Financial Risks, Stability, and Globalization


Omotunde Johnson
Published Date:
April 2002
  • ShareShare
Show Summary Details

In general, the paper by Sundararajan, Ariyoshi, and Ötker-Robe succeeds in what it sets out to do—that is, it provides a very comprehensive overview of the issues revolving around capital flows and liberalization of the capital account and the impact on an economy’s financial system. This comment is arranged in four main sections, based on points they covered: linkages between capital mobility and financial stability; systemic responses; sequencing of capital account liberalization and supporting policies; and concluding remarks.

The various sections have been extensively researched and provide a survey of recent literature, much of which has been written by the IMF staff. More important, the literature, in most cases, involves empirical work rather than just explorations of theoretical models. As such, the paper by Sundararajan, Ariyoshi, and Ötker-Robe should be viewed as a reference document, providing insights into the latest work on financial crises and their links with international capital mobility.

However, by the same token, the paper does not lend itself to comments regarding its contents precisely because it adheres to its aims so well. It does not appear to be intended as a diagnostic tool for predicting financial crises nor as a prescription for domestic financial policy. Rather, it is simply what it aims to be: it is indeed a comprehensive survey of issues surrounding capital mobility and financial system stability. In this respect, however, the value of the paper as a survey of issues would be enhanced if it had included assessment of prevailing conditions when policy responses have been more effective.

Inclusion of such an assessment on policy responses would not necessarily make the section regarding systemic responses a contentious part of the paper. Instead, such an assessment would provide more insights on the efficacy of measures on management of capital mobility, thereby making more complete the survey of issues surrounding capital mobility and financial stability. In other words, the paper would provide a survey of “lessons” to policymakers, instead of merely enumerating the various policy options that are available to governments. For example, capital controls are mentioned as possible aids in achieving macroeconomic policy objectives. It is important, however, to make an assessment of the macroeconomic and other conditions that must exist for capital controls to be effective with minimal distortionary effects.

The remainder of this comment will explain in more detail sections of the paper, with reference, where possible, to their impact on and implications for small, open economies such as Malaysia and to policy options available to governments.

Introduction and Linkages Between Capital Mobility and Financial Stability

The first sections of the paper by Sundararajan, Ariyoshi, and Ötker-Robe provide a background to justify the focus on international capital mobility as a source of systemic risk to financial systems as well as on the way in which these flows interact with domestic financial systems. However, the authors quite correctly point out that the risks that come from opening the financial system to external capital flows in general are no different than those that may be encountered when a highly regulated financial system is progressively deregulated. In fact, the authors note, “While in some aspects the differences between domestic and external liberalization are a matter of magnitude, the effect can be dramatically larger so as to be qualitatively different” (emphasis added).

In general the risks are as follows.

  • Investment opportunities: The additional opportunities for investment can reduce margins between borrowers and lenders and improve risk diversification but can also lead to excessive risk-taking behavior.
  • Increased liquidity in markets: In general, increased liquidity allows the price mechanism to signal and transmit information more efficiently, but excessive liquidity could increase volatility.
  • Linkages to other economic sectors: The intermediation process allows for more efficient allocation of resources. In a closed but liberalized system, this means greater linkages and risk sharing among participants in markets as well as at the industrial, regional, and even macroeconomic levels. In the matter of cross-border flows, this includes the rest of the world. However, by the same token, the risk of contagion effects is also increased.

However, beyond the matter of magnitude, there are also characteristics of international flows that bring both opportunities and risk, which include the following.

  • Macroeconomic policy can be affected by policies and conditions in other countries.
  • Cross-border flows introduce new risks such as foreign exchange risks, and transfer and settlement risks, all of which may not have been the case in a domestic system.
  • Differences in regulatory regimes could also permit circumvention of regulation and supervision.
  • An open capital account permits the possibility of capital flight.

In enumerating characteristics of cross-border flows, the paper has not distinguished between debt and non-debt flows. This separation is quite important in terms of the different impact that debt and non-debt flows have on management of risks.

The next part of the paper enumerates in detail the advantages and risks associated with capital flows. It is not disputed that one of the benefits of liberalized capital flows is that the market imposes discipline on policymakers as flows are “no longer captive in a country.” However, some qualification is required in that markets also do not assess risks appropriately. As such, measures to avoid overheating of the economy are often thwarted by euphoria of investors and continued inflows, despite macroeconomic measures being instituted to contain the overheating.

The section on risks first discusses risks to macroeconomic stability in relation to both capital inflows and outflows. Of course, problems associated with massive inflows are likely to also see the problems associated with sudden reversal of these inflows, but the effects are qualitatively different. Two important points were made in the studies cited by the authors. First, periods of high inflows that could lead to future instability were characterized by a contemporaneous increase in bank lending as a proportion of GDP because these inflows often cause external reserves to rise, feeding a credit boom.

Second, outflows, which tend to be associated with speculative currency attacks, often cause banking crises due to the sudden withdrawal of liquidity at a time when the financial system is at its most vulnerable. In this regard, some attention should be accorded to recent developments in Asia, where speculation is not confined to the currency markets. Instead, speculation in both the currency and stock markets reinforced each other, rapidly transforming a currency crisis into an economic and then a banking crisis and of course leading to a vicious circle. This development can be directly attributed to liberalization of capital account transactions.

This, in fact, is exactly what Malaysia went through, first in 1993-94 and again in 1997-98. In the first instance, short-term capital inflows could not be sterilized sufficiently quickly and the prudent monetary policy stance at the time was, paradoxically, a pull factor for these inflows. An adjustment in the exchange rate was also considered as a means to discourage inflows. However, allowing the ringgit to appreciate sharply, arising from the inflows of funds that were of a short-term nature, would run the risk of overshooting the exchange rate. This episode eventually culminated in several exchange control measures, such as restriction on sales of short-term monetary instruments to nonresidents and requiring banks to place with the central bank funds of foreign banks in non-interest-bearing vostro accounts. The measures were imposed in January and February 1994 to deal with the highly destabilizing speculative activity and to reassert control over monetary policy.

In 1998, the speculative attacks on regional currencies had led to the withdrawal of liquidity from the domestic financial system. However, use of exchange control measures in Malaysia was to address not excessive capital inflows, but rather outflows of the domestic currency by residents. (Unlike other crisis countries, Malaysia did not suffer outflows of foreign currency from residents. Such outflows were mainly nonresident funds.) Outflows of ringgit were the direct result of early liberalization of capital account transactions leading to an internationalization of the ringgit. During the crisis, the ringgit outflow was attracted by higher interest rates of 20 to 40 percent offered by regional offshore centers, while onshore rates were in the region of 11 percent. The strong demand for offshore ringgit and the consequent buildup of offshore ringgit increased the vulnerability of ringgit and threatened to further destabilize the economy. As the economy was already contracting, raising domestic interest rates to match rates offshore was not tenable. Imposition of selective exchange controls in September 1998 to avoid further internationalizing the ringgit was a final policy option when other measures failed to restore stability to financial markets.

The second major risk is the mismanagement of financial risk. This includes problems that occur with cross-border transactions as enumerated in Table 14.2 in the paper as well as the moral hazard problems that arise from implicit guarantees that have been mentioned earlier in this comment. Even where the source of the excessive lending (and in most cases, the problem manifests itself through excessive risk taking through indiscriminate credit expansion) is private sector activity, the risk is borne by governments. This occurs for two reasons.

  • Removing the implicit guarantee that caused the risk would hurt the private sector at the worst possible time (i.e., the right long-run policy may turn out to be short-term suboptimal), thereby reducing the authorities’ room to maneuver; and
  • Quite often, fixing the problem requires public expenditure (e.g., acquiring nonperforming loans, recapitalization, etc.).

The final risk is the risk of international risks and contagion. The highly interconnected nature of global financial markets demonstrates the importance of events that occur outside the country. Unfortunately, there appears to be little individual governments can do to avoid the impact of these events. The best that seems possible given the current international financial arrangements is for governments to be vigilant regarding regional and global events and thereby take preventative steps. However, the authors acknowledge that international cooperation has a role to play. More important, greater attention is needed to ensure effectiveness of regional surveillance by the IMF.

Another important distinction the authors have made is to differentiate the implications for different countries. The most important distinction to point out is that small, open economies are more likely to be adversely affected by the capital flows while simultaneously losing a large degree of their capacity to institute effective policy responses. Industrial countries, the authors also point out, are less susceptible to the worst effects of crisis, due to their overall creditworthiness, which in turn gives credibility to the continued convertibility of their currencies.

Creditworthiness of a nation in the context of this paper is its ability to honor its guarantee based on its track record. And this is but one of several factors identified in the paper that distinguish emerging and industrial countries. In the final analysis, during a crisis, capital will flee when in doubt of the ability of the country to honor its liabilities. When the private sector defaults, the sovereign’s creditworthiness becomes the primary consideration, as any response mounted by the authorities to restore confidence is dependent on its credibility. To suggest that industrial countries have some “built-in stability” due to “creditworthiness” seems to ignore the capability these economies have to put up shields and institute policies that protect themselves and investors from the worst effects of financial crises.

Capital flight by residents reflects the total erosion of confidence. It is important to establish the cause of the breakdown of the banking system. Is it a reflection or manifestation of the collapse of financial markets and deteriorating economic conditions, or the result of reckless and unsound banking practices? In the case of the former, which was the experience of the Asian crisis, prudential supervision alone may not have been effective in preventing the economic crisis. Nevertheless, robust financial systems would help the banking system to better absorb shocks and keep the problem from blowing up into a crisis, thereby minimizing the risk of capital outflows. In the case of bad banking by individual banks or a group of banks, the problem/banking crisis is likely to be localized. In this situation, effective banking supervision would have a better chance of keeping the crisis from spilling over into an economic downturn. This is particularly the case where funding for the lending is from local sources. However, once there are cross-border capital flows, the threat of reversal is real, including capital flight by residents.

In this regard, it would be interesting to assess the value of high reserves in influencing views of creditworthiness. In the same light, can high reserves offset poor prudential supervision and by itself limit speculative contagion?

Systemic Responses

The main systemic responses listed in the paper are

  • consistent and sustainable macroeconomic policies (i.e., sustainable current and fiscal account position, low inflation, etc.);
  • the institution of strong, market-based prudential measures to ensure adequate risk management;
  • international cooperation and coordination; and
  • capital controls.

As pointed out in the introduction, the final two systemic responses are more important to consider since the macroeconomic stability and adequate prudential safeguards have been universally accepted as necessary (if not sufficient) conditions for reducing risks from financial crises. It is disappointing that international cooperation as explored by the authors does not envision anything more ambitious than cooperation to standardize regulatory regimes, information sharing among regulators, and more transparent data dissemination by authorities (in particular, central banks). While these recommendations are undoubtedly on the right track, the paper pays passing mention to the role of hedge funds in amplifying the severity of the crisis in Asian countries by taking large positions and leverage levels. Hedge funds, by their ability to leverage heavily and attract others to follow, are able to corner or move markets, particularly the smaller and thinner emerging markets. Hence, it is important for developing countries, such as Malaysia, that the paper be more forthcoming with recommendations for the release of information such as informing authorities and other market participants of large positions and leverage levels and a regulatory framework for hedge funds, because hedge funds, unlike other players in the international financial system, are exempted from the securities and exchange oversight requirements.

Second, the other blanket recommendation appears to be that a greater stake in the financial system should be given to foreigners who could then provide a relatively stable presence should things go wrong. This statement is highly prejudicial and unsubstantiated. The more important factor is the health of financial institutions and not whether they are foreign or locally owned. The argument in the paper is valid to the extent that foreigners help to underwrite the risks in a domestic economy. As their well-being is dependent on the economic health of the country in which they are operating, it is in their interest to support the recovery of the economy. Foreign participation does help to raise the level of efficiency and governance in the local industries, including the banking industry. To the extent that about 25 percent of the commercial banks’ assets are held by the 13 foreign banks (100 percent foreign ownership), their customers enjoyed greater certainty during the crisis, as credit lines were not withdrawn by their correspondent banks. Also, as they are governed by parent policies, they are less likely to have misdirected lending and lower nonperforming loans. The better performance of the foreign banks during the crisis may have helped to contain the nonperforming loan problem in the banking system to a more manageable level. Credit lines from banks outside Malaysia were intact for these banks and that helped business to continue as usual. The premiums for these foreign banks would not have been affected if they had been branches rather than locally incorporated banks.

Notwithstanding the above, the larger market presence of foreign banks would not necessarily insulate the financial sector from adverse effects from the crisis. The Malaysian financial sector suffered inefficiencies in bank intermediation during the crisis although foreign banks account for more than 25 percent of assets of the financial sector. This was because decisions on loans tended to be influenced by views at the headquarters. Such views also tended to be based on media reports rather than appropriate assessment of risks based on available data.

Finally, the section concludes that capital controls have a place in the policy options available to governments. Controls are seen to have two purposes—namely, to achieve macroeconomic and prudential objectives. In the paper, controls on capital inflows are seen more positively when they relate to prudential objectives. However, countries that are in crisis situations do not have the luxury of instituting preventative measures when the immediate aim is to deal with the crisis. The paper, however, is silent on the use of controls as a crisis management tool. It would be useful to delve more into the merits and demerits of controls as a “firewall” to mitigate the severity of capital flows during crisis periods. In particular, as mentioned earlier, assessments should also be made on conditions that must exist for controls to be effective “circuit breakers” to overcome financial instability created by volatile capital flows.

I concur that controls, once instituted, are not a cure but merely provide relief for further financial reforms to be carried out. This does not preclude their usefulness. The Malaysian experience shows that authorities have used the breathing space provided by controls to strengthen macroeconomic fundamentals and deepen financial reforms.

Generalizations should not be too broad when assessing controls (e.g., controls on inflows are good and controls on outflows are bad). Analysis of controls should be on a disaggregated basis (disaggregated by specific measures and their implementation details) to fully understand their implications on the economy. Bearing in mind the financial and economic upheaval that short-term capital flows can create, the use of controls (and attendant shortcomings) represent the lesser of two evils.

Sequencing of Capital Account Liberalization and Supporting Policies

While according the importance of sequencing liberalization, the paper acknowledges that opening the capital account for opening’s sake is not good policy. The pace of liberalization is largely one of judgment, and the paper asserts that a controlled approach where authorities are able to identify vulnerabilities and address them through prudential policies is difficult to sustain. Interestingly, the authors do point out that the control over the pace of liberalization becomes weaker as institutions become more sophisticated,1 a point that ought to be noted. Therefore, the big question is how does a regulatory authority make the transition from looking directly at the conduct of business to looking at risk management?

The paper also emphasized that sequencing of liberalization of the various components may not itself be important, but rather it is the establishment of supporting policies that would determine whether liberalization is successful and orderly. While one cannot dispute this, appropriate attention to sequencing itself should not be given less priority. This is because, on practical grounds, it can be risky and administratively impossible to implement too many measures at one time. A step-by-step liberalization allows authorities to measure success and also take necessary additional steps should new risks emerge. This approach allows flexibility in managing adverse effects of liberalization. However, success of a gradualist approach is highly dependent on appropriate sequencing of liberalization.

The authors’ view that it would be difficult to maintain effective capital controls in an open economy with a developed financial system is debatable. Generalizations on controls run the risks of wrong conclusions on the efficacy or sustainability of specific control measures. Controls on internationalization of the domestic currency, for example, can be effectively implemented even in developed financial markets (e.g., Singapore). This is because such measures do not have an impact on the real economy and trade-related transactions.

Concluding Remarks

Creating awareness of work being done is important in evaluating the ability of economies, most notably those of developing countries, in facing crises. This paper, therefore, plays an important role. The authors conclude quite rightly that the important contributor to the inception of crises is not liberalization per se, but rather the inadequate preparation for this opening of the capital account. However, the nondiscriminatory method by which institutional investors assess developing countries during periods of extremely volatile flows must also be given some recognition. Second, the privilege of instituting emergency measures such as targeted controls on flows must be respected. In this context, multilateral institutions such as the IMF have a role to play in standing by good borrowers even when they take unpopular and unorthodox measures if they can show how the measures are to be used. It is premature to react negatively to policy measures that run contrary to the accepted orthodoxy on the basis of what they might do rather than assessing what they will do.

The Malaysian authorities in implementing capital controls had to trade efficiency gains for stability of the economy. In this regard, it would not be too wise to advise authorities “not to be overzealous in attempts to ensure stability.” To the authorities, the cost of giving up efficiency gains is small when compared with the benefit of achieving a stable economy. Moreover, once the economy has stabilized and recovery has occurred, a step-by-step approach to liberalization can be pursued to reinstill the “dynamism and risk taking that ensures economic development.”


Institutions here are taken to mean not only financial institutions such as banks but also investors in capital markets, regulators, and the markets themselves.

    Other Resources Citing This Publication