Source: Carnegie
Based on the First Meeting of the Carnegie Economic Reform Network, Miami, Florida, February 27-28, 1998
The Carnegie Economic Reform Network met for the first time on February 27-28, 1998, in Miami. During that meeting, CERN conducted a discussion on "Latin America in the Wake of the Asia Crisis: Lessons from and for Asia." In that context, participants in the meeting put forward broad-ranging ideas about how domestic and international policies might be improved to cope with the volatility of international capital flows.
This Working Paper summarizes the many ideas presented at the meeting, adding references and explanatory material as appropriate. It is not a CERN "position paper", and the ideas and recommendations presented here do not necessarily command widespread agreement within the group. Instead, it is a "work in progress" that provides a starting point for the next round of discussion within CERN.
Introduction
The Mexican crisis of 1994-95 and the Asian crises of 1997-98 demonstrate that coping with the volatility of international capital flows has become the most difficult problem of economic management for developing and transitional economies. These economic crises in Mexico, Thailand, South Korea, and Indonesia were caused by a complex combination of factors involving domestic banking systems, exchange rate management, and political events. The relative importance of these factors varied a great deal among the countries. But in every case the large inflow of foreign capital in the preceding period, which magnified domestic credit booms, and the sudden rush for the exits were critical factors. (See Annex 2 for data on the magnitude of these capital flows.) Undoubtedly, the current size, speed, and volatility of capital flows among open economies vastly complicate the problems of economic policy-making for developing and transitional economies and allow little margin for policy error.
To a significant extent, the participation of many developing and transitional economies?the emerging markets?in international capital markets creates difficulties simply because of disparities in size. Most of these economies and their domestic financial sectors are very small relative to the international capital markets. Capital flows around the world are like the oceanic tides: in deep bays, tidal movements are little noticed, but in shallow bays, the ebb and flow of the global ocean create huge effects. This may be dubbed "the Brittany effect:" the shallow financial systems of an emerging market economy can be left completely dry when the tide goes out. But the problems caused by size disparities are greatly compounded by the weaknesses in domestic banking systems. It is this combination of large, volatile international capital markets and small, weak domestic banking systems that is so dangerous.
In this new "instant" world of massive capital market flows, the government in an emerging market may be reduced to a marginal actor. Even if well endowed with foreign exchange reserves, its financial resources may easily be dwarfed by international capital movements. Moreover, the dynamics of capital markets are incredibly complex?driven by political events and changes in perception and expectation that may or not be solidly based on economic reality. Even an astute policymaker with good cards in his hand (not only ample international reserves but also high credibility) may not be able to play them successfully. Recognizing these limitations, government policy-makers ought to be humble about their ability to design policies that prevent financial crisis altogether. However, it is a realistic objective is to reduce the probability of and minimize the damage from such crises in emerging and transitional economies.
Designing policies to minimize the probability and the impact of financial crisis in this era of international capital mobility is critically important because the economic volatility generated by such crises constitutes a potential threat to political support for open, market-oriented policies. As stated by Ricardo Hausmann, Chief Economist at the Inter-American Development Bank: "The many benefits of globalization will not compensate for the costs of volatile capitalism quickly enough to build, much less sustain, a strong social consensus around the policies that facilitate global integration. If volatility remains unchecked, a backlash will build against the market-oriented, democratic reforms that are essential to consolidating the remarkable political and economic progress of the last decade." In testimony to the US Congress, Paul Volcker expressed a similar concern: "If?the turbulence of markets persistently undercuts strong and consistent growth in emerging markets, then temptations to reject the ideology of open markets and multilateralism will increase."
Where to Search for Remedies?
Debate about what ought to be done to reduce the risk of crises like those in Mexico, Thailand, South Korea, and Indonesia often revolves around these questions: to what extent were these crises the result of poor domestic policies? or are they best understood as the consequence of an inherently and excessively volatile international capital market? Thinking in prescriptive terms, should remedies focus on better domestic policies or an improved "international architecture"? It is equally important to ask: were these crises the result of inadequate theory about how to manage emerging economies in this new era of international capital markets? Or were they a consequence of failure to apply known theory and the lessons of experience? If so, were these failures due to political constraints on policy-makers, i.e. "a politically-induced learning disability"?
The dichotomies posed by these questions are, of course, too rigid. It was the interaction of domestic policy choices and international capital markets that spawned these crises. Clearly there were important errors of domestic policy that contributed to the crises, and these errors were severely punished by rapid outflows of capital. Some of these errors resulted from weakness in the dominant theoretical framework for managing small, open economies in a world of huge, electronic capital markets, and others reflected a political inability to apply established theory and proven lessons. From the perspective of national policy-makers who seek to avoid similar crises in the future, the first-line of defense ought to be identifying and avoiding domestic policy errors of both kinds. There must also be ways in which the international system can be improved so that emerging markets can take advantage of international capital inflows while reducing their volatility. This paper considers both domestic and international responses and tries to identify both theoretical inadequacies and common political failures to apply known theory and lessons.
Possible Domestic Responses
If there is a simple explanation for the complex set of events labeled the Asian crisis, it is this: "weak banks plus good times." Growing like a cancer during the boom years of the Asian miracle, poorly regulated and weakly supervised banks were a major factor in excessive debt financing by firms, inefficient investments, risky reliance on short-term debt, and extreme asset price inflation. When the asset price bubble burst in Thailand, similar weaknesses became unsustainable elsewhere. Both domestic and international capital fled, compelling devaluations. In important respects, the roots of the Asian crises were in the banking systems.
The Mexican crisis of 1994-95 is probably best understood as a currency crisis aggravated by a banking crisis, although some argue that the real roots of the crisis were in the banking system. The sharp devaluation in December 1994 exposed the existing vulnerabilities of the banking system. The consequent bank insolvencies contributed substantially to the depth of the recession in 1995.
These experiences make clear that weak banking systems in an era of international capital mobility can either trigger severe economic crisis or seriously aggravate a crisis arising from some other trigger. Conversely, better management of banking systems is a necessary?but certainly not sufficient?means to reduce both the volatility of international capital flows and the impact of remaining volatility. Given its central importance, this section focuses on domestic policies to strengthen banking systems.
Correcting Errors in Prevailing Theory. In part, recent banking crises reflect several inadequacies of the prevailing theory or conventional wisdom for managing banking systems in developing economies. First, the conventional wisdom that favors financial liberalization has placed insufficient emphasis on the critical need for strong prudential regulation and close supervision. In Chile, for example, which pioneered financial liberalization in the late 1970s, there was an overwhelming bias against any attempt by the government to regulate markets, including the financial markets. Even in Mexico and South Korea in the late 1990s, the absence of a strong bank supervisory capability was not viewed with alarm by many policy-makers. When applied to financial markets, this lack of recognition of the importance of good regulation and supervision?prior to or at least in parallel with financial liberalization?is clearly a conceptual error that must be avoided. In particular, it is crucial to create a reasonably good legal and institutional capacity for regulation and supervision of banks prior to opening the capital account.
Second, the guiding theory of financial liberalization had been ambiguous about the importance of good policies with respect to deposit insurance and bank insolvency, and in many countries these policies remain poorly defined and/or weakly applied. This puts economies at severe risk in several ways. For bank owners and managers, it creates serious moral hazard: a presumption that the government will bail out an insolvent bank, which is based on a long tradition in Latin America, induces risky lending practices. For depositors, the presumption that the government will guarantee all deposits eliminates the incentive for large depositors to monitor the health of their banks. And, when confidence in a bank or the banking system is in doubt, uncertainty about the willingness or capacity of a government to protect depositors? money can induce bank runs that would otherwise be avoidable. Conventional wisdom must now unambiguously embrace the need for clear laws and regulations governing deposit insurance and bank insolvency, and these rules must limit the potential fiscal burden on the government by putting a cap on deposit insurance and by imposing losses on bank shareholders.
Third, economic policy-makers and bank regulators need to monitor short-term, foreign-denominated liabilities much more closely and to recognize that such debt involves high rollover risk. Conventionally, attention is directed toward the level and source of debt, but timely data is usually quite inadequate about its maturity structure. The massive buildup of the Mexican government?s Tesobonos during 1994 and of short-term external credits to the South Korean banking system in 1996 and 1997 became fatal vulnerabilities in each case.
Fourth, the guidelines for the privatization of publicly owned banks need to include strict requirements with respect to the qualifications and financial capacities of new owners. In Mexico, very large bank privatizations proceeded quickly in 1991-92 with little concern about the prior experience and capacities of new bank owners and managers, and as a result these banks were weaker than necessary when the crisis hit in late 1994.
Overcoming Political Constraints to Good Policy. Although these errors in knowing the right thing to do have played a role in past banking crises, probably the greatest problem has been not doing the right thing. Policy-makers in many cases have been politically unable or unwilling to apply the lessons of experience from other countries and to implement policies that they know to be correct. As a general proposition, this "politically-induced learning disability" is the consequence of the structure of incentives facing policy-makers. In previous Latin American crises, such as in Chile (1982) and Mexico (1994), numerous policy-makers and non-governmental analysts were aware that many banks had become overextended and extremely vulnerable, if not technically insolvent. But, when an economy is growing rapidly, fueled by a boom in credit, there are obvious incentives to avoid "spoiling the party". Everyone has a bias to believe that the good times will go on indefinitely, and the skeptics and pessimists are ignored. These perverse incentives can be characterized in different ways: "incentives for denial," "a complicity of silence," "listening to interior decorators", and "the deceptive effect of success." By whatever name, there is a strong political disincentive for taking action early enough to avoid?or at least limit the magnitude of?a financial collapse.
The political incentive for inaction may be strongest in the latter year or so of the normal term of an elected government. The administration in power can always hope that any financial crisis can be postponed until after its term in office. These incentives for inaction are especially strong when a government faces the need to devalue a fixed (or tightly managed) exchange rate because such action directly hurts "those who believed in you" and rewards those who did not.
Political paralysis in the face of a credit-driven, unsustainable boom may not be inevitable, but it does seem to be the norm. How might this incentive problem be solved? Are there any devices which might help "save a government from itself" by increasing the likelihood of timely, corrective action to avoid a banking crisis? Here are some ideas, which are not mutually exclusive:
1. Create and/or strengthen "technocratic" institutions that have a mandate for political independence and a bureaucratic incentive to take a long-term view, to do the right thing, and to endure the political "heat". This, of course, is the theory underlying independent central banks, which are increasingly in favor in Latin America. The principle applies equally to the superintendency of banks. This is not, however, something than can be achieved quickly (on this point, see Box 1).
2. Improve the transparency of financial system and foreign exchange data. To some extent such transparency will automatically trigger timely and hopefully gradual market corrections. Markets react much more quickly than governments, and making relevant statistics readily available to market participants is crucial so that problems emerge and are resolved before they grow dangerously large. Recent examples of such improvements in transparency include Mexico?s daily update of its foreign exchange reserve data on the Internet and the newspaper publication of the independent auditors? ratings of Chilean banks.
Lags in Institutional Development
In some cases, weakness in the supervision of banks does not result from lack of understanding about the right thing to do or from the political inability to do the right thing. Instead, it is a consequence of the unavoidable difficulty in developing the institutional capacities of the bank superintendency. Even if the best laws and regulations governing banks are in place, a superintendency needs a strong cadre of trained professionals to enforce them. And, even if there are both the political will and budget resources to hire such people, creation of a well-managed and well-staffed institution is not instantaneous. However, a determined effort can reduce the time required to create such institutional capacity.
This illustrates a general problem plaguing market-oriented reforms: liberalization of markets can be achieved much more quickly than development of the institutional capacities that are often needed to support efficient and more stable markets. During this lag, newly liberalized economies are vulnerable to market failures. It is clearly critical to work toward reducing these time lags by focusing on institutional development.
3. Put in place market mechanisms so that market participants will be better able to recognize and sanction poor bank performance. The recent innovations in Argentina were designed for this purpose. For example, under new guidelines, banks issue subordinated debt as part of their capital, and a market for these debt instruments has been created. Investors in these instruments have a strong incentive to monitor the health of the bank concerned, and perceived weaknesses are reflected in the market price of this debt.
4. Define clear guidelines and processes for resolution of failed banks and for deposit insurance. In the absence of solid laws and regulations in this area?and an institutional capacity to enforce them, the superintendent of banks and the government as a whole will become embroiled in political conflict about the distribution of losses among shareholders, depositors, and the government. This is a major disincentive to the government for taking timely and correct action. Conflict avoidance leads to delay, even though the problem will likely get worse over time. Without good laws and processes, a larger share of these losses will likely be borne by the public treasury since this spreads the pain over a longer period. Furthermore, a larger part of the assets of wealthy shareholders will likely be protected at the expense of future taxpayers. But having well-defined laws and processes to distribute the costs of a crisis makes it less politically difficult for a government to face reality.
5. Ensure that the government itself is well organized to recognize and respond to crisis. There must be a locus of responsibility to initiate action and a reasonably compact decision-making structure. For example, in recent years the South Korean government had six chief economic advisers to the President and seven deputy prime ministers involved in economic policy. As a result, Korean economic policy was less consistent and predictable. Inter-ministerial policy coordination suffered, and this may have impeded the government?s capacity to recognize and respond to the growing crisis during 1997.
6. Make use of an external "whistle blower." Even if all of the above measures could be taken, the political incentives for inaction are so strong that the intervention by a party from outside the domestic political sphere may be necessary to induce timely actions. A far-sighted government may therefore wish to seek prior, informal agreement?with the IMF or the BIS, in particular?that the external party will issue a public warning about increasing risk. The appeal of such agreements would be greater if many countries participate. Since a banking crisis in any significant emerging market creates a risk of contagion to other emerging markets, there may be a sufficient international community of interest in an agreement for the creation of an early warning system implemented by an international institution.
Possible International Responses
Improving the Role of the IMF. In the realm of international policies that may help emerging markets cope with international capital market volatility, the key issues relate to the role and operation of the International Monetary Fund (IMF). The following recommendations would not involve fundamental changes in current practice:
1. In recognition of the critical importance of better banking system management, the IMF should more commonly apply stringent conditions with respect to the development of prudential regulations and bank supervisory capacity. Such institutional development has often not received sufficient priority. Whether the IMF takes a direct technical assistance role or relies on the World Bank or regional development banks for that purpose, it should insist on such institutional development as strongly as on standard macroeconomic policy measures. Also, since domestic supervision cannot be fully effective without collaboration from other authorities, the IMF (together with the Basle Committee and the BIS) should strongly encourage international coordination of bank supervision.
2. The IMF should be very strict about the fundamentals of macroeconomic management but more flexible about the specific approaches by which governments achieve agreed outcomes. Feedback from borrowing governments suggests that the IMF bureaucratic culture too often encourages a rigid approach. This seems especially true in the case of the weaker, poorer governments that are eligible to borrow under the Enhanced Structural Adjustment Facility.
3. Taking into account the strong political disincentives that prevent governments from taking early action to avoid a financial crisis, the IMF needs a proactive policy with respect to warnings about vulnerability and possible crisis. There are various possible modalities for doing so?both privately with governments and more publicly. The IMF and its member governments should continue its active exploration of alternatives and should begin implementing a stronger system of early warnings. Of course, the risk is that the IMF, by its warning, would precipitate a crisis that would not have otherwise happened. This risk needs to be balanced against the likelihood that, in the absence of timely corrective action, weakness in the economy would continue to grow, leading to a more severe and more damaging crisis later on.
4. The IMF should have sufficient financial resources to act as a lender of last resort when a country faces a liquidity crisis. Ironically, the growth of international capital markets makes the IMF?s role as a lender of last resort more necessary but also much more problematic. As noted above, emerging markets can easily suffer from the "Brittany effect" in which outflows of capital are disproportionate to any changes in its domestic situation. Systemic crises can happen. There needs to be a strong liquidity facility to help small countries cope with these tides, and the IMF is at present the most potent institution to play this role. However, to be effective, a lender of last resort must meet two criteria: it must have sufficient financial resources to be credible; and it must be able in most circumstances to distinguish between liquidity and solvency problems and to act accordingly. Whether or not the IMF can meet these criteria is debatable. First of all, the size of its current financial resources is dwarfed by the potential magnitude of international capital flows. As a result, the IMF response to an apparent liquidity crisis in a large country requires rapidly mobilizing complementary resources from G-7 governments and/or persuading private creditors to maintain their exposures. Secondly, the IMF may not be able to have a better view about the solvency of a banking system?which of course depends on the solvency of its borrowers?than the superintendent and central bank authorities. When these domestic institutions are weak, the IMF as lender of last resort is almost blind and can easily make mistakes?providing liquidity when instead bankruptcy proceedings of some sort are needed.
More fundamental changes that deserve serious discussion are:
1. The IMF could change its basic approach for determining a country?s eligibility for its financial assistance in favor of "Maastricht-like" criteria that must be satisfied in advance rather than ad hoc conditionality that is designed at the time of a crisis. At present, the IMF generally makes its financial assistance available during a time of crisis, and this assistance is conditional on a set of measures that promise to restore confidence and a viable balance of payments outlook. Although IMF programs typically contain a common set of measures?usually fiscal retrenchment and often devaluation, they are designed ad hoc in response to the circumstances and dimensions of the crisis. In the absence of crisis, the IMF has semi-annual Article IV consultations with member governments about their economic policies, but these tend to be treated as routine. Despite its efforts at "surveillance," the IMF can exert leverage only during a crisis and, in a sense, ends up practicing curative medicine rather than preventive care. An alternative vision of the IMF role might give much more weight to a preventive medicine approach. Its goal would be to tacitly guide international capital flows toward those countries that have a policy and institutional framework that allows resources to be used efficiently. One way to design such an approach would be to define a set of standards for economic policy management?for example, regarding fiscal balance and fiscal management practices, current account balances, external debt structure, inflation, bank regulations and supervisory capacity, tariff regimes, etc.?that would merit IMF approval. Those countries that have met the standards would be declared eligible for large-scale IMF financial support in the event of crisis arising from factors other than failure to maintain the necessary standards. Other countries would be ineligible even if willing to take ex post corrective measures in response to a crisis ?or, less harshly, would have access only to very limited financial assistance under strict conditionality. This "tough love" approach would be based on the premise that those countries that meet the standards but nonetheless face a balance of payments crisis are victims of unfortunate circumstances or of the sins of others. And, conversely, when countries that have not met the standard fall into crisis, it is most likely their own fault and, however repentant, they deserve no help.
2. The IMF could focus its energies on the systemic issues of the international monetary system rather than national economic policies. Such a shift in emphasis would reflect recognition that increasingly many governments have national economists who have a much deeper understanding of their economy than the IMF staff has. While these governments may still value the IMF?s independent viewpoint based on its international perspective, they tend to regard the IMF as one of several sources of such advice. With this in mind, the IMF?s comparative advantage might increasingly lie in its capacity to analyze the international system as a whole, to take a long-term view, and to push the system toward new norms and goals. This could involve analysis of?and public pronouncements about?the consistency of current account imbalances among countries. More ambitiously, as discussed in Box 2, it could involve design, advocacy, and eventually monitoring of a new international regime based on currency blocs (e.g., a US dollar area and a Yen area) in which other regions seek to imitate the recent creation of the Euro.
A World of Euros?
Arguably the new world of massive and instantaneous international capital flows has made it impossible for small nations to manage independent monetary policies. Crises such as those in Mexico and in Asia will be repeated around the world because there are too many separate monies and too many monetary policies. The European Union?s creation of a single currency may point the way to a future in which progressively larger sets of countries adopt a common currency. Achieving such a vision would take both time and intellectual leadership. Perhaps the IMF should give high priority toward providing that intellectual leadership and then creating a framework for its implementation.
Improving Standard Debt Contracts. Currently international capital markets suffer from what may be regarded as an important market imperfection: standard debt contracts do not contain clear provisions governing
.jpg)
