and other borrowers. To the extent that state banks are required to undertake lending or provision of other services on nonmarket terms in order to meet government policy objectives, this should be explicitly acknowledged and undertaken transparently, preferably with a government subsidy or guarantee.

Once a state-owned bank has been given a clear commercial mandate, the governance structure is important in ensuring that the mandate can be fulfilled. As with other state entities, a balance of independence and accountability is important. A board comprised of independent directors serving for fixed terms can serve as important buffer between government and the state-owned bank. Directors need to be clearly charged with stewardship of the public funds invested in the bank, so that their fiduciary responsibility should take precedence over any partisan affiliation. Directors of state-owned banks, taking seriously the responsibility for oversight of public funds, with fixed terms to preclude summary dismissal by the government of the day, may provide similar stewardship to that provided by directors of privately-owned banks.

One potential market distortion is that state-owned banks, even if operating on a commercial basis relatively well-insulated from political pressure, may have cost of fund advantages over private banks, arising from an implicit (or explicit) government guarantee of the deposits of state-owned banks. This may be more pronounced if the state-owned banks are not required to meet regulatory capital or other prudential requirements. Application of the same supervisory regime to state-owned banks and private banks can help to minimize the distortions introduced in the market by state-owned banks.

While it can be challenging in practice, treating government in the same way as other bank owners are treated—requiring all prudential norms to be observed, and that capital be restored in the event of losses—the supervisory regime can provide additional incentives for state-owned banks to operate on a commercial basis. This approach is a general principle for banking supervision, but in practice, there are inevitable complications in dealing with state-owned banks. Nevertheless, it is important in ensuring that the competitive playing field remains level and to maintain credibility in the financial sector that state-owned banks are not dealt with in a more favorable manner than privately-owned banks.40

VII. Conclusion

The question of how state-owned banks and their privatization affect financial sector stability and growth will continue to be an important issue for policymakers. Despite numerous privatizations in recent years, many countries continue to have financial sectors featuring significant roles for state-owned banks.

State-owned banks are often associated with significant shortcomings in the preconditions for an effective banking system, such as the rule of law and strong government infrastructure, so any particular problems introduced by state banks may be obscured by these important institutional weaknesses. This is consistent with the finding that large privatizations immediately precede crises in only a few instances. In these countries, failure to establish the institutional preconditions for sound banking prior to, or at least concurrently with, the privatizations is more likely to have been a proximate cause of the crisis than the privatizations themselves.

Nationalization of banks is a policy response often used in dealing with a banking crisis, raising the possibility of a temporary increase in state ownership and subsequent divestiture. Only about one-third of privatizations in postcrisis countries are explained by this phenomenon, suggesting that in the wake of a crisis, policymakers opt to divest government ownership in banks as part of the reforms intended to strengthen the financial sector. This is consistent with the growing preference for private ownership of banks, likely due to an expectation of greater financial stability and higher growth.

The prudential dimension distinguishes bank privatizations from the privatization of nonfinancial enterprises. The experience of failed privatizations illustrates that policymakers ignore at their peril the key prudential concerns of having fit and proper owners, adequate capital, competent management, and viable business plans. Too often, either through lack of capacity on the part of the supervisory authority, or a failure to conduct an appropriate prudential review, privatized banks subsequently face distress due to issues that could have been foreseen at the time of the privatization.

Governance structures can significantly mitigate the pitfalls of state ownership, although state-owned banks will inevitably be more susceptible to political suasion than their private sector counterparts. Greater focus on state bank governance is important given the significant presence of state ownership that will persist for many years in many countries. In some countries, a strong philosophical commitment to state-owned banks remains an integral part of public policy, and in other countries, even though there is some support for privatization, it will take many years to achieve the objective. In either case, enhanced governance can help to avoid the expensive cycle of losses and recapitalizations.

Considerably more research could assist policymakers in dealing with state-owned banks. A systematic series of case studies organized around themes such as the institutional infrastructure, public policy objectives, preparing banks for privatization, methods of privatization, and the prudential review, could lead to improved “how to” recommendations for policymakers. Further work on banking crises, including more precise delineation of crisis and identification of observable indicators would permit more valuable work on capturing the interaction between crises and possible causal factors, including state ownership.

Despite the absence of empirical proof, it is clear that policymakers believe it does matter whether the state owns some or all of a country’s commercial banks. The widespread trend of privatization is likely to continue, and further research can improve the practical policy advice provided to government officials undertaking bank privatizations.

APPENDIX I

Bank Privatizations

(Mid-1970s – 2003)

article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
Sources: Information on privatizations has been complied from a review of the bank privatization literature, publicly available IMF Staff Country Reports and Financial Sector Stability Assessments, press reports, and various occasional papers. Details on privatizations are often not readily available, and various sources often provide conflicting details. The author would be especially grateful for information to complete or correct the cases noted, and for details of additional bank privatizations.

APPENDIX II

Privatization and Crisis Dates

Appendix II below presents bank privatization data from Appendix I juxtaposed against the dates of banking crises. There are a total of 39 countries, among the 65 total countries included in Appendix I, that have experienced banking crises with specific dates as identified in the banking crisis literature.

There is no universal definition of banking crises, and determining the start and end date of crises requires judgment (for a discussion of the issues, see Bell and Pain (2000)). To determine the dates of banking crises, Bell and Pain’s chronology of banking crisis drawn from seven studies (excluding the Hardy and Pazarbasioglu cases of “distress”) was expanded to include the Caprio and Klingebiel (2003) data set (excluding borderline and nonsystemic crises), and the determination of government intervention from De Nicolò and others (2003).41 Crises lacking specific dates were excluded, and for those where different sources provide different dates, a consensus date was adopted which generally encompassed the longest indicated period of crisis.

“Crisis 1” has the broadest inclusion, being defined as any crisis identified by specific dates in at least one of the studies. Because of the subjective nature of identifying crisis, “Crisis 2” adopts the more stringent requirement that the crisis must be identified by at least two of the studies. While this does reduce by 10 the number of identified crises, it does not substantially change the finding that privatization preceding the onset of a crisis by five years or less is rare, that privatizations are common concurrently or within three years of the end of a crisis, and that divestiture of banks nationalized as part of crisis management accounts for only a small portion of post-crisis privatization.

Bank Privatization and Banking Crises

(Mid-1970s – 2003)

article image
article image
article image
article image
article image
article image

References

  • Abarbanell, Jeffery S., and John P. Bonin, 1997, “Bank Privatization in Poland: The Case of Bank Śląski,” Journal of Comparative Economics, Vol. 25, pp. 3161.

    • Search Google Scholar
    • Export Citation
  • Abarbanell, Jeffery S., and Anna Meyendorff, 1997, “Bank Privatization in Post-Communist Russia: The Case of Zhilsotsbank,” Journal of Comparative Economics, Vol. 25, pp. 6296.

    • Search Google Scholar
    • Export Citation
  • Aiginger, Karl, and Peter Mooslechner, 1997, “Framework Conditions, Methods and Processes of Privatization: Lessons from the Trade Sale of an Austrian Bank,” Paper presented at the Eleventh Plenary Session of the OECD Advisory Group on Privatization: Banks and Privatization, Rome, September 18–19.

    • Search Google Scholar
    • Export Citation
  • Aliber, Robert J., 1984, “International Banking: A Survey,” Journal of Money, Credit, and Banking, Vol. 16, pp. 66178.

  • Andrews, Michael, 2002, “Addressing the Prudential and Antitrust Aspects of Financial Sector Mergers and Acquisitions,” in Building Strong Banks, ed. by Charles Enoch, David Marston, and Michael Taylor (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Andrews, Michael, 2003, “Issuing Government Bonds to Finance Bank Recapitalization and Restructuring: Design Factors That Affect Banks’ Financial Performance,” IMF Policy Discussion Paper 03/4 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Asian Development Bank, 2001, Special Evaluation Study on the Privatization of Public Sector Enterprises: Lessons for Developing Member Countries (Manila: Asian Development Bank).

    • Search Google Scholar
    • Export Citation
  • Baer, Werner, and Nader Nazmi, 2000, “Privatization and Restructuring of Banks in Brazil,” Quarterly Review of Economics and Finance, Vol. 40, pp. 324.

    • Search Google Scholar
    • Export Citation
  • Barnes, Guillermo, 1994, “Lessons From Bank Privatization in Mexico,” in Building Sound Finance in Emerging Market Economies, ed. by Gerard Caprio, David Folkerts-Landau and Timothy D. Lane (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Barnett, Steven, 2004, “Banking Sector Developments,” in China’s Growth and Integration into the World Economy, ed. by Eswar Prasad, IMF Occasional Paper 232 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Barth, James R., Gerard Caprio Jr., and Ross Levine, 2000, “Banking Systems Around the Globe: Do Regulation and Ownership Affect Performance and Stability?World Bank Working Paper No. 2325 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Basel Committee on Banking Supervision, 2002, Supervisory Guidance on Dealing With Weak Banks: Report of the Task Force on Dealing with Weak Banks (Basel: Bank for International Settlements).

    • Search Google Scholar
    • Export Citation
  • Bell, James, and Darren Pain, 2000, “Leading Indicator Models of Banking Crises—A Critical Review,” Financial Stability Review, Bank of England, December 2000, pp. 113129.

    • Search Google Scholar
    • Export Citation
  • Beyer, Han-Joachim, Claudia Dziobek, and John R. Garrett, 1999, “Economic and Legal Considerations of Optimal Privatization: Case Studies of Mortgage Firms (DePfa Group and Fannie Mae),” IMF Working Paper 99/69 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bonin, John, Iftekhar Hasan, and Paul Wachtel, 2003, “Privatization Matters: Bank Performance in Transition EconomiesPaper presented at the World Bank Conference on Bank Privatization, Washington, November 20–21.

    • Search Google Scholar
    • Export Citation
  • Bonin, John, and Paul Wachtel, 1999, “Lessons from Bank Privatization in Central Europe,” paper presented at the World Bank/Federal Reserve Bank of Dallas Conference on Banking Privatization, Washington, March 15–16.

    • Search Google Scholar
    • Export Citation
  • Borish, Michael S., Wei Ding, and Michel Noël, 1997, “Restructuring the State-Owned Banking Sector: A Comparative Assessment of Approaches in Central Europe,” MOCT-MOST Economic Policy in Transitional Economies, Vol. 7, pp. 4977.

    • Search Google Scholar
    • Export Citation
  • Boubakri, Narjess, and Jean-Claude Cosset, 1999, “Does Privatization Meet Expectations? Evidence from African Countries,” paper presented at the African Economic Research Consortium biannual research workshop, Nairobi, December 4–9.

    • Search Google Scholar
    • Export Citation
  • Brownbridge, Martin, and Charles Harvey, 1998, Banking in Africa: The Impact of Financial Sector Reform Since Independence (Oxford: James Currey Limited).

    • Search Google Scholar
    • Export Citation
  • Clark, George R.G., and Robert Cull, 1997, “The Political Economy of Privatization: An Empirical Analysis of Bank Privatization in Argentina,” World Bank Working Paper No. 1962 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Clark, George R.G., and Maria Soledad Martinez Peria, 2001, “Does Foreign Bank Penetration Reduce Access to Credit in Developing Countries?World Bank Working No. Paper 2716 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Caprio, Gerard, and Daniela Klingebiel, 2003, “Episodes of Systemic and Borderline Financial Crises,” World Bank mimeo available via the Internet: http://wbln0018.worldbank.org/html/FinancialSectorWeb.nsf/SearchGeneral3?openform&Banking+Systems.

    • Search Google Scholar
    • Export Citation
  • Caprio, Gerard, and Maria Soledad Martinez Peria, 2000, “Avoiding Disaster: Policies to Reduce the Risk of Banking Crises,” paper presented at the conference on Monetary and Exchange Rate Policies: Options for Egypt, Cairo, November 19–21.

    • Search Google Scholar
    • Export Citation
  • Classens, Stijn, Asli Demirgüç-Kunt, and Harry Huizinga, 2001, “How does foreign entry affect domestic banking markets?Journal of Banking and Finance, Vol. 25, pp. 891911.

    • Search Google Scholar
    • Export Citation
  • Cull, Robert J., 1997, “Financial Sector Adjustment Lending: A Mid-Course Analysis,” World Bank Policy Research Working Paper No. 1804 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Das, Udaibir S., Marc Quintyn, and Kena Chenard, 2004, “Does Regulatory Governance Matter for Financial System Stability? An Empirical Analysis,” IMF Working Paper 04/89 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • De Nicolò, Gianni, Philip Bartholomew, Jahanara Zaman, and Mary Zephirin, 2003, “Bank Consolidation, Internationalization and Conglomeration: Trends and Implications for Financial Risk,” IMF Working Paper 03/158 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Dinç, Serdar, 2002, “Politicians and Banks: Political Influences on Government-Owned Banks in Emerging Countries,” mimeo, University of Michigan Business School.

    • Search Google Scholar
    • Export Citation
  • Drees, Burkhard, and Ceyla Pazarbasioglu, 1995, “The Nordic Banking Crises: Pitfalls in Financial Liberalization,” IMF Working Paper 95/61 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Dziobek, Claudia, and Ceyla Pazarbasioglu, 1997a, “Lessons from Systemic Bank Restructuring: A Survey of 24 Countries,” IMF Working Paper 97/161 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Dziobek, Claudia, and Ceyla Pazarbasioglu, 1997b, “Lessons and Elements of Best Practices,” in Systemic Bank Restructuring and Macroeconomic Policy, ed. by William E. Alexander and others, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Enoch, Charles, Anne-Marie Gulde, and Daniel Hardy, 2002, “Banking Crises and Bank Resolution: Experiences in Some Transition Economies,” IMF Working Paper 02/56 (Washington